crash-bt-user-space
by Yulong TANG 汤玉龙
Hi,
Is it possible to display the user stack trace of a process (Not kernel stack
trace of a process) from kernel core dump using crash utility.
#bt <pid>
will display the kernel stack trace of the process pid.
I know extension 'gcore' can generate an process coredump,then use gdb bt to print user stack trace
Is there any command can more quickly display this in crash shell?
Or if I want to develop a command for this, How should I do? To call embeded gdb function?
--Regards
jie ming
10 months, 2 weeks
- 2
- 1
[PATCH v2] add "files -n" command for an inode
by Huang Shijie
In the NUMA machine, it is useful to know the memory distribution of
an inode page cache:
How many pages in the node 0?
How many pages in the node 1?
Add "files -n" command to get the memory distribution information:
1.) Add new argument for dump_inode_page_cache_info()
2.) make page_to_nid() a global function.
3.) Add summary_inode_page() to check each page's node
information.
4.) Use print_inode_summary_info() to print the
memory distribution information of an inode.
Tested with the /proc/kcore.
Signed-off-by: Huang Shijie <shijie(a)os.amperecomputing.com>
---
v1 --> v2:
1.) rebased the code on the latest tree.
2.) added return value for summary_inode_page().
3.) Changed the output format.
4.) Changed the ulong.
5.) others.
---
defs.h | 1 +
filesys.c | 55 ++++++++++++++++++++++++++++++++++++++++++++++++++-----
help.c | 12 +++++++++++-
memory.c | 3 +--
4 files changed, 63 insertions(+), 8 deletions(-)
diff --git a/defs.h b/defs.h
index 788f63a..1fe2d0b 100644
--- a/defs.h
+++ b/defs.h
@@ -5750,6 +5750,7 @@ int dump_inode_page(ulong);
ulong valid_section_nr(ulong);
void display_memory_from_file_offset(ulonglong, long, void *);
void swap_info_init(void);
+int page_to_nid(ulong);
/*
* filesys.c
diff --git a/filesys.c b/filesys.c
index 1d0ee7f..81be108 100644
--- a/filesys.c
+++ b/filesys.c
@@ -49,7 +49,7 @@ static int match_file_string(char *, char *, char *);
static ulong get_root_vfsmount(char *);
static void check_live_arch_mismatch(void);
static long get_inode_nrpages(ulong);
-static void dump_inode_page_cache_info(ulong);
+static void dump_inode_page_cache_info(ulong, void *callback);
#define DENTRY_CACHE (20)
#define INODE_CACHE (20)
@@ -2192,8 +2192,33 @@ get_inode_nrpages(ulong i_mapping)
return nrpages;
}
+/* Used to collect the numa information for an inode */
+static ulong *numa_node;
+
+static void
+print_inode_summary_info(void)
+{
+ int i;
+
+ fprintf(fp, " NODE PAGES\n");
+ for (i = 0; i < vt->numnodes; i++)
+ fprintf(fp, " %2d %8ld\n", i, numa_node[i]);
+}
+
+static int
+summary_inode_page(ulong page)
+{
+ int node = page_to_nid(page);
+
+ if (0 <= node && node < vt->numnodes) {
+ numa_node[node]++;
+ return 1;
+ }
+ return 0;
+}
+
static void
-dump_inode_page_cache_info(ulong inode)
+dump_inode_page_cache_info(ulong inode, void *callback)
{
char *inode_buf;
ulong i_mapping, nrpages, root_rnode, xarray, count;
@@ -2236,7 +2261,7 @@ dump_inode_page_cache_info(ulong inode)
root_rnode = i_mapping + OFFSET(address_space_page_tree);
lp.index = 0;
- lp.value = (void *)&dump_inode_page;
+ lp.value = callback;
if (root_rnode)
count = do_radix_tree(root_rnode, RADIX_TREE_DUMP_CB, &lp);
@@ -2276,7 +2301,7 @@ cmd_files(void)
ref = NULL;
refarg = NULL;
- while ((c = getopt(argcnt, args, "d:R:p:c")) != EOF) {
+ while ((c = getopt(argcnt, args, "d:n:R:p:c")) != EOF) {
switch(c)
{
case 'R':
@@ -2295,11 +2320,31 @@ cmd_files(void)
display_dentry_info(value);
return;
+ case 'n':
+ if (VALID_MEMBER(address_space_page_tree) &&
+ VALID_MEMBER(inode_i_mapping)) {
+ value = htol(optarg, FAULT_ON_ERROR, NULL);
+
+ /* Allocate the array for this inode */
+ numa_node = malloc(sizeof(ulong) * vt->numnodes);
+ BZERO(numa_node, sizeof(ulong) * vt->numnodes);
+
+ dump_inode_page_cache_info(value, (void *)&summary_inode_page);
+
+ /* Print out the NUMA node information for this inode */
+ print_inode_summary_info();
+
+ free(numa_node);
+ numa_node = NULL;
+ } else
+ option_not_supported('n');
+ return;
+
case 'p':
if (VALID_MEMBER(address_space_page_tree) &&
VALID_MEMBER(inode_i_mapping)) {
value = htol(optarg, FAULT_ON_ERROR, NULL);
- dump_inode_page_cache_info(value);
+ dump_inode_page_cache_info(value, (void *)&dump_inode_page);
} else
option_not_supported('p');
return;
diff --git a/help.c b/help.c
index cc7ab20..f0f9139 100644
--- a/help.c
+++ b/help.c
@@ -7850,7 +7850,7 @@ NULL
char *help_files[] = {
"files",
"open files",
-"[-d dentry] | [-p inode] | [-c] [-R reference] [pid | taskp] ... ",
+"[-d dentry] | [-p inode] | [-n inode] | [-c] [-R reference] [pid | taskp] ... ",
" This command displays information about open files of a context.",
" It prints the context's current root directory and current working",
" directory, and then for each open file descriptor it prints a pointer",
@@ -7863,6 +7863,8 @@ char *help_files[] = {
" specific, and only shows the data requested.\n",
" -d dentry given a hexadecimal dentry address, display its inode,",
" super block, file type, and full pathname.",
+" -n inode given a hexadecimal inode address, check all the pages",
+" in the page cache, and display a NUMA node distribution.",
" -p inode given a hexadecimal inode address, dump all of its pages",
" that are in the page cache.",
" -c for each open file descriptor, prints a pointer to its",
@@ -7974,6 +7976,14 @@ char *help_files[] = {
" ca1ddde0 2eeef000 f59b91ac 3 2 82c referenced,uptodate,lru,private",
" ca36b300 3b598000 f59b91ac 4 2 82c referenced,uptodate,lru,private",
" ca202680 30134000 f59b91ac 5 2 82c referenced,uptodate,lru,private",
+" ",
+" %s> files -n ffff07ff8c6f97f8",
+" INODE NRPAGES",
+" ffff07ff8c6f97f8 25240",
+" ",
+" NODE PAGES",
+" 0 25240",
+" 1 0",
" ",
NULL
};
diff --git a/memory.c b/memory.c
index 86ccec5..ed1a4fb 100644
--- a/memory.c
+++ b/memory.c
@@ -300,7 +300,6 @@ static int dump_vm_event_state(void);
static int dump_page_states(void);
static int generic_read_dumpfile(ulonglong, void *, long, char *, ulong);
static int generic_write_dumpfile(ulonglong, void *, long, char *, ulong);
-static int page_to_nid(ulong);
static int get_kmem_cache_list(ulong **);
static int get_kmem_cache_root_list(ulong **);
static int get_kmem_cache_child_list(ulong **, ulong);
@@ -19846,7 +19845,7 @@ is_kmem_cache_addr_common(ulong vaddr, char *kbuf)
/*
* Kernel-config-neutral page-to-node evaluator.
*/
-static int
+int
page_to_nid(ulong page)
{
int i;
--
2.40.1
10 months, 2 weeks
- 1
- 0
[PATCH] gdb: avoid buffer overflow in ada_decode
by Lianbo Jiang
This is a partial backport patch from gdb commit 033bc52bb619 ("Avoid
buffer overflow in ada_decode").
The AddressSanitizer reports a dynamic-stack-buffer-overflow error as
below:
gdb/ada-lang.c:1388:16 in ada_decode[abi:cxx11](char const*, bool, bool)
Add a missing bounds check to fix the current issue.
Link: https://sourceware.org/bugzilla/show_bug.cgi?id=30639
Signed-off-by: Lianbo Jiang <lijiang(a)redhat.com>
---
Please see the CVE-2023-39128.
gdb-10.2.patch | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/gdb-10.2.patch b/gdb-10.2.patch
index 16228b1dbf73..3098c1c3aaaa 100644
--- a/gdb-10.2.patch
+++ b/gdb-10.2.patch
@@ -13,7 +13,8 @@ tar xvzmf gdb-10.2.tar.gz \
gdb-10.2/gdb/printcmd.c \
gdb-10.2/gdb/symfile.c \
gdb-10.2/gdb/Makefile.in \
- gdb-10.2/gdb/dwarf2/read.c
+ gdb-10.2/gdb/dwarf2/read.c \
+ gdb-10.2/gdb/ada-lang.c
exit 0
@@ -3145,3 +3146,14 @@ exit 0
strcat(req->buf, buf);
}
}
+--- gdb-10.2//gdb/ada-lang.c.orig
++++ gdb-10.2/gdb/ada-lang.c
+@@ -1158,7 +1158,7 @@ ada_decode (const char *encoded)
+ i -= 1;
+ if (i > 1 && encoded[i] == '_' && encoded[i - 1] == '_')
+ len0 = i - 1;
+- else if (encoded[i] == '$')
++ else if (i >= 0 && encoded[i] == '$')
+ len0 = i;
+ }
+
--
2.41.0
10 months, 2 weeks
- 2
- 1
[PATCH] add "files -n" command for an inode
by Huang Shijie
In the NUMA machine, it is useful to know the memory distribution of
an inode page cache:
How many pages in the node 0?
How many pages in the node 1?
Add "files -n" command to get the memory distribution information:
1.) Add new argument for dump_inode_page_cache_info()
2.) make page_to_nid() a global function.
3.) Add sumary_inode_page() to check each page's node
information.
4.) Use print_inode_sumary_info() to print the
memory distribution information of an inode.
Signed-off-by: Huang Shijie <shijie(a)os.amperecomputing.com>
---
defs.h | 1 +
filesys.c | 52 +++++++++++++++++++++++++++++++++++++++++++++++-----
help.c | 10 ++++++++++
memory.c | 3 +--
4 files changed, 59 insertions(+), 7 deletions(-)
diff --git a/defs.h b/defs.h
index 96a7a2a..dfcc33d 100644
--- a/defs.h
+++ b/defs.h
@@ -5748,6 +5748,7 @@ int dump_inode_page(ulong);
ulong valid_section_nr(ulong);
void display_memory_from_file_offset(ulonglong, long, void *);
void swap_info_init(void);
+int page_to_nid(ulong);
/*
* filesys.c
diff --git a/filesys.c b/filesys.c
index 1d0ee7f..3c6a6bd 100644
--- a/filesys.c
+++ b/filesys.c
@@ -49,7 +49,7 @@ static int match_file_string(char *, char *, char *);
static ulong get_root_vfsmount(char *);
static void check_live_arch_mismatch(void);
static long get_inode_nrpages(ulong);
-static void dump_inode_page_cache_info(ulong);
+static void dump_inode_page_cache_info(ulong, void *callback);
#define DENTRY_CACHE (20)
#define INODE_CACHE (20)
@@ -2192,8 +2192,30 @@ get_inode_nrpages(ulong i_mapping)
return nrpages;
}
+/* Used to collect the numa information for an inode */
+static unsigned long *numa_node;
+
+static void
+print_inode_sumary_info(void)
+{
+ int i;
+
+ fprintf(fp, "The physical memory in the page cache:\n");
+ for (i = 0; i < vt->numnodes; i++)
+ fprintf(fp, "\tNode[%d]: %16ld pages\n", i, numa_node[i]);
+}
+
+static int
+sumary_inode_page(ulong page)
+{
+ int node = page_to_nid(page);
+
+ if (node >= 0)
+ numa_node[node]++;
+}
+
static void
-dump_inode_page_cache_info(ulong inode)
+dump_inode_page_cache_info(ulong inode, void *callback)
{
char *inode_buf;
ulong i_mapping, nrpages, root_rnode, xarray, count;
@@ -2236,7 +2258,7 @@ dump_inode_page_cache_info(ulong inode)
root_rnode = i_mapping + OFFSET(address_space_page_tree);
lp.index = 0;
- lp.value = (void *)&dump_inode_page;
+ lp.value = callback;
if (root_rnode)
count = do_radix_tree(root_rnode, RADIX_TREE_DUMP_CB, &lp);
@@ -2276,7 +2298,7 @@ cmd_files(void)
ref = NULL;
refarg = NULL;
- while ((c = getopt(argcnt, args, "d:R:p:c")) != EOF) {
+ while ((c = getopt(argcnt, args, "d:n:R:p:c")) != EOF) {
switch(c)
{
case 'R':
@@ -2295,11 +2317,31 @@ cmd_files(void)
display_dentry_info(value);
return;
+ case 'n':
+ if (VALID_MEMBER(address_space_page_tree) &&
+ VALID_MEMBER(inode_i_mapping)) {
+ value = htol(optarg, FAULT_ON_ERROR, NULL);
+
+ /* Allocate the array for this inode */
+ numa_node = malloc(sizeof(unsigned long) * vt->numnodes);
+ BZERO(numa_node, sizeof(unsigned long) * vt->numnodes);
+
+ dump_inode_page_cache_info(value, (void *)&sumary_inode_page);
+
+ /* Print out the NUMA node information for this inode */
+ print_inode_sumary_info();
+
+ free(numa_node);
+ numa_node = NULL;
+ } else
+ option_not_supported('n');
+ return;
+
case 'p':
if (VALID_MEMBER(address_space_page_tree) &&
VALID_MEMBER(inode_i_mapping)) {
value = htol(optarg, FAULT_ON_ERROR, NULL);
- dump_inode_page_cache_info(value);
+ dump_inode_page_cache_info(value, (void *)&dump_inode_page);
} else
option_not_supported('p');
return;
diff --git a/help.c b/help.c
index cc7ab20..9b70940 100644
--- a/help.c
+++ b/help.c
@@ -7863,6 +7863,8 @@ char *help_files[] = {
" specific, and only shows the data requested.\n",
" -d dentry given a hexadecimal dentry address, display its inode,",
" super block, file type, and full pathname.",
+" -n inode given a hexadecimal inode address, check all the pages",
+" in the page cache, and give a NUMA node distribution.",
" -p inode given a hexadecimal inode address, dump all of its pages",
" that are in the page cache.",
" -c for each open file descriptor, prints a pointer to its",
@@ -7974,6 +7976,14 @@ char *help_files[] = {
" ca1ddde0 2eeef000 f59b91ac 3 2 82c referenced,uptodate,lru,private",
" ca36b300 3b598000 f59b91ac 4 2 82c referenced,uptodate,lru,private",
" ca202680 30134000 f59b91ac 5 2 82c referenced,uptodate,lru,private",
+" ",
+" %s> files -n ffff000118c49140",
+" INODE NRPAGES",
+" ffff000118c49140 985",
+" ",
+" The physical memory in the page cache:",
+" Node[0]: 0 pages",
+" Node[1]: 985 pages",
" ",
NULL
};
diff --git a/memory.c b/memory.c
index 86ccec5..ed1a4fb 100644
--- a/memory.c
+++ b/memory.c
@@ -300,7 +300,6 @@ static int dump_vm_event_state(void);
static int dump_page_states(void);
static int generic_read_dumpfile(ulonglong, void *, long, char *, ulong);
static int generic_write_dumpfile(ulonglong, void *, long, char *, ulong);
-static int page_to_nid(ulong);
static int get_kmem_cache_list(ulong **);
static int get_kmem_cache_root_list(ulong **);
static int get_kmem_cache_child_list(ulong **, ulong);
@@ -19846,7 +19845,7 @@ is_kmem_cache_addr_common(ulong vaddr, char *kbuf)
/*
* Kernel-config-neutral page-to-node evaluator.
*/
-static int
+int
page_to_nid(ulong page)
{
int i;
--
2.39.2
10 months, 2 weeks
- 3
- 2
Re: [Crash-utility] [PATCH v1 01/10] Add LoongArch64 framework code support
by lijiang
Hi, Ming
So sorry for the late reply.
On Thu, Sep 7, 2023 at 9:56 AM <crash-utility-request(a)redhat.com> wrote:
> Date: Thu, 7 Sep 2023 09:46:42 +0800
> From: Ming Wang <wangming01(a)loongson.cn>
> To: crash-utility(a)redhat.com
> Cc: yangtiezhu(a)loongson.cn, lixuefeng(a)loongson.cn,
> chenhuacai(a)kernel.org
> Subject: [Crash-utility] [PATCH v1 01/10] Add LoongArch64 framework
> code support
> Message-ID: <20230907014651.167784-2-wangming01(a)loongson.cn>
> Content-Type: text/plain; charset=UTF-8
>
> Mainly added some environment configurations, macro definitions, specific
> architecture structures and some function declarations supported by the
> LoongArch64 architecture.
>
> Co-developed-by: Youling Tang <tangyouling(a)loongson.cn>
> Signed-off-by: Youling Tang <tangyouling(a)loongson.cn>
> Signed-off-by: Ming Wang <wangming01(a)loongson.cn>
> ---
> Makefile | 9 +-
> configure.c | 27 +-
> defs.h | 185 +-
> diskdump.c | 19 +-
> gdb-10.2-loongarch.patch | 15206 +++++++++++++++++++++++++++++++++++++
> lkcd_vmdump_v1.h | 2 +-
> lkcd_vmdump_v2_v3.h | 5 +-
> loongarch64.c | 46 +
> netdump.c | 23 +
> ramdump.c | 2 +
> symbols.c | 23 +-
> 11 files changed, 15538 insertions(+), 9 deletions(-)
> create mode 100644 gdb-10.2-loongarch.patch
> create mode 100644 loongarch64.c
>
> diff --git a/Makefile b/Makefile
> index a94a243..471929a 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -65,7 +65,7 @@ CFILES=main.c tools.c global_data.c memory.c filesys.c
> help.c task.c \
> kernel.c test.c gdb_interface.c configure.c net.c dev.c bpf.c \
> printk.c \
> alpha.c x86.c ppc.c ia64.c s390.c s390x.c s390dbf.c ppc64.c
> x86_64.c \
> - arm.c arm64.c mips.c mips64.c riscv64.c sparc64.c \
> + arm.c arm64.c mips.c mips64.c riscv64.c loongarch64.c sparc64.c \
> extensions.c remote.c va_server.c va_server_v1.c symbols.c
> cmdline.c \
> lkcd_common.c lkcd_v1.c lkcd_v2_v3.c lkcd_v5.c lkcd_v7.c lkcd_v8.c\
> lkcd_fix_mem.c s390_dump.c lkcd_x86_trace.c \
> @@ -85,7 +85,7 @@ OBJECT_FILES=main.o tools.o global_data.o memory.o
> filesys.o help.o task.o \
> build_data.o kernel.o test.o gdb_interface.o net.o dev.o bpf.o \
> printk.o \
> alpha.o x86.o ppc.o ia64.o s390.o s390x.o s390dbf.o ppc64.o
> x86_64.o \
> - arm.o arm64.o mips.o mips64.o riscv64.o sparc64.o \
> + arm.o arm64.o mips.o mips64.o riscv64.o loongarch64.o sparc64.o \
> extensions.o remote.o va_server.o va_server_v1.o symbols.o
> cmdline.o \
> lkcd_common.o lkcd_v1.o lkcd_v2_v3.o lkcd_v5.o lkcd_v7.o lkcd_v8.o
> \
> lkcd_fix_mem.o s390_dump.o netdump.o diskdump.o makedumpfile.o
> xendump.o \
> @@ -293,6 +293,8 @@ gdb_unzip:
> gdb_patch:
> if [ -f ${GDB}.patch ] && [ -s ${GDB}.patch ]; then \
> patch -p0 < ${GDB}.patch; cp ${GDB}.patch ${GDB}; fi
> + if [ -f ${GDB}-loongarch.patch ] && [ -s ${GDB}-loongarch.patch ];
> then \
> + patch -p0 < ${GDB}-loongarch.patch; cp
> ${GDB}-loongarch.patch ${GDB}; fi
>
>
This means we have to maintain a specific gdb patch for loongarch. Although
a similar method still exists in the crash-7-branch, I would not suggest
doing that in crash-8(unless this is the optimal solution).
Ideally, you could backport the relevant patches based on the current
gdb-10.2.patch, and remove redundant(or unnecessary) changes, which makes
the gdb patches less changes.
library: ${OBJECT_FILES}
> ar -rs ${PROGRAM}lib.a ${OBJECT_FILES}
> @@ -445,6 +447,9 @@ riscv64.o: ${GENERIC_HFILES} ${REDHAT_HFILES} riscv64.c
> sparc64.o: ${GENERIC_HFILES} ${REDHAT_HFILES} sparc64.c
> ${CC} -c ${CRASH_CFLAGS} sparc64.c ${WARNING_OPTIONS}
> ${WARNING_ERROR}
>
> +loongarch64.o: ${GENERIC_HFILES} ${REDHAT_HFILES} loongarch64.c
> + ${CC} -c ${CRASH_CFLAGS} loongarch64.c ${WARNING_OPTIONS}
> ${WARNING_ERROR}
> +
> s390.o: ${GENERIC_HFILES} ${IBM_HFILES} s390.c
> ${CC} -c ${CRASH_CFLAGS} s390.c ${WARNING_OPTIONS} ${WARNING_ERROR}
>
> diff --git a/configure.c b/configure.c
> index 08b52be..d6bf738 100644
> --- a/configure.c
> +++ b/configure.c
> @@ -108,6 +108,7 @@ void add_extra_lib(char *);
> #undef SPARC64
> #undef MIPS64
> #undef RISCV64
> +#undef LOONGARCH64
>
> #define UNKNOWN 0
> #define X86 1
> @@ -124,6 +125,7 @@ void add_extra_lib(char *);
> #define SPARC64 12
> #define MIPS64 13
> #define RISCV64 14
> +#define LOONGARCH64 15
>
> #define TARGET_X86 "TARGET=X86"
> #define TARGET_ALPHA "TARGET=ALPHA"
> @@ -139,6 +141,7 @@ void add_extra_lib(char *);
> #define TARGET_MIPS64 "TARGET=MIPS64"
> #define TARGET_SPARC64 "TARGET=SPARC64"
> #define TARGET_RISCV64 "TARGET=RISCV64"
> +#define TARGET_LOONGARCH64 "TARGET=LOONGARCH64"
>
> #define TARGET_CFLAGS_X86 "TARGET_CFLAGS=-D_FILE_OFFSET_BITS=64"
> #define TARGET_CFLAGS_ALPHA "TARGET_CFLAGS="
> @@ -163,6 +166,7 @@ void add_extra_lib(char *);
> #define TARGET_CFLAGS_SPARC64 "TARGET_CFLAGS="
> #define TARGET_CFLAGS_RISCV64 "TARGET_CFLAGS="
> #define TARGET_CFLAGS_RISCV64_ON_X86_64 "TARGET_CFLAGS="
> +#define TARGET_CFLAGS_LOONGARCH64 "TARGET_CFLAGS="
>
> #define GDB_TARGET_DEFAULT "GDB_CONF_FLAGS="
> #define GDB_TARGET_ARM_ON_X86 "GDB_CONF_FLAGS=--target=arm-elf-linux"
> @@ -413,6 +417,9 @@ get_current_configuration(struct supported_gdb_version
> *sp)
> #if defined(__riscv) && (__riscv_xlen == 64)
> target_data.target = RISCV64;
> #endif
> +#ifdef __loongarch64
> + target_data.target = LOONGARCH64;
> +#endif
>
> set_initial_target(sp);
>
> @@ -512,6 +519,10 @@ get_current_configuration(struct
> supported_gdb_version *sp)
> (target_data.target != MIPS64))
> arch_mismatch(sp);
>
> + if ((target_data.initial_gdb_target == LOONGARCH64) &&
> + (target_data.target != LOONGARCH64))
> + arch_mismatch(sp);
> +
> if ((target_data.initial_gdb_target == RISCV64) &&
> (target_data.target != RISCV64)) {
> if (target_data.target == X86_64)
> @@ -686,6 +697,9 @@ show_configuration(void)
> case RISCV64:
> printf("TARGET: RISCV64\n");
> break;
> + case LOONGARCH64:
> + printf("TARGET: LOONGARCH64\n");
> + break;
> }
>
> if (strlen(target_data.program)) {
> @@ -811,6 +825,10 @@ build_configure(struct supported_gdb_version *sp)
> } else
> target_CFLAGS = TARGET_CFLAGS_RISCV64;
> break;
> + case LOONGARCH64:
> + target = TARGET_LOONGARCH64;
> + target_CFLAGS = TARGET_CFLAGS_LOONGARCH64;
> + break;
> }
>
> ldflags = get_extra_flags("LDFLAGS.extra", NULL);
> @@ -1408,7 +1426,7 @@ make_spec_file(struct supported_gdb_version *sp)
> printf("Vendor: Red Hat, Inc.\n");
> printf("Packager: Dave Anderson <anderson(a)redhat.com>\n");
> printf("ExclusiveOS: Linux\n");
> - printf("ExclusiveArch: %%{ix86} alpha ia64 ppc ppc64 ppc64pseries
> ppc64iseries x86_64 s390 s390x arm aarch64 ppc64le mips mipsel mips64el
> sparc64 riscv64\n");
> + printf("ExclusiveArch: %%{ix86} alpha ia64 ppc ppc64 ppc64pseries
> ppc64iseries x86_64 s390 s390x arm aarch64 ppc64le mips mipsel mips64el
> sparc64 riscv64 loongarch64\n");
> printf("Buildroot: %%{_tmppath}/%%{name}-root\n");
> printf("BuildRequires: ncurses-devel zlib-devel bison\n");
> printf("Requires: binutils\n");
> @@ -1649,6 +1667,8 @@ set_initial_target(struct supported_gdb_version *sp)
> target_data.initial_gdb_target = SPARC64;
> else if (strncmp(buf, "RISCV64", strlen("RISCV64")) == 0)
> target_data.initial_gdb_target = RISCV64;
> + else if (strncmp(buf, "LOONGARCH64", strlen("LOONGARCH64")) == 0)
> + target_data.initial_gdb_target = LOONGARCH64;
> }
>
> char *
> @@ -1670,6 +1690,7 @@ target_to_name(int target)
> case MIPS64: return("MIPS64");
> case SPARC64: return("SPARC64");
> case RISCV64: return("RISCV64");
> + case LOONGARCH64: return("LOONGARCH64");
> }
>
>
It doesn't support building the loongarch crash binary on an X86 64 host,
right?
> return "UNKNOWN";
> @@ -1738,6 +1759,10 @@ name_to_target(char *name)
> return RISCV64;
> else if (strncmp(name, "riscv64", strlen("riscv64")) == 0)
> return RISCV64;
> + else if (strncmp(name, "loongarch64", strlen("loongarch64")) == 0)
> + return LOONGARCH64;
> + else if (strncmp(name, "LOONGARCH64", strlen("LOONGARCH64")) == 0)
> + return LOONGARCH64;
>
> return UNKNOWN;
> }
> diff --git a/defs.h b/defs.h
> index 96a7a2a..54d5141 100644
> --- a/defs.h
> +++ b/defs.h
> @@ -76,7 +76,7 @@
> #if !defined(X86) && !defined(X86_64) && !defined(ALPHA) && !defined(PPC)
> && \
> !defined(IA64) && !defined(PPC64) && !defined(S390) &&
> !defined(S390X) && \
> !defined(ARM) && !defined(ARM64) && !defined(MIPS) &&
> !defined(MIPS64) && \
> - !defined(RISCV64) && !defined(SPARC64)
> + !defined(RISCV64) && !defined(LOONGARCH64) && !defined(SPARC64)
> #ifdef __alpha__
> #define ALPHA
> #endif
> @@ -121,6 +121,9 @@
> #if defined(__riscv) && (__riscv_xlen == 64)
> #define RISCV64
> #endif
> +#ifdef __loongarch64
> +#define LOONGARCH64
> +#endif
> #endif
>
> #ifdef X86
> @@ -165,6 +168,9 @@
> #ifdef RISCV64
> #define NR_CPUS (256)
> #endif
> +#ifdef LOONGARCH64
> +#define NR_CPUS (256)
> +#endif
>
> #define NR_DEVICE_DUMPS (64)
>
> @@ -2015,6 +2021,8 @@ struct offset_table { /* stash of
> commonly-used offsets */
> long atomic_t_counter;
> long percpu_counter_count;
> long mm_struct_mm_count;
> + long task_struct_thread_reg01;
> + long task_struct_thread_reg03;
>
They have to be appended to the end of the tables, see the writing patch
section:
https://github.com/crash-utility/crash/wiki
> long task_struct_thread_reg29;
> long task_struct_thread_reg31;
> long pt_regs_regs;
> @@ -3713,6 +3721,48 @@ typedef signed int s32;
>
> #endif /* RISCV64 */
>
> +/* fix compilation errors due to elf.h version. */
> +#ifndef EM_LOONGARCH
> +#define EM_LOONGARCH 258
> +#endif
> +
> +#ifdef LOONGARCH64
> +#define _64BIT_
> +#define MACHINE_TYPE "LOONGARCH64"
> +
> +#define PAGEBASE(X) (((ulong)(X)) & (ulong)machdep->pagemask)
> +
> +#define IS_XKPRANGE(X) (((X) >= 0x8000000000000000lu) && \
> + ((X) < 0xc000000000000000lu))
> +
> +#define PTOV(X) ((ulong)(X) + 0x9000000000000000lu)
> +#define VTOP(X) ((ulong)(X) & 0x0000fffffffffffflu)
> +
> +#define IS_VMALLOC_ADDR(X) (vt->vmalloc_start && (ulong)(X) >=
> vt->vmalloc_start)
> +
> +#define DEFAULT_MODULES_VADDR 0xffff800000000000lu
> +#define MODULES_VADDR (machdep->machspec->modules_vaddr)
> +#define MODULES_END (machdep->machspec->modules_end)
> +#define VMALLOC_START (machdep->machspec->vmalloc_start_addr)
> +#define VMALLOC_END (machdep->machspec->vmalloc_end)
> +
> +#define __SWP_TYPE_SHIFT 16
> +#define __SWP_TYPE_BITS 8
> +#define __SWP_TYPE_MASK ((1 << __SWP_TYPE_BITS) - 1)
> +#define __SWP_OFFSET_SHIFT (__SWP_TYPE_BITS + __SWP_TYPE_SHIFT)
> +
> +#define SWP_TYPE(entry) (((entry) >> __SWP_TYPE_SHIFT) &
> __SWP_TYPE_MASK)
> +#define SWP_OFFSET(entry) ((entry) >> __SWP_OFFSET_SHIFT)
> +
> +#define __swp_type(entry) SWP_TYPE(entry)
> +#define __swp_offset(entry) SWP_OFFSET(entry)
> +
> +#define TIF_SIGPENDING (1)
> +
> +#define _SECTION_SIZE_BITS 28
> +#define _MAX_PHYSMEM_BITS 48
> +#endif /* LOONGARCH64 */
> +
> #ifdef X86
> #define _32BIT_
> #define MACHINE_TYPE "X86"
> @@ -4765,6 +4815,10 @@ struct machine_specific {
> #define MAX_HEXADDR_STRLEN (16)
> #define UVADDR_PRLEN (16)
> #endif
> +#ifdef LOONGARCH64
> +#define MAX_HEXADDR_STRLEN (16)
> +#define UVADDR_PRLEN (16)
> +#endif
>
> #define BADADDR ((ulong)(-1))
> #define BADVAL ((ulong)(-1))
> @@ -5386,6 +5440,9 @@ void dump_build_data(void);
> #ifdef SPARC64
> #define machdep_init(X) sparc64_init(X)
> #endif
> +#ifdef LOONGARCH64
> +#define machdep_init(X) loongarch64_init(X)
> +#endif
> int clean_exit(int);
> int untrusted_file(FILE *, char *);
> char *readmem_function_name(void);
> @@ -5877,6 +5934,10 @@ void display_help_screen(char *);
> #ifdef RISCV64
> #define dump_machdep_table(X) riscv64_dump_machdep_table(X)
> #endif
> +#ifdef LOONGARCH64
> +#define dump_machdep_table(X) loongarch64_dump_machdep_table(X)
> +#endif
> +
> extern char *help_pointer[];
> extern char *help_alias[];
> extern char *help_ascii[];
> @@ -7081,6 +7142,128 @@ int sparc64_vmalloc_addr(ulong);
> error(FATAL, "The -d option is not applicable to sparc64.\n")
> #endif
>
> +/*
> + * loongarch64.c
> + */
> +void loongarch64_display_regs_from_elf_notes(int, FILE *);
> +#ifdef LOONGARCH64
> +void loongarch64_init(int);
> +void loongarch64_dump_machdep_table(ulong);
> +
> +#define display_idt_table() \
> + error(FATAL, "-d option is not applicable to LOONGARCH64
> architecture\n")
> +
> +/* from arch/loongarch/include/asm/ptrace.h */
> +struct loongarch64_pt_regs {
> + /* Saved main processor registers. */
> + unsigned long regs[32];
> +
> + /* Saved special registers. */
> + unsigned long csr_crmd;
> + unsigned long csr_prmd;
> + unsigned long csr_euen;
> + unsigned long csr_ecfg;
> + unsigned long csr_estat;
> + unsigned long csr_epc;
> + unsigned long csr_badvaddr;
> + unsigned long orig_a0;
> +};
> +
> +struct loongarch64_unwind_frame {
> + unsigned long sp;
> + unsigned long pc;
> + unsigned long ra;
> +};
>
If the above definitions are only used for the loongarch64.c, I would
suggest moving them somewhere in the loongarch64.c.
+
> +#define KSYMS_START (0x1)
> +
> +struct machine_specific {
> + ulong phys_base;
> + ulong vmalloc_start_addr;
> + ulong modules_vaddr;
> + ulong modules_end;
> +
> + struct loongarch64_pt_regs *crash_task_regs;
> +};
> +
> +/*
> + * Basic page table format:
> + *
> + * 63 62 61 PALEN-1 12 10 9 8 7 6 5 4 3 2 1 0
> + * +----+--+--+------+--------------------+----+--+--+-+-+-+---+---+-+-+
> + * |RPLV|NX|NR| | PA[PALEN-1:12] | |SP|PN|W|P|G|MAT|PLV|D|V|
> + * +----+--+--+------+--------------------+----+--+--+-+-+-+---+---+-+-+
> + *
> + *
> + * Huge page table format:
> + *
> + * 63 62 61 PALEN-1 12 10 9 8 7 6 5 4 3 2 1 0
> + * +----+--+--+------+-----------------+--+----+--+--+-+-+-+---+---+-+-+
> + * |RPLV|NX|NR| | PA[PALEN-1:12] | G| |SP|PN|W|P|H|MAT|PLV|D|V|
> + * +----+--+--+------+-----------------+--+----+--+--+-+-+-+---+---+-+-+
> + *
> + */
> +/* from arch/loongarch/include/asm/pgtable-bits.h */
> +
> +/* Page table bits */
> +#define _PAGE_VALID_SHIFT 0
> +#define _PAGE_ACCESSED_SHIFT 0 /* Reuse Valid for Accessed */
> +#define _PAGE_DIRTY_SHIFT 1
> +#define _PAGE_PLV_SHIFT 2 /* 2~3, two bits */
> +#define _CACHE_SHIFT 4 /* 4~5, two bits */
> +#define _PAGE_GLOBAL_SHIFT 6
> +#define _PAGE_HUGE_SHIFT 6 /* HUGE is a PMD bit */
> +#define _PAGE_PRESENT_SHIFT 7
> +#define _PAGE_WRITE_SHIFT 8
> +#define _PAGE_MODIFIED_SHIFT 9
> +#define _PAGE_PROTNONE_SHIFT 10
> +#define _PAGE_SPECIAL_SHIFT 11
> +#define _PAGE_HGLOBAL_SHIFT 12 /* HGlobal is a PMD bit */
> +#define _PAGE_PFN_SHIFT 12
> +#define _PAGE_SWP_EXCLUSIVE_SHIFT 23
> +#define _PAGE_PFN_END_SHIFT 48
> +#define _PAGE_PRESENT_INVALID_SHIFT 60
> +#define _PAGE_NO_READ_SHIFT 61
> +#define _PAGE_NO_EXEC_SHIFT 62
> +#define _PAGE_RPLV_SHIFT 63
> +
> +#ifndef _ULCAST_
> +#define _ULCAST_ (unsigned long)
> +#endif
> +
> +/* Used by software */
> +#define _PAGE_PRESENT (_ULCAST_(1) << _PAGE_PRESENT_SHIFT)
> +#define _PAGE_PRESENT_INVALID (_ULCAST_(1) <<
> _PAGE_PRESENT_INVALID_SHIFT)
> +#define _PAGE_WRITE (_ULCAST_(1) << _PAGE_WRITE_SHIFT)
> +#define _PAGE_ACCESSED (_ULCAST_(1) << _PAGE_ACCESSED_SHIFT)
> +#define _PAGE_MODIFIED (_ULCAST_(1) << _PAGE_MODIFIED_SHIFT)
> +#define _PAGE_PROTNONE (_ULCAST_(1) << _PAGE_PROTNONE_SHIFT)
> +#define _PAGE_SPECIAL (_ULCAST_(1) << _PAGE_SPECIAL_SHIFT)
> +
> +/* We borrow bit 23 to store the exclusive marker in swap PTEs. */
> +#define _PAGE_SWP_EXCLUSIVE (_ULCAST_(1) << _PAGE_SWP_EXCLUSIVE_SHIFT)
> +
> +/* Used by TLB hardware (placed in EntryLo*) */
> +#define _PAGE_VALID (_ULCAST_(1) << _PAGE_VALID_SHIFT)
> +#define _PAGE_DIRTY (_ULCAST_(1) << _PAGE_DIRTY_SHIFT)
> +#define _PAGE_PLV (_ULCAST_(3) << _PAGE_PLV_SHIFT)
> +#define _PAGE_GLOBAL (_ULCAST_(1) << _PAGE_GLOBAL_SHIFT)
> +#define _PAGE_HUGE (_ULCAST_(1) << _PAGE_HUGE_SHIFT)
> +#define _PAGE_HGLOBAL (_ULCAST_(1) << _PAGE_HGLOBAL_SHIFT)
> +#define _PAGE_NO_READ (_ULCAST_(1) << _PAGE_NO_READ_SHIFT)
> +#define _PAGE_NO_EXEC (_ULCAST_(1) << _PAGE_NO_EXEC_SHIFT)
> +#define _PAGE_RPLV (_ULCAST_(1) << _PAGE_RPLV_SHIFT)
> +#define _CACHE_MASK (_ULCAST_(3) << _CACHE_SHIFT)
> +#define _PFN_SHIFT (PAGESHIFT() - 12 + _PAGE_PFN_SHIFT)
> +
> +#define _PAGE_USER (PLV_USER << _PAGE_PLV_SHIFT)
> +#define _PAGE_KERN (PLV_KERN << _PAGE_PLV_SHIFT)
> +
> +#define _PFN_MASK (~((_ULCAST_(1) << (_PFN_SHIFT)) - 1) & \
> + ((_ULCAST_(1) << (_PAGE_PFN_END_SHIFT)) - 1))
> +
> +#endif /* LOONGARCH64 */
> +
> /*
> * netdump.c
> */
> diff --git a/diskdump.c b/diskdump.c
> index 2c284ff..94b98b5 100644
> --- a/diskdump.c
> +++ b/diskdump.c
> @@ -674,6 +674,9 @@ restart:
> else if (STRNEQ(header->utsname.machine, "riscv64") &&
> machine_type_mismatch(file, "RISCV64", NULL, 0))
> goto err;
> + else if (STRNEQ(header->utsname.machine, "loongarch64") &&
> + machine_type_mismatch(file, "LOONGARCH64", NULL, 0))
> + goto err;
>
> if (header->block_size != block_size) {
> block_size = header->block_size;
> @@ -834,6 +837,8 @@ restart:
> dd->machine_type = EM_SPARCV9;
> else if (machine_type("RISCV64"))
> dd->machine_type = EM_RISCV;
> + else if (machine_type("LOONGARCH64"))
> + dd->machine_type = EM_LOONGARCH;
> else {
> error(INFO, "%s: unsupported machine type: %s\n",
> DISKDUMP_VALID() ? "diskdump" : "compressed kdump",
> @@ -1593,6 +1598,12 @@ get_diskdump_regs_riscv64(struct bt_info *bt, ulong
> *eip, ulong *esp)
> machdep->get_stack_frame(bt, eip, esp);
> }
>
> +static void
> +get_diskdump_regs_loongarch64(struct bt_info *bt, ulong *eip, ulong *esp)
> +{
> + machdep->get_stack_frame(bt, eip, esp);
> +}
> +
> static void
> get_diskdump_regs_sparc64(struct bt_info *bt, ulong *eip, ulong *esp)
> {
> @@ -1676,6 +1687,10 @@ get_diskdump_regs(struct bt_info *bt, ulong *eip,
> ulong *esp)
> get_diskdump_regs_riscv64(bt, eip, esp);
> break;
>
> + case EM_LOONGARCH:
> + get_diskdump_regs_loongarch64(bt, eip, esp);
> + break;
> +
> default:
> error(FATAL, "%s: unsupported machine type: %s\n",
> DISKDUMP_VALID() ? "diskdump" : "compressed kdump",
> @@ -1823,7 +1838,7 @@ dump_note_offsets(FILE *fp)
> if (machine_type("X86_64") ||
> machine_type("S390X") ||
> machine_type("ARM64") || machine_type("PPC64")
> ||
> machine_type("SPARC64") ||
> machine_type("MIPS64") ||
> - machine_type("RISCV64")) {
> + machine_type("RISCV64") ||
> machine_type("LOONGARCH64")) {
> note64 = (void *)dd->notes_buf + tot;
> len = sizeof(Elf64_Nhdr);
> if (STRNEQ((char *)note64 + len, "QEMU"))
> @@ -1934,6 +1949,8 @@ __diskdump_memory_dump(FILE *fp)
> fprintf(fp, "(EM_AARCH64)\n"); break;
> case EM_SPARCV9:
> fprintf(fp, "(EM_SPARCV9)\n"); break;
> + case EM_LOONGARCH:
> + fprintf(fp, "(EM_LOONGARCH)\n"); break;
> default:
> fprintf(fp, "(unknown)\n"); break;
> }
> diff --git a/gdb-10.2-loongarch.patch b/gdb-10.2-loongarch.patch
> new file mode 100644
> index 0000000..82c1640
> --- /dev/null
> +++ b/gdb-10.2-loongarch.patch
> @@ -0,0 +1,15206 @@
> +From 6a84c2ee3967892e476338f8f4b85617ced2c4bb Mon Sep 17 00:00:00 2001
> +From: Ming Wang <wangming01(a)loongson.cn>
> +Date: Sat, 8 Jul 2023 15:07:21 +0800
> +Subject: [PATCH] gdb: Add LoongArch support.
> +
> +Signed-off-by: Ming Wang <wangming01(a)loongson.cn>
> +---
> + bfd/Makefile.am | 17 +-
> + bfd/Makefile.in | 21 +-
> + bfd/archures.c | 5 +
> + bfd/bfd-in2.h | 85 +
> + bfd/config.bfd | 15 +
> + bfd/configure | 2 +
> + bfd/configure.ac | 2 +
> + bfd/cpu-loongarch.c | 61 +
> + bfd/elf-bfd.h | 9 +
> + bfd/elf.c | 114 +-
> + bfd/elfnn-loongarch.c | 4128 ++++++++++++++++++++++
> + bfd/elfxx-loongarch.c | 1618 +++++++++
> + bfd/elfxx-loongarch.h | 45 +
> + bfd/libbfd.h | 80 +
> + bfd/po/BLD-POTFILES.in | 2 +
> + bfd/po/SRC-POTFILES.in | 1 +
> + bfd/reloc.c | 171 +
> + bfd/targets.c | 8 +
> + config.guess | 3 +
> + config.sub | 1 +
> + gdb/Makefile.in | 13 +
> + gdb/arch/loongarch-linux-nat.c | 93 +
> + gdb/arch/loongarch-linux-nat.h | 35 +
> + gdb/arch/loongarch.c | 75 +
> + gdb/arch/loongarch.h | 35 +
> + gdb/configure.host | 3 +
> + gdb/configure.nat | 4 +
> + gdb/configure.tgt | 8 +
> + gdb/doc/gdb.texinfo | 10 +
> + gdb/features/Makefile | 10 +
> + gdb/features/loongarch/base32.c | 48 +
> + gdb/features/loongarch/base32.xml | 46 +
> + gdb/features/loongarch/base64.c | 48 +
> + gdb/features/loongarch/base64.xml | 46 +
> + gdb/features/loongarch/fpu32.c | 54 +
> + gdb/features/loongarch/fpu32.xml | 53 +
> + gdb/features/loongarch/fpu64.c | 62 +
> + gdb/features/loongarch/fpu64.xml | 58 +
> + gdb/features/loongarch/lasx.c | 80 +
> + gdb/features/loongarch/lasx.xml | 59 +
> + gdb/features/loongarch/lbt32.c | 19 +
> + gdb/features/loongarch/lbt32.xml | 17 +
> + gdb/features/loongarch/lbt64.c | 19 +
> + gdb/features/loongarch/lbt64.xml | 17 +
> + gdb/features/loongarch/lsx.c | 80 +
> + gdb/features/loongarch/lsx.xml | 59 +
> + gdb/loongarch-linux-nat.c | 878 +++++
> + gdb/loongarch-linux-tdep.c | 709 ++++
> + gdb/loongarch-linux-tdep.h | 48 +
> + gdb/loongarch-tdep.c | 1926 ++++++++++
> + gdb/loongarch-tdep.h | 61 +
> + gdb/nat/loongarch-linux-watch.c | 330 ++
> + gdb/nat/loongarch-linux-watch.h | 132 +
> + gdb/remote.c | 25 +
> + gdb/target.h | 3 +
> + gdb/testsuite/gdb.base/dump.exp | 4 +
> + gdb/testsuite/gdb.base/float.exp | 2 +
> + gdb/testsuite/gdb.trace/entry-values.exp | 2 +
> + gdb/testsuite/gdb.xml/tdesc-regs.exp | 5 +
> + gdbserver/Makefile.in | 2 +
> + gdbserver/configure.srv | 7 +
> + gdbserver/linux-loongarch-low.cc | 284 ++
> + include/dis-asm.h | 1 +
> + include/elf/common.h | 14 +
> + include/elf/loongarch.h | 267 ++
> + include/opcode/loongarch.h | 239 ++
> + opcodes/Makefile.am | 3 +
> + opcodes/Makefile.in | 6 +
> + opcodes/configure | 1 +
> + opcodes/configure.ac | 1 +
> + opcodes/disassemble.c | 9 +
> + opcodes/disassemble.h | 1 +
> + opcodes/loongarch-coder.c | 481 +++
> + opcodes/loongarch-dis.c | 342 ++
> + opcodes/loongarch-opc.c | 870 +++++
> + opcodes/po/POTFILES.in | 3 +
> + 76 files changed, 14091 insertions(+), 4 deletions(-)
> + create mode 100644 bfd/cpu-loongarch.c
> + create mode 100644 bfd/elfnn-loongarch.c
> + create mode 100644 bfd/elfxx-loongarch.c
> + create mode 100644 bfd/elfxx-loongarch.h
> + create mode 100644 gdb/arch/loongarch-linux-nat.c
> + create mode 100644 gdb/arch/loongarch-linux-nat.h
> + create mode 100644 gdb/arch/loongarch.c
> + create mode 100644 gdb/arch/loongarch.h
> + create mode 100644 gdb/features/loongarch/base32.c
> + create mode 100644 gdb/features/loongarch/base32.xml
> + create mode 100644 gdb/features/loongarch/base64.c
> + create mode 100644 gdb/features/loongarch/base64.xml
> + create mode 100644 gdb/features/loongarch/fpu32.c
> + create mode 100644 gdb/features/loongarch/fpu32.xml
> + create mode 100644 gdb/features/loongarch/fpu64.c
> + create mode 100644 gdb/features/loongarch/fpu64.xml
> + create mode 100644 gdb/features/loongarch/lasx.c
> + create mode 100644 gdb/features/loongarch/lasx.xml
> + create mode 100644 gdb/features/loongarch/lbt32.c
> + create mode 100644 gdb/features/loongarch/lbt32.xml
> + create mode 100644 gdb/features/loongarch/lbt64.c
> + create mode 100644 gdb/features/loongarch/lbt64.xml
> + create mode 100644 gdb/features/loongarch/lsx.c
> + create mode 100644 gdb/features/loongarch/lsx.xml
> + create mode 100644 gdb/loongarch-linux-nat.c
> + create mode 100644 gdb/loongarch-linux-tdep.c
> + create mode 100644 gdb/loongarch-linux-tdep.h
> + create mode 100644 gdb/loongarch-tdep.c
> + create mode 100644 gdb/loongarch-tdep.h
> + create mode 100644 gdb/nat/loongarch-linux-watch.c
> + create mode 100644 gdb/nat/loongarch-linux-watch.h
> + create mode 100644 gdbserver/linux-loongarch-low.cc
> + create mode 100644 include/elf/loongarch.h
> + create mode 100644 include/opcode/loongarch.h
> + create mode 100644 opcodes/loongarch-coder.c
> + create mode 100644 opcodes/loongarch-dis.c
> + create mode 100644 opcodes/loongarch-opc.c
> +
>
The gdb patches have their own format(style), for more details, please
refer to the current gdb-10.2.patch.
Could you please try to simplify the gdb patches? Or is it possible to drop
unnecessary changes when backporting the related patches from upstream?
Thanks.
Lianbo
> +diff --git gdb-10.2/bfd/Makefile.am gdb-10.2/bfd/Makefile.am
> +index d07c960..d494ebf 100644
> +--- gdb-10.2/bfd/Makefile.am
> ++++ gdb-10.2/bfd/Makefile.am
> +@@ -118,6 +118,7 @@ ALL_MACHINES = \
> + cpu-ip2k.lo \
> + cpu-iq2000.lo \
> + cpu-lm32.lo \
> ++ cpu-loongarch.lo \
> + cpu-m10200.lo \
> + cpu-m10300.lo \
> + cpu-m32c.lo \
> +@@ -202,6 +203,7 @@ ALL_MACHINES_CFILES = \
> + cpu-ip2k.c \
> + cpu-iq2000.c \
> + cpu-lm32.c \
> ++ cpu-loongarch.c \
> + cpu-m10200.c \
> + cpu-m10300.c \
> + cpu-m32c.c \
> +@@ -548,6 +550,9 @@ BFD64_BACKENDS = \
> + elf64-ia64.lo \
> + elf64-ia64-vms.lo \
> + elfxx-ia64.lo \
> ++ elf32-loongarch.lo \
> ++ elf64-loongarch.lo \
> ++ elfxx-loongarch.lo \
> + elfn32-mips.lo \
> + elf64-mips.lo \
> + elfxx-mips.lo \
> +@@ -601,6 +606,7 @@ BFD64_BACKENDS_CFILES = \
> + elfn32-mips.c \
> + elfxx-aarch64.c \
> + elfxx-ia64.c \
> ++ elfxx-loongarch.c \
> + elfxx-mips.c \
> + elfxx-riscv.c \
> + mach-o-aarch64.c \
> +@@ -665,6 +671,7 @@ SOURCE_CFILES = \
> + BUILD_CFILES = \
> + elf32-aarch64.c elf64-aarch64.c \
> + elf32-ia64.c elf64-ia64.c \
> ++ elf32-loongarch.c elf64-loongarch.c \
> + elf32-riscv.c elf64-riscv.c \
> + peigen.c pepigen.c pex64igen.c
> +
> +@@ -686,7 +693,7 @@ SOURCE_HFILES = \
> + elf-bfd.h elfcode.h elfcore.h elf-hppa.h elf-linker-x86.h \
> + elf-linux-core.h elf-nacl.h elf-s390.h elf-vxworks.h \
> + elfxx-aarch64.h elfxx-ia64.h elfxx-mips.h elfxx-riscv.h \
> +- elfxx-sparc.h elfxx-tilegx.h elfxx-x86.h \
> ++ elfxx-sparc.h elfxx-tilegx.h elfxx-x86.h elfxx-loongarch.h \
> + genlink.h go32stub.h \
> + libaout.h libbfd.h libcoff.h libecoff.h libhppa.h \
> + libpei.h libxcoff.h \
> +@@ -842,6 +849,14 @@ elf64-ia64.c : elfnn-ia64.c
> + echo "#line 1 \"elfnn-ia64.c\"" > $@
> + $(SED) -e s/NN/64/g < $< >> $@
> +
> ++elf32-loongarch.c : elfnn-loongarch.c
> ++ echo "#line 1 \"elfnn-loongarch.c\"" > $@
> ++ $(SED) -e s/NN/32/g < $< >> $@
> ++
> ++elf64-loongarch.c : elfnn-loongarch.c
> ++ echo "#line 1 \"elfnn-loongarch.c\"" > $@
> ++ $(SED) -e s/NN/64/g < $< >> $@
> ++
> + elf32-riscv.c : elfnn-riscv.c
> + echo "#line 1 \"elfnn-riscv.c\"" > $@
> + $(SED) -e s/NN/32/g < $< >> $@
> +diff --git gdb-10.2/bfd/Makefile.in gdb-10.2/bfd/Makefile.in
> +index 9cad4da..ecf2c20 100644
> +--- gdb-10.2/bfd/Makefile.in
> ++++ gdb-10.2/bfd/Makefile.in
> +@@ -543,6 +543,7 @@ ALL_MACHINES = \
> + cpu-ip2k.lo \
> + cpu-iq2000.lo \
> + cpu-lm32.lo \
> ++ cpu-loongarch.lo \
> + cpu-m10200.lo \
> + cpu-m10300.lo \
> + cpu-m32c.lo \
> +@@ -627,6 +628,7 @@ ALL_MACHINES_CFILES = \
> + cpu-ip2k.c \
> + cpu-iq2000.c \
> + cpu-lm32.c \
> ++ cpu-loongarch.c \
> + cpu-m10200.c \
> + cpu-m10300.c \
> + cpu-m32c.c \
> +@@ -975,6 +977,9 @@ BFD64_BACKENDS = \
> + elf64-ia64.lo \
> + elf64-ia64-vms.lo \
> + elfxx-ia64.lo \
> ++ elf32-loongarch.lo \
> ++ elf64-loongarch.lo \
> ++ elfxx-loongarch.lo \
> + elfn32-mips.lo \
> + elf64-mips.lo \
> + elfxx-mips.lo \
> +@@ -1028,6 +1033,7 @@ BFD64_BACKENDS_CFILES = \
> + elfn32-mips.c \
> + elfxx-aarch64.c \
> + elfxx-ia64.c \
> ++ elfxx-loongarch.c \
> + elfxx-mips.c \
> + elfxx-riscv.c \
> + mach-o-aarch64.c \
> +@@ -1091,6 +1097,7 @@ SOURCE_CFILES = \
> + BUILD_CFILES = \
> + elf32-aarch64.c elf64-aarch64.c \
> + elf32-ia64.c elf64-ia64.c \
> ++ elf32-loongarch.c elf64-loongarch.c \
> + elf32-riscv.c elf64-riscv.c \
> + peigen.c pepigen.c pex64igen.c
> +
> +@@ -1109,7 +1116,7 @@ SOURCE_HFILES = \
> + elf-bfd.h elfcode.h elfcore.h elf-hppa.h elf-linker-x86.h \
> + elf-linux-core.h elf-nacl.h elf-s390.h elf-vxworks.h \
> + elfxx-aarch64.h elfxx-ia64.h elfxx-mips.h elfxx-riscv.h \
> +- elfxx-sparc.h elfxx-tilegx.h elfxx-x86.h \
> ++ elfxx-sparc.h elfxx-tilegx.h elfxx-x86.h elfxx-loongarch.h \
> + genlink.h go32stub.h \
> + libaout.h libbfd.h libcoff.h libecoff.h libhppa.h \
> + libpei.h libxcoff.h \
> +@@ -1349,6 +1356,7 @@ distclean-compile:
> + @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/cpu-k1om.Plo(a)am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/cpu-l1om.Plo(a)am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/cpu-lm32.Plo(a)am__quote@
> ++@AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/cpu-loongarch.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/cpu-m10200.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/cpu-m10300.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/cpu-m32c.Plo(a)am__quote@
> +@@ -1442,6 +1450,7 @@ distclean-compile:
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf32-ip2k.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf32-iq2000.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf32-lm32.Plo@am__quote@
> ++@AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf32-loongarch.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf32-m32c.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf32-m32r.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf32-m68hc11.Plo@am__quote@
> +@@ -1492,6 +1501,7 @@ distclean-compile:
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf64-hppa.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf64-ia64-vms.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf64-ia64.Plo@am__quote@
> ++@AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf64-loongarch.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf64-mips.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elf64-mmix.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/elf64-nfp.Plo(a)am__quote
> @
> +@@ -1506,6 +1516,7 @@ distclean-compile:
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elfn32-mips.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elfxx-aarch64.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elfxx-ia64.Plo@am__quote@
> ++@AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elfxx-loongarch.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elfxx-mips.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elfxx-riscv.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/elfxx-sparc.Plo@am__quote@
> +@@ -1972,6 +1983,14 @@ elf64-ia64.c : elfnn-ia64.c
> + echo "#line 1 \"elfnn-ia64.c\"" > $@
> + $(SED) -e s/NN/64/g < $< >> $@
> +
> ++elf32-loongarch.c : elfnn-loongarch.c
> ++ echo "#line 1 \"elfnn-loongarch.c\"" > $@
> ++ $(SED) -e s/NN/32/g < $< >> $@
> ++
> ++elf64-loongarch.c : elfnn-loongarch.c
> ++ echo "#line 1 \"elfnn-loongarch.c\"" > $@
> ++ $(SED) -e s/NN/64/g < $< >> $@
> ++
> + elf32-riscv.c : elfnn-riscv.c
> + echo "#line 1 \"elfnn-riscv.c\"" > $@
> + $(SED) -e s/NN/32/g < $< >> $@
> +diff --git gdb-10.2/bfd/archures.c gdb-10.2/bfd/archures.c
> +index 5069864..36ccb1c 100644
> +--- gdb-10.2/bfd/archures.c
> ++++ gdb-10.2/bfd/archures.c
> +@@ -555,6 +555,9 @@ DESCRIPTION
> + .#define bfd_mach_ck807 6
> + .#define bfd_mach_ck810 7
> + .#define bfd_mach_ck860 8
> ++. bfd_arch_loongarch, {* LoongArch *}
> ++.#define bfd_mach_loongarch32 1
> ++.#define bfd_mach_loongarch64 2
> + . bfd_arch_last
> + . };
> + */
> +@@ -636,6 +639,7 @@ extern const bfd_arch_info_type bfd_iq2000_arch;
> + extern const bfd_arch_info_type bfd_k1om_arch;
> + extern const bfd_arch_info_type bfd_l1om_arch;
> + extern const bfd_arch_info_type bfd_lm32_arch;
> ++extern const bfd_arch_info_type bfd_loongarch_arch;
> + extern const bfd_arch_info_type bfd_m32c_arch;
> + extern const bfd_arch_info_type bfd_m32r_arch;
> + extern const bfd_arch_info_type bfd_m68hc11_arch;
> +@@ -725,6 +729,7 @@ static const bfd_arch_info_type * const
> bfd_archures_list[] =
> + &bfd_k1om_arch,
> + &bfd_l1om_arch,
> + &bfd_lm32_arch,
> ++ &bfd_loongarch_arch,
> + &bfd_m32c_arch,
> + &bfd_m32r_arch,
> + &bfd_m68hc11_arch,
> +diff --git gdb-10.2/bfd/bfd-in2.h gdb-10.2/bfd/bfd-in2.h
> +index 935ba53..e8a38b8 100644
> +--- gdb-10.2/bfd/bfd-in2.h
> ++++ gdb-10.2/bfd/bfd-in2.h
> +@@ -1955,6 +1955,9 @@ enum bfd_architecture
> + #define bfd_mach_ck807 6
> + #define bfd_mach_ck810 7
> + #define bfd_mach_ck860 8
> ++ bfd_arch_loongarch, /* LoongArch */
> ++#define bfd_mach_loongarch32 1
> ++#define bfd_mach_loongarch64 2
> + bfd_arch_last
> + };
> +
> +@@ -6273,6 +6276,88 @@ assembler and not (currently) written to any
> object files. */
> +
> + /* S12Z relocations. */
> + BFD_RELOC_S12Z_OPR,
> ++
> ++/* LARCH relocations. */
> ++ BFD_RELOC_LARCH_TLS_DTPMOD32,
> ++ BFD_RELOC_LARCH_TLS_DTPREL32,
> ++ BFD_RELOC_LARCH_TLS_DTPMOD64,
> ++ BFD_RELOC_LARCH_TLS_DTPREL64,
> ++ BFD_RELOC_LARCH_TLS_TPREL32,
> ++ BFD_RELOC_LARCH_TLS_TPREL64,
> ++ BFD_RELOC_LARCH_MARK_LA,
> ++ BFD_RELOC_LARCH_MARK_PCREL,
> ++ BFD_RELOC_LARCH_SOP_PUSH_PCREL,
> ++ BFD_RELOC_LARCH_SOP_PUSH_ABSOLUTE,
> ++ BFD_RELOC_LARCH_SOP_PUSH_DUP,
> ++ BFD_RELOC_LARCH_SOP_PUSH_GPREL,
> ++ BFD_RELOC_LARCH_SOP_PUSH_TLS_TPREL,
> ++ BFD_RELOC_LARCH_SOP_PUSH_TLS_GOT,
> ++ BFD_RELOC_LARCH_SOP_PUSH_TLS_GD,
> ++ BFD_RELOC_LARCH_SOP_PUSH_PLT_PCREL,
> ++ BFD_RELOC_LARCH_SOP_ASSERT,
> ++ BFD_RELOC_LARCH_SOP_NOT,
> ++ BFD_RELOC_LARCH_SOP_SUB,
> ++ BFD_RELOC_LARCH_SOP_SL,
> ++ BFD_RELOC_LARCH_SOP_SR,
> ++ BFD_RELOC_LARCH_SOP_ADD,
> ++ BFD_RELOC_LARCH_SOP_AND,
> ++ BFD_RELOC_LARCH_SOP_IF_ELSE,
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_5,
> ++ BFD_RELOC_LARCH_SOP_POP_32_U_10_12,
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_12,
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_16,
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_16_S2,
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_5_20,
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_0_5_10_16_S2,
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_0_10_10_16_S2,
> ++ BFD_RELOC_LARCH_SOP_POP_32_U,
> ++ BFD_RELOC_LARCH_ADD8,
> ++ BFD_RELOC_LARCH_ADD16,
> ++ BFD_RELOC_LARCH_ADD24,
> ++ BFD_RELOC_LARCH_ADD32,
> ++ BFD_RELOC_LARCH_ADD64,
> ++ BFD_RELOC_LARCH_SUB8,
> ++ BFD_RELOC_LARCH_SUB16,
> ++ BFD_RELOC_LARCH_SUB24,
> ++ BFD_RELOC_LARCH_SUB32,
> ++ BFD_RELOC_LARCH_SUB64,
> ++ BFD_RELOC_LARCH_B16,
> ++ BFD_RELOC_LARCH_B21,
> ++ BFD_RELOC_LARCH_B26,
> ++ BFD_RELOC_LARCH_ABS_HI20,
> ++ BFD_RELOC_LARCH_ABS_LO12,
> ++ BFD_RELOC_LARCH_ABS64_LO20,
> ++ BFD_RELOC_LARCH_ABS64_HI12,
> ++ BFD_RELOC_LARCH_PCALA_HI20,
> ++ BFD_RELOC_LARCH_PCALA_LO12,
> ++ BFD_RELOC_LARCH_PCALA64_LO20,
> ++ BFD_RELOC_LARCH_PCALA64_HI12,
> ++ BFD_RELOC_LARCH_GOT_PC_HI20,
> ++ BFD_RELOC_LARCH_GOT_PC_LO12,
> ++ BFD_RELOC_LARCH_GOT64_PC_LO20,
> ++ BFD_RELOC_LARCH_GOT64_PC_HI12,
> ++ BFD_RELOC_LARCH_GOT_HI20,
> ++ BFD_RELOC_LARCH_GOT_LO12,
> ++ BFD_RELOC_LARCH_GOT64_LO20,
> ++ BFD_RELOC_LARCH_GOT64_HI12,
> ++ BFD_RELOC_LARCH_TLS_LE_HI20,
> ++ BFD_RELOC_LARCH_TLS_LE_LO12,
> ++ BFD_RELOC_LARCH_TLS_LE64_LO20,
> ++ BFD_RELOC_LARCH_TLS_LE64_HI12,
> ++ BFD_RELOC_LARCH_TLS_IE_PC_HI20,
> ++ BFD_RELOC_LARCH_TLS_IE_PC_LO12,
> ++ BFD_RELOC_LARCH_TLS_IE64_PC_LO20,
> ++ BFD_RELOC_LARCH_TLS_IE64_PC_HI12,
> ++ BFD_RELOC_LARCH_TLS_IE_HI20,
> ++ BFD_RELOC_LARCH_TLS_IE_LO12,
> ++ BFD_RELOC_LARCH_TLS_IE64_LO20,
> ++ BFD_RELOC_LARCH_TLS_IE64_HI12,
> ++ BFD_RELOC_LARCH_TLS_LD_PC_HI20,
> ++ BFD_RELOC_LARCH_TLS_LD_HI20,
> ++ BFD_RELOC_LARCH_TLS_GD_PC_HI20,
> ++ BFD_RELOC_LARCH_TLS_GD_HI20,
> ++ BFD_RELOC_LARCH_32_PCREL,
> ++ BFD_RELOC_LARCH_RELAX,
> + BFD_RELOC_UNUSED };
> +
> + typedef enum bfd_reloc_code_real bfd_reloc_code_real_type;
> +diff --git gdb-10.2/bfd/config.bfd gdb-10.2/bfd/config.bfd
> +index 6c2919e..417bdb0 100644
> +--- gdb-10.2/bfd/config.bfd
> ++++ gdb-10.2/bfd/config.bfd
> +@@ -183,6 +183,7 @@ hppa*) targ_archs=bfd_hppa_arch ;;
> + i[3-7]86) targ_archs=bfd_i386_arch ;;
> + ia16) targ_archs=bfd_i386_arch ;;
> + lm32) targ_archs=bfd_lm32_arch ;;
> ++loongarch*) targ_archs=bfd_loongarch_arch ;;
> + m6811*|m68hc11*) targ_archs="bfd_m68hc11_arch bfd_m68hc12_arch
> bfd_m9s12x_arch bfd_m9s12xg_arch" ;;
> + m6812*|m68hc12*) targ_archs="bfd_m68hc12_arch bfd_m68hc11_arch
> bfd_m9s12x_arch bfd_m9s12xg_arch" ;;
> + m68*) targ_archs=bfd_m68k_arch ;;
> +@@ -1401,6 +1402,20 @@ case "${targ}" in
> + targ_underscore=yes
> + ;;
> +
> ++#ifdef BFD64
> ++ loongarch32-*)
> ++ targ_defvec=loongarch_elf32_vec
> ++ targ_selvecs="loongarch_elf32_vec"
> ++ want64=true
> ++ ;;
> ++
> ++ loongarch64-*)
> ++ targ_defvec=loongarch_elf64_vec
> ++ targ_selvecs="loongarch_elf32_vec loongarch_elf64_vec"
> ++ want64=true
> ++ ;;
> ++#endif
> ++
> + # END OF targmatch.h
> + bpf-*-*)
> + echo "*** Configuration $targ is not fully supported." >&2
> +diff --git gdb-10.2/bfd/configure gdb-10.2/bfd/configure
> +index a9c4fd9..251439e 100755
> +--- gdb-10.2/bfd/configure
> ++++ gdb-10.2/bfd/configure
> +@@ -14836,6 +14836,8 @@ do
> + l1om_elf64_fbsd_vec) tb="$tb elf64-x86-64.lo elfxx-x86.lo
> elf-ifunc.lo elf64.lo $elf"; target_size=64 ;;
> + lm32_elf32_vec) tb="$tb elf32-lm32.lo elf32.lo $elf" ;;
> + lm32_elf32_fdpic_vec) tb="$tb elf32-lm32.lo elf32.lo $elf" ;;
> ++ loongarch_elf32_vec) tb="$tb elf32-loongarch.lo
> elfxx-loongarch.lo elf32.lo elf-ifunc.lo $elf" ;;
> ++ loongarch_elf64_vec) tb="$tb elf64-loongarch.lo elf64.lo
> elfxx-loongarch.lo elf32.lo elf-ifunc.lo $elf"; target_size=64 ;;
> + m32c_elf32_vec) tb="$tb elf32-m32c.lo elf32.lo $elf" ;;
> + m32r_elf32_vec) tb="$tb elf32-m32r.lo elf32.lo $elf" ;;
> + m32r_elf32_le_vec) tb="$tb elf32-m32r.lo elf32.lo
> $elf" ;;
> +diff --git gdb-10.2/bfd/configure.ac gdb-10.2/bfd/configure.ac
> +index f62659a..e97c9fc 100644
> +--- gdb-10.2/bfd/configure.ac
> ++++ gdb-10.2/bfd/configure.ac
> +@@ -542,6 +542,8 @@ do
> + l1om_elf64_fbsd_vec) tb="$tb elf64-x86-64.lo elfxx-x86.lo
> elf-ifunc.lo elf64.lo $elf"; target_size=64 ;;
> + lm32_elf32_vec) tb="$tb elf32-lm32.lo elf32.lo $elf" ;;
> + lm32_elf32_fdpic_vec) tb="$tb elf32-lm32.lo elf32.lo $elf" ;;
> ++ loongarch_elf32_vec) tb="$tb elf32-loongarch.lo
> elfxx-loongarch.lo elf32.lo elf-ifunc.lo $elf" ;;
> ++ loongarch_elf64_vec) tb="$tb elf64-loongarch.lo elf64.lo
> elfxx-loongarch.lo elf32.lo elf-ifunc.lo $elf"; target_size=64 ;;
> + m32c_elf32_vec) tb="$tb elf32-m32c.lo elf32.lo $elf" ;;
> + m32r_elf32_vec) tb="$tb elf32-m32r.lo elf32.lo $elf" ;;
> + m32r_elf32_le_vec) tb="$tb elf32-m32r.lo elf32.lo
> $elf" ;;
> +diff --git gdb-10.2/bfd/cpu-loongarch.c gdb-10.2/bfd/cpu-loongarch.c
> +new file mode 100644
> +index 0000000..01d6c70
> +--- /dev/null
> ++++ gdb-10.2/bfd/cpu-loongarch.c
> +@@ -0,0 +1,61 @@
> ++/* BFD support for LoongArch.
> ++ Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of BFD, the Binary File Descriptor library.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program; see the file COPYING3. If not,
> ++ see <http://www.gnu.org/licenses/>. */
> ++
> ++#include "sysdep.h"
> ++#include "bfd.h"
> ++#include "libbfd.h"
> ++
> ++static const bfd_arch_info_type bfd_loongarch32_arch =
> ++{
> ++ 32, /* 32 bits in a word. */
> ++ 32, /* 64 bits in an address. */
> ++ 8, /* 8 bits in a byte. */
> ++ bfd_arch_loongarch, /* Architecture. */
> ++ bfd_mach_loongarch32, /* Machine number - 0 for now. */
> ++ "loongarch32", /* Architecture name. */
> ++ "Loongarch32", /* Printable name. */
> ++ 3, /* Section align power. */
> ++ TRUE, /* This is the default architecture. */
> ++ bfd_default_compatible, /* Architecture comparison function. */
> ++ bfd_default_scan, /* String to architecture conversion. */
> ++ bfd_arch_default_fill, /* Default fill. */
> ++ NULL, /* Next in list. */
> ++ 0,
> ++};
> ++
> ++const bfd_arch_info_type bfd_loongarch_arch =
> ++{
> ++ 32, /* 32 bits in a word. */
> ++ 64, /* 64 bits in an address. */
> ++ 8, /* 8 bits in a byte. */
> ++ bfd_arch_loongarch, /* Architecture. */
> ++ /* Machine number of LoongArch64 is larger
> ++ * so that LoongArch64 is compatible to LoongArch32. */
> ++ bfd_mach_loongarch64,
> ++ "loongarch64", /* Architecture name. */
> ++ "Loongarch64", /* Printable name. */
> ++ 3, /* Section align power. */
> ++ TRUE, /* This is the default architecture. */
> ++ bfd_default_compatible, /* Architecture comparison function. */
> ++ bfd_default_scan, /* String to architecture conversion. */
> ++ bfd_arch_default_fill, /* Default fill. */
> ++ &bfd_loongarch32_arch, /* Next in list. */
> ++ 0,
> ++};
> +diff --git gdb-10.2/bfd/elf-bfd.h gdb-10.2/bfd/elf-bfd.h
> +index eebdf9a..148c08c 100644
> +--- gdb-10.2/bfd/elf-bfd.h
> ++++ gdb-10.2/bfd/elf-bfd.h
> +@@ -500,6 +500,7 @@ enum elf_target_id
> + I386_ELF_DATA,
> + IA64_ELF_DATA,
> + LM32_ELF_DATA,
> ++ LARCH_ELF_DATA,
> + M32R_ELF_DATA,
> + M68HC11_ELF_DATA,
> + M68K_ELF_DATA,
> +@@ -2799,6 +2800,14 @@ extern char *elfcore_write_lwpstatus
> + (bfd *, char *, int *, long, int, const void *);
> + extern char *elfcore_write_register_note
> + (bfd *, char *, int *, const char *, const void *, int);
> ++extern char *elfcore_write_loongarch_cpucfg
> ++ (bfd *, char *, int *, const void*, int);
> ++extern char *elfcore_write_loongarch_lbt
> ++ (bfd *, char *, int *, const void*, int);
> ++extern char *elfcore_write_loongarch_lsx
> ++ (bfd *, char *, int *, const void*, int);
> ++extern char *elfcore_write_loongarch_lasx
> ++ (bfd *, char *, int *, const void*, int);
> +
> + /* Internal structure which holds information to be included in the
> + PRPSINFO section of Linux core files.
> +diff --git gdb-10.2/bfd/elf.c gdb-10.2/bfd/elf.c
> +index 5a02f8d..580d7c5 100644
> +--- gdb-10.2/bfd/elf.c
> ++++ gdb-10.2/bfd/elf.c
> +@@ -4391,7 +4391,7 @@ get_program_header_size (bfd *abfd, struct
> bfd_link_info *info)
> + {
> + /* If we have a loadable interpreter section, we need a
> + PT_INTERP segment. In this case, assume we also need a
> +- PT_PHDR segment, although that may not be true for all
> ++ PT_PHDR segment, although that may not be TRUE for all
> + targets. */
> + segs += 2;
> + }
> +@@ -9903,6 +9903,30 @@ elfcore_grok_arc_v2 (bfd *abfd, Elf_Internal_Note
> *note)
> + return elfcore_make_note_pseudosection (abfd, ".reg-arc-v2", note);
> + }
> +
> ++static bfd_boolean
> ++elfcore_grok_loongarch_cpucfg (bfd *abfd, Elf_Internal_Note *note)
> ++{
> ++ return elfcore_make_note_pseudosection (abfd, ".reg-loongarch-cpucfg",
> note);
> ++}
> ++
> ++static bfd_boolean
> ++elfcore_grok_loongarch_lbt (bfd *abfd, Elf_Internal_Note *note)
> ++{
> ++ return elfcore_make_note_pseudosection (abfd, ".reg-loongarch-lbt",
> note);
> ++}
> ++
> ++static bfd_boolean
> ++elfcore_grok_loongarch_lsx (bfd *abfd, Elf_Internal_Note *note)
> ++{
> ++ return elfcore_make_note_pseudosection (abfd, ".reg-loongarch-lsx",
> note);
> ++}
> ++
> ++static bfd_boolean
> ++elfcore_grok_loongarch_lasx (bfd *abfd, Elf_Internal_Note *note)
> ++{
> ++ return elfcore_make_note_pseudosection (abfd, ".reg-loongarch-lasx",
> note);
> ++}
> ++
> + #if defined (HAVE_PRPSINFO_T)
> + typedef prpsinfo_t elfcore_psinfo_t;
> + #if defined (HAVE_PRPSINFO32_T) /* Sparc64 cross Sparc32 */
> +@@ -10560,6 +10584,34 @@ elfcore_grok_note (bfd *abfd, Elf_Internal_Note
> *note)
> + else
> + return TRUE;
> +
> ++ case NT_LARCH_CPUCFG:
> ++ if (note->namesz == 6
> ++ && strcmp (note->namedata, "LINUX") == 0)
> ++ return elfcore_grok_loongarch_cpucfg (abfd, note);
> ++ else
> ++ return TRUE;
> ++
> ++ case NT_LARCH_LBT:
> ++ if (note->namesz == 6
> ++ && strcmp (note->namedata, "LINUX") == 0)
> ++ return elfcore_grok_loongarch_lbt (abfd, note);
> ++ else
> ++ return TRUE;
> ++
> ++ case NT_LARCH_LSX:
> ++ if (note->namesz == 6
> ++ && strcmp (note->namedata, "LINUX") == 0)
> ++ return elfcore_grok_loongarch_lsx (abfd, note);
> ++ else
> ++ return TRUE;
> ++
> ++ case NT_LARCH_LASX:
> ++ if (note->namesz == 6
> ++ && strcmp (note->namedata, "LINUX") == 0)
> ++ return elfcore_grok_loongarch_lasx (abfd, note);
> ++ else
> ++ return TRUE;
> ++
> + case NT_PRPSINFO:
> + case NT_PSINFO:
> + if (bed->elf_backend_grok_psinfo)
> +@@ -11941,6 +11993,55 @@ elfcore_write_arc_v2 (bfd *abfd,
> + note_name, NT_ARC_V2, arc_v2, size);
> + }
> +
> ++char *
> ++elfcore_write_loongarch_cpucfg (bfd *abfd,
> ++ char *buf,
> ++ int *bufsiz,
> ++ const void *loongarch_cpucfg,
> ++ int size)
> ++{
> ++ char *note_name = "LINUX";
> ++ return elfcore_write_note (abfd, buf, bufsiz,
> ++ note_name, NT_LARCH_CPUCFG,
> ++ loongarch_cpucfg, size);
> ++}
> ++
> ++char *
> ++elfcore_write_loongarch_lbt (bfd *abfd,
> ++ char *buf,
> ++ int *bufsiz,
> ++ const void *loongarch_lbt,
> ++ int size)
> ++{
> ++ char *note_name = "LINUX";
> ++ return elfcore_write_note (abfd, buf, bufsiz,
> ++ note_name, NT_LARCH_LBT, loongarch_lbt, size);
> ++}
> ++
> ++char *
> ++elfcore_write_loongarch_lsx (bfd *abfd,
> ++ char *buf,
> ++ int *bufsiz,
> ++ const void *loongarch_lsx,
> ++ int size)
> ++{
> ++ char *note_name = "LINUX";
> ++ return elfcore_write_note (abfd, buf, bufsiz,
> ++ note_name, NT_LARCH_LSX, loongarch_lsx, size);
> ++}
> ++
> ++char *
> ++elfcore_write_loongarch_lasx (bfd *abfd,
> ++ char *buf,
> ++ int *bufsiz,
> ++ const void *loongarch_lasx,
> ++ int size)
> ++{
> ++ char *note_name = "LINUX";
> ++ return elfcore_write_note (abfd, buf, bufsiz,
> ++ note_name, NT_LARCH_LASX, loongarch_lasx,
> size);
> ++}
> ++
> + char *
> + elfcore_write_register_note (bfd *abfd,
> + char *buf,
> +@@ -12025,6 +12126,15 @@ elfcore_write_register_note (bfd *abfd,
> + return elfcore_write_aarch_pauth (abfd, buf, bufsiz, data, size);
> + if (strcmp (section, ".reg-arc-v2") == 0)
> + return elfcore_write_arc_v2 (abfd, buf, bufsiz, data, size);
> ++ if (strcmp (section, ".reg-loongarch-cpucfg") == 0)
> ++ return elfcore_write_loongarch_cpucfg (abfd, buf, bufsiz, data,
> size);
> ++ if (strcmp (section, ".reg-loongarch-lbt") == 0)
> ++ return elfcore_write_loongarch_lbt (abfd, buf, bufsiz, data, size);
> ++ if (strcmp (section, ".reg-loongarch-lsx") == 0)
> ++ return elfcore_write_loongarch_lsx (abfd, buf, bufsiz, data, size);
> ++ if (strcmp (section, ".reg-loongarch-lasx") == 0)
> ++ return elfcore_write_loongarch_lasx (abfd, buf, bufsiz, data, size);
> ++
> + return NULL;
> + }
> +
> +@@ -12491,7 +12601,7 @@ _bfd_elf_final_write_processing (bfd *abfd)
> +
> + /* Return TRUE for ELF symbol types that represent functions.
> + This is the default version of this function, which is sufficient for
> +- most targets. It returns true if TYPE is STT_FUNC or STT_GNU_IFUNC.
> */
> ++ most targets. It returns TRUE if TYPE is STT_FUNC or STT_GNU_IFUNC.
> */
> +
> + bfd_boolean
> + _bfd_elf_is_function_type (unsigned int type)
> +diff --git gdb-10.2/bfd/elfnn-loongarch.c gdb-10.2/bfd/elfnn-loongarch.c
> +new file mode 100644
> +index 0000000..226cd59
> +--- /dev/null
> ++++ gdb-10.2/bfd/elfnn-loongarch.c
> +@@ -0,0 +1,4128 @@
> ++/* LoongArch-specific support for NN-bit ELF.
> ++ Copyright (C) 2021-2022 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of BFD, the Binary File Descriptor library.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program; see the file COPYING3. If not,
> ++ see <http://www.gnu.org/licenses/>. */
> ++
> ++#include "ansidecl.h"
> ++#include "sysdep.h"
> ++#include "bfd.h"
> ++#include "libbfd.h"
> ++#define ARCH_SIZE NN
> ++#include "elf-bfd.h"
> ++#include "objalloc.h"
> ++#include "elf/loongarch.h"
> ++#include "elfxx-loongarch.h"
> ++
> ++static bfd_boolean
> ++loongarch_info_to_howto_rela (bfd *abfd, arelent *cache_ptr,
> ++ Elf_Internal_Rela *dst)
> ++{
> ++ cache_ptr->howto = loongarch_elf_rtype_to_howto (abfd,
> ++ ELFNN_R_TYPE
> (dst->r_info));
> ++ return cache_ptr->howto != NULL;
> ++}
> ++
> ++/* LoongArch ELF linker hash entry. */
> ++struct loongarch_elf_link_hash_entry
> ++{
> ++ struct elf_link_hash_entry elf;
> ++
> ++#define GOT_UNKNOWN 0
> ++#define GOT_NORMAL 1
> ++#define GOT_TLS_GD 2
> ++#define GOT_TLS_IE 4
> ++#define GOT_TLS_LE 8
> ++ char tls_type;
> ++};
> ++
> ++#define loongarch_elf_hash_entry(ent) \
> ++ ((struct loongarch_elf_link_hash_entry *) (ent))
> ++
> ++struct _bfd_loongarch_elf_obj_tdata
> ++{
> ++ struct elf_obj_tdata root;
> ++
> ++ /* The tls_type for each local got entry. */
> ++ char *local_got_tls_type;
> ++};
> ++
> ++#define _bfd_loongarch_elf_tdata(abfd) \
> ++ ((struct _bfd_loongarch_elf_obj_tdata *) (abfd)->tdata.any)
> ++
> ++#define _bfd_loongarch_elf_local_got_tls_type(abfd) \
> ++ (_bfd_loongarch_elf_tdata (abfd)->local_got_tls_type)
> ++
> ++#define _bfd_loongarch_elf_tls_type(abfd, h, symndx) \
> ++ (*((h) != NULL \
> ++ ? &loongarch_elf_hash_entry (h)->tls_type
> \
> ++ : &_bfd_loongarch_elf_local_got_tls_type (abfd)[symndx]))
> ++
> ++#define is_loongarch_elf(bfd) \
> ++ (bfd_get_flavour (bfd) == bfd_target_elf_flavour \
> ++ && elf_tdata (bfd) != NULL \
> ++ && elf_object_id (bfd) == LARCH_ELF_DATA)
> ++
> ++struct loongarch_elf_link_hash_table
> ++{
> ++ struct elf_link_hash_table elf;
> ++
> ++ /* Short-cuts to get to dynamic linker sections. */
> ++ asection *sdyntdata;
> ++
> ++ /* Small local sym to section mapping cache. */
> ++ struct sym_cache sym_cache;
> ++
> ++ /* Used by local STT_GNU_IFUNC symbols. */
> ++ htab_t loc_hash_table;
> ++ void *loc_hash_memory;
> ++
> ++ /* The max alignment of output sections. */
> ++ bfd_vma max_alignment;
> ++};
> ++
> ++/* Get the LoongArch ELF linker hash table from a link_info structure.
> */
> ++#define loongarch_elf_hash_table(p) \
> ++ (elf_hash_table_id (elf_hash_table (p)) == LARCH_ELF_DATA \
> ++ ? ((struct loongarch_elf_link_hash_table *) ((p)->hash)) \
> ++ : NULL)
> ++
> ++#define MINUS_ONE ((bfd_vma) 0 - 1)
> ++
> ++#define sec_addr(sec) ((sec)->output_section->vma + (sec)->output_offset)
> ++
> ++#define LARCH_ELF_LOG_WORD_BYTES (ARCH_SIZE == 32 ? 2 : 3)
> ++#define LARCH_ELF_WORD_BYTES (1 << LARCH_ELF_LOG_WORD_BYTES)
> ++
> ++#define PLT_HEADER_INSNS 8
> ++#define PLT_HEADER_SIZE (PLT_HEADER_INSNS * 4)
> ++
> ++#define PLT_ENTRY_INSNS 4
> ++#define PLT_ENTRY_SIZE (PLT_ENTRY_INSNS * 4)
> ++
> ++#define GOT_ENTRY_SIZE (LARCH_ELF_WORD_BYTES)
> ++
> ++#define GOTPLT_HEADER_SIZE (GOT_ENTRY_SIZE * 2)
> ++
> ++#define elf_backend_want_got_plt 1
> ++
> ++#define elf_backend_plt_readonly 1
> ++
> ++#define elf_backend_want_plt_sym 1
> ++#define elf_backend_plt_alignment 4
> ++#define elf_backend_can_gc_sections 1
> ++#define elf_backend_can_refcount 1
> ++#define elf_backend_want_got_sym 1
> ++
> ++#define elf_backend_got_header_size (GOT_ENTRY_SIZE * 1)
> ++
> ++#define elf_backend_want_dynrelro 1
> ++#define elf_backend_rela_normal 1
> ++#define elf_backend_default_execstack 0
> ++
> ++/* Generate a PLT header. */
> ++
> ++static bfd_boolean
> ++loongarch_make_plt_header (bfd_vma got_plt_addr, bfd_vma plt_header_addr,
> ++ uint32_t *entry)
> ++{
> ++ bfd_vma pcrel = got_plt_addr - plt_header_addr;
> ++ bfd_vma hi, lo;
> ++
> ++ if (pcrel + 0x80000800 > 0xffffffff)
> ++ {
> ++ _bfd_error_handler (_("%#" PRIx64 " invaild imm"), (uint64_t)
> pcrel);
> ++ bfd_set_error (bfd_error_bad_value);
> ++ return FALSE;
> ++ }
> ++ hi = ((pcrel + 0x800) >> 12) & 0xfffff;
> ++ lo = pcrel & 0xfff;
> ++
> ++ /* pcaddu12i $t2, %hi(%pcrel(.got.plt))
> ++ sub.[wd] $t1, $t1, $t3
> ++ ld.[wd] $t3, $t2, %lo(%pcrel(.got.plt)) # _dl_runtime_resolve
> ++ addi.[wd] $t1, $t1, -(PLT_HEADER_SIZE + 12)
> ++ addi.[wd] $t0, $t2, %lo(%pcrel(.got.plt))
> ++ srli.[wd] $t1, $t1, log2(16 / GOT_ENTRY_SIZE)
> ++ ld.[wd] $t0, $t0, GOT_ENTRY_SIZE
> ++ jirl $r0, $t3, 0 */
> ++
> ++ if (GOT_ENTRY_SIZE == 8)
> ++ {
> ++ entry[0] = 0x1c00000e | (hi & 0xfffff) << 5;
> ++ entry[1] = 0x0011bdad;
> ++ entry[2] = 0x28c001cf | (lo & 0xfff) << 10;
> ++ entry[3] = 0x02c001ad | ((-(PLT_HEADER_SIZE + 12)) & 0xfff) << 10;
> ++ entry[4] = 0x02c001cc | (lo & 0xfff) << 10;
> ++ entry[5] = 0x004501ad | (4 - LARCH_ELF_LOG_WORD_BYTES) << 10;
> ++ entry[6] = 0x28c0018c | GOT_ENTRY_SIZE << 10;
> ++ entry[7] = 0x4c0001e0;
> ++ }
> ++ else
> ++ {
> ++ entry[0] = 0x1c00000e | (hi & 0xfffff) << 5;
> ++ entry[1] = 0x00113dad;
> ++ entry[2] = 0x288001cf | (lo & 0xfff) << 10;
> ++ entry[3] = 0x028001ad | ((-(PLT_HEADER_SIZE + 12)) & 0xfff) << 10;
> ++ entry[4] = 0x028001cc | (lo & 0xfff) << 10;
> ++ entry[5] = 0x004481ad | (4 - LARCH_ELF_LOG_WORD_BYTES) << 10;
> ++ entry[6] = 0x2880018c | GOT_ENTRY_SIZE << 10;
> ++ entry[7] = 0x4c0001e0;
> ++ }
> ++ return TRUE;
> ++}
> ++
> ++/* Generate a PLT entry. */
> ++
> ++static bfd_boolean
> ++loongarch_make_plt_entry (bfd_vma got_plt_entry_addr, bfd_vma
> plt_entry_addr,
> ++ uint32_t *entry)
> ++{
> ++ bfd_vma pcrel = got_plt_entry_addr - plt_entry_addr;
> ++ bfd_vma hi, lo;
> ++
> ++ if (pcrel + 0x80000800 > 0xffffffff)
> ++ {
> ++ _bfd_error_handler (_("%#" PRIx64 " invaild imm"), (uint64_t)
> pcrel);
> ++ bfd_set_error (bfd_error_bad_value);
> ++ return FALSE;
> ++ }
> ++ hi = ((pcrel + 0x800) >> 12) & 0xfffff;
> ++ lo = pcrel & 0xfff;
> ++
> ++ entry[0] = 0x1c00000f | (hi & 0xfffff) << 5;
> ++ entry[1] = ((GOT_ENTRY_SIZE == 8 ? 0x28c001ef : 0x288001ef)
> ++ | (lo & 0xfff) << 10);
> ++ entry[2] = 0x4c0001ed; /* jirl $r13, $15, 0 */
> ++ entry[3] = 0x03400000; /* nop */
> ++
> ++ return TRUE;
> ++}
> ++
> ++/* Create an entry in an LoongArch ELF linker hash table. */
> ++
> ++static struct bfd_hash_entry *
> ++link_hash_newfunc (struct bfd_hash_entry *entry, struct bfd_hash_table
> *table,
> ++ const char *string)
> ++{
> ++ struct loongarch_elf_link_hash_entry *eh;
> ++
> ++ /* Allocate the structure if it has not already been allocated by a
> ++ subclass. */
> ++ if (entry == NULL)
> ++ {
> ++ entry = bfd_hash_allocate (table, sizeof (*eh));
> ++ if (entry == NULL)
> ++ return entry;
> ++ }
> ++
> ++ /* Call the allocation method of the superclass. */
> ++ entry = _bfd_elf_link_hash_newfunc (entry, table, string);
> ++ if (entry != NULL)
> ++ {
> ++ eh = (struct loongarch_elf_link_hash_entry *) entry;
> ++ eh->tls_type = GOT_UNKNOWN;
> ++ }
> ++
> ++ return entry;
> ++}
> ++
> ++/* Compute a hash of a local hash entry. We use elf_link_hash_entry
> ++ for local symbol so that we can handle local STT_GNU_IFUNC symbols
> ++ as global symbol. We reuse indx and dynstr_index for local symbol
> ++ hash since they aren't used by global symbols in this backend. */
> ++
> ++static hashval_t
> ++elfNN_loongarch_local_htab_hash (const void *ptr)
> ++{
> ++ struct elf_link_hash_entry *h = (struct elf_link_hash_entry *) ptr;
> ++ return ELF_LOCAL_SYMBOL_HASH (h->indx, h->dynstr_index);
> ++}
> ++
> ++/* Compare local hash entries. */
> ++
> ++static int
> ++elfNN_loongarch_local_htab_eq (const void *ptr1, const void *ptr2)
> ++{
> ++ struct elf_link_hash_entry *h1 = (struct elf_link_hash_entry *) ptr1;
> ++ struct elf_link_hash_entry *h2 = (struct elf_link_hash_entry *) ptr2;
> ++
> ++ return h1->indx == h2->indx && h1->dynstr_index == h2->dynstr_index;
> ++}
> ++
> ++/* Find and/or create a hash entry for local symbol. */
> ++static struct elf_link_hash_entry *
> ++elfNN_loongarch_get_local_sym_hash (struct loongarch_elf_link_hash_table
> *htab,
> ++ bfd *abfd, const Elf_Internal_Rela
> *rel,
> ++ bfd_boolean create)
> ++{
> ++ struct loongarch_elf_link_hash_entry e, *ret;
> ++ asection *sec = abfd->sections;
> ++ hashval_t h = ELF_LOCAL_SYMBOL_HASH (sec->id, ELFNN_R_SYM
> (rel->r_info));
> ++ void **slot;
> ++
> ++ e.elf.indx = sec->id;
> ++ e.elf.dynstr_index = ELFNN_R_SYM (rel->r_info);
> ++ slot = htab_find_slot_with_hash (htab->loc_hash_table, &e, h,
> ++ create ? INSERT : NO_INSERT);
> ++
> ++ if (!slot)
> ++ return NULL;
> ++
> ++ if (*slot)
> ++ {
> ++ ret = (struct loongarch_elf_link_hash_entry *) *slot;
> ++ return &ret->elf;
> ++ }
> ++
> ++ ret = ((struct loongarch_elf_link_hash_entry *)
> ++ objalloc_alloc ((struct objalloc *) htab->loc_hash_memory,
> ++ sizeof (struct loongarch_elf_link_hash_entry)));
> ++ if (ret)
> ++ {
> ++ memset (ret, 0, sizeof (*ret));
> ++ ret->elf.indx = sec->id;
> ++ ret->elf.pointer_equality_needed = 0;
> ++ ret->elf.dynstr_index = ELFNN_R_SYM (rel->r_info);
> ++ ret->elf.dynindx = -1;
> ++ ret->elf.needs_plt = 0;
> ++ ret->elf.plt.refcount = -1;
> ++ ret->elf.got.refcount = -1;
> ++ ret->elf.def_dynamic = 0;
> ++ ret->elf.def_regular = 1;
> ++ ret->elf.ref_dynamic = 0; /* This should be always 0 for local. */
> ++ ret->elf.ref_regular = 0;
> ++ ret->elf.forced_local = 1;
> ++ ret->elf.root.type = bfd_link_hash_defined;
> ++ *slot = ret;
> ++ }
> ++ return &ret->elf;
> ++}
> ++
> ++/* Destroy an LoongArch elf linker hash table. */
> ++
> ++static void
> ++elfNN_loongarch_link_hash_table_free (bfd *obfd)
> ++{
> ++ struct loongarch_elf_link_hash_table *ret;
> ++ ret = (struct loongarch_elf_link_hash_table *) obfd->link.hash;
> ++
> ++ if (ret->loc_hash_table)
> ++ htab_delete (ret->loc_hash_table);
> ++ if (ret->loc_hash_memory)
> ++ objalloc_free ((struct objalloc *) ret->loc_hash_memory);
> ++
> ++ _bfd_elf_link_hash_table_free (obfd);
> ++}
> ++
> ++/* Create a LoongArch ELF linker hash table. */
> ++
> ++static struct bfd_link_hash_table *
> ++loongarch_elf_link_hash_table_create (bfd *abfd)
> ++{
> ++ struct loongarch_elf_link_hash_table *ret;
> ++ bfd_size_type amt = sizeof (struct loongarch_elf_link_hash_table);
> ++
> ++ ret = (struct loongarch_elf_link_hash_table *) bfd_zmalloc (amt);
> ++ if (ret == NULL)
> ++ return NULL;
> ++
> ++ if (!_bfd_elf_link_hash_table_init
> ++ (&ret->elf, abfd, link_hash_newfunc,
> ++ sizeof (struct loongarch_elf_link_hash_entry), LARCH_ELF_DATA))
> ++ {
> ++ free (ret);
> ++ return NULL;
> ++ }
> ++
> ++ ret->max_alignment = MINUS_ONE;
> ++
> ++ ret->loc_hash_table = htab_try_create (1024,
> elfNN_loongarch_local_htab_hash,
> ++ elfNN_loongarch_local_htab_eq,
> NULL);
> ++ ret->loc_hash_memory = objalloc_create ();
> ++ if (!ret->loc_hash_table || !ret->loc_hash_memory)
> ++ {
> ++ elfNN_loongarch_link_hash_table_free (abfd);
> ++ return NULL;
> ++ }
> ++ ret->elf.root.hash_table_free = elfNN_loongarch_link_hash_table_free;
> ++
> ++ return &ret->elf.root;
> ++}
> ++
> ++/* Merge backend specific data from an object file to the output
> ++ object file when linking. */
> ++
> ++static bfd_boolean
> ++elfNN_loongarch_merge_private_bfd_data (bfd *ibfd, struct bfd_link_info
> *info)
> ++{
> ++ bfd *obfd = info->output_bfd;
> ++ flagword in_flags = elf_elfheader (ibfd)->e_flags;
> ++ flagword out_flags = elf_elfheader (obfd)->e_flags;
> ++
> ++ if (!is_loongarch_elf (ibfd) || !is_loongarch_elf (obfd))
> ++ return TRUE;
> ++
> ++ if (strcmp (bfd_get_target (ibfd), bfd_get_target (obfd)) != 0)
> ++ {
> ++ _bfd_error_handler (_("%pB: ABI is incompatible with that of "
> ++ "the selected emulation:\n"
> ++ " target emulation `%s' does not match `%s'"),
> ++ ibfd, bfd_get_target (ibfd), bfd_get_target
> (obfd));
> ++ return FALSE;
> ++ }
> ++
> ++ if (!_bfd_elf_merge_object_attributes (ibfd, info))
> ++ return FALSE;
> ++
> ++ /* If the input BFD is not a dynamic object and it does not contain any
> ++ non-data sections, do not account its ABI. For example, various
> ++ packages produces such data-only relocatable objects with
> ++ `ld -r -b binary` or `objcopy`, and these objects have zero e_flags.
> ++ But they are compatible with all ABIs. */
> ++ if (!(ibfd->flags & DYNAMIC))
> ++ {
> ++ asection *sec;
> ++ bfd_boolean have_code_sections = FALSE;
> ++ for (sec = ibfd->sections; sec != NULL; sec = sec->next)
> ++ if ((bfd_section_flags (sec)
> ++ & (SEC_LOAD | SEC_CODE | SEC_HAS_CONTENTS))
> ++ == (SEC_LOAD | SEC_CODE | SEC_HAS_CONTENTS))
> ++ {
> ++ have_code_sections = TRUE;
> ++ break;
> ++ }
> ++ if (!have_code_sections)
> ++ return TRUE;
> ++ }
> ++
> ++ if (!elf_flags_init (obfd))
> ++ {
> ++ elf_flags_init (obfd) = TRUE;
> ++ elf_elfheader (obfd)->e_flags = in_flags;
> ++ return TRUE;
> ++ }
> ++
> ++ /* Disallow linking different ABIs. */
> ++ if (EF_LOONGARCH_ABI(out_flags ^ in_flags) & EF_LOONGARCH_ABI_MASK)
> ++ {
> ++ _bfd_error_handler (_("%pB: can't link different ABI object."),
> ibfd);
> ++ goto fail;
> ++ }
> ++
> ++ return TRUE;
> ++
> ++ fail:
> ++ bfd_set_error (bfd_error_bad_value);
> ++ return FALSE;
> ++}
> ++
> ++/* Create the .got section. */
> ++
> ++static bfd_boolean
> ++loongarch_elf_create_got_section (bfd *abfd, struct bfd_link_info *info)
> ++{
> ++ flagword flags;
> ++ char *name;
> ++ asection *s, *s_got;
> ++ struct elf_link_hash_entry *h;
> ++ const struct elf_backend_data *bed = get_elf_backend_data (abfd);
> ++ struct elf_link_hash_table *htab = elf_hash_table (info);
> ++
> ++ /* This function may be called more than once. */
> ++ if (htab->sgot != NULL)
> ++ return TRUE;
> ++
> ++ flags = bed->dynamic_sec_flags;
> ++ name = bed->rela_plts_and_copies_p ? ".rela.got" : ".rel.got";
> ++ s = bfd_make_section_anyway_with_flags (abfd, name, flags |
> SEC_READONLY);
> ++
> ++ if (s == NULL || !bfd_set_section_alignment (s,
> bed->s->log_file_align))
> ++ return FALSE;
> ++ htab->srelgot = s;
> ++
> ++ s = s_got = bfd_make_section_anyway_with_flags (abfd, ".got", flags);
> ++ if (s == NULL || !bfd_set_section_alignment (s,
> bed->s->log_file_align))
> ++ return FALSE;
> ++ htab->sgot = s;
> ++
> ++ /* The first bit of the global offset table is the header. */
> ++ s->size += bed->got_header_size;
> ++
> ++ if (bed->want_got_plt)
> ++ {
> ++ s = bfd_make_section_anyway_with_flags (abfd, ".got.plt", flags);
> ++ if (s == NULL || !bfd_set_section_alignment (s,
> bed->s->log_file_align))
> ++ return FALSE;
> ++ htab->sgotplt = s;
> ++
> ++ /* Reserve room for the header. */
> ++ s->size = GOTPLT_HEADER_SIZE;
> ++ }
> ++
> ++ if (bed->want_got_sym)
> ++ {
> ++ /* Define the symbol _GLOBAL_OFFSET_TABLE_ at the start of the .got
> ++ section. We don't do this in the linker script because we don't
> want
> ++ to define the symbol if we are not creating a global offset
> table. */
> ++ h = _bfd_elf_define_linkage_sym (abfd, info, s_got,
> ++ "_GLOBAL_OFFSET_TABLE_");
> ++ elf_hash_table (info)->hgot = h;
> ++ if (h == NULL)
> ++ return FALSE;
> ++ }
> ++ return TRUE;
> ++}
> ++
> ++/* Create .plt, .rela.plt, .got, .got.plt, .rela.got, .dynbss, and
> ++ .rela.bss sections in DYNOBJ, and set up shortcuts to them in our
> ++ hash table. */
> ++
> ++static bfd_boolean
> ++loongarch_elf_create_dynamic_sections (bfd *dynobj, struct bfd_link_info
> *info)
> ++{
> ++ struct loongarch_elf_link_hash_table *htab;
> ++
> ++ htab = loongarch_elf_hash_table (info);
> ++ BFD_ASSERT (htab != NULL);
> ++
> ++ if (!loongarch_elf_create_got_section (dynobj, info))
> ++ return FALSE;
> ++
> ++ if (!_bfd_elf_create_dynamic_sections (dynobj, info))
> ++ return FALSE;
> ++
> ++ if (!bfd_link_pic (info))
> ++ htab->sdyntdata
> ++ = bfd_make_section_anyway_with_flags (dynobj, ".tdata.dyn",
> ++ SEC_ALLOC | SEC_THREAD_LOCAL);
> ++
> ++ if (!htab->elf.splt || !htab->elf.srelplt || !htab->elf.sdynbss
> ++ || (!bfd_link_pic (info) && (!htab->elf.srelbss ||
> !htab->sdyntdata)))
> ++ abort ();
> ++
> ++ return TRUE;
> ++}
> ++
> ++static bfd_boolean
> ++loongarch_elf_record_tls_and_got_reference (bfd *abfd,
> ++ struct bfd_link_info *info,
> ++ struct elf_link_hash_entry *h,
> ++ unsigned long symndx,
> ++ char tls_type)
> ++{
> ++ struct loongarch_elf_link_hash_table *htab = loongarch_elf_hash_table
> (info);
> ++ Elf_Internal_Shdr *symtab_hdr = &elf_tdata (abfd)->symtab_hdr;
> ++
> ++ /* This is a global offset table entry for a local symbol. */
> ++ if (elf_local_got_refcounts (abfd) == NULL)
> ++ {
> ++ bfd_size_type size =
> ++ symtab_hdr->sh_info * (sizeof (bfd_vma) + sizeof (tls_type));
> ++ if (!(elf_local_got_refcounts (abfd) = bfd_zalloc (abfd, size)))
> ++ return FALSE;
> ++ _bfd_loongarch_elf_local_got_tls_type (abfd) =
> ++ (char *) (elf_local_got_refcounts (abfd) + symtab_hdr->sh_info);
> ++ }
> ++
> ++ switch (tls_type)
> ++ {
> ++ case GOT_NORMAL:
> ++ case GOT_TLS_GD:
> ++ case GOT_TLS_IE:
> ++ /* Need GOT. */
> ++ if (htab->elf.sgot == NULL
> ++ && !loongarch_elf_create_got_section (htab->elf.dynobj, info))
> ++ return FALSE;
> ++ if (h)
> ++ {
> ++ if (h->got.refcount < 0)
> ++ h->got.refcount = 0;
> ++ h->got.refcount++;
> ++ }
> ++ else
> ++ elf_local_got_refcounts (abfd)[symndx]++;
> ++ break;
> ++ case GOT_TLS_LE:
> ++ /* No need for GOT. */
> ++ break;
> ++ default:
> ++ _bfd_error_handler (_("Internal error: unreachable."));
> ++ return FALSE;
> ++ }
> ++
> ++ char *new_tls_type = &_bfd_loongarch_elf_tls_type (abfd, h, symndx);
> ++ *new_tls_type |= tls_type;
> ++ if ((*new_tls_type & GOT_NORMAL) && (*new_tls_type & ~GOT_NORMAL))
> ++ {
> ++ _bfd_error_handler (_("%pB: `%s' accessed both as normal and "
> ++ "thread local symbol"),
> ++ abfd,
> ++ h ? h->root.root.string : "<local>");
> ++ return FALSE;
> ++ }
> ++
> ++ return TRUE;
> ++}
> ++
> ++/* Look through the relocs for a section during the first phase, and
> ++ allocate space in the global offset table or procedure linkage
> ++ table. */
> ++
> ++static bfd_boolean
> ++loongarch_elf_check_relocs (bfd *abfd, struct bfd_link_info *info,
> ++ asection *sec, const Elf_Internal_Rela *relocs)
> ++{
> ++ struct loongarch_elf_link_hash_table *htab;
> ++ Elf_Internal_Shdr *symtab_hdr;
> ++ struct elf_link_hash_entry **sym_hashes;
> ++ const Elf_Internal_Rela *rel;
> ++ asection *sreloc = NULL;
> ++
> ++ if (bfd_link_relocatable (info))
> ++ return TRUE;
> ++
> ++ htab = loongarch_elf_hash_table (info);
> ++ symtab_hdr = &elf_tdata (abfd)->symtab_hdr;
> ++ sym_hashes = elf_sym_hashes (abfd);
> ++
> ++ if (htab->elf.dynobj == NULL)
> ++ htab->elf.dynobj = abfd;
> ++
> ++ for (rel = relocs; rel < relocs + sec->reloc_count; rel++)
> ++ {
> ++ unsigned int r_type;
> ++ unsigned int r_symndx;
> ++ struct elf_link_hash_entry *h;
> ++ Elf_Internal_Sym *isym = NULL;
> ++
> ++ r_symndx = ELFNN_R_SYM (rel->r_info);
> ++ r_type = ELFNN_R_TYPE (rel->r_info);
> ++
> ++ if (r_symndx >= NUM_SHDR_ENTRIES (symtab_hdr))
> ++ {
> ++ _bfd_error_handler (_("%pB: bad symbol index: %d"), abfd,
> r_symndx);
> ++ return FALSE;
> ++ }
> ++
> ++ if (r_symndx < symtab_hdr->sh_info)
> ++ {
> ++ /* A local symbol. */
> ++ isym = bfd_sym_from_r_symndx (&htab->elf.sym_cache, abfd,
> r_symndx);
> ++ if (isym == NULL)
> ++ return FALSE;
> ++
> ++ if (ELF_ST_TYPE (isym->st_info) == STT_GNU_IFUNC)
> ++ {
> ++ h = elfNN_loongarch_get_local_sym_hash (htab, abfd, rel,
> TRUE);
> ++ if (h == NULL)
> ++ return FALSE;
> ++
> ++ h->type = STT_GNU_IFUNC;
> ++ h->ref_regular = 1;
> ++ }
> ++ else
> ++ h = NULL;
> ++ }
> ++ else
> ++ {
> ++ h = sym_hashes[r_symndx - symtab_hdr->sh_info];
> ++ while (h->root.type == bfd_link_hash_indirect
> ++ || h->root.type == bfd_link_hash_warning)
> ++ h = (struct elf_link_hash_entry *) h->root.u.i.link;
> ++ }
> ++
> ++ /* It is referenced by a non-shared object. */
> ++ if (h != NULL)
> ++ h->ref_regular = 1;
> ++
> ++ if (h && h->type == STT_GNU_IFUNC)
> ++ {
> ++ if (htab->elf.dynobj == NULL)
> ++ htab->elf.dynobj = abfd;
> ++
> ++ /* Create 'irelifunc' in PIC object. */
> ++ if (bfd_link_pic (info)
> ++ && !_bfd_elf_create_ifunc_sections (htab->elf.dynobj, info))
> ++ return FALSE;
> ++ /* If '.plt' not represent, create '.iplt' to deal with ifunc.
> */
> ++ else if (!htab->elf.splt
> ++ && !_bfd_elf_create_ifunc_sections (htab->elf.dynobj,
> info))
> ++ return FALSE;
> ++ /* Create the ifunc sections, iplt and ipltgot, for static
> ++ executables. */
> ++ if ((r_type == R_LARCH_64 || r_type == R_LARCH_32)
> ++ && !_bfd_elf_create_ifunc_sections (htab->elf.dynobj, info))
> ++ return FALSE;
> ++
> ++ if (h->plt.refcount < 0)
> ++ h->plt.refcount = 0;
> ++ h->plt.refcount++;
> ++ h->needs_plt = 1;
> ++
> ++ elf_tdata (info->output_bfd)->has_gnu_osabi |=
> elf_gnu_osabi_ifunc;
> ++ }
> ++
> ++ int need_dynreloc = 0;
> ++ int only_need_pcrel = 0;
> ++
> ++ switch (r_type)
> ++ {
> ++ case R_LARCH_GOT_PC_HI20:
> ++ case R_LARCH_GOT_HI20:
> ++ case R_LARCH_SOP_PUSH_GPREL:
> ++ /* For la.global. */
> ++ if (h)
> ++ h->pointer_equality_needed = 1;
> ++ if (!loongarch_elf_record_tls_and_got_reference (abfd, info, h,
> ++ r_symndx,
> ++ GOT_NORMAL))
> ++ return FALSE;
> ++ break;
> ++
> ++ case R_LARCH_TLS_LD_PC_HI20:
> ++ case R_LARCH_TLS_LD_HI20:
> ++ case R_LARCH_TLS_GD_PC_HI20:
> ++ case R_LARCH_TLS_GD_HI20:
> ++ case R_LARCH_SOP_PUSH_TLS_GD:
> ++ if (!loongarch_elf_record_tls_and_got_reference (abfd, info, h,
> ++ r_symndx,
> ++ GOT_TLS_GD))
> ++ return FALSE;
> ++ break;
> ++
> ++ case R_LARCH_TLS_IE_PC_HI20:
> ++ case R_LARCH_TLS_IE_HI20:
> ++ case R_LARCH_SOP_PUSH_TLS_GOT:
> ++ if (bfd_link_pic (info))
> ++ /* May fail for lazy-bind. */
> ++ info->flags |= DF_STATIC_TLS;
> ++
> ++ if (!loongarch_elf_record_tls_and_got_reference (abfd, info, h,
> ++ r_symndx,
> ++ GOT_TLS_IE))
> ++ return FALSE;
> ++ break;
> ++
> ++ case R_LARCH_TLS_LE_HI20:
> ++ case R_LARCH_SOP_PUSH_TLS_TPREL:
> ++ if (!bfd_link_executable (info))
> ++ return FALSE;
> ++
> ++ info->flags |= DF_STATIC_TLS;
> ++
> ++ if (!loongarch_elf_record_tls_and_got_reference (abfd, info, h,
> ++ r_symndx,
> ++ GOT_TLS_LE))
> ++ return FALSE;
> ++ break;
> ++
> ++ case R_LARCH_ABS_HI20:
> ++ case R_LARCH_SOP_PUSH_ABSOLUTE:
> ++ if (h != NULL)
> ++ /* If this reloc is in a read-only section, we might
> ++ need a copy reloc. We can't check reliably at this
> ++ stage whether the section is read-only, as input
> ++ sections have not yet been mapped to output sections.
> ++ Tentatively set the flag for now, and correct in
> ++ adjust_dynamic_symbol. */
> ++ h->non_got_ref = 1;
> ++ break;
> ++
> ++ case R_LARCH_PCALA_HI20:
> ++ if (h != NULL)
> ++ {
> ++ h->non_got_ref = 1;
> ++ h->pointer_equality_needed = 1;
> ++ }
> ++
> ++ break;
> ++
> ++ case R_LARCH_B21:
> ++ case R_LARCH_B16:
> ++ case R_LARCH_B26:
> ++ if (h != NULL)
> ++ {
> ++ h->needs_plt = 1;
> ++ if (!bfd_link_pic (info))
> ++ h->non_got_ref = 1;
> ++
> ++ /* We try to create PLT stub for all non-local function. */
> ++ if (h->plt.refcount < 0)
> ++ h->plt.refcount = 0;
> ++ h->plt.refcount++;
> ++ }
> ++
> ++ break;
> ++
> ++ case R_LARCH_SOP_PUSH_PCREL:
> ++ if (h != NULL)
> ++ {
> ++ if (!bfd_link_pic (info))
> ++ h->non_got_ref = 1;
> ++
> ++ /* We try to create PLT stub for all non-local function. */
> ++ if (h->plt.refcount < 0)
> ++ h->plt.refcount = 0;
> ++ h->plt.refcount++;
> ++ h->pointer_equality_needed = 1;
> ++ }
> ++
> ++ break;
> ++
> ++ case R_LARCH_SOP_PUSH_PLT_PCREL:
> ++ /* This symbol requires a procedure linkage table entry. We
> ++ actually build the entry in adjust_dynamic_symbol,
> ++ because this might be a case of linking PIC code without
> ++ linking in any dynamic objects, in which case we don't
> ++ need to generate a procedure linkage table after all. */
> ++ if (h != NULL)
> ++ {
> ++ h->needs_plt = 1;
> ++ if (h->plt.refcount < 0)
> ++ h->plt.refcount = 0;
> ++ h->plt.refcount++;
> ++ }
> ++ break;
> ++
> ++ case R_LARCH_TLS_DTPREL32:
> ++ case R_LARCH_TLS_DTPREL64:
> ++ need_dynreloc = 1;
> ++ only_need_pcrel = 1;
> ++ break;
> ++
> ++ case R_LARCH_JUMP_SLOT:
> ++ case R_LARCH_32:
> ++ case R_LARCH_64:
> ++
> ++ need_dynreloc = 1;
> ++
> ++ /* If resolved symbol is defined in this object,
> ++ 1. Under pie, the symbol is known. We convert it
> ++ into R_LARCH_RELATIVE and need load-addr still.
> ++ 2. Under pde, the symbol is known and we can discard
> R_LARCH_NN.
> ++ 3. Under dll, R_LARCH_NN can't be changed normally, since
> ++ its defination could be covered by the one in executable.
> ++ For symbolic, we convert it into R_LARCH_RELATIVE.
> ++ Thus, only under pde, it needs pcrel only. We discard it. */
> ++ only_need_pcrel = bfd_link_pde (info);
> ++
> ++ if (h != NULL
> ++ && (!bfd_link_pic (info)
> ++ || h->type == STT_GNU_IFUNC))
> ++ {
> ++ /* This reloc might not bind locally. */
> ++ h->non_got_ref = 1;
> ++ h->pointer_equality_needed = 1;
> ++
> ++ if (!h->def_regular
> ++ || (sec->flags & (SEC_CODE | SEC_READONLY)) != 0)
> ++ {
> ++ /* We may need a .plt entry if the symbol is a function
> ++ defined in a shared lib or is a function referenced
> ++ from the code or read-only section. */
> ++ h->plt.refcount += 1;
> ++ }
> ++ }
> ++ break;
> ++
> ++ case R_LARCH_GNU_VTINHERIT:
> ++ if (!bfd_elf_gc_record_vtinherit (abfd, sec, h, rel->r_offset))
> ++ return FALSE;
> ++ break;
> ++
> ++ case R_LARCH_GNU_VTENTRY:
> ++ if (!bfd_elf_gc_record_vtentry (abfd, sec, h, rel->r_addend))
> ++ return FALSE;
> ++ break;
> ++
> ++ default:
> ++ break;
> ++ }
> ++
> ++ /* Record some info for sizing and allocating dynamic entry. */
> ++ if (need_dynreloc && (sec->flags & SEC_ALLOC))
> ++ {
> ++ /* When creating a shared object, we must copy these
> ++ relocs into the output file. We create a reloc
> ++ section in dynobj and make room for the reloc. */
> ++ struct elf_dyn_relocs *p;
> ++ struct elf_dyn_relocs **head;
> ++
> ++ if (sreloc == NULL)
> ++ {
> ++ sreloc
> ++ = _bfd_elf_make_dynamic_reloc_section (sec,
> htab->elf.dynobj,
> ++
> LARCH_ELF_LOG_WORD_BYTES,
> ++ abfd, /*rela?*/
> TRUE);
> ++ if (sreloc == NULL)
> ++ return FALSE;
> ++ }
> ++
> ++ /* If this is a global symbol, we count the number of
> ++ relocations we need for this symbol. */
> ++ if (h != NULL)
> ++ head = &h->dyn_relocs;
> ++ else
> ++ {
> ++ /* Track dynamic relocs needed for local syms too.
> ++ We really need local syms available to do this
> ++ easily. Oh well. */
> ++
> ++ asection *s;
> ++ void *vpp;
> ++
> ++ s = bfd_section_from_elf_index (abfd, isym->st_shndx);
> ++ if (s == NULL)
> ++ s = sec;
> ++
> ++ vpp = &elf_section_data (s)->local_dynrel;
> ++ head = (struct elf_dyn_relocs **) vpp;
> ++ }
> ++
> ++ p = *head;
> ++ if (p == NULL || p->sec != sec)
> ++ {
> ++ bfd_size_type amt = sizeof *p;
> ++ p = (struct elf_dyn_relocs *) bfd_alloc (htab->elf.dynobj,
> amt);
> ++ if (p == NULL)
> ++ return FALSE;
> ++ p->next = *head;
> ++ *head = p;
> ++ p->sec = sec;
> ++ p->count = 0;
> ++ p->pc_count = 0;
> ++ }
> ++
> ++ p->count++;
> ++ p->pc_count += only_need_pcrel;
> ++ }
> ++ }
> ++
> ++ return TRUE;
> ++}
> ++
> ++/* Find dynamic relocs for H that apply to read-only sections. */
> ++
> ++static asection *
> ++readonly_dynrelocs (struct elf_link_hash_entry *h)
> ++{
> ++ struct elf_dyn_relocs *p;
> ++
> ++ for (p = h->dyn_relocs; p != NULL; p = p->next)
> ++ {
> ++ asection *s = p->sec->output_section;
> ++
> ++ if (s != NULL && (s->flags & SEC_READONLY) != 0)
> ++ return p->sec;
> ++ }
> ++ return NULL;
> ++}
> ++
> ++/* Adjust a symbol defined by a dynamic object and referenced by a
> ++ regular object. The current definition is in some section of the
> ++ dynamic object, but we're not including those sections. We have to
> ++ change the definition to something the rest of the link can
> ++ understand. */
> ++static bfd_boolean
> ++loongarch_elf_adjust_dynamic_symbol (struct bfd_link_info *info,
> ++ struct elf_link_hash_entry *h)
> ++{
> ++ struct loongarch_elf_link_hash_table *htab;
> ++ bfd *dynobj;
> ++
> ++ htab = loongarch_elf_hash_table (info);
> ++ BFD_ASSERT (htab != NULL);
> ++
> ++ dynobj = htab->elf.dynobj;
> ++
> ++ /* Make sure we know what is going on here. */
> ++ BFD_ASSERT (dynobj != NULL
> ++ && (h->needs_plt || h->type == STT_GNU_IFUNC ||
> h->is_weakalias
> ++ || (h->def_dynamic && h->ref_regular &&
> !h->def_regular)));
> ++
> ++ /* If this is a function, put it in the procedure linkage table. We
> ++ will fill in the contents of the procedure linkage table later
> ++ (although we could actually do it here). */
> ++ if (h->type == STT_FUNC || h->type == STT_GNU_IFUNC || h->needs_plt)
> ++ {
> ++ if (h->plt.refcount < 0
> ++ || (h->type != STT_GNU_IFUNC
> ++ && (SYMBOL_REFERENCES_LOCAL (info, h)
> ++ || (ELF_ST_VISIBILITY (h->other) != STV_DEFAULT
> ++ && h->root.type == bfd_link_hash_undefweak))))
> ++ {
> ++ /* This case can occur if we saw a R_LARCH_SOP_PUSH_PLT_PCREL
> reloc
> ++ in an input file, but the symbol was never referred to by a
> ++ dynamic object, or if all references were garbage collected.
> ++ In such a case, we don't actually need to build a PLT entry.
> */
> ++ h->plt.offset = MINUS_ONE;
> ++ h->needs_plt = 0;
> ++ }
> ++ else
> ++ h->needs_plt = 1;
> ++
> ++ return TRUE;
> ++ }
> ++ else
> ++ h->plt.offset = MINUS_ONE;
> ++
> ++ /* If this is a weak symbol, and there is a real definition, the
> ++ processor independent code will have arranged for us to see the
> ++ real definition first, and we can just use the same value. */
> ++ if (h->is_weakalias)
> ++ {
> ++ struct elf_link_hash_entry *def = weakdef (h);
> ++ BFD_ASSERT (def->root.type == bfd_link_hash_defined);
> ++ h->root.u.def.section = def->root.u.def.section;
> ++ h->root.u.def.value = def->root.u.def.value;
> ++ return TRUE;
> ++ }
> ++
> ++ /* R_LARCH_COPY is not adept glibc, not to generate. */
> ++ /* Can not print anything, because make check ld. */
> ++ return TRUE;
> ++}
> ++
> ++/* Allocate space in .plt, .got and associated reloc sections for
> ++ dynamic relocs. */
> ++
> ++static bfd_boolean
> ++allocate_dynrelocs (struct elf_link_hash_entry *h, void *inf)
> ++{
> ++ struct bfd_link_info *info;
> ++ struct loongarch_elf_link_hash_table *htab;
> ++ struct elf_dyn_relocs *p;
> ++
> ++ if (h->root.type == bfd_link_hash_indirect)
> ++ return TRUE;
> ++
> ++ if (h->type == STT_GNU_IFUNC
> ++ && h->def_regular)
> ++ return TRUE;
> ++
> ++ info = (struct bfd_link_info *) inf;
> ++ htab = loongarch_elf_hash_table (info);
> ++ bfd_boolean dyn = htab->elf.dynamic_sections_created;
> ++ BFD_ASSERT (htab != NULL);
> ++
> ++ do
> ++ {
> ++ asection *plt, *gotplt, *relplt;
> ++
> ++ if (!h->needs_plt)
> ++ break;
> ++
> ++ h->needs_plt = 0;
> ++
> ++ if (htab->elf.splt)
> ++ {
> ++ if (h->dynindx == -1 && !h->forced_local && dyn
> ++ && h->root.type == bfd_link_hash_undefweak)
> ++ {
> ++ if (!bfd_elf_link_record_dynamic_symbol (info, h))
> ++ return FALSE;
> ++ }
> ++
> ++ if (!WILL_CALL_FINISH_DYNAMIC_SYMBOL (1, bfd_link_pic (info), h)
> ++ && h->type != STT_GNU_IFUNC)
> ++ break;
> ++
> ++ plt = htab->elf.splt;
> ++ gotplt = htab->elf.sgotplt;
> ++ relplt = htab->elf.srelplt;
> ++ }
> ++ else if (htab->elf.iplt)
> ++ {
> ++ /* .iplt only for IFUNC. */
> ++ if (h->type != STT_GNU_IFUNC)
> ++ break;
> ++
> ++ plt = htab->elf.iplt;
> ++ gotplt = htab->elf.igotplt;
> ++ relplt = htab->elf.irelplt;
> ++ }
> ++ else
> ++ break;
> ++
> ++ if (plt->size == 0)
> ++ plt->size = PLT_HEADER_SIZE;
> ++
> ++ h->plt.offset = plt->size;
> ++ plt->size += PLT_ENTRY_SIZE;
> ++ gotplt->size += GOT_ENTRY_SIZE;
> ++ relplt->size += sizeof (ElfNN_External_Rela);
> ++
> ++ /* If this symbol is not defined in a regular file, and we are
> ++ not generating a shared library, then set the symbol to this
> ++ location in the .plt. This is required to make function
> ++ pointers compare as equal between the normal executable and
> ++ the shared library. */
> ++ if (!bfd_link_pic (info)
> ++ && !h->def_regular)
> ++ {
> ++ h->root.u.def.section = plt;
> ++ h->root.u.def.value = h->plt.offset;
> ++ }
> ++
> ++ h->needs_plt = 1;
> ++ }
> ++ while (0);
> ++
> ++ if (!h->needs_plt)
> ++ h->plt.offset = MINUS_ONE;
> ++
> ++ if (0 < h->got.refcount)
> ++ {
> ++ asection *s;
> ++ int tls_type = loongarch_elf_hash_entry (h)->tls_type;
> ++
> ++ /* Make sure this symbol is output as a dynamic symbol.
> ++ Undefined weak syms won't yet be marked as dynamic. */
> ++ if (h->dynindx == -1 && !h->forced_local && dyn
> ++ && h->root.type == bfd_link_hash_undefweak)
> ++ {
> ++ if (!bfd_elf_link_record_dynamic_symbol (info, h))
> ++ return FALSE;
> ++ }
> ++
> ++ s = htab->elf.sgot;
> ++ h->got.offset = s->size;
> ++ if (tls_type & (GOT_TLS_GD | GOT_TLS_IE))
> ++ {
> ++ /* TLS_GD needs two dynamic relocs and two GOT slots. */
> ++ if (tls_type & GOT_TLS_GD)
> ++ {
> ++ s->size += 2 * GOT_ENTRY_SIZE;
> ++ if (bfd_link_executable (info))
> ++ {
> ++ /* Link exe and not defined local. */
> ++ if (!SYMBOL_REFERENCES_LOCAL (info, h))
> ++ htab->elf.srelgot->size += 2 * sizeof
> (ElfNN_External_Rela);
> ++ }
> ++ else
> ++ {
> ++ if (SYMBOL_REFERENCES_LOCAL (info, h))
> ++ htab->elf.srelgot->size += sizeof
> (ElfNN_External_Rela);
> ++ else
> ++ htab->elf.srelgot->size += 2 * sizeof
> (ElfNN_External_Rela);
> ++ }
> ++ }
> ++
> ++ /* TLS_IE needs one dynamic reloc and one GOT slot. */
> ++ if (tls_type & GOT_TLS_IE)
> ++ {
> ++ s->size += GOT_ENTRY_SIZE;
> ++
> ++ if (bfd_link_executable (info))
> ++ {
> ++ /* Link exe and not defined local. */
> ++ if (!SYMBOL_REFERENCES_LOCAL (info, h))
> ++ htab->elf.srelgot->size += sizeof
> (ElfNN_External_Rela);
> ++ }
> ++ else
> ++ {
> ++ htab->elf.srelgot->size += sizeof (ElfNN_External_Rela);
> ++ }
> ++ }
> ++ }
> ++ else
> ++ {
> ++ s->size += GOT_ENTRY_SIZE;
> ++ if ((ELF_ST_VISIBILITY (h->other) == STV_DEFAULT
> ++ || h->root.type != bfd_link_hash_undefweak)
> ++ && (bfd_link_pic (info)
> ++ || WILL_CALL_FINISH_DYNAMIC_SYMBOL (dyn, bfd_link_pic
> (info),
> ++ h))
> ++ && !UNDEFWEAK_NO_DYNAMIC_RELOC (info, h))
> ++ /* Undefined weak symbol in static PIE resolves to 0 without
> ++ any dynamic relocations. */
> ++ htab->elf.srelgot->size += sizeof (ElfNN_External_Rela);
> ++ }
> ++ }
> ++ else
> ++ h->got.offset = MINUS_ONE;
> ++
> ++ if (h->dyn_relocs == NULL)
> ++ return TRUE;
> ++
> ++ /* Extra dynamic relocate,
> ++ * R_LARCH_64
> ++ * R_LARCH_TLS_DTPRELNN
> ++ * R_LARCH_JUMP_SLOT
> ++ * R_LARCH_NN. */
> ++
> ++ if (SYMBOL_CALLS_LOCAL (info, h))
> ++ {
> ++ struct elf_dyn_relocs **pp;
> ++
> ++ for (pp = &h->dyn_relocs; (p = *pp) != NULL;)
> ++ {
> ++ p->count -= p->pc_count;
> ++ p->pc_count = 0;
> ++ if (p->count == 0)
> ++ *pp = p->next;
> ++ else
> ++ pp = &p->next;
> ++ }
> ++ }
> ++
> ++ if (h->root.type == bfd_link_hash_undefweak)
> ++ {
> ++ if (UNDEFWEAK_NO_DYNAMIC_RELOC (info, h)
> ++ || ELF_ST_VISIBILITY (h->other) != STV_DEFAULT
> ++ || (!bfd_link_pic (info) && h->non_got_ref))
> ++ h->dyn_relocs = NULL;
> ++ else if (h->dynindx == -1 && !h->forced_local)
> ++ {
> ++ /* Make sure this symbol is output as a dynamic symbol.
> ++ Undefined weak syms won't yet be marked as dynamic. */
> ++ if (!bfd_elf_link_record_dynamic_symbol (info, h))
> ++ return FALSE;
> ++
> ++ if (h->dynindx == -1)
> ++ h->dyn_relocs = NULL;
> ++ }
> ++ }
> ++
> ++ for (p = h->dyn_relocs; p != NULL; p = p->next)
> ++ {
> ++ asection *sreloc = elf_section_data (p->sec)->sreloc;
> ++ sreloc->size += p->count * sizeof (ElfNN_External_Rela);
> ++ }
> ++
> ++ return TRUE;
> ++}
> ++
> ++/* A modified version of _bfd_elf_allocate_ifunc_dyn_relocs.
> ++ For local def and ref ifunc,
> ++ dynamic relocations are stored in
> ++ 1. rela.srelgot section in dynamic object (dll or exec).
> ++ 2. rela.irelplt section in static executable.
> ++ Unlike _bfd_elf_allocate_ifunc_dyn_relocs, rela.srelgot is used
> ++ instead of rela.srelplt. Glibc ELF loader will not support
> ++ R_LARCH_IRELATIVE relocation in rela.plt. */
> ++
> ++static bfd_boolean
> ++local_allocate_ifunc_dyn_relocs (struct bfd_link_info *info,
> ++ struct elf_link_hash_entry *h,
> ++ struct elf_dyn_relocs **head,
> ++ unsigned int plt_entry_size,
> ++ unsigned int plt_header_size,
> ++ unsigned int got_entry_size,
> ++ bfd_boolean avoid_plt)
> ++{
> ++ asection *plt, *gotplt, *relplt;
> ++ struct elf_dyn_relocs *p;
> ++ unsigned int sizeof_reloc;
> ++ const struct elf_backend_data *bed;
> ++ struct elf_link_hash_table *htab;
> ++ /* If AVOID_PLT is TRUE, don't use PLT if possible. */
> ++ bfd_boolean use_plt = !avoid_plt || h->plt.refcount > 0;
> ++ bfd_boolean need_dynreloc = !use_plt || bfd_link_pic (info);
> ++
> ++ /* When a PIC object references a STT_GNU_IFUNC symbol defined
> ++ in executable or it isn't referenced via PLT, the address of
> ++ the resolved function may be used. But in non-PIC executable,
> ++ the address of its plt slot may be used. Pointer equality may
> ++ not work correctly. PIE or non-PLT reference should be used if
> ++ pointer equality is required here.
> ++
> ++ If STT_GNU_IFUNC symbol is defined in position-dependent executable,
> ++ backend should change it to the normal function and set its address
> ++ to its PLT entry which should be resolved by R_*_IRELATIVE at
> ++ run-time. All external references should be resolved to its PLT in
> ++ executable. */
> ++ if (!need_dynreloc
> ++ && !(bfd_link_pde (info) && h->def_regular)
> ++ && (h->dynindx != -1
> ++ || info->export_dynamic)
> ++ && h->pointer_equality_needed)
> ++ {
> ++ info->callbacks->einfo
> ++ /* xgettext:c-format. */
> ++ (_("%F%P: dynamic STT_GNU_IFUNC symbol `%s' with pointer "
> ++ "equality in `%pB' can not be used when making an "
> ++ "executable; recompile with -fPIE and relink with -pie\n"),
> ++ h->root.root.string,
> ++ h->root.u.def.section->owner);
> ++ bfd_set_error (bfd_error_bad_value);
> ++ return FALSE;
> ++ }
> ++
> ++ htab = elf_hash_table (info);
> ++
> ++ /* When the symbol is marked with regular reference, if PLT isn't used
> ++ or we are building a PIC object, we must keep dynamic relocation
> ++ if there is non-GOT reference and use PLT if there is PC-relative
> ++ reference. */
> ++ if (need_dynreloc && h->ref_regular)
> ++ {
> ++ bfd_boolean keep = FALSE;
> ++ for (p = *head; p != NULL; p = p->next)
> ++ if (p->count)
> ++ {
> ++ h->non_got_ref = 1;
> ++ /* Need dynamic relocations for non-GOT reference. */
> ++ keep = TRUE;
> ++ if (p->pc_count)
> ++ {
> ++ /* Must use PLT for PC-relative reference. */
> ++ use_plt = TRUE;
> ++ need_dynreloc = bfd_link_pic (info);
> ++ break;
> ++ }
> ++ }
> ++ if (keep)
> ++ goto keep;
> ++ }
> ++
> ++ /* Support garbage collection against STT_GNU_IFUNC symbols. */
> ++ if (h->plt.refcount <= 0 && h->got.refcount <= 0)
> ++ {
> ++ h->got = htab->init_got_offset;
> ++ h->plt = htab->init_plt_offset;
> ++ *head = NULL;
> ++ return TRUE;
> ++ }
> ++
> ++ /* Return and discard space for dynamic relocations against it if
> ++ it is never referenced. */
> ++ if (!h->ref_regular)
> ++ {
> ++ if (h->plt.refcount > 0
> ++ || h->got.refcount > 0)
> ++ abort ();
> ++ h->got = htab->init_got_offset;
> ++ h->plt = htab->init_plt_offset;
> ++ *head = NULL;
> ++ return TRUE;
> ++ }
> ++
> ++ keep:
> ++ bed = get_elf_backend_data (info->output_bfd);
> ++ if (bed->rela_plts_and_copies_p)
> ++ sizeof_reloc = bed->s->sizeof_rela;
> ++ else
> ++ sizeof_reloc = bed->s->sizeof_rel;
> ++
> ++ /* When building a static executable, use iplt, igot.plt and
> ++ rela.iplt sections for STT_GNU_IFUNC symbols. */
> ++ if (htab->splt != NULL)
> ++ {
> ++ plt = htab->splt;
> ++ gotplt = htab->sgotplt;
> ++ /* Change dynamic info of ifunc gotplt from srelplt to srelgot. */
> ++ relplt = htab->srelgot;
> ++
> ++ /* If this is the first plt entry and PLT is used, make room for
> ++ the special first entry. */
> ++ if (plt->size == 0 && use_plt)
> ++ plt->size += plt_header_size;
> ++ }
> ++ else
> ++ {
> ++ plt = htab->iplt;
> ++ gotplt = htab->igotplt;
> ++ relplt = htab->irelplt;
> ++ }
> ++
> ++ if (use_plt)
> ++ {
> ++ /* Don't update value of STT_GNU_IFUNC symbol to PLT. We need
> ++ the original value for R_*_IRELATIVE. */
> ++ h->plt.offset = plt->size;
> ++
> ++ /* Make room for this entry in the plt/iplt section. */
> ++ plt->size += plt_entry_size;
> ++
> ++ /* We also need to make an entry in the got.plt/got.iplt section,
> ++ which will be placed in the got section by the linker script. */
> ++ gotplt->size += got_entry_size;
> ++ }
> ++
> ++ /* We also need to make an entry in the rela.plt/.rela.iplt
> ++ section for GOTPLT relocation if PLT is used. */
> ++ if (use_plt)
> ++ {
> ++ relplt->size += sizeof_reloc;
> ++ relplt->reloc_count++;
> ++ }
> ++
> ++ /* We need dynamic relocation for STT_GNU_IFUNC symbol only when
> ++ there is a non-GOT reference in a PIC object or PLT isn't used. */
> ++ if (!need_dynreloc || !h->non_got_ref)
> ++ *head = NULL;
> ++
> ++ /* Finally, allocate space. */
> ++ p = *head;
> ++ if (p != NULL)
> ++ {
> ++ bfd_size_type count = 0;
> ++ do
> ++ {
> ++ count += p->count;
> ++ p = p->next;
> ++ }
> ++ while (p != NULL);
> ++
> ++ htab->ifunc_resolvers = count != 0;
> ++
> ++ /* Dynamic relocations are stored in
> ++ 1. rela.srelgot section in PIC object.
> ++ 2. rela.srelgot section in dynamic executable.
> ++ 3. rela.irelplt section in static executable. */
> ++ if (htab->splt != NULL)
> ++ htab->srelgot->size += count * sizeof_reloc;
> ++ else
> ++ {
> ++ relplt->size += count * sizeof_reloc;
> ++ relplt->reloc_count += count;
> ++ }
> ++ }
> ++
> ++ /* For STT_GNU_IFUNC symbol, got.plt has the real function address
> ++ and got has the PLT entry adddress. We will load the GOT entry
> ++ with the PLT entry in finish_dynamic_symbol if it is used. For
> ++ branch, it uses got.plt. For symbol value, if PLT is used,
> ++ 1. Use got.plt in a PIC object if it is forced local or not
> ++ dynamic.
> ++ 2. Use got.plt in a non-PIC object if pointer equality isn't
> ++ needed.
> ++ 3. Use got.plt in PIE.
> ++ 4. Use got.plt if got isn't used.
> ++ 5. Otherwise use got so that it can be shared among different
> ++ objects at run-time.
> ++ If PLT isn't used, always use got for symbol value.
> ++ We only need to relocate got entry in PIC object or in dynamic
> ++ executable without PLT. */
> ++ if (use_plt
> ++ && (h->got.refcount <= 0
> ++ || (bfd_link_pic (info)
> ++ && (h->dynindx == -1
> ++ || h->forced_local))
> ++ || (
> ++ !h->pointer_equality_needed)
> ++ || htab->sgot == NULL))
> ++ {
> ++ /* Use got.plt. */
> ++ h->got.offset = (bfd_vma) -1;
> ++ }
> ++ else
> ++ {
> ++ if (!use_plt)
> ++ {
> ++ /* PLT isn't used. */
> ++ h->plt.offset = (bfd_vma) -1;
> ++ }
> ++ if (h->got.refcount <= 0)
> ++ {
> ++ /* GOT isn't need when there are only relocations for static
> ++ pointers. */
> ++ h->got.offset = (bfd_vma) -1;
> ++ }
> ++ else
> ++ {
> ++ h->got.offset = htab->sgot->size;
> ++ htab->sgot->size += got_entry_size;
> ++ /* Need to relocate the GOT entry in a PIC object or PLT isn't
> ++ used. Otherwise, the GOT entry will be filled with the PLT
> ++ entry and dynamic GOT relocation isn't needed. */
> ++ if (need_dynreloc)
> ++ {
> ++ /* For non-static executable, dynamic GOT relocation is in
> ++ rela.got section, but for static executable, it is
> ++ in rela.iplt section. */
> ++ if (htab->splt != NULL)
> ++ htab->srelgot->size += sizeof_reloc;
> ++ else
> ++ {
> ++ relplt->size += sizeof_reloc;
> ++ relplt->reloc_count++;
> ++ }
> ++ }
> ++ }
> ++ }
> ++
> ++ return TRUE;
> ++}
> ++
> ++/* Allocate space in .plt, .got and associated reloc sections for
> ++ ifunc dynamic relocs. */
> ++
> ++static bfd_boolean
> ++elfNN_allocate_ifunc_dynrelocs (struct elf_link_hash_entry *h, void *inf)
> ++{
> ++ struct bfd_link_info *info;
> ++ /* An example of a bfd_link_hash_indirect symbol is versioned
> ++ symbol. For example: __gxx_personality_v0(bfd_link_hash_indirect)
> ++ -> __gxx_personality_v0(bfd_link_hash_defined)
> ++
> ++ There is no need to process bfd_link_hash_indirect symbols here
> ++ because we will also be presented with the concrete instance of
> ++ the symbol and loongarch_elf_copy_indirect_symbol () will have been
> ++ called to copy all relevant data from the generic to the concrete
> ++ symbol instance. */
> ++ if (h->root.type == bfd_link_hash_indirect)
> ++ return TRUE;
> ++
> ++ if (h->root.type == bfd_link_hash_warning)
> ++ h = (struct elf_link_hash_entry *) h->root.u.i.link;
> ++
> ++ info = (struct bfd_link_info *) inf;
> ++
> ++ /* Since STT_GNU_IFUNC symbol must go through PLT, we handle it
> ++ here if it is defined and referenced in a non-shared object. */
> ++ if (h->type == STT_GNU_IFUNC && h->def_regular)
> ++ {
> ++ if (SYMBOL_REFERENCES_LOCAL (info, h))
> ++ return local_allocate_ifunc_dyn_relocs (info, h,
> ++ &h->dyn_relocs,
> ++ PLT_ENTRY_SIZE,
> ++ PLT_HEADER_SIZE,
> ++ GOT_ENTRY_SIZE,
> ++ FALSE);
> ++ else
> ++ return _bfd_elf_allocate_ifunc_dyn_relocs (info, h,
> ++ &h->dyn_relocs,
> ++ PLT_ENTRY_SIZE,
> ++ PLT_HEADER_SIZE,
> ++ GOT_ENTRY_SIZE,
> ++ FALSE);
> ++ }
> ++
> ++ return TRUE;
> ++}
> ++
> ++/* Allocate space in .plt, .got and associated reloc sections for
> ++ ifunc dynamic relocs. */
> ++
> ++static bfd_boolean
> ++elfNN_allocate_local_ifunc_dynrelocs (void **slot, void *inf)
> ++{
> ++ struct elf_link_hash_entry *h = (struct elf_link_hash_entry *) *slot;
> ++
> ++ if (h->type != STT_GNU_IFUNC
> ++ || !h->def_regular
> ++ || !h->ref_regular
> ++ || !h->forced_local
> ++ || h->root.type != bfd_link_hash_defined)
> ++ abort ();
> ++
> ++ return elfNN_allocate_ifunc_dynrelocs (h, inf);
> ++}
> ++
> ++/* Set DF_TEXTREL if we find any dynamic relocs that apply to
> ++ read-only sections. */
> ++
> ++static bfd_boolean
> ++maybe_set_textrel (struct elf_link_hash_entry *h, void *info_p)
> ++{
> ++ asection *sec;
> ++
> ++ if (h->root.type == bfd_link_hash_indirect)
> ++ return TRUE;
> ++
> ++ sec = readonly_dynrelocs (h);
> ++ if (sec != NULL)
> ++ {
> ++ struct bfd_link_info *info = (struct bfd_link_info *) info_p;
> ++
> ++ info->flags |= DF_TEXTREL;
> ++ info->callbacks->minfo (_("%pB: dynamic relocation against `%pT'
> in "
> ++ "read-only section `%pA'\n"),
> ++ sec->owner, h->root.root.string, sec);
> ++
> ++ /* Not an error, just cut short the traversal. */
> ++ return FALSE;
> ++ }
> ++ return TRUE;
> ++}
> ++
> ++static bfd_boolean
> ++loongarch_elf_size_dynamic_sections (bfd *output_bfd,
> ++ struct bfd_link_info *info)
> ++{
> ++ struct loongarch_elf_link_hash_table *htab;
> ++ bfd *dynobj;
> ++ asection *s;
> ++ bfd *ibfd;
> ++
> ++ htab = loongarch_elf_hash_table (info);
> ++ BFD_ASSERT (htab != NULL);
> ++ dynobj = htab->elf.dynobj;
> ++ BFD_ASSERT (dynobj != NULL);
> ++
> ++ if (htab->elf.dynamic_sections_created)
> ++ {
> ++ /* Set the contents of the .interp section to the interpreter. */
> ++ if (bfd_link_executable (info) && !info->nointerp)
> ++ {
> ++ const char *interpreter;
> ++ flagword flags = elf_elfheader (output_bfd)->e_flags;
> ++ s = bfd_get_linker_section (dynobj, ".interp");
> ++ BFD_ASSERT (s != NULL);
> ++ if (EF_LOONGARCH_IS_ILP32 (flags))
> ++ interpreter = "/lib32/ld.so.1";
> ++ else if (EF_LOONGARCH_IS_LP64 (flags))
> ++ interpreter = "/lib64/ld.so.1";
> ++ else
> ++ interpreter = "/lib/ld.so.1";
> ++ s->contents = (unsigned char *) interpreter;
> ++ s->size = strlen (interpreter) + 1;
> ++ }
> ++ }
> ++
> ++ /* Set up .got offsets for local syms, and space for local dynamic
> ++ relocs. */
> ++ for (ibfd = info->input_bfds; ibfd != NULL; ibfd = ibfd->link.next)
> ++ {
> ++ bfd_signed_vma *local_got;
> ++ bfd_signed_vma *end_local_got;
> ++ char *local_tls_type;
> ++ bfd_size_type locsymcount;
> ++ Elf_Internal_Shdr *symtab_hdr;
> ++ asection *srel;
> ++
> ++ if (!is_loongarch_elf (ibfd))
> ++ continue;
> ++
> ++ for (s = ibfd->sections; s != NULL; s = s->next)
> ++ {
> ++ struct elf_dyn_relocs *p;
> ++
> ++ for (p = elf_section_data (s)->local_dynrel; p != NULL; p =
> p->next)
> ++ {
> ++ p->count -= p->pc_count;
> ++ if (!bfd_is_abs_section (p->sec)
> ++ && bfd_is_abs_section (p->sec->output_section))
> ++ {
> ++ /* Input section has been discarded, either because
> ++ it is a copy of a linkonce section or due to
> ++ linker script /DISCARD/, so we'll be discarding
> ++ the relocs too. */
> ++ }
> ++ else if (0 < p->count)
> ++ {
> ++ srel = elf_section_data (p->sec)->sreloc;
> ++ srel->size += p->count * sizeof (ElfNN_External_Rela);
> ++ if ((p->sec->output_section->flags & SEC_READONLY) != 0)
> ++ info->flags |= DF_TEXTREL;
> ++ }
> ++ }
> ++ }
> ++
> ++ local_got = elf_local_got_refcounts (ibfd);
> ++ if (!local_got)
> ++ continue;
> ++
> ++ symtab_hdr = &elf_symtab_hdr (ibfd);
> ++ locsymcount = symtab_hdr->sh_info;
> ++ end_local_got = local_got + locsymcount;
> ++ local_tls_type = _bfd_loongarch_elf_local_got_tls_type (ibfd);
> ++ s = htab->elf.sgot;
> ++ srel = htab->elf.srelgot;
> ++ for (; local_got < end_local_got; ++local_got, ++local_tls_type)
> ++ {
> ++ if (0 < *local_got)
> ++ {
> ++ *local_got = s->size;
> ++
> ++ /* TLS gd use two got. */
> ++ if (*local_tls_type & GOT_TLS_GD)
> ++ s->size += GOT_ENTRY_SIZE * 2;
> ++ else
> ++ /* Normal got, tls ie/ld use one got. */
> ++ s->size += GOT_ENTRY_SIZE;
> ++
> ++ if (bfd_link_executable (info)
> ++ && (*local_tls_type & (GOT_TLS_GD| GOT_TLS_IE)))
> ++ ;/* Do nothing. */
> ++ else
> ++ {
> ++ srel->size += sizeof (ElfNN_External_Rela);
> ++ }
> ++ }
> ++ else
> ++ *local_got = MINUS_ONE;
> ++ }
> ++ }
> ++
> ++ /* Allocate global sym .plt and .got entries, and space for global
> ++ sym dynamic relocs. */
> ++ elf_link_hash_traverse (&htab->elf, allocate_dynrelocs, info);
> ++
> ++ /* Allocate global ifunc sym .plt and .got entries, and space for
> global
> ++ ifunc sym dynamic relocs. */
> ++ elf_link_hash_traverse (&htab->elf, elfNN_allocate_ifunc_dynrelocs,
> info);
> ++
> ++ /* Allocate .plt and .got entries, and space for local ifunc symbols.
> */
> ++ htab_traverse (htab->loc_hash_table,
> ++ (void *) elfNN_allocate_local_ifunc_dynrelocs, info);
> ++
> ++ /* Don't allocate .got.plt section if there are no PLT. */
> ++ if (htab->elf.sgotplt && htab->elf.sgotplt->size == GOTPLT_HEADER_SIZE
> ++ && (htab->elf.splt == NULL || htab->elf.splt->size == 0))
> ++ htab->elf.sgotplt->size = 0;
> ++
> ++ /* The check_relocs and adjust_dynamic_symbol entry points have
> ++ determined the sizes of the various dynamic sections. Allocate
> ++ memory for them. */
> ++ for (s = dynobj->sections; s != NULL; s = s->next)
> ++ {
> ++ if ((s->flags & SEC_LINKER_CREATED) == 0)
> ++ continue;
> ++
> ++ if (s == htab->elf.splt || s == htab->elf.iplt || s ==
> htab->elf.sgot
> ++ || s == htab->elf.sgotplt || s == htab->elf.igotplt
> ++ || s == htab->elf.sdynbss || s == htab->elf.sdynrelro)
> ++ {
> ++ /* Strip this section if we don't need it; see the
> ++ comment below. */
> ++ }
> ++ else if (strncmp (s->name, ".rela", 5) == 0)
> ++ {
> ++ if (s->size != 0)
> ++ {
> ++ /* We use the reloc_count field as a counter if we need
> ++ to copy relocs into the output file. */
> ++ s->reloc_count = 0;
> ++ }
> ++ }
> ++ else
> ++ {
> ++ /* It's not one of our sections. */
> ++ continue;
> ++ }
> ++
> ++ if (s->size == 0)
> ++ {
> ++ /* If we don't need this section, strip it from the
> ++ output file. This is mostly to handle .rela.bss and
> ++ .rela.plt. We must create both sections in
> ++ create_dynamic_sections, because they must be created
> ++ before the linker maps input sections to output
> ++ sections. The linker does that before
> ++ adjust_dynamic_symbol is called, and it is that
> ++ function which decides whether anything needs to go
> ++ into these sections. */
> ++ s->flags |= SEC_EXCLUDE;
> ++ continue;
> ++ }
> ++
> ++ if ((s->flags & SEC_HAS_CONTENTS) == 0)
> ++ continue;
> ++
> ++ /* Allocate memory for the section contents. Zero the memory
> ++ for the benefit of .rela.plt, which has 4 unused entries
> ++ at the beginning, and we don't want garbage. */
> ++ s->contents = (bfd_byte *) bfd_zalloc (dynobj, s->size);
> ++ if (s->contents == NULL)
> ++ return FALSE;
> ++ }
> ++
> ++ if (elf_hash_table (info)->dynamic_sections_created)
> ++ {
> ++ /* Add some entries to the .dynamic section. We fill in the
> ++ values later, in loongarch_elf_finish_dynamic_sections, but we
> ++ must add the entries now so that we get the correct size for
> ++ the .dynamic section. The DT_DEBUG entry is filled in by the
> ++ dynamic linker and used by the debugger. */
> ++#define add_dynamic_entry(TAG, VAL) _bfd_elf_add_dynamic_entry (info,
> TAG, VAL)
> ++
> ++ if (bfd_link_executable (info))
> ++ {
> ++ if (!add_dynamic_entry (DT_DEBUG, 0))
> ++ return FALSE;
> ++ }
> ++
> ++ if (htab->elf.srelplt->size != 0)
> ++ {
> ++ if (!add_dynamic_entry (DT_PLTGOT, 0)
> ++ || !add_dynamic_entry (DT_PLTRELSZ, 0)
> ++ || !add_dynamic_entry (DT_PLTREL, DT_RELA)
> ++ || !add_dynamic_entry (DT_JMPREL, 0))
> ++ return FALSE;
> ++ }
> ++
> ++ if (!add_dynamic_entry (DT_RELA, 0)
> ++ || !add_dynamic_entry (DT_RELASZ, 0)
> ++ || !add_dynamic_entry (DT_RELAENT, sizeof (ElfNN_External_Rela)))
> ++ return FALSE;
> ++
> ++ /* If any dynamic relocs apply to a read-only section,
> ++ then we need a DT_TEXTREL entry. */
> ++ if ((info->flags & DF_TEXTREL) == 0)
> ++ elf_link_hash_traverse (&htab->elf, maybe_set_textrel, info);
> ++
> ++ if (info->flags & DF_TEXTREL)
> ++ {
> ++ if (!add_dynamic_entry (DT_TEXTREL, 0))
> ++ return FALSE;
> ++ /* Clear the DF_TEXTREL flag. It will be set again if we
> ++ write out an actual text relocation; we may not, because
> ++ at this point we do not know whether e.g. any .eh_frame
> ++ absolute relocations have been converted to PC-relative. */
> ++ info->flags &= ~DF_TEXTREL;
> ++ }
> ++ }
> ++#undef add_dynamic_entry
> ++
> ++ return TRUE;
> ++}
> ++
> ++#define LARCH_LD_STACK_DEPTH 16
> ++static int64_t larch_opc_stack[LARCH_LD_STACK_DEPTH];
> ++static size_t larch_stack_top = 0;
> ++
> ++static bfd_reloc_status_type
> ++loongarch_push (int64_t val)
> ++{
> ++ if (LARCH_LD_STACK_DEPTH <= larch_stack_top)
> ++ return bfd_reloc_outofrange;
> ++ larch_opc_stack[larch_stack_top++] = val;
> ++ return bfd_reloc_ok;
> ++}
> ++
> ++static bfd_reloc_status_type
> ++loongarch_pop (int64_t *val)
> ++{
> ++ if (larch_stack_top == 0)
> ++ return bfd_reloc_outofrange;
> ++ BFD_ASSERT (val);
> ++ *val = larch_opc_stack[--larch_stack_top];
> ++ return bfd_reloc_ok;
> ++}
> ++
> ++static bfd_reloc_status_type
> ++loongarch_top (int64_t *val)
> ++{
> ++ if (larch_stack_top == 0)
> ++ return bfd_reloc_outofrange;
> ++ BFD_ASSERT (val);
> ++ *val = larch_opc_stack[larch_stack_top - 1];
> ++ return bfd_reloc_ok;
> ++}
> ++
> ++static void
> ++loongarch_elf_append_rela (bfd *abfd, asection *s, Elf_Internal_Rela
> *rel)
> ++{
> ++ BFD_ASSERT (s && s->contents);
> ++ const struct elf_backend_data *bed;
> ++ bfd_byte *loc;
> ++
> ++ bed = get_elf_backend_data (abfd);
> ++ if (!(s->size > s->reloc_count * bed->s->sizeof_rela))
> ++ BFD_ASSERT (s->size > s->reloc_count * bed->s->sizeof_rela);
> ++ loc = s->contents + (s->reloc_count++ * bed->s->sizeof_rela);
> ++ bed->s->swap_reloca_out (abfd, rel, loc);
> ++}
> ++
> ++/* Check rel->r_offset in range of contents. */
> ++static bfd_reloc_status_type
> ++loongarch_check_offset (const Elf_Internal_Rela *rel,
> ++ const asection *input_section)
> ++{
> ++ if (0 == strcmp(input_section->name, ".text")
> ++ && rel->r_offset > input_section->size)
> ++ return bfd_reloc_overflow;
> ++
> ++ return bfd_reloc_ok;
> ++}
> ++
> ++#define LARCH_RELOC_PERFORM_3OP(op1, op2, op3) \
> ++ ({ \
> ++ bfd_reloc_status_type ret = loongarch_pop (&op2); \
> ++ if (ret == bfd_reloc_ok) \
> ++ { \
> ++ ret = loongarch_pop (&op1); \
> ++ if (ret == bfd_reloc_ok) \
> ++ ret = loongarch_push (op3); \
> ++ } \
> ++ ret; \
> ++ })
> ++
> ++static bfd_reloc_status_type
> ++loongarch_reloc_rewrite_imm_insn (const Elf_Internal_Rela *rel,
> ++ const asection *input_section
> ATTRIBUTE_UNUSED,
> ++ reloc_howto_type *howto, bfd *input_bfd,
> ++ bfd_byte *contents, bfd_vma reloc_val)
> ++{
> ++ int bits = bfd_get_reloc_size (howto) * 8;
> ++ uint32_t insn = bfd_get (bits, input_bfd, contents + rel->r_offset);
> ++
> ++ if (!loongarch_adjust_reloc_bitsfield(howto, &reloc_val))
> ++ return bfd_reloc_overflow;
> ++
> ++ insn = (insn & (uint32_t)howto->src_mask)
> ++ | ((insn & (~(uint32_t)howto->dst_mask)) | reloc_val);
> ++
> ++ bfd_put (bits, input_bfd, insn, contents + rel->r_offset);
> ++
> ++ return bfd_reloc_ok;
> ++}
> ++
> ++static bfd_reloc_status_type
> ++perform_relocation (const Elf_Internal_Rela *rel, asection
> *input_section,
> ++ reloc_howto_type *howto, bfd_vma value,
> ++ bfd *input_bfd, bfd_byte *contents)
> ++{
> ++ int64_t opr1, opr2, opr3;
> ++ bfd_reloc_status_type r = bfd_reloc_ok;
> ++ int bits = bfd_get_reloc_size (howto) * 8;
> ++
> ++ switch (ELFNN_R_TYPE (rel->r_info))
> ++ {
> ++ case R_LARCH_SOP_PUSH_PCREL:
> ++ case R_LARCH_SOP_PUSH_ABSOLUTE:
> ++ case R_LARCH_SOP_PUSH_GPREL:
> ++ case R_LARCH_SOP_PUSH_TLS_TPREL:
> ++ case R_LARCH_SOP_PUSH_TLS_GOT:
> ++ case R_LARCH_SOP_PUSH_TLS_GD:
> ++ case R_LARCH_SOP_PUSH_PLT_PCREL:
> ++ r = loongarch_push (value);
> ++ break;
> ++
> ++ case R_LARCH_SOP_PUSH_DUP:
> ++ r = loongarch_pop (&opr1);
> ++ if (r == bfd_reloc_ok)
> ++ {
> ++ r = loongarch_push (opr1);
> ++ if (r == bfd_reloc_ok)
> ++ r = loongarch_push (opr1);
> ++ }
> ++ break;
> ++
> ++ case R_LARCH_SOP_ASSERT:
> ++ r = loongarch_pop (&opr1);
> ++ if (r != bfd_reloc_ok || !opr1)
> ++ r = bfd_reloc_notsupported;
> ++ break;
> ++
> ++ case R_LARCH_SOP_NOT:
> ++ r = loongarch_pop (&opr1);
> ++ if (r == bfd_reloc_ok)
> ++ r = loongarch_push (!opr1);
> ++ break;
> ++
> ++ case R_LARCH_SOP_SUB:
> ++ r = LARCH_RELOC_PERFORM_3OP (opr1, opr2, opr1 - opr2);
> ++ break;
> ++
> ++ case R_LARCH_SOP_SL:
> ++ r = LARCH_RELOC_PERFORM_3OP (opr1, opr2, opr1 << opr2);
> ++ break;
> ++
> ++ case R_LARCH_SOP_SR:
> ++ r = LARCH_RELOC_PERFORM_3OP (opr1, opr2, opr1 >> opr2);
> ++ break;
> ++
> ++ case R_LARCH_SOP_AND:
> ++ r = LARCH_RELOC_PERFORM_3OP (opr1, opr2, opr1 & opr2);
> ++ break;
> ++
> ++ case R_LARCH_SOP_ADD:
> ++ r = LARCH_RELOC_PERFORM_3OP (opr1, opr2, opr1 + opr2);
> ++ break;
> ++
> ++ case R_LARCH_SOP_IF_ELSE:
> ++ r = loongarch_pop (&opr3);
> ++ if (r == bfd_reloc_ok)
> ++ {
> ++ r = loongarch_pop (&opr2);
> ++ if (r == bfd_reloc_ok)
> ++ {
> ++ r = loongarch_pop (&opr1);
> ++ if (r == bfd_reloc_ok)
> ++ r = loongarch_push (opr1 ? opr2 : opr3);
> ++ }
> ++ }
> ++ break;
> ++
> ++ case R_LARCH_SOP_POP_32_S_10_5:
> ++ case R_LARCH_SOP_POP_32_S_10_12:
> ++ case R_LARCH_SOP_POP_32_S_10_16:
> ++ case R_LARCH_SOP_POP_32_S_10_16_S2:
> ++ case R_LARCH_SOP_POP_32_S_0_5_10_16_S2:
> ++ case R_LARCH_SOP_POP_32_S_0_10_10_16_S2:
> ++ case R_LARCH_SOP_POP_32_S_5_20:
> ++ case R_LARCH_SOP_POP_32_U_10_12:
> ++ case R_LARCH_SOP_POP_32_U:
> ++ r = loongarch_pop (&opr1);
> ++ if (r != bfd_reloc_ok)
> ++ break;
> ++ r = loongarch_check_offset (rel, input_section);
> ++ if (r != bfd_reloc_ok)
> ++ break;
> ++
> ++ r = loongarch_reloc_rewrite_imm_insn (rel, input_section,
> ++ howto, input_bfd,
> ++ contents, (bfd_vma)opr1);
> ++ break;
> ++
> ++ case R_LARCH_TLS_DTPREL32:
> ++ case R_LARCH_32:
> ++ case R_LARCH_TLS_DTPREL64:
> ++ case R_LARCH_64:
> ++ r = loongarch_check_offset (rel, input_section);
> ++ if (r != bfd_reloc_ok)
> ++ break;
> ++
> ++ bfd_put (bits, input_bfd, value, contents + rel->r_offset);
> ++ break;
> ++
> ++ case R_LARCH_ADD8:
> ++ case R_LARCH_ADD16:
> ++ case R_LARCH_ADD24:
> ++ case R_LARCH_ADD32:
> ++ case R_LARCH_ADD64:
> ++ r = loongarch_check_offset (rel, input_section);
> ++ if (r != bfd_reloc_ok)
> ++ break;
> ++
> ++ opr1 = bfd_get (bits, input_bfd, contents + rel->r_offset);
> ++ bfd_put (bits, input_bfd, opr1 + value, contents + rel->r_offset);
> ++ break;
> ++
> ++ case R_LARCH_SUB8:
> ++ case R_LARCH_SUB16:
> ++ case R_LARCH_SUB24:
> ++ case R_LARCH_SUB32:
> ++ case R_LARCH_SUB64:
> ++ r = loongarch_check_offset (rel, input_section);
> ++ if (r != bfd_reloc_ok)
> ++ break;
> ++
> ++ opr1 = bfd_get (bits, input_bfd, contents + rel->r_offset);
> ++ bfd_put (bits, input_bfd, opr1 - value, contents + rel->r_offset);
> ++ break;
> ++
> ++ /* For eh_frame and debug info. */
> ++ case R_LARCH_32_PCREL:
> ++ value -= sec_addr (input_section) + rel->r_offset;
> ++ value += rel->r_addend;
> ++ bfd_vma word = bfd_get (howto->bitsize, input_bfd,
> ++ contents + rel->r_offset);
> ++ word = (word & ~howto->dst_mask) | (value & howto->dst_mask);
> ++ bfd_put (howto->bitsize, input_bfd, word, contents +
> rel->r_offset);
> ++ r = bfd_reloc_ok;
> ++ break;
> ++
> ++ /* New reloc type.
> ++ R_LARCH_B16 ~ R_LARCH_TLS_GD_HI20. */
> ++ case R_LARCH_B16:
> ++ case R_LARCH_B21:
> ++ case R_LARCH_B26:
> ++ case R_LARCH_ABS_HI20:
> ++ case R_LARCH_ABS_LO12:
> ++ case R_LARCH_ABS64_LO20:
> ++ case R_LARCH_ABS64_HI12:
> ++ case R_LARCH_PCALA_HI20:
> ++ case R_LARCH_PCALA_LO12:
> ++ case R_LARCH_PCALA64_LO20:
> ++ case R_LARCH_PCALA64_HI12:
> ++ case R_LARCH_GOT_PC_HI20:
> ++ case R_LARCH_GOT_PC_LO12:
> ++ case R_LARCH_GOT64_PC_LO20:
> ++ case R_LARCH_GOT64_PC_HI12:
> ++ case R_LARCH_GOT_HI20:
> ++ case R_LARCH_GOT_LO12:
> ++ case R_LARCH_GOT64_LO20:
> ++ case R_LARCH_GOT64_HI12:
> ++ case R_LARCH_TLS_LE_HI20:
> ++ case R_LARCH_TLS_LE_LO12:
> ++ case R_LARCH_TLS_LE64_LO20:
> ++ case R_LARCH_TLS_LE64_HI12:
> ++ case R_LARCH_TLS_IE_PC_HI20:
> ++ case R_LARCH_TLS_IE_PC_LO12:
> ++ case R_LARCH_TLS_IE64_PC_LO20:
> ++ case R_LARCH_TLS_IE64_PC_HI12:
> ++ case R_LARCH_TLS_IE_HI20:
> ++ case R_LARCH_TLS_IE_LO12:
> ++ case R_LARCH_TLS_IE64_LO20:
> ++ case R_LARCH_TLS_IE64_HI12:
> ++ case R_LARCH_TLS_LD_PC_HI20:
> ++ case R_LARCH_TLS_LD_HI20:
> ++ case R_LARCH_TLS_GD_PC_HI20:
> ++ case R_LARCH_TLS_GD_HI20:
> ++ r = loongarch_check_offset (rel, input_section);
> ++ if (r != bfd_reloc_ok)
> ++ break;
> ++
> ++ r = loongarch_reloc_rewrite_imm_insn (rel, input_section,
> ++ howto, input_bfd,
> ++ contents, value);
> ++ break;
> ++
> ++ case R_LARCH_RELAX:
> ++ break;
> ++
> ++ default:
> ++ r = bfd_reloc_notsupported;
> ++ }
> ++ return r;
> ++}
> ++
> ++#define LARCH_RECENT_RELOC_QUEUE_LENGTH 72
> ++static struct
> ++{
> ++ bfd *bfd;
> ++ asection *section;
> ++ bfd_vma r_offset;
> ++ int r_type;
> ++ bfd_vma relocation;
> ++ Elf_Internal_Sym *sym;
> ++ struct elf_link_hash_entry *h;
> ++ bfd_vma addend;
> ++ int64_t top_then;
> ++} larch_reloc_queue[LARCH_RECENT_RELOC_QUEUE_LENGTH];
> ++static size_t larch_reloc_queue_head = 0;
> ++static size_t larch_reloc_queue_tail = 0;
> ++
> ++static const char *
> ++loongarch_sym_name (bfd *input_bfd, struct elf_link_hash_entry *h,
> ++ Elf_Internal_Sym *sym)
> ++{
> ++ const char *ret = NULL;
> ++ if (sym)
> ++ ret = bfd_elf_string_from_elf_section (input_bfd,
> ++ elf_symtab_hdr
> (input_bfd).sh_link,
> ++ sym->st_name);
> ++ else if (h)
> ++ ret = h->root.root.string;
> ++
> ++ if (ret == NULL || *ret == '\0')
> ++ ret = "<nameless>";
> ++ return ret;
> ++}
> ++
> ++static void
> ++loongarch_record_one_reloc (bfd *abfd, asection *section, int r_type,
> ++ bfd_vma r_offset, Elf_Internal_Sym *sym,
> ++ struct elf_link_hash_entry *h, bfd_vma addend)
> ++{
> ++ if ((larch_reloc_queue_head == 0
> ++ && larch_reloc_queue_tail == LARCH_RECENT_RELOC_QUEUE_LENGTH - 1)
> ++ || larch_reloc_queue_head == larch_reloc_queue_tail + 1)
> ++ larch_reloc_queue_head =
> ++ (larch_reloc_queue_head + 1) % LARCH_RECENT_RELOC_QUEUE_LENGTH;
> ++ larch_reloc_queue[larch_reloc_queue_tail].bfd = abfd;
> ++ larch_reloc_queue[larch_reloc_queue_tail].section = section;
> ++ larch_reloc_queue[larch_reloc_queue_tail].r_offset = r_offset;
> ++ larch_reloc_queue[larch_reloc_queue_tail].r_type = r_type;
> ++ larch_reloc_queue[larch_reloc_queue_tail].sym = sym;
> ++ larch_reloc_queue[larch_reloc_queue_tail].h = h;
> ++ larch_reloc_queue[larch_reloc_queue_tail].addend = addend;
> ++ loongarch_top (&larch_reloc_queue[larch_reloc_queue_tail].top_then);
> ++ larch_reloc_queue_tail =
> ++ (larch_reloc_queue_tail + 1) % LARCH_RECENT_RELOC_QUEUE_LENGTH;
> ++}
> ++
> ++static void
> ++loongarch_dump_reloc_record (void (*p) (const char *fmt, ...))
> ++{
> ++ size_t i = larch_reloc_queue_head;
> ++ bfd *a_bfd = NULL;
> ++ asection *section = NULL;
> ++ bfd_vma r_offset = 0;
> ++ int inited = 0;
> ++ p ("Dump relocate record:\n");
> ++ p ("stack top\t\trelocation name\t\tsymbol");
> ++ while (i != larch_reloc_queue_tail)
> ++ {
> ++ if (a_bfd != larch_reloc_queue[i].bfd
> ++ || section != larch_reloc_queue[i].section
> ++ || r_offset != larch_reloc_queue[i].r_offset)
> ++ {
> ++ a_bfd = larch_reloc_queue[i].bfd;
> ++ section = larch_reloc_queue[i].section;
> ++ r_offset = larch_reloc_queue[i].r_offset;
> ++ p ("\nat %pB(%pA+0x%v):\n", larch_reloc_queue[i].bfd,
> ++ larch_reloc_queue[i].section, larch_reloc_queue[i].r_offset);
> ++ }
> ++
> ++ if (!inited)
> ++ inited = 1, p ("...\n");
> ++
> ++ reloc_howto_type *howto =
> ++ loongarch_elf_rtype_to_howto (larch_reloc_queue[i].bfd,
> ++ larch_reloc_queue[i].r_type);
> ++ p ("0x%V %s\t`%s'", (bfd_vma) larch_reloc_queue[i].top_then,
> ++ howto ? howto->name : "<unknown reloc>",
> ++ loongarch_sym_name (larch_reloc_queue[i].bfd,
> larch_reloc_queue[i].h,
> ++ larch_reloc_queue[i].sym));
> ++
> ++ long addend = larch_reloc_queue[i].addend;
> ++ if (addend < 0)
> ++ p (" - %ld", -addend);
> ++ else if (0 < addend)
> ++ p (" + %ld(0x%v)", addend, larch_reloc_queue[i].addend);
> ++
> ++ p ("\n");
> ++ i = (i + 1) % LARCH_RECENT_RELOC_QUEUE_LENGTH;
> ++ }
> ++ p ("\n"
> ++ "-- Record dump end --\n\n");
> ++}
> ++
> ++static bfd_boolean
> ++loongarch_reloc_is_fatal (struct bfd_link_info *info,
> ++ bfd *input_bfd,
> ++ asection *input_section,
> ++ Elf_Internal_Rela *rel,
> ++ reloc_howto_type *howto,
> ++ bfd_reloc_status_type rtype,
> ++ bfd_boolean is_undefweak,
> ++ const char *name,
> ++ const char *msg)
> ++{
> ++ bfd_boolean fatal = TRUE;
> ++ switch (rtype)
> ++ {
> ++ /* 'dangerous' means we do it but can't promise it's ok
> ++ 'unsupport' means out of ability of relocation type
> ++ 'undefined' means we can't deal with the undefined symbol. */
> ++ case bfd_reloc_undefined:
> ++ info->callbacks->undefined_symbol (info, name, input_bfd,
> input_section,
> ++ rel->r_offset, TRUE);
> ++ info->callbacks->info ("%X%pB(%pA+0x%v): error: %s against
> %s`%s':\n%s\n",
> ++ input_bfd, input_section, rel->r_offset,
> ++ howto->name,
> ++ is_undefweak ? "[undefweak] " : "", name,
> msg);
> ++ break;
> ++ case bfd_reloc_dangerous:
> ++ info->callbacks->info ("%pB(%pA+0x%v): warning: %s against
> %s`%s':\n%s\n",
> ++ input_bfd, input_section, rel->r_offset,
> ++ howto->name,
> ++ is_undefweak ? "[undefweak] " : "", name,
> msg);
> ++ fatal = FALSE;
> ++ break;
> ++ case bfd_reloc_notsupported:
> ++ info->callbacks->info ("%X%pB(%pA+0x%v): error: %s against
> %s`%s':\n%s\n",
> ++ input_bfd, input_section, rel->r_offset,
> ++ howto->name,
> ++ is_undefweak ? "[undefweak] " : "", name,
> msg);
> ++ break;
> ++ default:
> ++ break;
> ++ }
> ++ return fatal;
> ++}
> ++
> ++#define RELOCATE_CALC_PC32_HI20(relocation, pc) \
> ++ ({ \
> ++ bfd_vma lo = (relocation) & ((bfd_vma)0xfff); \
> ++ pc = pc & (~(bfd_vma)0xfff); \
> ++ if (lo > 0x7ff) \
> ++ { \
> ++ relocation += 0x1000; \
> ++ } \
> ++ relocation &= ~(bfd_vma)0xfff; \
> ++ relocation -= pc; \
> ++ })
> ++
> ++#define RELOCATE_CALC_PC64_HI32(relocation, pc) \
> ++ ({ \
> ++ bfd_vma lo = (relocation) & ((bfd_vma)0xfff); \
> ++ if (lo > 0x7ff) \
> ++ { \
> ++ relocation -= 0x100000000; \
> ++ } \
> ++ relocation -= (pc & ~(bfd_vma)0xffffffff); \
> ++ })
> ++
> ++static int
> ++loongarch_elf_relocate_section (bfd *output_bfd, struct bfd_link_info
> *info,
> ++ bfd *input_bfd, asection *input_section,
> ++ bfd_byte *contents, Elf_Internal_Rela
> *relocs,
> ++ Elf_Internal_Sym *local_syms,
> ++ asection **local_sections)
> ++{
> ++ Elf_Internal_Rela *rel;
> ++ Elf_Internal_Rela *relend;
> ++ bfd_boolean fatal = FALSE;
> ++ asection *sreloc = elf_section_data (input_section)->sreloc;
> ++ struct loongarch_elf_link_hash_table *htab = loongarch_elf_hash_table
> (info);
> ++ Elf_Internal_Shdr *symtab_hdr = &elf_symtab_hdr (input_bfd);
> ++ struct elf_link_hash_entry **sym_hashes = elf_sym_hashes (input_bfd);
> ++ bfd_vma *local_got_offsets = elf_local_got_offsets (input_bfd);
> ++ bfd_boolean is_pic = bfd_link_pic (info);
> ++ bfd_boolean is_dyn = elf_hash_table (info)->dynamic_sections_created;
> ++ asection *plt = htab->elf.splt ? htab->elf.splt : htab->elf.iplt;
> ++ asection *got = htab->elf.sgot;
> ++
> ++ relend = relocs + input_section->reloc_count;
> ++ for (rel = relocs; rel < relend; rel++)
> ++ {
> ++ int r_type = ELFNN_R_TYPE (rel->r_info);
> ++ unsigned long r_symndx = ELFNN_R_SYM (rel->r_info);
> ++ bfd_vma pc = sec_addr (input_section) + rel->r_offset;
> ++ reloc_howto_type *howto = NULL;
> ++ asection *sec = NULL;
> ++ Elf_Internal_Sym *sym = NULL;
> ++ struct elf_link_hash_entry *h = NULL;
> ++ const char *name;
> ++ bfd_reloc_status_type r = bfd_reloc_ok;
> ++ bfd_boolean is_ie, is_undefweak, unresolved_reloc, defined_local;
> ++ bfd_boolean resolved_local, resolved_dynly, resolved_to_const;
> ++ char tls_type;
> ++ bfd_vma relocation, off, ie_off;
> ++ int i, j;
> ++
> ++ howto = loongarch_elf_rtype_to_howto (input_bfd, r_type);
> ++ if (howto == NULL || r_type == R_LARCH_GNU_VTINHERIT
> ++ || r_type == R_LARCH_GNU_VTENTRY)
> ++ continue;
> ++
> ++ /* This is a final link. */
> ++ if (r_symndx < symtab_hdr->sh_info)
> ++ {
> ++ is_undefweak = FALSE;
> ++ unresolved_reloc = FALSE;
> ++ sym = local_syms + r_symndx;
> ++ sec = local_sections[r_symndx];
> ++ relocation = _bfd_elf_rela_local_sym (output_bfd, sym, &sec,
> rel);
> ++
> ++ /* Relocate against local STT_GNU_IFUNC symbol. */
> ++ if (!bfd_link_relocatable (info)
> ++ && ELF_ST_TYPE (sym->st_info) == STT_GNU_IFUNC)
> ++ {
> ++ h = elfNN_loongarch_get_local_sym_hash (htab, input_bfd, rel,
> ++ FALSE);
> ++ if (h == NULL)
> ++ abort ();
> ++
> ++ /* Set STT_GNU_IFUNC symbol value. */
> ++ h->root.u.def.value = sym->st_value;
> ++ h->root.u.def.section = sec;
> ++ }
> ++ defined_local = TRUE;
> ++ resolved_local = TRUE;
> ++ resolved_dynly = FALSE;
> ++ resolved_to_const = FALSE;
> ++
> ++ /* Calc in funtion elf_link_input_bfd,
> ++ * if #define elf_backend_rela_normal to 1. */
> ++ if (bfd_link_relocatable (info)
> ++ && ELF_ST_TYPE (sym->st_info) == STT_SECTION)
> ++ continue;
> ++ }
> ++ else
> ++ {
> ++ bfd_boolean warned, ignored;
> ++
> ++ RELOC_FOR_GLOBAL_SYMBOL (info, input_bfd, input_section, rel,
> ++ r_symndx, symtab_hdr, sym_hashes,
> ++ h, sec, relocation,
> ++ unresolved_reloc, warned, ignored);
> ++ /* Here means symbol isn't local symbol only and 'h != NULL'. */
> ++
> ++ /* The 'unresolved_syms_in_objects' specify how to deal with
> undefined
> ++ symbol. And 'dynamic_undefined_weak' specify what to do when
> ++ meeting undefweak. */
> ++
> ++ if ((is_undefweak = h->root.type == bfd_link_hash_undefweak))
> ++ {
> ++ defined_local = FALSE;
> ++ resolved_local = FALSE;
> ++ resolved_to_const = (!is_dyn || h->dynindx == -1
> ++ || UNDEFWEAK_NO_DYNAMIC_RELOC (info,
> h));
> ++ resolved_dynly = !resolved_local && !resolved_to_const;
> ++ }
> ++ else if (warned)
> ++ {
> ++ /* Symbol undefined offen means failed already. I don't
> know why
> ++ 'warned' here but I guess it want to continue relocating
> as if
> ++ no error occures to find other errors as more as
> possible. */
> ++
> ++ /* To avoid generating warning messages about truncated
> ++ relocations, set the relocation's address to be the same
> as
> ++ the start of this section. */
> ++ relocation = (input_section->output_section
> ++ ? input_section->output_section->vma
> ++ : 0);
> ++
> ++ defined_local = relocation != 0;
> ++ resolved_local = defined_local;
> ++ resolved_to_const = !resolved_local;
> ++ resolved_dynly = FALSE;
> ++ }
> ++ else
> ++ {
> ++ defined_local = !unresolved_reloc && !ignored;
> ++ resolved_local =
> ++ defined_local && SYMBOL_REFERENCES_LOCAL (info, h);
> ++ resolved_dynly = !resolved_local;
> ++ resolved_to_const = !resolved_local && !resolved_dynly;
> ++ }
> ++ }
> ++
> ++ name = loongarch_sym_name (input_bfd, h, sym);
> ++
> ++ if (sec != NULL && discarded_section (sec))
> ++ RELOC_AGAINST_DISCARDED_SECTION (info, input_bfd, input_section,
> rel,
> ++ 1, relend, howto, 0, contents);
> ++
> ++ if (bfd_link_relocatable (info))
> ++ continue;
> ++
> ++ /* The r_symndx will be STN_UNDEF (zero) only for relocs against
> symbols
> ++ from removed linkonce sections, or sections discarded by a linker
> ++ script. Also for R_*_SOP_PUSH_ABSOLUTE and PCREL to specify
> const. */
> ++ if (r_symndx == STN_UNDEF || bfd_is_abs_section (sec))
> ++ {
> ++ defined_local = FALSE;
> ++ resolved_local = FALSE;
> ++ resolved_dynly = FALSE;
> ++ resolved_to_const = TRUE;
> ++ }
> ++
> ++ /* The ifunc reference generate plt. */
> ++ if (h && h->type == STT_GNU_IFUNC && h->plt.offset != MINUS_ONE)
> ++ {
> ++ defined_local = TRUE;
> ++ resolved_local = TRUE;
> ++ resolved_dynly = FALSE;
> ++ resolved_to_const = FALSE;
> ++ relocation = sec_addr (plt) + h->plt.offset;
> ++ }
> ++
> ++ unresolved_reloc = resolved_dynly;
> ++
> ++ BFD_ASSERT (resolved_local + resolved_dynly + resolved_to_const ==
> 1);
> ++
> ++ /* BFD_ASSERT (!resolved_dynly || (h && h->dynindx != -1));. */
> ++
> ++ BFD_ASSERT (!resolved_local || defined_local);
> ++
> ++ is_ie = FALSE;
> ++ switch (r_type)
> ++ {
> ++ case R_LARCH_MARK_PCREL:
> ++ case R_LARCH_MARK_LA:
> ++ case R_LARCH_NONE:
> ++ r = bfd_reloc_continue;
> ++ unresolved_reloc = FALSE;
> ++ break;
> ++
> ++ case R_LARCH_32:
> ++ case R_LARCH_64:
> ++ if (resolved_dynly || (is_pic && resolved_local))
> ++ {
> ++ Elf_Internal_Rela outrel;
> ++
> ++ /* When generating a shared object, these relocations are
> copied
> ++ into the output file to be resolved at run time. */
> ++
> ++ outrel.r_offset = _bfd_elf_section_offset (output_bfd, info,
> ++ input_section,
> ++ rel->r_offset);
> ++
> ++ unresolved_reloc = (!((bfd_vma) -2 <= outrel.r_offset)
> ++ && (input_section->flags & SEC_ALLOC));
> ++
> ++ outrel.r_offset += sec_addr (input_section);
> ++
> ++ /* A pointer point to a ifunc symbol. */
> ++ if (h && h->type == STT_GNU_IFUNC)
> ++ {
> ++ if (h->dynindx == -1)
> ++ {
> ++ outrel.r_info = ELFNN_R_INFO (0, R_LARCH_IRELATIVE);
> ++ outrel.r_addend = (h->root.u.def.value
> ++ +
> h->root.u.def.section->output_section->vma
> ++ + h->root.u.def.section->output_offset);
> ++ }
> ++ else
> ++ {
> ++ outrel.r_info = ELFNN_R_INFO (h->dynindx,
> R_LARCH_NN);
> ++ outrel.r_addend = 0;
> ++ }
> ++
> ++ if (SYMBOL_REFERENCES_LOCAL (info, h))
> ++ {
> ++
> ++ if (htab->elf.splt != NULL)
> ++ sreloc = htab->elf.srelgot;
> ++ else
> ++ sreloc = htab->elf.irelplt;
> ++ }
> ++ else
> ++ {
> ++
> ++ if (bfd_link_pic (info))
> ++ sreloc = htab->elf.irelifunc;
> ++ else if (htab->elf.splt != NULL)
> ++ sreloc = htab->elf.srelgot;
> ++ else
> ++ sreloc = htab->elf.irelplt;
> ++ }
> ++ }
> ++ else if (resolved_dynly)
> ++ {
> ++ if (h->dynindx == -1)
> ++ {
> ++ if (h->root.type == bfd_link_hash_undefined)
> ++ (*info->callbacks->undefined_symbol)
> ++ (info, name, input_bfd, input_section,
> ++ rel->r_offset, TRUE);
> ++
> ++ outrel.r_info = ELFNN_R_INFO (0, r_type);
> ++ }
> ++ else
> ++ outrel.r_info = ELFNN_R_INFO (h->dynindx, r_type);
> ++
> ++ outrel.r_addend = rel->r_addend;
> ++ }
> ++ else
> ++ {
> ++ outrel.r_info = ELFNN_R_INFO (0, R_LARCH_RELATIVE);
> ++ outrel.r_addend = relocation + rel->r_addend;
> ++ }
> ++
> ++ /* No alloc space of func allocate_dynrelocs. */
> ++ if (unresolved_reloc
> ++ && !(h && (h->is_weakalias || !h->dyn_relocs)))
> ++ loongarch_elf_append_rela (output_bfd, sreloc, &outrel);
> ++ }
> ++
> ++ relocation += rel->r_addend;
> ++ break;
> ++
> ++ case R_LARCH_ADD8:
> ++ case R_LARCH_ADD16:
> ++ case R_LARCH_ADD24:
> ++ case R_LARCH_ADD32:
> ++ case R_LARCH_ADD64:
> ++ case R_LARCH_SUB8:
> ++ case R_LARCH_SUB16:
> ++ case R_LARCH_SUB24:
> ++ case R_LARCH_SUB32:
> ++ case R_LARCH_SUB64:
> ++ if (resolved_dynly)
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_undefined, is_undefweak, name,
> ++ "Can't be resolved dynamically. "
> ++ "If this procedure is hand-written assembly,\n"
> ++ "there must be something like '.dword sym1 - sym2' "
> ++ "to generate these relocs\n"
> ++ "and we can't get known link-time address of "
> ++ "these symbols."));
> ++ else
> ++ relocation += rel->r_addend;
> ++ break;
> ++
> ++ case R_LARCH_TLS_DTPREL32:
> ++ case R_LARCH_TLS_DTPREL64:
> ++ if (resolved_dynly)
> ++ {
> ++ Elf_Internal_Rela outrel;
> ++
> ++ outrel.r_offset = _bfd_elf_section_offset (output_bfd, info,
> ++ input_section,
> ++ rel->r_offset);
> ++ unresolved_reloc = (!((bfd_vma) -2 <= outrel.r_offset)
> ++ && (input_section->flags & SEC_ALLOC));
> ++ outrel.r_info = ELFNN_R_INFO (h->dynindx, r_type);
> ++ outrel.r_offset += sec_addr (input_section);
> ++ outrel.r_addend = rel->r_addend;
> ++ if (unresolved_reloc)
> ++ loongarch_elf_append_rela (output_bfd, sreloc, &outrel);
> ++ break;
> ++ }
> ++
> ++ if (resolved_to_const)
> ++ fatal = loongarch_reloc_is_fatal (info, input_bfd,
> input_section,
> ++ rel, howto,
> ++ bfd_reloc_notsupported,
> ++ is_undefweak, name,
> ++ "Internal:");
> ++ if (resolved_local)
> ++ {
> ++ if (!elf_hash_table (info)->tls_sec)
> ++ {
> ++ fatal = loongarch_reloc_is_fatal (info, input_bfd,
> ++ input_section, rel, howto,
> bfd_reloc_notsupported,
> ++ is_undefweak, name, "TLS section not be
> created");
> ++ }
> ++ else
> ++ relocation -= elf_hash_table (info)->tls_sec->vma;
> ++ }
> ++ else
> ++ {
> ++ fatal = loongarch_reloc_is_fatal (info, input_bfd,
> ++ input_section, rel, howto, bfd_reloc_undefined,
> ++ is_undefweak, name,
> ++ "TLS LE just can be resolved local only.");
> ++ }
> ++
> ++ break;
> ++
> ++ case R_LARCH_SOP_PUSH_TLS_TPREL:
> ++ if (resolved_local)
> ++ {
> ++ if (!elf_hash_table (info)->tls_sec)
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_notsupported, is_undefweak, name,
> ++ "TLS section not be created"));
> ++ else
> ++ relocation -= elf_hash_table (info)->tls_sec->vma;
> ++ }
> ++ else
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_undefined, is_undefweak, name,
> ++ "TLS LE just can be resolved local only."));
> ++ break;
> ++
> ++ case R_LARCH_SOP_PUSH_ABSOLUTE:
> ++ if (is_undefweak)
> ++ {
> ++ if (resolved_dynly)
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_dangerous, is_undefweak, name,
> ++ "Someone require us to resolve undefweak "
> ++ "symbol dynamically. \n"
> ++ "But this reloc can't be done. "
> ++ "I think I can't throw error "
> ++ "for this\n"
> ++ "so I resolved it to 0. "
> ++ "I suggest to re-compile with '-fpic'."));
> ++
> ++ relocation = 0;
> ++ unresolved_reloc = FALSE;
> ++ break;
> ++ }
> ++
> ++ if (resolved_to_const)
> ++ {
> ++ relocation += rel->r_addend;
> ++ break;
> ++ }
> ++
> ++ if (is_pic)
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_notsupported, is_undefweak, name,
> ++ "Under PIC we don't know load address. Re-compile
> "
> ++ "with '-fpic'?"));
> ++ break;
> ++ }
> ++
> ++ if (resolved_dynly)
> ++ {
> ++ if (!(plt && h && h->plt.offset != MINUS_ONE))
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_undefined, is_undefweak, name,
> ++ "Can't be resolved dynamically. Try to
> re-compile "
> ++ "with '-fpic'?"));
> ++ break;
> ++ }
> ++
> ++ if (rel->r_addend != 0)
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_notsupported, is_undefweak, name,
> ++ "Shouldn't be with r_addend."));
> ++ break;
> ++ }
> ++
> ++ relocation = sec_addr (plt) + h->plt.offset;
> ++ unresolved_reloc = FALSE;
> ++ break;
> ++ }
> ++
> ++ if (resolved_local)
> ++ {
> ++ relocation += rel->r_addend;
> ++ break;
> ++ }
> ++
> ++ break;
> ++
> ++ case R_LARCH_SOP_PUSH_PCREL:
> ++ case R_LARCH_SOP_PUSH_PLT_PCREL:
> ++ unresolved_reloc = FALSE;
> ++
> ++ if (is_undefweak)
> ++ {
> ++ i = 0, j = 0;
> ++ relocation = 0;
> ++ if (resolved_dynly)
> ++ {
> ++ if (h && h->plt.offset != MINUS_ONE)
> ++ i = 1, j = 2;
> ++ else
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_dangerous, is_undefweak, name,
> ++ "Undefweak need to be resolved dynamically, "
> ++ "but PLT stub doesn't represent."));
> ++ }
> ++ }
> ++ else
> ++ {
> ++ if (!(defined_local || (h && h->plt.offset != MINUS_ONE)))
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_undefined, is_undefweak, name,
> ++ "PLT stub does not represent and "
> ++ "symbol not defined."));
> ++ break;
> ++ }
> ++
> ++ if (resolved_local)
> ++ i = 0, j = 2;
> ++ else /* if (resolved_dynly) */
> ++ {
> ++ if (!(h && h->plt.offset != MINUS_ONE))
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_dangerous, is_undefweak, name,
> ++ "Internal: PLT stub doesn't represent. "
> ++ "Resolve it with pcrel"));
> ++ i = 1, j = 3;
> ++ }
> ++ }
> ++
> ++ for (; i < j; i++)
> ++ {
> ++ if ((i & 1) == 0 && defined_local)
> ++ {
> ++ relocation -= pc;
> ++ relocation += rel->r_addend;
> ++ break;
> ++ }
> ++
> ++ if ((i & 1) && h && h->plt.offset != MINUS_ONE)
> ++ {
> ++ if (rel->r_addend != 0)
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_notsupported, is_undefweak, name,
> ++ "PLT shouldn't be with r_addend."));
> ++ break;
> ++ }
> ++ relocation = sec_addr (plt) + h->plt.offset - pc;
> ++ break;
> ++ }
> ++ }
> ++ break;
> ++
> ++ case R_LARCH_SOP_PUSH_GPREL:
> ++ unresolved_reloc = FALSE;
> ++
> ++ if (rel->r_addend != 0)
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_notsupported, is_undefweak, name,
> ++ "Shouldn't be with r_addend."));
> ++ break;
> ++ }
> ++
> ++ if (h != NULL)
> ++ {
> ++ off = h->got.offset & (~1);
> ++
> ++ if (h->got.offset == MINUS_ONE && h->type != STT_GNU_IFUNC)
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_notsupported, is_undefweak, name,
> ++ "Internal: GOT entry doesn't represent."));
> ++ break;
> ++ }
> ++
> ++ /* Hidden symbol not has .got entry, only .got.plt entry
> ++ so gprel is (plt - got). */
> ++ if (h->got.offset == MINUS_ONE && h->type == STT_GNU_IFUNC)
> ++ {
> ++ if (h->plt.offset == (bfd_vma) -1)
> ++ {
> ++ abort();
> ++ }
> ++
> ++ bfd_vma plt_index = h->plt.offset / PLT_ENTRY_SIZE;
> ++ off = plt_index * GOT_ENTRY_SIZE;
> ++
> ++ if (htab->elf.splt != NULL)
> ++ {
> ++ /* Section .plt header is 2 times of plt entry. */
> ++ off = sec_addr (htab->elf.sgotplt) + off
> ++ - sec_addr (htab->elf.sgot);
> ++ }
> ++ else
> ++ {
> ++ /* Section iplt not has plt header. */
> ++ off = sec_addr (htab->elf.igotplt) + off
> ++ - sec_addr (htab->elf.sgot);
> ++ }
> ++ }
> ++
> ++ if ((h->got.offset & 1) == 0)
> ++ {
> ++ if (!WILL_CALL_FINISH_DYNAMIC_SYMBOL (is_dyn,
> ++ bfd_link_pic
> (info), h)
> ++ && ((bfd_link_pic (info)
> ++ && SYMBOL_REFERENCES_LOCAL (info, h))))
> ++ {
> ++ /* This is actually a static link, or it is a
> ++ -Bsymbolic link and the symbol is defined
> ++ locally, or the symbol was forced to be local
> ++ because of a version file. We must initialize
> ++ this entry in the global offset table. Since the
> ++ offset must always be a multiple of the word size,
> ++ we use the least significant bit to record whether
> ++ we have initialized it already.
> ++
> ++ When doing a dynamic link, we create a rela.got
> ++ relocation entry to initialize the value. This
> ++ is done in the finish_dynamic_symbol routine. */
> ++
> ++ if (resolved_dynly)
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel,
> howto,
> ++ bfd_reloc_dangerous, is_undefweak,
> name,
> ++ "Internal: here shouldn't dynamic."));
> ++ }
> ++
> ++ if (!(defined_local || resolved_to_const))
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel,
> howto,
> ++ bfd_reloc_undefined, is_undefweak,
> name,
> ++ "Internal: "));
> ++ break;
> ++ }
> ++
> ++ asection *s;
> ++ Elf_Internal_Rela outrel;
> ++ /* We need to generate a R_LARCH_RELATIVE reloc
> ++ for the dynamic linker. */
> ++ s = htab->elf.srelgot;
> ++ if (!s)
> ++ {
> ++ fatal = loongarch_reloc_is_fatal
> ++ (info, input_bfd,
> ++ input_section, rel, howto,
> ++ bfd_reloc_notsupported, is_undefweak, name,
> ++ "Internal: '.rel.got' not represent");
> ++ break;
> ++ }
> ++
> ++ outrel.r_offset = sec_addr (got) + off;
> ++ outrel.r_info = ELFNN_R_INFO (0, R_LARCH_RELATIVE);
> ++ outrel.r_addend = relocation; /* Link-time addr. */
> ++ loongarch_elf_append_rela (output_bfd, s, &outrel);
> ++ }
> ++ bfd_put_NN (output_bfd, relocation, got->contents + off);
> ++ h->got.offset |= 1;
> ++ }
> ++ }
> ++ else
> ++ {
> ++ if (!local_got_offsets)
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_notsupported, is_undefweak, name,
> ++ "Internal: local got offsets not
> reporesent."));
> ++ break;
> ++ }
> ++
> ++ off = local_got_offsets[r_symndx] & (~1);
> ++
> ++ if (local_got_offsets[r_symndx] == MINUS_ONE)
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_notsupported, is_undefweak, name,
> ++ "Internal: GOT entry doesn't represent."));
> ++ break;
> ++ }
> ++
> ++ /* The offset must always be a multiple of the word size.
> ++ So, we can use the least significant bit to record
> ++ whether we have already processed this entry. */
> ++ if (local_got_offsets[r_symndx] == 0)
> ++ {
> ++ if (is_pic)
> ++ {
> ++ asection *s;
> ++ Elf_Internal_Rela outrel;
> ++ /* We need to generate a R_LARCH_RELATIVE reloc
> ++ for the dynamic linker. */
> ++ s = htab->elf.srelgot;
> ++ if (!s)
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel,
> howto,
> ++ bfd_reloc_notsupported, is_undefweak,
> name,
> ++ "Internal: '.rel.got' not represent"));
> ++ break;
> ++ }
> ++
> ++ outrel.r_offset = sec_addr (got) + off;
> ++ outrel.r_info = ELFNN_R_INFO (0, R_LARCH_RELATIVE);
> ++ outrel.r_addend = relocation; /* Link-time addr. */
> ++ loongarch_elf_append_rela (output_bfd, s, &outrel);
> ++ }
> ++
> ++ bfd_put_NN (output_bfd, relocation, got->contents + off);
> ++ local_got_offsets[r_symndx] |= 1;
> ++ }
> ++ }
> ++ relocation = off;
> ++
> ++ break;
> ++
> ++ case R_LARCH_SOP_PUSH_TLS_GOT:
> ++ case R_LARCH_SOP_PUSH_TLS_GD:
> ++ {
> ++ unresolved_reloc = FALSE;
> ++ if (r_type == R_LARCH_SOP_PUSH_TLS_GOT)
> ++ is_ie = TRUE;
> ++
> ++ bfd_vma got_off = 0;
> ++ if (h != NULL)
> ++ {
> ++ got_off = h->got.offset;
> ++ h->got.offset |= 1;
> ++ }
> ++ else
> ++ {
> ++ got_off = local_got_offsets[r_symndx];
> ++ local_got_offsets[r_symndx] |= 1;
> ++ }
> ++
> ++ BFD_ASSERT (got_off != MINUS_ONE);
> ++
> ++ ie_off = 0;
> ++ tls_type = _bfd_loongarch_elf_tls_type (input_bfd, h,
> r_symndx);
> ++ if ((tls_type & GOT_TLS_GD) && (tls_type & GOT_TLS_IE))
> ++ ie_off = 2 * GOT_ENTRY_SIZE;
> ++
> ++ if ((got_off & 1) == 0)
> ++ {
> ++ Elf_Internal_Rela rela;
> ++ asection *srel = htab->elf.srelgot;
> ++ bfd_vma tls_block_off = 0;
> ++
> ++ if (SYMBOL_REFERENCES_LOCAL (info, h))
> ++ {
> ++ BFD_ASSERT (elf_hash_table (info)->tls_sec);
> ++ tls_block_off = relocation
> ++ - elf_hash_table (info)->tls_sec->vma;
> ++ }
> ++
> ++ if (tls_type & GOT_TLS_GD)
> ++ {
> ++ rela.r_offset = sec_addr (got) + got_off;
> ++ rela.r_addend = 0;
> ++ if (SYMBOL_REFERENCES_LOCAL (info, h))
> ++ {
> ++ /* Local sym, used in exec, set module id 1. */
> ++ if (bfd_link_executable (info))
> ++ bfd_put_NN (output_bfd, 1, got->contents +
> got_off);
> ++ else
> ++ {
> ++ rela.r_info = ELFNN_R_INFO (0,
> ++
> R_LARCH_TLS_DTPMODNN);
> ++ loongarch_elf_append_rela (output_bfd, srel,
> &rela);
> ++ }
> ++
> ++ bfd_put_NN (output_bfd, tls_block_off,
> ++ got->contents + got_off +
> GOT_ENTRY_SIZE);
> ++ }
> ++ /* Dynamic resolved. */
> ++ else
> ++ {
> ++ /* Dynamic relocate module id. */
> ++ rela.r_info = ELFNN_R_INFO (h->dynindx,
> ++ R_LARCH_TLS_DTPMODNN);
> ++ loongarch_elf_append_rela (output_bfd, srel,
> &rela);
> ++
> ++ /* Dynamic relocate offset of block. */
> ++ rela.r_offset += GOT_ENTRY_SIZE;
> ++ rela.r_info = ELFNN_R_INFO (h->dynindx,
> ++ R_LARCH_TLS_DTPRELNN);
> ++ loongarch_elf_append_rela (output_bfd, srel,
> &rela);
> ++ }
> ++ }
> ++ if (tls_type & GOT_TLS_IE)
> ++ {
> ++ rela.r_offset = sec_addr (got) + got_off + ie_off;
> ++ if (SYMBOL_REFERENCES_LOCAL (info, h))
> ++ {
> ++ /* Local sym, used in exec, set module id 1. */
> ++ if (!bfd_link_executable (info))
> ++ {
> ++ rela.r_info = ELFNN_R_INFO (0,
> R_LARCH_TLS_TPRELNN);
> ++ rela.r_addend = tls_block_off;
> ++ loongarch_elf_append_rela (output_bfd, srel,
> &rela);
> ++ }
> ++
> ++ bfd_put_NN (output_bfd, tls_block_off,
> ++ got->contents + got_off + ie_off);
> ++ }
> ++ /* Dynamic resolved. */
> ++ else
> ++ {
> ++ /* Dynamic relocate offset of block. */
> ++ rela.r_info = ELFNN_R_INFO (h->dynindx,
> ++ R_LARCH_TLS_TPRELNN);
> ++ rela.r_addend = 0;
> ++ loongarch_elf_append_rela (output_bfd, srel,
> &rela);
> ++ }
> ++ }
> ++ }
> ++
> ++ relocation = (got_off & (~(bfd_vma)1)) + (is_ie ? ie_off : 0);
> ++ }
> ++ break;
> ++
> ++ /* New reloc types. */
> ++ case R_LARCH_B21:
> ++ case R_LARCH_B26:
> ++ case R_LARCH_B16:
> ++ unresolved_reloc = FALSE;
> ++ if (is_undefweak)
> ++ {
> ++ relocation = 0;
> ++ }
> ++
> ++ if (resolved_local)
> ++ {
> ++ relocation -= pc;
> ++ relocation += rel->r_addend;
> ++ }
> ++ else if (resolved_dynly)
> ++ {
> ++ BFD_ASSERT (h
> ++ && (h->plt.offset != MINUS_ONE
> ++ || ELF_ST_VISIBILITY (h->other) !=
> STV_DEFAULT)
> ++ && rel->r_addend == 0);
> ++ if (h && h->plt.offset == MINUS_ONE
> ++ && ELF_ST_VISIBILITY (h->other) != STV_DEFAULT)
> ++ {
> ++ relocation -= pc;
> ++ relocation += rel->r_addend;
> ++ }
> ++ else
> ++ relocation = sec_addr (plt) + h->plt.offset - pc;
> ++ }
> ++
> ++ break;
> ++
> ++ case R_LARCH_ABS_HI20:
> ++ case R_LARCH_ABS_LO12:
> ++ case R_LARCH_ABS64_LO20:
> ++ case R_LARCH_ABS64_HI12:
> ++ BFD_ASSERT (!is_pic);
> ++
> ++ if (is_undefweak)
> ++ {
> ++ BFD_ASSERT (resolved_dynly);
> ++ relocation = 0;
> ++ break;
> ++ }
> ++ else if (resolved_to_const || resolved_local)
> ++ {
> ++ relocation += rel->r_addend;
> ++ }
> ++ else if (resolved_dynly)
> ++ {
> ++ unresolved_reloc = FALSE;
> ++ BFD_ASSERT ((plt && h && h->plt.offset != MINUS_ONE)
> ++ && rel->r_addend == 0);
> ++ relocation = sec_addr (plt) + h->plt.offset;
> ++ }
> ++
> ++ break;
> ++
> ++ case R_LARCH_PCALA_HI20:
> ++ unresolved_reloc = FALSE;
> ++ if (h && h->plt.offset != MINUS_ONE)
> ++ relocation = sec_addr (plt) + h->plt.offset;
> ++ else
> ++ relocation += rel->r_addend;
> ++
> ++ RELOCATE_CALC_PC32_HI20 (relocation, pc);
> ++
> ++ break;
> ++
> ++ case R_LARCH_PCALA_LO12:
> ++ /* Not support if sym_addr in 2k page edge.
> ++ pcalau12i pc_hi20 (sym_addr)
> ++ ld.w/d pc_lo12 (sym_addr)
> ++ ld.w/d pc_lo12 (sym_addr + x)
> ++ ...
> ++ can not calc correct address
> ++ if sym_addr < 0x800 && sym_addr + x >= 0x800. */
> ++
> ++ if (h && h->plt.offset != MINUS_ONE)
> ++ relocation = sec_addr (plt) + h->plt.offset;
> ++ else
> ++ relocation += rel->r_addend;
> ++
> ++ {
> ++ relocation &= 0xfff;
> ++ /* Signed extend. */
> ++ relocation = (relocation ^ 0x800) - 0x800;
> ++
> ++ /* For 2G jump, generate pcalau12i, jirl. */
> ++ /* If use jirl, turns to R_LARCH_B16. */
> ++ uint32_t insn = bfd_get (32, input_bfd, contents +
> rel->r_offset);
> ++ if ((insn & 0x4c000000) == 0x4c000000)
> ++ {
> ++ rel->r_info = ELFNN_R_INFO (r_symndx, R_LARCH_B16);
> ++ howto = loongarch_elf_rtype_to_howto (input_bfd,
> R_LARCH_B16);
> ++ }
> ++ }
> ++ break;
> ++
> ++ case R_LARCH_PCALA64_LO20:
> ++ case R_LARCH_PCALA64_HI12:
> ++ if (h && h->plt.offset != MINUS_ONE)
> ++ relocation = sec_addr (plt) + h->plt.offset;
> ++ else
> ++ relocation += rel->r_addend;
> ++
> ++ RELOCATE_CALC_PC64_HI32 (relocation, pc);
> ++
> ++ break;
> ++
> ++ case R_LARCH_GOT_PC_HI20:
> ++ case R_LARCH_GOT_HI20:
> ++ /* Calc got offset. */
> ++ {
> ++ unresolved_reloc = FALSE;
> ++ BFD_ASSERT (rel->r_addend == 0);
> ++
> ++ bfd_vma got_off = 0;
> ++ if (h != NULL)
> ++ {
> ++ /* GOT ref or ifunc. */
> ++ BFD_ASSERT (h->got.offset != MINUS_ONE
> ++ || h->type == STT_GNU_IFUNC);
> ++
> ++ got_off = h->got.offset & (~(bfd_vma)1);
> ++ /* Hidden symbol not has got entry,
> ++ * only got.plt entry so it is (plt - got). */
> ++ if (h->got.offset == MINUS_ONE && h->type ==
> STT_GNU_IFUNC)
> ++ {
> ++ bfd_vma idx;
> ++ if (htab->elf.splt != NULL)
> ++ {
> ++ idx = (h->plt.offset - PLT_HEADER_SIZE)
> ++ / PLT_ENTRY_SIZE;
> ++ got_off = sec_addr (htab->elf.sgotplt)
> ++ + GOTPLT_HEADER_SIZE
> ++ + (idx * GOT_ENTRY_SIZE)
> ++ - sec_addr (htab->elf.sgot);
> ++ }
> ++ else
> ++ {
> ++ idx = h->plt.offset / PLT_ENTRY_SIZE;
> ++ got_off = sec_addr (htab->elf.sgotplt)
> ++ + (idx * GOT_ENTRY_SIZE)
> ++ - sec_addr (htab->elf.sgot);
> ++ }
> ++ }
> ++
> ++ if ((h->got.offset & 1) == 0)
> ++ {
> ++ /* We need to generate a R_LARCH_RELATIVE reloc once
> ++ * in loongarch_elf_finish_dynamic_symbol or now,
> ++ * call finish_dyn && nopic
> ++ * or !call finish_dyn && pic. */
> ++ if (!WILL_CALL_FINISH_DYNAMIC_SYMBOL (is_dyn,
> ++ bfd_link_pic
> (info),
> ++ h)
> ++ && bfd_link_pic (info)
> ++ && SYMBOL_REFERENCES_LOCAL (info, h))
> ++ {
> ++ Elf_Internal_Rela rela;
> ++ rela.r_offset = sec_addr (got) + got_off;
> ++ rela.r_info = ELFNN_R_INFO (0, R_LARCH_RELATIVE);
> ++ rela.r_addend = relocation;
> ++ loongarch_elf_append_rela (output_bfd,
> ++ htab->elf.srelgot,
> &rela);
> ++ }
> ++ h->got.offset |= 1;
> ++ }
> ++ }
> ++ else
> ++ {
> ++ BFD_ASSERT (local_got_offsets
> ++ && local_got_offsets[r_symndx] != MINUS_ONE);
> ++
> ++ got_off = local_got_offsets[r_symndx] & (~(bfd_vma)1);
> ++ if ((local_got_offsets[r_symndx] & 1) == 0)
> ++ {
> ++ if (bfd_link_pic (info))
> ++ {
> ++ Elf_Internal_Rela rela;
> ++ rela.r_offset = sec_addr (got) + got_off;
> ++ rela.r_info = ELFNN_R_INFO (0, R_LARCH_RELATIVE);
> ++ rela.r_addend = relocation;
> ++ loongarch_elf_append_rela (output_bfd,
> ++ htab->elf.srelgot,
> &rela);
> ++ }
> ++ local_got_offsets[r_symndx] |= 1;
> ++ }
> ++ }
> ++
> ++ bfd_put_NN (output_bfd, relocation, got->contents + got_off);
> ++
> ++ relocation = got_off + sec_addr (got);
> ++ }
> ++
> ++ if (r_type == R_LARCH_GOT_PC_HI20)
> ++ RELOCATE_CALC_PC32_HI20 (relocation, pc);
> ++
> ++ break;
> ++
> ++ case R_LARCH_GOT_PC_LO12:
> ++ case R_LARCH_GOT64_PC_LO20:
> ++ case R_LARCH_GOT64_PC_HI12:
> ++ case R_LARCH_GOT_LO12:
> ++ case R_LARCH_GOT64_LO20:
> ++ case R_LARCH_GOT64_HI12:
> ++ {
> ++ unresolved_reloc = FALSE;
> ++ bfd_vma got_off;
> ++ if (h)
> ++ got_off = h->got.offset & (~(bfd_vma)1);
> ++ else
> ++ got_off = local_got_offsets[r_symndx] & (~(bfd_vma)1);
> ++
> ++ if (h && h->got.offset == MINUS_ONE && h->type ==
> STT_GNU_IFUNC)
> ++ {
> ++ bfd_vma idx;
> ++ if (htab->elf.splt != NULL)
> ++ idx = (h->plt.offset - PLT_HEADER_SIZE) /
> PLT_ENTRY_SIZE;
> ++ else
> ++ idx = h->plt.offset / PLT_ENTRY_SIZE;
> ++
> ++ got_off = sec_addr (htab->elf.sgotplt)
> ++ + GOTPLT_HEADER_SIZE
> ++ + (idx * GOT_ENTRY_SIZE)
> ++ - sec_addr (htab->elf.sgot);
> ++ }
> ++ relocation = got_off + sec_addr (got);
> ++ }
> ++
> ++ if (r_type == R_LARCH_GOT_PC_LO12)
> ++ relocation &= (bfd_vma)0xfff;
> ++ else if (r_type == R_LARCH_GOT64_PC_LO20
> ++ || r_type == R_LARCH_GOT64_PC_HI12)
> ++ RELOCATE_CALC_PC64_HI32 (relocation, pc);
> ++
> ++ break;
> ++
> ++ case R_LARCH_TLS_LE_HI20:
> ++ case R_LARCH_TLS_LE_LO12:
> ++ case R_LARCH_TLS_LE64_LO20:
> ++ case R_LARCH_TLS_LE64_HI12:
> ++ BFD_ASSERT (resolved_local && elf_hash_table (info)->tls_sec);
> ++
> ++ relocation -= elf_hash_table (info)->tls_sec->vma;
> ++ break;
> ++
> ++ /* TLS IE LD/GD process separately is troublesome.
> ++ When a symbol is both ie and LD/GD, h->got.off |= 1
> ++ make only one type be relocated. We must use
> ++ h->got.offset |= 1 and h->got.offset |= 2
> ++ diff IE and LD/GD. And all (got_off & (~(bfd_vma)1))
> ++ (IE LD/GD and reusable GOT reloc) must change to
> ++ (got_off & (~(bfd_vma)3)), beause we use lowest 2 bits
> ++ as a tag.
> ++ Now, LD and GD is both GOT_TLS_GD type, LD seems to
> ++ can be omitted. */
> ++ case R_LARCH_TLS_IE_PC_HI20:
> ++ case R_LARCH_TLS_IE_HI20:
> ++ case R_LARCH_TLS_LD_PC_HI20:
> ++ case R_LARCH_TLS_LD_HI20:
> ++ case R_LARCH_TLS_GD_PC_HI20:
> ++ case R_LARCH_TLS_GD_HI20:
> ++ BFD_ASSERT (rel->r_addend == 0);
> ++ unresolved_reloc = FALSE;
> ++
> ++ if (r_type == R_LARCH_TLS_IE_PC_HI20
> ++ || r_type == R_LARCH_TLS_IE_HI20)
> ++ is_ie = TRUE;
> ++
> ++ bfd_vma got_off = 0;
> ++ if (h != NULL)
> ++ {
> ++ got_off = h->got.offset;
> ++ h->got.offset |= 1;
> ++ }
> ++ else
> ++ {
> ++ got_off = local_got_offsets[r_symndx];
> ++ local_got_offsets[r_symndx] |= 1;
> ++ }
> ++
> ++ BFD_ASSERT (got_off != MINUS_ONE);
> ++
> ++ ie_off = 0;
> ++ tls_type = _bfd_loongarch_elf_tls_type (input_bfd, h, r_symndx);
> ++ if ((tls_type & GOT_TLS_GD) && (tls_type & GOT_TLS_IE))
> ++ ie_off = 2 * GOT_ENTRY_SIZE;
> ++
> ++ if ((got_off & 1) == 0)
> ++ {
> ++ Elf_Internal_Rela rela;
> ++ asection *relgot = htab->elf.srelgot;
> ++ bfd_vma tls_block_off = 0;
> ++
> ++ if (SYMBOL_REFERENCES_LOCAL (info, h))
> ++ {
> ++ BFD_ASSERT (elf_hash_table (info)->tls_sec);
> ++ tls_block_off = relocation
> ++ - elf_hash_table (info)->tls_sec->vma;
> ++ }
> ++
> ++ if (tls_type & GOT_TLS_GD)
> ++ {
> ++ rela.r_offset = sec_addr (got) + got_off;
> ++ rela.r_addend = 0;
> ++ if (SYMBOL_REFERENCES_LOCAL (info, h))
> ++ {
> ++ /* Local sym, used in exec, set module id 1. */
> ++ if (bfd_link_executable (info))
> ++ bfd_put_NN (output_bfd, 1, got->contents +
> got_off);
> ++ else
> ++ {
> ++ rela.r_info = ELFNN_R_INFO (0,
> R_LARCH_TLS_DTPMODNN);
> ++ loongarch_elf_append_rela (output_bfd, relgot,
> &rela);
> ++ }
> ++
> ++ bfd_put_NN (output_bfd, tls_block_off,
> ++ got->contents + got_off +
> GOT_ENTRY_SIZE);
> ++ }
> ++ /* Dynamic resolved. */
> ++ else
> ++ {
> ++ /* Dynamic relocate module id. */
> ++ rela.r_info = ELFNN_R_INFO (h->dynindx,
> ++ R_LARCH_TLS_DTPMODNN);
> ++ loongarch_elf_append_rela (output_bfd, relgot,
> &rela);
> ++
> ++ /* Dynamic relocate offset of block. */
> ++ rela.r_offset += GOT_ENTRY_SIZE;
> ++ rela.r_info = ELFNN_R_INFO (h->dynindx,
> ++ R_LARCH_TLS_DTPRELNN);
> ++ loongarch_elf_append_rela (output_bfd, relgot,
> &rela);
> ++ }
> ++ }
> ++ if (tls_type & GOT_TLS_IE)
> ++ {
> ++ rela.r_offset = sec_addr (got) + got_off + ie_off;
> ++ if (SYMBOL_REFERENCES_LOCAL (info, h))
> ++ {
> ++ /* Local sym, used in exec, set module id 1. */
> ++ if (!bfd_link_executable (info))
> ++ {
> ++ rela.r_info = ELFNN_R_INFO (0,
> R_LARCH_TLS_TPRELNN);
> ++ rela.r_addend = tls_block_off;
> ++ loongarch_elf_append_rela (output_bfd, relgot,
> &rela);
> ++ }
> ++
> ++ bfd_put_NN (output_bfd, tls_block_off,
> ++ got->contents + got_off + ie_off);
> ++ }
> ++ /* Dynamic resolved. */
> ++ else
> ++ {
> ++ /* Dynamic relocate offset of block. */
> ++ rela.r_info = ELFNN_R_INFO (h->dynindx,
> ++ R_LARCH_TLS_TPRELNN);
> ++ rela.r_addend = 0;
> ++ loongarch_elf_append_rela (output_bfd, relgot,
> &rela);
> ++ }
> ++ }
> ++ }
> ++ relocation = (got_off & (~(bfd_vma)1)) + sec_addr (got)
> ++ + (is_ie ? ie_off : 0);
> ++
> ++ if (r_type == R_LARCH_TLS_LD_PC_HI20
> ++ || r_type == R_LARCH_TLS_GD_PC_HI20
> ++ || r_type == R_LARCH_TLS_IE_PC_HI20)
> ++ RELOCATE_CALC_PC32_HI20 (relocation, pc);
> ++
> ++ break;
> ++
> ++ case R_LARCH_TLS_IE_PC_LO12:
> ++ case R_LARCH_TLS_IE64_PC_LO20:
> ++ case R_LARCH_TLS_IE64_PC_HI12:
> ++ case R_LARCH_TLS_IE_LO12:
> ++ case R_LARCH_TLS_IE64_LO20:
> ++ case R_LARCH_TLS_IE64_HI12:
> ++ unresolved_reloc = FALSE;
> ++
> ++ if (h)
> ++ relocation = sec_addr (got) + (h->got.offset & (~(bfd_vma)3));
> ++ else
> ++ relocation = sec_addr (got)
> ++ + (local_got_offsets[r_symndx] & (~(bfd_vma)3));
> ++
> ++ tls_type = _bfd_loongarch_elf_tls_type (input_bfd, h, r_symndx);
> ++ /* Use both TLS_GD and TLS_IE. */
> ++ if ((tls_type & GOT_TLS_GD) && (tls_type & GOT_TLS_IE))
> ++ relocation += 2 * GOT_ENTRY_SIZE;
> ++
> ++ if (r_type == R_LARCH_TLS_IE_PC_LO12)
> ++ relocation &= (bfd_vma)0xfff;
> ++ else if (r_type == R_LARCH_TLS_IE64_PC_LO20
> ++ || r_type == R_LARCH_TLS_IE64_PC_HI12)
> ++ RELOCATE_CALC_PC64_HI32 (relocation, pc);
> ++
> ++ break;
> ++
> ++ case R_LARCH_RELAX:
> ++ break;
> ++
> ++ default:
> ++ break;
> ++ }
> ++
> ++ if (fatal)
> ++ break;
> ++
> ++ do
> ++ {
> ++ /* 'unresolved_reloc' means we haven't done it yet.
> ++ We need help of dynamic linker to fix this memory location
> up. */
> ++ if (!unresolved_reloc)
> ++ break;
> ++
> ++ if (_bfd_elf_section_offset (output_bfd, info, input_section,
> ++ rel->r_offset) == MINUS_ONE)
> ++ /* WHY? May because it's invalid so skip checking.
> ++ But why dynamic reloc a invalid section? */
> ++ break;
> ++
> ++ if (input_section->output_section->flags & SEC_DEBUGGING)
> ++ {
> ++ fatal = (loongarch_reloc_is_fatal
> ++ (info, input_bfd, input_section, rel, howto,
> ++ bfd_reloc_dangerous, is_undefweak, name,
> ++ "Seems dynamic linker not process "
> ++ "sections 'SEC_DEBUGGING'."));
> ++ }
> ++ if (!is_dyn)
> ++ break;
> ++
> ++ if ((info->flags & DF_TEXTREL) == 0)
> ++ if (input_section->output_section->flags & SEC_READONLY)
> ++ info->flags |= DF_TEXTREL;
> ++ }
> ++ while (0);
> ++
> ++ if (fatal)
> ++ break;
> ++
> ++ loongarch_record_one_reloc (input_bfd, input_section, r_type,
> ++ rel->r_offset, sym, h, rel->r_addend);
> ++
> ++ if (r != bfd_reloc_continue)
> ++ r = perform_relocation (rel, input_section, howto, relocation,
> ++ input_bfd, contents);
> ++
> ++ switch (r)
> ++ {
> ++ case bfd_reloc_dangerous:
> ++ case bfd_reloc_continue:
> ++ case bfd_reloc_ok:
> ++ continue;
> ++
> ++ case bfd_reloc_overflow:
> ++ /* Overflow value can't be filled in. */
> ++ loongarch_dump_reloc_record (info->callbacks->info);
> ++ info->callbacks->reloc_overflow
> ++ (info, h ? &h->root : NULL, name, howto->name, rel->r_addend,
> ++ input_bfd, input_section, rel->r_offset);
> ++ break;
> ++
> ++ case bfd_reloc_outofrange:
> ++ /* Stack state incorrect. */
> ++ loongarch_dump_reloc_record (info->callbacks->info);
> ++ info->callbacks->info
> ++ ("%X%H: Internal stack state is incorrect.\n"
> ++ "Want to push to full stack or pop from empty stack?\n",
> ++ input_bfd, input_section, rel->r_offset);
> ++ break;
> ++
> ++ case bfd_reloc_notsupported:
> ++ info->callbacks->info ("%X%H: Unknown relocation type.\n",
> input_bfd,
> ++ input_section, rel->r_offset);
> ++ break;
> ++
> ++ default:
> ++ info->callbacks->info ("%X%H: Internal: unknown error.\n",
> input_bfd,
> ++ input_section, rel->r_offset);
> ++ break;
> ++ }
> ++
> ++ fatal = TRUE;
> ++ }
> ++
> ++ return !fatal;
> ++}
> ++
> ++/* Finish up dynamic symbol handling. We set the contents of various
> ++ dynamic sections here. */
> ++
> ++static bfd_boolean
> ++loongarch_elf_finish_dynamic_symbol (bfd *output_bfd,
> ++ struct bfd_link_info *info,
> ++ struct elf_link_hash_entry *h,
> ++ Elf_Internal_Sym *sym)
> ++{
> ++ struct loongarch_elf_link_hash_table *htab = loongarch_elf_hash_table
> (info);
> ++ const struct elf_backend_data *bed = get_elf_backend_data (output_bfd);
> ++
> ++ if (h->plt.offset != MINUS_ONE)
> ++ {
> ++ size_t i, plt_idx;
> ++ asection *plt, *gotplt, *relplt;
> ++ bfd_vma got_address;
> ++ uint32_t plt_entry[PLT_ENTRY_INSNS];
> ++ bfd_byte *loc;
> ++ Elf_Internal_Rela rela;
> ++
> ++ if (htab->elf.splt)
> ++ {
> ++ BFD_ASSERT ((h->type == STT_GNU_IFUNC
> ++ && SYMBOL_REFERENCES_LOCAL (info, h))
> ++ || h->dynindx != -1);
> ++
> ++ plt = htab->elf.splt;
> ++ gotplt = htab->elf.sgotplt;
> ++ if (h->type == STT_GNU_IFUNC && SYMBOL_REFERENCES_LOCAL (info,
> h))
> ++ relplt = htab->elf.srelgot;
> ++ else
> ++ relplt = htab->elf.srelplt;
> ++ plt_idx = (h->plt.offset - PLT_HEADER_SIZE) / PLT_ENTRY_SIZE;
> ++ got_address =
> ++ sec_addr (gotplt) + GOTPLT_HEADER_SIZE + plt_idx *
> GOT_ENTRY_SIZE;
> ++ }
> ++ else /* if (htab->elf.iplt) */
> ++ {
> ++ BFD_ASSERT (h->type == STT_GNU_IFUNC
> ++ && SYMBOL_REFERENCES_LOCAL (info, h));
> ++
> ++ plt = htab->elf.iplt;
> ++ gotplt = htab->elf.igotplt;
> ++ relplt = htab->elf.irelplt;
> ++ plt_idx = h->plt.offset / PLT_ENTRY_SIZE;
> ++ got_address = sec_addr (gotplt) + plt_idx * GOT_ENTRY_SIZE;
> ++ }
> ++
> ++ /* Find out where the .plt entry should go. */
> ++ loc = plt->contents + h->plt.offset;
> ++
> ++ /* Fill in the PLT entry itself. */
> ++ if (!loongarch_make_plt_entry (got_address,
> ++ sec_addr (plt) + h->plt.offset,
> ++ plt_entry))
> ++ return FALSE;
> ++
> ++ for (i = 0; i < PLT_ENTRY_INSNS; i++)
> ++ bfd_put_32 (output_bfd, plt_entry[i], loc + 4 * i);
> ++
> ++ /* Fill in the initial value of the got.plt entry. */
> ++ loc = gotplt->contents + (got_address - sec_addr (gotplt));
> ++ bfd_put_NN (output_bfd, sec_addr (plt), loc);
> ++
> ++ rela.r_offset = got_address;
> ++
> ++ /* TRUE if this is a PLT reference to a local IFUNC. */
> ++ if (PLT_LOCAL_IFUNC_P (info, h)
> ++ && (relplt == htab->elf.srelgot
> ++ || relplt == htab->elf.irelplt))
> ++ {
> ++ {
> ++ rela.r_info = ELFNN_R_INFO (0, R_LARCH_IRELATIVE);
> ++ rela.r_addend = (h->root.u.def.value
> ++ + h->root.u.def.section->output_section->vma
> ++ + h->root.u.def.section->output_offset);
> ++ }
> ++
> ++ /* Find the space after dyn sort. */
> ++ {
> ++ Elf_Internal_Rela *dyn = (Elf_Internal_Rela
> *)relplt->contents;
> ++ bfd_boolean fill = FALSE;
> ++ for (;dyn < dyn + relplt->size / sizeof (*dyn); dyn++)
> ++ {
> ++ if (0 == dyn->r_offset)
> ++ {
> ++ bed->s->swap_reloca_out (output_bfd, &rela,
> ++ (bfd_byte *)dyn);
> ++ relplt->reloc_count++;
> ++ fill = TRUE;
> ++ break;
> ++ }
> ++ }
> ++ BFD_ASSERT (fill);
> ++ }
> ++
> ++ }
> ++ else
> ++ {
> ++ /* Fill in the entry in the rela.plt section. */
> ++ rela.r_info = ELFNN_R_INFO (h->dynindx, R_LARCH_JUMP_SLOT);
> ++ rela.r_addend = 0;
> ++ loc = relplt->contents + plt_idx * sizeof (ElfNN_External_Rela);
> ++ bed->s->swap_reloca_out (output_bfd, &rela, loc);
> ++ }
> ++
> ++ if (!h->def_regular)
> ++ {
> ++ /* Mark the symbol as undefined, rather than as defined in
> ++ the .plt section. Leave the value alone. */
> ++ sym->st_shndx = SHN_UNDEF;
> ++ /* If the symbol is weak, we do need to clear the value.
> ++ Otherwise, the PLT entry would provide a definition for
> ++ the symbol even if the symbol wasn't defined anywhere,
> ++ and so the symbol would never be NULL. */
> ++ if (!h->ref_regular_nonweak)
> ++ sym->st_value = 0;
> ++ }
> ++ }
> ++
> ++ if (h->got.offset != MINUS_ONE
> ++ /* TLS got entry have been handled in elf_relocate_section. */
> ++ && !(loongarch_elf_hash_entry (h)->tls_type & (GOT_TLS_GD |
> GOT_TLS_IE))
> ++ /* Have allocated got entry but not allocated rela before. */
> ++ && !UNDEFWEAK_NO_DYNAMIC_RELOC (info, h))
> ++ {
> ++ asection *sgot, *srela;
> ++ Elf_Internal_Rela rela;
> ++ bfd_vma off = h->got.offset & ~(bfd_vma)1;
> ++
> ++ /* This symbol has an entry in the GOT. Set it up. */
> ++ sgot = htab->elf.sgot;
> ++ srela = htab->elf.srelgot;
> ++ BFD_ASSERT (sgot && srela);
> ++
> ++ rela.r_offset = sec_addr (sgot) + off;
> ++
> ++ if (h->def_regular
> ++ && h->type == STT_GNU_IFUNC)
> ++ {
> ++ if(h->plt.offset == MINUS_ONE)
> ++ {
> ++ if (htab->elf.splt == NULL)
> ++ srela = htab->elf.irelplt;
> ++
> ++ if (SYMBOL_REFERENCES_LOCAL (info, h))
> ++ {
> ++ asection *sec = h->root.u.def.section;
> ++ rela.r_info = ELFNN_R_INFO (0, R_LARCH_IRELATIVE);
> ++ rela.r_addend = h->root.u.def.value +
> sec->output_section->vma
> ++ + sec->output_offset;
> ++ bfd_put_NN (output_bfd, 0, sgot->contents + off);
> ++ }
> ++ else
> ++ {
> ++ BFD_ASSERT (h->dynindx != -1);
> ++ rela.r_info = ELFNN_R_INFO (h->dynindx, R_LARCH_NN);
> ++ rela.r_addend = 0;
> ++ bfd_put_NN (output_bfd, (bfd_vma) 0, sgot->contents +
> off);
> ++ }
> ++ }
> ++ else if(bfd_link_pic (info))
> ++ {
> ++ rela.r_info = ELFNN_R_INFO (h->dynindx, R_LARCH_NN);
> ++ rela.r_addend = 0;
> ++ bfd_put_NN (output_bfd, rela.r_addend, sgot->contents + off);
> ++ }
> ++ else
> ++ {
> ++ asection *plt;
> ++ /* For non-shared object, we can't use .got.plt, which
> ++ contains the real function address if we need pointer
> ++ equality. We load the GOT entry with the PLT entry. */
> ++ plt = htab->elf.splt ? htab->elf.splt : htab->elf.iplt;
> ++ bfd_put_NN (output_bfd,
> ++ (plt->output_section->vma
> ++ + plt->output_offset
> ++ + h->plt.offset),
> ++ sgot->contents + off);
> ++ return TRUE;
> ++ }
> ++ }
> ++ else if (bfd_link_pic (info) && SYMBOL_REFERENCES_LOCAL (info, h))
> ++ {
> ++ asection *sec = h->root.u.def.section;
> ++ rela.r_info = ELFNN_R_INFO (0, R_LARCH_RELATIVE);
> ++ rela.r_addend = (h->root.u.def.value + sec->output_section->vma
> ++ + sec->output_offset);
> ++ }
> ++ else
> ++ {
> ++ BFD_ASSERT (h->dynindx != -1);
> ++ rela.r_info = ELFNN_R_INFO (h->dynindx, R_LARCH_NN);
> ++ rela.r_addend = 0;
> ++ }
> ++
> ++ loongarch_elf_append_rela (output_bfd, srela, &rela);
> ++ }
> ++
> ++ /* Mark some specially defined symbols as absolute. */
> ++ if (h == htab->elf.hdynamic || h == htab->elf.hgot || h ==
> htab->elf.hplt)
> ++ sym->st_shndx = SHN_ABS;
> ++
> ++ return TRUE;
> ++}
> ++
> ++/* Finish up the dynamic sections. */
> ++
> ++static bfd_boolean
> ++loongarch_finish_dyn (bfd *output_bfd, struct bfd_link_info *info, bfd
> *dynobj,
> ++ asection *sdyn)
> ++{
> ++ struct loongarch_elf_link_hash_table *htab = loongarch_elf_hash_table
> (info);
> ++ const struct elf_backend_data *bed = get_elf_backend_data (output_bfd);
> ++ size_t dynsize = bed->s->sizeof_dyn, skipped_size = 0;
> ++ bfd_byte *dyncon, *dynconend;
> ++
> ++ dynconend = sdyn->contents + sdyn->size;
> ++ for (dyncon = sdyn->contents; dyncon < dynconend; dyncon += dynsize)
> ++ {
> ++ Elf_Internal_Dyn dyn;
> ++ asection *s;
> ++ int skipped = 0;
> ++
> ++ bed->s->swap_dyn_in (dynobj, dyncon, &dyn);
> ++
> ++ switch (dyn.d_tag)
> ++ {
> ++ case DT_PLTGOT:
> ++ s = htab->elf.sgotplt;
> ++ dyn.d_un.d_ptr = s->output_section->vma + s->output_offset;
> ++ break;
> ++ case DT_JMPREL:
> ++ s = htab->elf.srelplt;
> ++ dyn.d_un.d_ptr = s->output_section->vma + s->output_offset;
> ++ break;
> ++ case DT_PLTRELSZ:
> ++ s = htab->elf.srelplt;
> ++ dyn.d_un.d_val = s->size;
> ++ break;
> ++ case DT_TEXTREL:
> ++ if ((info->flags & DF_TEXTREL) == 0)
> ++ skipped = 1;
> ++ break;
> ++ case DT_FLAGS:
> ++ if ((info->flags & DF_TEXTREL) == 0)
> ++ dyn.d_un.d_val &= ~DF_TEXTREL;
> ++ break;
> ++ }
> ++ if (skipped)
> ++ skipped_size += dynsize;
> ++ else
> ++ bed->s->swap_dyn_out (output_bfd, &dyn, dyncon - skipped_size);
> ++ }
> ++ /* Wipe out any trailing entries if we shifted down a dynamic tag. */
> ++ memset (dyncon - skipped_size, 0, skipped_size);
> ++ return TRUE;
> ++}
> ++
> ++/* Finish up local dynamic symbol handling. We set the contents of
> ++ various dynamic sections here. */
> ++
> ++static bfd_boolean
> ++elfNN_loongarch_finish_local_dynamic_symbol (void **slot, void *inf)
> ++{
> ++ struct elf_link_hash_entry *h = (struct elf_link_hash_entry *) *slot;
> ++ struct bfd_link_info *info = (struct bfd_link_info *) inf;
> ++
> ++ return loongarch_elf_finish_dynamic_symbol (info->output_bfd, info, h,
> NULL);
> ++}
> ++
> ++static bfd_boolean
> ++loongarch_elf_finish_dynamic_sections (bfd *output_bfd,
> ++ struct bfd_link_info *info)
> ++{
> ++ bfd *dynobj;
> ++ asection *sdyn, *plt, *gotplt = NULL;
> ++ struct loongarch_elf_link_hash_table *htab;
> ++
> ++ htab = loongarch_elf_hash_table (info);
> ++ BFD_ASSERT (htab);
> ++ dynobj = htab->elf.dynobj;
> ++ sdyn = bfd_get_linker_section (dynobj, ".dynamic");
> ++
> ++ if (elf_hash_table (info)->dynamic_sections_created)
> ++ {
> ++ BFD_ASSERT (htab->elf.splt && sdyn);
> ++
> ++ if (!loongarch_finish_dyn (output_bfd, info, dynobj, sdyn))
> ++ return FALSE;
> ++ }
> ++
> ++ plt = htab->elf.splt;
> ++ gotplt = htab->elf.sgotplt;
> ++
> ++ if (plt && 0 < plt->size)
> ++ {
> ++ size_t i;
> ++ uint32_t plt_header[PLT_HEADER_INSNS];
> ++ if (!loongarch_make_plt_header (sec_addr (gotplt), sec_addr (plt),
> ++ plt_header))
> ++ return FALSE;
> ++
> ++ for (i = 0; i < PLT_HEADER_INSNS; i++)
> ++ bfd_put_32 (output_bfd, plt_header[i], plt->contents + 4 * i);
> ++
> ++ elf_section_data (plt->output_section)->this_hdr.sh_entsize =
> ++ PLT_ENTRY_SIZE;
> ++ }
> ++
> ++ if (htab->elf.sgotplt)
> ++ {
> ++ asection *output_section = htab->elf.sgotplt->output_section;
> ++
> ++ if (bfd_is_abs_section (output_section))
> ++ {
> ++ _bfd_error_handler (_("discarded output section: `%pA'"),
> ++ htab->elf.sgotplt);
> ++ return FALSE;
> ++ }
> ++
> ++ if (0 < htab->elf.sgotplt->size)
> ++ {
> ++ /* Write the first two entries in .got.plt, needed for the
> dynamic
> ++ linker. */
> ++ bfd_put_NN (output_bfd, MINUS_ONE, htab->elf.sgotplt->contents);
> ++
> ++ bfd_put_NN (output_bfd, (bfd_vma) 0,
> ++ htab->elf.sgotplt->contents + GOT_ENTRY_SIZE);
> ++ }
> ++
> ++ elf_section_data (output_section)->this_hdr.sh_entsize =
> GOT_ENTRY_SIZE;
> ++ }
> ++
> ++ if (htab->elf.sgot)
> ++ {
> ++ asection *output_section = htab->elf.sgot->output_section;
> ++
> ++ if (0 < htab->elf.sgot->size)
> ++ {
> ++ /* Set the first entry in the global offset table to the address
> of
> ++ the dynamic section. */
> ++ bfd_vma val = sdyn ? sec_addr (sdyn) : 0;
> ++ bfd_put_NN (output_bfd, val, htab->elf.sgot->contents);
> ++ }
> ++
> ++ elf_section_data (output_section)->this_hdr.sh_entsize =
> GOT_ENTRY_SIZE;
> ++ }
> ++
> ++ /* Fill PLT and GOT entries for local STT_GNU_IFUNC symbols. */
> ++ htab_traverse (htab->loc_hash_table,
> ++ (void *) elfNN_loongarch_finish_local_dynamic_symbol,
> info);
> ++
> ++ return TRUE;
> ++}
> ++
> ++/* Return address for Ith PLT stub in section PLT, for relocation REL
> ++ or (bfd_vma) -1 if it should not be included. */
> ++
> ++static bfd_vma
> ++loongarch_elf_plt_sym_val (bfd_vma i, const asection *plt,
> ++ const arelent *rel ATTRIBUTE_UNUSED)
> ++{
> ++ return plt->vma + PLT_HEADER_SIZE + i * PLT_ENTRY_SIZE;
> ++}
> ++
> ++static enum elf_reloc_type_class
> ++loongarch_reloc_type_class (const struct bfd_link_info *info
> ATTRIBUTE_UNUSED,
> ++ const asection *rel_sec ATTRIBUTE_UNUSED,
> ++ const Elf_Internal_Rela *rela)
> ++{
> ++ struct loongarch_elf_link_hash_table *htab;
> ++ htab = loongarch_elf_hash_table (info);
> ++
> ++ if (htab->elf.dynsym != NULL && htab->elf.dynsym->contents != NULL)
> ++ {
> ++ /* Check relocation against STT_GNU_IFUNC symbol if there are
> ++ dynamic symbols. */
> ++ bfd *abfd = info->output_bfd;
> ++ const struct elf_backend_data *bed = get_elf_backend_data (abfd);
> ++ unsigned long r_symndx = ELFNN_R_SYM (rela->r_info);
> ++ if (r_symndx != STN_UNDEF)
> ++ {
> ++ Elf_Internal_Sym sym;
> ++ if (!bed->s->swap_symbol_in (abfd,
> ++ htab->elf.dynsym->contents
> ++ + r_symndx * bed->s->sizeof_sym,
> ++ 0, &sym))
> ++ {
> ++ /* xgettext:c-format */
> ++ _bfd_error_handler (_("%pB symbol number %lu references"
> ++ " nonexistent SHT_SYMTAB_SHNDX
> section"),
> ++ abfd, r_symndx);
> ++ /* Ideally an error class should be returned here. */
> ++ }
> ++ else if (ELF_ST_TYPE (sym.st_info) == STT_GNU_IFUNC)
> ++ return reloc_class_ifunc;
> ++ }
> ++ }
> ++
> ++ switch (ELFNN_R_TYPE (rela->r_info))
> ++ {
> ++ case R_LARCH_IRELATIVE:
> ++ return reloc_class_ifunc;
> ++ case R_LARCH_RELATIVE:
> ++ return reloc_class_relative;
> ++ case R_LARCH_JUMP_SLOT:
> ++ return reloc_class_plt;
> ++ case R_LARCH_COPY:
> ++ return reloc_class_copy;
> ++ default:
> ++ return reloc_class_normal;
> ++ }
> ++}
> ++
> ++/* Copy the extra info we tack onto an elf_link_hash_entry. */
> ++
> ++static void
> ++loongarch_elf_copy_indirect_symbol (struct bfd_link_info *info,
> ++ struct elf_link_hash_entry *dir,
> ++ struct elf_link_hash_entry *ind)
> ++{
> ++ struct elf_link_hash_entry *edir, *eind;
> ++
> ++ edir = dir;
> ++ eind = ind;
> ++
> ++ if (eind->dyn_relocs != NULL)
> ++ {
> ++ if (edir->dyn_relocs != NULL)
> ++ {
> ++ struct elf_dyn_relocs **pp;
> ++ struct elf_dyn_relocs *p;
> ++
> ++ /* Add reloc counts against the indirect sym to the direct sym
> ++ list. Merge any entries against the same section. */
> ++ for (pp = &eind->dyn_relocs; (p = *pp) != NULL;)
> ++ {
> ++ struct elf_dyn_relocs *q;
> ++
> ++ for (q = edir->dyn_relocs; q != NULL; q = q->next)
> ++ if (q->sec == p->sec)
> ++ {
> ++ q->pc_count += p->pc_count;
> ++ q->count += p->count;
> ++ *pp = p->next;
> ++ break;
> ++ }
> ++ if (q == NULL)
> ++ pp = &p->next;
> ++ }
> ++ *pp = edir->dyn_relocs;
> ++ }
> ++
> ++ edir->dyn_relocs = eind->dyn_relocs;
> ++ eind->dyn_relocs = NULL;
> ++ }
> ++
> ++ if (ind->root.type == bfd_link_hash_indirect && dir->got.refcount < 0)
> ++ {
> ++ loongarch_elf_hash_entry(edir)->tls_type
> ++ = loongarch_elf_hash_entry(eind)->tls_type;
> ++ loongarch_elf_hash_entry(eind)->tls_type = GOT_UNKNOWN;
> ++ }
> ++ _bfd_elf_link_hash_copy_indirect (info, dir, ind);
> ++}
> ++
> ++#define PRSTATUS_SIZE 0x1d8
> ++#define PRSTATUS_OFFSET_PR_CURSIG 0xc
> ++#define PRSTATUS_OFFSET_PR_PID 0x20
> ++#define ELF_GREGSET_T_SIZE 0x168
> ++#define PRSTATUS_OFFSET_PR_REG 0x70
> ++
> ++/* Support for core dump NOTE sections. */
> ++
> ++static bfd_boolean
> ++loongarch_elf_grok_prstatus (bfd *abfd, Elf_Internal_Note *note)
> ++{
> ++ switch (note->descsz)
> ++ {
> ++ default:
> ++ return FALSE;
> ++
> ++ /* The sizeof (struct elf_prstatus) on Linux/LoongArch. */
> ++ case PRSTATUS_SIZE:
> ++ /* pr_cursig */
> ++ elf_tdata (abfd)->core->signal =
> ++ bfd_get_16 (abfd, note->descdata + PRSTATUS_OFFSET_PR_CURSIG);
> ++
> ++ /* pr_pid */
> ++ elf_tdata (abfd)->core->lwpid =
> ++ bfd_get_32 (abfd, note->descdata + PRSTATUS_OFFSET_PR_PID);
> ++ break;
> ++ }
> ++
> ++ /* Make a ".reg/999" section. */
> ++ return _bfd_elfcore_make_pseudosection (abfd, ".reg",
> ELF_GREGSET_T_SIZE,
> ++ note->descpos
> ++ + PRSTATUS_OFFSET_PR_REG);
> ++}
> ++
> ++#define PRPSINFO_SIZE 0x88
> ++#define PRPSINFO_OFFSET_PR_PID 0x18
> ++#define PRPSINFO_OFFSET_PR_FNAME 0x28
> ++#define PRPSINFO_SIZEOF_PR_FNAME 0x10
> ++#define PRPSINFO_OFFSET_PR_PS_ARGS 0x38
> ++#define PRPSINFO_SIZEOF_PR_PS_ARGS 0x50
> ++
> ++static bfd_boolean
> ++loongarch_elf_grok_psinfo (bfd *abfd, Elf_Internal_Note *note)
> ++{
> ++ switch (note->descsz)
> ++ {
> ++ default:
> ++ return FALSE;
> ++
> ++ /* The sizeof (prpsinfo_t) on Linux/LoongArch. */
> ++ case PRPSINFO_SIZE:
> ++ /* pr_pid */
> ++ elf_tdata (abfd)->core->pid =
> ++ bfd_get_32 (abfd, note->descdata + PRPSINFO_OFFSET_PR_PID);
> ++
> ++ /* pr_fname */
> ++ elf_tdata (abfd)->core->program =
> ++ _bfd_elfcore_strndup (abfd, note->descdata +
> PRPSINFO_OFFSET_PR_FNAME,
> ++ PRPSINFO_SIZEOF_PR_FNAME);
> ++
> ++ /* pr_psargs */
> ++ elf_tdata (abfd)->core->command =
> ++ _bfd_elfcore_strndup (abfd, note->descdata +
> PRPSINFO_OFFSET_PR_PS_ARGS,
> ++ PRPSINFO_SIZEOF_PR_PS_ARGS);
> ++ break;
> ++ }
> ++
> ++ /* Note that for some reason, a spurious space is tacked
> ++ onto the end of the args in some (at least one anyway)
> ++ implementations, so strip it off if it exists. */
> ++
> ++ {
> ++ char *command = elf_tdata (abfd)->core->command;
> ++ int n = strlen (command);
> ++
> ++ if (0 < n && command[n - 1] == ' ')
> ++ command[n - 1] = '\0';
> ++ }
> ++
> ++ return TRUE;
> ++}
> ++
> ++/* Set the right mach type. */
> ++static bfd_boolean
> ++loongarch_elf_object_p (bfd *abfd)
> ++{
> ++ /* There are only two mach types in LoongArch currently. */
> ++ if (strcmp (abfd->xvec->name, "elf64-loongarch") == 0)
> ++ bfd_default_set_arch_mach (abfd, bfd_arch_loongarch,
> bfd_mach_loongarch64);
> ++ else
> ++ bfd_default_set_arch_mach (abfd, bfd_arch_loongarch,
> bfd_mach_loongarch32);
> ++ return TRUE;
> ++}
> ++
> ++static asection *
> ++loongarch_elf_gc_mark_hook (asection *sec, struct bfd_link_info *info,
> ++ Elf_Internal_Rela *rel,
> ++ struct elf_link_hash_entry *h,
> ++ Elf_Internal_Sym *sym)
> ++{
> ++ if (h != NULL)
> ++ switch (ELFNN_R_TYPE (rel->r_info))
> ++ {
> ++ case R_LARCH_GNU_VTINHERIT:
> ++ case R_LARCH_GNU_VTENTRY:
> ++ return NULL;
> ++ }
> ++
> ++ return _bfd_elf_gc_mark_hook (sec, info, rel, h, sym);
> ++}
> ++
> ++/* Return TRUE if symbol H should be hashed in the `.gnu.hash' section.
> For
> ++ executable PLT slots where the executable never takes the address of
> those
> ++ functions, the function symbols are not added to the hash table. */
> ++
> ++static bfd_boolean
> ++elf_loongarch64_hash_symbol (struct elf_link_hash_entry *h)
> ++{
> ++ if (h->plt.offset != (bfd_vma) -1
> ++ && !h->def_regular
> ++ && !h->pointer_equality_needed)
> ++ return FALSE;
> ++
> ++ return _bfd_elf_hash_symbol (h);
> ++}
> ++
> ++#define TARGET_LITTLE_SYM loongarch_elfNN_vec
> ++#define TARGET_LITTLE_NAME "elfNN-loongarch"
> ++#define ELF_ARCH bfd_arch_loongarch
> ++#define ELF_TARGET_ID LARCH_ELF_DATA
> ++#define ELF_MACHINE_CODE EM_LOONGARCH
> ++#define ELF_MAXPAGESIZE 0x4000
> ++#define bfd_elfNN_bfd_reloc_type_lookup loongarch_reloc_type_lookup
> ++#define bfd_elfNN_bfd_link_hash_table_create \
> ++ loongarch_elf_link_hash_table_create
> ++#define bfd_elfNN_bfd_reloc_name_lookup loongarch_reloc_name_lookup
> ++#define elf_info_to_howto_rel NULL /* Fall through to
> elf_info_to_howto. */
> ++#define elf_info_to_howto loongarch_info_to_howto_rela
> ++#define bfd_elfNN_bfd_merge_private_bfd_data \
> ++ elfNN_loongarch_merge_private_bfd_data
> ++
> ++#define elf_backend_reloc_type_class loongarch_reloc_type_class
> ++#define elf_backend_copy_indirect_symbol
> loongarch_elf_copy_indirect_symbol
> ++#define elf_backend_create_dynamic_sections
> \
> ++ loongarch_elf_create_dynamic_sections
> ++#define elf_backend_check_relocs loongarch_elf_check_relocs
> ++#define elf_backend_adjust_dynamic_symbol
> loongarch_elf_adjust_dynamic_symbol
> ++#define elf_backend_size_dynamic_sections
> loongarch_elf_size_dynamic_sections
> ++#define elf_backend_relocate_section loongarch_elf_relocate_section
> ++#define elf_backend_finish_dynamic_symbol
> loongarch_elf_finish_dynamic_symbol
> ++#define elf_backend_finish_dynamic_sections
> \
> ++ loongarch_elf_finish_dynamic_sections
> ++#define elf_backend_object_p loongarch_elf_object_p
> ++#define elf_backend_gc_mark_hook loongarch_elf_gc_mark_hook
> ++#define elf_backend_plt_sym_val loongarch_elf_plt_sym_val
> ++#define elf_backend_grok_prstatus loongarch_elf_grok_prstatus
> ++#define elf_backend_grok_psinfo loongarch_elf_grok_psinfo
> ++#define elf_backend_hash_symbol elf_loongarch64_hash_symbol
> ++
> ++#include "elfNN-target.h"
> +diff --git gdb-10.2/bfd/elfxx-loongarch.c gdb-10.2/bfd/elfxx-loongarch.c
> +new file mode 100644
> +index 0000000..9f54b00
> +--- /dev/null
> ++++ gdb-10.2/bfd/elfxx-loongarch.c
> +@@ -0,0 +1,1618 @@
> ++/* LoongArch-specific support for ELF.
> ++ Copyright (C) 2021-2022 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ Based on LoongArch target.
> ++
> ++ This file is part of BFD, the Binary File Descriptor library.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program; see the file COPYING3. If not,
> ++ see <http://www.gnu.org/licenses/>. */
> ++
> ++#include "sysdep.h"
> ++#include "bfd.h"
> ++#include "libbfd.h"
> ++#include "elf-bfd.h"
> ++#include "elf/loongarch.h"
> ++#include "elfxx-loongarch.h"
> ++
> ++#define ALL_ONES (~ (bfd_vma) 0)
> ++
> ++typedef struct loongarch_reloc_howto_type_struct
> ++{
> ++ /* The first must be reloc_howto_type! */
> ++ reloc_howto_type howto;
> ++ bfd_reloc_code_real_type bfd_type;
> ++ bfd_boolean (*adjust_reloc_bits)(reloc_howto_type *, bfd_vma *);
> ++ const char *larch_reloc_type_name;
> ++} loongarch_reloc_howto_type;
> ++
> ++#define LOONGARCH_DEFAULT_HOWTO(r_name)
> \
> ++ { HOWTO (R_LARCH_##r_name, 0, 2, 32, FALSE, 0,
> complain_overflow_signed, \
> ++ bfd_elf_generic_reloc, "R_LARCH_" #r_name, FALSE, 0, ALL_ONES,
> \
> ++ FALSE), BFD_RELOC_LARCH_##r_name, NULL, NULL }
> ++
> ++#define LOONGARCH_HOWTO(type, right, size, bits, pcrel, left, ovf,
> func, \
> ++ name, inplace, src_mask, dst_mask, pcrel_off, btype,
> afunc,lname) \
> ++ { HOWTO(type, right, size, bits, pcrel, left, ovf, func, name, \
> ++ inplace, src_mask, dst_mask, pcrel_off), btype, afunc, lname }
> ++
> ++#define LOONGARCH_EMPTY_HOWTO(C) \
> ++ { EMPTY_HOWTO (C), BFD_RELOC_NONE, NULL, NULL }
> ++
> ++static bfd_boolean
> ++reloc_bits (reloc_howto_type *howto, bfd_vma *val);
> ++static bfd_boolean
> ++reloc_bits_b16 (reloc_howto_type *howto, bfd_vma *fix_val);
> ++static bfd_boolean
> ++reloc_bits_b21 (reloc_howto_type *howto, bfd_vma *fix_val);
> ++static bfd_boolean
> ++reloc_bits_b26 (reloc_howto_type *howto, bfd_vma *val);
> ++
> ++/* This does not include any relocation information, but should be
> ++ good enough for GDB or objdump to read the file. */
> ++static loongarch_reloc_howto_type loongarch_howto_table[] =
> ++{
> ++ /* No relocation. */
> ++ LOONGARCH_HOWTO (R_LARCH_NONE, /* type (0). */
> ++ 0, /* rightshift */
> ++ 3, /* size */
> ++ 0, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_dont, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_NONE", /* name */
> ++ FALSE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ 0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_NONE, /* bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ /* 32 bit relocation. */
> ++ LOONGARCH_HOWTO (R_LARCH_32, /* type (1). */
> ++ 0, /* rightshift */
> ++ 2, /* size */
> ++ 32, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_dont, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_32", /* name */
> ++ FALSE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_32, /* bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ /* 64 bit relocation. */
> ++ LOONGARCH_HOWTO (R_LARCH_64, /* type (2). */
> ++ 0, /* rightshift */
> ++ 4, /* size */
> ++ 64, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_dont, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_64", /* name */
> ++ FALSE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_64, /* bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_RELATIVE, /* type (3). */
> ++ 0, /* rightshift */
> ++ 2, /* size */
> ++ 32, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_dont, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_RELATIVE", /* name */
> ++ FALSE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_NONE, /* undefined? */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_COPY, /* type (4). */
> ++ 0, /* rightshift */
> ++ 3, /* this one is variable size */
> ++ 0, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_bitfield, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_COPY", /* name */
> ++ FALSE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ 0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_NONE, /* undefined? */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_JUMP_SLOT, /* type (5). */
> ++ 0, /* rightshift */
> ++ 4, /* size */
> ++ 64, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_bitfield, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_JUMP_SLOT", /* name */
> ++ FALSE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ 0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_NONE, /* undefined? */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ /* Dynamic TLS relocations. */
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_DTPMOD32, /* type (6). */
> ++ 0, /* rightshift */
> ++ 2, /* size */
> ++ 32, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_dont, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_TLS_DTPMOD32", /* name */
> ++ FALSE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_DTPMOD32, /* bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_DTPMOD64, /* type (7). */
> ++ 0, /* rightshift */
> ++ 4, /* size */
> ++ 64, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_dont, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_TLS_DTPMOD64", /* name */
> ++ FALSE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_DTPMOD64, /* bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_DTPREL32, /* type (8). */
> ++ 0, /* rightshift */
> ++ 2, /* size */
> ++ 32, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_dont, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_TLS_DTPREL32", /* name */
> ++ TRUE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_DTPREL32, /* bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_DTPREL64, /* type (9). */
> ++ 0, /* rightshift */
> ++ 4, /* size */
> ++ 64, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_dont, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_TLS_DTPREL64", /* name */
> ++ TRUE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_DTPREL64, /* bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_TPREL32, /* type (10). */
> ++ 0, /* rightshift */
> ++ 2, /* size */
> ++ 32, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_dont, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_TLS_TPREL32", /* name */
> ++ FALSE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_TPREL32, /* bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_TPREL64, /* type (11). */
> ++ 0, /* rightshift */
> ++ 4, /* size */
> ++ 64, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_dont, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_TLS_TPREL64", /* name */
> ++ FALSE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_TPREL64, /* bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_IRELATIVE, /* type (12). */
> ++ 0, /* rightshift */
> ++ 2, /* size */
> ++ 32, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_dont, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_IRELATIVE", /* name */
> ++ FALSE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_NONE, /* undefined? */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_EMPTY_HOWTO (13),
> ++ LOONGARCH_EMPTY_HOWTO (14),
> ++ LOONGARCH_EMPTY_HOWTO (15),
> ++ LOONGARCH_EMPTY_HOWTO (16),
> ++ LOONGARCH_EMPTY_HOWTO (17),
> ++ LOONGARCH_EMPTY_HOWTO (18),
> ++ LOONGARCH_EMPTY_HOWTO (19),
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_MARK_LA, /* type (20). */
> ++ 0, /* rightshift. */
> ++ 3, /* size. */
> ++ 0, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_MARK_LA", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask. */
> ++ 0, /* dst_mask. */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_MARK_LA, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_MARK_PCREL, /* type (21). */
> ++ 0, /* rightshift. */
> ++ 3, /* size. */
> ++ 0, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_MARK_PCREL", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask. */
> ++ 0, /* dst_mask. */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_MARK_PCREL, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SOP_PUSH_PCREL, /* type (22). */
> ++ 2, /* rightshift. */
> ++ 2, /* size. */
> ++ 32, /* bitsize. */
> ++ TRUE /* FIXME: somewhat use this. */, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SOP_PUSH_PCREL", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0x03ffffff, /* src_mask. */
> ++ 0x03ffffff, /* dst_mask. */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SOP_PUSH_PCREL, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ /* type 23-37. */
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_PUSH_ABSOLUTE),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_PUSH_DUP),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_PUSH_GPREL),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_PUSH_TLS_TPREL),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_PUSH_TLS_GOT),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_PUSH_TLS_GD),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_PUSH_PLT_PCREL),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_ASSERT),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_NOT),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_SUB),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_SL),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_SR),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_ADD),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_AND),
> ++ LOONGARCH_DEFAULT_HOWTO (SOP_IF_ELSE),
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SOP_POP_32_S_10_5, /* type (38). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 5, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /*
> complain_on_overflow. */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SOP_POP_32_S_10_5", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x7c00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_5, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name
> */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SOP_POP_32_U_10_12, /* type (39). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_unsigned, /*
> complain_on_overflow. */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SOP_POP_32_U_10_12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SOP_POP_32_U_10_12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name
> */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SOP_POP_32_S_10_12, /* type (40). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /*
> complain_on_overflow. */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SOP_POP_32_S_10_12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name
> */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SOP_POP_32_S_10_16, /* type (41). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 16, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /*
> complain_on_overflow. */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SOP_POP_32_S_10_16", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3fffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_16, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name
> */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SOP_POP_32_S_10_16_S2, /* type (42). */
> ++ 2, /* rightshift. */
> ++ 2, /* size. */
> ++ 16, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /*
> complain_on_overflow. */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SOP_POP_32_S_10_16_S2", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3fffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_16_S2, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits_b16, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name
> */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SOP_POP_32_S_5_20, /* type (43). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /*
> complain_on_overflow. */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SOP_POP_32_S_5_20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_5_20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name
> */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SOP_POP_32_S_0_5_10_16_S2,
> ++ /* type (44). */
> ++ 2, /* rightshift. */
> ++ 2, /* size. */
> ++ 21, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /*
> complain_on_overflow. */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SOP_POP_32_S_0_5_10_16_S2", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0xfc0003e0, /* src_mask */
> ++ 0xfc0003e0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_0_5_10_16_S2,
> ++ /*
> bfd_reloc_code_real_type */
> ++ reloc_bits_b21, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name
> */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SOP_POP_32_S_0_10_10_16_S2, /* type
> (45). */
> ++ 2, /* rightshift. */
> ++ 2, /* size. */
> ++ 26, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SOP_POP_32_S_0_10_10_16_S2", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x03ffffff, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_0_10_10_16_S2,
> ++ /*
> bfd_reloc_code_real_type */
> ++ reloc_bits_b26, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SOP_POP_32_U, /* type (46). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 32, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_unsigned, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SOP_POP_32_S_U", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0xffffffff00000000, /* src_mask */
> ++ 0x00000000ffffffff, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SOP_POP_32_U, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_ADD8, /* type (47). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 8, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_ADD8", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_ADD8, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_ADD16, /* type (48). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 16, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_ADD16", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_ADD16, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_ADD24, /* type (49). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 24, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_ADD24", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_ADD24, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_ADD32, /* type (50). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 32, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_ADD32", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_ADD32, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_ADD64, /* type (51). */
> ++ 0, /* rightshift. */
> ++ 4, /* size. */
> ++ 64, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_ADD64", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_ADD64, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SUB8, /* type (52). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 8, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SUB8", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SUB8, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SUB16, /* type (53). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 16, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SUB16", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SUB16, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SUB24, /* type (54). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 24, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SUB24", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SUB24, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SUB32, /* type (55). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 32, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SUB32", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SUB32, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_SUB64, /* type (56). */
> ++ 0, /* rightshift. */
> ++ 4, /* size. */
> ++ 64, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_SUB64", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ ALL_ONES, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_SUB64, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_GNU_VTINHERIT, /* type (57). */
> ++ 0, /* rightshift. */
> ++ 3, /* size. */
> ++ 0, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_GNU_VTINHERIT", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_NONE, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_GNU_VTENTRY, /* type (58). */
> ++ 0, /* rightshift. */
> ++ 3, /* size. */
> ++ 0, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ NULL, /* special_function. */
> ++ "R_LARCH_GNU_VTENTRY", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_NONE, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_EMPTY_HOWTO (59),
> ++ LOONGARCH_EMPTY_HOWTO (60),
> ++ LOONGARCH_EMPTY_HOWTO (61),
> ++ LOONGARCH_EMPTY_HOWTO (62),
> ++ LOONGARCH_EMPTY_HOWTO (63),
> ++
> ++ /* New reloc types. */
> ++ LOONGARCH_HOWTO (R_LARCH_B16, /* type (64). */
> ++ 2, /* rightshift. */
> ++ 2, /* size. */
> ++ 16, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_B16", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0x3fffc00, /* src_mask */
> ++ 0x3fffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_B16, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits_b16, /* adjust_reloc_bits */
> ++ "b16"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_B21, /* type (65). */
> ++ 2, /* rightshift. */
> ++ 2, /* size. */
> ++ 21, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_B21", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0xfc0003e0, /* src_mask */
> ++ 0xfc0003e0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_B21, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits_b21, /* adjust_reloc_bits */
> ++ "b21"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_B26, /* type (66). */
> ++ 2, /* rightshift. */
> ++ 2, /* size. */
> ++ 26, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_B26", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x03ffffff, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_B26, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits_b26, /* adjust_reloc_bits */
> ++ "b26"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_ABS_HI20, /* type (67). */
> ++ 12, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_ABS_HI20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_ABS_HI20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "abs_hi20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_ABS_LO12, /* type (68). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_unsigned, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_ABS_LO12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_ABS_LO12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "abs_lo12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_ABS64_LO20, /* type (69). */
> ++ 32, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_ABS64_LO20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_ABS64_LO20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "abs64_lo20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_ABS64_HI12, /* type (70). */
> ++ 52, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_ABS64_HI12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_ABS64_HI12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "abs64_hi12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_PCALA_HI20, /* type (71). */
> ++ 12, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_PCALA_HI20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_PCALA_HI20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "pc_hi20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_PCALA_LO12, /* type (72). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_PCALA_LO12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_PCALA_LO12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "pc_lo12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_PCALA64_LO20, /* type (73). */
> ++ 32, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_PCALA64_LO20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_PCALA64_LO20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "pc64_lo20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_PCALA64_HI12, /* type (74). */
> ++ 52, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_PCALA64_HI12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_PCALA64_HI12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "pc64_hi12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_GOT_PC_HI20, /* type (75). */
> ++ 12, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_GOT_PC_HI20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_GOT_PC_HI20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "got_pc_hi20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_GOT_PC_LO12, /* type (76). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_GOT_PC_LO12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_GOT_PC_LO12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "got_pc_lo12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_GOT64_PC_LO20, /* type (77). */
> ++ 32, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_GOT64_PC_LO20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_GOT64_PC_LO20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "got64_pc_lo20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_GOT64_PC_HI12, /* type (78). */
> ++ 52, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_GOT64_PC_HI12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_GOT64_PC_HI12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "got64_pc_hi12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_GOT_HI20, /* type (79). */
> ++ 12, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_GOT_HI20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_GOT_HI20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "got_hi20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_GOT_LO12, /* type (80). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_GOT_LO12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_GOT_LO12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "got_lo12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_GOT64_LO20, /* type (81). */
> ++ 32, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_GOT64_LO20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_GOT64_LO20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "got64_lo20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_GOT64_HI12, /* type (82). */
> ++ 52, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_GOT64_HI12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_GOT64_HI12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "got64_hi12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_LE_HI20, /* type (83). */
> ++ 12, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_LE_HI20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_LE_HI20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "le_hi20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_LE_LO12, /* type (84). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_LE_LO12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_LE_LO12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "le_lo12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_LE64_LO20, /* type (85). */
> ++ 32, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_LE64_LO20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_LE64_LO20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "le64_lo20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_LE64_HI12, /* type (86). */
> ++ 52, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_LE64_HI12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_LE64_HI12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "le64_hi12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_IE_PC_HI20, /* type (87). */
> ++ 12, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_IE_PC_HI20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_IE_PC_HI20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "ie_pc_hi20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_IE_PC_LO12, /* type (88). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_unsigned, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_IE_PC_LO12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_IE_PC_LO12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "ie_pc_lo12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_IE64_PC_LO20, /* type (89). */
> ++ 32, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_IE64_PC_LO20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_IE64_PC_LO20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "ie64_pc_lo20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_IE64_PC_HI12, /* type (90). */
> ++ 52, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_IE64_PC_HI12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_IE64_PC_HI12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "ie64_pc_hi12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_IE_HI20, /* type (91). */
> ++ 12, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_IE_HI20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_IE_HI20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "ie_hi20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_IE_LO12, /* type (92). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_IE_LO12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_IE_LO12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "ie_lo12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_IE64_LO20, /* type (93). */
> ++ 32, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_IE64_LO20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_IE64_LO20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "ie64_lo20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_IE64_HI12, /* type (94). */
> ++ 52, /* rightshift. */
> ++ 2, /* size. */
> ++ 12, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 10, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_IE64_HI12", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x3ffc00, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_IE64_HI12, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "ie64_hi12"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_LD_PC_HI20, /* type (95). */
> ++ 12, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_LD_PC_HI20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_LD_PC_HI20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "ld_pc_hi20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_LD_HI20, /* type (96). */
> ++ 12, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_LD_HI20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_LD_HI20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "ld_hi20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_GD_PC_HI20, /* type (97). */
> ++ 12, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_GD_PC_HI20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_GD_PC_HI20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "gd_pc_hi20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_TLS_GD_HI20, /* type (98). */
> ++ 12, /* rightshift. */
> ++ 2, /* size. */
> ++ 20, /* bitsize. */
> ++ FALSE, /* pc_relative. */
> ++ 5, /* bitpos. */
> ++ complain_overflow_signed, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_TLS_GD_HI20", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0x1ffffe0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_TLS_GD_HI20, /*
> bfd_reloc_code_real_type */
> ++ reloc_bits, /* adjust_reloc_bits */
> ++ "gd_hi20"), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_32_PCREL, /* type (99). */
> ++ 0, /* rightshift. */
> ++ 2, /* size. */
> ++ 32, /* bitsize. */
> ++ TRUE, /* pc_relative. */
> ++ 0, /* bitpos. */
> ++ complain_overflow_dont, /* complain_on_overflow.
> */
> ++ bfd_elf_generic_reloc, /* special_function. */
> ++ "R_LARCH_32_PCREL", /* name. */
> ++ FALSE, /* partial_inplace. */
> ++ 0, /* src_mask */
> ++ 0xffffffff, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_32_PCREL, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++ LOONGARCH_HOWTO (R_LARCH_RELAX, /* type (100). */
> ++ 0, /* rightshift */
> ++ 0, /* size */
> ++ 0, /* bitsize */
> ++ FALSE, /* pc_relative */
> ++ 0, /* bitpos */
> ++ complain_overflow_dont, /* complain_on_overflow */
> ++ bfd_elf_generic_reloc, /* special_function */
> ++ "R_LARCH_RELAX", /* name */
> ++ FALSE, /* partial_inplace */
> ++ 0, /* src_mask */
> ++ 0, /* dst_mask */
> ++ FALSE, /* pcrel_offset */
> ++ BFD_RELOC_LARCH_RELAX, /*
> bfd_reloc_code_real_type */
> ++ NULL, /* adjust_reloc_bits */
> ++ NULL), /* larch_reloc_type_name */
> ++
> ++};
> ++
> ++reloc_howto_type *
> ++loongarch_elf_rtype_to_howto (bfd *abfd, unsigned int r_type)
> ++{
> ++ if(r_type < R_LARCH_count)
> ++ {
> ++ /* For search table fast. */
> ++ BFD_ASSERT (ARRAY_SIZE (loongarch_howto_table) == R_LARCH_count);
> ++
> ++ if (loongarch_howto_table[r_type].howto.type == r_type)
> ++ return (reloc_howto_type *)&loongarch_howto_table[r_type];
> ++
> ++ for (size_t i = 0; i < ARRAY_SIZE (loongarch_howto_table); i++)
> ++ if (loongarch_howto_table[i].howto.type == r_type)
> ++ return (reloc_howto_type *)&loongarch_howto_table[i];
> ++ }
> ++
> ++ (*_bfd_error_handler) (_("%pB: unsupported relocation type %#x"),
> ++ abfd, r_type);
> ++ bfd_set_error (bfd_error_bad_value);
> ++ return NULL;
> ++}
> ++
> ++reloc_howto_type *
> ++loongarch_reloc_name_lookup (bfd *abfd ATTRIBUTE_UNUSED, const char
> *r_name)
> ++{
> ++ BFD_ASSERT (ARRAY_SIZE (loongarch_howto_table) == R_LARCH_count);
> ++
> ++ for (size_t i = 0; i < ARRAY_SIZE (loongarch_howto_table); i++)
> ++ if (loongarch_howto_table[i].howto.name
> ++ && strcasecmp (loongarch_howto_table[i].howto.name, r_name) == 0)
> ++ return (reloc_howto_type *)&loongarch_howto_table[i];
> ++
> ++ (*_bfd_error_handler) (_("%pB: unsupported relocation type %s"),
> ++ abfd, r_name);
> ++ bfd_set_error (bfd_error_bad_value);
> ++
> ++ return NULL;
> ++}
> ++
> ++/* Cost so much. */
> ++reloc_howto_type *
> ++loongarch_reloc_type_lookup (bfd *abfd ATTRIBUTE_UNUSED,
> ++ bfd_reloc_code_real_type code)
> ++{
> ++ BFD_ASSERT (ARRAY_SIZE (loongarch_howto_table) == R_LARCH_count);
> ++
> ++ /* Fast search for new reloc types. */
> ++ if (BFD_RELOC_LARCH_B16 <= code && code < BFD_RELOC_LARCH_RELAX)
> ++ {
> ++ BFD_ASSERT (BFD_RELOC_LARCH_RELAX - BFD_RELOC_LARCH_B16
> ++ == R_LARCH_RELAX - R_LARCH_B16);
> ++ loongarch_reloc_howto_type *ht = NULL;
> ++ ht = &loongarch_howto_table[code - BFD_RELOC_LARCH_B16 +
> R_LARCH_B16];
> ++ BFD_ASSERT (ht->bfd_type == code);
> ++ return (reloc_howto_type *)ht;
> ++ }
> ++
> ++ for (size_t i = 0; i < ARRAY_SIZE (loongarch_howto_table); i++)
> ++ if (loongarch_howto_table[i].bfd_type == code)
> ++ return (reloc_howto_type *)&loongarch_howto_table[i];
> ++
> ++ (*_bfd_error_handler) (_("%pB: unsupported bfd relocation type %#x"),
> ++ abfd, code);
> ++ bfd_set_error (bfd_error_bad_value);
> ++
> ++ return NULL;
> ++}
> ++
> ++bfd_reloc_code_real_type
> ++loongarch_larch_reloc_name_lookup (bfd *abfd ATTRIBUTE_UNUSED,
> ++ const char *l_r_name)
> ++{
> ++ for (size_t i = 0; i < ARRAY_SIZE (loongarch_howto_table); i++)
> ++ {
> ++ loongarch_reloc_howto_type *lht = &loongarch_howto_table[i];
> ++ if ((NULL != lht->larch_reloc_type_name)
> ++ && (0 == strcmp (lht->larch_reloc_type_name, l_r_name)))
> ++ return lht->bfd_type;
> ++ }
> ++
> ++ (*_bfd_error_handler) (_("%pB: unsupported relocation type name %s"),
> ++ abfd, l_r_name);
> ++ bfd_set_error (bfd_error_bad_value);
> ++ return BFD_RELOC_NONE;
> ++}
> ++
> ++
> ++/* Functions for reloc bits field.
> ++ 1. Signed extend *fix_val.
> ++ 2. Return FALSE if overflow. */
> ++
> ++#define LARCH_RELOC_BFD_VMA_BIT_MASK(bitsize) \
> ++ (~((((bfd_vma)0x1) << (bitsize)) - 1))
> ++
> ++/* Adjust val to perform insn
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_5
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_12
> ++ BFD_RELOC_LARCH_SOP_POP_32_U_10_12
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_16
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_5_20
> ++ BFD_RELOC_LARCH_SOP_POP_32_U. */
> ++static bfd_boolean
> ++reloc_bits (reloc_howto_type *howto, bfd_vma *fix_val)
> ++{
> ++ bfd_signed_vma val = ((bfd_signed_vma)(*fix_val)) >> howto->rightshift;
> ++
> ++ /* Perform insn bits field. */
> ++ val = val & (((bfd_vma)0x1 << howto->bitsize) - 1);
> ++ val <<= howto->bitpos;
> ++
> ++ *fix_val = (bfd_vma)val;
> ++
> ++ return TRUE;
> ++}
> ++
> ++/* Adjust val to perform insn
> ++ R_LARCH_SOP_POP_32_S_10_16_S2
> ++ R_LARCH_B16. */
> ++static bfd_boolean
> ++reloc_bits_b16 (reloc_howto_type *howto, bfd_vma *fix_val)
> ++{
> ++ if (howto->complain_on_overflow != complain_overflow_signed)
> ++ return FALSE;
> ++
> ++ bfd_signed_vma val = *fix_val;
> ++
> ++ /* Judge whether 4 bytes align. */
> ++ if (val & ((0x1UL << howto->rightshift) - 1))
> ++ return FALSE;
> ++
> ++ int bitsize = howto->bitsize + howto->rightshift;
> ++ bfd_signed_vma sig_bit = (val >> (bitsize - 1)) & 0x1;
> ++
> ++ /* If val < 0, sign bit is 1. */
> ++ if (sig_bit)
> ++ {
> ++ /* Signed bits is 1. */
> ++ if ((LARCH_RELOC_BFD_VMA_BIT_MASK (bitsize - 1) & val)
> ++ != LARCH_RELOC_BFD_VMA_BIT_MASK (bitsize - 1))
> ++ return FALSE;
> ++ }
> ++ else
> ++ {
> ++ /* Signed bits is 0. */
> ++ if (LARCH_RELOC_BFD_VMA_BIT_MASK (bitsize) & val)
> ++ return FALSE;
> ++ }
> ++
> ++ /* Perform insn bits field. */
> ++ val >>= howto->rightshift;
> ++ val = val & (((bfd_vma)0x1 << howto->bitsize) - 1);
> ++ val <<= howto->bitpos;
> ++
> ++ *fix_val = val;
> ++
> ++ return TRUE;
> ++}
> ++
> ++/* Reloc type :
> ++ R_LARCH_SOP_POP_32_S_0_5_10_16_S2
> ++ R_LARCH_B21. */
> ++static bfd_boolean
> ++reloc_bits_b21 (reloc_howto_type *howto,
> ++ bfd_vma *fix_val)
> ++{
> ++ if (howto->complain_on_overflow != complain_overflow_signed)
> ++ return FALSE;
> ++
> ++ bfd_signed_vma val = *fix_val;
> ++
> ++ if (val & ((0x1UL << howto->rightshift) - 1))
> ++ return FALSE;
> ++
> ++ int bitsize = howto->bitsize + howto->rightshift;
> ++ bfd_signed_vma sig_bit = (val >> (bitsize - 1)) & 0x1;
> ++
> ++ /* If val < 0. */
> ++ if (sig_bit)
> ++ {
> ++ if ((LARCH_RELOC_BFD_VMA_BIT_MASK (bitsize - 1) & val)
> ++ != LARCH_RELOC_BFD_VMA_BIT_MASK (bitsize - 1))
> ++ return FALSE;
> ++ }
> ++ else
> ++ {
> ++ if (LARCH_RELOC_BFD_VMA_BIT_MASK (bitsize) & val)
> ++ return FALSE;
> ++ }
> ++
> ++ /* Perform insn bits field. */
> ++ val >>= howto->rightshift;
> ++ val = val & (((bfd_vma)0x1 << howto->bitsize) - 1);
> ++
> ++ /* Perform insn bits field. 15:0<<10, 20:16>>16. */
> ++ val = ((val & 0xffff) << 10) | ((val >> 16) & 0x1f);
> ++
> ++ *fix_val = val;
> ++
> ++ return TRUE;
> ++}
> ++
> ++/* Reloc type:
> ++ R_LARCH_SOP_POP_32_S_0_10_10_16_S2
> ++ R_LARCH_B26. */
> ++static bfd_boolean
> ++reloc_bits_b26 (reloc_howto_type *howto,
> ++ bfd_vma *fix_val)
> ++{
> ++ /* Return FALSE if overflow. */
> ++ if (howto->complain_on_overflow != complain_overflow_signed)
> ++ return FALSE;
> ++
> ++ bfd_signed_vma val = *fix_val;
> ++
> ++ if (val & ((0x1UL << howto->rightshift) - 1))
> ++ return FALSE;
> ++
> ++ int bitsize = howto->bitsize + howto->rightshift;
> ++ bfd_signed_vma sig_bit = (val >> (bitsize - 1)) & 0x1;
> ++
> ++ /* If val < 0. */
> ++ if (sig_bit)
> ++ {
> ++ if ((LARCH_RELOC_BFD_VMA_BIT_MASK (bitsize - 1) & val)
> ++ != LARCH_RELOC_BFD_VMA_BIT_MASK (bitsize - 1))
> ++ return FALSE;
> ++ }
> ++ else
> ++ {
> ++ if (LARCH_RELOC_BFD_VMA_BIT_MASK (bitsize) & val)
> ++ return FALSE;
> ++ }
> ++
> ++ /* Perform insn bits field. */
> ++ val >>= howto->rightshift;
> ++ val = val & (((bfd_vma)0x1 << howto->bitsize) - 1);
> ++
> ++ /* Perform insn bits field. 25:16>>16, 15:0<<10. */
> ++ val = ((val & 0xffff) << 10) | ((val >> 16) & 0x3ff);
> ++
> ++ *fix_val = val;
> ++
> ++ return TRUE;
> ++}
> ++
> ++bfd_boolean
> ++loongarch_adjust_reloc_bitsfield (reloc_howto_type *howto,
> ++ bfd_vma *fix_val)
> ++{
> ++ BFD_ASSERT (((loongarch_reloc_howto_type *)howto)->adjust_reloc_bits);
> ++ return ((loongarch_reloc_howto_type *)
> ++ howto)->adjust_reloc_bits(howto, fix_val);
> ++}
> +diff --git gdb-10.2/bfd/elfxx-loongarch.h gdb-10.2/bfd/elfxx-loongarch.h
> +new file mode 100644
> +index 0000000..49c1515
> +--- /dev/null
> ++++ gdb-10.2/bfd/elfxx-loongarch.h
> +@@ -0,0 +1,45 @@
> ++/* LoongArch-specific backend routines.
> ++ Copyright (C) 2021-2022 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of BFD, the Binary File Descriptor library.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program; see the file COPYING3. If not,
> ++ see <http://www.gnu.org/licenses/>. */
> ++
> ++#include "elf/common.h"
> ++#include "elf/internal.h"
> ++
> ++extern reloc_howto_type *
> ++loongarch_elf_rtype_to_howto (bfd *abfd, unsigned int r_type);
> ++
> ++extern reloc_howto_type *
> ++loongarch_reloc_type_lookup (bfd *abfd, bfd_reloc_code_real_type code);
> ++
> ++extern reloc_howto_type *
> ++loongarch_reloc_name_lookup (bfd *abfd ATTRIBUTE_UNUSED, const char
> *r_name);
> ++
> ++extern bfd_reloc_code_real_type
> ++loongarch_larch_reloc_name_lookup (bfd *abfd ATTRIBUTE_UNUSED,
> ++ const char *l_r_name);
> ++
> ++bfd_boolean loongarch_adjust_reloc_bitsfield (reloc_howto_type *howto,
> bfd_vma *fix_val);
> ++
> ++/* TRUE if this is a PLT reference to a local IFUNC. */
> ++#define PLT_LOCAL_IFUNC_P(INFO, H) \
> ++ ((H)->dynindx == -1 \
> ++ || ((bfd_link_executable (INFO) \
> ++ || ELF_ST_VISIBILITY ((H)->other) != STV_DEFAULT) \
> ++ && (H)->def_regular \
> ++ && (H)->type == STT_GNU_IFUNC))
> +diff --git gdb-10.2/bfd/libbfd.h gdb-10.2/bfd/libbfd.h
> +index 74d7e41..a3b40ec 100644
> +--- gdb-10.2/bfd/libbfd.h
> ++++ gdb-10.2/bfd/libbfd.h
> +@@ -3396,6 +3396,86 @@ static const char *const
> bfd_reloc_code_real_names[] = { "@@uninitialized@@",
> + "BFD_RELOC_CKCORE_PCREL_BLOOP_IMM4BY4",
> + "BFD_RELOC_CKCORE_PCREL_BLOOP_IMM12BY4",
> + "BFD_RELOC_S12Z_OPR",
> ++ "BFD_RELOC_LARCH_TLS_DTPMOD32",
> ++ "BFD_RELOC_LARCH_TLS_DTPREL32",
> ++ "BFD_RELOC_LARCH_TLS_DTPMOD64",
> ++ "BFD_RELOC_LARCH_TLS_DTPREL64",
> ++ "BFD_RELOC_LARCH_TLS_TPREL32",
> ++ "BFD_RELOC_LARCH_TLS_TPREL64",
> ++ "BFD_RELOC_LARCH_MARK_LA",
> ++ "BFD_RELOC_LARCH_MARK_PCREL",
> ++ "BFD_RELOC_LARCH_SOP_PUSH_PCREL",
> ++ "BFD_RELOC_LARCH_SOP_PUSH_ABSOLUTE",
> ++ "BFD_RELOC_LARCH_SOP_PUSH_DUP",
> ++ "BFD_RELOC_LARCH_SOP_PUSH_GPREL",
> ++ "BFD_RELOC_LARCH_SOP_PUSH_TLS_TPREL",
> ++ "BFD_RELOC_LARCH_SOP_PUSH_TLS_GOT",
> ++ "BFD_RELOC_LARCH_SOP_PUSH_TLS_GD",
> ++ "BFD_RELOC_LARCH_SOP_PUSH_PLT_PCREL",
> ++ "BFD_RELOC_LARCH_SOP_ASSERT",
> ++ "BFD_RELOC_LARCH_SOP_NOT",
> ++ "BFD_RELOC_LARCH_SOP_SUB",
> ++ "BFD_RELOC_LARCH_SOP_SL",
> ++ "BFD_RELOC_LARCH_SOP_SR",
> ++ "BFD_RELOC_LARCH_SOP_ADD",
> ++ "BFD_RELOC_LARCH_SOP_AND",
> ++ "BFD_RELOC_LARCH_SOP_IF_ELSE",
> ++ "BFD_RELOC_LARCH_SOP_POP_32_S_10_5",
> ++ "BFD_RELOC_LARCH_SOP_POP_32_U_10_12",
> ++ "BFD_RELOC_LARCH_SOP_POP_32_S_10_12",
> ++ "BFD_RELOC_LARCH_SOP_POP_32_S_10_16",
> ++ "BFD_RELOC_LARCH_SOP_POP_32_S_10_16_S2",
> ++ "BFD_RELOC_LARCH_SOP_POP_32_S_5_20",
> ++ "BFD_RELOC_LARCH_SOP_POP_32_S_0_5_10_16_S2",
> ++ "BFD_RELOC_LARCH_SOP_POP_32_S_0_10_10_16_S2",
> ++ "BFD_RELOC_LARCH_SOP_POP_32_U",
> ++ "BFD_RELOC_LARCH_ADD8",
> ++ "BFD_RELOC_LARCH_ADD16",
> ++ "BFD_RELOC_LARCH_ADD24",
> ++ "BFD_RELOC_LARCH_ADD32",
> ++ "BFD_RELOC_LARCH_ADD64",
> ++ "BFD_RELOC_LARCH_SUB8",
> ++ "BFD_RELOC_LARCH_SUB16",
> ++ "BFD_RELOC_LARCH_SUB24",
> ++ "BFD_RELOC_LARCH_SUB32",
> ++ "BFD_RELOC_LARCH_SUB64",
> ++ "BFD_RELOC_LARCH_B16",
> ++ "BFD_RELOC_LARCH_B21",
> ++ "BFD_RELOC_LARCH_B26",
> ++ "BFD_RELOC_LARCH_ABS_HI20",
> ++ "BFD_RELOC_LARCH_ABS_LO12",
> ++ "BFD_RELOC_LARCH_ABS64_LO20",
> ++ "BFD_RELOC_LARCH_ABS64_HI12",
> ++ "BFD_RELOC_LARCH_PCALA_HI20",
> ++ "BFD_RELOC_LARCH_PCALA_LO12",
> ++ "BFD_RELOC_LARCH_PCALA64_LO20",
> ++ "BFD_RELOC_LARCH_PCALA64_HI12",
> ++ "BFD_RELOC_LARCH_GOT_PC_HI20",
> ++ "BFD_RELOC_LARCH_GOT_PC_LO12",
> ++ "BFD_RELOC_LARCH_GOT64_PC_LO20",
> ++ "BFD_RELOC_LARCH_GOT64_PC_HI12",
> ++ "BFD_RELOC_LARCH_GOT_HI20",
> ++ "BFD_RELOC_LARCH_GOT_LO12",
> ++ "BFD_RELOC_LARCH_GOT64_LO20",
> ++ "BFD_RELOC_LARCH_GOT64_HI12",
> ++ "BFD_RELOC_LARCH_TLS_LE_HI20",
> ++ "BFD_RELOC_LARCH_TLS_LE_LO12",
> ++ "BFD_RELOC_LARCH_TLS_LE64_LO20",
> ++ "BFD_RELOC_LARCH_TLS_LE64_HI12",
> ++ "BFD_RELOC_LARCH_TLS_IE_PC_HI20",
> ++ "BFD_RELOC_LARCH_TLS_IE_PC_LO12",
> ++ "BFD_RELOC_LARCH_TLS_IE64_PC_LO20",
> ++ "BFD_RELOC_LARCH_TLS_IE64_PC_HI12",
> ++ "BFD_RELOC_LARCH_TLS_IE_HI20",
> ++ "BFD_RELOC_LARCH_TLS_IE_LO12",
> ++ "BFD_RELOC_LARCH_TLS_IE64_LO20",
> ++ "BFD_RELOC_LARCH_TLS_IE64_HI12",
> ++ "BFD_RELOC_LARCH_TLS_LD_PC_HI20",
> ++ "BFD_RELOC_LARCH_TLS_LD_HI20",
> ++ "BFD_RELOC_LARCH_TLS_GD_PC_HI20",
> ++ "BFD_RELOC_LARCH_TLS_GD_HI20",
> ++ "BFD_RELOC_LARCH_32_PCREL",
> ++ "BFD_RELOC_LARCH_RELAX",
> + "@@overflow: BFD_RELOC_UNUSED@@",
> + };
> + #endif
> +diff --git gdb-10.2/bfd/po/BLD-POTFILES.in gdb-10.2/bfd/po/BLD-POTFILES.in
> +index 5fc39cf..8385565 100644
> +--- gdb-10.2/bfd/po/BLD-POTFILES.in
> ++++ gdb-10.2/bfd/po/BLD-POTFILES.in
> +@@ -2,10 +2,12 @@ bfd_stdint.h
> + bfdver.h
> + elf32-aarch64.c
> + elf32-ia64.c
> ++elf32-loongarch.c
> + elf32-riscv.c
> + elf32-target.h
> + elf64-aarch64.c
> + elf64-ia64.c
> ++elf64-loongarch.c
> + elf64-riscv.c
> + elf64-target.h
> + peigen.c
> +diff --git gdb-10.2/bfd/po/SRC-POTFILES.in gdb-10.2/bfd/po/SRC-POTFILES.in
> +index 83530b2..96edf89 100644
> +--- gdb-10.2/bfd/po/SRC-POTFILES.in
> ++++ gdb-10.2/bfd/po/SRC-POTFILES.in
> +@@ -72,6 +72,7 @@ cpu-iq2000.c
> + cpu-k1om.c
> + cpu-l1om.c
> + cpu-lm32.c
> ++cpu-loongarch.c
> + cpu-m10200.c
> + cpu-m10300.c
> + cpu-m32c.c
> +diff --git gdb-10.2/bfd/reloc.c gdb-10.2/bfd/reloc.c
> +index dc923fe..9d70bb1 100644
> +--- gdb-10.2/bfd/reloc.c
> ++++ gdb-10.2/bfd/reloc.c
> +@@ -8142,6 +8142,177 @@ ENUM
> + ENUMDOC
> + S12Z relocations.
> +
> ++ENUM
> ++ BFD_RELOC_LARCH_TLS_DTPMOD32
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_DTPREL32
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_DTPMOD64
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_DTPREL64
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_TPREL32
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_TPREL64
> ++ENUMX
> ++ BFD_RELOC_LARCH_MARK_LA
> ++ENUMX
> ++ BFD_RELOC_LARCH_MARK_PCREL
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_PUSH_PCREL
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_PUSH_ABSOLUTE
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_PUSH_DUP
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_PUSH_GPREL
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_PUSH_TLS_TPREL
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_PUSH_TLS_GOT
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_PUSH_TLS_GD
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_PUSH_PLT_PCREL
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_ASSERT
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_NOT
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_SUB
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_SL
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_SR
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_ADD
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_AND
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_IF_ELSE
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_5
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_POP_32_U_10_12
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_12
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_16
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_10_16_S2
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_5_20
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_0_5_10_16_S2
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_POP_32_S_0_10_10_16_S2
> ++ENUMX
> ++ BFD_RELOC_LARCH_SOP_POP_32_U
> ++ENUMX
> ++ BFD_RELOC_LARCH_ADD8
> ++ENUMX
> ++ BFD_RELOC_LARCH_ADD16
> ++ENUMX
> ++ BFD_RELOC_LARCH_ADD24
> ++ENUMX
> ++ BFD_RELOC_LARCH_ADD32
> ++ENUMX
> ++ BFD_RELOC_LARCH_ADD64
> ++ENUMX
> ++ BFD_RELOC_LARCH_SUB8
> ++ENUMX
> ++ BFD_RELOC_LARCH_SUB16
> ++ENUMX
> ++ BFD_RELOC_LARCH_SUB24
> ++ENUMX
> ++ BFD_RELOC_LARCH_SUB32
> ++ENUMX
> ++ BFD_RELOC_LARCH_SUB64
> ++
> ++ENUMX
> ++ BFD_RELOC_LARCH_B16
> ++ENUMX
> ++ BFD_RELOC_LARCH_B21
> ++ENUMX
> ++ BFD_RELOC_LARCH_B26
> ++
> ++ENUMX
> ++ BFD_RELOC_LARCH_ABS_HI20
> ++ENUMX
> ++ BFD_RELOC_LARCH_ABS_LO12
> ++ENUMX
> ++ BFD_RELOC_LARCH_ABS64_LO20
> ++ENUMX
> ++ BFD_RELOC_LARCH_ABS64_HI12
> ++
> ++ENUMX
> ++ BFD_RELOC_LARCH_PCALA_HI20
> ++ENUMX
> ++ BFD_RELOC_LARCH_PCALA_LO12
> ++ENUMX
> ++ BFD_RELOC_LARCH_PCALA64_LO20
> ++ENUMX
> ++ BFD_RELOC_LARCH_PCALA64_HI12
> ++
> ++ENUMX
> ++ BFD_RELOC_LARCH_GOT_PC_HI20
> ++ENUMX
> ++ BFD_RELOC_LARCH_GOT_PC_LO12
> ++ENUMX
> ++ BFD_RELOC_LARCH_GOT64_PC_LO20
> ++ENUMX
> ++ BFD_RELOC_LARCH_GOT64_PC_HI12
> ++ENUMX
> ++ BFD_RELOC_LARCH_GOT_HI20
> ++ENUMX
> ++ BFD_RELOC_LARCH_GOT_LO12
> ++ENUMX
> ++ BFD_RELOC_LARCH_GOT64_LO20
> ++ENUMX
> ++ BFD_RELOC_LARCH_GOT64_HI12
> ++
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_LE_HI20
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_LE_LO12
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_LE64_LO20
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_LE64_HI12
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_IE_PC_HI20
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_IE_PC_LO12
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_IE64_PC_LO20
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_IE64_PC_HI12
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_IE_HI20
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_IE_LO12
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_IE64_LO20
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_IE64_HI12
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_LD_PC_HI20
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_LD_HI20
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_GD_PC_HI20
> ++ENUMX
> ++ BFD_RELOC_LARCH_TLS_GD_HI20
> ++
> ++ENUMX
> ++ BFD_RELOC_LARCH_32_PCREL
> ++
> ++ENUMX
> ++ BFD_RELOC_LARCH_RELAX
> ++
> ++ENUMDOC
> ++ LARCH relocations.
> ++
> + ENDSENUM
> + BFD_RELOC_UNUSED
> + CODE_FRAGMENT
> +diff --git gdb-10.2/bfd/targets.c gdb-10.2/bfd/targets.c
> +index 35492b9..5f2147f 100644
> +--- gdb-10.2/bfd/targets.c
> ++++ gdb-10.2/bfd/targets.c
> +@@ -759,6 +759,8 @@ extern const bfd_target l1om_elf64_vec;
> + extern const bfd_target l1om_elf64_fbsd_vec;
> + extern const bfd_target lm32_elf32_vec;
> + extern const bfd_target lm32_elf32_fdpic_vec;
> ++extern const bfd_target loongarch_elf64_vec;
> ++extern const bfd_target loongarch_elf32_vec;
> + extern const bfd_target m32c_elf32_vec;
> + extern const bfd_target m32r_elf32_vec;
> + extern const bfd_target m32r_elf32_le_vec;
> +@@ -1346,6 +1348,12 @@ static const bfd_target * const
> _bfd_target_vector[] =
> + &z80_elf32_vec,
> +
> + &z8k_coff_vec,
> ++
> ++#ifdef BFD64
> ++ &loongarch_elf32_vec,
> ++ &loongarch_elf64_vec,
> ++#endif
> ++
> + #endif /* not SELECT_VECS */
> +
> + /* Always support S-records, for convenience. */
> +diff --git gdb-10.2/config.guess gdb-10.2/config.guess
> +index 45001cf..1155f0a 100755
> +--- gdb-10.2/config.guess
> ++++ gdb-10.2/config.guess
> +@@ -985,6 +985,9 @@ EOF
> + k1om:Linux:*:*)
> + echo "$UNAME_MACHINE"-unknown-linux-"$LIBC"
> + exit ;;
> ++ loongarch32:Linux:*:* | loongarch64:Linux:*:* |
> loongarchx32:Linux:*:*)
> ++ echo "$UNAME_MACHINE"-unknown-linux-"$LIBC"
> ++ exit ;;
> + m32r*:Linux:*:*)
> + echo "$UNAME_MACHINE"-unknown-linux-"$LIBC"
> + exit ;;
> +diff --git gdb-10.2/config.sub gdb-10.2/config.sub
> +index f02d43a..df57c18 100755
> +--- gdb-10.2/config.sub
> ++++ gdb-10.2/config.sub
> +@@ -1183,6 +1183,7 @@ case $cpu-$vendor in
> + | k1om \
> + | le32 | le64 \
> + | lm32 \
> ++ | loongarch32 | loongarch64 | loongarchx32 \
> + | m32c | m32r | m32rle \
> + | m5200 | m68000 | m680[012346]0 | m68360 | m683?2
> | m68k \
> + | m6811 | m68hc11 | m6812 | m68hc12 | m68hcs12x \
> +diff --git gdb-10.2/gdb/Makefile.in gdb-10.2/gdb/Makefile.in
> +index ec371fc..67af7d0 100644
> +--- gdb-10.2/gdb/Makefile.in
> ++++ gdb-10.2/gdb/Makefile.in
> +@@ -712,6 +712,8 @@ ALL_TARGET_OBS = \
> + arch/i386.o \
> + arch/ppc-linux-common.o \
> + arch/riscv.o \
> ++ arch/loongarch.o \
> ++ arch/loongarch-linux-nat.o \
> + arm-bsd-tdep.o \
> + arm-fbsd-tdep.o \
> + arm-linux-tdep.o \
> +@@ -760,6 +762,9 @@ ALL_TARGET_OBS = \
> + linux-record.o \
> + linux-tdep.o \
> + lm32-tdep.o \
> ++ loongarch-tdep.o \
> ++ loongarch-linux-tdep.o \
> ++ loongarch-linux-nat.o \
> + m32c-tdep.o \
> + m32r-linux-tdep.o \
> + m32r-tdep.o \
> +@@ -1324,6 +1329,8 @@ HFILES_NO_SRCDIR = \
> + linux-record.h \
> + linux-tdep.h \
> + location.h \
> ++ loongarch-tdep.h \
> ++ loongarch-linux-tdep.h \
> + m2-lang.h \
> + m32r-tdep.h \
> + m68k-tdep.h \
> +@@ -1456,6 +1463,8 @@ HFILES_NO_SRCDIR = \
> + arch/arc.h \
> + arch/arm.h \
> + arch/i386.h \
> ++ arch/loongarch.h \
> ++ arch/loongarch-linux-nat.h \
> + arch/ppc-linux-common.h \
> + arch/ppc-linux-tdesc.h \
> + arch/riscv.h \
> +@@ -1500,6 +1509,7 @@ HFILES_NO_SRCDIR = \
> + nat/linux-personality.h \
> + nat/linux-ptrace.h \
> + nat/linux-waitpid.h \
> ++ nat/loongarch-linux-watch.h \
> + nat/mips-linux-watch.h \
> + nat/ppc-linux.h \
> + nat/x86-cpuid.h \
> +@@ -2205,6 +2215,9 @@ ALLDEPFILES = \
> + linux-record.c \
> + linux-tdep.c \
> + lm32-tdep.c \
> ++ loongarch-tdep.c \
> ++ loongarch-linux-tdep.c \
> ++ loongarch-linux-nat.c \
> + m32r-linux-nat.c \
> + m32r-linux-tdep.c \
> + m32r-tdep.c \
> +diff --git gdb-10.2/gdb/arch/loongarch-linux-nat.c
> gdb-10.2/gdb/arch/loongarch-linux-nat.c
> +new file mode 100644
> +index 0000000..baf59f8
> +--- /dev/null
> ++++ gdb-10.2/gdb/arch/loongarch-linux-nat.c
> +@@ -0,0 +1,93 @@
> ++/* Copyright (C) 2021 Free Software Foundation, Inc.
> ++
> ++ This file is part of GDB.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program. If not, see <http://www.gnu.org/licenses/>.
> */
> ++
> ++/* For external ptrace. */
> ++#ifdef GDBSERVER
> ++#include "server.h"
> ++#include "nat/gdb_ptrace.h"
> ++#else
> ++#include "defs.h"
> ++#include "nat/gdb_ptrace.h"
> ++#endif
> ++
> ++#include "arch/loongarch.h"
> ++#include "arch/loongarch-linux-nat.h"
> ++#include "loongarch-linux-tdep.h"
> ++#include "elf/common.h"
> ++#include <sys/uio.h>
> ++
> ++static uint32_t
> ++loongarch_cpucfg_may_ptrace (uint64_t rj, int tid)
> ++{
> ++ char t_buf[rj * 4 + 4];
> ++ struct iovec iovec = { .iov_base = &t_buf, .iov_len = sizeof (t_buf) };
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_LARCH_CPUCFG, &iovec) < 0)
> ++ ((uint32_t *) t_buf)[rj] = loongarch_cpucfg (rj);
> ++ return ((uint32_t *) t_buf)[rj];
> ++}
> ++
> ++struct target_desc *
> ++loongarch_linux_read_description_runtime (int tid)
> ++{
> ++ int rlen, flen, lbt, lsx, lasx;
> ++
> ++ uint32_t cpucfg1 = loongarch_cpucfg_may_ptrace (1, tid);
> ++ rlen = cpucfg1 & 0x2 /* LA64 */ ? 64 : 32;
> ++
> ++ uint32_t cpucfg2 = loongarch_cpucfg_may_ptrace (2, tid);
> ++ flen = 0;
> ++ if (cpucfg2 & 0x4 /* FP_DP */)
> ++ flen = 64;
> ++ else if (cpucfg2 & 0x2 /* FP_SP */)
> ++ flen = 32;
> ++ if (flen)
> ++ {
> ++ loongarch_elf_fpregset_t regset;
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_FPREGSET, &iovec) < 0)
> ++ flen = 0;
> ++ }
> ++
> ++ lbt = 0;
> ++ if (cpucfg2 & 0x1c0000 /* LBT_X86 || LBT_ARM || LBT_MIPS */)
> ++ {
> ++ loongarch_elf_lbtregset_t regset;
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_LARCH_LBT, &iovec) == 0)
> ++ lbt = 1;
> ++ }
> ++
> ++ lsx = 0;
> ++ if (cpucfg2 & 0x40 /* LSX */)
> ++ {
> ++ loongarch_elf_lsxregset_t regset;
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_LARCH_LSX, &iovec) == 0)
> ++ lsx = 1;
> ++ }
> ++
> ++ lasx = 0;
> ++ if (cpucfg2 & 0x80 /* LASX */)
> ++ {
> ++ loongarch_elf_lasxregset_t regset;
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_LARCH_LASX, &iovec) == 0)
> ++ lasx = 1;
> ++ }
> ++
> ++ return loongarch_create_target_description (rlen, flen, lbt, lsx,
> lasx);
> ++}
> +diff --git gdb-10.2/gdb/arch/loongarch-linux-nat.h
> gdb-10.2/gdb/arch/loongarch-linux-nat.h
> +new file mode 100644
> +index 0000000..a9cd453
> +--- /dev/null
> ++++ gdb-10.2/gdb/arch/loongarch-linux-nat.h
> +@@ -0,0 +1,35 @@
> ++/*
> ++ Copyright (C) 2021 Free Software Foundation, Inc.
> ++
> ++ This file is part of GDB.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program. If not, see <http://www.gnu.org/licenses/>.
> */
> ++
> ++#ifndef LOONGARCH_LINUX_NAT_H
> ++#define LOONGARCH_LINUX_NAT_H
> ++#include <stdint.h>
> ++
> ++static inline uint32_t
> ++loongarch_cpucfg (uint64_t rj)
> ++{
> ++ uint32_t ret;
> ++ asm ("cpucfg %0,%1" : "=r"(ret) : "r"(rj));
> ++ return ret;
> ++}
> ++
> ++struct target_desc;
> ++
> ++extern struct target_desc *loongarch_linux_read_description_runtime (int
> tid);
> ++
> ++#endif
> +diff --git gdb-10.2/gdb/arch/loongarch.c gdb-10.2/gdb/arch/loongarch.c
> +new file mode 100644
> +index 0000000..cc6b767
> +--- /dev/null
> ++++ gdb-10.2/gdb/arch/loongarch.c
> +@@ -0,0 +1,75 @@
> ++/* Copyright (C) 2021 Free Software Foundation, Inc.
> ++
> ++ This file is part of GDB.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program. If not, see <http://www.gnu.org/licenses/>.
> */
> ++
> ++#include "gdbsupport/common-defs.h"
> ++#include "gdbsupport/common-regcache.h"
> ++#include "arch/loongarch.h"
> ++
> ++unsigned int loongarch_debug = 0;
> ++
> ++#include <../features/loongarch/base32.c>
> ++#include <../features/loongarch/base64.c>
> ++#include <../features/loongarch/fpu32.c>
> ++#include <../features/loongarch/fpu64.c>
> ++#include <../features/loongarch/lbt32.c>
> ++#include <../features/loongarch/lbt64.c>
> ++#include <../features/loongarch/lsx.c>
> ++#include <../features/loongarch/lasx.c>
> ++
> ++target_desc *
> ++loongarch_create_target_description (int rlen, int flen, int lbt,
> ++ int lsx, int lasx)
> ++{
> ++ gdb_assert (rlen == 32 || rlen == 64);
> ++
> ++ target_desc *tdesc = allocate_target_description ();
> ++
> ++ set_tdesc_architecture (tdesc, rlen == 64 ? "loongarch64" :
> "loongarch32");
> ++
> ++ int regnum = 0;
> ++
> ++ if (rlen == 64)
> ++ regnum = create_feature_loongarch_base64 (tdesc, regnum);
> ++ else if (rlen == 32)
> ++ regnum = create_feature_loongarch_base32 (tdesc, regnum);
> ++ else
> ++ gdb_assert_not_reached ("rlen unknown");
> ++
> ++ if (flen == 64)
> ++ regnum = create_feature_loongarch_fpu64 (tdesc, regnum);
> ++ else if (flen == 32)
> ++ regnum = create_feature_loongarch_fpu32 (tdesc, regnum);
> ++
> ++ if (lbt && rlen == 32)
> ++ regnum = create_feature_loongarch_lbt32 (tdesc, regnum);
> ++ else if (lbt && rlen == 64)
> ++ regnum = create_feature_loongarch_lbt64 (tdesc, regnum);
> ++
> ++ if (lsx)
> ++ regnum = create_feature_loongarch_lsx (tdesc, regnum);
> ++
> ++ if (lasx)
> ++ regnum = create_feature_loongarch_lasx (tdesc, regnum);
> ++
> ++ return tdesc;
> ++}
> ++
> ++target_desc *
> ++loongarch_get_base_target_description (int rlen, int flen)
> ++{
> ++ return loongarch_create_target_description (rlen, flen, 0, 0, 0);
> ++}
> +diff --git gdb-10.2/gdb/arch/loongarch.h gdb-10.2/gdb/arch/loongarch.h
> +new file mode 100644
> +index 0000000..997afab
> +--- /dev/null
> ++++ gdb-10.2/gdb/arch/loongarch.h
> +@@ -0,0 +1,35 @@
> ++/*
> ++ Copyright (C) 2021 Free Software Foundation, Inc.
> ++
> ++ This file is part of GDB.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program. If not, see <http://www.gnu.org/licenses/>.
> */
> ++
> ++#ifndef ARCH_LOONGARCH_H
> ++#define ARCH_LOONGARCH_H
> ++
> ++#include "elf/loongarch.h"
> ++#include "opcode/loongarch.h"
> ++
> ++extern unsigned int loongarch_debug;
> ++
> ++struct target_desc;
> ++
> ++extern struct target_desc *loongarch_get_base_target_description (int
> rlen, int flen);
> ++
> ++extern struct target_desc *
> ++loongarch_create_target_description (int rlen, int flen, int lbt,
> ++ int lsx, int lasx);
> ++
> ++#endif
> +diff --git gdb-10.2/gdb/configure.host gdb-10.2/gdb/configure.host
> +index ce52823..046bc85 100644
> +--- gdb-10.2/gdb/configure.host
> ++++ gdb-10.2/gdb/configure.host
> +@@ -63,6 +63,7 @@ alpha*) gdb_host_cpu=alpha ;;
> + arm*) gdb_host_cpu=arm ;;
> + hppa*) gdb_host_cpu=pa ;;
> + i[34567]86*) gdb_host_cpu=i386 ;;
> ++loongarch*) gdb_host_cpu=loongarch ;;
> + m68*) gdb_host_cpu=m68k ;;
> + mips*) gdb_host_cpu=mips ;;
> + powerpc* | rs6000) gdb_host_cpu=powerpc ;;
> +@@ -120,6 +121,8 @@ i[34567]86-*-cygwin*) gdb_host=cygwin ;;
> +
> + ia64-*-linux*) gdb_host=linux ;;
> +
> ++loongarch*-linux*) gdb_host=linux ;;
> ++
> + m68*-*-linux*) gdb_host=linux ;;
> + m68*-*-netbsdelf* | m68*-*-knetbsd*-gnu)
> + gdb_host=nbsdelf ;;
> +diff --git gdb-10.2/gdb/configure.nat gdb-10.2/gdb/configure.nat
> +index bb70e30..fad1631 100644
> +--- gdb-10.2/gdb/configure.nat
> ++++ gdb-10.2/gdb/configure.nat
> +@@ -253,6 +253,10 @@ case ${gdb_host} in
> + # Host: Intel IA-64 running GNU/Linux
> + NATDEPFILES="${NATDEPFILES} ia64-linux-nat.o"
> + ;;
> ++ loongarch)
> ++ # Host: LoongArch, running GNU/Linux.
> ++ NATDEPFILES="${NATDEPFILES} loongarch-linux-nat.o
> arch/loongarch-linux-nat.o linux-nat-trad.o nat/loongarch-linux-watch.o"
> ++ ;;
> + m32r)
> + # Host: M32R based machine running GNU/Linux
> + NATDEPFILES="${NATDEPFILES} m32r-linux-nat.o"
> +diff --git gdb-10.2/gdb/configure.tgt gdb-10.2/gdb/configure.tgt
> +index a3e11c4..4454838 100644
> +--- gdb-10.2/gdb/configure.tgt
> ++++ gdb-10.2/gdb/configure.tgt
> +@@ -95,6 +95,9 @@ xtensa*)
> + cpu_obs="xtensa-tdep.o xtensa-config.o solib-svr4.o"
> + ;;
> +
> ++loongarch*)
> ++ cpu_obs="arch/loongarch.o";;
> ++
> + esac
> +
> + # 2. Get the objects per os in $TARG.
> +@@ -346,6 +349,11 @@ lm32-*-*)
> + gdb_sim=../sim/lm32/libsim.a
> + ;;
> +
> ++loongarch*-linux*)
> ++ gdb_target_obs="loongarch-tdep.o loongarch-linux-tdep.o
> glibc-tdep.o linux-tdep.o solib-svr4.o symfile-mem.o"
> ++ build_gdbserver=yes
> ++ ;;
> ++
> + m32c-*-*)
> + # Target: Renesas M32C family
> + gdb_target_obs="m32c-tdep.o"
> +diff --git gdb-10.2/gdb/doc/gdb.texinfo gdb-10.2/gdb/doc/gdb.texinfo
> +index 097aacd..94a8537 100644
> +--- gdb-10.2/gdb/doc/gdb.texinfo
> ++++ gdb-10.2/gdb/doc/gdb.texinfo
> +@@ -45227,6 +45227,7 @@ registers using the capitalization used in the
> description.
> + * ARC Features::
> + * ARM Features::
> + * i386 Features::
> ++* LoongArch Features::
> + * MicroBlaze Features::
> + * MIPS Features::
> + * M68K Features::
> +@@ -45444,6 +45445,15 @@ The @samp{org.gnu.gdb.i386.pkeys} feature is
> optional. It should
> + describe a single register, @samp{pkru}. It is a 32-bit register
> + valid for i386 and amd64.
> +
> ++@node LoongArch Features
> ++@subsection LoongArch Features
> ++@cindex target descriptions, LoongArch Features
> ++
> ++The @samp{org.gnu.gdb.loongarch.base} feature is required for LoongArch
> ++targets. It should contain the registers @samp{r0} through @samp{r31},
> ++@samp{pc}, and @samp{badvaddr}. Either the architectural names
> (@samp{r0},
> ++@samp{r1}, etc) can be used, or the ABI names (@samp{zero}, @samp{ra},
> etc).
> ++
> + @node MicroBlaze Features
> + @subsection MicroBlaze Features
> + @cindex target descriptions, MicroBlaze features
> +diff --git gdb-10.2/gdb/features/Makefile gdb-10.2/gdb/features/Makefile
> +index 67c3ae1..a6f8ea0 100644
> +--- gdb-10.2/gdb/features/Makefile
> ++++ gdb-10.2/gdb/features/Makefile
> +@@ -74,6 +74,7 @@ arm-expedite = r11,sp,pc
> + i386-expedite = ebp,esp,eip
> + amd64-expedite = rbp,rsp,rip
> + x32-expedite = rbp,rsp,rip
> ++loongarch-expedite = r3,pc
> + mips-expedite = r29,pc
> + mips-dsp-expedite = r29,pc
> + mips64-expedite = r29,pc
> +@@ -178,6 +179,7 @@ GDB = false
> + aarch64-feature = 1
> + arm-feature = 1
> + i386-feature = 1
> ++loongarch-feature = 1
> + riscv-feature = 1
> + tic6x-feature = 1
> +
> +@@ -228,6 +230,14 @@ FEATURE_XMLFILES = aarch64-core.xml \
> + i386/64bit-pkeys.xml \
> + i386/64bit-sse.xml \
> + i386/x32-core.xml \
> ++ loongarch/base32.xml \
> ++ loongarch/base64.xml \
> ++ loongarch/fpu32.xml \
> ++ loongarch/fpu64.xml \
> ++ loongarch/lbt32.xml \
> ++ loongarch/lbt64.xml \
> ++ loongarch/lsx.xml \
> ++ loongarch/lasx.xml \
> + riscv/32bit-cpu.xml \
> + riscv/32bit-fpu.xml \
> + riscv/64bit-cpu.xml \
> +diff --git gdb-10.2/gdb/features/loongarch/base32.c
> gdb-10.2/gdb/features/loongarch/base32.c
> +new file mode 100644
> +index 0000000..3fb35ef
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/base32.c
> +@@ -0,0 +1,48 @@
> ++/* THIS FILE IS GENERATED. -*- buffer-read-only: t -*- vi:set ro:
> ++ Original: base32.xml */
> ++
> ++#include "gdbsupport/tdesc.h"
> ++
> ++static int
> ++create_feature_loongarch_base32 (struct target_desc *result, long regnum)
> ++{
> ++ struct tdesc_feature *feature;
> ++
> ++ feature = tdesc_create_feature (result, "org.gnu.gdb.loongarch.base");
> ++ tdesc_create_reg (feature, "r0", regnum++, 1, "general", 32, "uint32");
> ++ tdesc_create_reg (feature, "r1", regnum++, 1, "general", 32,
> "code_ptr");
> ++ tdesc_create_reg (feature, "r2", regnum++, 1, "general", 32,
> "data_ptr");
> ++ tdesc_create_reg (feature, "r3", regnum++, 1, "general", 32,
> "data_ptr");
> ++ tdesc_create_reg (feature, "r4", regnum++, 1, "general", 32, "uint32");
> ++ tdesc_create_reg (feature, "r5", regnum++, 1, "general", 32, "uint32");
> ++ tdesc_create_reg (feature, "r6", regnum++, 1, "general", 32, "uint32");
> ++ tdesc_create_reg (feature, "r7", regnum++, 1, "general", 32, "uint32");
> ++ tdesc_create_reg (feature, "r8", regnum++, 1, "general", 32, "uint32");
> ++ tdesc_create_reg (feature, "r9", regnum++, 1, "general", 32, "uint32");
> ++ tdesc_create_reg (feature, "r10", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r11", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r12", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r13", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r14", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r15", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r16", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r17", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r18", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r19", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r20", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r21", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r22", regnum++, 1, "general", 32,
> "data_ptr");
> ++ tdesc_create_reg (feature, "r23", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r24", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r25", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r26", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r27", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r28", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r29", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r30", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "r31", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "orig_a0", regnum++, 1, "general", 32,
> "uint32");
> ++ tdesc_create_reg (feature, "pc", regnum++, 1, "general", 32,
> "code_ptr");
> ++ tdesc_create_reg (feature, "badv", regnum++, 1, "general", 32,
> "code_ptr");
> ++ return regnum;
> ++}
> +diff --git gdb-10.2/gdb/features/loongarch/base32.xml
> gdb-10.2/gdb/features/loongarch/base32.xml
> +new file mode 100644
> +index 0000000..4c44dbe
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/base32.xml
> +@@ -0,0 +1,46 @@
> ++<?xml version="1.0"?>
> ++<!-- Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ Copying and distribution of this file, with or without modification,
> ++ are permitted in any medium without royalty provided the copyright
> ++ notice and this notice are preserved. -->
> ++
> ++<!DOCTYPE feature SYSTEM "gdb-target.dtd">
> ++<feature name="org.gnu.gdb.loongarch.base">
> ++ <reg name="r0" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r1" bitsize="32" type="code_ptr" group="general"/>
> ++ <reg name="r2" bitsize="32" type="data_ptr" group="general"/>
> ++ <reg name="r3" bitsize="32" type="data_ptr" group="general"/>
> ++ <reg name="r4" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r5" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r6" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r7" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r8" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r9" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r10" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r11" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r12" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r13" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r14" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r15" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r16" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r17" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r18" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r19" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r20" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r21" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r22" bitsize="32" type="data_ptr" group="general"/>
> ++ <reg name="r23" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r24" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r25" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r26" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r27" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r28" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r29" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r30" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="r31" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="orig_a0" bitsize="32" type="uint32" group="general"/>
> ++ <reg name="pc" bitsize="32" type="code_ptr" group="general"/>
> ++ <reg name="badv" bitsize="32" type="code_ptr" group="general"/>
> ++</feature>
> +diff --git gdb-10.2/gdb/features/loongarch/base64.c
> gdb-10.2/gdb/features/loongarch/base64.c
> +new file mode 100644
> +index 0000000..d84d425
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/base64.c
> +@@ -0,0 +1,48 @@
> ++/* THIS FILE IS GENERATED. -*- buffer-read-only: t -*- vi:set ro:
> ++ Original: base64.xml */
> ++
> ++#include "gdbsupport/tdesc.h"
> ++
> ++static int
> ++create_feature_loongarch_base64 (struct target_desc *result, long regnum)
> ++{
> ++ struct tdesc_feature *feature;
> ++
> ++ feature = tdesc_create_feature (result, "org.gnu.gdb.loongarch.base");
> ++ tdesc_create_reg (feature, "r0", regnum++, 1, "general", 64, "uint64");
> ++ tdesc_create_reg (feature, "r1", regnum++, 1, "general", 64,
> "code_ptr");
> ++ tdesc_create_reg (feature, "r2", regnum++, 1, "general", 64,
> "data_ptr");
> ++ tdesc_create_reg (feature, "r3", regnum++, 1, "general", 64,
> "data_ptr");
> ++ tdesc_create_reg (feature, "r4", regnum++, 1, "general", 64, "uint64");
> ++ tdesc_create_reg (feature, "r5", regnum++, 1, "general", 64, "uint64");
> ++ tdesc_create_reg (feature, "r6", regnum++, 1, "general", 64, "uint64");
> ++ tdesc_create_reg (feature, "r7", regnum++, 1, "general", 64, "uint64");
> ++ tdesc_create_reg (feature, "r8", regnum++, 1, "general", 64, "uint64");
> ++ tdesc_create_reg (feature, "r9", regnum++, 1, "general", 64, "uint64");
> ++ tdesc_create_reg (feature, "r10", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r11", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r12", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r13", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r14", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r15", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r16", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r17", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r18", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r19", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r20", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r21", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r22", regnum++, 1, "general", 64,
> "data_ptr");
> ++ tdesc_create_reg (feature, "r23", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r24", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r25", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r26", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r27", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r28", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r29", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r30", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "r31", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "orig_a0", regnum++, 1, "general", 64,
> "uint64");
> ++ tdesc_create_reg (feature, "pc", regnum++, 1, "general", 64,
> "code_ptr");
> ++ tdesc_create_reg (feature, "badv", regnum++, 1, "general", 64,
> "code_ptr");
> ++ return regnum;
> ++}
> +diff --git gdb-10.2/gdb/features/loongarch/base64.xml
> gdb-10.2/gdb/features/loongarch/base64.xml
> +new file mode 100644
> +index 0000000..05d766e
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/base64.xml
> +@@ -0,0 +1,46 @@
> ++<?xml version="1.0"?>
> ++<!-- Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ Copying and distribution of this file, with or without modification,
> ++ are permitted in any medium without royalty provided the copyright
> ++ notice and this notice are preserved. -->
> ++
> ++<!DOCTYPE feature SYSTEM "gdb-target.dtd">
> ++<feature name="org.gnu.gdb.loongarch.base">
> ++ <reg name="r0" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r1" bitsize="64" type="code_ptr" group="general"/>
> ++ <reg name="r2" bitsize="64" type="data_ptr" group="general"/>
> ++ <reg name="r3" bitsize="64" type="data_ptr" group="general"/>
> ++ <reg name="r4" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r5" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r6" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r7" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r8" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r9" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r10" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r11" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r12" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r13" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r14" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r15" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r16" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r17" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r18" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r19" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r20" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r21" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r22" bitsize="64" type="data_ptr" group="general"/>
> ++ <reg name="r23" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r24" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r25" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r26" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r27" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r28" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r29" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r30" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="r31" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="orig_a0" bitsize="64" type="uint64" group="general"/>
> ++ <reg name="pc" bitsize="64" type="code_ptr" group="general"/>
> ++ <reg name="badv" bitsize="64" type="code_ptr" group="general"/>
> ++</feature>
> +diff --git gdb-10.2/gdb/features/loongarch/fpu32.c
> gdb-10.2/gdb/features/loongarch/fpu32.c
> +new file mode 100644
> +index 0000000..bf8964a
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/fpu32.c
> +@@ -0,0 +1,54 @@
> ++/* THIS FILE IS GENERATED. -*- buffer-read-only: t -*- vi:set ro:
> ++ Original: fpu32.xml */
> ++
> ++#include "gdbsupport/tdesc.h"
> ++
> ++static int
> ++create_feature_loongarch_fpu32 (struct target_desc *result, long regnum)
> ++{
> ++ struct tdesc_feature *feature;
> ++
> ++ feature = tdesc_create_feature (result, "org.gnu.gdb.loongarch.fpu");
> ++ tdesc_create_reg (feature, "f0", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f1", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f2", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f3", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f4", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f5", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f6", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f7", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f8", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f9", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f10", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f11", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f12", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f13", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f14", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f15", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f16", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f17", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f18", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f19", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f20", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f21", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f22", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f23", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f24", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f25", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f26", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f27", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f28", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f29", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f30", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "f31", regnum++, 1, "float", 32,
> "ieee_single");
> ++ tdesc_create_reg (feature, "fcc0", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc1", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc2", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc3", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc4", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc5", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc6", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc7", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcsr", regnum++, 1, "float", 32, "uint32");
> ++ return regnum;
> ++}
> +diff --git gdb-10.2/gdb/features/loongarch/fpu32.xml
> gdb-10.2/gdb/features/loongarch/fpu32.xml
> +new file mode 100644
> +index 0000000..8421730
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/fpu32.xml
> +@@ -0,0 +1,53 @@
> ++<?xml version="1.0"?>
> ++<!-- Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ Copying and distribution of this file, with or without modification,
> ++ are permitted in any medium without royalty provided the copyright
> ++ notice and this notice are preserved. -->
> ++
> ++<!DOCTYPE feature SYSTEM "gdb-target.dtd">
> ++<feature name="org.gnu.gdb.loongarch.fpu">
> ++
> ++ <reg name="f0" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f1" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f2" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f3" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f4" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f5" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f6" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f7" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f8" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f9" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f10" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f11" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f12" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f13" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f14" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f15" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f16" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f17" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f18" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f19" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f20" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f21" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f22" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f23" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f24" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f25" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f26" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f27" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f28" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f29" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f30" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="f31" bitsize="32" type="ieee_single" group="float"/>
> ++ <reg name="fcc0" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc1" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc2" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc3" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc4" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc5" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc6" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc7" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcsr" bitsize="32" type="uint32" group="float"/>
> ++</feature>
> +diff --git gdb-10.2/gdb/features/loongarch/fpu64.c
> gdb-10.2/gdb/features/loongarch/fpu64.c
> +new file mode 100644
> +index 0000000..f9e24c3
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/fpu64.c
> +@@ -0,0 +1,62 @@
> ++/* THIS FILE IS GENERATED. -*- buffer-read-only: t -*- vi:set ro:
> ++ Original: fpu64.xml */
> ++
> ++#include "gdbsupport/tdesc.h"
> ++
> ++static int
> ++create_feature_loongarch_fpu64 (struct target_desc *result, long regnum)
> ++{
> ++ struct tdesc_feature *feature;
> ++
> ++ feature = tdesc_create_feature (result, "org.gnu.gdb.loongarch.fpu");
> ++ tdesc_type_with_fields *type_with_fields;
> ++ type_with_fields = tdesc_create_union (feature, "fpu64type");
> ++ tdesc_type *field_type;
> ++ field_type = tdesc_named_type (feature, "ieee_single");
> ++ tdesc_add_field (type_with_fields, "f", field_type);
> ++ field_type = tdesc_named_type (feature, "ieee_double");
> ++ tdesc_add_field (type_with_fields, "d", field_type);
> ++
> ++ tdesc_create_reg (feature, "f0", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f1", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f2", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f3", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f4", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f5", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f6", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f7", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f8", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f9", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f10", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f11", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f12", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f13", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f14", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f15", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f16", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f17", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f18", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f19", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f20", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f21", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f22", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f23", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f24", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f25", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f26", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f27", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f28", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f29", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f30", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "f31", regnum++, 1, "float", 64,
> "fpu64type");
> ++ tdesc_create_reg (feature, "fcc0", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc1", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc2", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc3", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc4", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc5", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc6", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcc7", regnum++, 1, "float", 8, "uint8");
> ++ tdesc_create_reg (feature, "fcsr", regnum++, 1, "float", 32, "uint32");
> ++ return regnum;
> ++}
> +diff --git gdb-10.2/gdb/features/loongarch/fpu64.xml
> gdb-10.2/gdb/features/loongarch/fpu64.xml
> +new file mode 100644
> +index 0000000..420be8b
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/fpu64.xml
> +@@ -0,0 +1,58 @@
> ++<?xml version="1.0"?>
> ++<!-- Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ Copying and distribution of this file, with or without modification,
> ++ are permitted in any medium without royalty provided the copyright
> ++ notice and this notice are preserved. -->
> ++
> ++<!DOCTYPE feature SYSTEM "gdb-target.dtd">
> ++<feature name="org.gnu.gdb.loongarch.fpu">
> ++
> ++ <union id="fpu64type">
> ++ <field name="f" type="ieee_single"/>
> ++ <field name="d" type="ieee_double"/>
> ++ </union>
> ++
> ++ <reg name="f0" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f1" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f2" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f3" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f4" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f5" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f6" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f7" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f8" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f9" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f10" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f11" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f12" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f13" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f14" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f15" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f16" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f17" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f18" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f19" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f20" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f21" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f22" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f23" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f24" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f25" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f26" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f27" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f28" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f29" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f30" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="f31" bitsize="64" type="fpu64type" group="float"/>
> ++ <reg name="fcc0" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc1" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc2" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc3" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc4" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc5" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc6" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcc7" bitsize="8" type="uint8" group="float"/>
> ++ <reg name="fcsr" bitsize="32" type="uint32" group="float"/>
> ++</feature>
> +diff --git gdb-10.2/gdb/features/loongarch/lasx.c
> gdb-10.2/gdb/features/loongarch/lasx.c
> +new file mode 100644
> +index 0000000..96e3ea9
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/lasx.c
> +@@ -0,0 +1,80 @@
> ++/* THIS FILE IS GENERATED. -*- buffer-read-only: t -*- vi:set ro:
> ++ Original: lasx.xml */
> ++
> ++#include "gdbsupport/tdesc.h"
> ++
> ++static int
> ++create_feature_loongarch_lasx (struct target_desc *result, long regnum)
> ++{
> ++ struct tdesc_feature *feature;
> ++
> ++ feature = tdesc_create_feature (result, "org.gnu.gdb.loongarch.lasx");
> ++ tdesc_type *element_type;
> ++ element_type = tdesc_named_type (feature, "int8");
> ++ tdesc_create_vector (feature, "v32i8", element_type, 32);
> ++
> ++ element_type = tdesc_named_type (feature, "int16");
> ++ tdesc_create_vector (feature, "v16i16", element_type, 16);
> ++
> ++ element_type = tdesc_named_type (feature, "int32");
> ++ tdesc_create_vector (feature, "v8i32", element_type, 8);
> ++
> ++ element_type = tdesc_named_type (feature, "int64");
> ++ tdesc_create_vector (feature, "v4i64", element_type, 4);
> ++
> ++ element_type = tdesc_named_type (feature, "ieee_single");
> ++ tdesc_create_vector (feature, "v8f32", element_type, 8);
> ++
> ++ element_type = tdesc_named_type (feature, "ieee_double");
> ++ tdesc_create_vector (feature, "v4f64", element_type, 4);
> ++
> ++ tdesc_type_with_fields *type_with_fields;
> ++ type_with_fields = tdesc_create_union (feature, "lasxv");
> ++ tdesc_type *field_type;
> ++ field_type = tdesc_named_type (feature, "v32i8");
> ++ tdesc_add_field (type_with_fields, "v32i8", field_type);
> ++ field_type = tdesc_named_type (feature, "v16i16");
> ++ tdesc_add_field (type_with_fields, "v16i16", field_type);
> ++ field_type = tdesc_named_type (feature, "v8i32");
> ++ tdesc_add_field (type_with_fields, "v8i32", field_type);
> ++ field_type = tdesc_named_type (feature, "v4i64");
> ++ tdesc_add_field (type_with_fields, "v4i64", field_type);
> ++ field_type = tdesc_named_type (feature, "v8f32");
> ++ tdesc_add_field (type_with_fields, "v8f32", field_type);
> ++ field_type = tdesc_named_type (feature, "v4f64");
> ++ tdesc_add_field (type_with_fields, "v4f64", field_type);
> ++
> ++ tdesc_create_reg (feature, "xr0", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr1", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr2", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr3", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr4", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr5", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr6", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr7", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr8", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr9", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr10", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr11", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr12", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr13", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr14", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr15", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr16", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr17", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr18", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr19", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr20", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr21", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr22", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr23", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr24", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr25", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr26", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr27", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr28", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr29", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr30", regnum++, 1, "lasx", 256, "lasxv");
> ++ tdesc_create_reg (feature, "xr31", regnum++, 1, "lasx", 256, "lasxv");
> ++ return regnum;
> ++}
> +diff --git gdb-10.2/gdb/features/loongarch/lasx.xml
> gdb-10.2/gdb/features/loongarch/lasx.xml
> +new file mode 100644
> +index 0000000..6f73df0
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/lasx.xml
> +@@ -0,0 +1,59 @@
> ++<?xml version="1.0"?>
> ++<!-- Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ Copying and distribution of this file, with or without modification,
> ++ are permitted in any medium without royalty provided the copyright
> ++ notice and this notice are preserved. -->
> ++
> ++<!DOCTYPE feature SYSTEM "gdb-target.dtd">
> ++<feature name="org.gnu.gdb.loongarch.lasx">
> ++ <vector id="v32i8" type="int8" count="32"/>
> ++ <vector id="v16i16" type="int16" count="16"/>
> ++ <vector id="v8i32" type="int32" count="8"/>
> ++ <vector id="v4i64" type="int64" count="4"/>
> ++ <vector id="v8f32" type="ieee_single" count="8"/>
> ++ <vector id="v4f64" type="ieee_double" count="4"/>
> ++
> ++ <union id="lasxv">
> ++ <field name="v32i8" type="v32i8"/>
> ++ <field name="v16i16" type="v16i16"/>
> ++ <field name="v8i32" type="v8i32"/>
> ++ <field name="v4i64" type="v4i64"/>
> ++ <field name="v8f32" type="v8f32"/>
> ++ <field name="v4f64" type="v4f64"/>
> ++ </union>
> ++
> ++ <reg name="xr0" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr1" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr2" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr3" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr4" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr5" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr6" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr7" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr8" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr9" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr10" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr11" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr12" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr13" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr14" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr15" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr16" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr17" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr18" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr19" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr20" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr21" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr22" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr23" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr24" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr25" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr26" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr27" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr28" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr29" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr30" bitsize="256" type="lasxv" group="lasx"/>
> ++ <reg name="xr31" bitsize="256" type="lasxv" group="lasx"/>
> ++</feature>
> +diff --git gdb-10.2/gdb/features/loongarch/lbt32.c
> gdb-10.2/gdb/features/loongarch/lbt32.c
> +new file mode 100644
> +index 0000000..d245c75
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/lbt32.c
> +@@ -0,0 +1,19 @@
> ++/* THIS FILE IS GENERATED. -*- buffer-read-only: t -*- vi:set ro:
> ++ Original: lbt32.xml */
> ++
> ++#include "gdbsupport/tdesc.h"
> ++
> ++static int
> ++create_feature_loongarch_lbt32 (struct target_desc *result, long regnum)
> ++{
> ++ struct tdesc_feature *feature;
> ++
> ++ feature = tdesc_create_feature (result, "org.gnu.gdb.loongarch.lbt");
> ++ tdesc_create_reg (feature, "scr0", regnum++, 1, "lbt", 32, "uint32");
> ++ tdesc_create_reg (feature, "scr1", regnum++, 1, "lbt", 32, "uint32");
> ++ tdesc_create_reg (feature, "scr2", regnum++, 1, "lbt", 32, "uint32");
> ++ tdesc_create_reg (feature, "scr3", regnum++, 1, "lbt", 32, "uint32");
> ++ tdesc_create_reg (feature, "EFLAG", regnum++, 1, "lbt", 32, "uint32");
> ++ tdesc_create_reg (feature, "x86_top", regnum++, 1, "lbt", 8, "uint8");
> ++ return regnum;
> ++}
> +diff --git gdb-10.2/gdb/features/loongarch/lbt32.xml
> gdb-10.2/gdb/features/loongarch/lbt32.xml
> +new file mode 100644
> +index 0000000..1c0133e
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/lbt32.xml
> +@@ -0,0 +1,17 @@
> ++<?xml version="1.0"?>
> ++<!-- Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ Copying and distribution of this file, with or without modification,
> ++ are permitted in any medium without royalty provided the copyright
> ++ notice and this notice are preserved. -->
> ++
> ++<!DOCTYPE feature SYSTEM "gdb-target.dtd">
> ++<feature name="org.gnu.gdb.loongarch.lbt">
> ++ <reg name="scr0" bitsize="32" type="uint32" group="lbt"/>
> ++ <reg name="scr1" bitsize="32" type="uint32" group="lbt"/>
> ++ <reg name="scr2" bitsize="32" type="uint32" group="lbt"/>
> ++ <reg name="scr3" bitsize="32" type="uint32" group="lbt"/>
> ++ <reg name="EFLAG" bitsize="32" type="uint32" group="lbt"/>
> ++ <reg name="x86_top" bitsize="8" type="uint8" group="lbt"/>
> ++</feature>
> +diff --git gdb-10.2/gdb/features/loongarch/lbt64.c
> gdb-10.2/gdb/features/loongarch/lbt64.c
> +new file mode 100644
> +index 0000000..ecef330
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/lbt64.c
> +@@ -0,0 +1,19 @@
> ++/* THIS FILE IS GENERATED. -*- buffer-read-only: t -*- vi:set ro:
> ++ Original: lbt64.xml */
> ++
> ++#include "gdbsupport/tdesc.h"
> ++
> ++static int
> ++create_feature_loongarch_lbt64 (struct target_desc *result, long regnum)
> ++{
> ++ struct tdesc_feature *feature;
> ++
> ++ feature = tdesc_create_feature (result, "org.gnu.gdb.loongarch.lbt");
> ++ tdesc_create_reg (feature, "scr0", regnum++, 1, "lbt", 64, "uint64");
> ++ tdesc_create_reg (feature, "scr1", regnum++, 1, "lbt", 64, "uint64");
> ++ tdesc_create_reg (feature, "scr2", regnum++, 1, "lbt", 64, "uint64");
> ++ tdesc_create_reg (feature, "scr3", regnum++, 1, "lbt", 64, "uint64");
> ++ tdesc_create_reg (feature, "EFLAG", regnum++, 1, "lbt", 32, "uint32");
> ++ tdesc_create_reg (feature, "x86_top", regnum++, 1, "lbt", 8, "uint8");
> ++ return regnum;
> ++}
> +diff --git gdb-10.2/gdb/features/loongarch/lbt64.xml
> gdb-10.2/gdb/features/loongarch/lbt64.xml
> +new file mode 100644
> +index 0000000..1df26f5
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/lbt64.xml
> +@@ -0,0 +1,17 @@
> ++<?xml version="1.0"?>
> ++<!-- Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ Copying and distribution of this file, with or without modification,
> ++ are permitted in any medium without royalty provided the copyright
> ++ notice and this notice are preserved. -->
> ++
> ++<!DOCTYPE feature SYSTEM "gdb-target.dtd">
> ++<feature name="org.gnu.gdb.loongarch.lbt">
> ++ <reg name="scr0" bitsize="64" type="uint64" group="lbt"/>
> ++ <reg name="scr1" bitsize="64" type="uint64" group="lbt"/>
> ++ <reg name="scr2" bitsize="64" type="uint64" group="lbt"/>
> ++ <reg name="scr3" bitsize="64" type="uint64" group="lbt"/>
> ++ <reg name="EFLAG" bitsize="32" type="uint32" group="lbt"/>
> ++ <reg name="x86_top" bitsize="8" type="uint8" group="lbt"/>
> ++</feature>
> +diff --git gdb-10.2/gdb/features/loongarch/lsx.c
> gdb-10.2/gdb/features/loongarch/lsx.c
> +new file mode 100644
> +index 0000000..dd253f3
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/lsx.c
> +@@ -0,0 +1,80 @@
> ++/* THIS FILE IS GENERATED. -*- buffer-read-only: t -*- vi:set ro:
> ++ Original: lsx.xml */
> ++
> ++#include "gdbsupport/tdesc.h"
> ++
> ++static int
> ++create_feature_loongarch_lsx (struct target_desc *result, long regnum)
> ++{
> ++ struct tdesc_feature *feature;
> ++
> ++ feature = tdesc_create_feature (result, "org.gnu.gdb.loongarch.lsx");
> ++ tdesc_type *element_type;
> ++ element_type = tdesc_named_type (feature, "int8");
> ++ tdesc_create_vector (feature, "v16i8", element_type, 16);
> ++
> ++ element_type = tdesc_named_type (feature, "int16");
> ++ tdesc_create_vector (feature, "v8i16", element_type, 8);
> ++
> ++ element_type = tdesc_named_type (feature, "int32");
> ++ tdesc_create_vector (feature, "v4i32", element_type, 4);
> ++
> ++ element_type = tdesc_named_type (feature, "int64");
> ++ tdesc_create_vector (feature, "v2i64", element_type, 2);
> ++
> ++ element_type = tdesc_named_type (feature, "ieee_single");
> ++ tdesc_create_vector (feature, "v4f32", element_type, 4);
> ++
> ++ element_type = tdesc_named_type (feature, "ieee_double");
> ++ tdesc_create_vector (feature, "v2f64", element_type, 2);
> ++
> ++ tdesc_type_with_fields *type_with_fields;
> ++ type_with_fields = tdesc_create_union (feature, "lsxv");
> ++ tdesc_type *field_type;
> ++ field_type = tdesc_named_type (feature, "v16i8");
> ++ tdesc_add_field (type_with_fields, "v16i8", field_type);
> ++ field_type = tdesc_named_type (feature, "v8i16");
> ++ tdesc_add_field (type_with_fields, "v8i16", field_type);
> ++ field_type = tdesc_named_type (feature, "v4i32");
> ++ tdesc_add_field (type_with_fields, "v4i32", field_type);
> ++ field_type = tdesc_named_type (feature, "v2i64");
> ++ tdesc_add_field (type_with_fields, "v2i64", field_type);
> ++ field_type = tdesc_named_type (feature, "v4f32");
> ++ tdesc_add_field (type_with_fields, "v4f32", field_type);
> ++ field_type = tdesc_named_type (feature, "v2f64");
> ++ tdesc_add_field (type_with_fields, "v2f64", field_type);
> ++
> ++ tdesc_create_reg (feature, "vr0", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr1", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr2", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr3", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr4", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr5", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr6", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr7", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr8", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr9", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr10", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr11", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr12", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr13", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr14", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr15", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr16", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr17", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr18", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr19", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr20", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr21", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr22", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr23", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr24", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr25", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr26", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr27", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr28", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr29", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr30", regnum++, 1, "lsx", 128, "lsxv");
> ++ tdesc_create_reg (feature, "vr31", regnum++, 1, "lsx", 128, "lsxv");
> ++ return regnum;
> ++}
> +diff --git gdb-10.2/gdb/features/loongarch/lsx.xml
> gdb-10.2/gdb/features/loongarch/lsx.xml
> +new file mode 100644
> +index 0000000..a0a6ba4
> +--- /dev/null
> ++++ gdb-10.2/gdb/features/loongarch/lsx.xml
> +@@ -0,0 +1,59 @@
> ++<?xml version="1.0"?>
> ++<!-- Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ Copying and distribution of this file, with or without modification,
> ++ are permitted in any medium without royalty provided the copyright
> ++ notice and this notice are preserved. -->
> ++
> ++<!DOCTYPE feature SYSTEM "gdb-target.dtd">
> ++<feature name="org.gnu.gdb.loongarch.lsx">
> ++ <vector id="v16i8" type="int8" count="16"/>
> ++ <vector id="v8i16" type="int16" count="8"/>
> ++ <vector id="v4i32" type="int32" count="4"/>
> ++ <vector id="v2i64" type="int64" count="2"/>
> ++ <vector id="v4f32" type="ieee_single" count="4"/>
> ++ <vector id="v2f64" type="ieee_double" count="2"/>
> ++
> ++ <union id="lsxv">
> ++ <field name="v16i8" type="v16i8"/>
> ++ <field name="v8i16" type="v8i16"/>
> ++ <field name="v4i32" type="v4i32"/>
> ++ <field name="v2i64" type="v2i64"/>
> ++ <field name="v4f32" type="v4f32"/>
> ++ <field name="v2f64" type="v2f64"/>
> ++ </union>
> ++
> ++ <reg name="vr0" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr1" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr2" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr3" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr4" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr5" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr6" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr7" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr8" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr9" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr10" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr11" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr12" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr13" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr14" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr15" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr16" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr17" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr18" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr19" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr20" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr21" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr22" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr23" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr24" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr25" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr26" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr27" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr28" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr29" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr30" bitsize="128" type="lsxv" group="lsx"/>
> ++ <reg name="vr31" bitsize="128" type="lsxv" group="lsx"/>
> ++</feature>
> +diff --git gdb-10.2/gdb/loongarch-linux-nat.c
> gdb-10.2/gdb/loongarch-linux-nat.c
> +new file mode 100644
> +index 0000000..e79b660
> +--- /dev/null
> ++++ gdb-10.2/gdb/loongarch-linux-nat.c
> +@@ -0,0 +1,878 @@
> ++/* Target-dependent code for GNU/Linux LoongArch.
> ++
> ++ Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of GDB.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program. If not, see <http://www.gnu.org/licenses/>.
> */
> ++
> ++#include "defs.h"
> ++#include "command.h"
> ++#include "gdbcmd.h"
> ++#include "inferior.h"
> ++#include "loongarch-tdep.h"
> ++#include "target.h"
> ++#include "regcache.h"
> ++#include "linux-nat-trad.h"
> ++#include "loongarch-linux-tdep.h"
> ++#include "target-descriptions.h"
> ++
> ++#include "gdb_proc_service.h"
> ++#include "gregset.h"
> ++
> ++#include "nat/gdb_ptrace.h"
> ++#include <asm/ptrace.h>
> ++#include "inf-ptrace.h"
> ++#include "arch/loongarch-linux-nat.h"
> ++#include "elf/common.h"
> ++
> ++#include "nat/loongarch-linux-watch.h"
> ++
> ++class loongarch_linux_nat_target final : public linux_nat_trad_target
> ++{
> ++public:
> ++ void fetch_registers (struct regcache *, int) override;
> ++ void store_registers (struct regcache *, int) override;
> ++
> ++ void close () override;
> ++
> ++ int can_use_hw_breakpoint (enum bptype, int, int) override;
> ++
> ++ int insert_hw_breakpoint (struct gdbarch *,
> ++ struct bp_target_info *) override;
> ++
> ++ int remove_hw_breakpoint (struct gdbarch *,
> ++ struct bp_target_info *) override;
> ++
> ++ int remove_watchpoint (CORE_ADDR, int, enum target_hw_bp_type,
> ++ struct expression *) override;
> ++
> ++ int insert_watchpoint (CORE_ADDR, int, enum target_hw_bp_type,
> ++ struct expression *) override;
> ++
> ++ bool stopped_by_watchpoint () override;
> ++
> ++ bool stopped_data_address (CORE_ADDR *) override;
> ++
> ++ int region_ok_for_hw_watchpoint (CORE_ADDR, int) override;
> ++
> ++ const struct target_desc *read_description () override;
> ++
> ++ enum target_xfer_status xfer_partial (enum target_object object,
> ++ const char *annex, gdb_byte
> *readbuf,
> ++ const gdb_byte *writebuf,
> ++ ULONGEST offset, ULONGEST len,
> ++ ULONGEST *xfered_len) override;
> ++
> ++protected:
> ++ CORE_ADDR register_u_offset (struct gdbarch *gdbarch, int regno,
> ++ int store_p) override;
> ++
> ++ void low_new_thread (struct lwp_info *lp) override;
> ++};
> ++
> ++const struct target_desc *
> ++loongarch_linux_nat_target::read_description ()
> ++{
> ++ return loongarch_linux_read_description_runtime (inferior_ptid.pid ());
> ++}
> ++
> ++/* Fill GDB's register array with the general-purpose, orig_a0, pc and
> badv
> ++ register values from the current thread. */
> ++
> ++static void
> ++fetch_gregs_from_thread (struct regcache *regcache, int regno, pid_t tid)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ loongarch_elf_gregset_t regset;
> ++
> ++ if ((regno == -1) || (regs->r <= regno && regno < regs->r + 32) ||
> ++ (regs->orig_a0 == regno) || (regs->pc == regno) ||
> ++ (regs->badv == regno))
> ++ {
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_PRSTATUS, (long) &iovec) < 0)
> ++ perror_with_name (_("Couldn't get NT_PRSTATUS registers"));
> ++ else
> ++ loongarch_elf_gregset.supply_regset (NULL, regcache, regno,
> ®set,
> ++ sizeof (regset));
> ++ }
> ++}
> ++
> ++/* Store to the current thread the valid general-purpose, orig_a0, pc
> and badv
> ++ register values in the GDB's register array. */
> ++
> ++static void
> ++store_gregs_to_thread (struct regcache *regcache, int regno, pid_t tid)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ loongarch_elf_gregset_t regset;
> ++
> ++ if ((regno == -1) || (regs->r <= regno && regno < regs->r + 32) ||
> ++ (regs->orig_a0 == regno) || (regs->pc == regno) ||
> ++ (regs->badv == regno))
> ++ {
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_PRSTATUS, (long) &iovec) < 0)
> ++ perror_with_name (_("Couldn't get NT_PRSTATUS registers"));
> ++ else
> ++ {
> ++ loongarch_elf_gregset.collect_regset (NULL, regcache, regno,
> ®set,
> ++ sizeof (regset));
> ++ if (ptrace (PTRACE_SETREGSET, tid, NT_PRSTATUS, (long) &iovec) < 0)
> ++ perror_with_name (_("Couldn't set NT_PRSTATUS registers"));
> ++ }
> ++ }
> ++}
> ++
> ++/* Fill GDB's register array with the fp, fcc and fcsr
> ++ register values from the current thread. */
> ++
> ++static void
> ++fetch_fpregs_from_thread (struct regcache *regcache, int regno, pid_t
> tid)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ loongarch_elf_fpregset_t regset;
> ++
> ++ if ((regno == -1) || (regs->f <= regno && regno < regs->f + 32) ||
> ++ (regs->fcc <= regno && regno < regs->fcc + 8) || (regs->fcsr ==
> regno))
> ++ {
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_FPREGSET, (long) &iovec) < 0)
> ++ perror_with_name (_("Couldn't get NT_FPREGSET registers"));
> ++ else
> ++ loongarch_elf_fpregset.supply_regset (NULL, regcache, regno,
> ®set,
> ++ sizeof (regset));
> ++ }
> ++}
> ++
> ++/* Store to the current thread the valid fp, fcc and fcsr
> ++ register values in the GDB's register array. */
> ++
> ++static void
> ++store_fpregs_to_thread (struct regcache *regcache, int regno, pid_t tid)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ loongarch_elf_fpregset_t regset;
> ++
> ++ if ((regno == -1) || (regs->f <= regno && regno < regs->f + 32) ||
> ++ (regs->fcc <= regno && regno < regs->fcc + 8) || (regs->fcsr ==
> regno))
> ++ {
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_FPREGSET, (long) &iovec) < 0)
> ++ perror_with_name (_("Couldn't get NT_FPREGSET registers"));
> ++ else
> ++ {
> ++ loongarch_elf_fpregset.collect_regset (NULL, regcache, regno,
> ®set,
> ++ sizeof (regset));
> ++ if (ptrace (PTRACE_SETREGSET, tid, NT_FPREGSET, (long) &iovec) < 0)
> ++ perror_with_name (_("Couldn't set NT_FPREGSET registers"));
> ++ }
> ++ }
> ++}
> ++
> ++/* Fill GDB's register array with the Binary Translation
> ++ register values from the current thread. */
> ++
> ++static void
> ++fetch_lbtregs_from_thread (struct regcache *regcache, int regno, pid_t
> tid)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ loongarch_elf_lbtregset_t regset;
> ++
> ++ if ((regno == -1) || (regs->scr <= regno && regno < regs->scr + 4) ||
> ++ (regs->EFLAG == regno) || (regs->x86_top == regno))
> ++ {
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_LARCH_LBT, (long) &iovec) < 0)
> ++ perror_with_name (_("Couldn't get LBT registers"));
> ++ else
> ++ loongarch_elf_lbtregset.supply_regset (NULL, regcache, regno,
> ®set,
> ++ sizeof (regset));
> ++ }
> ++}
> ++
> ++/* Store to the current thread the valid Binary Translation
> ++ register values in the GDB's register array. */
> ++
> ++static void
> ++store_lbtregs_to_thread (struct regcache *regcache, int regno, pid_t tid)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ loongarch_elf_lbtregset_t regset;
> ++
> ++ if ((regno == -1) || (regs->scr <= regno && regno < regs->scr + 4) ||
> ++ (regs->EFLAG == regno) || (regs->x86_top == regno))
> ++ {
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_LARCH_LBT, (long) &iovec) < 0)
> ++ perror_with_name (_("Couldn't get LBT registers"));
> ++ else
> ++ {
> ++ loongarch_elf_lbtregset.collect_regset (NULL, regcache, regno,
> ®set,
> ++ sizeof (regset));
> ++ if (ptrace (PTRACE_SETREGSET, tid, NT_LARCH_LBT, (long) &iovec) <
> 0)
> ++ perror_with_name (_("Couldn't set LBT registers"));
> ++ }
> ++ }
> ++}
> ++
> ++/* Fill GDB's register array with the SIMD eXtension
> ++ register values from the current thread. */
> ++
> ++static void
> ++fetch_lsxregs_from_thread (struct regcache *regcache, int regno, pid_t
> tid)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ loongarch_elf_lsxregset_t regset;
> ++
> ++ if ((regno == -1) || (regs->vr <= regno && regno < regs->vr + 32))
> ++ {
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_LARCH_LSX, (long) &iovec) < 0)
> ++ perror_with_name (_("Couldn't get LSX registers"));
> ++ else
> ++ loongarch_elf_lsxregset.supply_regset (NULL, regcache, regno,
> ®set,
> ++ sizeof (regset));
> ++ }
> ++}
> ++
> ++/* Store to the current thread the valid SIMD eXtension
> ++ register values in the GDB's register array. */
> ++
> ++static void
> ++store_lsxregs_to_thread (struct regcache *regcache, int regno, pid_t tid)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ loongarch_elf_lsxregset_t regset;
> ++
> ++ if ((regno == -1) || (regs->vr <= regno && regno < regs->vr + 32))
> ++ {
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_LARCH_LSX, (long) &iovec) < 0)
> ++ perror_with_name (_("Couldn't get LSX registers"));
> ++ else
> ++ {
> ++ loongarch_elf_lsxregset.collect_regset (NULL, regcache, regno,
> ®set,
> ++ sizeof (regset));
> ++ if (ptrace (PTRACE_SETREGSET, tid, NT_LARCH_LSX, (long) &iovec) <
> 0)
> ++ perror_with_name (_("Couldn't set LSX registers"));
> ++ }
> ++ }
> ++}
> ++
> ++/* Fill GDB's register array with the Advanced SIMD eXtension
> ++ register values from the current thread. */
> ++
> ++static void
> ++fetch_lasxregs_from_thread (struct regcache *regcache, int regno, pid_t
> tid)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ loongarch_elf_lasxregset_t regset;
> ++
> ++ if ((regno == -1) || (regs->xr <= regno && regno < regs->xr + 32))
> ++ {
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_LARCH_LASX, (long) &iovec) < 0)
> ++ perror_with_name (_("Couldn't get LASX registers"));
> ++ else
> ++ loongarch_elf_lasxregset.supply_regset (NULL, regcache, regno,
> ®set,
> ++ sizeof (regset));
> ++ }
> ++}
> ++
> ++/* Store to the current thread the valid Advanced SIMD eXtension
> ++ register values in the GDB's register array. */
> ++
> ++static void
> ++store_lasxregs_to_thread (struct regcache *regcache, int regno, pid_t
> tid)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ loongarch_elf_lasxregset_t regset;
> ++
> ++ if ((regno == -1) || (regs->xr <= regno && regno < regs->xr + 32))
> ++ {
> ++ struct iovec iovec = { .iov_base = ®set, .iov_len = sizeof
> (regset) };
> ++
> ++ if (ptrace (PTRACE_GETREGSET, tid, NT_LARCH_LSX, (long) &iovec) < 0)
> ++ perror_with_name (_("Couldn't get LSX registers"));
> ++ else
> ++ {
> ++ loongarch_elf_lsxregset.collect_regset (NULL, regcache, regno,
> ®set,
> ++ sizeof (regset));
> ++ if (ptrace (PTRACE_SETREGSET, tid, NT_LARCH_LSX, (long) &iovec) <
> 0)
> ++ perror_with_name (_("Couldn't set LSX registers"));
> ++ }
> ++ }
> ++}
> ++
> ++/* Implement the "fetch_registers" target_ops method. */
> ++
> ++void
> ++loongarch_linux_nat_target::fetch_registers (struct regcache *regcache,
> ++ int regno)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ pid_t tid = get_ptrace_pid (regcache->ptid ());
> ++
> ++ fetch_gregs_from_thread(regcache, regno, tid);
> ++
> ++ if (regs->f > 0)
> ++ fetch_fpregs_from_thread(regcache, regno, tid);
> ++
> ++ if (regs->scr > 0)
> ++ fetch_lbtregs_from_thread(regcache, regno, tid);
> ++
> ++ if (regs->vr > 0)
> ++ fetch_lsxregs_from_thread(regcache, regno, tid);
> ++
> ++ if (regs->xr > 0)
> ++ fetch_lasxregs_from_thread(regcache, regno, tid);
> ++}
> ++
> ++/* Implement the "store_registers" target_ops method. */
> ++
> ++void
> ++loongarch_linux_nat_target::store_registers (struct regcache *regcache,
> ++ int regno)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ pid_t tid = get_ptrace_pid (regcache->ptid ());
> ++
> ++ store_gregs_to_thread(regcache, regno, tid);
> ++
> ++ if (regs->f > 0)
> ++ store_fpregs_to_thread(regcache, regno, tid);
> ++
> ++ if (regs->scr > 0)
> ++ store_lbtregs_to_thread(regcache, regno, tid);
> ++
> ++ if (regs->vr > 0)
> ++ store_lsxregs_to_thread(regcache, regno, tid);
> ++
> ++ if (regs->xr > 0)
> ++ store_lasxregs_to_thread(regcache, regno, tid);
> ++}
> ++
> ++/* Return the address in the core dump or inferior of register REGNO. */
> ++
> ++CORE_ADDR
> ++loongarch_linux_nat_target::register_u_offset (struct gdbarch *gdbarch,
> ++ int regno, int store_p)
> ++{
> ++ auto regs = &gdbarch_tdep (gdbarch)->regs;
> ++ /* according to <asm/ptrace.h> */
> ++ if (0 <= regs->r && regs->r <= regno && regno < regs->r + GPR_NUM)
> ++ return GPR_BASE + regno - regs->r;
> ++ else if (regs->pc == regno)
> ++ return PC;
> ++ return -1;
> ++}
> ++
> ++/* Implement the to_xfer_partial target_ops method. */
> ++
> ++enum target_xfer_status
> ++loongarch_linux_nat_target::xfer_partial (enum target_object object,
> ++ const char *annex, gdb_byte
> *readbuf,
> ++ const gdb_byte *writebuf,
> ++ ULONGEST offset, ULONGEST len,
> ++ ULONGEST *xfered_len)
> ++{
> ++ pid_t pid = inferior_ptid.pid ();
> ++
> ++ if (object != TARGET_OBJECT_LARCH)
> ++ return linux_nat_trad_target::xfer_partial (
> ++ object, annex, readbuf, writebuf, offset, len, xfered_len);
> ++
> ++ if (strcmp (annex, "cpucfg") == 0)
> ++ {
> ++ if (writebuf)
> ++ return TARGET_XFER_E_IO;
> ++ if (offset % 4 != 0 || offset % 4 != 0)
> ++ return TARGET_XFER_E_IO;
> ++ char t_buf[offset + len];
> ++ struct iovec iovec = { .iov_base = &t_buf, .iov_len = sizeof
> (t_buf) };
> ++ if (ptrace (PTRACE_GETREGSET, pid, NT_LARCH_CPUCFG, &iovec) < 0)
> ++ {
> ++ size_t i;
> ++ for (i = offset / 4; i < (offset + len) / 4 + 1; i++)
> ++ ((uint32_t *) t_buf)[i] = loongarch_cpucfg (i);
> ++ }
> ++ memcpy (readbuf, t_buf + offset, len);
> ++ *xfered_len = len;
> ++ return TARGET_XFER_OK;
> ++ }
> ++
> ++ return TARGET_XFER_E_IO;
> ++}
> ++
> ++static loongarch_linux_nat_target the_loongarch_linux_nat_target;
> ++
> ++/* Wrapper functions. These are only used by libthread_db. */
> ++
> ++void
> ++supply_gregset (struct regcache *regcache,
> ++ const gdb_gregset_t /* elf_gregset_t */ *gregset)
> ++{
> ++ loongarch_elf_gregset.supply_regset (NULL, regcache, -1, gregset,
> ++ sizeof (gdb_gregset_t));
> ++}
> ++
> ++void
> ++fill_gregset (const struct regcache *regcache,
> ++ gdb_gregset_t /* elf_gregset_t */ *gregset, int regno)
> ++{
> ++ loongarch_elf_gregset.collect_regset (NULL, regcache, regno, gregset,
> ++ sizeof (gdb_gregset_t));
> ++}
> ++
> ++void
> ++supply_fpregset (struct regcache *regcache,
> ++ const gdb_fpregset_t /* elf_fpregset_t */ *fpregset)
> ++{
> ++ loongarch_elf_fpregset.supply_regset (NULL, regcache, -1, fpregset,
> ++ sizeof (gdb_fpregset_t));
> ++}
> ++
> ++void
> ++fill_fpregset (const struct regcache *regcache,
> ++ gdb_fpregset_t /* elf_fpregset_t */ *fpregset, int regno)
> ++{
> ++ loongarch_elf_fpregset.collect_regset (NULL, regcache, regno, fpregset,
> ++ sizeof (gdb_fpregset_t));
> ++}
> ++
> ++/* -1 if the kernel and/or CPU do not support watch registers.
> ++ 1 if watch_readback is valid and we can read style, num_valid
> ++ and the masks.
> ++ 0 if we need to read the watch_readback. */
> ++
> ++static int watch_readback_valid;
> ++
> ++/* Cached watch register read values. */
> ++
> ++static struct pt_watch_regs watch_readback =
> ++{
> ++ .max_valid = MAX_DEBUG_REGISTER,
> ++};
> ++
> ++static struct loongarch_watchpoint *current_watches;
> ++
> ++/* The current set of watch register values for writing the
> ++ registers. */
> ++
> ++static struct pt_watch_regs watch_mirror =
> ++{
> ++ .max_valid = MAX_DEBUG_REGISTER,
> ++};
> ++
> ++static void
> ++loongarch_show_dr (const char *func, CORE_ADDR addr, int len,
> ++ enum target_hw_bp_type type)
> ++{
> ++ int i;
> ++
> ++ puts_unfiltered (func);
> ++ if (addr || len)
> ++ printf_unfiltered (
> ++ " (addr=%s, len=%d, type=%s)", paddress (target_gdbarch (), addr),
> len,
> ++ type == hw_write
> ++ ? "data-write"
> ++ : (type == hw_read
> ++ ? "data-read"
> ++ : (type == hw_access ? "data-read/write"
> ++ : (type == hw_execute ? "instruction-execute"
> ++ : "??unknown??"))));
> ++ puts_unfiltered (":\n");
> ++
> ++ for (i = 0; i < MAX_DEBUG_REGISTER; i++)
> ++ printf_unfiltered (
> ++ "\tDR%d: addr=%s, mask=%s\n", i,
> ++ paddress (target_gdbarch (),
> ++ loongarch_linux_watch_get_addr (&watch_mirror, i)),
> ++ paddress (target_gdbarch (),
> ++ loongarch_linux_watch_get_mask (&watch_mirror, i)));
> ++}
> ++
> ++/* Target to_can_use_hw_breakpoint implementation. Return 1 if we can
> ++ handle the specified watch type. */
> ++
> ++int
> ++loongarch_linux_nat_target::can_use_hw_breakpoint (enum bptype type, int
> cnt,
> ++ int ot)
> ++{
> ++ int i;
> ++ uint32_t wanted_mask, irw_mask;
> ++ long lwp = inferior_ptid.lwp ();
> ++
> ++ if (!loongarch_linux_read_watch_registers (lwp, &watch_readback,
> ++ &watch_readback_valid, 0))
> ++ return 0;
> ++
> ++ switch (type)
> ++ {
> ++ case bp_hardware_watchpoint:
> ++ wanted_mask = W_MASK;
> ++ break;
> ++ case bp_read_watchpoint:
> ++ wanted_mask = R_MASK;
> ++ break;
> ++ case bp_access_watchpoint:
> ++ wanted_mask = R_MASK | W_MASK;
> ++ break;
> ++ case bp_hardware_breakpoint:
> ++ wanted_mask = I_MASK;
> ++ break;
> ++ default:
> ++ return 0;
> ++ }
> ++
> ++ for (i = 0; i < loongarch_linux_watch_get_num_valid (&watch_readback)
> && cnt;
> ++ i++)
> ++ {
> ++ irw_mask = loongarch_linux_watch_get_irwmask (&watch_readback, i);
> ++ if ((irw_mask & wanted_mask) == wanted_mask)
> ++ cnt--;
> ++ }
> ++ return (cnt == 0) ? 1 : 0;
> ++}
> ++
> ++/* Target to_stopped_by_watchpoint implementation. Return 1 if
> ++ stopped by watchpoint. The watchhi R and W bits indicate the watch
> ++ register triggered. */
> ++
> ++bool
> ++loongarch_linux_nat_target::stopped_by_watchpoint ()
> ++{
> ++ int n;
> ++ int num_valid;
> ++
> ++ if (!loongarch_linux_read_watch_registers (
> ++ inferior_ptid.lwp (), &watch_readback, &watch_readback_valid, 1))
> ++ return false;
> ++
> ++ num_valid = loongarch_linux_watch_get_num_valid (&watch_readback);
> ++
> ++ for (n = 0; n < MAX_DEBUG_REGISTER && n < num_valid; n++)
> ++ if (loongarch_linux_watch_get_irwstat (&watch_readback, n)
> ++ & (R_MASK | W_MASK))
> ++ return true;
> ++
> ++ return false;
> ++}
> ++
> ++/* Target to_stopped_data_address implementation. Set the address
> ++ where the watch triggered (if known). Return 1 if the address was
> ++ known. */
> ++
> ++bool
> ++loongarch_linux_nat_target::stopped_data_address (CORE_ADDR *paddr)
> ++{
> ++ int num_valid, n;
> ++ if (!loongarch_linux_read_watch_registers (
> ++ inferior_ptid.lwp (), &watch_readback, &watch_readback_valid, 1))
> ++ return false;
> ++
> ++ num_valid = loongarch_linux_watch_get_num_valid (&watch_readback);
> ++
> ++ for (n = 0; n < MAX_DEBUG_REGISTER && n < num_valid; n++)
> ++ if (loongarch_linux_watch_get_irwstat (&watch_readback, n)
> ++ & (R_MASK | W_MASK))
> ++ {
> ++ CORE_ADDR t_addr, t_mask;
> ++ int t_irw;
> ++ struct loongarch_watchpoint *watch;
> ++
> ++ t_addr = loongarch_linux_watch_get_addr (&watch_readback, n);
> ++ t_irw = loongarch_linux_watch_get_irw (&watch_readback, n) &
> IRW_MASK;
> ++ t_mask = loongarch_linux_watch_get_mask (&watch_readback, n);
> ++
> ++ for (watch = current_watches; watch != NULL; watch = watch->next)
> ++ {
> ++ CORE_ADDR addr = watch->addr;
> ++ CORE_ADDR last_byte = addr + watch->len - 1;
> ++
> ++ if ((t_irw & loongarch_linux_watch_type_to_irw (watch->type))
> == 0)
> ++ {
> ++ /* Different type. */
> ++ continue;
> ++ }
> ++ /* Check for overlap of even a single byte. */
> ++ if (last_byte >= t_addr && addr <= t_addr + t_mask)
> ++ {
> ++ *paddr = addr;
> ++ return true;
> ++ }
> ++ }
> ++ }
> ++ return false;
> ++}
> ++
> ++/* Target to_region_ok_for_hw_watchpoint implementation. Return 1 if
> ++ the specified region can be covered by the watch registers. */
> ++
> ++int
> ++loongarch_linux_nat_target::region_ok_for_hw_watchpoint (CORE_ADDR addr,
> ++ int len)
> ++{
> ++ struct pt_watch_regs dummy_regs;
> ++ int i;
> ++
> ++ if (!loongarch_linux_read_watch_registers (
> ++ inferior_ptid.lwp (), &watch_readback, &watch_readback_valid, 0))
> ++ return 0;
> ++
> ++ dummy_regs = watch_readback;
> ++ /* Clear them out. */
> ++ for (i = 0; i < loongarch_linux_watch_get_num_valid (&dummy_regs); i++)
> ++ {
> ++ loongarch_linux_watch_set_addr (&dummy_regs, i, 0);
> ++ loongarch_linux_watch_set_mask (&dummy_regs, i, 0);
> ++ loongarch_linux_watch_set_irw (&dummy_regs, i, 0);
> ++ }
> ++ return loongarch_linux_watch_try_one_watch (&dummy_regs, addr, len, 0);
> ++}
> ++
> ++/* Write the mirrored watch register values for each thread. */
> ++
> ++static int
> ++write_watchpoint_regs (void)
> ++{
> ++ struct lwp_info *lp;
> ++ int tid;
> ++
> ++ ALL_LWPS (lp)
> ++ {
> ++ tid = lp->ptid.lwp ();
> ++ if (ptrace (PTRACE_SET_WATCH_REGS, tid, &watch_mirror, NULL) == -1)
> ++ perror_with_name (_ ("Couldn't write debug register"));
> ++ }
> ++ return 0;
> ++}
> ++
> ++/* linux_nat_target::low_new_thread implementation. */
> ++
> ++void
> ++loongarch_linux_nat_target::low_new_thread (struct lwp_info *lp)
> ++{
> ++ long tid = lp->ptid.lwp ();
> ++
> ++ if (!loongarch_linux_read_watch_registers (tid, &watch_readback,
> ++ &watch_readback_valid, 0))
> ++ return;
> ++
> ++ if (ptrace (PTRACE_SET_WATCH_REGS, tid, &watch_mirror, NULL) == -1)
> ++ perror_with_name (_ ("Couldn't write debug register"));
> ++}
> ++
> ++/* Target to_insert_watchpoint implementation. Try to insert a new
> ++ watch. Return zero on success. */
> ++
> ++int
> ++loongarch_linux_nat_target::insert_watchpoint (CORE_ADDR addr, int len,
> ++ enum target_hw_bp_type type,
> ++ struct expression *cond)
> ++{
> ++ struct pt_watch_regs regs =
> ++ {
> ++ .max_valid = MAX_DEBUG_REGISTER,
> ++ };
> ++ struct loongarch_watchpoint *new_watch;
> ++ struct loongarch_watchpoint **pw;
> ++
> ++ int retval;
> ++
> ++ if (!loongarch_linux_read_watch_registers (
> ++ inferior_ptid.lwp (), &watch_readback, &watch_readback_valid, 0))
> ++ return -1;
> ++
> ++ if (len <= 0)
> ++ return -1;
> ++
> ++ regs = watch_readback;
> ++ /* Add the current watches. */
> ++ loongarch_linux_watch_populate_regs (current_watches, ®s);
> ++
> ++ /* Now try to add the new watch. */
> ++ if (!loongarch_linux_watch_try_one_watch (
> ++ ®s, addr, len, loongarch_linux_watch_type_to_irw (type)))
> ++ return -1;
> ++
> ++ /* It fit. Stick it on the end of the list. */
> ++ new_watch = XNEW (struct loongarch_watchpoint);
> ++ new_watch->addr = addr;
> ++ new_watch->len = len;
> ++ new_watch->type = type;
> ++ new_watch->next = NULL;
> ++
> ++ pw = ¤t_watches;
> ++ while (*pw != NULL)
> ++ pw = &(*pw)->next;
> ++ *pw = new_watch;
> ++
> ++ watch_mirror = regs;
> ++ retval = write_watchpoint_regs ();
> ++
> ++ if (show_debug_regs)
> ++ loongarch_show_dr ("insert_watchpoint", addr, len, type);
> ++
> ++ return retval;
> ++}
> ++
> ++/* Target to_remove_watchpoint implementation. Try to remove a watch.
> ++ Return zero on success. */
> ++
> ++int
> ++loongarch_linux_nat_target::remove_watchpoint (CORE_ADDR addr, int len,
> ++ enum target_hw_bp_type type,
> ++ struct expression *cond)
> ++{
> ++ int retval;
> ++ int deleted_one;
> ++
> ++ struct loongarch_watchpoint **pw;
> ++ struct loongarch_watchpoint *w;
> ++
> ++ /* Search for a known watch that matches. Then unlink and free
> ++ it. */
> ++ deleted_one = 0;
> ++ pw = ¤t_watches;
> ++ while ((w = *pw))
> ++ {
> ++ if (w->addr == addr && w->len == len && w->type == type)
> ++ {
> ++ *pw = w->next;
> ++ xfree (w);
> ++ deleted_one = 1;
> ++ break;
> ++ }
> ++ pw = &(w->next);
> ++ }
> ++
> ++ if (!deleted_one)
> ++ return -1; /* We don't know about it, fail doing nothing. */
> ++
> ++ /* At this point watch_readback is known to be valid because we
> ++ could not have added the watch without reading it. */
> ++ gdb_assert (watch_readback_valid == 1);
> ++
> ++ watch_mirror = watch_readback;
> ++ loongarch_linux_watch_populate_regs (current_watches, &watch_mirror);
> ++
> ++ retval = write_watchpoint_regs ();
> ++
> ++ if (show_debug_regs)
> ++ loongarch_show_dr ("remove_watchpoint", addr, len, type);
> ++
> ++ return retval;
> ++}
> ++
> ++/* Insert a hardware-assisted breakpoint at BP_TGT->reqstd_address.
> ++ Return 0 on success, -1 on failure. */
> ++
> ++int
> ++loongarch_linux_nat_target::insert_hw_breakpoint (
> ++ struct gdbarch *gdbarch, struct bp_target_info *bp_tgt)
> ++{
> ++ int ret;
> ++ CORE_ADDR addr = bp_tgt->placed_address = bp_tgt->reqstd_address;
> ++ int len = 4;
> ++ const enum target_hw_bp_type type = hw_execute;
> ++
> ++ gdbarch_breakpoint_from_pc (gdbarch, &addr, &len);
> ++
> ++ if (show_debug_regs)
> ++ fprintf_unfiltered (
> ++ gdb_stdlog, "insert_hw_breakpoint on entry (addr=0x%08lx,
> len=%d))\n",
> ++ (unsigned long) addr, len);
> ++
> ++ ret = insert_watchpoint (addr, len, type, NULL);
> ++ return ret;
> ++}
> ++
> ++/* Remove a hardware-assisted breakpoint at BP_TGT->placed_address.
> ++ Return 0 on success, -1 on failure. */
> ++
> ++int
> ++loongarch_linux_nat_target::remove_hw_breakpoint (
> ++ struct gdbarch *gdbarch, struct bp_target_info *bp_tgt)
> ++{
> ++ int ret;
> ++ CORE_ADDR addr = bp_tgt->placed_address;
> ++ int len = 4;
> ++ const enum target_hw_bp_type type = hw_execute;
> ++
> ++ gdbarch_breakpoint_from_pc (gdbarch, &addr, &len);
> ++
> ++ if (show_debug_regs)
> ++ fprintf_unfiltered (
> ++ gdb_stdlog, "remove_hw_breakpoint on entry (addr=0x%08lx,
> len=%d))\n",
> ++ (unsigned long) addr, len);
> ++
> ++ ret = remove_watchpoint (addr, len, type, NULL);
> ++
> ++ return ret;
> ++}
> ++
> ++/* Target to_close implementation. Free any watches and call the
> ++ super implementation. */
> ++
> ++void
> ++loongarch_linux_nat_target::close ()
> ++{
> ++ struct loongarch_watchpoint *w;
> ++ struct loongarch_watchpoint *nw;
> ++
> ++ /* Clean out the current_watches list. */
> ++ w = current_watches;
> ++ while (w)
> ++ {
> ++ nw = w->next;
> ++ xfree (w);
> ++ w = nw;
> ++ }
> ++ current_watches = NULL;
> ++
> ++ linux_nat_trad_target::close ();
> ++}
> ++
> ++void _initialize_loongarch_linux_nat ();
> ++void
> ++_initialize_loongarch_linux_nat ()
> ++{
> ++ add_setshow_boolean_cmd (
> ++ "show-debug-regs", class_maintenance, &show_debug_regs, _ ("\
> ++Set whether to show variables that mirror the LoongArch debug
> registers."),
> ++ _ ("\
> ++Show whether to show variables that mirror the LoongArch debug
> registers."),
> ++ _ ("\
> ++Use \"on\" to enable, \"off\" to disable.\n\
> ++If enabled, the debug registers values are shown when GDB inserts\n\
> ++or removes a hardware breakpoint or watchpoint, and when the inferior\n\
> ++triggers a breakpoint or watchpoint."),
> ++ NULL, NULL, &maintenance_set_cmdlist, &maintenance_show_cmdlist);
> ++
> ++ linux_target = &the_loongarch_linux_nat_target;
> ++ add_inf_child_target (&the_loongarch_linux_nat_target);
> ++}
> +diff --git gdb-10.2/gdb/loongarch-linux-tdep.c
> gdb-10.2/gdb/loongarch-linux-tdep.c
> +new file mode 100644
> +index 0000000..5daabe3
> +--- /dev/null
> ++++ gdb-10.2/gdb/loongarch-linux-tdep.c
> +@@ -0,0 +1,709 @@
> ++/* Target-dependent code for GNU/Linux LoongArch.
> ++
> ++ Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of GDB.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program. If not, see <http://www.gnu.org/licenses/>.
> */
> ++
> ++#include "defs.h"
> ++#include "inferior.h"
> ++#include "gdbcore.h"
> ++#include "target.h"
> ++#include "solib-svr4.h"
> ++#include "osabi.h"
> ++#include "loongarch-tdep.h"
> ++#include "frame.h"
> ++#include "regcache.h"
> ++#include "trad-frame.h"
> ++#include "tramp-frame.h"
> ++#include "gdbtypes.h"
> ++#include "objfiles.h"
> ++#include "solib.h"
> ++#include "solist.h"
> ++#include "symtab.h"
> ++#include "target-descriptions.h"
> ++#include "loongarch-linux-tdep.h"
> ++#include "glibc-tdep.h"
> ++#include "linux-tdep.h"
> ++#include "xml-syscall.h"
> ++#include "gdbsupport/gdb_signals.h"
> ++
> ++static void
> ++loongarch_supply_elf_gregset (const struct regset *r,
> ++ struct regcache *regcache, int regno,
> ++ const void *gprs, size_t len)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ gdb_assert (0 <= regs->r && sizeof (loongarch_elf_gregset_t) <= len);
> ++
> ++ int regsize = register_size (regcache->arch (), regs->r);
> ++ const gdb_byte *buf = NULL;
> ++
> ++ if (regno == -1)
> ++ {
> ++ /* Set $r0 = 0. */
> ++ regcache->raw_supply_zeroed (regs->r);
> ++
> ++ for (int i = 1; i < 32; i++)
> ++ {
> ++ buf = (const gdb_byte*)gprs + regsize * i;
> ++ regcache->raw_supply (regs->r + i, (const void *)buf);
> ++ }
> ++
> ++ /* Size base (orig_a0) = regsize * regs->orig_a0. */
> ++ buf = (const gdb_byte*)gprs + regsize * regs->orig_a0;
> ++ regcache->raw_supply (regs->orig_a0, (const void *)buf);
> ++
> ++ /* Size base (pc) = regsize * regs->pc. */
> ++ buf = (const gdb_byte*)gprs + regsize * regs->pc;
> ++ regcache->raw_supply (regs->pc, (const void *)buf);
> ++
> ++ /* Size base (badv) = regsize * regs->badv. */
> ++ buf = (const gdb_byte*)gprs + regsize * regs->badv;
> ++ regcache->raw_supply (regs->badv, (const void *)buf);
> ++ }
> ++ else if (regs->r == regno)
> ++ regcache->raw_supply_zeroed (regno);
> ++ else if ((regs->r < regno && regno < regs->r + 32)
> ++ || (regs->orig_a0 == regno)
> ++ || (regs->pc == regno)
> ++ || (regs->badv == regno))
> ++ {
> ++ /* Offset offset (regno) = regsize * (regno - regs->r). */
> ++ buf = (const gdb_byte*)gprs + regsize * (regno - regs->r);
> ++ regcache->raw_supply (regno, (const void *)buf);
> ++ }
> ++}
> ++
> ++static void
> ++loongarch_fill_elf_gregset (const struct regset *r,
> ++ const struct regcache *regcache, int regno,
> ++ void *gprs, size_t len)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ gdb_assert (0 <= regs->r && sizeof (loongarch_elf_gregset_t) <= len);
> ++ int regsize = register_size (regcache->arch (), regs->r);
> ++ gdb_byte *buf = NULL;
> ++
> ++ if (regno == -1)
> ++ {
> ++ for (int i = 0; i < 32; i++)
> ++ {
> ++ buf = (gdb_byte *)gprs + regsize * i;
> ++ regcache->raw_collect (regs->r + i, (void *)buf);
> ++ }
> ++
> ++ /* Size base (orig_a0) = regsize * regs->orig_a0. */
> ++ buf = (gdb_byte *)gprs + regsize * regs->orig_a0;
> ++ regcache->raw_collect (regs->orig_a0, (void *)buf);
> ++
> ++ /* Size base (pc) = regsize * regs->pc. */
> ++ buf = (gdb_byte *)gprs + regsize * regs->pc;
> ++ regcache->raw_collect (regs->pc, (void *)buf);
> ++
> ++ /* Size base (badv) = regsize * regs->badv. */
> ++ buf = (gdb_byte *)gprs + regsize * regs->badv;
> ++ regcache->raw_collect (regs->badv, (void *)buf);
> ++ }
> ++ else if ((regs->r <= regno && regno < regs->r + 32)
> ++ ||(regs->orig_a0 == regno)
> ++ ||(regs->pc == regno)
> ++ ||(regs->badv == regno))
> ++ {
> ++ /* Offset offset (regno) = regsize * (regno - regs->r). */
> ++ buf = (gdb_byte *)gprs + regsize * (regno - regs->r);
> ++ regcache->raw_collect (regno, (void *)buf);
> ++ }
> ++}
> ++
> ++const struct regset loongarch_elf_gregset =
> ++{
> ++ NULL,
> ++ loongarch_supply_elf_gregset,
> ++ loongarch_fill_elf_gregset,
> ++};
> ++
> ++static void
> ++loongarch_supply_elf_fpregset (const struct regset *r,
> ++ struct regcache *regcache, int regno,
> ++ const void *fprs, size_t len)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ gdb_assert (0 <= regs->f && sizeof (loongarch_elf_fpregset_t) <= len);
> ++
> ++ const gdb_byte *buf = NULL;
> ++ int fprsize = register_size (regcache->arch (), regs->f);
> ++ int fccsize = register_size (regcache->arch (), regs->fcc);
> ++
> ++ if (regno == -1)
> ++ {
> ++ /* 32 fprs. */
> ++ for (int i = 0; i < 32; i++)
> ++ {
> ++ buf = (const gdb_byte *)fprs + fprsize * i;
> ++ regcache->raw_supply (regs->f + i, (const void *)buf);
> ++ }
> ++
> ++ /* 8 fccs , base (fcc) = 32 * sizeof (fpr). */
> ++ buf = (const gdb_byte *)fprs + fprsize * 32;
> ++ for (int i = 0; i < 8; i++)
> ++ {
> ++ regcache->raw_supply (regs->fcc + i, (const void *)buf);
> ++ buf += fccsize;
> ++ }
> ++
> ++ /* Size base (fcsr) = 32 * sizeof (fpr) + 8 * sizeof (fcc). */
> ++ buf = (const gdb_byte *)fprs + 32 * fprsize + 8 * fccsize;
> ++ regno = regs->fcsr;
> ++ }
> ++ else if (regs->f <= regno && regno < regs->f + 32)
> ++ {
> ++ /* Offset offset (regno - f) = (regno - regs->f) * sizeof (fpr). */
> ++ buf = (const gdb_byte *)fprs + fprsize * (regno - regs->f);
> ++ }
> ++ else if (regs->fcc <= regno && regno < regs->fcc + 8)
> ++ {
> ++ /* Size base (fcc) + offset (regno - fcc)
> ++ = 32 * sizeof (fpr) + (regno - regs->fcc) * sizeof (fcc). */
> ++ buf = (const gdb_byte *)fprs + 32 * fprsize
> ++ + (regno - regs->fcc) * fccsize;
> ++ }
> ++ else if (regs->fcsr == regno)
> ++ {
> ++ /* Size base (fcsr) = 32 * sizeof (fpr) + 8 * sizeof (fcc). */
> ++ buf = (const gdb_byte *)fprs + 32 * fprsize + 8 * fccsize;
> ++ }
> ++ else
> ++ {
> ++ return;
> ++ }
> ++
> ++ /* Supply register. */
> ++ regcache->raw_supply (regno, (const void *)buf);
> ++}
> ++
> ++static void
> ++loongarch_fill_elf_fpregset (const struct regset *r,
> ++ const struct regcache *regcache, int regno,
> ++ void *fprs, size_t len)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ gdb_assert (0 <= regs->f && sizeof (loongarch_elf_fpregset_t) <= len);
> ++ gdb_byte *buf = NULL;
> ++ int fprsize = register_size (regcache->arch (), regs->f);
> ++ int fccsize = register_size (regcache->arch (), regs->fcc);
> ++
> ++ if (regno == -1)
> ++ {
> ++ /* 32 fprs. */
> ++ for (int i = 0; i < 32; i++)
> ++ {
> ++ buf = (gdb_byte *)fprs + fprsize * i;
> ++ regcache->raw_collect (regs->f + i, (void *)buf);
> ++ }
> ++
> ++ /* 8 fccs , base (fcc) = 32 * sizeof (fpr). */
> ++ buf = (gdb_byte *)fprs + fprsize * 32;
> ++ for (int i = 0; i < 8; i++)
> ++ {
> ++ regcache->raw_collect (regs->fcc + i, (void *)buf);
> ++ buf += fccsize;
> ++ }
> ++
> ++ /* Size base (fcsr) = 32 * sizeof (fpr) + 8 * sizeof (fcc). */
> ++ buf = (gdb_byte *)fprs + fprsize * 32 + fccsize * 8;
> ++ regno = regs->fcsr;
> ++ }
> ++ else if (regs->f <= regno && regno < regs->f + 32)
> ++ {
> ++ /* Offset offset (regno - f) = (regno - regs->f) * sizeof (fpr). */
> ++ buf = (gdb_byte *)fprs + fprsize * (regno - regs->f);
> ++ }
> ++ else if (regs->fcc <= regno && regno < regs->fcc + 8)
> ++ {
> ++ /* Size base (fcc) + offset (regno - fcc)
> ++ = 32 * sizeof (fpr) + (regno - regs->fcc) * sizeof (fcc). */
> ++ buf = (gdb_byte *)fprs + 32 * fprsize + (regno - regs->fcc) *
> fccsize;
> ++ }
> ++ else if (regs->fcsr == regno)
> ++ {
> ++ /* Size base (fcsr) = 32 * sizeof (fpr) + 8 * sizeof (fcc). */
> ++ buf = (gdb_byte *)fprs + 32 * fprsize + 8 * fccsize;
> ++ }
> ++ else
> ++ {
> ++ return;
> ++ }
> ++
> ++ /* Supply register. */
> ++ regcache->raw_collect (regno, (void *)buf);
> ++}
> ++
> ++const struct regset loongarch_elf_fpregset =
> ++{
> ++ NULL,
> ++ loongarch_supply_elf_fpregset,
> ++ loongarch_fill_elf_fpregset,
> ++};
> ++
> ++static void
> ++loongarch_supply_elf_cpucfgregset (const struct regset *r,
> ++ struct regcache *regcache, int regno,
> ++ const void *cpucfgs, size_t len)
> ++{
> ++}
> ++
> ++static void
> ++loongarch_fill_elf_cpucfgregset (const struct regset *r,
> ++ const struct regcache *regcache, int
> regno,
> ++ void *cpucfgs, size_t len)
> ++{
> ++ ULONGEST xfered_len;
> ++ target_xfer_partial (current_inferior ()->top_target (),
> ++ /* current_top_target (),*/ TARGET_OBJECT_LARCH,
> ++ "cpucfg", (gdb_byte *) cpucfgs, NULL, 0, len,
> ++ &xfered_len);
> ++ memset ((gdb_byte *) cpucfgs + xfered_len, 0, len - xfered_len);
> ++}
> ++
> ++const struct regset loongarch_elf_cpucfgregset =
> ++{
> ++ NULL,
> ++ loongarch_supply_elf_cpucfgregset,
> ++ loongarch_fill_elf_cpucfgregset,
> ++};
> ++
> ++static void
> ++loongarch_supply_elf_lbtregset (const struct regset *r,
> ++ struct regcache *regcache, int regno,
> ++ const void *lbtrs, size_t len)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ gdb_assert (0 <= regs->scr && sizeof (loongarch_elf_lbtregset_t) <=
> len);
> ++
> ++ const gdb_byte *buf = NULL;
> ++ int scrsize = register_size (regcache->arch (), regs->scr);
> ++ int efsize = register_size (regcache->arch (), regs->EFLAG);
> ++
> ++ if (regno == -1)
> ++ {
> ++ /* 4 scrs. */
> ++ for (int i = 0; i < 4; i++)
> ++ {
> ++ buf = (const gdb_byte *)lbtrs + scrsize * i;
> ++ regcache->raw_supply (regs->scr + i, (const void *)buf);
> ++ }
> ++
> ++ /* Size base (EFLAG) = 4 * sizeof (scr). */
> ++ buf = (const gdb_byte *)lbtrs + scrsize * 4;
> ++ regcache->raw_supply (regs->EFLAG, (const void *)buf);
> ++
> ++ /* Size base (x86_top) = 4 * sizeof (scr) + sizeof (EFLAG). */
> ++ buf = (const gdb_byte *)lbtrs + scrsize * 4 + efsize;
> ++ regno = regs->x86_top;
> ++ }
> ++ else if (regs->scr <= regno && regno < regs->scr + 4)
> ++ /* Offset offset (EFLAG) = sizeof (scr) * (regno - regs->scr). */
> ++ buf = (const gdb_byte *)lbtrs + scrsize * (regno - regs->scr);
> ++ else if (regs->EFLAG == regno)
> ++ /* Size base (EFLAG) = 4 * sizeof (scr). */
> ++ buf = (const gdb_byte *)lbtrs + scrsize * 4;
> ++ else if (regs->x86_top == regno)
> ++ {
> ++ /* Size base (x86_top) = 4 * sizeof (scr) + sizeof (EFLAG). */
> ++ buf = (const gdb_byte *)lbtrs + scrsize * 4 + efsize;
> ++ }
> ++ else
> ++ {
> ++ return;
> ++ }
> ++
> ++ regcache->raw_supply (regno, (const void *)buf);
> ++}
> ++
> ++static void
> ++loongarch_fill_elf_lbtregset (const struct regset *r,
> ++ const struct regcache *regcache, int regno,
> ++ void *lbtrs, size_t len)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ gdb_assert (0 <= regs->scr && sizeof (loongarch_elf_lbtregset_t) <=
> len);
> ++
> ++ gdb_byte *buf = NULL;
> ++ int scrsize = register_size (regcache->arch (), regs->scr);
> ++ int efsize = register_size (regcache->arch (), regs->EFLAG);
> ++
> ++ if (regno == -1)
> ++ {
> ++
> ++ /* 4 scrs. */
> ++ for (int i = 0; i < 4; i++)
> ++ {
> ++ buf = (gdb_byte *)lbtrs + scrsize * i;
> ++ regcache->raw_collect (regs->scr + i, (void *)buf);
> ++ }
> ++
> ++ /* Size base (EFLAG) = 4 * sizeof (scr). */
> ++ buf = (gdb_byte *)lbtrs + scrsize * 4;
> ++ regcache->raw_collect (regs->EFLAG, (void *)buf);
> ++
> ++ /* Size base (x86_top) = 4 * sizeof (scr) + sizeof (EFLAG). */
> ++ buf = (gdb_byte *)lbtrs + scrsize * 4 + efsize;
> ++ regno = regs->x86_top;
> ++ }
> ++ else if (regs->scr <= regno && regno < regs->scr + 4)
> ++ /* Offset offset (EFLAG) = sizeof (scr) * (regno - regs->scr). */
> ++ buf = (gdb_byte *)lbtrs + scrsize * (regno - regs->scr);
> ++ else if (regs->EFLAG == regno)
> ++ /* Size base (EFLAG) = 4 * sizeof (scr). */
> ++ buf = (gdb_byte *)lbtrs + scrsize * 4;
> ++ else if (regs->x86_top == regno)
> ++ /* Size base (x86_top) = 4 * sizeof (scr) + sizeof (EFLAG). */
> ++ buf = (gdb_byte *)lbtrs + scrsize * 4 + efsize;
> ++ else
> ++ {
> ++ return;
> ++ }
> ++
> ++ regcache->raw_collect (regno, (void *)buf);
> ++}
> ++
> ++const struct regset loongarch_elf_lbtregset =
> ++{
> ++ NULL,
> ++ loongarch_supply_elf_lbtregset,
> ++ loongarch_fill_elf_lbtregset,
> ++};
> ++
> ++static void
> ++loongarch_supply_elf_lsxregset (const struct regset *r,
> ++ struct regcache *regcache, int regno,
> ++ const void *lsxrs, size_t len)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ gdb_assert (0 <= regs->vr && sizeof (loongarch_elf_lsxregset_t) <=
> len);
> ++
> ++ const gdb_byte *buf = NULL;
> ++ int regsize = register_size (regcache->arch (), regs->vr);
> ++
> ++ if (regno == -1)
> ++ {
> ++ for (int i = 0; i < 32; i++)
> ++ {
> ++ buf = (const gdb_byte *)lsxrs + regsize * i;
> ++ regcache->raw_supply (regs->vr + i, (const void *)buf);
> ++ }
> ++ }
> ++ else if (regs->vr <= regno && regno < regs->vr + 32)
> ++ {
> ++ buf = (const gdb_byte *)lsxrs + regsize * (regno - regs->vr);
> ++ regcache->raw_supply (regno, (const void *)buf);
> ++ }
> ++}
> ++
> ++static void
> ++loongarch_fill_elf_lsxregset (const struct regset *r,
> ++ const struct regcache *regcache, int regno,
> ++ void *lsxrs, size_t len)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ gdb_assert (0 <= regs->vr && sizeof (loongarch_elf_lsxregset_t) <=
> len);
> ++
> ++ gdb_byte *buf = NULL;
> ++ int regsize = register_size (regcache->arch (), regs->vr);
> ++
> ++ if (regno == -1)
> ++ {
> ++ for (int i = 0; i < 32; i++)
> ++ {
> ++ buf = (gdb_byte *)lsxrs + regsize * i;
> ++ regcache->raw_collect (regs->vr + i, (void *)buf);
> ++ }
> ++ }
> ++ else if (regs->vr <= regno && regno < regs->vr + 32)
> ++ {
> ++ buf = (gdb_byte *)lsxrs + regsize * (regno - regs->vr);
> ++ regcache->raw_collect (regno, (void *)buf);
> ++ }
> ++}
> ++
> ++const struct regset loongarch_elf_lsxregset =
> ++{
> ++ NULL,
> ++ loongarch_supply_elf_lsxregset,
> ++ loongarch_fill_elf_lsxregset,
> ++};
> ++
> ++static void
> ++loongarch_supply_elf_lasxregset (const struct regset *r,
> ++ struct regcache *regcache, int regno,
> ++ const void *lasxrs, size_t len)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ gdb_assert (0 <= regs->xr && sizeof (loongarch_elf_lasxregset_t) <=
> len);
> ++
> ++ const gdb_byte *buf = NULL;
> ++ int regsize = register_size (regcache->arch (), regs->xr);
> ++
> ++ if (regno == -1)
> ++ {
> ++ for (int i = 0; i < 32; i++)
> ++ {
> ++ buf = (const gdb_byte *)lasxrs + regsize * i;
> ++ regcache->raw_supply (regs->xr + i, (const void *)buf);
> ++ }
> ++ }
> ++ else if (regs->xr <= regno && regno < regs->xr + 32)
> ++ {
> ++ buf = (const gdb_byte *)lasxrs + regsize * (regno - regs->xr);
> ++ regcache->raw_supply (regno, (const void *)buf);
> ++ }
> ++}
> ++
> ++static void
> ++loongarch_fill_elf_lasxregset (const struct regset *r,
> ++ const struct regcache *regcache, int regno,
> ++ void *lasxrs, size_t len)
> ++{
> ++ auto regs = &gdbarch_tdep (regcache->arch ())->regs;
> ++ gdb_assert (0 <= regs->xr && sizeof (loongarch_elf_lasxregset_t) <=
> len);
> ++
> ++ gdb_byte *buf = NULL;
> ++ int regsize = register_size (regcache->arch (), regs->xr);
> ++
> ++ if (regno == -1)
> ++ {
> ++ for (int i = 0; i < 32; i++)
> ++ {
> ++ buf = (gdb_byte *)lasxrs + regsize * i;
> ++ regcache->raw_collect (regs->xr + i, (void *)buf);
> ++ }
> ++ }
> ++ else if (regs->xr <= regno && regno < regs->xr + 32)
> ++ {
> ++ buf = (gdb_byte *)lasxrs + regsize * (regno - regs->xr);
> ++ regcache->raw_collect (regno, (void *)buf);
> ++ }
> ++}
> ++
> ++const struct regset loongarch_elf_lasxregset =
> ++{
> ++ NULL,
> ++ loongarch_supply_elf_lasxregset,
> ++ loongarch_fill_elf_lasxregset,
> ++};
> ++
> ++static void
> ++loongarch_linux_iterate_over_regset_sections (
> ++ struct gdbarch *gdbarch, iterate_over_regset_sections_cb *cb, void
> *cb_data,
> ++ const struct regcache *regcache)
> ++{
> ++ auto regs = &gdbarch_tdep (gdbarch)->regs;
> ++ if (0 <= regs->r)
> ++ cb (".reg", sizeof (loongarch_elf_gregset_t),
> ++ sizeof (loongarch_elf_gregset_t), &loongarch_elf_gregset, NULL,
> ++ cb_data);
> ++ if (0 <= regs->f)
> ++ cb (".reg2", sizeof (loongarch_elf_fpregset_t),
> ++ sizeof (loongarch_elf_fpregset_t), &loongarch_elf_fpregset, NULL,
> ++ cb_data);
> ++ do
> ++ {
> ++ uint32_t t;
> ++ ULONGEST xfered_len;
> ++ if (target_xfer_partial (current_inferior ()->top_target (),
> ++ /* current_top_target (),*/
> TARGET_OBJECT_LARCH,
> ++ "cpucfg", (gdb_byte *) &t, NULL, 0, sizeof
> (t),
> ++ &xfered_len) != TARGET_XFER_OK)
> ++ break;
> ++ cb (".reg-loongarch-cpucfg", 64 * 4, 64 * 4,
> &loongarch_elf_cpucfgregset,
> ++ "LoongArch CPU config", cb_data);
> ++ }
> ++ while (0);
> ++ if (0 <= regs->scr)
> ++ cb (".reg-loongarch-lbt", sizeof (loongarch_elf_lbtregset_t),
> ++ sizeof (loongarch_elf_lbtregset_t), &loongarch_elf_lbtregset,
> ++ "LoongArch Binary Translation", cb_data);
> ++ if (0 <= regs->vr)
> ++ cb (".reg-loongarch-lsx", sizeof (loongarch_elf_lsxregset_t),
> ++ sizeof (loongarch_elf_lsxregset_t), &loongarch_elf_lsxregset,
> ++ "LoongArch SIMD Extension", cb_data);
> ++ if (0 <= regs->xr)
> ++ cb (".reg-loongarch-lasx", sizeof (loongarch_elf_lasxregset_t),
> ++ sizeof (loongarch_elf_lasxregset_t), &loongarch_elf_lasxregset,
> ++ "LoongArch Advanced SIMD Extension", cb_data);
> ++}
> ++
> ++static const struct target_desc *
> ++loongarch_linux_core_read_description (struct gdbarch *gdbarch,
> ++ struct target_ops *target, bfd
> *abfd)
> ++{
> ++ int rlen, flen, fpu64, lbt, lsx, lasx;
> ++
> ++ rlen = 64;
> ++
> ++ fpu64 = !!bfd_get_section_by_name (abfd, ".reg2");
> ++ lbt = !!bfd_get_section_by_name (abfd, ".reg-loongarch-lbt");
> ++ lsx = !!bfd_get_section_by_name (abfd, ".reg-loongarch-lsx");
> ++ lasx = !!bfd_get_section_by_name (abfd, ".reg-loongarch-lasx");
> ++
> ++ if (fpu64)
> ++ flen = 64;
> ++ else
> ++ flen = 0;
> ++
> ++ return loongarch_create_target_description (rlen, flen, lbt, lsx,
> lasx);
> ++}
> ++
> ++/* The RT signal frames look like this:
> ++ struct rt_sigframe {
> ++ struct siginfo rs_info;
> ++ struct ucontext rs_uctx;
> ++ };
> ++
> ++ struct ucontext {
> ++ unsigned long uc_flags;
> ++ struct ucontext *uc_link;
> ++ stack_t uc_stack;
> ++ sigset_t uc_sigmask;
> ++ _u8 __unused[1024 / 8 - sizeof(sigset_t)];
> ++ struct sigcontext uc_mcontext;
> ++ };
> ++}; */
> ++
> ++#define RTSIGFRAME_SIGINFO_SIZE 128
> ++#define UCONTEXT_SIGCONTEXT_OFFSET 176
> ++#define RTSIGFRAME_SIGCONTEXT_OFFSET (RTSIGFRAME_SIGINFO_SIZE \
> ++ + UCONTEXT_SIGCONTEXT_OFFSET)
> ++
> ++static void
> ++loongarch_linux_lp64_sigframe_init (const struct tramp_frame *self,
> ++ struct frame_info *this_frame,
> ++ struct trad_frame_cache *this_cache,
> ++ CORE_ADDR func)
> ++{
> ++ struct gdbarch *gdbarch = get_frame_arch (this_frame);
> ++ auto regs = &gdbarch_tdep (gdbarch)->regs;
> ++ CORE_ADDR frame_sp = get_frame_sp (this_frame);
> ++
> ++ CORE_ADDR sigcontext_base = frame_sp + RTSIGFRAME_SIGCONTEXT_OFFSET;
> ++ int i;
> ++
> ++ trad_frame_set_reg_addr (this_cache, regs->pc, sigcontext_base);
> ++ for (i = 0; i < 32; i++)
> ++ trad_frame_set_reg_addr (this_cache, regs->r + i,
> ++ sigcontext_base + 8 + i * 8);
> ++
> ++ trad_frame_set_id (this_cache, frame_id_build (frame_sp, func));
> ++}
> ++
> ++static const struct tramp_frame loongarch_linux_lp64_rt_sigframe =
> ++{
> ++ SIGTRAMP_FRAME,
> ++ 4,
> ++ { /* From $kernel/arch/loongarch/vdso/sigreturn.S. */
> ++ /* ori $r11, $r0, 0x8b(__NR_rt_sigreturn) */
> ++ { 0x03822c0b, ULONGEST_MAX },
> ++ { 0x002b0000, ULONGEST_MAX }, /* syscall 0 */
> ++ { TRAMP_SENTINEL_INSN, ULONGEST_MAX } },
> ++ loongarch_linux_lp64_sigframe_init,
> ++ NULL
> ++};
> ++
> ++/* Return the current system call's number present in the
> ++ a7 register. When the function fails, it returns -1. */
> ++
> ++static LONGEST
> ++loongarch_linux_get_syscall_number (struct gdbarch *gdbarch,
> ++ thread_info *thread)
> ++{
> ++ struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch);
> ++ auto regs = &tdep->regs;
> ++ struct regcache *regcache = get_thread_regcache (thread);
> ++ LONGEST ret;
> ++
> ++ if ((EF_LOONGARCH_IS_LP64 (tdep->ef_abi))
> ++ && (REG_VALID == regcache_cooked_read_signed (regcache, regs->r +
> 11, &ret)))
> ++ return ret;
> ++
> ++ return -1;
> ++}
> ++
> ++static CORE_ADDR
> ++loongarch_linux_syscall_next_pc (struct frame_info *frame)
> ++{
> ++ struct gdbarch *gdbarch = get_frame_arch (frame);
> ++ struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch);
> ++ auto regs = &tdep->regs;
> ++ ULONGEST a7 = get_frame_register_unsigned (frame, regs->r + 11);
> ++
> ++ /* If we are about to make a sigreturn syscall, use the unwinder to
> ++ decode the signal frame. */
> ++ if ((EF_LOONGARCH_IS_LP64 (tdep->ef_abi))
> ++ && (a7 == 0x8b /* LP64: __NR_rt_sigreturn. */))
> ++ return frame_unwind_caller_pc (get_current_frame ());
> ++
> ++ return -1;
> ++}
> ++
> ++static void
> ++loongarch_linux_init_abi (struct gdbarch_info info, struct gdbarch
> *gdbarch)
> ++{
> ++ struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch);
> ++
> ++ linux_init_abi (info, gdbarch); /* FIXME displaced step support. */
> ++
> ++ if (EF_LOONGARCH_IS_ILP32 (tdep->ef_abi))
> ++ set_solib_svr4_fetch_link_map_offsets (
> ++ gdbarch, svr4_ilp32_fetch_link_map_offsets);
> ++
> ++ else if (EF_LOONGARCH_IS_LP64 (tdep->ef_abi))
> ++ {
> ++ set_solib_svr4_fetch_link_map_offsets (gdbarch,
> ++
> svr4_lp64_fetch_link_map_offsets);
> ++ tramp_frame_prepend_unwinder (gdbarch,
> ++ &loongarch_linux_lp64_rt_sigframe);
> ++ tdep->syscall_next_pc = loongarch_linux_syscall_next_pc;
> ++
> ++ set_gdbarch_get_syscall_number (gdbarch,
> ++ loongarch_linux_get_syscall_number);
> ++ }
> ++
> ++ /* GNU/Linux uses the dynamic linker included in the GNU C Library. */
> ++ set_gdbarch_skip_solib_resolver (gdbarch, glibc_skip_solib_resolver);
> ++
> ++ /* Enable TLS support. */
> ++ set_gdbarch_fetch_tls_load_module_address (gdbarch,
> ++ svr4_fetch_objfile_link_map);
> ++
> ++ set_gdbarch_call_dummy_location (gdbarch, AT_ENTRY_POINT);
> ++
> ++ /* Core file support. */
> ++ set_gdbarch_iterate_over_regset_sections (
> ++ gdbarch, loongarch_linux_iterate_over_regset_sections);
> ++ set_gdbarch_core_read_description (gdbarch,
> ++
> loongarch_linux_core_read_description);
> ++}
> ++
> ++void _initialize_loongarch_linux_tdep ();
> ++void
> ++_initialize_loongarch_linux_tdep ()
> ++{
> ++ gdbarch_register_osabi (bfd_arch_loongarch, bfd_mach_loongarch32,
> ++ GDB_OSABI_LINUX, loongarch_linux_init_abi);
> ++ gdbarch_register_osabi (bfd_arch_loongarch, bfd_mach_loongarch64,
> ++ GDB_OSABI_LINUX, loongarch_linux_init_abi);
> ++}
> +diff --git gdb-10.2/gdb/loongarch-linux-tdep.h
> gdb-10.2/gdb/loongarch-linux-tdep.h
> +new file mode 100644
> +index 0000000..bb70043
> +--- /dev/null
> ++++ gdb-10.2/gdb/loongarch-linux-tdep.h
> +@@ -0,0 +1,48 @@
> ++/* GNU/Linux on LoongArch target support, prototypes.
> ++
> ++ Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of GDB.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program. If not, see <http://www.gnu.org/licenses/>.
> */
> ++
> ++#ifndef LOONGARCH_LINUX_TDEP_H
> ++#define LOONGARCH_LINUX_TDEP_H
> ++
> ++#include <regset.h>
> ++
> ++#define ELF_NGREG 45
> ++#define ELF_NFPREG 34
> ++
> ++typedef unsigned long loongarch_elf_gregset_t[ELF_NGREG];
> ++extern const struct regset loongarch_elf_gregset;
> ++
> ++typedef double loongarch_elf_fpregset_t[ELF_NFPREG];
> ++extern const struct regset loongarch_elf_fpregset;
> ++
> ++/* Regset variable size. */
> ++extern const struct regset loongarch_elf_cpucfg;
> ++
> ++/* 4 SCRs + 4-byte EFLAG + 1-byte x86_top. */
> ++typedef uint64_t loongarch_elf_lbtregset_t[5];
> ++extern const struct regset loongarch_elf_lbtregset;
> ++
> ++typedef uint64_t loongarch_elf_lsxregset_t[32 * 2];
> ++extern const struct regset loongarch_elf_lsxregset;
> ++
> ++typedef uint64_t loongarch_elf_lasxregset_t[32 * 4];
> ++extern const struct regset loongarch_elf_lasxregset;
> ++
> ++#endif
> +diff --git gdb-10.2/gdb/loongarch-tdep.c gdb-10.2/gdb/loongarch-tdep.c
> +new file mode 100644
> +index 0000000..422c80b
> +--- /dev/null
> ++++ gdb-10.2/gdb/loongarch-tdep.c
> +@@ -0,0 +1,1926 @@
> ++/* Target-dependent code for GNU/Linux LoongArch.
> ++
> ++ Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of GDB.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program. If not, see <http://www.gnu.org/licenses/>.
> */
> ++
> ++#include "defs.h"
> ++#include "frame.h"
> ++#include "inferior.h"
> ++#include "symtab.h"
> ++#include "value.h"
> ++#include "gdbcmd.h"
> ++#include "language.h"
> ++#include "gdbcore.h"
> ++#include "symfile.h"
> ++#include "objfiles.h"
> ++#include "gdbtypes.h"
> ++#include "target.h"
> ++#include "arch-utils.h"
> ++#include "regcache.h"
> ++#include "osabi.h"
> ++#include "block.h"
> ++#include "reggroups.h"
> ++#include "elf-bfd.h"
> ++#include "symcat.h"
> ++#include "dis-asm.h"
> ++#include "frame-unwind.h"
> ++#include "frame-base.h"
> ++#include "trad-frame.h"
> ++#include "infcall.h"
> ++#include "floatformat.h"
> ++#include "remote.h"
> ++#include "target-descriptions.h"
> ++#include "dwarf2/frame.h"
> ++#include "user-regs.h"
> ++#include "valprint.h"
> ++#include "gdbsupport/common-defs.h"
> ++#include "cli/cli-decode.h"
> ++#include "observable.h"
> ++#include "loongarch-tdep.h"
> ++#include "arch/loongarch.h"
> ++
> ++#include <algorithm>
> ++
> ++/* Figure out where the longjmp will land.
> ++ We expect the first arg to be a pointer to the jmp_buf structure
> ++ from which we extract the pc (LOONGARCH_JB_PC) that we will land
> ++ at. The pc is copied into PC. This routine returns 1 on
> ++ success. */
> ++#define LOONGARCH_JB_PC 0
> ++
> ++static int
> ++loongarch_rlen (struct gdbarch *gdbarch)
> ++{
> ++ if (EF_LOONGARCH_IS_LP64 (gdbarch_tdep (gdbarch)->ef_abi))
> ++ return 64;
> ++ else if (EF_LOONGARCH_IS_ILP32 (gdbarch_tdep (gdbarch)->ef_abi))
> ++ return 32;
> ++ else
> ++ gdb_assert_not_reached ("unknown ABI");
> ++ return 0;
> ++}
> ++
> ++static insn_t
> ++loongarch_fetch_instruction (CORE_ADDR addr, int *errp)
> ++{
> ++ size_t insnlen = loongarch_insn_length (0);
> ++ gdb_byte buf[insnlen];
> ++ int err;
> ++ ULONGEST ret;
> ++
> ++ err = target_read_memory (addr, buf, insnlen);
> ++ if (errp != NULL)
> ++ *errp = err;
> ++ if (err != 0)
> ++ {
> ++ if (errp == NULL)
> ++ memory_error (TARGET_XFER_E_IO, addr);
> ++ return 0;
> ++ }
> ++ ret = extract_unsigned_integer (buf, insnlen, BFD_ENDIAN_LITTLE);
> ++ return ret;
> ++}
> ++
> ++static int
> ++loongarch_insn_is_branch_and_must_branch (insn_t insn)
> ++{
> ++ if ((insn & 0xfc000000) == 0x4c000000 /* jirl
> r0:5,r5:5,s10:16<<2 */
> ++ || (insn & 0xfc000000) == 0x50000000 /* b sb0:10|10:16<<2 */
> ++ || (insn & 0xfc000000) == 0x54000000) /* bl sb0:10|10:16<<2 */
> ++ return 1;
> ++ return 0;
> ++}
> ++
> ++static int
> ++loongarch_insn_is_branch (insn_t insn)
> ++{
> ++ if (loongarch_insn_is_branch_and_must_branch (insn)
> ++ || (insn & 0xfc000000) == 0x40000000 /* beqz r5:5,sb0:5|10:16<<2
> */
> ++ || (insn & 0xfc000000) == 0x44000000 /* bnez r5:5,sb0:5|10:16<<2
> */
> ++ || (insn & 0xfc000300) == 0x48000000 /* bceqz c5:3,sb0:5|10:16<<2
> */
> ++ || (insn & 0xfc000300) == 0x48000100 /* bcnez c5:3,sb0:5|10:16<<2
> */
> ++ || (insn & 0xfc000000) == 0x58000000 /* beq r5:5,r0:5,sb10:16<<2
> */
> ++ || (insn & 0xfc000000) == 0x5c000000 /* bne r5:5,r0:5,sb10:16<<2
> */
> ++ || (insn & 0xfc000000) == 0x60000000 /* blt r5:5,r0:5,sb10:16<<2
> */
> ++ || (insn & 0xfc000000) == 0x64000000 /* bge r5:5,r0:5,sb10:16<<2
> */
> ++ || (insn & 0xfc000000) == 0x68000000 /* bltu r5:5,r0:5,sb10:16<<2
> */
> ++ || (insn & 0xfc000000) == 0x6c000000) /* bgeu r5:5,r0:5,sb10:16<<2
> */
> ++ return 1;
> ++ return 0;
> ++}
> ++
> ++static CORE_ADDR
> ++loongarch_next_pc_if_branch (struct regcache *regcache, CORE_ADDR cur_pc,
> ++ insn_t insn)
> ++{
> ++ struct gdbarch *gdbarch = regcache->arch ();
> ++ auto regs = &gdbarch_tdep (gdbarch)->regs;
> ++ CORE_ADDR next_pc;
> ++
> ++ if ((insn & 0xfc000000) == 0x40000000 /* beqz
> r5:5,sb0:5|10:16<<2 */
> ++ || (insn & 0xfc000000) == 0x44000000 /* bnez r5:5,sb0:5|10:16<<2
> */
> ++ || (insn & 0xfc000300) == 0x48000000 /* bceqz c5:3,sb0:5|10:16<<2
> */
> ++ || (insn & 0xfc000300) == 0x48000100) /* bcnez c5:3,sb0:5|10:16<<2
> */
> ++ next_pc = cur_pc + loongarch_decode_imm ("0:5|10:16<<2", insn, 1);
> ++ else if ((insn & 0xfc000000) == 0x4c000000) /* jirl
> r0:5,r5:5,s10:16<<2 */
> ++ next_pc = regcache_raw_get_signed (
> ++ regcache, regs->r + loongarch_decode_imm ("5:5", insn, 0))
> ++ + loongarch_decode_imm ("10:16<<2", insn, 1);
> ++ else if ((insn & 0xfc000000) == 0x50000000 /* b sb0:10|10:16<<2 */
> ++ || (insn & 0xfc000000) == 0x54000000) /* bl sb0:10|10:16<<2 */
> ++ next_pc = cur_pc + loongarch_decode_imm ("0:10|10:16<<2", insn, 1);
> ++ else if ((insn & 0xfc000000) == 0x58000000 /* beq
> r5:5,r0:5,sb10:16<<2 */
> ++ || (insn & 0xfc000000) == 0x5c000000 /* bne
> r5:5,r0:5,sb10:16<<2 */
> ++ || (insn & 0xfc000000) == 0x60000000 /* blt
> r5:5,r0:5,sb10:16<<2 */
> ++ || (insn & 0xfc000000) == 0x64000000 /* bge
> r5:5,r0:5,sb10:16<<2 */
> ++ || (insn & 0xfc000000) == 0x68000000 /* bltu
> r5:5,r0:5,sb10:16<<2 */
> ++ || (insn & 0xfc000000)
> ++ == 0x6c000000) /* bgeu r5:5,r0:5,sb10:16<<2 */
> ++ next_pc = cur_pc + loongarch_decode_imm ("10:16<<2", insn, 1);
> ++ else
> ++ gdb_assert_not_reached ("I don't know what branch is this");
> ++
> ++ return next_pc;
> ++}
> ++
> ++/* Checks for an atomic sequence of instructions beginning with a LL/LLD
> ++ instruction and ending with a SC/SCD instruction. If such a sequence
> ++ is found, attempt to step through it. A breakpoint is placed at the
> end of
> ++ the sequence. */
> ++
> ++static std::vector<CORE_ADDR>
> ++loongarch_deal_with_atomic_sequence (struct regcache *regcache,
> CORE_ADDR pc)
> ++{
> ++ struct gdbarch *gdbarch = regcache->arch ();
> ++ CORE_ADDR next_pc;
> ++ std::vector<CORE_ADDR> next_pcs;
> ++ insn_t insn = loongarch_fetch_instruction (pc, NULL);
> ++ size_t insnlen = loongarch_insn_length (insn);
> ++ int i, atomic_sequence_length, found_atomic_sequence_endpoint;
> ++
> ++ if ((insn & 0xff000000) != 0x20000000 /* ll.w */
> ++ && (insn & 0xff000000) != 0x22000000) /* ll.d */
> ++ return {};
> ++
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (gdb_stdlog,
> ++ "Single step: PC: %s OK, I found ll\\.[wd] here.
> It's "
> ++ "atomic sequence?\n",
> ++ paddress (gdbarch, pc));
> ++
> ++ atomic_sequence_length = 30; /* Magic. */
> ++ found_atomic_sequence_endpoint = 0;
> ++ for (pc += insnlen, i = 0; i < atomic_sequence_length; pc += insnlen,
> i++)
> ++ {
> ++ insn = loongarch_fetch_instruction (pc, NULL);
> ++ insnlen = loongarch_insn_length (insn);
> ++
> ++ if (loongarch_insn_is_branch_and_must_branch (insn))
> ++ {
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (
> ++ gdb_stdlog,
> ++ "Single step: PC: %s Must branch here. Treat it normally.\n",
> ++ paddress (gdbarch, pc));
> ++ break;
> ++ }
> ++ else if (loongarch_insn_is_branch (insn))
> ++ {
> ++ next_pc = loongarch_next_pc_if_branch (regcache, pc, insn);
> ++
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (gdb_stdlog,
> ++ "Single step: PC: %s May branch inside and
> "
> ++ "target is %s. Breakpoint there.\n",
> ++ paddress (gdbarch, pc),
> ++ paddress (gdbarch, next_pc));
> ++
> ++ next_pcs.push_back (next_pc);
> ++ }
> ++ else if ((insn & 0xff000000) == 0x21000000 /* sc.w */
> ++ || (insn & 0xff000000) == 0x23000000) /* sc.d */
> ++ {
> ++ found_atomic_sequence_endpoint = 1;
> ++ next_pc = pc + insnlen;
> ++
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (gdb_stdlog,
> ++ "Single step: PC: %s I found sc\\.[wd] and
> "
> ++ "atomic sequence ends at here.\n"
> ++ "Breakpoint next pc: %s.\n",
> ++ paddress (gdbarch, pc),
> ++ paddress (gdbarch, next_pc));
> ++
> ++ next_pcs.push_back (next_pc);
> ++ break;
> ++ }
> ++ }
> ++
> ++ if (!found_atomic_sequence_endpoint)
> ++ {
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (
> ++ gdb_stdlog,
> ++ "Single step: PC: %s Not ends with sc\\.[wd] in %d insns?\n"
> ++ "Treat it as not atomic sequence.\n",
> ++ paddress (gdbarch, pc), atomic_sequence_length);
> ++
> ++ return {};
> ++ }
> ++
> ++ return next_pcs;
> ++}
> ++
> ++/* Implement LoongArch software single step. */
> ++
> ++std::vector<CORE_ADDR>
> ++loongarch_software_single_step (struct regcache *regcache);
> ++std::vector<CORE_ADDR>
> ++loongarch_software_single_step (struct regcache *regcache)
> ++{
> ++ struct gdbarch *gdbarch = regcache->arch ();
> ++ struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch);
> ++ CORE_ADDR pc = regcache_read_pc (regcache);
> ++ std::vector<CORE_ADDR> next_pcs
> ++ = loongarch_deal_with_atomic_sequence (regcache, pc);
> ++
> ++ if (!next_pcs.empty ())
> ++ return next_pcs;
> ++
> ++ insn_t insn = loongarch_fetch_instruction (pc, NULL);
> ++ size_t insnlen = loongarch_insn_length (insn);
> ++ CORE_ADDR next = pc + insnlen;
> ++
> ++ if ((insn & 0xffff8000) == 0x002b0000 && tdep->syscall_next_pc)
> ++ {
> ++ CORE_ADDR syscall_next = tdep->syscall_next_pc (get_current_frame
> ());
> ++ if (syscall_next != -1)
> ++ {
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (gdb_stdlog,
> ++ "PC: %s Syscall found. Next pc is %s.\n",
> ++ paddress (gdbarch, pc),
> ++ paddress (gdbarch, syscall_next));
> ++ return {syscall_next};
> ++ }
> ++ }
> ++
> ++ if (loongarch_insn_is_branch (insn))
> ++ {
> ++ CORE_ADDR branch_tgt = loongarch_next_pc_if_branch (regcache, pc,
> insn);
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (
> ++ gdb_stdlog, "PC: %s Next pc is %s if branch, %s for
> non-branch.\n",
> ++ paddress (gdbarch, pc), paddress (gdbarch, branch_tgt),
> ++ paddress (gdbarch, next));
> ++ return {next, branch_tgt};
> ++ }
> ++ else
> ++ {
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (gdb_stdlog, "PC: %s Next pc is %s.\n",
> ++ paddress (gdbarch, pc), paddress (gdbarch,
> next));
> ++ return {next};
> ++ }
> ++}
> ++
> ++/* Callback function for user_reg_add. */
> ++
> ++static struct value *
> ++value_of_loongarch_user_reg (struct frame_info *frame, const void *baton)
> ++{
> ++ return value_of_register ((long long) baton, frame);
> ++}
> ++
> ++/* Implement the register_name gdbarch method. */
> ++
> ++static const char *
> ++loongarch_register_name (struct gdbarch *gdbarch, int regnum)
> ++{
> ++ auto regs = &gdbarch_tdep (gdbarch)->regs;
> ++
> ++ if ((0 <= regs->r && regs->r <= regnum && regnum < regs->r + 32)
> ++ && (EF_LOONGARCH_IS_LP64 (gdbarch_tdep (gdbarch)->ef_abi)))
> ++ return loongarch_r_lp64_name[regnum - regs->r] + 1;
> ++
> ++ else if ((0 <= regs->f && regs->f <= regnum && regnum < regs->f + 32)
> ++ && (EF_LOONGARCH_IS_LP64 (gdbarch_tdep (gdbarch)->ef_abi)))
> ++ return loongarch_f_lp64_name[regnum - regs->f] + 1;
> ++
> ++ return tdesc_register_name (gdbarch, regnum);
> ++}
> ++
> ++/* Analyze the function prologue from START_PC to LIMIT_PC. Builds
> ++ the associated FRAME_CACHE if not null.
> ++ Return the address of the first instruction past the prologue. */
> ++
> ++static CORE_ADDR
> ++loongarch_scan_prologue (struct gdbarch *gdbarch, CORE_ADDR start_pc,
> ++ CORE_ADDR limit_pc, struct frame_info *this_frame,
> ++ struct trad_frame_cache *this_cache)
> ++{
> ++ auto regs = &gdbarch_tdep (gdbarch)->regs;
> ++ int rlen_is_64b = (loongarch_rlen (gdbarch) == 64);
> ++
> ++ CORE_ADDR cur_pc, prologue_end = 0;
> ++ insn_t insn;
> ++ size_t insnlen;
> ++
> ++ int sp = regs->sp - regs->r;
> ++
> ++ long frame_offset = 0;
> ++ int non_prologue_insns = 0;
> ++ int cfa_unknown = 0;
> ++
> ++ /* Try to trace li. */
> ++ int64_t r_value[32] = {0};
> ++ int r_value_known[32] = {1, 0};
> ++
> ++ long r_cfa_offset[32] = {0};
> ++ int r_cfa_offset_p[32] = {0};
> ++
> ++ long f_cfa_offset[32] = {0};
> ++ int f_cfa_offset_p[32] = {0};
> ++
> ++ if (start_pc + 80 < limit_pc)
> ++ limit_pc = start_pc + 80;
> ++
> ++ for (cur_pc = start_pc; cur_pc < limit_pc; cur_pc += insnlen)
> ++ {
> ++ int rd, rj, rk;
> ++ int64_t si12, si20, si14;
> ++
> ++ insn = loongarch_fetch_instruction (cur_pc, NULL);
> ++ insnlen = loongarch_insn_length (insn);
> ++
> ++ rd = loongarch_decode_imm ("0:5", insn, 0);
> ++ rj = loongarch_decode_imm ("5:5", insn, 0);
> ++ rk = loongarch_decode_imm ("10:5", insn, 0);
> ++ si12 = loongarch_decode_imm ("10:12", insn, 1);
> ++ si20 = loongarch_decode_imm ("5:20", insn, 1);
> ++ si14 = loongarch_decode_imm ("10:14<<2", insn, 1);
> ++
> ++ if ((((insn & 0xffc00000) == 0x02800000 /* addi.w sp,sp,si12 */
> ++ && !rlen_is_64b)
> ++ || ((insn & 0xffc00000) == 0x02c00000 /* addi.d sp,sp,si12 */
> ++ && rlen_is_64b))
> ++ && rd == sp && rj == sp)
> ++ {
> ++ if (si12 < 0)
> ++ frame_offset -= si12;
> ++ else
> ++ /* Exit loop if a positive stack adjustment is found, which
> ++ usually means that the stack cleanup code in the function
> ++ epilogue is reached. */
> ++ break;
> ++ prologue_end = cur_pc + insnlen;
> ++ }
> ++ else if ((((insn & 0xffc00000) == 0x29800000 /* st.w rd,sp,si12 */
> ++ && !rlen_is_64b)
> ++ || ((insn & 0xffc00000) == 0x29c00000 /* st.d rd,sp,si12 */
> ++ && rlen_is_64b))
> ++ && rj == sp)
> ++ {
> ++ if (!r_cfa_offset_p[rd] && !r_value_known[rd])
> ++ r_cfa_offset[rd] = si12 - frame_offset, r_cfa_offset_p[rd] = 1;
> ++ prologue_end = cur_pc + insnlen;
> ++ }
> ++ else if ((((insn & 0xff000000) == 0x25000000 /* stptr.w rd,sp,si14
> */
> ++ && !rlen_is_64b)
> ++ || ((insn & 0xff000000) == 0x27000000 /* stptr.d
> rd,sp,si14 */
> ++ && rlen_is_64b))
> ++ && rj == sp)
> ++ {
> ++ if (!r_cfa_offset_p[rd] && !r_value_known[rd])
> ++ r_cfa_offset[rd] = si14 - frame_offset, r_cfa_offset_p[rd] = 1;
> ++ prologue_end = cur_pc + insnlen;
> ++ }
> ++ else if (((insn & 0xffc00000) == 0x2b400000 /* fst.s
> fd,sp,si12 */
> ++ || (insn & 0xffc00000) == 0x2bc00000) /* fst.d fd,sp,si12
> */
> ++ && rj == sp)
> ++ {
> ++ if (!f_cfa_offset_p[rd])
> ++ f_cfa_offset[rd] = si12 - frame_offset, f_cfa_offset_p[rd] = 1;
> ++ }
> ++ else if ((((insn & 0xffff8000) == 0x00110000 /* sub.w sp,sp,rk */
> ++ && !rlen_is_64b)
> ++ || ((insn & 0xffff8000) == 0x00118000 /* sub.d sp,sp,rk */
> ++ && rlen_is_64b))
> ++ && rd == sp && rj == sp)
> ++ {
> ++ if (r_value_known[rk])
> ++ {
> ++ frame_offset += r_value[rk];
> ++ prologue_end = cur_pc + insnlen;
> ++ }
> ++ else
> ++ cfa_unknown = 1;
> ++ }
> ++ else if ((((insn & 0xffff8000) == 0x00100000 /* add.w sp,sp,rk */
> ++ && !rlen_is_64b)
> ++ || ((insn & 0xffff8000) == 0x00108000 /* add.d sp,sp,rk */
> ++ && rlen_is_64b))
> ++ && rd == sp && rj == sp)
> ++ {
> ++ if (r_value_known[rk] && r_value[rk] < 0)
> ++ {
> ++ frame_offset -= r_value[rk];
> ++ prologue_end = cur_pc + insnlen;
> ++ }
> ++ else
> ++ cfa_unknown = 1;
> ++ }
> ++ else if ((insn & 0xffff8000) == 0x00150000 /* or rd,sp,$r0 */
> ++ && rj == sp && rk == 0)
> ++ {
> ++ sp = rd;
> ++ prologue_end = cur_pc + insnlen;
> ++ }
> ++ else if ((insn & 0xffc00000) == 0x02800000) /* addi.w rd,rj,si12 */
> ++ {
> ++ if (r_value_known[rj] && rd != 0)
> ++ r_value[rd] = (int32_t) (r_value[rj] + si12),
> ++ r_value_known[rd] = 1;
> ++ }
> ++ else if ((insn & 0xffc00000) == 0x03800000) /* ori rd,rj,si12 */
> ++ {
> ++ if (r_value_known[rj] && rd != 0)
> ++ r_value[rd] = r_value[rj] | (si12 & 0xfff), r_value_known[rd]
> = 1;
> ++ }
> ++ else if ((insn & 0xfe000000) == 0x14000000) /* lu12i.w rd,si20 */
> ++ {
> ++ if (rd != 0)
> ++ r_value[rd] = si20 << 12, r_value_known[rd] = 1;
> ++ }
> ++ else if ((insn & 0xfe000000) == 0x16000000) /* lu32i.d rd,si20 */
> ++ {
> ++ if (r_value_known[rd] && rd != 0)
> ++ r_value[rd] = (r_value[rd] & 0xffffffff) | (si20 << 32),
> ++ r_value_known[rd] = 1;
> ++ }
> ++ else if ((insn & 0xffc00000) == 0x03000000) /* lu52i.d rd,rj,si12
> */
> ++ {
> ++ if (r_value_known[rj] && rd != 0)
> ++ r_value[rd] = (r_value[rj] & 0xfffffffffffff) | (si12 << 52),
> ++ r_value_known[rd] = 1;
> ++ }
> ++ else if (loongarch_insn_is_branch (insn))
> ++ break; /* Shrink-wrap or end of prologue in a basic block. */
> ++ else
> ++ non_prologue_insns++;
> ++
> ++ /* 4 INSNs for 'la' and one for some other. */
> ++ if (5 < non_prologue_insns)
> ++ break;
> ++ }
> ++
> ++ if (loongarch_debug)
> ++ {
> ++ const char *fun_name;
> ++ find_pc_partial_function (start_pc, &fun_name, NULL, NULL);
> ++ fprintf_unfiltered (gdb_stdlog,
> ++ "Prologue Analyze: -- Start -- Callee [%s] %s\n",
> ++ fun_name ? fun_name : "<unknown>",
> ++ paddress (gdbarch, start_pc));
> ++ }
> ++
> ++ do
> ++ {
> ++ int i;
> ++ CORE_ADDR cfa = -1;
> ++
> ++ if (!(this_frame && this_cache))
> ++ break;
> ++
> ++ if (!cfa_unknown)
> ++ {
> ++ try
> ++ {
> ++ cfa = get_frame_register_signed (this_frame, regs->r + sp)
> ++ + frame_offset;
> ++ }
> ++ catch (const gdb_exception_error &ex)
> ++ {
> ++ cfa_unknown = 1;
> ++ if (ex.error != NOT_AVAILABLE_ERROR)
> ++ throw;
> ++ }
> ++
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (
> ++ gdb_stdlog,
> ++ "Prologue Analyze: CFA is (frame pointer $%s + 0x%lx) =
> %s\n",
> ++ gdbarch_register_name (gdbarch, regs->r + sp),
> ++ (long) frame_offset,
> ++ cfa_unknown ? "<unknown>" : paddress (gdbarch, cfa));
> ++ }
> ++ else if (loongarch_debug)
> ++ fprintf_unfiltered (gdb_stdlog,
> ++ "Prologue Analyze: Unknown stack frame size,
> so "
> ++ "can't get known CFA\n");
> ++
> ++ if (r_cfa_offset_p[1] && !cfa_unknown)
> ++ {
> ++ CORE_ADDR ret_saved = cfa + r_cfa_offset[1];
> ++ trad_frame_set_reg_addr (this_cache, gdbarch_pc_regnum (gdbarch),
> ++ ret_saved);
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (
> ++ gdb_stdlog,
> ++ "Prologue Analyze: Return addr saved in (CFA - 0x%lx) =
> %s\n",
> ++ -r_cfa_offset[1], paddress (gdbarch, ret_saved));
> ++ }
> ++ else if (r_cfa_offset_p[1] /* && cfa_unknown */)
> ++ {
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (gdb_stdlog,
> ++ "Prologue Analyze: Return addr saved in
> (CFA "
> ++ "- 0x%lx), but CFA is unknown\n",
> ++ -r_cfa_offset[1]);
> ++ }
> ++ else
> ++ {
> ++ trad_frame_set_reg_realreg (this_cache, gdbarch_pc_regnum
> (gdbarch),
> ++ regs->r + 1);
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (gdb_stdlog,
> ++ "Prologue Analyze: No found $r1 pushed in "
> ++ "stack. Return addr saved in $r1\n");
> ++ }
> ++
> ++ if (cfa_unknown)
> ++ {
> ++ trad_frame_set_this_base (this_cache, -1);
> ++ break;
> ++ }
> ++
> ++ trad_frame_set_reg_value (this_cache, gdbarch_sp_regnum (gdbarch),
> ++ (LONGEST) cfa);
> ++ trad_frame_set_this_base (this_cache, cfa);
> ++
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (
> ++ gdb_stdlog,
> ++ "Prologue Analyze: Where caller's registers saved as follow:\n");
> ++
> ++ for (i = 0; i < 32; i++)
> ++ if (r_cfa_offset_p[i] && i != 1)
> ++ {
> ++ trad_frame_set_reg_addr (this_cache, regs->r + i,
> ++ cfa + r_cfa_offset[i]);
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (
> ++ gdb_stdlog,
> ++ "Prologue Analyze: $%s: saved in (CFA - 0x%lx) = %s\n",
> ++ gdbarch_register_name (gdbarch, regs->r + i),
> -r_cfa_offset[i],
> ++ paddress (gdbarch, cfa + r_cfa_offset[i]));
> ++ }
> ++
> ++ if (regs->f <= 0)
> ++ for (i = 0; i < 32; i++)
> ++ {
> ++ if (f_cfa_offset_p[i])
> ++ trad_frame_set_reg_addr (this_cache, regs->f + i,
> ++ cfa + f_cfa_offset[i]);
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (
> ++ gdb_stdlog,
> ++ "Prologue Analyze: $%s: saved in (CFA - 0x%lx) = %s\n",
> ++ gdbarch_register_name (gdbarch, regs->f + i),
> -f_cfa_offset[i],
> ++ paddress (gdbarch, cfa + f_cfa_offset[i]));
> ++ }
> ++ }
> ++ while (0);
> ++
> ++ if (loongarch_debug)
> ++ fprintf_unfiltered (gdb_stdlog, "Prologue Analyze: -- End -- %s\n",
> ++ paddress (gdbarch, cur_pc));
> ++
> ++ return prologue_end ? prologue_end : cur_pc;
> ++}
> ++
> ++/* Implement the loongarch_skip_prologue gdbarch method. */
> ++
> ++/* To skip prologues, I use this predicate. Returns either PC itself
> ++ if the code at PC does not look like a function prologue; otherwise
> ++ returns an address that (if we're lucky) follows the prologue. If
> ++ LENIENT, then we must skip everything which is involved in setting
> ++ up the frame (it's OK to skip more, just so long as we don't skip
> ++ anything which might clobber the registers which are being saved.
> ++ We must skip more in the case where part of the prologue is in the
> ++ delay slot of a non-prologue instruction). */
> ++
> ++static CORE_ADDR
> ++loongarch_skip_prologue (struct gdbarch *gdbarch, CORE_ADDR pc)
> ++{
> ++ CORE_ADDR limit_pc;
> ++ CORE_ADDR func_addr;
> ++
> ++ /* See if we can determine the end of the prologue via the symbol
> table.
> ++ If so, then return either PC, or the PC after the prologue,
> whichever
> ++ is greater. */
> ++ if (find_pc_partial_function (pc, NULL, &func_addr, NULL))
> ++ {
> ++ CORE_ADDR post_prologue_pc
> ++ = skip_prologue_using_sal (gdbarch, func_addr);
> ++ if (post_prologue_pc != 0)
> ++ return std::max (pc, post_prologue_pc);
> ++ }
> ++
> ++ /* Can't determine prologue from the symbol table, need to examine
> ++ instructions. */
> ++
> ++ /* Find an upper limit on the function prologue using the debug
> ++ information. If the debug information could not be used to provide
> ++ that bound, then use an arbitrary large number as the upper bound.
> */
> ++ limit_pc = skip_prologue_using_sal (gdbarch, pc);
> ++ if (limit_pc == 0)
> ++ limit_pc = pc + 100; /* Magic. */
> ++
> ++ return loongarch_scan_prologue (gdbarch, pc, limit_pc, NULL, NULL);
> ++}
> ++
> ++/* Adjust the address downward (direction of stack growth) so that it
> ++ is correctly aligned for a new stack frame. */
> ++static CORE_ADDR
> ++loongarch_frame_align (struct gdbarch *gdbarch, CORE_ADDR addr)
> ++{
> ++ return align_down (addr, 16);
> ++}
> ++
> ++/* Implement the unwind_pc gdbarch method. */
> ++
> ++static CORE_ADDR
> ++loongarch_unwind_pc (struct gdbarch *gdbarch, struct frame_info
> *next_frame)
> ++{
> ++ return frame_unwind_register_signed (next_frame,
> ++ gdbarch_pc_regnum (gdbarch));
> ++}
> ++
> ++/* Implement the unwind_sp gdbarch method. */
> ++
> ++static CORE_ADDR
> ++loongarch_unwind_sp (struct gdbarch *gdbarch, struct frame_info
> *next_frame)
> ++{
> ++ return frame_unwind_register_signed (next_frame,
> ++ gdbarch_sp_regnum (gdbarch));
> ++}
> ++
> ++/* Implement the dummy_id gdbarch method. */
> ++
> ++static struct frame_id
> ++loongarch_dummy_id (struct gdbarch *gdbarch, struct frame_info
> *this_frame)
> ++{
> ++ return frame_id_build (
> ++ get_frame_register_signed (this_frame, gdbarch_sp_regnum (gdbarch)),
> ++ get_frame_pc (this_frame));
> ++}
> ++
> ++/* Generate, or return the cached frame cache for the loongarch frame
> ++ unwinder. */
> ++
> ++static struct trad_frame_cache *
> ++loongarch_frame_cache (struct frame_info *this_frame, void **this_cache)
> ++{
> ++ struct gdbarch *gdbarch = get_frame_arch (this_frame);
> ++ struct trad_frame_cache *cache;
> ++ CORE_ADDR pc, start_addr, stack_addr;
> ++
> ++ if (*this_cache != NULL)
> ++ return (struct trad_frame_cache *) *this_cache;
> ++ cache = trad_frame_cache_zalloc (this_frame);
> ++ *this_cache = cache;
> ++
> ++ pc = get_frame_address_in_block (this_frame);
> ++ if (find_pc_partial_function (pc, NULL, &start_addr, NULL))
> ++ {
> ++ loongarch_scan_prologue (gdbarch, start_addr, pc, this_frame,
> cache);
> ++ stack_addr = trad_frame_get_this_base (cache);
> ++ trad_frame_set_id (cache,
> ++ stack_addr == -1
> ++ ? frame_id_build_unavailable_stack (start_addr)
> ++ : frame_id_build (stack_addr, start_addr));
> ++ }
> ++ else
> ++ {
> ++ auto regs = &gdbarch_tdep (gdbarch)->regs;
> ++ trad_frame_set_reg_realreg (cache, regs->ra, -2 /* TF_REG_UNKNOWN
> */);
> ++ trad_frame_set_reg_realreg (cache, gdbarch_pc_regnum (gdbarch),
> ++ regs->ra);
> ++
> ++ trad_frame_set_id (cache, frame_id_build_unavailable_stack (pc));
> ++ }
> ++ return cache;
> ++}
> ++
> ++/* Implement the this_id callback for loongarch frame unwinder. */
> ++
> ++static void
> ++loongarch_frame_this_id (struct frame_info *this_frame, void
> **prologue_cache,
> ++ struct frame_id *this_id)
> ++{
> ++ struct trad_frame_cache *info;
> ++
> ++ info = loongarch_frame_cache (this_frame, prologue_cache);
> ++ trad_frame_get_id (info, this_id);
> ++}
> ++
> ++/* Implement the prev_register callback for loongarch frame unwinder. */
> ++
> ++static struct value *
> ++loongarch_frame_prev_register (struct frame_info *this_frame,
> ++ void **prologue_cache, int regnum)
> ++{
> ++ struct trad_frame_cache *info;
> ++
> ++ info = loongarch_frame_cache (this_frame, prologue_cache);
> ++ return trad_frame_get_register (info, this_frame, regnum);
> ++}
> ++
> ++static const struct frame_unwind loongarch_frame_unwind = {
> ++ /*.type =*/NORMAL_FRAME,
> ++ /*.stop_reason =*/default_frame_unwind_stop_reason,
> ++ /*.this_id =*/loongarch_frame_this_id,
> ++ /*.prev_register =*/loongarch_frame_prev_register,
> ++ /*.unwind_data =*/NULL,
> ++ /*.sniffer =*/default_frame_sniffer,
> ++};
> ++
> ++typedef struct stack_data_t
> ++{
> ++ const gdb_byte *addr = NULL;
> ++ int len = 0;
> ++ bool ref = false;
> ++} stack_data_t;
> ++
> ++static void
> ++pass_on_stack (std::vector<stack_data_t> &stack, const gdb_byte *val,
> int len,
> ++ int align, bool ref = false)
> ++{
> ++ stack_data_t buf;
> ++ buf.addr = val;
> ++ buf.len = align_up (len, align);
> ++ buf.ref = ref;
> ++
> ++ stack.push_back (buf);
> ++}
> ++
> ++static void
> ++pass_on_reg (struct regcache *regcache, int regno, const gdb_byte *val,
> ++ int len)
> ++{
> ++ gdb_byte reg[32];
> ++ memset (reg, 0, sizeof (reg));
> ++ memcpy (reg, val, len);
> ++ regcache->cooked_write (regno, reg);
> ++}
> ++
> ++static void
> ++compute_type_num(struct type *tp, int &complex_num, int &float_num,
> ++ int &other_num, int &counter, int &float_seq,
> ++ int &other_seq)
> ++{
> ++ if (tp->code () == TYPE_CODE_COMPLEX)
> ++ complex_num++;
> ++ else if (tp->code () == TYPE_CODE_FLT)
> ++ float_num++;
> ++ else if (tp->code () != TYPE_CODE_STRUCT)
> ++ other_num++;
> ++
> ++ /* When the function parameter or return value type is a structure,
> ++ traverse each member in the structure and make relevant marks. */
> ++ for (int i = 0; i < tp->num_fields (); i++)
> ++ {
> ++ field fd = tp->field (i);
> ++ struct type *t = fd.type ();
> ++
> ++ /* Call check_typedef(TYPE_TARGET_TYPE (TYPE)) on our type to make
> ++ sure that, if TYPE is a TYPE_CODE_TYPEDEF, its TYPE is set to
> ++ the target type instead of TYPE_CODE_TYPEDEF. */
> ++ if (t->code () == TYPE_CODE_TYPEDEF)
> ++ t = check_typedef (TYPE_TARGET_TYPE (t));
> ++
> ++ switch (t->code ())
> ++ {
> ++ case TYPE_CODE_STRUCT:
> ++ compute_type_num(t, complex_num, float_num, other_num,
> ++ counter, float_seq, other_seq);
> ++ break;
> ++ case TYPE_CODE_COMPLEX:
> ++ complex_num++;
> ++ break;
> ++ case TYPE_CODE_FLT:
> ++ float_num++;
> ++ float_seq = ++counter;
> ++ break;
> ++ default:
> ++ other_num++;
> ++ other_seq = ++counter;
> ++ break;
> ++ }
> ++ }
> ++}
> ++
> ++static void
> ++pass_small_struct_on_reg (struct gdbarch *gdbarch, struct type *tp,
> ++ const gdb_byte *data,
> std::vector<stack_data_t> &gp,
> ++ std::vector<stack_data_t> &fp)
> ++{
> ++ const int rlen = loongarch_rlen (gdbarch) / 8;
> ++ int len = TYPE_LENGTH (tp);
> ++ int complex_num = 0, float_num = 0, other_num = 0;
> ++ int counter = 0, float_seq = 0, other_seq = 0;
> ++ stack_data_t elm;
> ++
> ++ gdb_assert (len <= 2 * rlen);
> ++
> ++ /* Compute the relevant members and types in the function parameters
> ++ and mark them. */
> ++ compute_type_num(tp, complex_num, float_num, other_num,
> ++ counter, float_seq, other_seq);
> ++
> ++ if (other_num > 0 && float_num == 0 && len <= rlen)
> ++ {
> ++ /* For the small structure has only other types (like
> char/short/int/long
> ++ etc.), and the size does not exceed rlen, pass on one gp or
> stack. */
> ++ elm.addr = data;
> ++ elm.len = rlen;
> ++ gp.push_back (elm);
> ++ }
> ++ else if (float_num == 1 && complex_num == 0 && other_num == 0 && len
> <= rlen)
> ++ {
> ++ /* For the small structure has only floating-point (like
> float/double),
> ++ and the size does not exceed rlen, pass on one fp or stack. */
> ++ elm.addr = data;
> ++ elm.len = rlen;
> ++ fp.push_back (elm);
> ++ }
> ++ else if (float_num == 1 && other_num == 1)
> ++ {
> ++ /* For the small structure has only one floating-point type and
> ++ one other type(like float and int, char and double etc.), the
> ++ floating-point type passes through one fp or stack, and the
> ++ other types pass on one gp or stack. */
> ++ if (float_seq < other_seq)
> ++ {
> ++ /* Floating point first, first pass on fp, then gp. */
> ++ elm.addr = data;
> ++ if (len == rlen)
> ++ elm.len = rlen / 2;
> ++ else
> ++ elm.len = rlen;
> ++ fp.push_back (elm);
> ++ elm.addr += elm.len;
> ++ gp.push_back (elm);
> ++ }
> ++ else
> ++ {
> ++ /* Floating point after, first pass on gp, then fp. */
> ++ elm.addr = data;
> ++ if (len == rlen)
> ++ elm.len = rlen / 2;
> ++ else
> ++ elm.len = rlen;
> ++ gp.push_back (elm);
> ++ elm.addr += elm.len;
> ++ fp.push_back (elm);
> ++ }
> ++ }
> ++ else if ((complex_num == 1 && float_num == 0 && other_num == 0) ||
> ++ (float_num ==2 && other_num == 0))
> ++ {
> ++ /* For the small structure has only two floating-point types or
> ++ * one complex number type, pass on two fp or stack. */
> ++ elm.addr = data;
> ++ /* 2 float or 1 'float _Complex'. */
> ++ if (len == rlen)
> ++ elm.len = rlen / 2;
> ++ /* 2 double or 1 'double _Complex'. */
> ++ else
> ++ elm.len = rlen;
> ++ fp.push_back (elm);
> ++ elm.addr += elm.len;
> ++ fp.push_back (elm);
> ++ }
> ++ else
> ++ {
> ++ /* For other cases, pass on two gp or stack. */
> ++ /* For example, the small structure is of the following type,
> ++ 1. with more than 2 other types and the size is greater
> than rlen
> ++ (like struct{int; int; int;}; struct{long; int; short;
> char;}; etc.).
> ++ 2. with 'long double' on fpu64 or 'double' on fpu32
> ++ (like struct{long double;}; or struct{double;}; etc.).
> ++ 3. with more than 2 floating-point types
> ++ (like struct{float; float; float;}; struct{float; float;
> double;};
> ++ struct{float; float; float; float;}; etc.)
> ++ 4. with 2 'float _Complex'
> ++ (like struct{float _Complex; float _Complex;} etc.). */
> ++ elm.addr = data;
> ++ elm.len = rlen;
> ++ gp.push_back (elm);
> ++ elm.addr += elm.len;
> ++ gp.push_back (elm);
> ++ }
> ++}
> ++
> ++static bool
> ++try_pass_small_struct_on_reg (struct gdbarch *gdbarch,
> ++ struct regcache *regcache, struct value *arg,
> ++ int &gp, int &fp, int gp_max, int fp_max)
> ++{
> ++ const int rlen = loongarch_rlen (gdbarch) / 8;
> ++ struct type *a_type = check_typedef (value_type (arg));
> ++ int len = TYPE_LENGTH (a_type);
> ++ const gdb_byte *val = value_contents (arg);
> ++
> ++ std::vector<stack_data_t> gpreg;
> ++ std::vector<stack_data_t> fpreg;
> ++
> ++ gdb_assert (len <= 2 * rlen);
> ++ // gdb_assert (a_type->code () == TYPE_CODE_STRUCT);
> ++
> ++ pass_small_struct_on_reg (gdbarch, a_type, val, gpreg, fpreg);
> ++
> ++ if (gp + gpreg.size () - 1 < gp_max && fp + fpreg.size () - 1 < fp_max)
> ++ {
> ++ for (auto it : gpreg)
> ++ {
> ++ pass_on_reg (regcache, gp, it.addr, it.len);
> ++ gp++;
> ++ }
> ++ for (auto it : fpreg)
> ++ {
> ++ pass_on_reg (regcache, fp, it.addr, it.len);
> ++ fp++;
> ++ }
> ++ return true;
> ++ }
> ++ return false;
> ++}
> ++
> ++/* Implement the push dummy call gdbarch callback. */
> ++
> ++static CORE_ADDR
> ++loongarch_lp32lp64_push_dummy_call (
> ++ struct gdbarch *gdbarch, struct value *function, struct regcache
> *regcache,
> ++ CORE_ADDR bp_addr, int nargs, struct value **args, CORE_ADDR sp,
> ++ function_call_return_method return_method, CORE_ADDR struct_addr)
> ++{
> ++ const int rlen = loongarch_rlen (gdbarch) / 8;
> ++ auto regs = &gdbarch_tdep (gdbarch)->regs;
> ++ int gp = regs->r + 4; /* $a0 = $r4 = regs->r + 4 */
> ++ int fp = regs->f; /* $fa0 */
> ++ const int gp_max = gp + 8; /* gpr $a0 ~ $a7 ($r4 ~ $r11) */
> ++ const int fp_max = fp + 8; /* fpr $fa0 ~ $fa7 */
> ++ std::vector<stack_data_t> stack;
> ++ int vec_insn = 0;
> ++
> ++ {
> ++ if (return_method != return_method_normal)
> ++ {
> ++ regcache_cooked_write_unsigned (regcache, gp++, struct_addr);
> ++ }
> ++
> ++ if (return_method == return_method_hidden_param)
> ++ {
> ++ args++;
> ++ nargs--;
> ++ }
> ++ }
> ++ regcache_cooked_write_signed (regcache, regs->ra, bp_addr);
> ++
> ++ struct type *f_type = check_typedef (value_type (function));
> ++
> ++ for (int i = 0; i < nargs; i++)
> ++ {
> ++ struct value *arg = args[i];
> ++ struct type *a_type = check_typedef (value_type (arg));
> ++ int len = TYPE_LENGTH (a_type);
> ++ const gdb_byte *val = value_contents (arg);
> ++
> ++ switch (a_type->code ())
> ++ {
> ++ case TYPE_CODE_INT:
> ++ case TYPE_CODE_BOOL:
> ++ case TYPE_CODE_CHAR:
> ++ case TYPE_CODE_RANGE:
> ++ case TYPE_CODE_ENUM:
> ++ case TYPE_CODE_PTR:
> ++ if (gp < gp_max)
> ++ {
> ++ if (TYPE_UNSIGNED (a_type))
> ++ {
> ++ ULONGEST data
> ++ = extract_unsigned_integer (val, len,
> BFD_ENDIAN_LITTLE);
> ++ regcache_cooked_write_unsigned (regcache, gp++, data);
> ++ }
> ++ else
> ++ {
> ++ LONGEST data
> ++ = extract_signed_integer (val, len, BFD_ENDIAN_LITTLE);
> ++ regcache_cooked_write_signed (regcache, gp++, data);
> ++ }
> ++ }
> ++ else
> ++ {
> ++ pass_on_stack (stack, val, len, rlen);
> ++ }
> ++ break;
> ++ case TYPE_CODE_FLT:
> ++ if (len <= rlen)
> ++ {
> ++ if (!TYPE_VARARGS (f_type) && (fp < fp_max))
> ++ pass_on_reg (regcache, fp++, val, len);
> ++ else if (gp < gp_max)
> ++ pass_on_reg (regcache, gp++, val, len);
> ++ else
> ++ pass_on_stack (stack, val, len, rlen);
> ++ }
> ++ /* Long double like struct. */
> ++ else
> ++ {
> ++ if (gp < gp_max - 1)
> ++ {
> ++ pass_on_reg (regcache, gp++, val, rlen);
> ++ pass_on_reg (regcache, gp++, val + rlen, len - rlen);
> ++ }
> ++ else
> ++ pass_on_stack (stack, val, len, rlen);
> ++ }
> ++ break;
> ++ case TYPE_CODE_ARRAY:
> ++ /* lsx */
> ++ if (TYPE_VECTOR (a_type) && len == vec_insn && vec_insn == 16
> ++ && fp < fp_max)
> ++ {
> ++ pass_on_reg (regcache, regs->vr + (fp++ - regs->f), val,
> len);
> ++ }
> ++ /* lasx */
> ++ else if (TYPE_VECTOR (a_type) && len == vec_insn && vec_insn ==
> 32
> ++ && fp < fp_max)
> ++ {
> ++ pass_on_reg (regcache, regs->xr + (fp++ - regs->f), val,
> len);
> ++ }
> ++ /* scalar */
> ++ else
> ++ {
> ++ if (len > rlen * 2)
> ++ {
> ++ /* Address on register, data on stack. */
> ++ sp = align_down (sp - len, rlen);
> ++ write_memory (sp, val, len);
> ++ if (gp < gp_max)
> ++ pass_on_reg (regcache, gp++, (const gdb_byte *) &sp,
> rlen);
> ++ else
> ++ pass_on_stack (stack, (const gdb_byte *) sp, rlen,
> rlen,
> ++ true);
> ++ }
> ++ else
> ++ {
> ++ if (len <= rlen && gp < gp_max)
> ++ {
> ++ pass_on_reg (regcache, gp++, val, len);
> ++ }
> ++ else if (gp + 1 < gp_max)
> ++ {
> ++ pass_on_reg (regcache, gp++, val, rlen);
> ++ pass_on_reg (regcache, gp++, val + rlen, rlen);
> ++ }
> ++ else
> ++ {
> ++ pass_on_stack (stack, val, len, rlen);
> ++ }
> ++ }
> ++ }
> ++ break;
> ++ case TYPE_CODE_STRUCT:
> ++ case TYPE_CODE_UNION:
> ++ if (len > rlen * 2)
> ++ {
> ++ /* Address on register, data on stack. */
> ++ sp = align_down (sp - len, rlen);
> ++ write_memory (sp, val, len);
> ++ if (gp < gp_max)
> ++ pass_on_reg (regcache, gp++, (const gdb_byte *) &sp, rlen);
> ++ else
> ++ pass_on_stack (stack, (const gdb_byte *) sp, rlen, rlen,
> true);
> ++ }
> ++ else
> ++ {
> ++ if (!try_pass_small_struct_on_reg (gdbarch, regcache, arg,
> gp,
> ++ fp, gp_max, fp_max))
> ++ {
> ++ pass_on_stack (stack, val, len, rlen);
> ++ }
> ++ }
> ++ break;
> ++ case TYPE_CODE_COMPLEX:
> ++ {
> ++ /* Two fpr or mem. */
> ++ struct type *t_type = check_typedef (TYPE_TARGET_TYPE
> (a_type));
> ++ int tlen = TYPE_LENGTH (t_type);
> ++
> ++ if (tlen < rlen)
> ++ {
> ++ if (!TYPE_VARARGS (f_type) && fp + 1 < fp_max)
> ++ {
> ++ pass_on_reg (regcache, fp++, (const gdb_byte *) val,
> tlen);
> ++ pass_on_reg (regcache, fp++, (const gdb_byte *) val +
> tlen,
> ++ tlen);
> ++ }
> ++ else if (gp < gp_max)
> ++ {
> ++ pass_on_reg (regcache, gp++, (const gdb_byte *) val,
> rlen);
> ++ }
> ++ else
> ++ {
> ++ pass_on_stack (stack, val, len, rlen);
> ++ }
> ++ }
> ++ else if (tlen == rlen)
> ++ {
> ++ if (!TYPE_VARARGS (f_type) && fp + 1 < fp_max)
> ++ {
> ++ pass_on_reg (regcache, fp++, (const gdb_byte *) val,
> tlen);
> ++ pass_on_reg (regcache, fp++, (const gdb_byte *) val +
> tlen,
> ++ tlen);
> ++ }
> ++ else if (gp + 1 < gp_max)
> ++ {
> ++ pass_on_reg (regcache, gp++, (const gdb_byte *) val,
> rlen);
> ++ pass_on_reg (regcache, gp++, (const gdb_byte *) val +
> rlen,
> ++ rlen);
> ++ }
> ++ else if (gp + 1 == gp_max)
> ++ {
> ++ pass_on_reg (regcache, gp++, (const gdb_byte *) val,
> rlen);
> ++ pass_on_stack (stack, val, tlen, rlen);
> ++ }
> ++ else
> ++ {
> ++ pass_on_stack (stack, val, len, rlen);
> ++ }
> ++ }
> ++ else
> ++ {
> ++ sp = align_down (sp - len, rlen);
> ++ write_memory (sp, val, len);
> ++ if (gp < gp_max)
> ++ pass_on_reg (regcache, gp++, (const gdb_byte *) &sp,
> rlen);
> ++ else
> ++ {
> ++ pass_on_stack (stack, (const gdb_byte *) sp, rlen,
> rlen,
> ++ true);
> ++ }
> ++ }
> ++ }
> ++ break;
> ++ default:
> ++ break;
> ++ }
> ++ }
> ++
> ++ for (auto it : stack)
> ++ sp = align_down (sp - it.len, rlen);
> ++
> ++ sp = align_down (sp, 16);
> ++ CORE_ADDR tsp = sp;
> ++ for (auto it : stack)
> ++ {
> ++ if (it.ref)
> ++ write_memory (tsp, (const gdb_byte *) &it.addr, it.len);
> ++ else
> ++ write_memory (tsp, it.addr, it.len);
> ++ tsp += it.len;
> ++ stack.pop_back ();
> ++ }
> ++ regcache_cooked_write_unsigned (regcache, regs->sp, sp);
> ++ return sp;
> ++}
> ++
> ++static void
> ++loongarch_xfer_reg_part (struct regcache *regcache, int reg_num, int len,
> ++ gdb_byte *readbuf, size_t readbuf_off,
> ++ const gdb_byte *writebuf, size_t writebuf_off)
> ++{
> ++ if (readbuf)
> ++ regcache->cooked_read_part (reg_num, 0, len, readbuf + readbuf_off);
> ++ if (writebuf)
> ++ regcache->cooked_write_part (reg_num, 0, len, writebuf +
> writebuf_off);
> ++}
> ++
> ++static enum return_value_convention
> ++loongarch_lp64_return_value (struct gdbarch *gdbarch, struct value
> *function,
> ++ struct type *type, struct regcache *regcache,
> ++ gdb_byte *readbuf, const gdb_byte *writebuf)
> ++{
> ++ const size_t rlen = loongarch_rlen (gdbarch) / 8;
> ++ auto regs = &gdbarch_tdep (gdbarch)->regs;
> ++ size_t len = TYPE_LENGTH (type);
> ++ enum type_code typecode = type->code ();
> ++ int fpu_exist = 0 <= regs->f;
> ++ int fv = fpu_exist ? regs->f : regs->r + 4;
> ++
> ++ gdb_assert (8 <= sizeof (LONGEST));
> ++
> ++ gdb_assert (!fpu_exist || register_size (gdbarch, regs->f) == rlen);
> ++
> ++ if (2 * rlen < len)
> ++ return RETURN_VALUE_STRUCT_CONVENTION;
> ++
> ++ if (((typecode == TYPE_CODE_INT && TYPE_UNSIGNED (type))
> ++ || typecode == TYPE_CODE_ENUM)
> ++ && len <= rlen)
> ++ /* For unsigned scalar type, we have zero-extended one in $v0. */
> ++ if (writebuf)
> ++ {
> ++ gdb_byte buf[rlen];
> ++ store_signed_integer (buf, rlen, BFD_ENDIAN_LITTLE,
> ++ extract_unsigned_integer (writebuf, len,
> ++
> BFD_ENDIAN_LITTLE));
> ++ loongarch_xfer_reg_part (regcache, regs->r + 4, rlen, NULL, 0,
> ++ writebuf, 0);
> ++ }
> ++ else
> ++ loongarch_xfer_reg_part (regcache, regs->r + 4, len, readbuf, 0,
> NULL,
> ++ 0);
> ++ else if (((typecode == TYPE_CODE_INT && !TYPE_UNSIGNED (type))
> ++ || typecode == TYPE_CODE_PTR)
> ++ && len <= rlen)
> ++ /* For signed scalar type, we have sign-extended one in $v0. */
> ++ if (writebuf)
> ++ {
> ++ gdb_byte buf[rlen];
> ++ store_signed_integer (buf, rlen, BFD_ENDIAN_LITTLE,
> ++ extract_signed_integer (writebuf, len,
> ++ BFD_ENDIAN_LITTLE));
> ++ loongarch_xfer_reg_part (regcache, regs->r + 4, rlen, NULL, 0,
> ++ writebuf, 0);
> ++ }
> ++ else
> ++ loongarch_xfer_reg_part (regcache, regs->r + 4, len, readbuf, 0,
> NULL,
> ++ 0);
> ++ else
> ++ {
> ++ int complex_num = 0, float_num = 0, other_num = 0;
> ++ int counter = 0, float_seq = 0, other_seq = 0, tlen;
> ++ /* Calculate the relevant members and types in the return value
> ++ and mark them. */
> ++ compute_type_num(type, complex_num, float_num, other_num,
> ++ counter, float_seq, other_seq);
> ++
> ++ if (len == rlen)
> ++ tlen = rlen / 2;
> ++ else
> ++ tlen = rlen;
> ++
> ++ /* For the small structure has only other types members (like
> char/short/int/long
> ++ etc.), and the size does not exceed rlen, pass on $v0. */
> ++ /* For 'char/short/int/long' etc. pass on $v0. */
> ++ if (other_num > 0 && float_num == 0 && len <= rlen)
> ++ loongarch_xfer_reg_part (regcache, regs->r + 4, len, readbuf, 0,
> ++ writebuf, 0);
> ++ /* For small structure with only one floating-point member, (like
> float/double) pass on $fv0. */
> ++ /* For float/double pass on $fv0. */
> ++ else if (float_num == 1 && complex_num == 0 && other_num == 0 &&
> len <= rlen)
> ++ loongarch_xfer_reg_part (regcache, fv, len, readbuf, 0, writebuf,
> 0);
> ++ /* For small structure with one float/double member and one other
> member
> ++ (char/short/int/long etc.). If the float/dobule member is in the
> front
> ++ position, the float/dobule member pass on $fv0, the other member
> pass
> ++ on $v0, otherwise the opposite . */
> ++ else if (float_num == 1 && other_num == 1)
> ++ if (float_seq < other_seq)
> ++ loongarch_xfer_reg_part (regcache, fv, rlen, readbuf, 0,
> ++ writebuf, 0),
> ++ loongarch_xfer_reg_part (regcache, regs->r + 4, tlen, readbuf,
> ++ tlen, writebuf, rlen);
> ++ else
> ++ loongarch_xfer_reg_part (regcache, regs->r + 4, rlen, readbuf, 0,
> ++ writebuf, 0),
> ++ loongarch_xfer_reg_part (regcache, fv, tlen, readbuf,
> ++ tlen, writebuf, rlen);
> ++ /* For small structure with one 'float/double _Complex' member,
> ++ $fv0 is real and $fv1 is img. */
> ++ /* For small structure with only one float and double member or
> ++ or two float member , or two dobule member, $fv0 is the 1st
> ++ member and $fv1 is the 2nd member. */
> ++ /* For 'float/double _Complex', $fv0 is real and $fv1 is img. */
> ++ else if ((complex_num == 1 && float_num == 0 && other_num == 0) ||
> ++ (float_num ==2 && other_num == 0))
> ++ loongarch_xfer_reg_part (regcache, fv, rlen, readbuf, 0,
> ++ writebuf, 0),
> ++ loongarch_xfer_reg_part (regcache, fv + 1, tlen, readbuf,
> ++ tlen, writebuf, rlen);
> ++ /* For small structure with 'long double' member,
> ++ or when the small structure has more than two vaild members
> ++ and the size is greater than rlen, pass on $v0 and $v1. */
> ++ /* For small structure with two 'float _Complex' member,
> ++ $v0 is the 1st member and $v1 is the 2nd member. */
> ++ /* For 'long double' on fpu64 or 'double' on fpu32 pass on $v0 and
> $v1. */
> ++ else
> ++ loongarch_xfer_reg_part (regcache, regs->r + 4, rlen, readbuf, 0,
> ++ writebuf, 0),
> ++ loongarch_xfer_reg_part (regcache, regs->r + 5, len - rlen,
> readbuf,
> ++ rlen, writebuf, rlen);
> ++ }
> ++
> ++ return RETURN_VALUE_REGISTER_CONVENTION;
> ++}
> ++
> ++static int
> ++loongarch_dwarf2_reg_to_regnum (struct gdbarch *gdbarch, int num)
> ++{
> ++ auto regs = &gdbarch_tdep (gdbarch)->regs;
> ++ if (0 <= num && num < 32)
> ++ return regs->r + num;
> ++ else if (32 <= num && num < 64 && 0 <= regs->f)
> ++ return regs->f + num - 32;
> ++ else if (64 <= num && num < 72 && 0 <= regs->fcc)
> ++ return regs->fcc + num - 64;
> ++ else
> ++ return -1;
> ++}
> ++
> ++static std::string
> ++loongarch_gcc_target_options (struct gdbarch *gdbarch)
> ++{
> ++ return "";
> ++}
> ++
> ++static int
> ++loongarch_register_reggroup_p (struct gdbarch *gdbarch, int regnum,
> ++ struct reggroup *group)
> ++{
> ++ auto regs = &gdbarch_tdep (gdbarch)->regs;
> ++
> ++ if (gdbarch_register_name (gdbarch, regnum) == NULL
> ++ || *gdbarch_register_name (gdbarch, regnum) == '\0')
> ++ return 0;
> ++
> ++ int raw_p = regnum < gdbarch_num_regs (gdbarch);
> ++
> ++ if (group == save_reggroup || group == restore_reggroup)
> ++ return raw_p;
> ++ if (group == all_reggroup)
> ++ return 1;
> ++
> ++ if (group == general_reggroup
> ++ && (regs->orig_a0 == regnum || regs->pc == regnum
> ++ || regs->badv == regnum
> ++ || (regs->r <= regnum && regnum < regs->r + 32)))
> ++ return 1;
> ++
> ++ /* Only $rx and $pc in general_reggroup. */
> ++ if (group == general_reggroup)
> ++ return 0;
> ++
> ++ if (0 <= regs->f
> ++ && (regs->fcsr == regnum || (regs->f <= regnum && regnum < regs->f
> + 32)
> ++ || (regs->fcc <= regnum && regnum < regs->fcc + 8)))
> ++ return group == float_reggroup;
> ++
> ++ /* Only $fx / $fccx / $fcsr in float_reggroup. */
> ++ if (group == float_reggroup)
> ++ return 0;
> ++
> ++ if (0 <= regs->vr && regs->vr <= regnum && regnum < regs->vr + 32)
> ++ if (group == vector_reggroup)
> ++ return 1;
> ++
> ++ if (0 <= regs->xr && regs->xr <= regnum && regnum < regs->xr + 32)
> ++ if (group == vector_reggroup)
> ++ return 1;
> ++
> ++ int ret = tdesc_register_in_reggroup_p (gdbarch, regnum, group);
> ++ if (ret != -1)
> ++ return ret;
> ++
> ++ return default_register_reggroup_p (gdbarch, regnum, group);
> ++}
> ++
> ++constexpr gdb_byte loongarch_default_breakpoint[] = {0x05, 0x00, 0x2a,
> 0x00};
> ++typedef BP_MANIPULATION (loongarch_default_breakpoint)
> loongarch_breakpoint;
> ++
> ++/* Initialize the current architecture based on INFO. If possible,
> ++ re-use an architecture from ARCHES, which is a list of
> ++ architectures already created during this debugging session.
> ++
> ++ Called e.g. at program startup, when reading a core file, and when
> ++ reading a binary file. */
> ++
> ++/* This predicate tests whether we need to read lsx/lasx registers
> ++ (instead of fp registers with the same DWARF2 code
> ++ (thus the same internal code, though lasx/lsx/fp reg internal
> ++ codes are different)) according to the byte-size of requested type.
> */
> ++
> ++static int
> ++loongarch_fp_regnum_refers_to_lsx_lasx_p (struct gdbarch *gdbarch, int
> regnum,
> ++ struct type *type)
> ++{
> ++ /* Conditions:
> ++ 1) regnum is in "disputed" zone (fp/lsx/lasx, translated
> ++ from dwarf regnum).
> ++ 2) type is larger than 8 bytes.
> ++
> ++ (if specified type is larger than 8 bytes,
> ++ then regnum refers to lsx / lasx register instead of fp register).
> ++ */
> ++ return regnum >= gdbarch_tdep (gdbarch)->regs.f
> ++ && regnum < gdbarch_tdep (gdbarch)->regs.f + 32
> ++ && TYPE_LENGTH (type) > 8;
> ++}
> ++
> ++static int
> ++loongarch_convert_register_p (struct gdbarch *gdbarch, int regnum,
> ++ struct type *type)
> ++{
> ++ return loongarch_fp_regnum_refers_to_lsx_lasx_p (gdbarch, regnum,
> type);
> ++}
> ++
> ++static int
> ++loongarch_register_to_value (struct frame_info *frame, int regnum,
> ++ struct type *type, gdb_byte *to, int
> *optimizedp,
> ++ int *unavailablep)
> ++{
> ++ struct gdbarch *gdbarch = get_frame_arch (frame);
> ++
> ++ if (loongarch_fp_regnum_refers_to_lsx_lasx_p (gdbarch, regnum, type))
> ++ {
> ++ /* Add a displacement to regnum. */
> ++ switch (TYPE_LENGTH (type))
> ++ {
> ++ case 16: /* 16-byte types, access vr. */
> ++ if (!get_frame_register_bytes (frame,
> ++ regnum
> ++ + gdbarch_tdep
> (gdbarch)->regs.vr
> ++ - gdbarch_tdep
> (gdbarch)->regs.f,
> ++ 0, 16, to + 0, optimizedp,
> ++ unavailablep))
> ++ return 0;
> ++ break;
> ++
> ++ case 32: /* 32-byte types, access xr. */
> ++ if (!get_frame_register_bytes (frame,
> ++ regnum
> ++ + gdbarch_tdep
> (gdbarch)->regs.xr
> ++ - gdbarch_tdep
> (gdbarch)->regs.f,
> ++ 0, 32, to + 0, optimizedp,
> ++ unavailablep))
> ++ return 0;
> ++ break;
> ++
> ++ default:
> ++ goto fail;
> ++ }
> ++
> ++ *optimizedp = *unavailablep = 0;
> ++ return 1; /* 1 for success, 0 for fail. */
> ++ }
> ++
> ++fail:
> ++ internal_error (__FILE__, __LINE__,
> ++ _ ("loongarch_register_to_value: unrecognized case"));
> ++}
> ++
> ++static void
> ++loongarch_value_to_register (struct frame_info *frame, int regnum,
> ++ struct type *type, const gdb_byte *from)
> ++{
> ++ struct gdbarch *gdbarch = get_frame_arch (frame);
> ++ if (loongarch_fp_regnum_refers_to_lsx_lasx_p (gdbarch, regnum, type))
> ++ {
> ++ switch (TYPE_LENGTH (type))
> ++ {
> ++ case 16: /* 16-byte types, access vr. */
> ++ put_frame_register (frame,
> ++ regnum + gdbarch_tdep (gdbarch)->regs.vr
> ++ - gdbarch_tdep (gdbarch)->regs.f,
> ++ from);
> ++ return;
> ++
> ++ case 32: /* 32-byte types, access xr. */
> ++ put_frame_register (frame,
> ++ regnum + gdbarch_tdep (gdbarch)->regs.xr
> ++ - gdbarch_tdep (gdbarch)->regs.f,
> ++ from);
> ++ return;
> ++ }
> ++ }
> ++
> ++ internal_error (__FILE__, __LINE__,
> ++ _ ("loongarch_value_to_register: unrecognized case"));
> ++}
> ++
> ++static int
> ++loongarch_get_longjmp_target (struct frame_info *frame, CORE_ADDR *pc)
> ++{
> ++ CORE_ADDR jb_addr;
> ++ struct gdbarch *gdbarch = get_frame_arch (frame);
> ++ uint32_t ptr_size = gdbarch_ptr_bit (gdbarch) / TARGET_CHAR_BIT;
> ++ gdb_byte buf[ptr_size];
> ++
> ++ jb_addr = get_frame_register_unsigned (frame, LOONGARCH_A0_REGNUM);
> ++
> ++ if (target_read_memory ((jb_addr + LOONGARCH_JB_PC * ptr_size),
> ++ buf, ptr_size))
> ++ return 0;
> ++
> ++ *pc = extract_unsigned_integer (buf, ptr_size, BFD_ENDIAN_LITTLE);
> ++
> ++ return 1;
> ++}
> ++
> ++static struct gdbarch *
> ++loongarch_gdbarch_init (struct gdbarch_info info, struct gdbarch_list
> *arches)
> ++{
> ++ struct gdbarch *gdbarch;
> ++ struct gdbarch_tdep tdep_instant, *tdep;
> ++ struct tdesc_arch_data *tdesc_data = NULL;
> ++ const struct target_desc *tdesc = info.target_desc;
> ++ int i;
> ++ size_t regnum;
> ++
> ++ tdep = &tdep_instant;
> ++ memset (tdep, 0, sizeof (*tdep));
> ++ memset (&tdep->regs, -1, sizeof (tdep->regs));
> ++
> ++ /* If abfd is nullptr then a EF_LOONGARCH_ABI_LP64 |
> EF_LOONGARCH_FLOAT_ABI_DOUBLE
> ++ is returned in its default state. */
> ++ if (info.abfd != NULL
> ++ && bfd_get_flavour (info.abfd) == bfd_target_elf_flavour)
> ++ {
> ++ int eflags = elf_elfheader (info.abfd)->e_flags;
> ++ unsigned char eclass = elf_elfheader
> (info.abfd)->e_ident[EI_CLASS];
> ++
> ++ if (eflags) /* Executable file */
> ++ {
> ++ tdep->ef_abi = (EF_LOONGARCH_ABI(eflags) &
> EF_LOONGARCH_ABI_MASK);
> ++ }
> ++ else /* Core file */
> ++ {
> ++ if (eclass == ELFCLASS64)
> ++ tdep->ef_abi = EF_LOONGARCH_ABI_LP64_DOUBLE_FLOAT;
> ++ else
> ++ tdep->ef_abi = EF_LOONGARCH_ABI_ILP32_DOUBLE_FLOAT;
> ++ }
> ++ }
> ++ else
> ++ tdep->ef_abi = EF_LOONGARCH_ABI_LP64_DOUBLE_FLOAT;
> ++
> ++ /* Check any target description for validity. */
> ++ if (!tdesc_has_registers (tdesc))
> ++ tdesc = loongarch_get_base_target_description (
> ++ EF_LOONGARCH_IS_ILP32 (tdep->ef_abi) ? 32 : 64,
> ++ EF_LOONGARCH_IS_SINGLE_FLOAT (tdep->ef_abi) ? 32 : 64);
> ++
> ++ int valid_p = 1;
> ++ const struct tdesc_feature *feature;
> ++
> ++ feature = tdesc_find_feature (tdesc, "org.gnu.gdb.loongarch.base");
> ++ if (feature == NULL)
> ++ return NULL;
> ++ regnum = 0;
> ++ tdesc_data = tdesc_data_alloc ();
> ++
> ++ tdep->regs.r = regnum;
> ++ for (i = 0; i < 32; i++)
> ++ valid_p &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ loongarch_r_normal_name[i] + 1);
> ++ valid_p &= tdesc_numbered_register (feature, tdesc_data,
> ++ tdep->regs.orig_a0 = regnum++,
> "orig_a0");
> ++ valid_p &= tdesc_numbered_register (feature, tdesc_data,
> ++ tdep->regs.pc = regnum++, "pc");
> ++ valid_p
> ++ &= tdesc_numbered_register (feature, tdesc_data,
> ++ tdep->regs.badv = regnum++, "badv");
> ++
> ++ if ((feature = tdesc_find_feature (tdesc,
> "org.gnu.gdb.loongarch.fpu")))
> ++ {
> ++ tdep->regs.f = regnum;
> ++ for (i = 0; i < 32; i++)
> ++ valid_p
> ++ &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ loongarch_f_normal_name[i] + 1);
> ++ tdep->regs.fcc = regnum;
> ++ valid_p &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ "fcc0");
> ++ valid_p &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ "fcc1");
> ++ valid_p &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ "fcc2");
> ++ valid_p &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ "fcc3");
> ++ valid_p &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ "fcc4");
> ++ valid_p &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ "fcc5");
> ++ valid_p &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ "fcc6");
> ++ valid_p &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ "fcc7");
> ++ valid_p &= tdesc_numbered_register (feature, tdesc_data,
> ++ tdep->regs.fcsr = regnum++,
> "fcsr");
> ++ }
> ++
> ++ if ((feature = tdesc_find_feature (tdesc,
> "org.gnu.gdb.loongarch.lbt")))
> ++ {
> ++ tdep->regs.scr = regnum;
> ++ for (i = 0; i < 4; i++)
> ++ valid_p
> ++ &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ loongarch_cr_normal_name[i] + 1);
> ++ valid_p
> ++ &= tdesc_numbered_register (feature, tdesc_data,
> ++ tdep->regs.EFLAG = regnum++, "EFLAG");
> ++ valid_p
> ++ &= tdesc_numbered_register (feature, tdesc_data,
> ++ tdep->regs.x86_top = regnum++,
> "x86_top");
> ++ }
> ++
> ++ if ((feature = tdesc_find_feature (tdesc,
> "org.gnu.gdb.loongarch.lsx")))
> ++ {
> ++ tdep->regs.vr = regnum;
> ++ for (i = 0; i < 32; i++)
> ++ valid_p
> ++ &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ loongarch_v_normal_name[i] + 1);
> ++ }
> ++
> ++ if ((feature = tdesc_find_feature (tdesc,
> "org.gnu.gdb.loongarch.lasx")))
> ++ {
> ++ tdep->regs.xr = regnum;
> ++ for (i = 0; i < 32; i++)
> ++ valid_p
> ++ &= tdesc_numbered_register (feature, tdesc_data, regnum++,
> ++ loongarch_x_normal_name[i] + 1);
> ++ }
> ++
> ++ if (!valid_p)
> ++ {
> ++ return NULL;
> ++ }
> ++
> ++ info.byte_order_for_code = BFD_ENDIAN_LITTLE;
> ++
> ++ /* Find a candidate among the list of pre-declared architectures. */
> ++ for (arches = gdbarch_list_lookup_by_info (arches, &info); arches !=
> NULL;
> ++ arches = gdbarch_list_lookup_by_info (arches->next, &info))
> ++ {
> ++ if (gdbarch_tdep (arches->gdbarch)->ef_abi != tdep->ef_abi)
> ++ continue;
> ++
> ++ return arches->gdbarch;
> ++ }
> ++
> ++ /* None found, so create a new architecture from the information
> provided. */
> ++ tdep = (struct gdbarch_tdep *) xmalloc (sizeof (tdep_instant));
> ++ memcpy (tdep, &tdep_instant, sizeof (tdep_instant));
> ++ gdbarch = gdbarch_alloc (&info, tdep);
> ++
> ++ /* Target data types. */
> ++ if (EF_LOONGARCH_IS_ILP32 (tdep->ef_abi))
> ++ {
> ++ set_gdbarch_short_bit (gdbarch, 16);
> ++ set_gdbarch_int_bit (gdbarch, 32);
> ++ set_gdbarch_long_bit (gdbarch, 32);
> ++ set_gdbarch_long_long_bit (gdbarch, 32);
> ++ set_gdbarch_float_bit (gdbarch, 32);
> ++ set_gdbarch_double_bit (gdbarch, 64);
> ++ set_gdbarch_long_double_bit (gdbarch, 128);
> ++ set_gdbarch_long_double_format (gdbarch, floatformats_ia64_quad);
> ++ set_gdbarch_ptr_bit (gdbarch, 32);
> ++ set_gdbarch_char_signed (gdbarch, 0);
> ++ }
> ++ else if (EF_LOONGARCH_IS_LP64 (tdep->ef_abi))
> ++ {
> ++ set_gdbarch_short_bit (gdbarch, 16);
> ++ set_gdbarch_int_bit (gdbarch, 32);
> ++ set_gdbarch_long_bit (gdbarch, 64);
> ++ set_gdbarch_long_long_bit (gdbarch, 64);
> ++ set_gdbarch_float_bit (gdbarch, 32);
> ++ set_gdbarch_double_bit (gdbarch, 64);
> ++ set_gdbarch_long_double_bit (gdbarch, 128);
> ++ set_gdbarch_long_double_format (gdbarch, floatformats_ia64_quad);
> ++ set_gdbarch_ptr_bit (gdbarch, 64);
> ++ set_gdbarch_char_signed (gdbarch, 0);
> ++
> ++ tdep->regs.ra = tdep->regs.r + 1;
> ++ tdep->regs.sp = tdep->regs.r + 3;
> ++
> ++ for (i = 0; i < ARRAY_SIZE (loongarch_r_normal_name); ++i)
> ++ if (loongarch_r_normal_name[i][0] != '\0')
> ++ user_reg_add (gdbarch, loongarch_r_normal_name[i] + 1,
> ++ value_of_loongarch_user_reg,
> ++ (void *) (size_t) (tdep->regs.r + i));
> ++
> ++ for (i = 0; i < ARRAY_SIZE (loongarch_r_lp64_name); ++i)
> ++ if (loongarch_r_lp64_name[i][0] != '\0')
> ++ user_reg_add (gdbarch, loongarch_r_lp64_name[i] + 1,
> ++ value_of_loongarch_user_reg,
> ++ (void *) (size_t) (tdep->regs.r + i));
> ++
> ++ for (i = 0; i < ARRAY_SIZE (loongarch_r_lp64_name1); ++i)
> ++ if (loongarch_r_lp64_name[i][0] != '\0')
> ++ user_reg_add (gdbarch, loongarch_r_lp64_name1[i] + 1,
> ++ value_of_loongarch_user_reg,
> ++ (void *) (size_t) (tdep->regs.r + i));
> ++
> ++ /* Functions handling dummy frames. */
> ++ set_gdbarch_push_dummy_call (gdbarch,
> ++ loongarch_lp32lp64_push_dummy_call);
> ++ set_gdbarch_return_value (gdbarch, loongarch_lp64_return_value);
> ++
> ++ }
> ++ else
> ++ gdb_assert_not_reached ("unknown ABI");
> ++
> ++ /* Hook in OS ABI-specific overrides, if they have been registered. */
> ++ info.target_desc = tdesc;
> ++ info.tdesc_data = tdesc_data;
> ++
> ++ /* Register architecture. */
> ++ set_gdbarch_num_regs (gdbarch, regnum);
> ++ set_gdbarch_sp_regnum (gdbarch, tdep->regs.sp);
> ++ set_gdbarch_pc_regnum (gdbarch, tdep->regs.pc);
> ++
> ++ tdesc_use_registers (gdbarch, tdesc, std::move (tdesc_data));
> ++
> ++ /* Functions to supply register information. */
> ++ set_gdbarch_register_name (gdbarch, loongarch_register_name);
> ++
> ++ /* Handle overlapping dwarf2 register code for fp/lsx/lasx. */
> ++ set_gdbarch_convert_register_p (gdbarch, loongarch_convert_register_p);
> ++ set_gdbarch_register_to_value (gdbarch, loongarch_register_to_value);
> ++ set_gdbarch_value_to_register (gdbarch, loongarch_value_to_register);
> ++
> ++ /* Functions to analyze frames. */
> ++ set_gdbarch_skip_prologue (gdbarch, loongarch_skip_prologue);
> ++ set_gdbarch_inner_than (gdbarch, core_addr_lessthan);
> ++ set_gdbarch_frame_align (gdbarch, loongarch_frame_align);
> ++
> ++ /* Functions to access frame data. */
> ++ set_gdbarch_unwind_pc (gdbarch, loongarch_unwind_pc);
> ++ set_gdbarch_unwind_sp (gdbarch, loongarch_unwind_sp);
> ++
> ++ set_gdbarch_dummy_id (gdbarch, loongarch_dummy_id);
> ++
> ++ set_gdbarch_software_single_step (gdbarch,
> loongarch_software_single_step);
> ++
> ++ set_gdbarch_breakpoint_kind_from_pc (gdbarch,
> ++ loongarch_breakpoint::kind_from_pc);
> ++ set_gdbarch_sw_breakpoint_from_kind (gdbarch,
> ++ loongarch_breakpoint::bp_from_kind);
> ++
> ++ set_gdbarch_have_nonsteppable_watchpoint (gdbarch, 1);
> ++
> ++ /* Virtual tables. */
> ++ set_gdbarch_vbit_in_delta (gdbarch, 1);
> ++
> ++ set_gdbarch_gcc_target_options (gdbarch, loongarch_gcc_target_options);
> ++
> ++ gdbarch_init_osabi (info, gdbarch);
> ++ set_gdbarch_register_reggroup_p (gdbarch,
> loongarch_register_reggroup_p);
> ++ set_gdbarch_register_name (gdbarch, loongarch_register_name);
> ++
> ++ set_gdbarch_get_longjmp_target (gdbarch, loongarch_get_longjmp_target);
> ++
> ++ /* Frame unwinders. Use DWARF debug info if available, otherwise use
> our own
> ++ unwinder. */
> ++ set_gdbarch_dwarf2_reg_to_regnum (gdbarch,
> loongarch_dwarf2_reg_to_regnum);
> ++ dwarf2_append_unwinders (gdbarch);
> ++ frame_unwind_append_unwinder (gdbarch, &loongarch_frame_unwind);
> ++
> ++ return gdbarch;
> ++}
> ++
> ++static void
> ++info_loongarch (const char *addr_exp, int from_tty)
> ++{
> ++ char *buf, *t;
> ++ int set;
> ++ char *item;
> ++ unsigned long addr;
> ++ unsigned long long value;
> ++
> ++ if (addr_exp)
> ++ {
> ++ addr_exp = skip_spaces (addr_exp);
> ++ buf = (char *) alloca (strlen (addr_exp) + 1);
> ++ strcpy (buf, addr_exp);
> ++ loongarch_eliminate_adjacent_repeat_char (buf, ' ');
> ++ }
> ++ else
> ++ goto Empty;
> ++
> ++ if (!(t = strtok (buf, " ")))
> ++ goto Empty;
> ++ if (strcmp (t, "set") == 0)
> ++ {
> ++ t = strtok (NULL, " ");
> ++ set = 1;
> ++ }
> ++ else
> ++ {
> ++ if (strcmp (t, "get") == 0)
> ++ t = strtok (NULL, " ");
> ++ set = 0;
> ++ }
> ++ if (!(item = t))
> ++ goto Empty;
> ++ if (!(t = strtok (NULL, " ")))
> ++ goto Empty;
> ++ addr = strtoul (t, NULL, 0);
> ++ if (set && (t = strtok (NULL, " ")) == NULL)
> ++ goto Empty;
> ++ value = strtoll (t, NULL, 0);
> ++
> ++ if (set)
> ++ if (strcmp (item, "cpucfg") == 0)
> ++ {
> ++ uint32_t val32 = value;
> ++ ULONGEST xfered_len;
> ++ target_xfer_partial (current_inferior ()->top_target (),
> ++ TARGET_OBJECT_LARCH, "cpucfg", NULL,
> ++ (const gdb_byte *) &val32, addr * 4,
> ++ sizeof (val32), &xfered_len);
> ++ if (0 < xfered_len)
> ++ fprintf_unfiltered (gdb_stdout, "ok\n");
> ++ else
> ++ error ("Set failed");
> ++ }
> ++ else
> ++ {
> ++ uint64_t val64 = value;
> ++ ULONGEST xfered_len;
> ++ target_xfer_partial (current_inferior ()->top_target (),
> ++ TARGET_OBJECT_LARCH, item, NULL,
> ++ (const gdb_byte *) &val64, addr * 8,
> ++ sizeof (val64), &xfered_len);
> ++ if (0 < xfered_len)
> ++ fprintf_unfiltered (gdb_stdout, "ok\n");
> ++ else
> ++ error ("Set failed");
> ++ }
> ++ else if (strcmp (item, "cpucfg") == 0)
> ++ {
> ++ uint32_t val32;
> ++ ULONGEST xfered_len;
> ++ target_xfer_partial (current_inferior ()->top_target (),
> ++ TARGET_OBJECT_LARCH, "cpucfg", (gdb_byte *)
> &val32,
> ++ NULL, addr * 4, sizeof (val32), &xfered_len);
> ++ if (0 < xfered_len)
> ++ fprintf_unfiltered (gdb_stdout, "return is %x\n", val32);
> ++ else
> ++ error ("Get failed");
> ++ }
> ++ else
> ++ {
> ++ uint64_t val64;
> ++ ULONGEST xfered_len;
> ++ target_xfer_partial (current_inferior ()->top_target (),
> ++ TARGET_OBJECT_LARCH, item, (gdb_byte *) &val64,
> ++ NULL, addr * 8, sizeof (val64), &xfered_len);
> ++ if (0 < xfered_len)
> ++ fprintf_unfiltered (gdb_stdout, "return is %llx\n", (long long)
> val64);
> ++ else
> ++ error ("Get failed");
> ++ }
> ++
> ++ return;
> ++Empty:
> ++ error ("Empty. Should be 'info loongarch ([get]|set) item addr
> [value]'");
> ++}
> ++
> ++void _initialize_loongarch_tdep ();
> ++void
> ++_initialize_loongarch_tdep ()
> ++{
> ++ gdbarch_register (bfd_arch_loongarch, loongarch_gdbarch_init, NULL);
> ++
> ++ add_info ("loongarch", info_loongarch, _ ("Loongarch extra"));
> ++
> ++ /* Debug this files internals. */
> ++ add_setshow_zuinteger_cmd ("loongarch", class_maintenance,
> &loongarch_debug,
> ++ _ ("\
> ++Set loongarch debugging."),
> ++ _ ("\
> ++Show loongarch debugging."),
> ++ _ ("\
> ++When non-zero, loongarch specific debugging is enabled."),
> ++ NULL, NULL, &setdebuglist, &showdebuglist);
> ++}
> +diff --git gdb-10.2/gdb/loongarch-tdep.h gdb-10.2/gdb/loongarch-tdep.h
> +new file mode 100644
> +index 0000000..12bc78e
> +--- /dev/null
> ++++ gdb-10.2/gdb/loongarch-tdep.h
> +@@ -0,0 +1,61 @@
> ++/* Target-dependent code for GNU/Linux LoongArch.
> ++
> ++ Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of GDB.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program. If not, see <http://www.gnu.org/licenses/>.
> */
> ++
> ++#ifndef LOONGARCH_TDEP_H
> ++#define LOONGARCH_TDEP_H
> ++
> ++#include "arch/loongarch.h"
> ++
> ++/* Register numbers of important registers. Note that most of
> ++ these values are "real" register numbers, and correspond to the
> ++ general registers of the machine. */
> ++#define LOONGARCH_A0_REGNUM 4 /* Loc of first arg */
> ++
> ++struct gdbarch_tdep
> ++{
> ++ int ef_abi; /* EF_LOONGARCH_ABI */
> ++
> ++ struct
> ++ {
> ++ int r;
> ++ int ra;
> ++ int sp;
> ++ int orig_a0;
> ++ int pc;
> ++ int badv;
> ++
> ++ int f;
> ++ int fcc;
> ++ int fcsr;
> ++ int vr;
> ++ int xr;
> ++
> ++ int scr;
> ++ int EFLAG;
> ++ int x86_top;
> ++
> ++ } regs;
> ++
> ++ /* Return the expected next PC if FRAME is stopped at a syscall
> ++ instruction. */
> ++ CORE_ADDR (*syscall_next_pc) (struct frame_info *frame);
> ++};
> ++
> ++#endif /* LOONGARCH_TDEP_H */
> +diff --git gdb-10.2/gdb/nat/loongarch-linux-watch.c
> gdb-10.2/gdb/nat/loongarch-linux-watch.c
> +new file mode 100644
> +index 0000000..f7c0dbf
> +--- /dev/null
> ++++ gdb-10.2/gdb/nat/loongarch-linux-watch.c
> +@@ -0,0 +1,330 @@
> ++/* Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of GDB.
> ++
> ++ Based on MIPS target.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program. If not, see <http://www.gnu.org/licenses/>.
> */
> ++
> ++#include "gdbsupport/common-defs.h"
> ++#include "nat/gdb_ptrace.h"
> ++#include "loongarch-linux-watch.h"
> ++
> ++/* Assuming usable watch registers REGS, return the irwmask of
> ++ register N. */
> ++
> ++uint8_t
> ++loongarch_linux_watch_get_irwmask (struct pt_watch_regs *regs, int n)
> ++{
> ++ switch (regs->style)
> ++ {
> ++ case pt_watch_style_la32:
> ++ return regs->la32[n].irwmask & IRW_MASK;
> ++ case pt_watch_style_la64:
> ++ return regs->la64[n].irwmask & IRW_MASK;
> ++ default:
> ++ internal_error (__FILE__, __LINE__,
> ++ _ ("Unrecognized watch register style"));
> ++ }
> ++}
> ++
> ++/* Assuming usable watch registers REGS, return the irwstat of
> ++ register N. */
> ++
> ++uint8_t
> ++loongarch_linux_watch_get_irwstat (struct pt_watch_regs *regs, int n)
> ++{
> ++ switch (regs->style)
> ++ {
> ++ case pt_watch_style_la32:
> ++ return regs->la32[n].irwstat & IRW_MASK;
> ++ case pt_watch_style_la64:
> ++ return regs->la64[n].irwstat & IRW_MASK;
> ++ default:
> ++ internal_error (__FILE__, __LINE__,
> ++ _ ("Unrecognized watch register style"));
> ++ }
> ++}
> ++
> ++/* Assuming usable watch registers REGS, return the num_valid. */
> ++
> ++uint32_t
> ++loongarch_linux_watch_get_num_valid (struct pt_watch_regs *regs)
> ++{
> ++ return regs->num_valid;
> ++}
> ++
> ++/* Assuming usable watch registers REGS, return the addr of
> ++ register N. */
> ++
> ++CORE_ADDR
> ++loongarch_linux_watch_get_addr (struct pt_watch_regs *regs, int n)
> ++{
> ++ switch (regs->style)
> ++ {
> ++ case pt_watch_style_la32:
> ++ return regs->la32[n].addr;
> ++ case pt_watch_style_la64:
> ++ return regs->la64[n].addr;
> ++ default:
> ++ internal_error (__FILE__, __LINE__,
> ++ _ ("Unrecognized watch register style"));
> ++ }
> ++}
> ++
> ++/* Assuming usable watch registers REGS, set addr of register N to
> ++ VALUE. */
> ++
> ++void
> ++loongarch_linux_watch_set_addr (struct pt_watch_regs *regs, int n,
> ++ CORE_ADDR value)
> ++{
> ++ switch (regs->style)
> ++ {
> ++ case pt_watch_style_la32:
> ++ /* The cast will never throw away bits as 64 bit addresses can
> ++ never be used on a 32 bit kernel. */
> ++ regs->la32[n].addr = (uint32_t) value;
> ++ break;
> ++ case pt_watch_style_la64:
> ++ regs->la64[n].addr = value;
> ++ break;
> ++ default:
> ++ internal_error (__FILE__, __LINE__,
> ++ _ ("Unrecognized watch register style"));
> ++ }
> ++}
> ++
> ++/* Assuming usable watch registers REGS, return the mask of
> ++ register N. */
> ++
> ++CORE_ADDR
> ++loongarch_linux_watch_get_mask (struct pt_watch_regs *regs, int n)
> ++{
> ++ switch (regs->style)
> ++ {
> ++ case pt_watch_style_la32:
> ++ return regs->la32[n].mask;
> ++ case pt_watch_style_la64:
> ++ return regs->la64[n].mask;
> ++ default:
> ++ internal_error (__FILE__, __LINE__,
> ++ _ ("Unrecognized watch register style"));
> ++ }
> ++}
> ++
> ++/* Assuming usable watch registers REGS, set mask of register N to
> ++ VALUE. */
> ++
> ++void
> ++loongarch_linux_watch_set_mask (struct pt_watch_regs *regs, int n,
> ++ CORE_ADDR value)
> ++{
> ++ switch (regs->style)
> ++ {
> ++ case pt_watch_style_la32:
> ++ regs->la32[n].mask = value;
> ++ break;
> ++ case pt_watch_style_la64:
> ++ regs->la64[n].mask = value;
> ++ break;
> ++ default:
> ++ internal_error (__FILE__, __LINE__,
> ++ _ ("Unrecognized watch register style"));
> ++ }
> ++}
> ++
> ++/* Assuming usable watch registers REGS, return the irw of
> ++ register N. */
> ++
> ++uint8_t
> ++loongarch_linux_watch_get_irw (struct pt_watch_regs *regs, int n)
> ++{
> ++ switch (regs->style)
> ++ {
> ++ case pt_watch_style_la32:
> ++ return regs->la32[n].irw;
> ++ case pt_watch_style_la64:
> ++ return regs->la64[n].irw;
> ++ default:
> ++ internal_error (__FILE__, __LINE__,
> ++ _ ("Unrecognized watch register style"));
> ++ }
> ++}
> ++
> ++/* Assuming usable watch registers REGS, set irw of register N to
> ++ VALUE. */
> ++
> ++void
> ++loongarch_linux_watch_set_irw (struct pt_watch_regs *regs, int n,
> ++ uint8_t value)
> ++{
> ++ switch (regs->style)
> ++ {
> ++ case pt_watch_style_la32:
> ++ regs->la32[n].irw = value;
> ++ break;
> ++ case pt_watch_style_la64:
> ++ regs->la64[n].irw = value;
> ++ break;
> ++ default:
> ++ internal_error (__FILE__, __LINE__,
> ++ _ ("Unrecognized watch register style"));
> ++ }
> ++}
> ++
> ++/* Read the watch registers of process LWPID and store it in
> ++ WATCH_READBACK. Save true to *WATCH_READBACK_VALID if watch
> ++ registers are valid. Return 1 if watch registers are usable.
> ++ Cached information is used unless FORCE is true. */
> ++
> ++int
> ++loongarch_linux_read_watch_registers (long lwpid,
> ++ struct pt_watch_regs *watch_readback,
> ++ int *watch_readback_valid, int force)
> ++{
> ++ if (force || *watch_readback_valid == 0)
> ++ {
> ++ if (ptrace (PTRACE_GET_WATCH_REGS, lwpid, watch_readback, NULL) ==
> -1)
> ++ {
> ++ *watch_readback_valid = -1;
> ++ return 0;
> ++ }
> ++ if (watch_readback->num_valid == 0)
> ++ {
> ++ *watch_readback_valid = -1;
> ++ return 0;
> ++ }
> ++ /* Watch registers appear to be usable. */
> ++ *watch_readback_valid = 1;
> ++ }
> ++ return (*watch_readback_valid == 1) ? 1 : 0;
> ++}
> ++
> ++/* Convert GDB's TYPE to an IRW mask. */
> ++
> ++uint32_t
> ++loongarch_linux_watch_type_to_irw (enum target_hw_bp_type type)
> ++{
> ++ switch (type)
> ++ {
> ++ case hw_write:
> ++ return W_MASK;
> ++ case hw_read:
> ++ return R_MASK;
> ++ case hw_access:
> ++ return (W_MASK | R_MASK);
> ++ case hw_execute:
> ++ return I_MASK;
> ++ default:
> ++ return 0;
> ++ }
> ++}
> ++
> ++/* Set any low order bits in MASK that are not set. */
> ++
> ++static CORE_ADDR
> ++fill_mask (CORE_ADDR mask)
> ++{
> ++ CORE_ADDR f = 1;
> ++
> ++ while (f && f < mask)
> ++ {
> ++ mask |= f;
> ++ f <<= 1;
> ++ }
> ++ return mask;
> ++}
> ++
> ++/* Try to add a single watch to the specified registers REGS. The
> ++ address of added watch is ADDR, the length is LEN, and the mask
> ++ is IRW. Return 1 on success, 0 on failure. */
> ++
> ++int
> ++loongarch_linux_watch_try_one_watch (struct pt_watch_regs *regs,
> ++ CORE_ADDR addr, int len, uint32_t irw)
> ++{
> ++ CORE_ADDR base_addr, last_byte;
> ++ CORE_ADDR mask_bits, t_addr, t_mask;
> ++ uint8_t t_irw;
> ++ int i;
> ++
> ++ if (len <= 0)
> ++ return 0;
> ++
> ++ last_byte = addr + len - 1;
> ++ mask_bits = fill_mask (addr ^ last_byte);
> ++ base_addr = addr & ~mask_bits;
> ++
> ++ /* Check to see if it is covered by current registers. */
> ++ for (i = 0; i < loongarch_linux_watch_get_num_valid (regs); i++)
> ++ {
> ++ t_addr = loongarch_linux_watch_get_addr (regs, i);
> ++ t_irw = loongarch_linux_watch_get_irw (regs, i);
> ++ if (t_addr != 0 && irw == ((uint32_t) t_irw & irw))
> ++ {
> ++ t_mask = loongarch_linux_watch_get_mask (regs, i);
> ++ if (addr >= t_addr && last_byte <= (t_addr + t_mask))
> ++ return 1;
> ++ }
> ++ }
> ++ /* Try to find an empty register. */
> ++ for (i = 0; i < loongarch_linux_watch_get_num_valid (regs); i++)
> ++ {
> ++ t_addr = loongarch_linux_watch_get_addr (regs, i);
> ++ if (t_addr == 0
> ++ && irw == (loongarch_linux_watch_get_irwmask (regs, i) & irw))
> ++ {
> ++ /* It fits, we'll take it. */
> ++ loongarch_linux_watch_set_addr (regs, i, base_addr);
> ++ loongarch_linux_watch_set_mask (regs, i, mask_bits);
> ++ loongarch_linux_watch_set_irw (regs, i, irw);
> ++ return 1;
> ++ }
> ++ }
> ++ /* It didn't fit anywhere, we failed. */
> ++ return 0;
> ++}
> ++
> ++/* Fill in the watch registers REGS with the currently cached
> ++ watches CURRENT_WATCHES. */
> ++
> ++void
> ++loongarch_linux_watch_populate_regs (
> ++ struct loongarch_watchpoint *current_watches, struct pt_watch_regs
> *regs)
> ++{
> ++ struct loongarch_watchpoint *w;
> ++ int i;
> ++
> ++ /* Clear them out. */
> ++ for (i = 0; i < loongarch_linux_watch_get_num_valid (regs); i++)
> ++ {
> ++ loongarch_linux_watch_set_addr (regs, i, 0);
> ++ loongarch_linux_watch_set_mask (regs, i, 0);
> ++ loongarch_linux_watch_set_irw (regs, i, 0);
> ++ }
> ++
> ++ w = current_watches;
> ++ while (w)
> ++ {
> ++ uint32_t irw = loongarch_linux_watch_type_to_irw (w->type);
> ++
> ++ i = loongarch_linux_watch_try_one_watch (regs, w->addr, w->len,
> irw);
> ++ /* They must all fit, because we previously calculated that they
> ++ would. */
> ++ gdb_assert (i);
> ++ w = w->next;
> ++ }
> ++}
> +diff --git gdb-10.2/gdb/nat/loongarch-linux-watch.h
> gdb-10.2/gdb/nat/loongarch-linux-watch.h
> +new file mode 100644
> +index 0000000..ab80b44
> +--- /dev/null
> ++++ gdb-10.2/gdb/nat/loongarch-linux-watch.h
> +@@ -0,0 +1,132 @@
> ++/* Copyright (C) 2021 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of GDB.
> ++
> ++ Based on MIPS target.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program. If not, see <http://www.gnu.org/licenses/>.
> */
> ++
> ++#ifndef LOONGARCH_LINUX_WATCH_H
> ++#define LOONGARCH_LINUX_WATCH_H 1
> ++
> ++#include <asm/ptrace.h>
> ++#include "gdbsupport/break-common.h"
> ++
> ++#define MAX_DEBUG_REGISTER 16
> ++
> ++/* If macro PTRACE_GET_WATCH_REGS is not defined, kernel header doesn't
> ++ have hardware watchpoint-related structures. Define them below. */
> ++
> ++#ifndef PTRACE_GET_WATCH_REGS
> ++#define PTRACE_GET_WATCH_REGS 0xd0
> ++#define PTRACE_SET_WATCH_REGS 0xd1
> ++
> ++enum pt_watch_style
> ++{
> ++ pt_watch_style_la32,
> ++ pt_watch_style_la64
> ++};
> ++
> ++/* A value of zero in a watchlo indicates that it is available. */
> ++
> ++struct la32_watch_regs
> ++{
> ++ uint32_t addr;
> ++ /* Lower 16 bits of watchhi. */
> ++ uint32_t mask;
> ++ /* Valid mask and I R W bits.
> ++ * bit 0 -- 1 if W bit is usable.
> ++ * bit 1 -- 1 if R bit is usable.
> ++ * bit 2 -- 1 if I bit is usable.
> ++ * bits 3 - 11 -- Valid watchhi mask bits.
> ++ */
> ++ uint8_t irw;
> ++ uint8_t irwstat;
> ++ uint8_t irwmask;
> ++ /* There is confusion across gcc versions about structure alignment,
> ++ so we force 8 byte alignment for these structures so they match
> ++ the kernel even if it was build with a different gcc version. */
> ++} __attribute__ ((aligned (8)));
> ++
> ++struct la64_watch_regs
> ++{
> ++ uint64_t addr;
> ++ uint64_t mask;
> ++ uint8_t irw;
> ++ uint8_t irwstat;
> ++ uint8_t irwmask;
> ++} __attribute__ ((aligned (8)));
> ++
> ++struct pt_watch_regs
> ++{
> ++ uint16_t max_valid;
> ++ uint16_t num_valid;
> ++ enum pt_watch_style style;
> ++ union
> ++ {
> ++ struct la32_watch_regs la32[MAX_DEBUG_REGISTER];
> ++ struct la64_watch_regs la64[MAX_DEBUG_REGISTER];
> ++ };
> ++};
> ++
> ++#endif /* !PTRACE_GET_WATCH_REGS */
> ++
> ++#define W_BIT 0
> ++#define R_BIT 1
> ++#define I_BIT 2
> ++
> ++#define W_MASK (1 << W_BIT)
> ++#define R_MASK (1 << R_BIT)
> ++#define I_MASK (1 << I_BIT)
> ++
> ++#define IRW_MASK (I_MASK | R_MASK | W_MASK)
> ++
> ++/* We keep list of all watchpoints we should install and calculate the
> ++ watch register values each time the list changes. This allows for
> ++ easy sharing of watch registers for more than one watchpoint. */
> ++
> ++struct loongarch_watchpoint
> ++{
> ++ CORE_ADDR addr;
> ++ int len;
> ++ enum target_hw_bp_type type;
> ++ struct loongarch_watchpoint *next;
> ++};
> ++
> ++uint32_t loongarch_linux_watch_get_num_valid (struct pt_watch_regs
> *regs);
> ++uint8_t loongarch_linux_watch_get_irwmask (struct pt_watch_regs *regs,
> int n);
> ++uint8_t loongarch_linux_watch_get_irwstat (struct pt_watch_regs *regs,
> int n);
> ++CORE_ADDR loongarch_linux_watch_get_addr (struct pt_watch_regs *regs,
> int n);
> ++void loongarch_linux_watch_set_addr (struct pt_watch_regs *regs, int n,
> ++ CORE_ADDR value);
> ++CORE_ADDR loongarch_linux_watch_get_mask (struct pt_watch_regs *regs,
> int n);
> ++void loongarch_linux_watch_set_mask (struct pt_watch_regs *regs, int n,
> ++ CORE_ADDR value);
> ++uint8_t loongarch_linux_watch_get_irw (struct pt_watch_regs *regs, int
> n);
> ++void loongarch_linux_watch_set_irw (struct pt_watch_regs *regs, int n,
> ++ uint8_t value);
> ++int loongarch_linux_watch_try_one_watch (struct pt_watch_regs *regs,
> ++ CORE_ADDR addr, int len,
> ++ uint32_t irw);
> ++void loongarch_linux_watch_populate_regs (
> ++ struct loongarch_watchpoint *current_watches, struct pt_watch_regs
> *regs);
> ++uint32_t loongarch_linux_watch_type_to_irw (enum target_hw_bp_type type);
> ++
> ++int loongarch_linux_read_watch_registers (long lwpid,
> ++ struct pt_watch_regs
> *watch_readback,
> ++ int *watch_readback_valid,
> ++ int force);
> ++
> ++#endif /* #define LOONGARCH_LINUX_WATCH_H */
> +diff --git gdb-10.2/gdb/remote.c gdb-10.2/gdb/remote.c
> +index 4896c2a..67a9acd 100644
> +--- gdb-10.2/gdb/remote.c
> ++++ gdb-10.2/gdb/remote.c
> +@@ -1983,6 +1983,8 @@ enum {
> + PACKET_qXfer_statictrace_read,
> + PACKET_qXfer_traceframe_info,
> + PACKET_qXfer_uib,
> ++ PACKET_qXfer_loongarch_read,
> ++ PACKET_qXfer_loongarch_write,
> + PACKET_qGetTIBAddr,
> + PACKET_qGetTLSAddr,
> + PACKET_qSupported,
> +@@ -5154,6 +5156,10 @@ static const struct protocol_feature
> remote_protocol_features[] = {
> + PACKET_qXfer_threads },
> + { "qXfer:traceframe-info:read", PACKET_DISABLE,
> remote_supported_packet,
> + PACKET_qXfer_traceframe_info },
> ++ { "qXfer:loongarch:read", PACKET_DISABLE, remote_supported_packet,
> ++ PACKET_qXfer_loongarch_read },
> ++ { "qXfer:loongarch:write", PACKET_DISABLE, remote_supported_packet,
> ++ PACKET_qXfer_loongarch_write },
> + { "QPassSignals", PACKET_DISABLE, remote_supported_packet,
> + PACKET_QPassSignals },
> + { "QCatchSyscalls", PACKET_DISABLE, remote_supported_packet,
> +@@ -11013,6 +11019,18 @@ remote_target::xfer_partial (enum target_object
> object,
> + return TARGET_XFER_E_IO;
> + }
> +
> ++ if (object == TARGET_OBJECT_LARCH)
> ++ {
> ++ if (readbuf)
> ++ return remote_read_qxfer ("loongarch", annex, readbuf, offset, len,
> ++ xfered_len, &remote_protocol_packets
> ++ [PACKET_qXfer_loongarch_read]);
> ++ else
> ++ return remote_write_qxfer ("loongarch", annex, writebuf, offset,
> len,
> ++ xfered_len, &remote_protocol_packets
> ++ [PACKET_qXfer_loongarch_write]);
> ++ }
> ++
> + /* Only handle flash writes. */
> + if (writebuf != NULL)
> + {
> +@@ -14626,6 +14644,13 @@ Show the maximum size of the address (in bits)
> in a memory packet."), NULL,
> + add_packet_config_cmd (&remote_protocol_packets[PACKET_qXfer_uib],
> + "qXfer:uib:read", "unwind-info-block", 0);
> +
> ++ add_packet_config_cmd
> (&remote_protocol_packets[PACKET_qXfer_loongarch_read],
> ++ "qXfer:loongarch:read", "read-loongarch-object",
> 0);
> ++
> ++ add_packet_config_cmd
> ++ (&remote_protocol_packets[PACKET_qXfer_loongarch_write],
> ++ "qXfer:loongarch:write", "write-loongarch-object", 0);
> ++
> + add_packet_config_cmd (&remote_protocol_packets[PACKET_qGetTLSAddr],
> + "qGetTLSAddr", "get-thread-local-storage-address",
> + 0);
> +diff --git gdb-10.2/gdb/target.h gdb-10.2/gdb/target.h
> +index 9603912..40a3cce 100644
> +--- gdb-10.2/gdb/target.h
> ++++ gdb-10.2/gdb/target.h
> +@@ -135,6 +135,9 @@ enum inferior_event_type
> +
> + enum target_object
> + {
> ++ /* LARCH target specific transfer. See "loongarch-nat.c" "corelow.c"
> ++ and "remote.c". */
> ++ TARGET_OBJECT_LARCH,
> + /* AVR target specific transfer. See "avr-tdep.c" and "remote.c". */
> + TARGET_OBJECT_AVR,
> + /* Transfer up-to LEN bytes of memory starting at OFFSET. */
> +diff --git gdb-10.2/gdb/testsuite/gdb.base/dump.exp
> gdb-10.2/gdb/testsuite/gdb.base/dump.exp
> +index 2b79237..64d5b39 100644
> +--- gdb-10.2/gdb/testsuite/gdb.base/dump.exp
> ++++ gdb-10.2/gdb/testsuite/gdb.base/dump.exp
> +@@ -143,11 +143,13 @@ make_dump_file "dump srec val [set intarr1.srec]
> intarray" \
> + make_dump_file "dump srec val [set intstr1.srec] intstruct" \
> + "dump struct as value, srec"
> +
> ++if { ![istarget loongarch*-*-*] } {
> + make_dump_file "dump ihex val [set intarr1.ihex] intarray" \
> + "dump array as value, intel hex"
> +
> + make_dump_file "dump ihex val [set intstr1.ihex] intstruct" \
> + "dump struct as value, intel hex"
> ++}
> +
> + make_dump_file "dump tekhex val [set intarr1.tekhex] intarray" \
> + "dump array as value, tekhex"
> +@@ -244,11 +246,13 @@ make_dump_file "dump srec mem [set intarr2.srec]
> $array_start $array_end" \
> + make_dump_file "dump srec mem [set intstr2.srec] $struct_start
> $struct_end" \
> + "dump struct as memory, srec"
> +
> ++if { ![istarget loongarch*-*-*] } {
> + make_dump_file "dump ihex mem [set intarr2.ihex] $array_start
> $array_end" \
> + "dump array as memory, ihex"
> +
> + make_dump_file "dump ihex mem [set intstr2.ihex] $struct_start
> $struct_end" \
> + "dump struct as memory, ihex"
> ++}
> +
> + make_dump_file "dump tekhex mem [set intarr2.tekhex] $array_start
> $array_end" \
> + "dump array as memory, tekhex"
> +diff --git gdb-10.2/gdb/testsuite/gdb.base/float.exp
> gdb-10.2/gdb/testsuite/gdb.base/float.exp
> +index dc5e2fa..d179a8f 100644
> +--- gdb-10.2/gdb/testsuite/gdb.base/float.exp
> ++++ gdb-10.2/gdb/testsuite/gdb.base/float.exp
> +@@ -120,6 +120,8 @@ if { [is_aarch64_target] } then {
> + pass "info float (without FPU)"
> + }
> + }
> ++} elseif [istarget "loongarch*-*-*"] then {
> ++ gdb_test "info float" "f.*fcc0.*fcsr.*" "info float"
> + } else {
> + gdb_test "info float" "No floating.point info available for this
> processor." "info float (unknown target)"
> + }
> +diff --git gdb-10.2/gdb/testsuite/gdb.trace/entry-values.exp
> gdb-10.2/gdb/testsuite/gdb.trace/entry-values.exp
> +index 3695a1e..f19f5b9 100644
> +--- gdb-10.2/gdb/testsuite/gdb.trace/entry-values.exp
> ++++ gdb-10.2/gdb/testsuite/gdb.trace/entry-values.exp
> +@@ -62,6 +62,8 @@ if { [istarget "arm*-*-*"] || [istarget "aarch64*-*-*"]
> } {
> + # returns. The only exception is JALRC, in which case execution
> + # resumes from `insn1' instead.
> + set call_insn {jalrc|[jb]al[sxr]*[ \t][^\r\n]+\r\n}
> ++} elseif { [istarget "loongarch*-*-*"] } {
> ++ set call_insn "bl"
> + } else {
> + set call_insn "call"
> + }
> +diff --git gdb-10.2/gdb/testsuite/gdb.xml/tdesc-regs.exp
> gdb-10.2/gdb/testsuite/gdb.xml/tdesc-regs.exp
> +index 1ee6ae1..4d69fb3 100644
> +--- gdb-10.2/gdb/testsuite/gdb.xml/tdesc-regs.exp
> ++++ gdb-10.2/gdb/testsuite/gdb.xml/tdesc-regs.exp
> +@@ -82,6 +82,11 @@ switch -glob -- [istarget] {
> + set regdir "i386/"
> + set core-regs {64bit-core.xml 64bit-sse.xml}
> + }
> ++ "loongarch64-*-*" {
> ++ set architecture "loongarch64"
> ++ set regdir "loongarch/"
> ++ set core-regs {base64.xml fpu64.xml lbt64.xml lsx.xml lasx.xml}
> ++ }
> + }
> +
> + # If no core registers were specified, assume this target does not
> +diff --git gdb-10.2/gdbserver/Makefile.in gdb-10.2/gdbserver/Makefile.in
> +index 2bd3a57..3bd8970 100644
> +--- gdb-10.2/gdbserver/Makefile.in
> ++++ gdb-10.2/gdbserver/Makefile.in
> +@@ -182,6 +182,7 @@ SFILES = \
> + $(srcdir)/linux-arc-low.cc \
> + $(srcdir)/linux-arm-low.cc \
> + $(srcdir)/linux-ia64-low.cc \
> ++ $(srcdir)/linux-loongarch-low.cc \
> + $(srcdir)/linux-low.cc \
> + $(srcdir)/linux-m68k-low.cc \
> + $(srcdir)/linux-mips-low.cc \
> +@@ -214,6 +215,7 @@ SFILES = \
> + $(srcdir)/../gdb/arch/arm.c \
> + $(srcdir)/../gdb/arch/arm-get-next-pcs.c \
> + $(srcdir)/../gdb/arch/arm-linux.c \
> ++ $(srcdir)/../gdb/arch/loongarch.c \
> + $(srcdir)/../gdb/arch/ppc-linux-common.c \
> + $(srcdir)/../gdb/arch/riscv.c \
> + $(srcdir)/../gdb/nat/aarch64-sve-linux-ptrace.c \
> +diff --git gdb-10.2/gdbserver/configure.srv
> gdb-10.2/gdbserver/configure.srv
> +index 0cb5072..c4ca3a1 100644
> +--- gdb-10.2/gdbserver/configure.srv
> ++++ gdb-10.2/gdbserver/configure.srv
> +@@ -116,6 +116,13 @@ case "${gdbserver_host}" in
> + srv_tgtobj="$srv_linux_obj linux-ia64-low.o"
> + srv_linux_usrregs=yes
> + ;;
> ++ loongarch*-*-linux*) srv_tgtobj="arch/loongarch.o
> arch/loongarch-linux-nat.o"
> ++ srv_tgtobj="${srv_tgtobj} linux-loongarch-low.o"
> ++ srv_tgtobj="${srv_tgtobj} ${srv_linux_obj}"
> ++ srv_linux_regsets=yes
> ++ srv_linux_usrregs=yes
> ++ srv_linux_thread_db=yes
> ++ ;;
> + m68*-*-linux*) if test "$gdb_cv_m68k_is_coldfire" = yes; then
> + srv_regobj=reg-cf.o
> + else
> +diff --git gdb-10.2/gdbserver/linux-loongarch-low.cc
> gdb-10.2/gdbserver/linux-loongarch-low.cc
> +new file mode 100644
> +index 0000000..cd8ad6e
> +--- /dev/null
> ++++ gdb-10.2/gdbserver/linux-loongarch-low.cc
> +@@ -0,0 +1,284 @@
> ++/* GNU/Linux/LoongArch specific low level interface, for the remote
> server
> ++ for GDB.
> ++ Copyright (C) 2022 Free Software Foundation, Inc.
> ++
> ++ This file is part of GDB.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the License, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program. If not, see <http://www.gnu.org/licenses/>.
> */
> ++
> ++#include "server.h"
> ++#include "linux-low.h"
> ++#include "tdesc.h"
> ++#include "elf/common.h"
> ++#include "arch/loongarch-linux-nat.h"
> ++
> ++/* Linux target op definitions for the LoongArch architecture. */
> ++
> ++class loongarch_target : public linux_process_target
> ++{
> ++public:
> ++
> ++ const regs_info *get_regs_info () override;
> ++
> ++ int breakpoint_kind_from_pc (CORE_ADDR *pcptr) override;
> ++
> ++ const gdb_byte *sw_breakpoint_from_kind (int kind, int *size) override;
> ++
> ++protected:
> ++
> ++ void low_arch_setup () override;
> ++
> ++ bool low_cannot_fetch_register (int regno) override;
> ++
> ++ bool low_cannot_store_register (int regno) override;
> ++
> ++ bool low_fetch_register (regcache *regcache, int regno) override;
> ++
> ++ bool low_supports_breakpoints () override;
> ++
> ++ CORE_ADDR low_get_pc (regcache *regcache) override;
> ++
> ++ void low_set_pc (regcache *regcache, CORE_ADDR newpc) override;
> ++
> ++ bool low_breakpoint_at (CORE_ADDR pc) override;
> ++};
> ++
> ++/* The singleton target ops object. */
> ++
> ++static loongarch_target the_loongarch_target;
> ++
> ++bool
> ++loongarch_target::low_cannot_fetch_register (int regno)
> ++{
> ++ gdb_assert_not_reached ("linux target op low_cannot_fetch_register "
> ++ "is not implemented by the target");
> ++}
> ++
> ++bool
> ++loongarch_target::low_cannot_store_register (int regno)
> ++{
> ++ gdb_assert_not_reached ("linux target op low_cannot_store_register "
> ++ "is not implemented by the target");
> ++}
> ++
> ++/* Implementation of linux target ops method "low_arch_setup". */
> ++
> ++void
> ++loongarch_target::low_arch_setup ()
> ++{
> ++ static const char *expedite_regs[] = { "r3", "pc", NULL };
> ++ int pid = lwpid_of (current_thread);
> ++ struct target_desc *tdesc = loongarch_linux_read_description_runtime
> (pid);
> ++
> ++ if (!tdesc->expedite_regs)
> ++ init_target_desc (tdesc, expedite_regs);
> ++ current_process ()->tdesc = tdesc;
> ++}
> ++
> ++/* Collect GPRs from REGCACHE into BUF. */
> ++
> ++static void
> ++loongarch_fill_gregset (struct regcache *regcache, void *buf)
> ++{
> ++ const struct target_desc *tdesc = regcache->tdesc;
> ++ elf_gregset_t *regset = (elf_gregset_t *) buf;
> ++ int regno = find_regno (tdesc, "r0");
> ++ int i;
> ++
> ++ for (i = 1; i < 32; i++)
> ++ collect_register (regcache, regno + i, *regset + i);
> ++ collect_register_by_name (regcache, "orig_a0", *regset + 32);
> ++ collect_register_by_name (regcache, "pc", *regset + 33);
> ++ collect_register_by_name (regcache, "badv", *regset + 34);
> ++}
> ++
> ++/* Supply GPRs from BUF into REGCACHE. */
> ++
> ++static void
> ++loongarch_store_gregset (struct regcache *regcache, const void *buf)
> ++{
> ++ const struct target_desc *tdesc = regcache->tdesc;
> ++ const elf_gregset_t *regset = (const elf_gregset_t *) buf;
> ++ int regno = find_regno (tdesc, "r0");
> ++ int i;
> ++
> ++ supply_register_zeroed (regcache, regno);
> ++ for (i = 1; i < 32; i++)
> ++ supply_register (regcache, regno + i, *regset + i);
> ++ supply_register_by_name (regcache, "orig_a0", *regset + 32);
> ++ supply_register_by_name (regcache, "pc", *regset + 33);
> ++ supply_register_by_name (regcache, "badv", *regset + 34);
> ++}
> ++
> ++/* Collect FPRs from REGCACHE into BUF. */
> ++
> ++static void
> ++loongarch_fill_fpregset (struct regcache *regcache, void *buf)
> ++{
> ++ const struct target_desc *tdesc = regcache->tdesc;
> ++ int f = find_regno (tdesc, "f0");
> ++ int fcc = find_regno (tdesc, "fcc0");
> ++ int flen = register_size (regcache->tdesc, f);
> ++ gdb_byte *regbuf = (gdb_byte *) buf;
> ++ int i;
> ++
> ++ for (i = 0; i < ELF_NFPREG - 2; i++, regbuf += flen)
> ++ collect_register (regcache, f + i, regbuf);
> ++ for (i = 0; i < 8; i++)
> ++ collect_register (regcache, fcc + i, regbuf++);
> ++ collect_register_by_name (regcache, "fcsr", regbuf);
> ++}
> ++
> ++/* Supply FPRs from BUF into REGCACHE. */
> ++
> ++static void
> ++loongarch_store_fpregset (struct regcache *regcache, const void *buf)
> ++{
> ++ const struct target_desc *tdesc = regcache->tdesc;
> ++ int f = find_regno (tdesc, "f0");
> ++ int fcc = find_regno (tdesc, "fcc0");
> ++ int flen = register_size (regcache->tdesc, f);
> ++ const gdb_byte *regbuf = (const gdb_byte *) buf;
> ++ int i;
> ++
> ++ for (i = 0; i < ELF_NFPREG - 2; i++, regbuf += flen)
> ++ supply_register (regcache, f + i, regbuf);
> ++ for (i = 0; i < 8; i++)
> ++ supply_register (regcache, fcc + i, regbuf++);
> ++ supply_register_by_name (regcache, "fcsr", regbuf);
> ++}
> ++
> ++/* LoongArch/Linux regsets. */
> ++static struct regset_info loongarch_regsets[] = {
> ++ { PTRACE_GETREGSET, PTRACE_SETREGSET, NT_PRSTATUS, sizeof
> (elf_gregset_t),
> ++ GENERAL_REGS, loongarch_fill_gregset, loongarch_store_gregset },
> ++ { PTRACE_GETREGSET, PTRACE_SETREGSET, NT_FPREGSET, sizeof
> (elf_fpregset_t),
> ++ FP_REGS, loongarch_fill_fpregset, loongarch_store_fpregset },
> ++ NULL_REGSET
> ++};
> ++
> ++/* LoongArch/Linux regset information. */
> ++static struct regsets_info loongarch_regsets_info =
> ++ {
> ++ loongarch_regsets, /* regsets */
> ++ 0, /* num_regsets */
> ++ NULL, /* disabled_regsets */
> ++ };
> ++
> ++/* Definition of linux_target_ops data member "regs_info". */
> ++static struct regs_info loongarch_regs =
> ++ {
> ++ NULL, /* regset_bitmap */
> ++ NULL, /* usrregs */
> ++ &loongarch_regsets_info,
> ++ };
> ++
> ++/* Implementation of linux target ops method "get_regs_info". */
> ++
> ++const regs_info *
> ++loongarch_target::get_regs_info ()
> ++{
> ++ return &loongarch_regs;
> ++}
> ++
> ++/* Implementation of linux target ops method "low_fetch_register". */
> ++
> ++bool
> ++loongarch_target::low_fetch_register (regcache *regcache, int regno)
> ++{
> ++ const struct target_desc *tdesc = regcache->tdesc;
> ++
> ++ if (regno != find_regno (tdesc, "r0"))
> ++ return false;
> ++ supply_register_zeroed (regcache, regno);
> ++ return true;
> ++}
> ++
> ++bool
> ++loongarch_target::low_supports_breakpoints ()
> ++{
> ++ return true;
> ++}
> ++
> ++/* Implementation of linux target ops method "low_get_pc". */
> ++
> ++CORE_ADDR
> ++loongarch_target::low_get_pc (regcache *regcache)
> ++{
> ++ if (register_size (regcache->tdesc, 0) == 8)
> ++ return linux_get_pc_64bit (regcache);
> ++ else
> ++ return linux_get_pc_32bit (regcache);
> ++}
> ++
> ++/* Implementation of linux target ops method "low_set_pc". */
> ++
> ++void
> ++loongarch_target::low_set_pc (regcache *regcache, CORE_ADDR newpc)
> ++{
> ++ if (register_size (regcache->tdesc, 0) == 8)
> ++ linux_set_pc_64bit (regcache, newpc);
> ++ else
> ++ linux_set_pc_32bit (regcache, newpc);
> ++}
> ++
> ++#define loongarch_breakpoint_len 4
> ++
> ++/* LoongArch BRK software debug mode instruction.
> ++ This instruction needs to match gdb/loongarch-tdep.c
> ++ (loongarch_default_breakpoint). */
> ++static const gdb_byte loongarch_breakpoint[] = {0x05, 0x00, 0x2a, 0x00};
> ++
> ++/* Implementation of target ops method "breakpoint_kind_from_pc". */
> ++
> ++int
> ++loongarch_target::breakpoint_kind_from_pc (CORE_ADDR *pcptr)
> ++{
> ++ return loongarch_breakpoint_len;
> ++}
> ++
> ++/* Implementation of target ops method "sw_breakpoint_from_kind". */
> ++
> ++const gdb_byte *
> ++loongarch_target::sw_breakpoint_from_kind (int kind, int *size)
> ++{
> ++ *size = loongarch_breakpoint_len;
> ++ return (const gdb_byte *) &loongarch_breakpoint;
> ++}
> ++
> ++/* Implementation of linux target ops method "low_breakpoint_at". */
> ++
> ++bool
> ++loongarch_target::low_breakpoint_at (CORE_ADDR pc)
> ++{
> ++ gdb_byte insn[loongarch_breakpoint_len];
> ++
> ++ read_memory (pc, (unsigned char *) &insn, loongarch_breakpoint_len);
> ++ if (memcmp (insn, loongarch_breakpoint, loongarch_breakpoint_len) == 0)
> ++ return true;
> ++
> ++ return false;
> ++}
> ++
> ++/* The linux target ops object. */
> ++
> ++linux_process_target *the_linux_target = &the_loongarch_target;
> ++
> ++/* Initialize the LoongArch/Linux target. */
> ++
> ++void
> ++initialize_low_arch ()
> ++{
> ++ initialize_regsets_info (&loongarch_regsets_info);
> ++}
> +diff --git gdb-10.2/include/dis-asm.h gdb-10.2/include/dis-asm.h
> +index 0532cef..ce43595 100644
> +--- gdb-10.2/include/dis-asm.h
> ++++ gdb-10.2/include/dis-asm.h
> +@@ -303,6 +303,7 @@ extern void print_arm_disassembler_options (FILE *);
> + extern void print_arc_disassembler_options (FILE *);
> + extern void print_s390_disassembler_options (FILE *);
> + extern void print_wasm32_disassembler_options (FILE *);
> ++extern void print_loongarch_disassembler_options (FILE *);
> + extern bfd_boolean aarch64_symbol_is_valid (asymbol *, struct
> disassemble_info *);
> + extern bfd_boolean arm_symbol_is_valid (asymbol *, struct
> disassemble_info *);
> + extern bfd_boolean csky_symbol_is_valid (asymbol *, struct
> disassemble_info *);
> +diff --git gdb-10.2/include/elf/common.h gdb-10.2/include/elf/common.h
> +index 571e21a..9296080 100644
> +--- gdb-10.2/include/elf/common.h
> ++++ gdb-10.2/include/elf/common.h
> +@@ -342,6 +342,7 @@
> + #define EM_BPF 247 /* Linux BPF ? in-kernel virtual
> machine. */
> + #define EM_NFP 250 /* Netronome Flow Processor. */
> + #define EM_CSKY 252 /* C-SKY processor family. */
> ++#define EM_LOONGARCH 258 /* LoongArch */
> +
> + /* If it is necessary to assign new unofficial EM_* values, please pick
> large
> + random numbers (0x8523, 0xa7f2, etc.) to minimize the chances of
> collision
> +@@ -662,6 +663,19 @@
> + /* note name must be "LINUX". */
> + #define NT_ARC_V2 0x600 /* ARC HS accumulator/extra
> registers. */
> + /* note name must be "LINUX". */
> ++#define NT_LARCH_CPUCFG 0xa00 /* LoongArch CPU config registers
> */
> ++ /* note name must be "LINUX". */
> ++#define NT_LARCH_CSR 0xa01 /* LoongArch Control State
> Registers */
> ++ /* note name must be "LINUX". */
> ++#define NT_LARCH_LSX 0xa02 /* LoongArch SIMD eXtension
> registers */
> ++ /* note name must be "LINUX". */
> ++#define NT_LARCH_LASX 0xa03 /* LoongArch Advanced SIMD
> eXtension registers */
> ++ /* note name must be "LINUX". */
> ++#define NT_LARCH_LBT 0xa04 /* LoongArch Binary Translation
> registers */
> ++ /* note name must be "CORE". */
> ++#define NT_RISCV_CSR 0x900 /* RISC-V Control and Status
> Registers */
> ++ /* note name must be "LINUX". */
> ++
> + #define NT_SIGINFO 0x53494749 /* Fields of siginfo_t. */
> + #define NT_FILE 0x46494c45 /* Description of mapped
> files. */
> +
> +diff --git gdb-10.2/include/elf/loongarch.h
> gdb-10.2/include/elf/loongarch.h
> +new file mode 100644
> +index 0000000..74757b8
> +--- /dev/null
> ++++ gdb-10.2/include/elf/loongarch.h
> +@@ -0,0 +1,267 @@
> ++/* Copyright (C) 2021-2022 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of GNU Binutils.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the license, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program; see the file COPYING3. If not,
> ++ see <http://www.gnu.org/licenses/>. */
> ++
> ++#ifndef _ELF_LOONGARCH_H
> ++#define _ELF_LOONGARCH_H
> ++
> ++#include "elf/reloc-macros.h"
> ++#include "libiberty.h"
> ++
> ++START_RELOC_NUMBERS (elf_loongarch_reloc_type)
> ++/* Used by the dynamic linker. */
> ++RELOC_NUMBER (R_LARCH_NONE, 0)
> ++RELOC_NUMBER (R_LARCH_32, 1)
> ++RELOC_NUMBER (R_LARCH_64, 2)
> ++RELOC_NUMBER (R_LARCH_RELATIVE, 3)
> ++RELOC_NUMBER (R_LARCH_COPY, 4)
> ++RELOC_NUMBER (R_LARCH_JUMP_SLOT, 5)
> ++RELOC_NUMBER (R_LARCH_TLS_DTPMOD32, 6)
> ++RELOC_NUMBER (R_LARCH_TLS_DTPMOD64, 7)
> ++RELOC_NUMBER (R_LARCH_TLS_DTPREL32, 8)
> ++RELOC_NUMBER (R_LARCH_TLS_DTPREL64, 9)
> ++RELOC_NUMBER (R_LARCH_TLS_TPREL32, 10)
> ++RELOC_NUMBER (R_LARCH_TLS_TPREL64, 11)
> ++RELOC_NUMBER (R_LARCH_IRELATIVE, 12)
> ++
> ++/* Reserved for future relocs that the dynamic linker must understand.
> */
> ++
> ++/* Used by the static linker for relocating .text. */
> ++RELOC_NUMBER (R_LARCH_MARK_LA, 20)
> ++RELOC_NUMBER (R_LARCH_MARK_PCREL, 21)
> ++
> ++RELOC_NUMBER (R_LARCH_SOP_PUSH_PCREL, 22)
> ++
> ++RELOC_NUMBER (R_LARCH_SOP_PUSH_ABSOLUTE, 23)
> ++
> ++RELOC_NUMBER (R_LARCH_SOP_PUSH_DUP, 24)
> ++RELOC_NUMBER (R_LARCH_SOP_PUSH_GPREL, 25)
> ++RELOC_NUMBER (R_LARCH_SOP_PUSH_TLS_TPREL, 26)
> ++RELOC_NUMBER (R_LARCH_SOP_PUSH_TLS_GOT, 27)
> ++RELOC_NUMBER (R_LARCH_SOP_PUSH_TLS_GD, 28)
> ++RELOC_NUMBER (R_LARCH_SOP_PUSH_PLT_PCREL, 29)
> ++
> ++RELOC_NUMBER (R_LARCH_SOP_ASSERT, 30)
> ++RELOC_NUMBER (R_LARCH_SOP_NOT, 31)
> ++RELOC_NUMBER (R_LARCH_SOP_SUB, 32)
> ++RELOC_NUMBER (R_LARCH_SOP_SL, 33)
> ++RELOC_NUMBER (R_LARCH_SOP_SR, 34)
> ++RELOC_NUMBER (R_LARCH_SOP_ADD, 35)
> ++RELOC_NUMBER (R_LARCH_SOP_AND, 36)
> ++RELOC_NUMBER (R_LARCH_SOP_IF_ELSE, 37)
> ++RELOC_NUMBER (R_LARCH_SOP_POP_32_S_10_5, 38)
> ++RELOC_NUMBER (R_LARCH_SOP_POP_32_U_10_12, 39)
> ++RELOC_NUMBER (R_LARCH_SOP_POP_32_S_10_12, 40)
> ++RELOC_NUMBER (R_LARCH_SOP_POP_32_S_10_16, 41)
> ++RELOC_NUMBER (R_LARCH_SOP_POP_32_S_10_16_S2, 42)
> ++RELOC_NUMBER (R_LARCH_SOP_POP_32_S_5_20, 43)
> ++RELOC_NUMBER (R_LARCH_SOP_POP_32_S_0_5_10_16_S2, 44)
> ++RELOC_NUMBER (R_LARCH_SOP_POP_32_S_0_10_10_16_S2, 45)
> ++RELOC_NUMBER (R_LARCH_SOP_POP_32_U, 46)
> ++
> ++/* Used by the static linker for relocating non .text. */
> ++RELOC_NUMBER (R_LARCH_ADD8, 47)
> ++RELOC_NUMBER (R_LARCH_ADD16, 48)
> ++RELOC_NUMBER (R_LARCH_ADD24, 49)
> ++RELOC_NUMBER (R_LARCH_ADD32, 50)
> ++RELOC_NUMBER (R_LARCH_ADD64, 51)
> ++RELOC_NUMBER (R_LARCH_SUB8, 52)
> ++RELOC_NUMBER (R_LARCH_SUB16, 53)
> ++RELOC_NUMBER (R_LARCH_SUB24, 54)
> ++RELOC_NUMBER (R_LARCH_SUB32, 55)
> ++RELOC_NUMBER (R_LARCH_SUB64, 56)
> ++
> ++/* I don't know what it is. Existing in almost all other arch. */
> ++RELOC_NUMBER (R_LARCH_GNU_VTINHERIT, 57)
> ++RELOC_NUMBER (R_LARCH_GNU_VTENTRY, 58)
> ++
> ++
> ++/* B16:
> ++ beq/bne/blt/bge/bltu/bgeu/jirl
> ++ %b16 (sym). */
> ++RELOC_NUMBER (R_LARCH_B16, 64)
> ++/* B21:
> ++ beqz/bnez
> ++ %b16 (sym). */
> ++RELOC_NUMBER (R_LARCH_B21, 65)
> ++/* B26:
> ++ b/bl
> ++ %b26 (sym) or %plt (sym). */
> ++RELOC_NUMBER (R_LARCH_B26, 66)
> ++
> ++/* ABS: 32/64
> ++ lu12i.w
> ++ %abs_hi20 (sym). */
> ++RELOC_NUMBER (R_LARCH_ABS_HI20, 67)
> ++/* ABS: 32/64
> ++ ori
> ++ %abs_lo12 (sym). */
> ++RELOC_NUMBER (R_LARCH_ABS_LO12, 68)
> ++
> ++/* ABS: 64
> ++ lu32i.d
> ++ %abs64_lo20 (sym). */
> ++RELOC_NUMBER (R_LARCH_ABS64_LO20, 69)
> ++/* ABS: 64
> ++ lu52i.d
> ++ %abs64_hi12 (sym). */
> ++RELOC_NUMBER (R_LARCH_ABS64_HI12, 70)
> ++
> ++/* PCREL: 32/64
> ++ pcalau12i
> ++ %pc_hi20 (sym). */
> ++RELOC_NUMBER (R_LARCH_PCALA_HI20, 71)
> ++/* PCREL: 32/64
> ++ addi.w/addi.d
> ++ %pc_lo12 (sym). */
> ++RELOC_NUMBER (R_LARCH_PCALA_LO12, 72)
> ++/* PCREL: 64
> ++ lu32i.d
> ++ %pc64_lo20 (sym). */
> ++RELOC_NUMBER (R_LARCH_PCALA64_LO20, 73)
> ++/* PCREL: 64
> ++ lu52i.d
> ++ %pc64_hi12 (sym). */
> ++RELOC_NUMBER (R_LARCH_PCALA64_HI12, 74)
> ++
> ++/* GOT: 32/64
> ++ pcalau12i
> ++ %got_pc_hi20 (got). */
> ++RELOC_NUMBER (R_LARCH_GOT_PC_HI20, 75)
> ++/* GOT: 32/64
> ++ ld.w/ld.d
> ++ %got_pc_lo12 (got). */
> ++RELOC_NUMBER (R_LARCH_GOT_PC_LO12, 76)
> ++/* GOT: 32/64
> ++ lu32i.d
> ++ %got_pc_lo12 (got). */
> ++RELOC_NUMBER (R_LARCH_GOT64_PC_LO20, 77)
> ++/* GOT64: PCREL
> ++ lu52i.d
> ++ %got64_pc_hi12 (got). */
> ++RELOC_NUMBER (R_LARCH_GOT64_PC_HI12, 78)
> ++/* GOT32/64: ABS
> ++ lu12i.w
> ++ %got_hi20 (got). */
> ++RELOC_NUMBER (R_LARCH_GOT_HI20, 79)
> ++/* GOT: 32/64: ABS
> ++ ori
> ++ %got_lo12 (got). */
> ++RELOC_NUMBER (R_LARCH_GOT_LO12, 80)
> ++/* GOT64: ABS
> ++ lu32i.d
> ++ %got64_lo20 (got). */
> ++RELOC_NUMBER (R_LARCH_GOT64_LO20, 81)
> ++/* GOT64: ABS
> ++ lu52i.d
> ++ %got64_hi12 (got). */
> ++RELOC_NUMBER (R_LARCH_GOT64_HI12, 82)
> ++
> ++/* TLS-LE: 32/64
> ++ lu12i.w
> ++ %le_hi20 (sym). */
> ++RELOC_NUMBER (R_LARCH_TLS_LE_HI20, 83)
> ++/* TLS-LE: 32/64
> ++ ori
> ++ %le_lo12 (sym). */
> ++RELOC_NUMBER (R_LARCH_TLS_LE_LO12, 84)
> ++/* TLS-LE: 64
> ++ lu32i.d
> ++ %le64_lo20 (sym). */
> ++RELOC_NUMBER (R_LARCH_TLS_LE64_LO20, 85)
> ++/* TLS-LE: 64
> ++ lu52i.d
> ++ %le64_hi12 (sym). */
> ++RELOC_NUMBER (R_LARCH_TLS_LE64_HI12, 86)
> ++
> ++/* TLS-IE: 32/64
> ++ pcalau12i
> ++ %ie_pc_hi20 (sym). */
> ++RELOC_NUMBER (R_LARCH_TLS_IE_PC_HI20, 87)
> ++RELOC_NUMBER (R_LARCH_TLS_IE_PC_LO12, 88)
> ++RELOC_NUMBER (R_LARCH_TLS_IE64_PC_LO20, 89)
> ++RELOC_NUMBER (R_LARCH_TLS_IE64_PC_HI12, 90)
> ++
> ++/* TLS-IE: 32/64: ABS
> ++ lu12i.w
> ++ %ie_hi20 (sym). */
> ++RELOC_NUMBER (R_LARCH_TLS_IE_HI20, 91)
> ++RELOC_NUMBER (R_LARCH_TLS_IE_LO12, 92)
> ++RELOC_NUMBER (R_LARCH_TLS_IE64_LO20, 93)
> ++RELOC_NUMBER (R_LARCH_TLS_IE64_HI12, 94)
> ++
> ++/* TLS-LD: 32/64
> ++ pcalau12i
> ++ %ld_pc_hi20 (sym). */
> ++RELOC_NUMBER (R_LARCH_TLS_LD_PC_HI20, 95)
> ++/* TLS-LD: 32/64: ABS
> ++ lu12i.w
> ++ %ld_hi20 (sym). */
> ++RELOC_NUMBER (R_LARCH_TLS_LD_HI20, 96)
> ++
> ++/* TLS-GD: 32/64
> ++ pcalau12i
> ++ %gd_pc_hi20 (sym). */
> ++RELOC_NUMBER (R_LARCH_TLS_GD_PC_HI20, 97)
> ++/* TLS-GD: 32/64: ABS
> ++ lu12i.w
> ++ %gd_hi20 (sym). */
> ++RELOC_NUMBER (R_LARCH_TLS_GD_HI20, 98)
> ++
> ++/* For eh_frame and debug info. */
> ++RELOC_NUMBER (R_LARCH_32_PCREL, 99)
> ++
> ++/* RELAX. */
> ++RELOC_NUMBER (R_LARCH_RELAX, 100)
> ++
> ++END_RELOC_NUMBERS (R_LARCH_count)
> ++
> ++/* Processor specific flags for the ELF header e_flags field. */
> ++/*The flag lp64s/lp64f/lp64d/ilp32s/ilp32f/ilp32d 3bits. */
> ++#define EF_LOONGARCH_ABI_LP64_SOFT_FLOAT 0x1
> ++#define EF_LOONGARCH_ABI_LP64_SINGLE_FLOAT 0x2
> ++#define EF_LOONGARCH_ABI_LP64_DOUBLE_FLOAT 0x3
> ++
> ++#define EF_LOONGARCH_ABI_ILP32_SOFT_FLOAT 0x5
> ++#define EF_LOONGARCH_ABI_ILP32_SINGLE_FLOAT 0x6
> ++#define EF_LOONGARCH_ABI_ILP32_DOUBLE_FLOAT 0x7
> ++
> ++#define EF_LOONGARCH_ABI_MASK 0x7
> ++#define EF_LOONGARCH_ABI_ILP32_MASK 0x4
> ++#define EF_LOONGARCH_ABI_FLOAT_MASK 0x3
> ++#define EF_LOONGARCH_ABI_SOFT_FLOAT_MASK 0x1
> ++#define EF_LOONGARCH_ABI_SINGLE_FLOAT_MASK 0x2
> ++#define EF_LOONGARCH_ABI_DOUBLE_FLOAT_MASK 0x3
> ++
> ++#define EF_LOONGARCH_ABI(abi) (EF_LOONGARCH_ABI_MASK & (abi))
> ++
> ++#define EF_LOONGARCH_IS_LP64(abi) \
> ++ (EF_LOONGARCH_ABI(abi) && (!(EF_LOONGARCH_ABI(abi) &
> EF_LOONGARCH_ABI_ILP32_MASK)))
> ++#define EF_LOONGARCH_IS_ILP32(abi) \
> ++ (EF_LOONGARCH_ABI(abi) && (EF_LOONGARCH_ABI(abi) &
> EF_LOONGARCH_ABI_ILP32_MASK))
> ++
> ++#define EF_LOONGARCH_IS_SOFT_FLOAT(abi) \
> ++ (!((EF_LOONGARCH_ABI(abi) & EF_LOONGARCH_ABI_FLOAT_MASK) ^
> EF_LOONGARCH_ABI_SOFT_FLOAT_MASK))
> ++
> ++#define EF_LOONGARCH_IS_SINGLE_FLOAT(abi) \
> ++ (!((EF_LOONGARCH_ABI(abi) & EF_LOONGARCH_ABI_FLOAT_MASK) ^
> EF_LOONGARCH_ABI_SINGLE_FLOAT_MASK))
> ++
> ++#define EF_LOONGARCH_IS_DOUBLE_FLOAT(abi) \
> ++ (!((EF_LOONGARCH_ABI(abi) & EF_LOONGARCH_ABI_FLOAT_MASK) ^
> EF_LOONGARCH_ABI_DOUBLE_FLOAT_MASK))
> ++
> ++#endif /* _ELF_LOONGARCH_H */
> +diff --git gdb-10.2/include/opcode/loongarch.h
> gdb-10.2/include/opcode/loongarch.h
> +new file mode 100644
> +index 0000000..c392234
> +--- /dev/null
> ++++ gdb-10.2/include/opcode/loongarch.h
> +@@ -0,0 +1,239 @@
> ++/* LoongArch assembler/disassembler support.
> ++
> ++ Copyright (C) 2021-2022 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of GNU Binutils.
> ++
> ++ This program is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3 of the license, or
> ++ (at your option) any later version.
> ++
> ++ This program is distributed in the hope that it will be useful,
> ++ but WITHOUT ANY WARRANTY; without even the implied warranty of
> ++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> ++ GNU General Public License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program; see the file COPYING3. If not,
> ++ see <http://www.gnu.org/licenses/>. */
> ++
> ++#ifndef _LOONGARCH_H_
> ++#define _LOONGARCH_H_
> ++#include <stdint.h>
> ++
> ++#ifdef __cplusplus
> ++extern "C"
> ++{
> ++#endif
> ++
> ++ typedef uint32_t insn_t;
> ++
> ++ struct loongarch_opcode
> ++ {
> ++ const insn_t match;
> ++ const insn_t mask; /* High 1 byte is main opcode and it must be
> 0xf. */
> ++#define LARCH_INSN_OPC(insn) ((insn & 0xf0000000) >> 28)
> ++ const char *const name;
> ++
> ++ /* ACTUAL PARAMETER:
> ++
> ++ // BNF with regular expression.
> ++args : token* end
> ++
> ++ // just few char separate 'iden'
> ++token : ','
> ++| '('
> ++| ')'
> ++| iden // maybe a label (include at least one alphabet),
> ++ maybe a number, maybe a expr
> ++| regname
> ++
> ++regname : '$' iden
> ++
> ++iden : [a-zA-Z0-9\.\+\-]+
> ++
> ++end : '\0'
> ++
> ++
> ++FORMAT: A string to describe the format of actual parameter including
> ++bit field infomation. For example, "r5:5,r0:5,sr10:16<<2" matches
> ++"$12,$13,12345" and "$4,$7,a_label". That 'sr' means the instruction
> ++may need relocate. '10:16' means bit field of instruction.
> ++In a 'format', every 'escape's can be replaced to 'iden' or 'regname'
> ++acrroding to its meaning. We fill all information needed by
> ++disassembing and assembing to 'format'.
> ++
> ++ // BNF with regular expression.
> ++format : escape (literal+ escape)* literal* end
> ++| (literal+ escape)* literal* end
> ++
> ++end : '\0' // Get here means parse end.
> ++
> ++ // The intersection between any two among FIRST (end), FIRST
> ++ // (literal) and FIRST (escape) must be empty.
> ++ // So we can build a simple parser.
> ++literal : ','
> ++| '('
> ++| ')'
> ++
> ++ // Double '<'s means the real number is the immediate after shifting
> left.
> ++escape : esc_ch bit_field '<' '<' dec2
> ++| esc_ch bit_field
> ++| esc_ch // for MACRO. non-macro format must indicate 'bit_field'
> ++
> ++ // '|' means to concatenate nonadjacent bit fields
> ++ // For example, "10:16|0:4" means
> ++ // "16 bits starting from the 10th bit concatenating with 4 bits
> ++ // starting from the 0th bit".
> ++ // This is to say "[25..10]||[3..0]" (little endian).
> ++b_field : dec2 ':' dec2
> ++| dec2 ':' dec2 '|' bit_field
> ++
> ++esc_ch : 's' 'r' // signed immediate or label need relocate
> ++| 's' // signed immediate no need relocate
> ++| 'u' // unsigned immediate
> ++| 'l' // label needed relocate
> ++| 'r' // general purpose registers
> ++| 'f' // FPU registers
> ++| 'v' // 128 bit SIMD register
> ++| 'x' // 256 bit SIMD register
> ++
> ++dec2 : [1-9][0-9]?
> ++| 0
> ++
> ++*/
> ++ const char *const format;
> ++
> ++ /* MACRO: Indicate how a macro instruction expand for assembling.
> ++ The main is to replace the '%num'(means the 'num'th 'escape' in
> ++ 'format') in 'macro' string to get the real instruction.
> ++
> ++ Maybe need
> ++ */
> ++ const char *const macro;
> ++ const int *include;
> ++ const int *exclude;
> ++
> ++ const unsigned long pinfo;
> ++#define USELESS 0x0l
> ++ };
> ++
> ++ struct hash_control;
> ++
> ++ struct loongarch_ase
> ++ {
> ++ const int *enabled;
> ++ struct loongarch_opcode *const opcodes;
> ++ const int *include;
> ++ const int *exclude;
> ++
> ++ /* For disassemble to create main opcode hash table. */
> ++ const struct loongarch_opcode *opc_htab[16];
> ++ unsigned char opc_htab_inited;
> ++
> ++ /* For GAS to create hash table. */
> ++ struct htab *name_hash_entry;
> ++ };
> ++
> ++ extern int is_unsigned (const char *);
> ++ extern int is_signed (const char *);
> ++ extern int is_branch_label (const char *);
> ++
> ++ extern int loongarch_get_bit_field_width (const char *bit_field, char
> **end);
> ++ extern int32_t loongarch_decode_imm (const char *bit_field, insn_t
> insn,
> ++ int si);
> ++
> ++#define MAX_ARG_NUM_PLUS_2 9
> ++
> ++ extern size_t loongarch_split_args_by_comma (char *args,
> ++ const char *arg_strs[]);
> ++ extern char *loongarch_cat_splited_strs (const char *arg_strs[]);
> ++ extern insn_t loongarch_foreach_args (
> ++ const char *format, const char *arg_strs[],
> ++ int32_t (*helper) (char esc1, char esc2, const char *bit_field,
> ++ const char *arg, void *context),
> ++ void *context);
> ++
> ++ extern int loongarch_check_format (const char *format);
> ++ extern int loongarch_check_macro (const char *format, const char
> *macro);
> ++
> ++ extern char *loongarch_expand_macro_with_format_map (
> ++ const char *format, const char *macro, const char *const arg_strs[],
> ++ const char *(*map) (char esc1, char esc2, const char *arg),
> ++ char *(*helper) (const char *const arg_strs[], void *context),
> ++ void *context, size_t len_str);
> ++ extern char *loongarch_expand_macro (
> ++ const char *macro, const char *const arg_strs[],
> ++ char *(*helper) (const char *const arg_strs[], void *context),
> ++ void *context, size_t len_str);
> ++ extern size_t loongarch_bits_imm_needed (int64_t imm, int si);
> ++
> ++ extern void loongarch_eliminate_adjacent_repeat_char (char *dest, char
> c);
> ++
> ++ extern int loongarch_parse_dis_options (const char *opts_in);
> ++ extern void loongarch_disassemble_one (
> ++ int64_t pc, insn_t insn,
> ++ int (*fprintf_func) (void *stream, const char *format, ...), void
> *stream);
> ++
> ++ extern const char *const loongarch_r_normal_name[32];
> ++ extern const char *const loongarch_r_lp64_name[32];
> ++ extern const char *const loongarch_r_lp64_name1[32];
> ++ extern const char *const loongarch_f_normal_name[32];
> ++ extern const char *const loongarch_f_lp64_name[32];
> ++ extern const char *const loongarch_f_lp64_name1[32];
> ++ extern const char *const loongarch_c_normal_name[8];
> ++ extern const char *const loongarch_cr_normal_name[4];
> ++ extern const char *const loongarch_v_normal_name[32];
> ++ extern const char *const loongarch_x_normal_name[32];
> ++
> ++ extern struct loongarch_ase loongarch_ASEs[];
> ++
> ++ extern struct loongarch_ASEs_option
> ++ {
> ++ struct opt_abi
> ++ {
> ++ int elf_abi;
> ++ } abi;
> ++#define ase_abi abi.elf_abi
> ++
> ++ struct opt_isa
> ++ {
> ++ int use_ilp32;
> ++ int use_lp64;
> ++
> ++ int use_soft_float;
> ++ int use_single_float;
> ++ int use_double_float;
> ++
> ++ int use_lsx;
> ++ int use_lasx;
> ++
> ++ int use_la_local_with_abs;
> ++ int use_la_global_with_pcrel;
> ++ int use_la_global_with_abs;
> ++ } isa;
> ++#define ase_ilp32 isa.use_ilp32
> ++#define ase_lp64 isa.use_lp64
> ++
> ++#define ase_nf isa.use_soft_float
> ++#define ase_sf isa.use_single_float
> ++#define ase_df isa.use_double_float
> ++
> ++#define ase_lsx isa.use_lsx
> ++#define ase_lasx isa.use_lasx
> ++
> ++#define ase_labs isa.use_la_local_with_abs
> ++#define ase_gpcr isa.use_la_global_with_pcrel
> ++#define ase_gabs isa.use_la_global_with_abs
> ++
> ++ } LARCH_opts;
> ++
> ++ extern size_t loongarch_insn_length (insn_t insn);
> ++
> ++#ifdef __cplusplus
> ++}
> ++#endif
> ++
> ++#endif /* _LOONGARCH_H_ */
> +diff --git gdb-10.2/opcodes/Makefile.am gdb-10.2/opcodes/Makefile.am
> +index 7318bf0..c9139e5 100644
> +--- gdb-10.2/opcodes/Makefile.am
> ++++ gdb-10.2/opcodes/Makefile.am
> +@@ -162,6 +162,9 @@ TARGET_LIBOPCODES_CFILES = \
> + lm32-ibld.c \
> + lm32-opc.c \
> + lm32-opinst.c \
> ++ loongarch-opc.c \
> ++ loongarch-dis.c \
> ++ loongarch-coder.c \
> + m10200-dis.c \
> + m10200-opc.c \
> + m10300-dis.c \
> +diff --git gdb-10.2/opcodes/Makefile.in gdb-10.2/opcodes/Makefile.in
> +index ddb9346..19a90fb 100644
> +--- gdb-10.2/opcodes/Makefile.in
> ++++ gdb-10.2/opcodes/Makefile.in
> +@@ -552,6 +552,9 @@ TARGET_LIBOPCODES_CFILES = \
> + lm32-ibld.c \
> + lm32-opc.c \
> + lm32-opinst.c \
> ++ loongarch-opc.c \
> ++ loongarch-dis.c \
> ++ loongarch-coder.c \
> + m10200-dis.c \
> + m10200-opc.c \
> + m10300-dis.c \
> +@@ -966,6 +969,9 @@ distclean-compile:
> + @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/lm32-ibld.Plo(a)am__quote
> @
> + @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/lm32-opc.Plo(a)am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/lm32-opinst.Plo@am__quote@
> ++@AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/loongarch-coder.Plo@am__quote@
> ++@AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/loongarch-dis.Plo@am__quote@
> ++@AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/loongarch-opc.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/m10200-dis.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/m10200-opc.Plo@am__quote@
> + @AMDEP_TRUE@@am__include@ @am__quote@
> ./$(DEPDIR)/m10300-dis.Plo@am__quote@
> +diff --git gdb-10.2/opcodes/configure gdb-10.2/opcodes/configure
> +index e448c9e..68380ae 100755
> +--- gdb-10.2/opcodes/configure
> ++++ gdb-10.2/opcodes/configure
> +@@ -12949,6 +12949,7 @@ if test x${all_targets} = xfalse ; then
> + bfd_z80_arch) ta="$ta z80-dis.lo" ;;
> + bfd_z8k_arch) ta="$ta z8k-dis.lo" ;;
> + bfd_bpf_arch) ta="$ta bpf-asm.lo bpf-desc.lo bpf-dis.lo
> bpf-ibld.lo bpf-opc.lo" using_cgen=yes ;;
> ++ bfd_loongarch_arch) ta="$ta loongarch-dis.lo
> loongarch-opc.lo loongarch-coder.lo" ;;
> +
> + "") ;;
> + *) as_fn_error $? "*** unknown target architecture
> $arch" "$LINENO" 5 ;;
> +diff --git gdb-10.2/opcodes/configure.ac gdb-10.2/opcodes/configure.ac
> +index 00be9c8..434cc2f 100644
> +--- gdb-10.2/opcodes/configure.ac
> ++++ gdb-10.2/opcodes/configure.ac
> +@@ -340,6 +340,7 @@ if test x${all_targets} = xfalse ; then
> + bfd_z80_arch) ta="$ta z80-dis.lo" ;;
> + bfd_z8k_arch) ta="$ta z8k-dis.lo" ;;
> + bfd_bpf_arch) ta="$ta bpf-asm.lo bpf-desc.lo bpf-dis.lo
> bpf-ibld.lo bpf-opc.lo" using_cgen=yes ;;
> ++ bfd_loongarch_arch) ta="$ta loongarch-dis.lo loongarch-opc.lo
> loongarch-coder.lo" ;;
> +
> + "") ;;
> + *) AC_MSG_ERROR(*** unknown target architecture
> $arch) ;;
> +diff --git gdb-10.2/opcodes/disassemble.c gdb-10.2/opcodes/disassemble.c
> +index 290dcdd..eec560e 100644
> +--- gdb-10.2/opcodes/disassemble.c
> ++++ gdb-10.2/opcodes/disassemble.c
> +@@ -49,6 +49,7 @@
> + #define ARCH_ip2k
> + #define ARCH_iq2000
> + #define ARCH_lm32
> ++#define ARCH_loongarch
> + #define ARCH_m32c
> + #define ARCH_m32r
> + #define ARCH_m68hc11
> +@@ -551,6 +552,11 @@ disassembler (enum bfd_architecture a,
> + case bfd_arch_tilepro:
> + disassemble = print_insn_tilepro;
> + break;
> ++#endif
> ++#ifdef ARCH_loongarch
> ++ case bfd_arch_loongarch:
> ++ disassemble = print_insn_loongarch;
> ++ break;
> + #endif
> + default:
> + return 0;
> +@@ -591,6 +597,9 @@ disassembler_usage (FILE *stream ATTRIBUTE_UNUSED)
> + #ifdef ARCH_wasm32
> + print_wasm32_disassembler_options (stream);
> + #endif
> ++#ifdef ARCH_loongarch
> ++ print_loongarch_disassembler_options (stream);
> ++#endif
> +
> + return;
> + }
> +diff --git gdb-10.2/opcodes/disassemble.h gdb-10.2/opcodes/disassemble.h
> +index 89db886..cc529c9 100644
> +--- gdb-10.2/opcodes/disassemble.h
> ++++ gdb-10.2/opcodes/disassemble.h
> +@@ -100,6 +100,7 @@ extern int print_insn_xtensa (bfd_vma,
> disassemble_info *);
> + extern int print_insn_z80 (bfd_vma, disassemble_info *);
> + extern int print_insn_z8001 (bfd_vma, disassemble_info *);
> + extern int print_insn_z8002 (bfd_vma, disassemble_info *);
> ++extern int print_insn_loongarch (bfd_vma, disassemble_info
> *);
> +
> + extern disassembler_ftype csky_get_disassembler (bfd *);
> + extern disassembler_ftype rl78_get_disassembler (bfd *);
> +diff --git gdb-10.2/opcodes/loongarch-coder.c
> gdb-10.2/opcodes/loongarch-coder.c
> +new file mode 100644
> +index 0000000..f5e10b9
> +--- /dev/null
> ++++ gdb-10.2/opcodes/loongarch-coder.c
> +@@ -0,0 +1,481 @@
> ++/* LoongArch opcode support.
> ++ Copyright (C) 2021-2022 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of the GNU opcodes library.
> ++
> ++ This library is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3, or (at your option)
> ++ any later version.
> ++
> ++ It is distributed in the hope that it will be useful, but WITHOUT
> ++ ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
> ++ or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
> ++ License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program; see the file COPYING3. If not,
> ++ see <http://www.gnu.org/licenses/>. */
> ++#include "sysdep.h"
> ++#include "opcode/loongarch.h"
> ++
> ++int
> ++is_unsigned (const char *c_str)
> ++{
> ++ if (c_str[0] == '0' && (c_str[1] == 'x' || c_str[1] == 'X'))
> ++ {
> ++ c_str += 2;
> ++ while (('a' <= *c_str && *c_str <= 'f')
> ++ || ('A' <= *c_str && *c_str <= 'F')
> ++ || ('0' <= *c_str && *c_str <= '9'))
> ++ c_str++;
> ++ }
> ++ else if (*c_str == '\0')
> ++ return 0;
> ++ else
> ++ while ('0' <= *c_str && *c_str <= '9')
> ++ c_str++;
> ++ return *c_str == '\0';
> ++}
> ++
> ++int
> ++is_signed (const char *c_str)
> ++{
> ++ return *c_str == '-' ? is_unsigned (c_str + 1) : is_unsigned (c_str);
> ++}
> ++
> ++int
> ++loongarch_get_bit_field_width (const char *bit_field, char **end)
> ++{
> ++ int width = 0;
> ++ char has_specify = 0, *bit_field_1 = (char *) bit_field;
> ++ if (bit_field_1 && *bit_field_1 != '\0')
> ++ while (1)
> ++ {
> ++ strtol (bit_field_1, &bit_field_1, 10);
> ++
> ++ if (*bit_field_1 != ':')
> ++ break;
> ++ bit_field_1++;
> ++
> ++ width += strtol (bit_field_1, &bit_field_1, 10);
> ++ has_specify = 1;
> ++
> ++ if (*bit_field_1 != '|')
> ++ break;
> ++ bit_field_1++;
> ++ }
> ++ if (end)
> ++ *end = bit_field_1;
> ++ return has_specify ? width : -1;
> ++}
> ++
> ++int32_t
> ++loongarch_decode_imm (const char *bit_field, insn_t insn, int si)
> ++{
> ++ int32_t ret = 0;
> ++ uint32_t t;
> ++ int len = 0, width, b_start;
> ++ char *bit_field_1 = (char *) bit_field;
> ++ while (1)
> ++ {
> ++ b_start = strtol (bit_field_1, &bit_field_1, 10);
> ++ if (*bit_field_1 != ':')
> ++ break;
> ++ width = strtol (bit_field_1 + 1, &bit_field_1, 10);
> ++ len += width;
> ++
> ++ t = insn;
> ++ t <<= sizeof (t) * 8 - width - b_start;
> ++ t >>= sizeof (t) * 8 - width;
> ++ ret <<= width;
> ++ ret |= t;
> ++
> ++ if (*bit_field_1 != '|')
> ++ break;
> ++ bit_field_1++;
> ++ }
> ++
> ++ if (*bit_field_1 == '<' && *(++bit_field_1) == '<')
> ++ {
> ++ width = atoi (bit_field_1 + 1);
> ++ ret <<= width;
> ++ len += width;
> ++ }
> ++ else if (*bit_field_1 == '+')
> ++ ret += atoi (bit_field_1 + 1);
> ++
> ++ /* Extend signed bit. */
> ++ if (si)
> ++ {
> ++ uint32_t sign = 1u << (len - 1);
> ++ ret = (ret ^ sign) - sign;
> ++ }
> ++
> ++ return ret;
> ++}
> ++
> ++static insn_t
> ++loongarch_encode_imm (const char *bit_field, int32_t imm)
> ++{
> ++ char *bit_field_1 = (char *) bit_field;
> ++ char *t = bit_field_1;
> ++ int width, b_start;
> ++ insn_t ret = 0;
> ++ uint32_t i;
> ++ uint32_t uimm = (uint32_t)imm;
> ++
> ++ width = loongarch_get_bit_field_width (t, &t);
> ++ if (width == -1)
> ++ return ret;
> ++
> ++ if (*t == '<' && *(++t) == '<')
> ++ width += atoi (t + 1);
> ++ else if (*t == '+')
> ++ uimm -= atoi (t + 1);
> ++
> ++ uimm = width ? (uimm << (sizeof (uimm) * 8 - width)) : 0;
> ++
> ++ while (1)
> ++ {
> ++ b_start = strtol (bit_field_1, &bit_field_1, 10);
> ++ if (*bit_field_1 != ':')
> ++ break;
> ++ width = strtol (bit_field_1 + 1, &bit_field_1, 10);
> ++ i = uimm;
> ++ i = width ? (i >> (sizeof (i) * 8 - width)) : 0;
> ++ i = (b_start == 32) ? 0 : (i << b_start);
> ++ ret |= i;
> ++ uimm = (width == 32) ? 0 : (uimm << width);
> ++
> ++ if (*bit_field_1 != '|')
> ++ break;
> ++ bit_field_1++;
> ++ }
> ++ return ret;
> ++}
> ++
> ++/* Parse such FORMAT
> ++ ""
> ++ "u"
> ++ "v0:5,r5:5,s10:10<<2"
> ++ "r0:5,r5:5,r10:5,u15:2+1"
> ++ "r,r,u0:5+32,u0:5+1"
> ++*/
> ++static int
> ++loongarch_parse_format (const char *format, char *esc1s, char *esc2s,
> ++ const char **bit_fields)
> ++{
> ++ size_t arg_num = 0;
> ++
> ++ if (*format == '\0')
> ++ goto end;
> ++
> ++ while (1)
> ++ {
> ++ /* esc1 esc2
> ++ for "[a-zA-Z][a-zA-Z]?" */
> ++ if (('a' <= *format && *format <= 'z')
> ++ || ('A' <= *format && *format <= 'Z'))
> ++ {
> ++ *esc1s++ = *format++;
> ++ if (('a' <= *format && *format <= 'z')
> ++ || ('A' <= *format && *format <= 'Z'))
> ++ *esc2s++ = *format++;
> ++ else
> ++ *esc2s++ = '\0';
> ++ }
> ++ else
> ++ return -1;
> ++
> ++ arg_num++;
> ++ if (MAX_ARG_NUM_PLUS_2 - 2 < arg_num)
> ++ /* Need larger MAX_ARG_NUM_PLUS_2. */
> ++ return -1;
> ++
> ++ *bit_fields++ = format;
> ++
> ++ if ('0' <= *format && *format <= '9')
> ++ {
> ++ /* For "[0-9]+:[0-9]+(\|[0-9]+:[0-9]+)*". */
> ++ while (1)
> ++ {
> ++ while ('0' <= *format && *format <= '9')
> ++ format++;
> ++
> ++ if (*format != ':')
> ++ return -1;
> ++ format++;
> ++
> ++ if (!('0' <= *format && *format <= '9'))
> ++ return -1;
> ++ while ('0' <= *format && *format <= '9')
> ++ format++;
> ++
> ++ if (*format != '|')
> ++ break;
> ++ format++;
> ++ }
> ++
> ++ /* For "((\+|<<)[1-9][0-9]*)?". */
> ++ do
> ++ {
> ++ if (*format == '+')
> ++ format++;
> ++ else if (format[0] == '<' && format[1] == '<')
> ++ format += 2;
> ++ else
> ++ break;
> ++
> ++ if (!('1' <= *format && *format <= '9'))
> ++ return -1;
> ++ while ('0' <= *format && *format <= '9')
> ++ format++;
> ++ }
> ++ while (0);
> ++ }
> ++
> ++ if (*format == ',')
> ++ format++;
> ++ else if (*format == '\0')
> ++ break;
> ++ else
> ++ return -1;
> ++ }
> ++
> ++ end:
> ++ *esc1s = '\0';
> ++ return 0;
> ++}
> ++
> ++size_t
> ++loongarch_split_args_by_comma (char *args, const char *arg_strs[])
> ++{
> ++ size_t num = 0;
> ++
> ++ if (*args)
> ++ arg_strs[num++] = args;
> ++ for (; *args; args++)
> ++ if (*args == ',')
> ++ {
> ++ if (MAX_ARG_NUM_PLUS_2 - 1 == num)
> ++ break;
> ++ else
> ++ *args = '\0', arg_strs[num++] = args + 1;
> ++ }
> ++ arg_strs[num] = NULL;
> ++ return num;
> ++}
> ++
> ++char *
> ++loongarch_cat_splited_strs (const char *arg_strs[])
> ++{
> ++ char *ret;
> ++ size_t n, l;
> ++
> ++ for (l = 0, n = 0; arg_strs[n]; n++)
> ++ l += strlen (arg_strs[n]);
> ++ ret = malloc (l + n + 1);
> ++ if (!ret)
> ++ return ret;
> ++
> ++ ret[0] = '\0';
> ++ if (0 < n)
> ++ strcat (ret, arg_strs[0]);
> ++ for (l = 1; l < n; l++)
> ++ strcat (ret, ","), strcat (ret, arg_strs[l]);
> ++ return ret;
> ++}
> ++
> ++insn_t
> ++loongarch_foreach_args (const char *format, const char *arg_strs[],
> ++ int32_t (*helper) (char esc1, char esc2,
> ++ const char *bit_field,
> ++ const char *arg, void *context),
> ++ void *context)
> ++{
> ++ char esc1s[MAX_ARG_NUM_PLUS_2 - 1], esc2s[MAX_ARG_NUM_PLUS_2 - 1];
> ++ const char *bit_fields[MAX_ARG_NUM_PLUS_2 - 1];
> ++ size_t i;
> ++ insn_t ret = 0;
> ++ int ok;
> ++
> ++ ok = loongarch_parse_format (format, esc1s, esc2s, bit_fields) == 0;
> ++
> ++ /* Make sure the num of actual args is equal to the num of escape. */
> ++ for (i = 0; esc1s[i] && arg_strs[i]; i++)
> ++ ;
> ++ ok = ok && !esc1s[i] && !arg_strs[i];
> ++
> ++ if (ok && helper)
> ++ {
> ++ for (i = 0; arg_strs[i]; i++)
> ++ ret |= loongarch_encode_imm (bit_fields[i],
> ++ helper (esc1s[i], esc2s[i],
> ++ bit_fields[i], arg_strs[i],
> ++ context));
> ++ ret |= helper ('\0', '\0', NULL, NULL, context);
> ++ }
> ++
> ++ return ret;
> ++}
> ++
> ++int
> ++loongarch_check_format (const char *format)
> ++{
> ++ char esc1s[MAX_ARG_NUM_PLUS_2 - 1], esc2s[MAX_ARG_NUM_PLUS_2 - 1];
> ++ const char *bit_fields[MAX_ARG_NUM_PLUS_2 - 1];
> ++
> ++ if (!format)
> ++ return -1;
> ++
> ++ return loongarch_parse_format (format, esc1s, esc2s, bit_fields);
> ++}
> ++
> ++int
> ++loongarch_check_macro (const char *format, const char *macro)
> ++{
> ++ int num_of_args;
> ++ char esc1s[MAX_ARG_NUM_PLUS_2 - 1], esc2s[MAX_ARG_NUM_PLUS_2 - 1];
> ++ const char *bit_fields[MAX_ARG_NUM_PLUS_2 - 1];
> ++
> ++ if (!format || !macro
> ++ || loongarch_parse_format (format, esc1s, esc2s, bit_fields) != 0)
> ++ return -1;
> ++
> ++ for (num_of_args = 0; esc1s[num_of_args]; num_of_args++)
> ++ ;
> ++
> ++ for (; macro[0]; macro++)
> ++ if (macro[0] == '%')
> ++ {
> ++ macro++;
> ++ if ('1' <= macro[0] && macro[0] <= '9')
> ++ {
> ++ if (num_of_args < macro[0] - '0')
> ++ /* Out of args num. */
> ++ return -1;
> ++ }
> ++ else if (macro[0] == 'f')
> ++ ;
> ++ else if (macro[0] == '%')
> ++ ;
> ++ else
> ++ return -1;
> ++ }
> ++ return 0;
> ++}
> ++
> ++static const char *
> ++I (char esc_ch1 ATTRIBUTE_UNUSED, char esc_ch2 ATTRIBUTE_UNUSED,
> ++ const char *c_str)
> ++{
> ++ return c_str;
> ++}
> ++
> ++char *
> ++loongarch_expand_macro_with_format_map (
> ++ const char *format, const char *macro, const char *const arg_strs[],
> ++ const char *(*map) (char esc1, char esc2, const char *arg),
> ++ char *(*helper) (const char *const arg_strs[], void *context), void
> *context,
> ++ size_t len_str)
> ++{
> ++ char esc1s[MAX_ARG_NUM_PLUS_2 - 1], esc2s[MAX_ARG_NUM_PLUS_2 - 1];
> ++ const char *bit_fields[MAX_ARG_NUM_PLUS_2 - 1];
> ++ const char *src;
> ++ char *dest;
> ++
> ++ /* The expanded macro character length does not exceed 1000, and
> number of
> ++ label is 6 at most in the expanded macro. The len_str is the length
> of
> ++ str. */
> ++ char *buffer =(char *) malloc(1024 + 6 * len_str);
> ++
> ++ if (format)
> ++ loongarch_parse_format (format, esc1s, esc2s, bit_fields);
> ++
> ++ src = macro;
> ++ dest = buffer;
> ++
> ++ while (*src)
> ++ if (*src == '%')
> ++ {
> ++ src++;
> ++ if ('1' <= *src && *src <= '9')
> ++ {
> ++ size_t i = *src - '1';
> ++ const char *t = map (esc1s[i], esc2s[i], arg_strs[i]);
> ++ while (*t)
> ++ *dest++ = *t++;
> ++ }
> ++ else if (*src == '%')
> ++ *dest++ = '%';
> ++ else if (*src == 'f' && helper)
> ++ {
> ++ char *b, *t;
> ++ t = b = (*helper) (arg_strs, context);
> ++ if (b)
> ++ {
> ++ while (*t)
> ++ *dest++ = *t++;
> ++ free (b);
> ++ }
> ++ }
> ++ src++;
> ++ }
> ++ else
> ++ *dest++ = *src++;
> ++
> ++ *dest = '\0';
> ++ return buffer;
> ++}
> ++
> ++char *
> ++loongarch_expand_macro (const char *macro, const char *const arg_strs[],
> ++ char *(*helper) (const char *const arg_strs[],
> ++ void *context),
> ++ void *context, size_t len_str)
> ++{
> ++ return loongarch_expand_macro_with_format_map (NULL, macro, arg_strs,
> I,
> ++ helper, context, len_str);
> ++}
> ++
> ++size_t
> ++loongarch_bits_imm_needed (int64_t imm, int si)
> ++{
> ++ size_t ret;
> ++ if (si)
> ++ {
> ++ if (imm < 0)
> ++ {
> ++ uint64_t uimm = (uint64_t) imm;
> ++ uint64_t uimax = UINT64_C (1) << 63;
> ++ for (ret = 0; (uimm & uimax) != 0; uimm <<= 1, ret++)
> ++ ;
> ++ ret = 64 - ret + 1;
> ++ }
> ++ else
> ++ ret = loongarch_bits_imm_needed (imm, 0) + 1;
> ++ }
> ++ else
> ++ {
> ++ uint64_t t = imm;
> ++ for (ret = 0; t; t >>= 1, ret++)
> ++ ;
> ++ }
> ++ return ret;
> ++}
> ++
> ++void
> ++loongarch_eliminate_adjacent_repeat_char (char *dest, char c)
> ++{
> ++ if (c == '\0')
> ++ return;
> ++ char *src = dest;
> ++ while (*dest)
> ++ {
> ++ while (src[0] == c && src[0] == src[1])
> ++ src++;
> ++ *dest++ = *src++;
> ++ }
> ++}
> +diff --git gdb-10.2/opcodes/loongarch-dis.c
> gdb-10.2/opcodes/loongarch-dis.c
> +new file mode 100644
> +index 0000000..9dcf989
> +--- /dev/null
> ++++ gdb-10.2/opcodes/loongarch-dis.c
> +@@ -0,0 +1,342 @@
> ++/* LoongArch opcode support.
> ++ Copyright (C) 2021-2022 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of the GNU opcodes library.
> ++
> ++ This library is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3, or (at your option)
> ++ any later version.
> ++
> ++ It is distributed in the hope that it will be useful, but WITHOUT
> ++ ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
> ++ or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
> ++ License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program; see the file COPYING3. If not,
> ++ see <http://www.gnu.org/licenses/>. */
> ++
> ++#include "sysdep.h"
> ++#include "disassemble.h"
> ++#include "opintl.h"
> ++#include "opcode/loongarch.h"
> ++#include "libiberty.h"
> ++#include <stdlib.h>
> ++
> ++static const struct loongarch_opcode *
> ++get_loongarch_opcode_by_binfmt (insn_t insn)
> ++{
> ++ const struct loongarch_opcode *it;
> ++ struct loongarch_ase *ase;
> ++ size_t i;
> ++ for (ase = loongarch_ASEs; ase->enabled; ase++)
> ++ {
> ++ if (!*ase->enabled || (ase->include && !*ase->include)
> ++ || (ase->exclude && *ase->exclude))
> ++ continue;
> ++
> ++ if (!ase->opc_htab_inited)
> ++ {
> ++ for (it = ase->opcodes; it->mask; it++)
> ++ if (!ase->opc_htab[LARCH_INSN_OPC (it->match)]
> ++ && it->macro == NULL)
> ++ ase->opc_htab[LARCH_INSN_OPC (it->match)] = it;
> ++ for (i = 0; i < 16; i++)
> ++ if (!ase->opc_htab[i])
> ++ ase->opc_htab[i] = it;
> ++ ase->opc_htab_inited = 1;
> ++ }
> ++
> ++ it = ase->opc_htab[LARCH_INSN_OPC (insn)];
> ++ for (; it->name; it++)
> ++ if ((insn & it->mask) == it->match && it->mask
> ++ && !(it->include && !*it->include)
> ++ && !(it->exclude && *it->exclude))
> ++ return it;
> ++ }
> ++ return NULL;
> ++}
> ++
> ++static const char *const *loongarch_r_disname = NULL;
> ++static const char *const *loongarch_f_disname = NULL;
> ++static const char *const *loongarch_c_disname = NULL;
> ++static const char *const *loongarch_cr_disname = NULL;
> ++static const char *const *loongarch_v_disname = NULL;
> ++static const char *const *loongarch_x_disname = NULL;
> ++
> ++static void
> ++set_default_loongarch_dis_options (void)
> ++{
> ++ LARCH_opts.ase_ilp32 = 1;
> ++ LARCH_opts.ase_lp64 = 1;
> ++ LARCH_opts.ase_sf = 1;
> ++ LARCH_opts.ase_df = 1;
> ++ LARCH_opts.ase_lsx = 1;
> ++ LARCH_opts.ase_lasx = 1;
> ++
> ++ loongarch_r_disname = loongarch_r_lp64_name;
> ++ loongarch_f_disname = loongarch_f_lp64_name;
> ++ loongarch_c_disname = loongarch_c_normal_name;
> ++ loongarch_cr_disname = loongarch_cr_normal_name;
> ++ loongarch_v_disname = loongarch_v_normal_name;
> ++ loongarch_x_disname = loongarch_x_normal_name;
> ++}
> ++
> ++static int
> ++parse_loongarch_dis_option (const char *option)
> ++{
> ++ if (strcmp (option, "numeric") == 0)
> ++ {
> ++ loongarch_r_disname = loongarch_r_normal_name;
> ++ loongarch_f_disname = loongarch_f_normal_name;
> ++ }
> ++ return -1;
> ++}
> ++
> ++static int
> ++parse_loongarch_dis_options (const char *opts_in)
> ++{
> ++ set_default_loongarch_dis_options ();
> ++
> ++ if (opts_in == NULL)
> ++ return 0;
> ++
> ++ char *opts, *opt, *opt_end;
> ++ opts = xmalloc (strlen (opts_in) + 1);
> ++ strcpy (opts, opts_in);
> ++
> ++ for (opt = opt_end = opts; opt_end != NULL; opt = opt_end + 1)
> ++ {
> ++ if ((opt_end = strchr (opt, ',')) != NULL)
> ++ *opt_end = 0;
> ++ if (parse_loongarch_dis_option (opt) != 0)
> ++ return -1;
> ++ }
> ++ free (opts);
> ++ return 0;
> ++}
> ++
> ++static int32_t
> ++dis_one_arg (char esc1, char esc2, const char *bit_field,
> ++ const char *arg ATTRIBUTE_UNUSED, void *context)
> ++{
> ++ static int need_comma = 0;
> ++ struct disassemble_info *info = context;
> ++ insn_t insn = *(insn_t *) info->private_data;
> ++ int32_t imm, u_imm;
> ++
> ++ if (esc1)
> ++ {
> ++ if (need_comma)
> ++ info->fprintf_func (info->stream, ", ");
> ++ need_comma = 1;
> ++ imm = loongarch_decode_imm (bit_field, insn, 1);
> ++ u_imm = loongarch_decode_imm (bit_field, insn, 0);
> ++ }
> ++
> ++ switch (esc1)
> ++ {
> ++ case 'r':
> ++ info->fprintf_func (info->stream, "%s",
> loongarch_r_disname[u_imm]);
> ++ break;
> ++ case 'f':
> ++ info->fprintf_func (info->stream, "%s",
> loongarch_f_disname[u_imm]);
> ++ break;
> ++ case 'c':
> ++ switch (esc2)
> ++ {
> ++ case 'r':
> ++ info->fprintf_func (info->stream, "%s",
> loongarch_cr_disname[u_imm]);
> ++ break;
> ++ default:
> ++ info->fprintf_func (info->stream, "%s",
> loongarch_c_disname[u_imm]);
> ++ }
> ++ break;
> ++ case 'v':
> ++ info->fprintf_func (info->stream, "%s",
> loongarch_v_disname[u_imm]);
> ++ break;
> ++ case 'x':
> ++ info->fprintf_func (info->stream, "%s",
> loongarch_x_disname[u_imm]);
> ++ break;
> ++ case 'u':
> ++ info->fprintf_func (info->stream, "0x%x", u_imm);
> ++ break;
> ++ case 's':
> ++ if (imm == 0)
> ++ info->fprintf_func (info->stream, "%d", imm);
> ++ else
> ++ info->fprintf_func (info->stream, "%d(0x%x)", imm, u_imm);
> ++ switch (esc2)
> ++ {
> ++ case 'b':
> ++ info->insn_type = dis_branch;
> ++ info->target += imm;
> ++ }
> ++ break;
> ++ case '\0':
> ++ need_comma = 0;
> ++ }
> ++ return 0;
> ++}
> ++
> ++static void
> ++disassemble_one (insn_t insn, struct disassemble_info *info)
> ++{
> ++ const struct loongarch_opcode *opc = get_loongarch_opcode_by_binfmt
> (insn);
> ++
> ++#ifdef LOONGARCH_DEBUG
> ++ char have_space[32] = { 0 };
> ++ insn_t t;
> ++ int i;
> ++ const char *t_f = opc ? opc->format : NULL;
> ++ if (t_f)
> ++ while (*t_f)
> ++ {
> ++ while (('a' <= t_f[0] && t_f[0] <= 'z')
> ++ || ('A' <= t_f[0] && t_f[0] <= 'Z')
> ++ || t_f[0] == ',')
> ++ t_f++;
> ++ while (1)
> ++ {
> ++ i = strtol (t_f, &t_f, 10);
> ++ have_space[i] = 1;
> ++ t_f++; /* ':' */
> ++ i += strtol (t_f, &t_f, 10);
> ++ have_space[i] = 1;
> ++ if (t_f[0] == '|')
> ++ t_f++;
> ++ else
> ++ break;
> ++ }
> ++ if (t_f[0] == '<')
> ++ t_f += 2; /* '<' '<' */
> ++ strtol (t_f, &t_f, 10);
> ++ }
> ++
> ++ have_space[28] = 1;
> ++ have_space[0] = 0;
> ++ t = ~((insn_t) -1 >> 1);
> ++ for (i = 31; 0 <= i; i--)
> ++ {
> ++ if (t & insn)
> ++ info->fprintf_func (info->stream, "1");
> ++ else
> ++ info->fprintf_func (info->stream, "0");
> ++ if (have_space[i])
> ++ info->fprintf_func (info->stream, " ");
> ++ t = t >> 1;
> ++ }
> ++ info->fprintf_func (info->stream, "\t");
> ++#endif
> ++
> ++ if (!opc)
> ++ {
> ++ info->insn_type = dis_noninsn;
> ++ info->fprintf_func (info->stream, "0x%08x", insn);
> ++ return;
> ++ }
> ++
> ++ info->insn_type = dis_nonbranch;
> ++ info->fprintf_func (info->stream, "%-12s", opc->name);
> ++
> ++ {
> ++ char *fake_args = xmalloc (strlen (opc->format) + 1);
> ++ const char *fake_arg_strs[MAX_ARG_NUM_PLUS_2];
> ++ strcpy (fake_args, opc->format);
> ++ if (0 < loongarch_split_args_by_comma (fake_args, fake_arg_strs))
> ++ info->fprintf_func (info->stream, "\t");
> ++ info->private_data = &insn;
> ++ loongarch_foreach_args (opc->format, fake_arg_strs, dis_one_arg,
> info);
> ++ free (fake_args);
> ++ }
> ++
> ++ if (info->insn_type == dis_branch || info->insn_type == dis_condbranch
> ++ /* Someother if we have extra info to print. */)
> ++ info->fprintf_func (info->stream, "\t#");
> ++
> ++ if (info->insn_type == dis_branch || info->insn_type == dis_condbranch)
> ++ {
> ++ info->fprintf_func (info->stream, " ");
> ++ info->print_address_func (info->target, info);
> ++ }
> ++}
> ++
> ++int
> ++print_insn_loongarch (bfd_vma memaddr, struct disassemble_info *info)
> ++{
> ++ insn_t insn;
> ++ int status;
> ++
> ++ static int not_init_yet = 1;
> ++ if (not_init_yet)
> ++ {
> ++ parse_loongarch_dis_options (info->disassembler_options);
> ++ not_init_yet = 0;
> ++ }
> ++
> ++ info->bytes_per_chunk = 4;
> ++ info->bytes_per_line = 4;
> ++ info->display_endian = BFD_ENDIAN_LITTLE;
> ++ info->insn_info_valid = 1;
> ++ info->target = memaddr;
> ++
> ++ if ((status = info->read_memory_func (memaddr, (bfd_byte *) &insn,
> ++ sizeof (insn), info)) != 0)
> ++ {
> ++ info->memory_error_func (status, memaddr, info);
> ++ return -1; /* loongarch_insn_length (0); */
> ++ }
> ++
> ++ disassemble_one (insn, info);
> ++
> ++ return loongarch_insn_length (insn);
> ++}
> ++
> ++void
> ++print_loongarch_disassembler_options (FILE *stream)
> ++{
> ++ fprintf (stream, _("\n\
> ++The following LoongArch disassembler options are supported for use\n\
> ++with the -M switch (multiple options should be separated by
> commas):\n"));
> ++
> ++ fprintf (stream, _("\n\
> ++ numeric Print numeric register names, rather than ABI
> names.\n"));
> ++ fprintf (stream, _("\n"));
> ++}
> ++
> ++int
> ++loongarch_parse_dis_options (const char *opts_in)
> ++{
> ++ return parse_loongarch_dis_options (opts_in);
> ++}
> ++
> ++static void
> ++my_print_address_func (bfd_vma addr, struct disassemble_info *dinfo)
> ++{
> ++ dinfo->fprintf_func (dinfo->stream, "0x%llx", (long long) addr);
> ++}
> ++
> ++void
> ++loongarch_disassemble_one (int64_t pc, insn_t insn,
> ++ int (*fprintf_func) (void *stream,
> ++ const char *format, ...),
> ++ void *stream)
> ++{
> ++ static struct disassemble_info my_disinfo =
> ++ {
> ++ .print_address_func = my_print_address_func,
> ++ };
> ++ static int not_init_yet = 1;
> ++ if (not_init_yet)
> ++ {
> ++ loongarch_parse_dis_options (NULL);
> ++ not_init_yet = 0;
> ++ }
> ++
> ++ my_disinfo.fprintf_func = fprintf_func;
> ++ my_disinfo.stream = stream;
> ++ my_disinfo.target = pc;
> ++ disassemble_one (insn, &my_disinfo);
> ++}
> +diff --git gdb-10.2/opcodes/loongarch-opc.c
> gdb-10.2/opcodes/loongarch-opc.c
> +new file mode 100644
> +index 0000000..be0de61
> +--- /dev/null
> ++++ gdb-10.2/opcodes/loongarch-opc.c
> +@@ -0,0 +1,870 @@
> ++/* LoongArch opcode support.
> ++ Copyright (C) 2021-2022 Free Software Foundation, Inc.
> ++ Contributed by Loongson Ltd.
> ++
> ++ This file is part of the GNU opcodes library.
> ++
> ++ This library is free software; you can redistribute it and/or modify
> ++ it under the terms of the GNU General Public License as published by
> ++ the Free Software Foundation; either version 3, or (at your option)
> ++ any later version.
> ++
> ++ It is distributed in the hope that it will be useful, but WITHOUT
> ++ ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
> ++ or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
> ++ License for more details.
> ++
> ++ You should have received a copy of the GNU General Public License
> ++ along with this program; see the file COPYING3. If not,
> ++ see <http://www.gnu.org/licenses/>. */
> ++
> ++#include <stddef.h>
> ++#include "opcode/loongarch.h"
> ++#include "libiberty.h"
> ++
> ++struct loongarch_ASEs_option LARCH_opts;
> ++
> ++size_t
> ++loongarch_insn_length (insn_t insn ATTRIBUTE_UNUSED)
> ++{
> ++ return 4;
> ++}
> ++
> ++const char *const loongarch_r_normal_name[32] =
> ++{
> ++ "$r0", "$r1", "$r2", "$r3", "$r4", "$r5", "$r6", "$r7",
> ++ "$r8", "$r9", "$r10", "$r11", "$r12", "$r13", "$r14", "$r15",
> ++ "$r16", "$r17", "$r18", "$r19", "$r20", "$r21", "$r22", "$r23",
> ++ "$r24", "$r25", "$r26", "$r27", "$r28", "$r29", "$r30", "$r31",
> ++};
> ++
> ++const char *const loongarch_r_lp64_name[32] =
> ++{
> ++ "$zero", "$ra", "$tp", "$sp", "$a0", "$a1", "$a2", "$a3",
> ++ "$a4", "$a5", "$a6", "$a7", "$t0", "$t1", "$t2", "$t3",
> ++ "$t4", "$t5", "$t6", "$t7", "$t8", "$x", "$fp", "$s0",
> ++ "$s1", "$s2", "$s3", "$s4", "$s5", "$s6", "$s7", "$s8",
> ++};
> ++
> ++const char *const loongarch_r_lp64_name1[32] =
> ++{
> ++ "", "", "", "", "$v0", "$v1", "", "", "", "", "", "", "", "", "", "",
> ++ "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "",
> ++};
> ++
> ++const char *const loongarch_f_normal_name[32] =
> ++{
> ++ "$f0", "$f1", "$f2", "$f3", "$f4", "$f5", "$f6", "$f7",
> ++ "$f8", "$f9", "$f10", "$f11", "$f12", "$f13", "$f14", "$f15",
> ++ "$f16", "$f17", "$f18", "$f19", "$f20", "$f21", "$f22", "$f23",
> ++ "$f24", "$f25", "$f26", "$f27", "$f28", "$f29", "$f30", "$f31",
> ++};
> ++
> ++const char *const loongarch_f_lp64_name[32] =
> ++{
> ++ "$fa0", "$fa1", "$fa2", "$fa3", "$fa4", "$fa5", "$fa6", "$fa7",
> ++ "$ft0", "$ft1", "$ft2", "$ft3", "$ft4", "$ft5", "$ft6", "$ft7",
> ++ "$ft8", "$ft9", "$ft10", "$ft11", "$ft12", "$ft13", "$ft14", "$ft15",
> ++ "$fs0", "$fs1", "$fs2", "$fs3", "$fs4", "$fs5", "$fs6", "$fs7",
> ++};
> ++
> ++const char *const loongarch_f_lp64_name1[32] =
> ++{
> ++ "$fv0", "$fv1", "", "", "", "", "", "", "", "", "", "", "", "", "", "",
> ++ "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "",
> ++};
> ++
> ++const char *const loongarch_c_normal_name[8] =
> ++{
> ++ "$fcc0", "$fcc1", "$fcc2", "$fcc3", "$fcc4", "$fcc5", "$fcc6", "$fcc7",
> ++};
> ++
> ++const char *const loongarch_cr_normal_name[4] =
> ++{
> ++ "$scr0",
> ++ "$scr1",
> ++ "$scr2",
> ++ "$scr3",
> ++};
> ++
> ++const char *const loongarch_v_normal_name[32] =
> ++{
> ++ "$vr0", "$vr1", "$vr2", "$vr3", "$vr4", "$vr5", "$vr6", "$vr7",
> ++ "$vr8", "$vr9", "$vr10", "$vr11", "$vr12", "$vr13", "$vr14", "$vr15",
> ++ "$vr16", "$vr17", "$vr18", "$vr19", "$vr20", "$vr21", "$vr22", "$vr23",
> ++ "$vr24", "$vr25", "$vr26", "$vr27", "$vr28", "$vr29", "$vr30", "$vr31",
> ++};
> ++
> ++const char *const loongarch_x_normal_name[32] =
> ++{
> ++ "$xr0", "$xr1", "$xr2", "$xr3", "$xr4", "$xr5", "$xr6", "$xr7",
> ++ "$xr8", "$xr9", "$xr10", "$xr11", "$xr12", "$xr13", "$xr14", "$xr15",
> ++ "$xr16", "$xr17", "$xr18", "$xr19", "$xr20", "$xr21", "$xr22", "$xr23",
> ++ "$xr24", "$xr25", "$xr26", "$xr27", "$xr28", "$xr29", "$xr30", "$xr31",
> ++};
> ++
> ++/* Can not use xx_pa for abs. */
> ++
> ++/* For LoongArch32 abs. */
> ++#define INSN_LA_ABS32 \
> ++ "lu12i.w %1,%%abs_hi20(%2);" \
> ++ "ori %1,%1,%%abs_lo12(%2);", \
> ++ &LARCH_opts.ase_ilp32, \
> ++ &LARCH_opts.ase_lp64
> ++#define INSN_LA_ABS64 \
> ++ "lu12i.w %1,%%abs_hi20(%2);" \
> ++ "ori %1,%1,%%abs_lo12(%2);" \
> ++ "lu32i.d %1,%%abs64_lo20(%2);" \
> ++ "lu52i.d %1,%1,%%abs64_hi12(%2);", \
> ++ &LARCH_opts.ase_lp64, 0
> ++
> ++#define INSN_LA_PCREL32 \
> ++ "pcalau12i %1,%%pc_hi20(%2);" \
> ++ "addi.w %1,%1,%%pc_lo12(%2);", \
> ++ &LARCH_opts.ase_ilp32, \
> ++ &LARCH_opts.ase_lp64
> ++#define INSN_LA_PCREL64 \
> ++ "pcalau12i %1,%%pc_hi20(%2);" \
> ++ "addi.d %1,%1,%%pc_lo12(%2);", \
> ++ &LARCH_opts.ase_lp64, 0
> ++#define INSN_LA_PCREL64_LARGE \
> ++ "pcalau12i %1,%%pc_hi20(%3);" \
> ++ "addi.d %2,$r0,%%pc_lo12(%3);" \
> ++ "lu32i.d %2,%%pc64_lo20(%3);" \
> ++ "lu52i.d %2,%2,%%pc64_hi12(%3);" \
> ++ "add.d %1,%1,%2;", \
> ++ &LARCH_opts.ase_lp64, 0
> ++
> ++#define INSN_LA_GOT32 \
> ++ "pcalau12i %1,%%got_pc_hi20(%2);" \
> ++ "ld.w %1,%1,%%got_pc_lo12(%2);", \
> ++ &LARCH_opts.ase_ilp32, \
> ++ &LARCH_opts.ase_lp64
> ++/* got32 abs. */
> ++#define INSN_LA_GOT32_ABS \
> ++ "lu12i.w %1,%%got_hi20(%2);" \
> ++ "ori %1,%1,%%got_lo12(%2);" \
> ++ "ld.w %1,%1,0;", \
> ++ &LARCH_opts.ase_gabs, \
> ++ &LARCH_opts.ase_lp64
> ++#define INSN_LA_GOT64 \
> ++ "pcalau12i %1,%%got_pc_hi20(%2);" \
> ++ "ld.d %1,%1,%%got_pc_lo12(%2);", \
> ++ &LARCH_opts.ase_lp64, 0
> ++/* got64 abs. */
> ++#define INSN_LA_GOT64_LARGE_ABS \
> ++ "lu12i.w %1,%%got_hi20(%2);" \
> ++ "ori %1,%1,%%got_lo12(%2);" \
> ++ "lu32i.d %1,%%got64_lo20(%2);" \
> ++ "lu52i.d %1,%1,%%got64_hi12(%2);" \
> ++ "ld.d %1,%1,0", \
> ++ &LARCH_opts.ase_lp64, \
> ++ &LARCH_opts.ase_gpcr
> ++/* got64 pic. */
> ++#define INSN_LA_GOT64_LARGE_PCREL \
> ++ "pcalau12i %1,%%got_pc_hi20(%3);" \
> ++ "addi.d %2,$r0,%%got_pc_lo12(%3);" \
> ++ "lu32i.d %2,%%got64_pc_lo20(%3);" \
> ++ "lu52i.d %2,%2,%%got64_pc_hi12(%3);"\
> ++ "ldx.d %1,%1,%2;", \
> ++ &LARCH_opts.ase_lp64, \
> ++ &LARCH_opts.ase_gabs
> ++
> ++/* For LoongArch32/64 cmode=normal. */
> ++#define INSN_LA_TLS_LE \
> ++ "lu12i.w %1,%%le_hi20(%2);" \
> ++ "ori %1,%1,%%le_lo12(%2);", \
> ++ &LARCH_opts.ase_ilp32, 0
> ++
> ++/* For LoongArch64 cmode=large. */
> ++#define INSN_LA_TLS_LE64_LARGE \
> ++ "lu12i.w %1,%%le_hi20(%2);" \
> ++ "ori %1,%1,%%le_lo12(%2);" \
> ++ "lu32i.d %1,%%le64_lo20(%2);" \
> ++ "lu52i.d %1,%1,%%le64_hi12(%2);", \
> ++ &LARCH_opts.ase_lp64, 0
> ++
> ++#define INSN_LA_TLS_IE32 \
> ++ "pcalau12i %1,%%ie_pc_hi20(%2);" \
> ++ "ld.w %1,%1,%%ie_pc_lo12(%2);", \
> ++ &LARCH_opts.ase_ilp32, \
> ++ &LARCH_opts.ase_lp64
> ++/* For ie32 abs. */
> ++#define INSN_LA_TLS_IE32_ABS \
> ++ "lu12i.w %1,%%ie_hi20(%2);" \
> ++ "ori %1,%1,%%ie_lo12(%2);" \
> ++ "ld.w %1,%1,0", \
> ++ &LARCH_opts.ase_gabs, \
> ++ &LARCH_opts.ase_lp64
> ++#define INSN_LA_TLS_IE64 \
> ++ "pcalau12i %1,%%ie_pc_hi20(%2);" \
> ++ "ld.d %1,%1,%%ie_pc_lo12(%2);", \
> ++ &LARCH_opts.ase_lp64, 0
> ++/* For ie64 pic. */
> ++#define INSN_LA_TLS_IE64_LARGE_PCREL \
> ++ "pcalau12i %1,%%ie_pc_hi20(%3);" \
> ++ "addi.d %2,$r0,%%ie_pc_lo12(%3);" \
> ++ "lu32i.d %2,%%ie64_pc_lo20(%3);" \
> ++ "lu52i.d %2,%2,%%ie64_pc_hi12(%3);"\
> ++ "ldx.d %1,%1,%2;", \
> ++ &LARCH_opts.ase_lp64, \
> ++ &LARCH_opts.ase_gabs
> ++/* For ie64 abs. */
> ++#define INSN_LA_TLS_IE64_LARGE_ABS \
> ++ "lu12i.w %1,%%ie_hi20(%2);" \
> ++ "ori %1,%1,%%ie_lo12(%2);" \
> ++ "lu32i.d %1,%%ie64_lo20(%2);" \
> ++ "lu52i.d %1,%1,%%ie64_hi12(%2);" \
> ++ "ld.d %1,%1,0", \
> ++ &LARCH_opts.ase_lp64, \
> ++ &LARCH_opts.ase_gpcr
> ++
> ++/* For LoongArch32/64 cmode=normal. */
> ++#define INSN_LA_TLS_LD32 \
> ++ "pcalau12i %1,%%ld_pc_hi20(%2);" \
> ++ "addi.w %1,%1,%%got_pc_lo12(%2);", \
> ++ &LARCH_opts.ase_ilp32, \
> ++ &LARCH_opts.ase_lp64
> ++#define INSN_LA_TLS_LD32_ABS \
> ++ "lu12i.w %1,%%ld_hi20(%2);" \
> ++ "ori %1,%1,%%got_lo12(%2);", \
> ++ &LARCH_opts.ase_gabs, \
> ++ &LARCH_opts.ase_lp64
> ++#define INSN_LA_TLS_LD64 \
> ++ "pcalau12i %1,%%ld_pc_hi20(%2);" \
> ++ "addi.d %1,%1,%%got_pc_lo12(%2);", \
> ++ &LARCH_opts.ase_lp64, 0
> ++#define INSN_LA_TLS_LD64_LARGE_PCREL \
> ++ "pcalau12i %1,%%ld_pc_hi20(%3);" \
> ++ "addi.d %2,$r0,%%got_pc_lo12(%3);" \
> ++ "lu32i.d %2,%%got64_pc_lo20(%3);" \
> ++ "lu52i.d %2,%2,%%got64_pc_hi12(%3);"\
> ++ "add.d %1,%1,%2;", \
> ++ &LARCH_opts.ase_lp64, \
> ++ &LARCH_opts.ase_gabs
> ++#define INSN_LA_TLS_LD64_LARGE_ABS \
> ++ "lu12i.w %1,%%ld_hi20(%2);" \
> ++ "ori %1,%1,%%got_lo12(%2);" \
> ++ "lu32i.d %1,%%got64_lo20(%2);" \
> ++ "lu52i.d %1,%1,%%got64_hi12(%2);", \
> ++ &LARCH_opts.ase_lp64, \
> ++ &LARCH_opts.ase_gpcr
> ++
> ++#define INSN_LA_TLS_GD32 \
> ++ "pcalau12i %1,%%gd_pc_hi20(%2);" \
> ++ "addi.w %1,%1,%%got_pc_lo12(%2);", \
> ++ &LARCH_opts.ase_ilp32, \
> ++ &LARCH_opts.ase_lp64
> ++#define INSN_LA_TLS_GD32_ABS \
> ++ "lu12i.w %1,%%gd_hi20(%2);" \
> ++ "ori %1,%1,%%got_lo12(%2);", \
> ++ &LARCH_opts.ase_gabs, \
> ++ &LARCH_opts.ase_lp64
> ++#define INSN_LA_TLS_GD64 \
> ++ "pcalau12i %1,%%gd_pc_hi20(%2);" \
> ++ "addi.d %1,%1,%%got_pc_lo12(%2);", \
> ++ &LARCH_opts.ase_lp64, 0
> ++#define INSN_LA_TLS_GD64_LARGE_PCREL \
> ++ "pcalau12i %1,%%gd_pc_hi20(%3);" \
> ++ "addi.d %2,$r0,%%got_pc_lo12(%3);" \
> ++ "lu32i.d %2,%%got64_pc_lo20(%3);" \
> ++ "lu52i.d %2,%2,%%got64_pc_hi12(%3);"\
> ++ "add.d %1,%1,%2;", \
> ++ &LARCH_opts.ase_lp64, \
> ++ &LARCH_opts.ase_gabs
> ++#define INSN_LA_TLS_GD64_LARGE_ABS \
> ++ "lu12i.w %1,%%gd_hi20(%2);" \
> ++ "ori %1,%1,%%got_lo12(%2);" \
> ++ "lu32i.d %1,%%got64_lo20(%2);" \
> ++ "lu52i.d %1,%1,%%got64_hi12(%2);", \
> ++ &LARCH_opts.ase_lp64, \
> ++ &LARCH_opts.ase_gpcr
> ++
> ++
> ++static struct loongarch_opcode loongarch_macro_opcodes[] =
> ++{
> ++ /* match, mask, name, format, macro, include, exclude, pinfo.
> */
> ++ { 0, 0, "li.w", "r,sc", "%f", 0, 0, 0 },
> ++ { 0, 0, "li.d", "r,sc", "%f", 0, 0, 0 },
> ++
> ++ { 0, 0, "la", "r,la", "la.global %1,%2", 0,
> 0, 0 },
> ++ { 0, 0, "la.global", "r,la", "la.pcrel %1,%2",
> &LARCH_opts.ase_gpcr, 0, 0 },
> ++ { 0, 0, "la.global", "r,r,la", "la.pcrel %1,%2,%3",
> &LARCH_opts.ase_gpcr, 0, 0 },
> ++ { 0, 0, "la.global", "r,la", "la.abs %1,%2",
> &LARCH_opts.ase_gabs, 0, 0 },
> ++ { 0, 0, "la.global", "r,r,la", "la.abs %1,%3",
> &LARCH_opts.ase_gabs, 0, 0 },
> ++ { 0, 0, "la.global", "r,la", "la.got %1,%2", 0,
> 0, 0 },
> ++ { 0, 0, "la.global", "r,r,la", "la.got %1,%2,%3",
> &LARCH_opts.ase_lp64, 0, 0 },
> ++
> ++ { 0, 0, "la.local", "r,la", "la.abs %1,%2",
> &LARCH_opts.ase_labs, 0, 0 },
> ++ { 0, 0, "la.local", "r,r,la", "la.abs %1,%3",
> &LARCH_opts.ase_labs, 0, 0 },
> ++ { 0, 0, "la.local", "r,la", "la.pcrel %1,%2", 0,
> 0, 0 },
> ++ { 0, 0, "la.local", "r,r,la", "la.pcrel %1,%2,%3",
> &LARCH_opts.ase_lp64, 0, 0 },
> ++
> ++ { 0, 0, "la.abs", "r,la", INSN_LA_ABS32, 0 },
> ++ { 0, 0, "la.abs", "r,la", INSN_LA_ABS64, 0 },
> ++ { 0, 0, "la.pcrel", "r,la", INSN_LA_PCREL32, 0 },
> ++ { 0, 0, "la.pcrel", "r,la", INSN_LA_PCREL64, 0 },
> ++ { 0, 0, "la.pcrel", "r,r,la", INSN_LA_PCREL64_LARGE, 0 },
> ++ { 0, 0, "la.got", "r,la", INSN_LA_GOT32, 0 },
> ++ { 0, 0, "la.got", "r,la", INSN_LA_GOT32_ABS, 0 },
> ++ { 0, 0, "la.got", "r,la", INSN_LA_GOT64, 0 },
> ++ { 0, 0, "la.got", "r,la", INSN_LA_GOT64_LARGE_ABS, 0 },
> ++ { 0, 0, "la.got", "r,r,la", INSN_LA_GOT64_LARGE_PCREL, 0 },
> ++ { 0, 0, "la.tls.le", "r,l", INSN_LA_TLS_LE, 0
> },
> ++ { 0, 0, "la.tls.le", "r,l", INSN_LA_TLS_LE64_LARGE, 0
> },
> ++ { 0, 0, "la.tls.ie", "r,l", INSN_LA_TLS_IE32, 0
> },
> ++ { 0, 0, "la.tls.ie", "r,l", INSN_LA_TLS_IE32_ABS, 0
> },
> ++ { 0, 0, "la.tls.ie", "r,l", INSN_LA_TLS_IE64, 0
> },
> ++ { 0, 0, "la.tls.ie", "r,l", INSN_LA_TLS_IE64_LARGE_ABS, 0
> },
> ++ { 0, 0, "la.tls.ie", "r,r,l", INSN_LA_TLS_IE64_LARGE_PCREL, 0
> },
> ++ { 0, 0, "la.tls.ld", "r,l", INSN_LA_TLS_LD32, 0
> },
> ++ { 0, 0, "la.tls.ld", "r,l", INSN_LA_TLS_LD32_ABS, 0
> },
> ++ { 0, 0, "la.tls.ld", "r,l", INSN_LA_TLS_LD64, 0
> },
> ++ { 0, 0, "la.tls.ld", "r,l", INSN_LA_TLS_LD64_LARGE_ABS, 0
> },
> ++ { 0, 0, "la.tls.ld", "r,r,l", INSN_LA_TLS_LD64_LARGE_PCREL, 0
> },
> ++ { 0, 0, "la.tls.gd", "r,l", INSN_LA_TLS_GD32, 0
> },
> ++ { 0, 0, "la.tls.gd", "r,l", INSN_LA_TLS_GD32_ABS, 0
> },
> ++ { 0, 0, "la.tls.gd", "r,l", INSN_LA_TLS_GD64, 0
> },
> ++ { 0, 0, "la.tls.gd", "r,l", INSN_LA_TLS_GD64_LARGE_ABS, 0
> },
> ++ { 0, 0, "la.tls.gd", "r,r,l", INSN_LA_TLS_GD64_LARGE_PCREL, 0
> },
> ++
> ++ { 0 } /* Terminate the list. */
> ++};
> ++
> ++static struct loongarch_opcode loongarch_fix_opcodes[] =
> ++{
> ++ /* match, mask, name, format,
> macro, include, exclude, pinfo. */
> ++ { 0x00001000, 0xfffffc00, "clo.w", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00001400, 0xfffffc00, "clz.w", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00001800, 0xfffffc00, "cto.w", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00001c00, 0xfffffc00, "ctz.w", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00002000, 0xfffffc00, "clo.d", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00002400, 0xfffffc00, "clz.d", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00002800, 0xfffffc00, "cto.d", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00002c00, 0xfffffc00, "ctz.d", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00003000, 0xfffffc00, "revb.2h", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00003400, 0xfffffc00, "revb.4h", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00003800, 0xfffffc00, "revb.2w", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00003c00, 0xfffffc00, "revb.d", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00004000, 0xfffffc00, "revh.2w", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00004400, 0xfffffc00, "revh.d", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00004800, 0xfffffc00, "bitrev.4b", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00004c00, 0xfffffc00, "bitrev.8b", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00005000, 0xfffffc00, "bitrev.w", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00005400, 0xfffffc00, "bitrev.d", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00005800, 0xfffffc00, "ext.w.h", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00005c00, 0xfffffc00, "ext.w.b", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ /* or %1,%2,$r0 */
> ++ { 0x00150000, 0xfffffc00, "move", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00006000, 0xfffffc00, "rdtimel.w", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00006400, 0xfffffc00, "rdtimeh.w", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00006800, 0xfffffc00, "rdtime.d", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00006c00, 0xfffffc00, "cpucfg", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x00010000, 0xffff801f, "asrtle.d", "r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00018000, 0xffff801f, "asrtgt.d", "r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00040000, 0xfffe0000, "alsl.w",
> "r0:5,r5:5,r10:5,u15:2+1", 0, 0, 0, 0
> },
> ++ { 0x00060000, 0xfffe0000, "alsl.wu",
> "r0:5,r5:5,r10:5,u15:2+1", 0, 0, 0, 0 },
> ++ { 0x00080000, 0xfffe0000, "bytepick.w", "r0:5,r5:5,r10:5,u15:2",
> 0, 0, 0, 0 },
> ++ { 0x000c0000, 0xfffc0000, "bytepick.d", "r0:5,r5:5,r10:5,u15:3",
> 0, 0, 0, 0 },
> ++ { 0x00100000, 0xffff8000, "add.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00108000, 0xffff8000, "add.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00110000, 0xffff8000, "sub.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00118000, 0xffff8000, "sub.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00120000, 0xffff8000, "slt", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00128000, 0xffff8000, "sltu", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00130000, 0xffff8000, "maskeqz", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00138000, 0xffff8000, "masknez", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00140000, 0xffff8000, "nor", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00148000, 0xffff8000, "and", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00150000, 0xffff8000, "or", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00158000, 0xffff8000, "xor", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00160000, 0xffff8000, "orn", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00168000, 0xffff8000, "andn", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00170000, 0xffff8000, "sll.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00178000, 0xffff8000, "srl.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00180000, 0xffff8000, "sra.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00188000, 0xffff8000, "sll.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00190000, 0xffff8000, "srl.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00198000, 0xffff8000, "sra.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x001b0000, 0xffff8000, "rotr.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x001b8000, 0xffff8000, "rotr.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x001c0000, 0xffff8000, "mul.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x001c8000, 0xffff8000, "mulh.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x001d0000, 0xffff8000, "mulh.wu", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x001d8000, 0xffff8000, "mul.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x001e0000, 0xffff8000, "mulh.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x001e8000, 0xffff8000, "mulh.du", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x001f0000, 0xffff8000, "mulw.d.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x001f8000, 0xffff8000, "mulw.d.wu", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00200000, 0xffff8000, "div.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00208000, 0xffff8000, "mod.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00210000, 0xffff8000, "div.wu", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00218000, 0xffff8000, "mod.wu", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00220000, 0xffff8000, "div.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00228000, 0xffff8000, "mod.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00230000, 0xffff8000, "div.du", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00238000, 0xffff8000, "mod.du", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00240000, 0xffff8000, "crc.w.b.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00248000, 0xffff8000, "crc.w.h.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00250000, 0xffff8000, "crc.w.w.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00258000, 0xffff8000, "crc.w.d.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00260000, 0xffff8000, "crcc.w.b.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00268000, 0xffff8000, "crcc.w.h.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00270000, 0xffff8000, "crcc.w.w.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x00278000, 0xffff8000, "crcc.w.d.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x002a0000, 0xffff8000, "break", "u0:15",
> 0, 0, 0, 0 },
> ++ { 0x002a8000, 0xffff8000, "dbcl", "u0:15",
> 0, 0, 0, 0 },
> ++ { 0x002b0000, 0xffff8000, "syscall", "u0:15",
> 0, 0, 0, 0 },
> ++ { 0x002c0000, 0xfffe0000, "alsl.d",
> "r0:5,r5:5,r10:5,u15:2+1", 0, 0, 0, 0
> },
> ++ { 0x00408000, 0xffff8000, "slli.w", "r0:5,r5:5,u10:5",
> 0, 0, 0, 0 },
> ++ { 0x00410000, 0xffff0000, "slli.d", "r0:5,r5:5,u10:6",
> 0, 0, 0, 0 },
> ++ { 0x00448000, 0xffff8000, "srli.w", "r0:5,r5:5,u10:5",
> 0, 0, 0, 0 },
> ++ { 0x00450000, 0xffff0000, "srli.d", "r0:5,r5:5,u10:6",
> 0, 0, 0, 0 },
> ++ { 0x00488000, 0xffff8000, "srai.w", "r0:5,r5:5,u10:5",
> 0, 0, 0, 0 },
> ++ { 0x00490000, 0xffff0000, "srai.d", "r0:5,r5:5,u10:6",
> 0, 0, 0, 0 },
> ++ { 0x004c8000, 0xffff8000, "rotri.w", "r0:5,r5:5,u10:5",
> 0, 0, 0, 0 },
> ++ { 0x004d0000, 0xffff0000, "rotri.d", "r0:5,r5:5,u10:6",
> 0, 0, 0, 0 },
> ++ { 0x00600000, 0xffe08000, "bstrins.w", "r0:5,r5:5,u16:5,u10:5",
> 0, 0, 0, 0 },
> ++ { 0x00608000, 0xffe08000, "bstrpick.w", "r0:5,r5:5,u16:5,u10:5",
> 0, 0, 0, 0 },
> ++ { 0x00800000, 0xffc00000, "bstrins.d", "r0:5,r5:5,u16:6,u10:6",
> 0, 0, 0, 0 },
> ++ { 0x00c00000, 0xffc00000, "bstrpick.d", "r0:5,r5:5,u16:6,u10:6",
> 0, 0, 0, 0 },
> ++ { 0 } /* Terminate the list. */
> ++};
> ++
> ++static struct loongarch_opcode loongarch_single_float_opcodes[] =
> ++{
> ++ /* match, mask, name, format,
> macro, include, exclude, pinfo. */
> ++ { 0x01008000, 0xffff8000, "fadd.s", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01028000, 0xffff8000, "fsub.s", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01048000, 0xffff8000, "fmul.s", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01068000, 0xffff8000, "fdiv.s", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01088000, 0xffff8000, "fmax.s", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x010a8000, 0xffff8000, "fmin.s", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x010c8000, 0xffff8000, "fmaxa.s", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x010e8000, 0xffff8000, "fmina.s", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01108000, 0xffff8000, "fscaleb.s", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01128000, 0xffff8000, "fcopysign.s", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01140400, 0xfffffc00, "fabs.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01141400, 0xfffffc00, "fneg.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01142400, 0xfffffc00, "flogb.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01143400, 0xfffffc00, "fclass.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01144400, 0xfffffc00, "fsqrt.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01145400, 0xfffffc00, "frecip.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01146400, 0xfffffc00, "frsqrt.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01149400, 0xfffffc00, "fmov.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0114a400, 0xfffffc00, "movgr2fr.w", "f0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0114ac00, 0xfffffc00, "movgr2frh.w", "f0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0114b400, 0xfffffc00, "movfr2gr.s", "r0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0114bc00, 0xfffffc00, "movfrh2gr.s", "r0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0114c000, 0xfffffc00, "movgr2fcsr", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0114c800, 0xfffffc00, "movfcsr2gr", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0114d000, 0xfffffc18, "movfr2cf", "c0:3,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0114d400, 0xffffff00, "movcf2fr", "f0:5,c5:3",
> 0, 0, 0, 0 },
> ++ { 0x0114d800, 0xfffffc18, "movgr2cf", "c0:3,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0114dc00, 0xffffff00, "movcf2gr", "r0:5,c5:3",
> 0, 0, 0, 0 },
> ++ { 0x011a0400, 0xfffffc00, "ftintrm.w.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011a2400, 0xfffffc00, "ftintrm.l.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011a4400, 0xfffffc00, "ftintrp.w.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011a6400, 0xfffffc00, "ftintrp.l.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011a8400, 0xfffffc00, "ftintrz.w.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011aa400, 0xfffffc00, "ftintrz.l.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011ac400, 0xfffffc00, "ftintrne.w.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011ae400, 0xfffffc00, "ftintrne.l.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011b0400, 0xfffffc00, "ftint.w.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011b2400, 0xfffffc00, "ftint.l.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011d1000, 0xfffffc00, "ffint.s.w", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011d1800, 0xfffffc00, "ffint.s.l", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011e4400, 0xfffffc00, "frint.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0 } /* Terminate the list. */
> ++};
> ++static struct loongarch_opcode loongarch_double_float_opcodes[] =
> ++{
> ++ /* match, mask, name, format,
> macro, include, exclude, pinfo. */
> ++ { 0x01010000, 0xffff8000, "fadd.d", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01030000, 0xffff8000, "fsub.d", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01050000, 0xffff8000, "fmul.d", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01070000, 0xffff8000, "fdiv.d", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01090000, 0xffff8000, "fmax.d", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x010b0000, 0xffff8000, "fmin.d", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x010d0000, 0xffff8000, "fmaxa.d", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x010f0000, 0xffff8000, "fmina.d", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01110000, 0xffff8000, "fscaleb.d", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01130000, 0xffff8000, "fcopysign.d", "f0:5,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x01140800, 0xfffffc00, "fabs.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01141800, 0xfffffc00, "fneg.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01142800, 0xfffffc00, "flogb.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01143800, 0xfffffc00, "fclass.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01144800, 0xfffffc00, "fsqrt.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01145800, 0xfffffc00, "frecip.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01146800, 0xfffffc00, "frsqrt.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01149800, 0xfffffc00, "fmov.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0114a800, 0xfffffc00, "movgr2fr.d", "f0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0114b800, 0xfffffc00, "movfr2gr.d", "r0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01191800, 0xfffffc00, "fcvt.s.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x01192400, 0xfffffc00, "fcvt.d.s", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011a0800, 0xfffffc00, "ftintrm.w.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011a2800, 0xfffffc00, "ftintrm.l.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011a4800, 0xfffffc00, "ftintrp.w.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011a6800, 0xfffffc00, "ftintrp.l.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011a8800, 0xfffffc00, "ftintrz.w.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011aa800, 0xfffffc00, "ftintrz.l.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011ac800, 0xfffffc00, "ftintrne.w.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011ae800, 0xfffffc00, "ftintrne.l.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011b0800, 0xfffffc00, "ftint.w.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011b2800, 0xfffffc00, "ftint.l.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011d2000, 0xfffffc00, "ffint.d.w", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011d2800, 0xfffffc00, "ffint.d.l", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x011e4800, 0xfffffc00, "frint.d", "f0:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0 } /* Terminate the list. */
> ++};
> ++
> ++static struct loongarch_opcode loongarch_imm_opcodes[] =
> ++{
> ++ /* match, mask, name, format,
> macro, include, exclude, pinfo. */
> ++ { 0x02000000, 0xffc00000, "slti", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x02400000, 0xffc00000, "sltui", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x02800000, 0xffc00000, "addi.w", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x02c00000, 0xffc00000, "addi.d", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x03000000, 0xffc00000, "lu52i.d", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "nop", "",
> "andi $r0,$r0,0", 0, 0, 0 },
> ++ { 0x03400000, 0xffc00000, "andi", "r0:5,r5:5,u10:12",
> 0, 0, 0, 0 },
> ++ { 0x03800000, 0xffc00000, "ori", "r0:5,r5:5,u10:12",
> 0, 0, 0, 0 },
> ++ { 0x03c00000, 0xffc00000, "xori", "r0:5,r5:5,u10:12",
> 0, 0, 0, 0 },
> ++ { 0x10000000, 0xfc000000, "addu16i.d", "r0:5,r5:5,s10:16",
> 0, 0, 0, 0 },
> ++ { 0x14000000, 0xfe000000, "lu12i.w", "r0:5,s5:20",
> 0, 0, 0, 0 },
> ++ { 0x16000000, 0xfe000000, "lu32i.d", "r0:5,s5:20",
> 0, 0, 0, 0 },
> ++ { 0x18000000, 0xfe000000, "pcaddi", "r0:5,s5:20",
> 0, 0, 0, 0 },
> ++ { 0x1a000000, 0xfe000000, "pcalau12i", "r0:5,s5:20",
> 0, 0, 0, 0 },
> ++ { 0x1c000000, 0xfe000000, "pcaddu12i", "r0:5,s5:20",
> 0, 0, 0, 0 },
> ++ { 0x1e000000, 0xfe000000, "pcaddu18i", "r0:5,s5:20",
> 0, 0, 0, 0 },
> ++ { 0 } /* Terminate the list. */
> ++};
> ++
> ++static struct loongarch_opcode loongarch_privilege_opcodes[] =
> ++{
> ++ /* match, mask, name, format,
> macro, include, exclude, pinfo. */
> ++ { 0x04000000, 0xff0003e0, "csrrd", "r0:5,u10:14",
> 0, 0, 0, 0 },
> ++ { 0x04000020, 0xff0003e0, "csrwr", "r0:5,u10:14",
> 0, 0, 0, 0 },
> ++ { 0x04000000, 0xff000000, "csrxchg", "r0:5,r5:5,u10:14",
> 0, 0, 0, 0 },
> ++ { 0x06000000, 0xffc00000, "cacop", "u0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x06400000, 0xfffc0000, "lddir", "r0:5,r5:5,u10:8",
> 0, 0, 0, 0 },
> ++ { 0x06440000, 0xfffc001f, "ldpte", "r5:5,u10:8",
> 0, 0, 0, 0 },
> ++ { 0x06480000, 0xfffffc00, "iocsrrd.b", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x06480400, 0xfffffc00, "iocsrrd.h", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x06480800, 0xfffffc00, "iocsrrd.w", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x06480c00, 0xfffffc00, "iocsrrd.d", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x06481000, 0xfffffc00, "iocsrwr.b", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x06481400, 0xfffffc00, "iocsrwr.h", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x06481800, 0xfffffc00, "iocsrwr.w", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x06481c00, 0xfffffc00, "iocsrwr.d", "r0:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x06482000, 0xffffffff, "tlbclr", "",
> 0, 0, 0, 0 },
> ++ { 0x06482400, 0xffffffff, "tlbflush", "",
> 0, 0, 0, 0 },
> ++ { 0x06482800, 0xffffffff, "tlbsrch", "",
> 0, 0, 0, 0 },
> ++ { 0x06482c00, 0xffffffff, "tlbrd", "",
> 0, 0, 0, 0 },
> ++ { 0x06483000, 0xffffffff, "tlbwr", "",
> 0, 0, 0, 0 },
> ++ { 0x06483400, 0xffffffff, "tlbfill", "",
> 0, 0, 0, 0 },
> ++ { 0x06483800, 0xffffffff, "ertn", "",
> 0, 0, 0, 0 },
> ++ { 0x06488000, 0xffff8000, "idle", "u0:15",
> 0, 0, 0, 0 },
> ++ { 0x06498000, 0xffff8000, "invtlb", "u0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0 } /* Terminate the list. */
> ++};
> ++
> ++static struct loongarch_opcode loongarch_4opt_single_float_opcodes[] =
> ++{
> ++ /* match, mask, name, format,
> macro, include, exclude, pinfo. */
> ++ { 0x08100000, 0xfff00000, "fmadd.s", "f0:5,f5:5,f10:5,f15:5",
> 0, 0, 0, 0 },
> ++ { 0x08500000, 0xfff00000, "fmsub.s", "f0:5,f5:5,f10:5,f15:5",
> 0, 0, 0, 0 },
> ++ { 0x08900000, 0xfff00000, "fnmadd.s", "f0:5,f5:5,f10:5,f15:5",
> 0, 0, 0, 0 },
> ++ { 0x08d00000, 0xfff00000, "fnmsub.s", "f0:5,f5:5,f10:5,f15:5",
> 0, 0, 0, 0 },
> ++ { 0x0c100000, 0xffff8018, "fcmp.caf.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c108000, 0xffff8018, "fcmp.saf.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c110000, 0xffff8018, "fcmp.clt.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c118000, 0xffff8018, "fcmp.slt.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c118000, 0xffff8018, "fcmp.sgt.s", "c0:3,f10:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0c120000, 0xffff8018, "fcmp.ceq.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c128000, 0xffff8018, "fcmp.seq.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c130000, 0xffff8018, "fcmp.cle.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c138000, 0xffff8018, "fcmp.sle.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c138000, 0xffff8018, "fcmp.sge.s", "c0:3,f10:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0c140000, 0xffff8018, "fcmp.cun.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c148000, 0xffff8018, "fcmp.sun.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c150000, 0xffff8018, "fcmp.cult.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c150000, 0xffff8018, "fcmp.cugt.s", "c0:3,f10:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0c158000, 0xffff8018, "fcmp.sult.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c160000, 0xffff8018, "fcmp.cueq.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c168000, 0xffff8018, "fcmp.sueq.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c170000, 0xffff8018, "fcmp.cule.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c170000, 0xffff8018, "fcmp.cuge.s", "c0:3,f10:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0c178000, 0xffff8018, "fcmp.sule.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c180000, 0xffff8018, "fcmp.cne.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c188000, 0xffff8018, "fcmp.sne.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c1a0000, 0xffff8018, "fcmp.cor.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c1a8000, 0xffff8018, "fcmp.sor.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c1c0000, 0xffff8018, "fcmp.cune.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c1c8000, 0xffff8018, "fcmp.sune.s", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0d000000, 0xfffc0000, "fsel", "f0:5,f5:5,f10:5,c15:3",
> 0, 0, 0, 0 },
> ++ { 0 } /* Terminate the list. */
> ++};
> ++
> ++static struct loongarch_opcode loongarch_4opt_double_float_opcodes[] =
> ++{
> ++ /* match, mask, name, format,
> macro, include, exclude, pinfo. */
> ++ { 0x08200000, 0xfff00000, "fmadd.d", "f0:5,f5:5,f10:5,f15:5",
> 0, 0, 0, 0 },
> ++ { 0x08600000, 0xfff00000, "fmsub.d", "f0:5,f5:5,f10:5,f15:5",
> 0, 0, 0, 0 },
> ++ { 0x08a00000, 0xfff00000, "fnmadd.d", "f0:5,f5:5,f10:5,f15:5",
> 0, 0, 0, 0 },
> ++ { 0x08e00000, 0xfff00000, "fnmsub.d", "f0:5,f5:5,f10:5,f15:5",
> 0, 0, 0, 0 },
> ++ { 0x0c200000, 0xffff8018, "fcmp.caf.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c208000, 0xffff8018, "fcmp.saf.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c210000, 0xffff8018, "fcmp.clt.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c218000, 0xffff8018, "fcmp.slt.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c218000, 0xffff8018, "fcmp.sgt.d", "c0:3,f10:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0c220000, 0xffff8018, "fcmp.ceq.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c228000, 0xffff8018, "fcmp.seq.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c230000, 0xffff8018, "fcmp.cle.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c238000, 0xffff8018, "fcmp.sle.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c238000, 0xffff8018, "fcmp.sge.d", "c0:3,f10:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0c240000, 0xffff8018, "fcmp.cun.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c248000, 0xffff8018, "fcmp.sun.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c250000, 0xffff8018, "fcmp.cult.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c250000, 0xffff8018, "fcmp.cugt.d", "c0:3,f10:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0c258000, 0xffff8018, "fcmp.sult.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c260000, 0xffff8018, "fcmp.cueq.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c268000, 0xffff8018, "fcmp.sueq.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c270000, 0xffff8018, "fcmp.cule.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c270000, 0xffff8018, "fcmp.cuge.d", "c0:3,f10:5,f5:5",
> 0, 0, 0, 0 },
> ++ { 0x0c278000, 0xffff8018, "fcmp.sule.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c280000, 0xffff8018, "fcmp.cne.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c288000, 0xffff8018, "fcmp.sne.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c2a0000, 0xffff8018, "fcmp.cor.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c2a8000, 0xffff8018, "fcmp.sor.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c2c0000, 0xffff8018, "fcmp.cune.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0x0c2c8000, 0xffff8018, "fcmp.sune.d", "c0:3,f5:5,f10:5",
> 0, 0, 0, 0 },
> ++ { 0 } /* Terminate the list. */
> ++};
> ++
> ++static struct loongarch_opcode loongarch_load_store_opcodes[] =
> ++{
> ++ /* match, mask, name, format,
> macro, include, exclude, pinfo. */
> ++ { 0x20000000, 0xff000000, "ll.w", "r0:5,r5:5,s10:14<<2",
> 0, 0, 0, 0 },
> ++ { 0x21000000, 0xff000000, "sc.w", "r0:5,r5:5,s10:14<<2",
> 0, 0, 0, 0 },
> ++ { 0x22000000, 0xff000000, "ll.d", "r0:5,r5:5,s10:14<<2",
> 0, 0, 0, 0 },
> ++ { 0x23000000, 0xff000000, "sc.d", "r0:5,r5:5,s10:14<<2",
> 0, 0, 0, 0 },
> ++ { 0x24000000, 0xff000000, "ldptr.w", "r0:5,r5:5,s10:14<<2",
> 0, 0, 0, 0 },
> ++ { 0x25000000, 0xff000000, "stptr.w", "r0:5,r5:5,s10:14<<2",
> 0, 0, 0, 0 },
> ++ { 0x26000000, 0xff000000, "ldptr.d", "r0:5,r5:5,s10:14<<2",
> 0, 0, 0, 0 },
> ++ { 0x27000000, 0xff000000, "stptr.d", "r0:5,r5:5,s10:14<<2",
> 0, 0, 0, 0 },
> ++ { 0x28000000, 0xffc00000, "ld.b", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x28400000, 0xffc00000, "ld.h", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x28800000, 0xffc00000, "ld.w", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x28c00000, 0xffc00000, "ld.d", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x29000000, 0xffc00000, "st.b", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x29400000, 0xffc00000, "st.h", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x29800000, 0xffc00000, "st.w", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x29c00000, 0xffc00000, "st.d", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x2a000000, 0xffc00000, "ld.bu", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x2a400000, 0xffc00000, "ld.hu", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x2a800000, 0xffc00000, "ld.wu", "r0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x2ac00000, 0xffc00000, "preld", "u0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x38000000, 0xffff8000, "ldx.b", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x38040000, 0xffff8000, "ldx.h", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x38080000, 0xffff8000, "ldx.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x380c0000, 0xffff8000, "ldx.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x38100000, 0xffff8000, "stx.b", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x38140000, 0xffff8000, "stx.h", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x38180000, 0xffff8000, "stx.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x381c0000, 0xffff8000, "stx.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x38200000, 0xffff8000, "ldx.bu", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x38240000, 0xffff8000, "ldx.hu", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x38280000, 0xffff8000, "ldx.wu", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x382c0000, 0xffff8000, "preldx", "u0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amswap.w", "r,r,r,u0:0",
> "amswap.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x38600000, 0xffff8000, "amswap.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amswap.d", "r,r,r,u0:0",
> "amswap.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x38608000, 0xffff8000, "amswap.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amadd.w", "r,r,r,u0:0",
> "amadd.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x38610000, 0xffff8000, "amadd.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amadd.d", "r,r,r,u0:0",
> "amadd.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x38618000, 0xffff8000, "amadd.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amand.w", "r,r,r,u0:0",
> "amand.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x38620000, 0xffff8000, "amand.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amand.d", "r,r,r,u0:0",
> "amand.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x38628000, 0xffff8000, "amand.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amor.w", "r,r,r,u0:0",
> "amor.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x38630000, 0xffff8000, "amor.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amor.d", "r,r,r,u0:0",
> "amor.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x38638000, 0xffff8000, "amor.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amxor.w", "r,r,r,u0:0",
> "amxor.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x38640000, 0xffff8000, "amxor.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amxor.d", "r,r,r,u0:0",
> "amxor.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x38648000, 0xffff8000, "amxor.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammax.w", "r,r,r,u0:0",
> "ammax.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x38650000, 0xffff8000, "ammax.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammax.d", "r,r,r,u0:0",
> "ammax.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x38658000, 0xffff8000, "ammax.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammin.w", "r,r,r,u0:0",
> "ammin.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x38660000, 0xffff8000, "ammin.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammin.d", "r,r,r,u0:0",
> "ammin.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x38668000, 0xffff8000, "ammin.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammax.wu", "r,r,r,u0:0",
> "ammax.wu %1,%2,%3", 0, 0, 0 },
> ++ { 0x38670000, 0xffff8000, "ammax.wu", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammax.du", "r,r,r,u0:0",
> "ammax.du %1,%2,%3", 0, 0, 0 },
> ++ { 0x38678000, 0xffff8000, "ammax.du", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammin.wu", "r,r,r,u0:0",
> "ammin.wu %1,%2,%3", 0, 0, 0 },
> ++ { 0x38680000, 0xffff8000, "ammin.wu", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammin.du", "r,r,r,u0:0",
> "ammin.du %1,%2,%3", 0, 0, 0 },
> ++ { 0x38688000, 0xffff8000, "ammin.du", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amswap_db.w", "r,r,r,u0:0",
> "amswap_db.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x38690000, 0xffff8000, "amswap_db.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amswap_db.d", "r,r,r,u0:0",
> "amswap_db.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x38698000, 0xffff8000, "amswap_db.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amadd_db.w", "r,r,r,u0:0",
> "amadd_db.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x386a0000, 0xffff8000, "amadd_db.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amadd_db.d", "r,r,r,u0:0",
> "amadd_db.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x386a8000, 0xffff8000, "amadd_db.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amand_db.w", "r,r,r,u0:0",
> "amand_db.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x386b0000, 0xffff8000, "amand_db.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amand_db.d", "r,r,r,u0:0",
> "amand_db.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x386b8000, 0xffff8000, "amand_db.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amor_db.w", "r,r,r,u0:0",
> "amor_db.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x386c0000, 0xffff8000, "amor_db.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amor_db.d", "r,r,r,u0:0",
> "amor_db.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x386c8000, 0xffff8000, "amor_db.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amxor_db.w", "r,r,r,u0:0",
> "amxor_db.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x386d0000, 0xffff8000, "amxor_db.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "amxor_db.d", "r,r,r,u0:0",
> "amxor_db.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x386d8000, 0xffff8000, "amxor_db.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammax_db.w", "r,r,r,u0:0",
> "ammax_db.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x386e0000, 0xffff8000, "ammax_db.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammax_db.d", "r,r,r,u0:0",
> "ammax_db.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x386e8000, 0xffff8000, "ammax_db.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammin_db.w", "r,r,r,u0:0",
> "ammin_db.w %1,%2,%3", 0, 0, 0 },
> ++ { 0x386f0000, 0xffff8000, "ammin_db.w", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammin_db.d", "r,r,r,u0:0",
> "ammin_db.d %1,%2,%3", 0, 0, 0 },
> ++ { 0x386f8000, 0xffff8000, "ammin_db.d", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammax_db.wu", "r,r,r,u0:0",
> "ammax_db.wu %1,%2,%3", 0, 0, 0 },
> ++ { 0x38700000, 0xffff8000, "ammax_db.wu", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammax_db.du", "r,r,r,u0:0",
> "ammax_db.du %1,%2,%3", 0, 0, 0 },
> ++ { 0x38708000, 0xffff8000, "ammax_db.du", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammin_db.wu", "r,r,r,u0:0",
> "ammin_db.wu %1,%2,%3", 0, 0, 0 },
> ++ { 0x38710000, 0xffff8000, "ammin_db.wu", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ammin_db.du", "r,r,r,u0:0",
> "ammin_db.du %1,%2,%3", 0, 0, 0 },
> ++ { 0x38718000, 0xffff8000, "ammin_db.du", "r0:5,r10:5,r5:5",
> 0, 0, 0, 0 },
> ++ { 0x38720000, 0xffff8000, "dbar", "u0:15",
> 0, 0, 0, 0 },
> ++ { 0x38728000, 0xffff8000, "ibar", "u0:15",
> 0, 0, 0, 0 },
> ++ { 0x38780000, 0xffff8000, "ldgt.b", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x38788000, 0xffff8000, "ldgt.h", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x38790000, 0xffff8000, "ldgt.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x38798000, 0xffff8000, "ldgt.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x387a0000, 0xffff8000, "ldle.b", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x387a8000, 0xffff8000, "ldle.h", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x387b0000, 0xffff8000, "ldle.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x387b8000, 0xffff8000, "ldle.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x387c0000, 0xffff8000, "stgt.b", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x387c8000, 0xffff8000, "stgt.h", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x387d0000, 0xffff8000, "stgt.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x387d8000, 0xffff8000, "stgt.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x387e0000, 0xffff8000, "stle.b", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x387e8000, 0xffff8000, "stle.h", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x387f0000, 0xffff8000, "stle.w", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0x387f8000, 0xffff8000, "stle.d", "r0:5,r5:5,r10:5",
> 0, 0, 0, 0 },
> ++ { 0 } /* Terminate the list. */
> ++};
> ++
> ++static struct loongarch_opcode
> loongarch_single_float_load_store_opcodes[] =
> ++{
> ++ /* match, mask, name, format,
> macro, include, exclude, pinfo. */
> ++ { 0x2b000000, 0xffc00000, "fld.s", "f0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x2b400000, 0xffc00000, "fst.s", "f0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x38300000, 0xffff8000, "fldx.s", "f0:5,r5:5,r10:5",
> 0, &LARCH_opts.ase_lp64, 0, 0 },
> ++ { 0x38380000, 0xffff8000, "fstx.s", "f0:5,r5:5,r10:5",
> 0, &LARCH_opts.ase_lp64, 0, 0 },
> ++ { 0x38740000, 0xffff8000, "fldgt.s", "f0:5,r5:5,r10:5",
> 0, &LARCH_opts.ase_lp64, 0, 0 },
> ++ { 0x38750000, 0xffff8000, "fldle.s", "f0:5,r5:5,r10:5",
> 0, &LARCH_opts.ase_lp64, 0, 0 },
> ++ { 0x38760000, 0xffff8000, "fstgt.s", "f0:5,r5:5,r10:5",
> 0, &LARCH_opts.ase_lp64, 0, 0 },
> ++ { 0x38770000, 0xffff8000, "fstle.s", "f0:5,r5:5,r10:5",
> 0, &LARCH_opts.ase_lp64, 0, 0 },
> ++ { 0 } /* Terminate the list. */
> ++};
> ++
> ++static struct loongarch_opcode
> loongarch_double_float_load_store_opcodes[] =
> ++{
> ++ /* match, mask, name, format,
> macro, include, exclude, pinfo. */
> ++ { 0x2b800000, 0xffc00000, "fld.d", "f0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x2bc00000, 0xffc00000, "fst.d", "f0:5,r5:5,s10:12",
> 0, 0, 0, 0 },
> ++ { 0x38340000, 0xffff8000, "fldx.d", "f0:5,r5:5,r10:5",
> 0, &LARCH_opts.ase_lp64, 0, 0 },
> ++ { 0x383c0000, 0xffff8000, "fstx.d", "f0:5,r5:5,r10:5",
> 0, &LARCH_opts.ase_lp64, 0, 0 },
> ++ { 0x38748000, 0xffff8000, "fldgt.d", "f0:5,r5:5,r10:5",
> 0, &LARCH_opts.ase_lp64, 0, 0 },
> ++ { 0x38758000, 0xffff8000, "fldle.d", "f0:5,r5:5,r10:5",
> 0, &LARCH_opts.ase_lp64, 0, 0 },
> ++ { 0x38768000, 0xffff8000, "fstgt.d", "f0:5,r5:5,r10:5",
> 0, &LARCH_opts.ase_lp64, 0, 0 },
> ++ { 0x38778000, 0xffff8000, "fstle.d", "f0:5,r5:5,r10:5",
> 0, &LARCH_opts.ase_lp64, 0, 0 },
> ++ { 0 } /* Terminate the list. */
> ++};
> ++
> ++static struct loongarch_opcode loongarch_float_jmp_opcodes[] =
> ++{
> ++ { 0x0, 0x0, "bceqz", "c,la",
> "bceqz %1,%%b21(%2)", 0, 0, 0 },
> ++ { 0x48000000, 0xfc000300, "bceqz", "c5:3,sb0:5|10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "bcnez", "c,la",
> "bcnez %1,%%b21(%2)", 0, 0, 0 },
> ++ { 0x48000100, 0xfc000300, "bcnez", "c5:3,sb0:5|10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0 } /* Terminate the list. */
> ++};
> ++
> ++static struct loongarch_opcode loongarch_jmp_opcodes[] =
> ++{
> ++ /* match, mask, name, format,
> macro, include, exclude, pinfo. */
> ++ { 0x0, 0x0, "bltz", "r,la",
> "bltz %1,%%b16(%2)", 0, 0, 0 },
> ++ { 0x60000000, 0xfc00001f, "bltz", "r5:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "bgtz", "r,la",
> "bgtz %1,%%b16(%2)", 0, 0, 0 },
> ++ { 0x60000000, 0xfc0003e0, "bgtz", "r0:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "bgez", "r,la",
> "bgez %1,%%b16(%2)", 0, 0, 0 },
> ++ { 0x64000000, 0xfc00001f, "bgez", "r5:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "blez", "r,la",
> "blez %1,%%b16(%2)", 0, 0, 0 },
> ++ { 0x64000000, 0xfc0003e0, "blez", "r0:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "beqz", "r,la",
> "beqz %1,%%b21(%2)", 0, 0, 0 },
> ++ { 0x40000000, 0xfc000000, "beqz", "r5:5,sb0:5|10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "bnez", "r,la",
> "bnez %1,%%b21(%2)", 0, 0, 0 },
> ++ { 0x44000000, 0xfc000000, "bnez", "r5:5,sb0:5|10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "jr", "r",
> "jirl $r0,%1,0", 0, 0, 0 },
> ++ { 0x50000000, 0xfc000000, "b", "sb0:10|10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "b", "la",
> "b %%b26(%1)", 0, 0, 0 },
> ++ { 0x4c000000, 0xfc000000, "jirl", "r0:5,r5:5,s10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "bl", "la",
> "bl %%b26(%1)", 0, 0, 0 },
> ++ { 0x54000000, 0xfc000000, "bl", "sb0:10|10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "beq", "r,r,la",
> "beq %1,%2,%%b16(%3)", 0, 0, 0 },
> ++ { 0x58000000, 0xfc000000, "beq", "r5:5,r0:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "bne", "r,r,la",
> "bne %1,%2,%%b16(%3)", 0, 0, 0 },
> ++ { 0x5c000000, 0xfc000000, "bne", "r5:5,r0:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "blt", "r,r,la",
> "blt %1,%2,%%b16(%3)", 0, 0, 0 },
> ++ { 0x60000000, 0xfc000000, "blt", "r5:5,r0:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "bgt", "r,r,la",
> "bgt %1,%2,%%b16(%3)", 0, 0, 0 },
> ++ { 0x60000000, 0xfc000000, "bgt", "r0:5,r5:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "bge", "r,r,la",
> "bge %1,%2,%%b16(%3)", 0, 0, 0 },
> ++ { 0x64000000, 0xfc000000, "bge", "r5:5,r0:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "ble", "r,r,la",
> "ble %1,%2,%%b16(%3)", 0, 0, 0 },
> ++ { 0x64000000, 0xfc000000, "ble", "r0:5,r5:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "bltu", "r,r,la",
> "bltu %1,%2,%%b16(%3)", 0, 0, 0 },
> ++ { 0x68000000, 0xfc000000, "bltu", "r5:5,r0:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "bgtu", "r,r,la",
> "bgtu %1,%2,%%b16(%3)", 0, 0, 0 },
> ++ { 0x68000000, 0xfc000000, "bgtu", "r0:5,r5:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "bgeu", "r,r,la",
> "bgeu %1,%2,%%b16(%3)", 0, 0, 0 },
> ++ { 0x6c000000, 0xfc000000, "bgeu", "r5:5,r0:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0x0, 0x0, "bleu", "r,r,la",
> "bleu %1,%2,%%b16(%3)", 0, 0, 0 },
> ++ { 0x6c000000, 0xfc000000, "bleu", "r0:5,r5:5,sb10:16<<2",
> 0, 0, 0, 0 },
> ++ { 0 } /* Terminate the list. */
> ++};
> ++
> ++struct loongarch_ase loongarch_ASEs[] =
> ++{
> ++ { &LARCH_opts.ase_ilp32, loongarch_macro_opcodes, 0, 0, { 0
> }, 0, 0 },
> ++ { &LARCH_opts.ase_ilp32, loongarch_imm_opcodes, 0, 0, { 0
> }, 0, 0 },
> ++ { &LARCH_opts.ase_ilp32, loongarch_privilege_opcodes, 0,
> 0, { 0 }, 0, 0 },
> ++ { &LARCH_opts.ase_ilp32, loongarch_load_store_opcodes, 0, 0, { 0
> }, 0, 0 },
> ++ { &LARCH_opts.ase_ilp32, loongarch_fix_opcodes, 0, 0, { 0
> }, 0, 0 },
> ++ { &LARCH_opts.ase_ilp32, loongarch_jmp_opcodes, 0, 0, { 0
> }, 0, 0 },
> ++ { &LARCH_opts.ase_sf, loongarch_float_jmp_opcodes, 0, 0, { 0
> }, 0, 0 },
> ++ { &LARCH_opts.ase_sf, loongarch_single_float_opcodes, 0,
> 0, { 0 }, 0, 0 },
> ++ { &LARCH_opts.ase_df, loongarch_double_float_opcodes, 0,
> 0, { 0 }, 0, 0 },
> ++ { &LARCH_opts.ase_sf, loongarch_4opt_single_float_opcodes, 0,
> 0, { 0 }, 0, 0 },
> ++ { &LARCH_opts.ase_df, loongarch_4opt_double_float_opcodes, 0,
> 0, { 0 }, 0, 0 },
> ++ { &LARCH_opts.ase_sf, loongarch_single_float_load_store_opcodes, 0,
> 0, { 0 }, 0, 0 },
> ++ { &LARCH_opts.ase_df, loongarch_double_float_load_store_opcodes, 0,
> 0, { 0 }, 0, 0 },
> ++ { 0 },
> ++};
> +diff --git gdb-10.2/opcodes/po/POTFILES.in gdb-10.2/opcodes/po/POTFILES.in
> +index 7572204..84625ac 100644
> +--- gdb-10.2/opcodes/po/POTFILES.in
> ++++ gdb-10.2/opcodes/po/POTFILES.in
> +@@ -108,6 +108,9 @@ lm32-ibld.c
> + lm32-opc.c
> + lm32-opc.h
> + lm32-opinst.c
> ++loongarch-coder.c
> ++loongarch-dis.c
> ++loongarch-opc.c
> + m10200-dis.c
> + m10200-opc.c
> + m10300-dis.c
> +--
> +2.33.0
> diff --git a/lkcd_vmdump_v1.h b/lkcd_vmdump_v1.h
> index 98ee094..3ffb219 100644
> --- a/lkcd_vmdump_v1.h
> +++ b/lkcd_vmdump_v1.h
> @@ -114,7 +114,7 @@ typedef struct _dump_header_s {
> struct new_utsname dh_utsname;
>
> /* the dump registers */
> -#if !defined(IA64) && !defined(S390) && !defined(S390X) &&
> !defined(ARM64) && !defined(RISCV64)
> +#if !defined(IA64) && !defined(S390) && !defined(S390X) &&
> !defined(ARM64) && !defined(RISCV64) && !defined(LOONGARCH64)
> struct pt_regs dh_regs;
> #endif
>
> diff --git a/lkcd_vmdump_v2_v3.h b/lkcd_vmdump_v2_v3.h
> index ef3067f..f215446 100644
> --- a/lkcd_vmdump_v2_v3.h
> +++ b/lkcd_vmdump_v2_v3.h
> @@ -37,7 +37,8 @@
>
> #if defined(ARM) || defined(X86) || defined(PPC) || defined(S390) || \
> defined(S390X) || defined(ARM64) || defined(MIPS) || \
> - defined(MIPS64) || defined(SPARC64) || defined(RISCV64)
> + defined(MIPS64) || defined(SPARC64) || defined(RISCV64) || \
> + defined(LOONGARCH64)
>
> /*
> * Kernel header file for Linux crash dumps.
> @@ -84,7 +85,7 @@ typedef struct _dump_header_asm_s {
> uint32_t dha_eip;
>
> /* the dump registers */
> -#if !defined(S390) && !defined(S390X) && !defined(ARM64) &&
> !defined(RISCV64)
> +#if !defined(S390) && !defined(S390X) && !defined(ARM64) &&
> !defined(RISCV64) && !defined(LOONGARCH64)
> struct pt_regs dha_regs;
> #endif
>
> diff --git a/loongarch64.c b/loongarch64.c
> new file mode 100644
> index 0000000..9791e43
> --- /dev/null
> +++ b/loongarch64.c
> @@ -0,0 +1,46 @@
> +/* loongarch64.c - core analysis suite
> + *
> + * Copyright (C) 2021 Loongson Technology Co., Ltd.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option) any later version.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> + * GNU General Public License for more details.
> + */
> +
> +#ifdef LOONGARCH64
> +
> +#include <elf.h>
> +#include "defs.h"
> +
> +void
> +loongarch64_dump_machdep_table(ulong arg)
> +{
> +}
> +
> +void
> +loongarch64_init(int when)
> +{
> +}
> +
> +void
> +loongarch64_display_regs_from_elf_notes(int cpu, FILE *ofp)
> +{
> +}
> +
> +#else /* !LOONGARCH64 */
> +
> +#include "defs.h"
> +
> +void
> +loongarch64_display_regs_from_elf_notes(int cpu, FILE *ofp)
> +{
> + return;
> +}
> +
> +#endif /* !LOONGARCH64 */
> diff --git a/netdump.c b/netdump.c
> index 61ddeaa..9ace50e 100644
> --- a/netdump.c
> +++ b/netdump.c
> @@ -43,6 +43,7 @@ static void get_netdump_regs_arm(struct bt_info *, ulong
> *, ulong *);
> static void get_netdump_regs_arm64(struct bt_info *, ulong *, ulong *);
> static void get_netdump_regs_mips(struct bt_info *, ulong *, ulong *);
> static void get_netdump_regs_riscv(struct bt_info *, ulong *, ulong *);
> +static void get_netdump_regs_loongarch64(struct bt_info *, ulong *, ulong
> *);
> static void check_dumpfile_size(char *);
> static int proc_kcore_init_32(FILE *, int);
> static int proc_kcore_init_64(FILE *, int);
> @@ -63,6 +64,7 @@ static char *vmcoreinfo_read_string(const char *);
> #define READ_PAGESIZE_FROM_VMCOREINFO() \
> (machine_type("IA64") || machine_type("PPC64") ||
> machine_type("PPC") || machine_type("ARM64"))
>
> +
> /*
> * kdump installs NT_PRSTATUS elf notes only to the cpus
> * that were online during dumping. Hence we call into
> @@ -312,6 +314,12 @@ is_netdump(char *file, ulong source_query)
> goto bailout;
> break;
>
> + case EM_LOONGARCH:
> + if (machine_type_mismatch(file, "LOONGARCH64",
> NULL,
> + source_query))
> + goto bailout;
> + break;
> +
> default:
> if (machine_type_mismatch(file, "(unknown)", NULL,
> source_query))
> @@ -1494,6 +1502,9 @@ dump_Elf32_Ehdr(Elf32_Ehdr *elf)
> case EM_MIPS:
> netdump_print("(EM_MIPS)\n");
> break;
> + case EM_LOONGARCH:
> + netdump_print("(EM_LOONGARCH)\n");
> + break;
> default:
> netdump_print("(unsupported)\n");
> break;
> @@ -1656,6 +1667,9 @@ dump_Elf64_Ehdr(Elf64_Ehdr *elf)
> case EM_AARCH64:
> netdump_print("(EM_AARCH64)\n");
> break;
> + case EM_LOONGARCH:
> + netdump_print("(EM_LOONGARCH)\n");
> + break;
> default:
> netdump_print("(unsupported)\n");
> break;
> @@ -2684,6 +2698,9 @@ get_netdump_regs(struct bt_info *bt, ulong *eip,
> ulong *esp)
> case EM_RISCV:
> get_netdump_regs_riscv(bt, eip, esp);
> break;
> + case EM_LOONGARCH:
> + return get_netdump_regs_loongarch64(bt, eip, esp);
> + break;
>
> default:
> error(FATAL,
> @@ -3895,6 +3912,12 @@ get_netdump_regs_riscv(struct bt_info *bt, ulong
> *eip, ulong *esp)
> machdep->get_stack_frame(bt, eip, esp);
> }
>
> +static void
> +get_netdump_regs_loongarch64(struct bt_info *bt, ulong *eip, ulong *esp)
> +{
> + machdep->get_stack_frame(bt, eip, esp);
> +}
> +
> int
> is_partial_netdump(void)
> {
> diff --git a/ramdump.c b/ramdump.c
> index d2bd7ff..b67ebeb 100644
> --- a/ramdump.c
> +++ b/ramdump.c
> @@ -190,6 +190,8 @@ char *ramdump_to_elf(void)
> e_machine = EM_X86_64;
> else if (machine_type("RISCV64"))
> e_machine = EM_RISCV;
> + else if (machine_type("LOONGARCH64"))
> + e_machine = EM_LOONGARCH;
> else
> error(FATAL, "ramdump: unsupported machine type: %s\n",
> MACHINE_TYPE);
> diff --git a/symbols.c b/symbols.c
> index 8e8b4c3..e1a3457 100644
> --- a/symbols.c
> +++ b/symbols.c
> @@ -2976,9 +2976,11 @@ store_module_kallsyms_v2(struct load_module *lm,
> int start, int curr,
> /*
> * On ARM/ARM64 we have linker mapping symbols like '$a'
> * or '$x' for ARM64, and '$d'.
> + * On LoongArch we have linker mapping symbols like '.L'
> + * or 'L0'.
> * Make sure that these don't end up into our symbol list.
> */
> - if ((machine_type("ARM") || machine_type("ARM64")) &&
> + if ((machine_type("ARM") || machine_type("ARM64") ||
> machine_type("LOONGARCH64")) &&
> !machdep->verify_symbol(nameptr, ec->st_value,
> ec->st_info))
> continue;
>
> @@ -4229,6 +4231,11 @@ is_kernel(char *file)
> goto bailout;
> break;
>
> + case EM_LOONGARCH:
> + if (machine_type_mismatch(file, "LOONGARCH64",
> NULL, 0))
> + goto bailout;
> + break;
> +
> default:
> if (machine_type_mismatch(file, "(unknown)", NULL,
> 0))
> goto bailout;
> @@ -4283,6 +4290,11 @@ is_kernel(char *file)
> goto bailout;
> break;
>
> + case EM_LOONGARCH:
> + if (machine_type_mismatch(file, "LOONGARCH64",
> NULL, 0))
> + goto bailout;
> + break;
> +
> default:
> if (machine_type_mismatch(file, "(unknown)", NULL,
> 0))
> goto bailout;
> @@ -4547,6 +4559,11 @@ is_shared_object(char *file)
> if (machine_type("RISCV64"))
> return TRUE;
> break;
> +
> + case EM_LOONGARCH:
> + if (machine_type("LOONGARCH64"))
> + return TRUE;
> + break;
> }
>
> if (CRASHDEBUG(1))
> @@ -9795,6 +9812,10 @@ dump_offset_table(char *spec, ulong makestruct)
> OFFSET(task_struct_thread_esp));
> fprintf(fp, " task_struct_thread_ksp: %ld\n",
> OFFSET(task_struct_thread_ksp));
> + fprintf(fp, " task_struct_thread_reg01: %ld\n",
> + OFFSET(task_struct_thread_reg01));
> + fprintf(fp, " task_struct_thread_reg03: %ld\n",
> + OFFSET(task_struct_thread_reg03));
> fprintf(fp, " task_struct_thread_reg29: %ld\n",
> OFFSET(task_struct_thread_reg29));
> fprintf(fp, " task_struct_thread_reg31: %ld\n",
> --
> 2.34.1
>
>
10 months, 2 weeks
- 2
- 2
[PATCH v4] Fix: move huge compressed obj from page to zspage
by 陈冠有
when zspage define 'huge', crash-utility zram decompress fail.
we need to be compatible with the previous kernel,
so we can't define 'huge' directly in zspage.
Before:
crash> mod -s zram zram.ko
MODULE NAME BASE SIZE OBJECT FILE
ffffffde224db800 zram ffffffde224d2000 57344 zram.ko
crash> mod -s zsmalloc zsmalloc.ko
MODULE NAME BASE SIZE OBJECT FILE
ffffffde224c5180 zsmalloc ffffffde224bf000 40960 zsmalloc.ko
crash> rd 0x13d89fb0
rd: zspage magic incorrect: b0
After:
crash> rd 0x13d89fb0
13d89fb0: c2b54f7170883b20 ;.pqO..
Link: https://lkml.kernel.org/r/20211115185909.3949505-6-minchan@kernel.org
Signed-off-by: chenguanyou <chenguanyou(a)xiaomi.com>
---
defs.h | 13 +++++++++++
diskdump.c | 67 ++++++++++++++++++++++++++++++++++++++++++++----------
2 files changed, 68 insertions(+), 12 deletions(-)
diff --git a/defs.h b/defs.h
index 96a7a2a..0af69cc 100644
--- a/defs.h
+++ b/defs.h
@@ -2225,6 +2225,7 @@ struct offset_table { /* stash of commonly-used offsets */
long module_memory_base;
long module_memory_size;
long irq_data_irq;
+ long zspage_huge;
};
struct size_table { /* stash of commonly-used sizes */
@@ -7221,6 +7222,18 @@ struct zspage {
unsigned int freeobj;
};
+struct zspage6 {
+ struct {
+ unsigned int huge : 1;
+ unsigned int fullness : 2;
+ unsigned int class : 9;
+ unsigned int isolated : 3;
+ unsigned int magic : 8;
+ };
+ unsigned int inuse;
+ unsigned int freeobj;
+};
+
/*
* makedumpfile.c
*/
diff --git a/diskdump.c b/diskdump.c
index 2c284ff..02aaf1b 100644
--- a/diskdump.c
+++ b/diskdump.c
@@ -2754,6 +2754,9 @@ zram_init(void)
if (INVALID_MEMBER(zram_table_flag))
MEMBER_OFFSET_INIT(zram_table_flag, "zram_table_entry", "value");
STRUCT_SIZE_INIT(zram_table_entry, "zram_table_entry");
+ MEMBER_OFFSET_INIT(zspoll_size_class, "zs_pool", "size_class");
+ MEMBER_OFFSET_INIT(size_class_size, "size_class", "size");
+ MEMBER_OFFSET_INIT(zspage_huge, "zspage", "huge");
}
static unsigned char *
@@ -2761,9 +2764,11 @@ zram_object_addr(ulong pool, ulong handle, unsigned char *zram_buf)
{
ulong obj, off, class, page, zspage;
struct zspage zspage_s;
+ struct zspage6 zspage6_s;
physaddr_t paddr;
unsigned int obj_idx, class_idx, size;
ulong pages[2], sizes[2];
+ ulong zs_magic;
readmem(handle, KVADDR, &obj, sizeof(void *), "zram entry", FAULT_ON_ERROR);
obj >>= OBJ_TAG_BITS;
@@ -2772,11 +2777,19 @@ zram_object_addr(ulong pool, ulong handle, unsigned char *zram_buf)
readmem(page + OFFSET(page_private), KVADDR, &zspage,
sizeof(void *), "page_private", FAULT_ON_ERROR);
- readmem(zspage, KVADDR, &zspage_s, sizeof(struct zspage), "zspage", FAULT_ON_ERROR);
- class_idx = zspage_s.class;
- if (zspage_s.magic != ZSPAGE_MAGIC)
- error(FATAL, "zspage magic incorrect: %x\n", zspage_s.magic);
+ if (VALID_MEMBER(zspage_huge)) {
+ readmem(zspage, KVADDR, &zspage6_s, sizeof(struct zspage6), "zspage6", FAULT_ON_ERROR);
+ class_idx = zspage6_s.class;
+ zs_magic = zspage6_s.magic;
+ } else {
+ readmem(zspage, KVADDR, &zspage_s, sizeof(struct zspage), "zspage", FAULT_ON_ERROR);
+ class_idx = zspage_s.class;
+ zs_magic = zspage_s.magic;
+ }
+
+ if (zs_magic != ZSPAGE_MAGIC)
+ error(FATAL, "zspage magic incorrect: %x\n", zs_magic);
class = pool + OFFSET(zspoll_size_class);
class += (class_idx * sizeof(void *));
@@ -2794,8 +2807,13 @@ zram_object_addr(ulong pool, ulong handle, unsigned char *zram_buf)
}
pages[0] = page;
- readmem(page + OFFSET(page_freelist), KVADDR, &pages[1],
- sizeof(void *), "page_freelist", FAULT_ON_ERROR);
+ if (VALID_MEMBER(page_freelist)) {
+ readmem(page + OFFSET(page_freelist), KVADDR, &pages[1],
+ sizeof(void *), "page_freelist", FAULT_ON_ERROR);
+ } else {
+ readmem(page + OFFSET(page_index), KVADDR, &pages[1],
+ sizeof(void *), "page_index", FAULT_ON_ERROR);
+ }
sizes[0] = PAGESIZE() - off;
sizes[1] = size - sizes[0];
if (!is_page_ptr(pages[0], &paddr)) {
@@ -2812,9 +2830,15 @@ zram_object_addr(ulong pool, ulong handle, unsigned char *zram_buf)
readmem(paddr, PHYSADDR, zram_buf + sizes[0], sizes[1], "zram buffer[1]", FAULT_ON_ERROR);
out:
- readmem(page, KVADDR, &obj, sizeof(void *), "page flags", FAULT_ON_ERROR);
- if (!(obj & (1<<10))) { //PG_OwnerPriv1 flag
- return (zram_buf + ZS_HANDLE_SIZE);
+ if (!VALID_MEMBER(zspage_huge)) {
+ readmem(page, KVADDR, &obj, sizeof(void *), "page flags", FAULT_ON_ERROR);
+ if (!(obj & (1<<10))) { //PG_OwnerPriv1 flag
+ return (zram_buf + ZS_HANDLE_SIZE);
+ }
+ } else {
+ if (!zspage6_s.huge) {
+ return (zram_buf + ZS_HANDLE_SIZE);
+ }
}
return zram_buf;
@@ -2929,6 +2953,11 @@ try_zram_decompress(ulonglong pte_val, unsigned char *buf, ulong len, ulonglong
unsigned char *outbuf = NULL;
ulong zram, zram_table_entry, sector, index, entry, flags, size,
outsize, off;
+ unsigned long *same_buf = NULL;
+ bool is_same = false;
+ ulong ZRAM_SAME_PAGEFLAG;
+ ulong ZRAM_WB_PAGEFLAG;
+ long ZRAM_LOCK_VALUE = 24;
if (INVALID_MEMBER(zram_compressor)) {
zram_init();
@@ -2992,11 +3021,25 @@ try_zram_decompress(ulonglong pte_val, unsigned char *buf, ulong len, ulonglong
sizeof(void *), "entry of table", FAULT_ON_ERROR);
readmem(zram_table_entry + OFFSET(zram_table_flag), KVADDR, &flags,
sizeof(void *), "zram_table_flag", FAULT_ON_ERROR);
- if (!entry || (flags & ZRAM_FLAG_SAME_BIT)) {
- memset(buf, entry, len);
+
+ if (!enumerator_value("ZRAM_LOCK", &ZRAM_LOCK_VALUE)
+ && THIS_KERNEL_VERSION >= LINUX(6,1,0))
+ ZRAM_LOCK_VALUE = PAGESHIFT() + 1;
+
+ ZRAM_SAME_PAGEFLAG = ZRAM_LOCK_VALUE + 1;
+ ZRAM_WB_PAGEFLAG = ZRAM_LOCK_VALUE + 2;
+
+ if (!entry || flags & (1 << ZRAM_SAME_PAGEFLAG)) {
+ same_buf = (unsigned long *)GETBUF(PAGESIZE());
+ for (int count = 0; count < PAGESIZE() / sizeof(unsigned long); count++) {
+ same_buf[count] = entry;
+ }
+ memcpy(buf, same_buf + off, PAGESIZE());
+ FREEBUF(same_buf);
goto out;
}
- size = flags & (ZRAM_FLAG_SHIFT -1);
+
+ size = flags & ((1 << ZRAM_LOCK_VALUE) - 1);
if (size == 0) {
len = 0;
goto out;
--
2.39.0
#/******本邮件及其附件含有小米公司的保密信息，仅限于发送给上面地址中列出的个人或群组。禁止任何其他人以任何形式使用（包括但不限于全部或部分地泄露、复制、或散发）本邮件中的信息。如果您错收了本邮件，请您立即电话或邮件通知发件人并删除本邮件！ This e-mail and its attachments contain confidential information from XIAOMI, which is intended only for the person or entity whose address is listed above. Any use of the information contained herein in any way (including, but not limited to, total or partial disclosure, reproduction, or dissemination) by persons other than the intended recipient(s) is prohibited. If you receive this e-mail in error, please notify the sender by phone or email immediately and delete it!******/#
10 months, 3 weeks
- 3
- 12
[PATCH RESEND v2] diskdump: add hook for additional checks on prstatus notes validity
by Aditya Gupta
Upstream crash reports these warnings on PowerPC64:
WARNING: cpu 0 invalid NT_PRSTATUS note (n_type != NT_PRSTATUS)
...
Apart from these warnings, register values are also invalid.
This warning was found in the commit:
commit db8c030857b4 ("diskdump/netdump: fix segmentation fault
caused by failure of stopping CPUs")
With above commit, crash checks whether 'crash_notes' is initialised,
before mapping PRSTATUS notes.
But some architectures such as PowerPC64, in fadump case
(firmware-assisted dump), don't populate 'crash_notes' since the
registers are already stored in the cpu notes in the vmcore.
Instead of checking 'crash_notes' for all architectures, introduce
a machdep hook ('is_cpu_prstatus_valid'), for architectures to
decide validity checks for PRSTATUS notes
A default hook ('diskdump_is_cpu_prstatus_valid') has also been provided
for all architectures other than PowerPC64, which checks if 'crash_notes'
for a given cpu is valid, maintaining the current behaviour
PowerPC64 doesn't utilise 'crash_notes' to get register values, so no
additional checks are required
Fixes: db8c030857b4 ("diskdump/netdump: fix segmentation fault caused by failure of stopping CPUs")
Signed-off-by: Aditya Gupta <adityag(a)linux.ibm.com>
---
Testing
=======
NOTE: To test this on PowerPC64 with upstream kernel dump, AND on system
with Radix MMU, following patch will also be needed to be applied:
Link: https://listman.redhat.com/archives/crash-utility/2023-September/010961.html
This is due to change in vmemmap address mapping for Radix MMU, since
following patch in the kernel:
368a0590d954: (powerpc/book3s64/vmemmap: switch radix to use a
different vmemmap handling function)
More details about the change are in the linked patch. Basically what
changed is, the address mapping for vmemmap address is now in kernel
page table, in case of Radix MMU, instead of 'vmemmap_list' which is currently
used in crash.
Git Tree for Testing
====================
1. With this patch (diskdump: add hook for additional ...) applied:
https://github.com/adi-g15-ibm/crash/tree/bugzilla-203256-list-v2
2. With both this and the linked patch (ppc64: do page traversal ...) applied:
https://github.com/adi-g15-ibm/crash/tree/bugzilla-203256-withupstreamradix
Changelog
=========
V2
+ added fix for netdump also, same as diskdump, as that was also modified by
db8c030857b4
+ fixed compile warnings
---
---
defs.h | 2 ++
diskdump.c | 15 ++++++++++++---
netdump.c | 5 ++---
ppc64.c | 10 ++++++++++
4 files changed, 26 insertions(+), 6 deletions(-)
diff --git a/defs.h b/defs.h
index 96a7a2a31471..23dee48759fb 100644
--- a/defs.h
+++ b/defs.h
@@ -1075,6 +1075,7 @@ struct machdep_table {
void (*show_interrupts)(int, ulong *);
int (*is_page_ptr)(ulong, physaddr_t *);
int (*get_cpu_reg)(int, int, const char *, int, void *);
+ int (*is_cpu_prstatus_valid)(int cpu);
};
/*
@@ -7181,6 +7182,7 @@ int dumpfile_is_split(void);
void show_split_dumpfiles(void);
void x86_process_elf_notes(void *, unsigned long);
void *diskdump_get_prstatus_percpu(int);
+int diskdump_is_cpu_prstatus_valid(int cpu);
int have_crash_notes(int cpu);
void map_cpus_to_prstatus_kdump_cmprs(void);
void diskdump_display_regs(int, FILE *);
diff --git a/diskdump.c b/diskdump.c
index 2c284ff3f97f..ad9a00b08ce1 100644
--- a/diskdump.c
+++ b/diskdump.c
@@ -142,13 +142,22 @@ int have_crash_notes(int cpu)
return TRUE;
}
+int diskdump_is_cpu_prstatus_valid(int cpu)
+{
+ static int crash_notes_exists = -1;
+
+ if (crash_notes_exists == -1)
+ crash_notes_exists = kernel_symbol_exists("crash_notes");
+
+ return (!crash_notes_exists || have_crash_notes(cpu));
+}
+
void
map_cpus_to_prstatus_kdump_cmprs(void)
{
void **nt_ptr;
int online, i, j, nrcpus;
size_t size;
- int crash_notes_exists;
if (pc->flags2 & QEMU_MEM_DUMP_COMPRESSED) /* notes exist for all cpus */
goto resize_note_pointers;
@@ -171,10 +180,9 @@ map_cpus_to_prstatus_kdump_cmprs(void)
* Re-populate the array with the notes mapping to online cpus
*/
nrcpus = (kt->kernel_NR_CPUS ? kt->kernel_NR_CPUS : NR_CPUS);
- crash_notes_exists = kernel_symbol_exists("crash_notes");
for (i = 0, j = 0; i < nrcpus; i++) {
- if (in_cpu_map(ONLINE_MAP, i) && (!crash_notes_exists || have_crash_notes(i))) {
+ if (in_cpu_map(ONLINE_MAP, i) && machdep->is_cpu_prstatus_valid(i)) {
dd->nt_prstatus_percpu[i] = nt_ptr[j++];
dd->num_prstatus_notes =
MAX(dd->num_prstatus_notes, i+1);
@@ -1076,6 +1084,7 @@ diskdump_init(char *unused, FILE *fptr)
if (!DISKDUMP_VALID() && !KDUMP_CMPRS_VALID())
return FALSE;
+ machdep->is_cpu_prstatus_valid = diskdump_is_cpu_prstatus_valid;
dd->ofp = fptr;
return TRUE;
}
diff --git a/netdump.c b/netdump.c
index 61ddeaa08831..390786364959 100644
--- a/netdump.c
+++ b/netdump.c
@@ -75,7 +75,6 @@ map_cpus_to_prstatus(void)
void **nt_ptr;
int online, i, j, nrcpus;
size_t size;
- int crash_notes_exists;
if (pc->flags2 & QEMU_MEM_DUMP_ELF) /* notes exist for all cpus */
return;
@@ -98,10 +97,9 @@ map_cpus_to_prstatus(void)
* Re-populate the array with the notes mapping to online cpus
*/
nrcpus = (kt->kernel_NR_CPUS ? kt->kernel_NR_CPUS : NR_CPUS);
- crash_notes_exists = kernel_symbol_exists("crash_notes");
for (i = 0, j = 0; i < nrcpus; i++) {
- if (in_cpu_map(ONLINE_MAP, i) && (!crash_notes_exists || have_crash_notes(i))) {
+ if (in_cpu_map(ONLINE_MAP, i) && machdep->is_cpu_prstatus_valid(i)) {
nd->nt_prstatus_percpu[i] = nt_ptr[j++];
nd->num_prstatus_notes =
MAX(nd->num_prstatus_notes, i+1);
@@ -735,6 +733,7 @@ netdump_init(char *unused, FILE *fptr)
if (!VMCORE_VALID())
return FALSE;
+ machdep->is_cpu_prstatus_valid = diskdump_is_cpu_prstatus_valid;
nd->ofp = fptr;
check_dumpfile_size(pc->dumpfile);
diff --git a/ppc64.c b/ppc64.c
index fc34006f4863..5a8ef9e58173 100644
--- a/ppc64.c
+++ b/ppc64.c
@@ -298,6 +298,15 @@ struct machine_specific book3e_machine_specific = {
.is_vmaddr = book3e_is_vmaddr,
};
+/**
+ * No additional checks are required on PPC64, for checking if PRSTATUS notes
+ * is valid
+ */
+static int ppc64_is_cpu_prstatus_valid(int cpu)
+{
+ return TRUE;
+}
+
#define SKIBOOT_BASE 0x30000000
/*
@@ -400,6 +409,7 @@ ppc64_init(int when)
machdep->value_to_symbol = generic_machdep_value_to_symbol;
machdep->get_kvaddr_ranges = ppc64_get_kvaddr_ranges;
machdep->init_kernel_pgd = NULL;
+ machdep->is_cpu_prstatus_valid = ppc64_is_cpu_prstatus_valid;
if (symbol_exists("vmemmap_populate")) {
if (symbol_exists("vmemmap")) {
--
2.41.0
10 months, 3 weeks
- 2
- 2
Re: [Crash-utility] [PATCH RESEND v2] diskdump: add hook for additional checks on prstatus notes validity
by lijiang
On Sat, Oct 14, 2023 at 9:39 PM <crash-utility-request(a)redhat.com> wrote:
> Date: Sat, 14 Oct 2023 19:09:30 +0530
> From: Aditya Gupta <adityag(a)linux.ibm.com>
> To: crash-utility(a)redhat.com, <lijiang(a)redhat.com>
> Cc: Hari Bathini <hbathini(a)linux.ibm.com>, Mahesh J Salgaonkar
> <mahesh(a)linux.ibm.com>, Sourabh Jain <sourabhjain(a)linux.ibm.com>,
> d.hatayama(a)fujitsu.com
> Subject: [Crash-utility] [PATCH RESEND v2] diskdump: add hook for
> additional checks on prstatus notes validity
> Message-ID: <20231014133930.147343-1-adityag(a)linux.ibm.com>
> Content-Type: text/plain; charset="US-ASCII"; x-default=true
>
> Upstream crash reports these warnings on PowerPC64:
>
> WARNING: cpu 0 invalid NT_PRSTATUS note (n_type != NT_PRSTATUS)
> ...
>
> Apart from these warnings, register values are also invalid.
>
> This warning was found in the commit:
>
> commit db8c030857b4 ("diskdump/netdump: fix segmentation fault
> caused by failure of stopping CPUs")
>
> With above commit, crash checks whether 'crash_notes' is initialised,
> before mapping PRSTATUS notes.
>
> But some architectures such as PowerPC64, in fadump case
> (firmware-assisted dump), don't populate 'crash_notes' since the
> registers are already stored in the cpu notes in the vmcore.
>
> Instead of checking 'crash_notes' for all architectures, introduce
> a machdep hook ('is_cpu_prstatus_valid'), for architectures to
> decide validity checks for PRSTATUS notes
>
> A default hook ('diskdump_is_cpu_prstatus_valid') has also been provided
> for all architectures other than PowerPC64, which checks if 'crash_notes'
> for a given cpu is valid, maintaining the current behaviour
>
> PowerPC64 doesn't utilise 'crash_notes' to get register values, so no
> additional checks are required
>
> Fixes: db8c030857b4 ("diskdump/netdump: fix segmentation fault caused by
> failure of stopping CPUs")
> Signed-off-by: Aditya Gupta <adityag(a)linux.ibm.com>
>
> ---
> Testing
> =======
>
> NOTE: To test this on PowerPC64 with upstream kernel dump, AND on system
> with Radix MMU, following patch will also be needed to be applied:
>
> Link:
> https://listman.redhat.com/archives/crash-utility/2023-September/010961.html
>
> This is due to change in vmemmap address mapping for Radix MMU, since
> following patch in the kernel:
>
> 368a0590d954: (powerpc/book3s64/vmemmap: switch radix to use a
> different vmemmap handling function)
>
> More details about the change are in the linked patch. Basically what
> changed is, the address mapping for vmemmap address is now in kernel
> page table, in case of Radix MMU, instead of 'vmemmap_list' which is
> currently
> used in crash.
>
> Git Tree for Testing
> ====================
>
> 1. With this patch (diskdump: add hook for additional ...) applied:
>
> https://github.com/adi-g15-ibm/crash/tree/bugzilla-203256-list-v2
>
> 2. With both this and the linked patch (ppc64: do page traversal ...)
> applied:
>
> https://github.com/adi-g15-ibm/crash/tree/bugzilla-203256-withupstreamradix
>
> Changelog
> =========
>
> V2
> + added fix for netdump also, same as diskdump, as that was also modified
> by
> db8c030857b4
> + fixed compile warnings
>
>
Thank you for the update, Aditya.
For the v2: Ack.
Thanks
Lianbo
---
>
> ---
> defs.h | 2 ++
> diskdump.c | 15 ++++++++++++---
> netdump.c | 5 ++---
> ppc64.c | 10 ++++++++++
> 4 files changed, 26 insertions(+), 6 deletions(-)
>
> diff --git a/defs.h b/defs.h
> index 96a7a2a31471..23dee48759fb 100644
> --- a/defs.h
> +++ b/defs.h
> @@ -1075,6 +1075,7 @@ struct machdep_table {
> void (*show_interrupts)(int, ulong *);
> int (*is_page_ptr)(ulong, physaddr_t *);
> int (*get_cpu_reg)(int, int, const char *, int, void *);
> + int (*is_cpu_prstatus_valid)(int cpu);
> };
>
> /*
> @@ -7181,6 +7182,7 @@ int dumpfile_is_split(void);
> void show_split_dumpfiles(void);
> void x86_process_elf_notes(void *, unsigned long);
> void *diskdump_get_prstatus_percpu(int);
> +int diskdump_is_cpu_prstatus_valid(int cpu);
> int have_crash_notes(int cpu);
> void map_cpus_to_prstatus_kdump_cmprs(void);
> void diskdump_display_regs(int, FILE *);
> diff --git a/diskdump.c b/diskdump.c
> index 2c284ff3f97f..ad9a00b08ce1 100644
> --- a/diskdump.c
> +++ b/diskdump.c
> @@ -142,13 +142,22 @@ int have_crash_notes(int cpu)
> return TRUE;
> }
>
> +int diskdump_is_cpu_prstatus_valid(int cpu)
> +{
> + static int crash_notes_exists = -1;
> +
> + if (crash_notes_exists == -1)
> + crash_notes_exists = kernel_symbol_exists("crash_notes");
> +
> + return (!crash_notes_exists || have_crash_notes(cpu));
> +}
> +
> void
> map_cpus_to_prstatus_kdump_cmprs(void)
> {
> void **nt_ptr;
> int online, i, j, nrcpus;
> size_t size;
> - int crash_notes_exists;
>
> if (pc->flags2 & QEMU_MEM_DUMP_COMPRESSED) /* notes exist for all
> cpus */
> goto resize_note_pointers;
> @@ -171,10 +180,9 @@ map_cpus_to_prstatus_kdump_cmprs(void)
> * Re-populate the array with the notes mapping to online cpus
> */
> nrcpus = (kt->kernel_NR_CPUS ? kt->kernel_NR_CPUS : NR_CPUS);
> - crash_notes_exists = kernel_symbol_exists("crash_notes");
>
> for (i = 0, j = 0; i < nrcpus; i++) {
> - if (in_cpu_map(ONLINE_MAP, i) && (!crash_notes_exists ||
> have_crash_notes(i))) {
> + if (in_cpu_map(ONLINE_MAP, i) &&
> machdep->is_cpu_prstatus_valid(i)) {
> dd->nt_prstatus_percpu[i] = nt_ptr[j++];
> dd->num_prstatus_notes =
> MAX(dd->num_prstatus_notes, i+1);
> @@ -1076,6 +1084,7 @@ diskdump_init(char *unused, FILE *fptr)
> if (!DISKDUMP_VALID() && !KDUMP_CMPRS_VALID())
> return FALSE;
>
> + machdep->is_cpu_prstatus_valid = diskdump_is_cpu_prstatus_valid;
> dd->ofp = fptr;
> return TRUE;
> }
> diff --git a/netdump.c b/netdump.c
> index 61ddeaa08831..390786364959 100644
> --- a/netdump.c
> +++ b/netdump.c
> @@ -75,7 +75,6 @@ map_cpus_to_prstatus(void)
> void **nt_ptr;
> int online, i, j, nrcpus;
> size_t size;
> - int crash_notes_exists;
>
> if (pc->flags2 & QEMU_MEM_DUMP_ELF) /* notes exist for all cpus */
> return;
> @@ -98,10 +97,9 @@ map_cpus_to_prstatus(void)
> * Re-populate the array with the notes mapping to online cpus
> */
> nrcpus = (kt->kernel_NR_CPUS ? kt->kernel_NR_CPUS : NR_CPUS);
> - crash_notes_exists = kernel_symbol_exists("crash_notes");
>
> for (i = 0, j = 0; i < nrcpus; i++) {
> - if (in_cpu_map(ONLINE_MAP, i) && (!crash_notes_exists ||
> have_crash_notes(i))) {
> + if (in_cpu_map(ONLINE_MAP, i) &&
> machdep->is_cpu_prstatus_valid(i)) {
> nd->nt_prstatus_percpu[i] = nt_ptr[j++];
> nd->num_prstatus_notes =
> MAX(nd->num_prstatus_notes, i+1);
> @@ -735,6 +733,7 @@ netdump_init(char *unused, FILE *fptr)
> if (!VMCORE_VALID())
> return FALSE;
>
> + machdep->is_cpu_prstatus_valid = diskdump_is_cpu_prstatus_valid;
> nd->ofp = fptr;
>
> check_dumpfile_size(pc->dumpfile);
> diff --git a/ppc64.c b/ppc64.c
> index fc34006f4863..5a8ef9e58173 100644
> --- a/ppc64.c
> +++ b/ppc64.c
> @@ -298,6 +298,15 @@ struct machine_specific book3e_machine_specific = {
> .is_vmaddr = book3e_is_vmaddr,
> };
>
> +/**
> + * No additional checks are required on PPC64, for checking if PRSTATUS
> notes
> + * is valid
> + */
> +static int ppc64_is_cpu_prstatus_valid(int cpu)
> +{
> + return TRUE;
> +}
> +
> #define SKIBOOT_BASE 0x30000000
>
> /*
> @@ -400,6 +409,7 @@ ppc64_init(int when)
> machdep->value_to_symbol = generic_machdep_value_to_symbol;
> machdep->get_kvaddr_ranges = ppc64_get_kvaddr_ranges;
> machdep->init_kernel_pgd = NULL;
> + machdep->is_cpu_prstatus_valid =
> ppc64_is_cpu_prstatus_valid;
>
> if (symbol_exists("vmemmap_populate")) {
> if (symbol_exists("vmemmap")) {
> --
> 2.41.0
>
10 months, 3 weeks
- 2
- 1
Re: [Crash-utility] [PATCH v1] diskdump: add hook for additional checks on prstatus notes validity
by lijiang
Hi, Aditya
Thank you for the fix.
On Fri, Sep 22, 2023 at 7:03 AM <crash-utility-request(a)redhat.com> wrote:
> Date: Thu, 21 Sep 2023 12:29:05 +0530
> From: Aditya Gupta <adityag(a)linux.ibm.com>
> To: crash-utility(a)redhat.com
> Cc: Hari Bathini <hbathini(a)linux.ibm.com>, Mahesh J Salgaonkar
> <mahesh(a)linux.ibm.com>, Sourabh Jain <sourabhjain(a)linux.ibm.com>,
> d.hatayama(a)fujitsu.com
> Subject: [Crash-utility] [PATCH v1] diskdump: add hook for additional
> checks on prstatus notes validity
> Message-ID: <20230921065905.1020839-1-adityag(a)linux.ibm.com>
> Content-Type: text/plain; charset="US-ASCII"; x-default=true
>
> Upstream crash reports these warnings on PowerPC64:
>
> WARNING: cpu 0 invalid NT_PRSTATUS note (n_type != NT_PRSTATUS)
> ...
>
> Apart from these warnings, register values are also invalid.
>
> This warning was found in the commit:
>
> commit db8c030857b4 ("diskdump/netdump: fix segmentation fault
> caused by failure of stopping CPUs")
>
> With above commit, crash checks whether 'crash_notes' is initialised,
> before mapping PRSTATUS notes.
>
> But some architectures such as PowerPC64, in fadump case
> (firmware-assisted dump), don't populate 'crash_notes' since the
> registers are already stored in the cpu notes in the vmcore.
>
> Instead of checking 'crash_notes' for all architectures, introduce
> a machdep hook ('is_cpu_prstatus_valid'), for architectures to
> decide validity checks for PRSTATUS notes
>
> A default hook ('diskdump_is_cpu_prstatus_valid') has also been provided
> for all architectures other than PowerPC64, which checks if 'crash_notes'
> for a given cpu is valid, maintaining the current behaviour
>
> PowerPC64 doesn't utilise 'crash_notes' to get register values, so no
> additional checks are required
>
> Fixes: db8c030857b4 ("diskdump/netdump: fix segmentation fault caused by
> failure of stopping CPUs")
> Signed-off-by: Aditya Gupta <adityag(a)linux.ibm.com>
>
> ---
> Testing
> =======
>
> NOTE: To test this on PowerPC64 with upstream kernel dump, AND on system
> with Radix MMU, following patch will also be needed to be applied:
>
> Link:
> https://listman.redhat.com/archives/crash-utility/2023-September/010961.html
>
> This is due to change in vmemmap address mapping for Radix MMU, since
> following patch in the kernel:
>
> 368a0590d954: (powerpc/book3s64/vmemmap: switch radix to use a
> different vmemmap handling function)
>
> More details about the change are in the linked patch. Basically what
> changed is, the address mapping for vmemmap address is now in kernel
> page table, in case of Radix MMU, instead of 'vmemmap_list' which is
> currently
> used in crash.
>
> Git Tree for Testing
> ====================
>
> 1. With this patch (diskdump: add hook for additional ...) applied:
>
> https://github.com/adi-g15-ibm/crash/tree/bugzilla-203256-list-v1
>
> 2. With both this and the linked patch (ppc64: do page traversal ...)
> applied:
>
> https://github.com/adi-g15-ibm/crash/tree/bugzilla-203256-withupstreamradix
>
> ---
> ---
> defs.h | 1 +
> diskdump.c | 15 ++++++++++++---
> ppc64.c | 10 ++++++++++
> 3 files changed, 23 insertions(+), 3 deletions(-)
>
> diff --git a/defs.h b/defs.h
> index 96a7a2a31471..f7f56947e5ac 100644
> --- a/defs.h
> +++ b/defs.h
> @@ -1073,6 +1073,7 @@ struct machdep_table {
> int (*verify_line_number)(ulong, ulong, ulong);
> void (*get_irq_affinity)(int);
> void (*show_interrupts)(int, ulong *);
> + int (*is_cpu_prstatus_valid)(int cpu);
>
I would suggest putting it at the end of this table. Although it may not
break the compatibility of the extension module, just like the
offset_table/size_table, I get used to doing that if there is no special
reason.
> int (*is_page_ptr)(ulong, physaddr_t *);
> int (*get_cpu_reg)(int, int, const char *, int, void *);
> };
> diff --git a/diskdump.c b/diskdump.c
> index 2c284ff3f97f..ad9a00b08ce1 100644
> --- a/diskdump.c
> +++ b/diskdump.c
> @@ -142,13 +142,22 @@ int have_crash_notes(int cpu)
> return TRUE;
> }
>
> +int diskdump_is_cpu_prstatus_valid(int cpu)
> +{
> + static int crash_notes_exists = -1;
> +
> + if (crash_notes_exists == -1)
> + crash_notes_exists = kernel_symbol_exists("crash_notes");
> +
> + return (!crash_notes_exists || have_crash_notes(cpu));
> +}
> +
>
Got a warning as below:
cc -c -g -DX86_64 -DLZO -DGDB_10_2 diskdump.c -Wall -O2
-Wstrict-prototypes -Wmissing-prototypes -fstack-protector
-Wformat-security
diskdump.c:145:5: warning: no previous prototype for
‘diskdump_is_cpu_prstatus_valid’ [-Wmissing-prototypes]
145 | int diskdump_is_cpu_prstatus_valid(int cpu)
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> void
> map_cpus_to_prstatus_kdump_cmprs(void)
> {
> void **nt_ptr;
> int online, i, j, nrcpus;
> size_t size;
> - int crash_notes_exists;
>
> if (pc->flags2 & QEMU_MEM_DUMP_COMPRESSED) /* notes exist for all
> cpus */
> goto resize_note_pointers;
> @@ -171,10 +180,9 @@ map_cpus_to_prstatus_kdump_cmprs(void)
> * Re-populate the array with the notes mapping to online cpus
> */
> nrcpus = (kt->kernel_NR_CPUS ? kt->kernel_NR_CPUS : NR_CPUS);
> - crash_notes_exists = kernel_symbol_exists("crash_notes");
>
> for (i = 0, j = 0; i < nrcpus; i++) {
> - if (in_cpu_map(ONLINE_MAP, i) && (!crash_notes_exists ||
> have_crash_notes(i))) {
> + if (in_cpu_map(ONLINE_MAP, i) &&
> machdep->is_cpu_prstatus_valid(i)) {
> dd->nt_prstatus_percpu[i] = nt_ptr[j++];
> dd->num_prstatus_notes =
> MAX(dd->num_prstatus_notes, i+1);
> @@ -1076,6 +1084,7 @@ diskdump_init(char *unused, FILE *fptr)
> if (!DISKDUMP_VALID() && !KDUMP_CMPRS_VALID())
> return FALSE;
>
> + machdep->is_cpu_prstatus_valid = diskdump_is_cpu_prstatus_valid;
> dd->ofp = fptr;
> return TRUE;
> }
> diff --git a/ppc64.c b/ppc64.c
> index fc34006f4863..1159b8c3a8e7 100644
> --- a/ppc64.c
> +++ b/ppc64.c
> @@ -298,6 +298,15 @@ struct machine_specific book3e_machine_specific = {
> .is_vmaddr = book3e_is_vmaddr,
> };
>
> +/**
> + * No additional checks are required on PPC64, for checking if PRSTATUS
> notes
> + * is valid
> + */
> +int ppc64_is_cpu_prstatus_valid(int cpu)
> +{
> + return TRUE;
> +}
> +
> #define SKIBOOT_BASE 0x30000000
>
> /*
> @@ -418,6 +427,7 @@ ppc64_init(int when)
> break;
>
> case POST_GDB:
> + machdep->is_cpu_prstatus_valid =
> ppc64_is_cpu_prstatus_valid;
>
The hook is set in the stage of POST_GDB, I'm wondering if the current
warning is still shown in the crash minimal mode(with option --minimal).
Can you help to confirm this one?
And other changes are fine to me.
Thanks.
Lianbo
ms = machdep->machspec;
>
> if (!(machdep->flags & BOOK3E)) {
> --
> 2.41.0
>
10 months, 3 weeks
- 2
- 5
[PATCH v2] diskdump: add hook for additional checks on prstatus notes validity
by Aditya Gupta
Upstream crash reports these warnings on PowerPC64:
WARNING: cpu 0 invalid NT_PRSTATUS note (n_type != NT_PRSTATUS)
...
Apart from these warnings, register values are also invalid.
This warning was found in the commit:
commit db8c030857b4 ("diskdump/netdump: fix segmentation fault
caused by failure of stopping CPUs")
With above commit, crash checks whether 'crash_notes' is initialised,
before mapping PRSTATUS notes.
But some architectures such as PowerPC64, in fadump case
(firmware-assisted dump), don't populate 'crash_notes' since the
registers are already stored in the cpu notes in the vmcore.
Instead of checking 'crash_notes' for all architectures, introduce
a machdep hook ('is_cpu_prstatus_valid'), for architectures to
decide validity checks for PRSTATUS notes
A default hook ('diskdump_is_cpu_prstatus_valid') has also been provided
for all architectures other than PowerPC64, which checks if 'crash_notes'
for a given cpu is valid, maintaining the current behaviour
PowerPC64 doesn't utilise 'crash_notes' to get register values, so no
additional checks are required
Fixes: db8c030857b4 ("diskdump/netdump: fix segmentation fault caused by failure of stopping CPUs")
Signed-off-by: Aditya Gupta <adityag(a)linux.ibm.com>
---
Testing
=======
NOTE: To test this on PowerPC64 with upstream kernel dump, AND on system
with Radix MMU, following patch will also be needed to be applied:
Link: https://listman.redhat.com/archives/crash-utility/2023-September/010961.html
This is due to change in vmemmap address mapping for Radix MMU, since
following patch in the kernel:
368a0590d954: (powerpc/book3s64/vmemmap: switch radix to use a
different vmemmap handling function)
More details about the change are in the linked patch. Basically what
changed is, the address mapping for vmemmap address is now in kernel
page table, in case of Radix MMU, instead of 'vmemmap_list' which is currently
used in crash.
Git Tree for Testing
====================
1. With this patch (diskdump: add hook for additional ...) applied:
https://github.com/adi-g15-ibm/crash/tree/bugzilla-203256-list-v2
2. With both this and the linked patch (ppc64: do page traversal ...) applied:
https://github.com/adi-g15-ibm/crash/tree/bugzilla-203256-withupstreamradix
---
---
defs.h | 2 ++
diskdump.c | 15 ++++++++++++---
netdump.c | 5 ++---
ppc64.c | 10 ++++++++++
4 files changed, 26 insertions(+), 6 deletions(-)
diff --git a/defs.h b/defs.h
index 96a7a2a31471..23dee48759fb 100644
--- a/defs.h
+++ b/defs.h
@@ -1075,6 +1075,7 @@ struct machdep_table {
void (*show_interrupts)(int, ulong *);
int (*is_page_ptr)(ulong, physaddr_t *);
int (*get_cpu_reg)(int, int, const char *, int, void *);
+ int (*is_cpu_prstatus_valid)(int cpu);
};
/*
@@ -7181,6 +7182,7 @@ int dumpfile_is_split(void);
void show_split_dumpfiles(void);
void x86_process_elf_notes(void *, unsigned long);
void *diskdump_get_prstatus_percpu(int);
+int diskdump_is_cpu_prstatus_valid(int cpu);
int have_crash_notes(int cpu);
void map_cpus_to_prstatus_kdump_cmprs(void);
void diskdump_display_regs(int, FILE *);
diff --git a/diskdump.c b/diskdump.c
index 2c284ff3f97f..ad9a00b08ce1 100644
--- a/diskdump.c
+++ b/diskdump.c
@@ -142,13 +142,22 @@ int have_crash_notes(int cpu)
return TRUE;
}
+int diskdump_is_cpu_prstatus_valid(int cpu)
+{
+ static int crash_notes_exists = -1;
+
+ if (crash_notes_exists == -1)
+ crash_notes_exists = kernel_symbol_exists("crash_notes");
+
+ return (!crash_notes_exists || have_crash_notes(cpu));
+}
+
void
map_cpus_to_prstatus_kdump_cmprs(void)
{
void **nt_ptr;
int online, i, j, nrcpus;
size_t size;
- int crash_notes_exists;
if (pc->flags2 & QEMU_MEM_DUMP_COMPRESSED) /* notes exist for all cpus */
goto resize_note_pointers;
@@ -171,10 +180,9 @@ map_cpus_to_prstatus_kdump_cmprs(void)
* Re-populate the array with the notes mapping to online cpus
*/
nrcpus = (kt->kernel_NR_CPUS ? kt->kernel_NR_CPUS : NR_CPUS);
- crash_notes_exists = kernel_symbol_exists("crash_notes");
for (i = 0, j = 0; i < nrcpus; i++) {
- if (in_cpu_map(ONLINE_MAP, i) && (!crash_notes_exists || have_crash_notes(i))) {
+ if (in_cpu_map(ONLINE_MAP, i) && machdep->is_cpu_prstatus_valid(i)) {
dd->nt_prstatus_percpu[i] = nt_ptr[j++];
dd->num_prstatus_notes =
MAX(dd->num_prstatus_notes, i+1);
@@ -1076,6 +1084,7 @@ diskdump_init(char *unused, FILE *fptr)
if (!DISKDUMP_VALID() && !KDUMP_CMPRS_VALID())
return FALSE;
+ machdep->is_cpu_prstatus_valid = diskdump_is_cpu_prstatus_valid;
dd->ofp = fptr;
return TRUE;
}
diff --git a/netdump.c b/netdump.c
index 61ddeaa08831..390786364959 100644
--- a/netdump.c
+++ b/netdump.c
@@ -75,7 +75,6 @@ map_cpus_to_prstatus(void)
void **nt_ptr;
int online, i, j, nrcpus;
size_t size;
- int crash_notes_exists;
if (pc->flags2 & QEMU_MEM_DUMP_ELF) /* notes exist for all cpus */
return;
@@ -98,10 +97,9 @@ map_cpus_to_prstatus(void)
* Re-populate the array with the notes mapping to online cpus
*/
nrcpus = (kt->kernel_NR_CPUS ? kt->kernel_NR_CPUS : NR_CPUS);
- crash_notes_exists = kernel_symbol_exists("crash_notes");
for (i = 0, j = 0; i < nrcpus; i++) {
- if (in_cpu_map(ONLINE_MAP, i) && (!crash_notes_exists || have_crash_notes(i))) {
+ if (in_cpu_map(ONLINE_MAP, i) && machdep->is_cpu_prstatus_valid(i)) {
nd->nt_prstatus_percpu[i] = nt_ptr[j++];
nd->num_prstatus_notes =
MAX(nd->num_prstatus_notes, i+1);
@@ -735,6 +733,7 @@ netdump_init(char *unused, FILE *fptr)
if (!VMCORE_VALID())
return FALSE;
+ machdep->is_cpu_prstatus_valid = diskdump_is_cpu_prstatus_valid;
nd->ofp = fptr;
check_dumpfile_size(pc->dumpfile);
diff --git a/ppc64.c b/ppc64.c
index fc34006f4863..5a8ef9e58173 100644
--- a/ppc64.c
+++ b/ppc64.c
@@ -298,6 +298,15 @@ struct machine_specific book3e_machine_specific = {
.is_vmaddr = book3e_is_vmaddr,
};
+/**
+ * No additional checks are required on PPC64, for checking if PRSTATUS notes
+ * is valid
+ */
+static int ppc64_is_cpu_prstatus_valid(int cpu)
+{
+ return TRUE;
+}
+
#define SKIBOOT_BASE 0x30000000
/*
@@ -400,6 +409,7 @@ ppc64_init(int when)
machdep->value_to_symbol = generic_machdep_value_to_symbol;
machdep->get_kvaddr_ranges = ppc64_get_kvaddr_ranges;
machdep->init_kernel_pgd = NULL;
+ machdep->is_cpu_prstatus_valid = ppc64_is_cpu_prstatus_valid;
if (symbol_exists("vmemmap_populate")) {
if (symbol_exists("vmemmap")) {
--
2.41.0
10 months, 4 weeks
- 1
- 1