From: Alan Modra <amodra@gmail.com>
To: Hannes Domani <ssbssa@yahoo.de>
Cc: "binutils@sourceware.org" <binutils@sourceware.org>
Subject: Re: PE objdump -x
Date: Wed, 3 Aug 2022 16:06:47 +0930 [thread overview]
Message-ID: <YuoXf9qmXGFSMlAg@squeak.grove.modra.org> (raw)
In-Reply-To: <45264354.7432971.1659437212970@mail.yahoo.com>
On Tue, Aug 02, 2022 at 10:46:52AM +0000, Hannes Domani wrote:
> >
> > - if (ex_dta >= ex_dta_end || ex_dta + 4 >= ex_dta_end)
> > + if (ex_dta >= ex_dta_end || ex_dta + 4 > ex_dta_end)
> > return false;
>
> Are here really both if-conditions necessary?
No, I was lazy in not cleaning that up. All of these tests are better
written as a comparison against size remaining, due to ISO C 9899
standard 6.5.2 para 8 regarding adding a constant to a pointer:
"If both the pointer operand and the result point to elements of the
same array object, or one past the last element of the array object,
the evaluation shall not produce an overflow; otherwise, the behavior
is undefined."
So "ex_dta + 4" might be undefined behaviour, if you interpret "the
array object" in this case to be the malloc'd section contents! This
sort of thing is quite a surprise to anyone used to the nice flat
memory models of current machines, and even with some of the weird old
machines the part of the spec I'm quoting likely doesn't apply to
malloc'd memory. But it's easy to avoid the potential UB.
* pei-x86_64.c (pex64_get_unwind_info): Tidy sanity checks.
(pex64_xdata_print_uwd_codes): Likewise.
diff --git a/bfd/pei-x86_64.c b/bfd/pei-x86_64.c
index 795bf66f8b4..9d0ff81ec4b 100644
--- a/bfd/pei-x86_64.c
+++ b/bfd/pei-x86_64.c
@@ -109,7 +109,7 @@ pex64_get_unwind_info (bfd *abfd, struct pex64_unwind_info *ui,
memset (ui, 0, sizeof (struct pex64_unwind_info));
- if (ex_dta >= ex_dta_end || ex_dta + 4 > ex_dta_end)
+ if (ex_dta_end - ex_dta < 4)
return false;
ui->Version = PEX64_UWI_VERSION (ex_ui->Version_Flags);
@@ -123,14 +123,14 @@ pex64_get_unwind_info (bfd *abfd, struct pex64_unwind_info *ui,
ui->rawUnwindCodes = ex_dta + 4;
ui->rawUnwindCodesEnd = ex_dta_end;
- ex_dta += ui->SizeOfBlock;
- if (ex_dta > ex_dta_end)
+ if ((size_t) (ex_dta_end - ex_dta) < ui->SizeOfBlock)
return false;
+ ex_dta += ui->SizeOfBlock;
switch (ui->Flags)
{
case UNW_FLAG_CHAININFO:
- if (ex_dta + 12 > ex_dta_end)
+ if (ex_dta_end - ex_dta < 12)
return false;
ui->rva_BeginAddress = bfd_get_32 (abfd, ex_dta + 0);
ui->rva_EndAddress = bfd_get_32 (abfd, ex_dta + 4);
@@ -140,7 +140,7 @@ pex64_get_unwind_info (bfd *abfd, struct pex64_unwind_info *ui,
case UNW_FLAG_EHANDLER:
case UNW_FLAG_UHANDLER:
case UNW_FLAG_FHANDLER:
- if (ex_dta + 4 > ex_dta_end)
+ if (ex_dta_end - ex_dta < 4)
return false;
ui->rva_ExceptionHandler = bfd_get_32 (abfd, ex_dta);
ui->SizeOfBlock += 4;
@@ -172,7 +172,8 @@ pex64_xdata_print_uwd_codes (FILE *file, bfd *abfd,
i = 0;
- if (ui->rawUnwindCodes + ui->CountOfCodes * 2 > ui->rawUnwindCodesEnd)
+ if ((size_t) (ui->rawUnwindCodesEnd - ui->rawUnwindCodes)
+ < ui->CountOfCodes * 2)
{
fprintf (file, _("warning: corrupt unwind data\n"));
return;
@@ -226,7 +227,7 @@ pex64_xdata_print_uwd_codes (FILE *file, bfd *abfd,
case UWOP_ALLOC_LARGE:
if (info == 0)
{
- if (dta + 4 > ui->rawUnwindCodesEnd)
+ if (ui->rawUnwindCodesEnd - dta < 4)
{
fprintf (file, _("warning: corrupt unwind data\n"));
return;
@@ -236,7 +237,7 @@ pex64_xdata_print_uwd_codes (FILE *file, bfd *abfd,
}
else
{
- if (dta + 6 > ui->rawUnwindCodesEnd)
+ if (ui->rawUnwindCodesEnd - dta < 6)
{
fprintf (file, _("warning: corrupt unwind data\n"));
return;
@@ -261,7 +262,7 @@ pex64_xdata_print_uwd_codes (FILE *file, bfd *abfd,
break;
case UWOP_SAVE_NONVOL:
- if (dta + 4 > ui->rawUnwindCodesEnd)
+ if (ui->rawUnwindCodesEnd - dta < 4)
{
fprintf (file, _("warning: corrupt unwind data\n"));
return;
@@ -273,7 +274,7 @@ pex64_xdata_print_uwd_codes (FILE *file, bfd *abfd,
break;
case UWOP_SAVE_NONVOL_FAR:
- if (dta + 6 > ui->rawUnwindCodesEnd)
+ if (ui->rawUnwindCodesEnd - dta < 6)
{
fprintf (file, _("warning: corrupt unwind data\n"));
return;
@@ -287,7 +288,7 @@ pex64_xdata_print_uwd_codes (FILE *file, bfd *abfd,
case UWOP_SAVE_XMM:
if (ui->Version == 1)
{
- if (dta + 4 > ui->rawUnwindCodesEnd)
+ if (ui->rawUnwindCodesEnd - dta < 4)
{
fprintf (file, _("warning: corrupt unwind data\n"));
return;
@@ -305,7 +306,7 @@ pex64_xdata_print_uwd_codes (FILE *file, bfd *abfd,
break;
case UWOP_SAVE_XMM_FAR:
- if (dta + 6 > ui->rawUnwindCodesEnd)
+ if (ui->rawUnwindCodesEnd - dta < 6)
{
fprintf (file, _("warning: corrupt unwind data\n"));
return;
@@ -317,7 +318,7 @@ pex64_xdata_print_uwd_codes (FILE *file, bfd *abfd,
break;
case UWOP_SAVE_XMM128:
- if (dta + 4 > ui->rawUnwindCodesEnd)
+ if (ui->rawUnwindCodesEnd - dta < 4)
{
fprintf (file, _("warning: corrupt unwind data\n"));
return;
@@ -329,7 +330,7 @@ pex64_xdata_print_uwd_codes (FILE *file, bfd *abfd,
break;
case UWOP_SAVE_XMM128_FAR:
- if (dta + 6 > ui->rawUnwindCodesEnd)
+ if (ui->rawUnwindCodesEnd - dta < 6)
{
fprintf (file, _("warning: corrupt unwind data\n"));
return;
--
Alan Modra
Australia Development Lab, IBM
prev parent reply other threads:[~2022-08-03 6:36 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-02 7:46 Alan Modra
2022-08-02 10:46 ` Hannes Domani
2022-08-03 6:36 ` Alan Modra [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=YuoXf9qmXGFSMlAg@squeak.grove.modra.org \
--to=amodra@gmail.com \
--cc=binutils@sourceware.org \
--cc=ssbssa@yahoo.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).