From: Jim Wilson <wilson@cygnus.com>
To: binutils@sourceware.cygnus.com
Cc: wilson@cygnus.com, rth@cygnus.com
Subject: patch for gas/dwarf2dbg.c, core dump and corrupted .debug_line section
Date: Wed, 25 Aug 1999 20:43:00 -0000 [thread overview]
Message-ID: <199908260343.UAA26025@ada.cygnus.com.cygnus.com> (raw)
This fixes 4 problems I noticed while trying to use this code from gcc.
1) The comments in get_filenum do not match the code. The code is correct.
2) get_filenum performs an out-of-bounds array access if filenum is 0, which
makes last == -1.
3) get_filenum does not update ls.last_filename when creating a new filetable
entry.
4) dwarf2_gen_line_info saves seg/subseg info after it has already been
clobbered by a subseg_new call. This causes gas to accidentally emit code
into the .debug_line section.
Wed Aug 25 20:20:33 1999 Jim Wilson <wilson@cygnus.com>
* dwarf2dbg.c (out_end_sequence): Correct comments. Set last to
ls.last_filename if last is less than zero. Set ls.last_filename
when allocating new entry.
(dwarf2_gen_line_info): Save seg and subseg info before subseg_new
call.
Index: dwarf2dbg.c
===================================================================
RCS file: /cvs/cvsfiles/devo/gas/dwarf2dbg.c,v
retrieving revision 1.7
diff -p -r1.7 dwarf2dbg.c
*** dwarf2dbg.c 1999/07/23 22:12:34 1.7
--- dwarf2dbg.c 1999/08/26 03:34:48
*************** out_end_sequence ()
*** 359,365 ****
/* Look up a filenumber either by filename or by filenumber. If both
a filenumber and a filename are specified, lookup by filename takes
precedence. If the filename cannot be found, it is added to the
! filetable the filenumber for the new entry is returned. */
static int
get_filenum (filenum, file)
int filenum;
--- 359,365 ----
/* Look up a filenumber either by filename or by filenumber. If both
a filenumber and a filename are specified, lookup by filename takes
precedence. If the filename cannot be found, it is added to the
! filetable and the filenumber for the new entry is returned. */
static int
get_filenum (filenum, file)
int filenum;
*************** get_filenum (filenum, file)
*** 368,377 ****
int i, last = filenum - 1;
char char0 = file[0];
! if (last >= ls.num_filenames)
last = ls.last_filename;
! /* do a quick check against the previously used filename: */
if (ls.num_filenames > 0 && ls.file[last].name[0] == char0
&& strcmp (ls.file[last].name + 1, file + 1) == 0)
return last + 1;
--- 368,379 ----
int i, last = filenum - 1;
char char0 = file[0];
! /* If filenum is out of range of the filename table, then try using the
! table entry returned from the previous call. */
! if (last >= ls.num_filenames || last < 0)
last = ls.last_filename;
! /* Do a quick check against the specified or previously used filenum. */
if (ls.num_filenames > 0 && ls.file[last].name[0] == char0
&& strcmp (ls.file[last].name + 1, file + 1) == 0)
return last + 1;
*************** get_filenum (filenum, file)
*** 395,400 ****
--- 397,403 ----
}
ls.file[ls.num_filenames].dir = 0;
ls.file[ls.num_filenames].name = file;
+ ls.last_filename = ls.num_filenames;
return ++ls.num_filenames;
}
*************** dwarf2_gen_line_info (addr, l)
*** 425,430 ****
--- 428,438 ----
else
return; /* no filename, no filnum => no play */
+ /* Must save these before the subseg_new call, as that call will change
+ them. */
+ saved_seg = now_seg;
+ saved_subseg = now_subseg;
+
if (!ls.line_seg)
{
#ifdef BFD_ASSEMBLER
*************** dwarf2_gen_line_info (addr, l)
*** 445,452 ****
#endif
}
- saved_seg = now_seg;
- saved_subseg = now_subseg;
subseg_set (ls.line_seg, DL_BODY);
if (ls.text_seg != saved_seg || ls.text_subseg != saved_subseg)
--- 453,458 ----
next reply other threads:[~1999-08-25 20:43 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
1999-08-25 20:43 Jim Wilson [this message]
1999-08-26 5:18 Nick Clifton
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=199908260343.UAA26025@ada.cygnus.com.cygnus.com \
--to=wilson@cygnus.com \
--cc=binutils@sourceware.cygnus.com \
--cc=rth@cygnus.com \
/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).