11:34 p.m.
crash> rd -u 0x400000
rd: invalid user virtual address: 400000 type: "64-bit UVADDR"
Ah, damn it -- it would also be printed later on in
readmem() if uvtop() fails:
while (size > 0) {
switch (memtype)
{
case UVADDR:
if (!uvtop(CURRENT_CONTEXT(), addr, &paddr, 0)) {
if (PRINT_ERROR_MESSAGE)
error(INFO, INVALID_UVADDR, addr, type);
goto readmem_error;
}
break;
That's probably where it's happening...
I would guess that x86_64_uvtop() needs to immediately
recognize if the new VM scheme is in place, and if so
it should veer off to a new "x86_64_uvtop_lev4()" routine.
We definitely want to keep the current one untouched, while
the new one must need a 4th layer of page table translation.
Note that x86_64_kvtop() has the extra FILL_PML4() stuff
before the 3-level breakdown -- I guess that's why vmalloc
addresses still translate. But there must be another layer
user virtual addresses need to go through.
Dave
12:14 a.m.
New subject: [PATCH} crash 4.0-2.8 fixes to make it workonx86_64 for 2.6.14
On Wed, 2005-11-02 at 17:34 -0500, Dave Anderson wrote:
> crash> rd -u 0x400000
> rd: invalid user virtual address: 400000 type: "64-bit UVADDR"
Ah, damn it -- it would also be printed later on in
readmem() if uvtop() fails:
while (size > 0) {
switch (memtype)
{
case UVADDR:
if (!uvtop(CURRENT_CONTEXT(), addr, &paddr,
0)) {
if (PRINT_ERROR_MESSAGE)
error(INFO, INVALID_UVADDR,
addr, type);
goto readmem_error;
}
break;
That's probably where it's happening...
I would guess that x86_64_uvtop() needs to immediately
recognize if the new VM scheme is in place, and if so
it should veer off to a new "x86_64_uvtop_lev4()" routine.
Yep. I realized that after sending you e-mail. Now we have
4-level pagetables which include "PUD". I cooked up one,
but its still not working :(
I tried reusing the FILL_PML4 code from x86_64_kvtop()
for x86_64_uvtop() -- it still doesn't work.
I need to spend more time on it to understand it better
before I hack it up blindly.
Thanks,
Badari
10:21 p.m.
New subject: [PATCH} crash 4.0-2.8 fixes to make it workonx86_64 for 2.6.14
Dave,
Here is the latest patch to make crash work on 2.6.14 kernel -
with fixes for 4-level pagetables also.
I changes & naming are kind of hacky (especially with naming)
-- but to cleanup it completely and have backward compatibility
needs more work. We should get rid of PML4 completely and
move to PGD/PUD/PMD/PTE.
Thanks,
Badari
elm3a242:~/crash-4.0-2.8 # ./crash
crash 4.0-2.8
Copyright (C) 2002, 2003, 2004, 2005 Red Hat, Inc.
Copyright (C) 2004, 2005 IBM Corporation
Copyright (C) 1999-2005 Hewlett-Packard Co
Copyright (C) 1999, 2002 Silicon Graphics, Inc.
Copyright (C) 1999, 2000, 2001, 2002 Mission Critical Linux, Inc.
This program is free software, covered by the GNU General Public
License,
and you are welcome to change it and/or distribute copies of it under
certain conditions. Enter "help copying" to see the conditions.
This program has absolutely no warranty. Enter "help warranty" for
details.
GNU gdb 6.1
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you
are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for
details.
This GDB was configured as "x86_64-unknown-linux-gnu"...
KERNEL: /usr/src/linux-2.6.14.madv/vmlinux
DUMPFILE: /dev/mem
CPUS: 2
DATE: Thu Nov 3 08:25:50 2005
UPTIME: 07:12:24
LOAD AVERAGE: 0.60, 0.43, 0.18
TASKS: 62
NODENAME: elm3a242
RELEASE: 2.6.14
VERSION: #3 SMP PREEMPT Wed Nov 2 03:08:36 PST 2005
MACHINE: x86_64 (3000 Mhz)
MEMORY: 4.6 GB
PID: 31320
COMMAND: "crash"
TASK: ffff810122f11860 [THREAD_INFO: ffff8100987f6000]
CPU: 1
STATE: TASK_RUNNING (ACTIVE)
crash> set
PID: 31320
COMMAND: "crash"
TASK: ffff810122f11860 [THREAD_INFO: ffff8100987f6000]
CPU: 1
STATE: TASK_RUNNING (ACTIVE)
crash> vm
PID: 31320 TASK: ffff810122f11860 CPU: 1 COMMAND: "crash"
MM PGD RSS TOTAL_VM
ffff810122e93700 ffff810099584000 148044k 86512k
VMA START END FLAGS FILE
ffff810122438cc0 400000 754000 1875 /root/crash-4.0-2.8/crash
ffff810122438590 853000 8b0000 101873 /root/crash-4.0-2.8/crash
ffff810122438d78 8b0000 1c2e000 100073
ffff810122438088 2aaaaaaab000 2aaaaaac0000 875 /lib64/ld-2.3.3.so
ffff8101224389e0 2aaaaaac0000 2aaaaaac2000 100073
ffff810123caa4d8 2aaaaaac2000 2aaaaaaf5000
71 /usr/lib/locale/en_US.utf8/LC_CTYPE
ffff810123caaee8 2aaaaaaf5000 2aaaaaafb000
d1 /usr/lib64/gconv/gconv-modules.cache
ffff8101136809e0 2aaaaaafb000 2aaaaaaff000 100073
ffff8101224387b8 2aaaaabc0000 2aaaaabc2000 100873 /lib64/ld-2.3.3.so
ffff810122438648 2aaaaabc2000 2aaaaac16000 75 /lib64/tls/libm.so.6
ffff810122438700 2aaaaac16000 2aaaaacc2000 70 /lib64/tls/libm.so.6
ffff8101276f4590 2aaaaacc2000 2aaaaad19000 100073 /lib64/tls/libm.so.6
ffff8101276f4a98 2aaaaad19000 2aaaaad1a000 100073
ffff810120ff1590 2aaaaad1a000 2aaaaad64000
75 /lib64/libncurses.so.5.4
ffff8101276f4b50 2aaaaad64000 2aaaaae1a000
70 /lib64/libncurses.so.5.4
ffff810120ff14d8 2aaaaae1a000 2aaaaae7a000
100073 /lib64/libncurses.so.5.4
ffff810120ff1140 2aaaaae7a000 2aaaaae7b000 100073
crash> rd -u 0x400000
400000: 00010102464c457f .ELF....
crash> rd -u 2aaaaaaab000
2aaaaaaab000: 00010102464c457f .ELF....
crash> mod
MODULE NAME SIZE OBJECT FILE
ffffffff88012600 dm_mod 70872 (not loaded) [CONFIG_KALLSYMS]
ffffffff8805ef80 ipv6 312832 (not loaded) [CONFIG_KALLSYMS]
ffffffff8806e200 parport 47244 (not loaded) [CONFIG_KALLSYMS]
ffffffff88073800 lp 17232 (not loaded) [CONFIG_KALLSYMS]
ffffffff8807fd00 parport_pc 33896 (not loaded) [CONFIG_KALLSYMS]
ffffffff8808ab00 usbserial 39280 (not loaded) [CONFIG_KALLSYMS]
ffffffff8808e900 hw_random 7968 (not loaded) [CONFIG_KALLSYMS]
ffffffff88098f80 uhci_hcd 38304 (not loaded) [CONFIG_KALLSYMS]
ffffffff880a4600 ehci_hcd 39944 (not loaded) [CONFIG_KALLSYMS]
ffffffff880acd80 i2c_core 29568 (not loaded) [CONFIG_KALLSYMS]
ffffffff880b1700 i2c_i801 11540 (not loaded) [CONFIG_KALLSYMS]
ffffffff880b6000 joydev 13952 (not loaded) [CONFIG_KALLSYMS]
ffffffff880bb000 edd 13984 (not loaded) [CONFIG_KALLSYMS]
10:29 p.m.
New subject: [PATCH} crash 4.0-2.8 fixes to make itworkonx86_64 for 2.6.14
Badari Pulavarty wrote:
Dave,
Here is the latest patch to make crash work on 2.6.14 kernel -
with fixes for 4-level pagetables also.
I changes & naming are kind of hacky (especially with naming)
-- but to cleanup it completely and have backward compatibility
needs more work. We should get rid of PML4 completely and
move to PGD/PUD/PMD/PTE.
Thanks,
Badari
That's some nice work you've done here -- I can't thank you enough!
But are you saying that with this patch, it does *not* have backwards-compatibility?
It looks like you've completely segregated the new code to me.
Dave
elm3a242:~/crash-4.0-2.8 # ./crash
crash 4.0-2.8
Copyright (C) 2002, 2003, 2004, 2005 Red Hat, Inc.
Copyright (C) 2004, 2005 IBM Corporation
Copyright (C) 1999-2005 Hewlett-Packard Co
Copyright (C) 1999, 2002 Silicon Graphics, Inc.
Copyright (C) 1999, 2000, 2001, 2002 Mission Critical Linux, Inc.
This program is free software, covered by the GNU General Public
License,
and you are welcome to change it and/or distribute copies of it under
certain conditions. Enter "help copying" to see the conditions.
This program has absolutely no warranty. Enter "help warranty" for
details.
GNU gdb 6.1
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you
are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for
details.
This GDB was configured as "x86_64-unknown-linux-gnu"...
KERNEL: /usr/src/linux-2.6.14.madv/vmlinux
DUMPFILE: /dev/mem
CPUS: 2
DATE: Thu Nov 3 08:25:50 2005
UPTIME: 07:12:24
LOAD AVERAGE: 0.60, 0.43, 0.18
TASKS: 62
NODENAME: elm3a242
RELEASE: 2.6.14
VERSION: #3 SMP PREEMPT Wed Nov 2 03:08:36 PST 2005
MACHINE: x86_64 (3000 Mhz)
MEMORY: 4.6 GB
PID: 31320
COMMAND: "crash"
TASK: ffff810122f11860 [THREAD_INFO: ffff8100987f6000]
CPU: 1
STATE: TASK_RUNNING (ACTIVE)
crash> set
PID: 31320
COMMAND: "crash"
TASK: ffff810122f11860 [THREAD_INFO: ffff8100987f6000]
CPU: 1
STATE: TASK_RUNNING (ACTIVE)
crash> vm
PID: 31320 TASK: ffff810122f11860 CPU: 1 COMMAND: "crash"
MM PGD RSS TOTAL_VM
ffff810122e93700 ffff810099584000 148044k 86512k
VMA START END FLAGS FILE
ffff810122438cc0 400000 754000 1875 /root/crash-4.0-2.8/crash
ffff810122438590 853000 8b0000 101873 /root/crash-4.0-2.8/crash
ffff810122438d78 8b0000 1c2e000 100073
ffff810122438088 2aaaaaaab000 2aaaaaac0000 875 /lib64/ld-2.3.3.so
ffff8101224389e0 2aaaaaac0000 2aaaaaac2000 100073
ffff810123caa4d8 2aaaaaac2000 2aaaaaaf5000
71 /usr/lib/locale/en_US.utf8/LC_CTYPE
ffff810123caaee8 2aaaaaaf5000 2aaaaaafb000
d1 /usr/lib64/gconv/gconv-modules.cache
ffff8101136809e0 2aaaaaafb000 2aaaaaaff000 100073
ffff8101224387b8 2aaaaabc0000 2aaaaabc2000 100873 /lib64/ld-2.3.3.so
ffff810122438648 2aaaaabc2000 2aaaaac16000 75 /lib64/tls/libm.so.6
ffff810122438700 2aaaaac16000 2aaaaacc2000 70 /lib64/tls/libm.so.6
ffff8101276f4590 2aaaaacc2000 2aaaaad19000 100073 /lib64/tls/libm.so.6
ffff8101276f4a98 2aaaaad19000 2aaaaad1a000 100073
ffff810120ff1590 2aaaaad1a000 2aaaaad64000
75 /lib64/libncurses.so.5.4
ffff8101276f4b50 2aaaaad64000 2aaaaae1a000
70 /lib64/libncurses.so.5.4
ffff810120ff14d8 2aaaaae1a000 2aaaaae7a000
100073 /lib64/libncurses.so.5.4
ffff810120ff1140 2aaaaae7a000 2aaaaae7b000 100073
crash> rd -u 0x400000
400000: 00010102464c457f .ELF....
crash> rd -u 2aaaaaaab000
2aaaaaaab000: 00010102464c457f .ELF....
crash> mod
MODULE NAME SIZE OBJECT FILE
ffffffff88012600 dm_mod 70872 (not loaded) [CONFIG_KALLSYMS]
ffffffff8805ef80 ipv6 312832 (not loaded) [CONFIG_KALLSYMS]
ffffffff8806e200 parport 47244 (not loaded) [CONFIG_KALLSYMS]
ffffffff88073800 lp 17232 (not loaded) [CONFIG_KALLSYMS]
ffffffff8807fd00 parport_pc 33896 (not loaded) [CONFIG_KALLSYMS]
ffffffff8808ab00 usbserial 39280 (not loaded) [CONFIG_KALLSYMS]
ffffffff8808e900 hw_random 7968 (not loaded) [CONFIG_KALLSYMS]
ffffffff88098f80 uhci_hcd 38304 (not loaded) [CONFIG_KALLSYMS]
ffffffff880a4600 ehci_hcd 39944 (not loaded) [CONFIG_KALLSYMS]
ffffffff880acd80 i2c_core 29568 (not loaded) [CONFIG_KALLSYMS]
ffffffff880b1700 i2c_i801 11540 (not loaded) [CONFIG_KALLSYMS]
ffffffff880b6000 joydev 13952 (not loaded) [CONFIG_KALLSYMS]
ffffffff880bb000 edd 13984 (not loaded) [CONFIG_KALLSYMS]
------------------------------------------------------------------------------------------------------------------------
Name: x86_64-crash-fixes2.patch
x86_64-crash-fixes2.patch Type: text/x-patch
Encoding: 7bit
------------------------------------------------------------------------------------------------------------------------
--
Crash-utility mailing list
Crash-utility(a)redhat.com
https://www.redhat.com/mailman/listinfo/crash-utility
10:40 p.m.
New subject: [PATCH} crash 4.0-2.8 fixes to make itworkonx86_64 for 2.6.14
On Thu, 2005-11-03 at 16:29 -0500, Dave Anderson wrote:
Badari Pulavarty wrote:
> Dave,
>
> Here is the latest patch to make crash work on 2.6.14 kernel -
> with fixes for 4-level pagetables also.
>
> I changes & naming are kind of hacky (especially with naming)
> -- but to cleanup it completely and have backward compatibility
> needs more work. We should get rid of PML4 completely and
> move to PGD/PUD/PMD/PTE.
>
> Thanks,
> Badari
>
That's some nice work you've done here -- I can't thank you enough!
But are you saying that with this patch, it does *not* have backwards-compatibility?
It looks like you've completely segregated the new code to me.
This patch does provide backward compatibility (I tested with 2.6.10 and
2.6.14).
My complaint is the naming convention, for longterm maintenance:
For user virtual - before 2.6.10 we have only 3-level pagetables
but now we have 4-level page tables. What used to be PGD (before
2.6.10) now became PUD and so on..
2.6.10 2.6.14
no 4 level == PGD
PGD == PUD
PMD == PMD
PTE == PTE
Thanks,
Badari
11:30 p.m.
New subject: [PATCH} crash 4.0-2.8 fixes to makeitworkonx86_64 for 2.6.14
Hi Badari,
When you can spare a few cycles, can you sanity check my
adaptation of your patch? If you go to:
http://people.redhat.com/anderson/pbadari
You'll find a "crash.tar.gz" file. Go into any recent crash build
directory, un-tar it on top of whatever's there, and do a make.
(It's quicker that rebuilding the whole thing...)
It works find on RHEL3 and RHEL4, but I just want make
absolutely sure it still works OK on your test kernel. I didn't
change any functionality, but I did move a few things around.
And many thanks again...
Dave
11:54 p.m.
New subject: [PATCH} crash 4.0-2.8 fixes to makeitworkonx86_64 for 2.6.14
On Fri, 2005-11-04 at 17:30 -0500, Dave Anderson wrote:
Hi Badari,
When you can spare a few cycles, can you sanity check my
adaptation of your patch? If you go to:
http://people.redhat.com/anderson/pbadari
You'll find a "crash.tar.gz" file. Go into any recent crash build
directory, un-tar it on top of whatever's there, and do a make.
(It's quicker that rebuilding the whole thing...)
It works find on RHEL3 and RHEL4, but I just want make
absolutely sure it still works OK on your test kernel. I didn't
change any functionality, but I did move a few things around.
And many thanks again...
It seems to work fine on 2.6.10 & 2.6.14 :)
BTW, I am looking at PPC64 4-level page table changes - its
even crazier. All PGD/PMD shifts & masks have changed :(
Thanks,
Badari
3:16 p.m.
New subject: [PATCH} crash 4.0-2.8 fixes tomakeitworkonx86_64 for 2.6.14
Badari Pulavarty wrote:
On Fri, 2005-11-04 at 17:30 -0500, Dave Anderson wrote:
> Hi Badari,
>
> When you can spare a few cycles, can you sanity check my
> adaptation of your patch? If you go to:
>
> http://people.redhat.com/anderson/pbadari
>
> You'll find a "crash.tar.gz" file. Go into any recent crash build
> directory, un-tar it on top of whatever's there, and do a make.
> (It's quicker that rebuilding the whole thing...)
>
> It works find on RHEL3 and RHEL4, but I just want make
> absolutely sure it still works OK on your test kernel. I didn't
> change any functionality, but I did move a few things around.
>
> And many thanks again...
>
It seems to work fine on 2.6.10 & 2.6.14 :)
BTW, I am looking at PPC64 4-level page table changes - its
even crazier. All PGD/PMD shifts & masks have changed :(
Thanks,
Badari
Thanks...
Re: ppc64, Haren mentioned that in our kexec/kdump conference call the
other day. At least now there's a "template" in place that you guys
can use to implement it.
Dave
6891
days inactive
6896
days old
devel@lists.crash-utility.osci.io
7 comments
2 participants
participants (2)
-
Badari Pulavarty -
Dave Anderson