[Crash-utility] Re: [PATCH] x86_64: check bt->bptr before calculate framesize

Date: Tue, 26 Dec 2023 06:42:55 +0000 From: HAGIO KAZUHITO(萩尾　一仁)&lt;k-hagio-ab(a)nec.com&gt; Subject: [Crash-utility] Re: [PATCH] x86_64: check bt->bptr before calculate framesize To: Tao Liu&lt;ltao@redhat.com&gt;,&quot;devel(a)lists.crash-utility.osci.io&quot; <devel(a)lists.crash-utility.osci.io&gt; Message-ID:<58d63b19-0d8a-3be0-234a-703a4c25dcb9@nec.com> Content-Type: text/plain; charset="utf-8" On 2023/12/26 10:19, Tao Liu wrote: > Previously the value of bt->bptr is not checked, which may led to a > wrong prev_sp and framesize. As a result, bt->stackbuf[] will be > accessed out of range, and segfault. > > Before: > crash> set debug 1 > crash> bt > ...snip... > --- <NMI exception stack> --- > #8 [ffffffff9a603e10] __switch_to_asm at ffffffff99800214 > rsp: ffffffff9a603e10 textaddr: ffffffff99800214 -> spo: 0 bpo: 0 spr: 0 bpr: 0 type: 0 end: 0 > #9 [ffffffff9a603e40] __schedule at ffffffff9960dfb1 > rsp: ffffffff9a603e40 textaddr: ffffffff9960dfb1 -> spo: 16 bpo: -16 spr: 4 bpr: 1 type: 0 end: 0 > rsp: ffffffff9a603e40 rbp: ffffb9ca076e7ca8 prev_sp: ffffb9ca076e7cb8 framesize: 1829650024 > Segmentation fault (core dumped) > > (gdb) p/x bt->stackbase > $1 = 0xffffffff9a600000 > (gdb) p/x bt->stacktop > $2 = 0xffffffff9a604000 > > After: > crash> set debug 1 > crash> bt > ...snip... > --- <NMI exception stack> --- > #8 [ffffffff9a603e10] __switch_to_asm at ffffffff99800214 > rsp: ffffffff9a603e10 textaddr: ffffffff99800214 -> spo: 0 bpo: 0 spr: 0 bpr: 0 type: 0 end: 0 > #9 [ffffffff9a603e40] __schedule at ffffffff9960dfb1 > rsp: ffffffff9a603e40 textaddr: ffffffff9960dfb1 -> spo: 16 bpo: -16 spr: 4 bpr: 1 type: 0 end: 0 > #10 [ffffffff9a603e98] schedule_idle at ffffffff9960e87c > rsp: ffffffff9a603e98 textaddr: ffffffff9960e87c -> spo: 8 bpo: 0 spr: 5 bpr: 0 type: 0 end: 0 > rsp: ffffffff9a603e98 prev_sp: ffffffff9a603ea8 framesize: 0 > ...snip... > > This patch will check bt->bptr value before calculate framesize. Only bt->bptr > falls into the range of bt->stackbase and bt->stacktop will be > regarded as valid. > > Signed-off-by: Tao Liu&lt;ltao(a)redhat.com&gt; > --- > x86_64.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/x86_64.c b/x86_64.c > index 8abe39d..ff1ba6e 100644 > --- a/x86_64.c > +++ b/x86_64.c > @@ -8707,7 +8707,8 @@ x86_64_get_framesize(struct bt_info *bt, ulong textaddr, ulong rsp, char *stack_ > if (CRASHDEBUG(1)) > fprintf(fp, "rsp: %lx prev_sp: %lx framesize: %d\n", > rsp, prev_sp, framesize); > - } else if ((korc->sp_reg == ORC_REG_BP) && bt->bptr) { > + } else if ((korc->sp_reg == ORC_REG_BP) && bt->bptr && > + bt->bptr >= bt->stackbase && bt->bptr <= bt->stacktop) { Thank you for looking into this! Looks good to me. Just a nit, INSTACK() can be used?

Thanks, Kazu > prev_sp = bt->bptr + korc->sp_offset; > framesize = (prev_sp - (rsp + 8) - 8); > if (CRASHDEBUG(1))