From: Thiago Jung Bauermann <bauerman@br.ibm.com>
To: gcc-patches <gcc-patches@gcc.gnu.org>
Cc: gdb-patches <gdb-patches@sourceware.org>
Subject: [libiberty] don't demangle functions named "."
Date: Wed, 09 Jan 2008 17:15:00 -0000 [thread overview]
Message-ID: <1199898911.15225.48.camel@localhost.localdomain> (raw)
[-- Attachment #1: Type: text/plain, Size: 1888 bytes --]
Hi folks,
This patch is to stop the C++ demangler from trying to demangle a symbol
which is not a C++ mangled function.
The problem is happening in Linux/ppc64 with symbols marking the start
of function instructions in the text section, which begin with a dot.
When the function name starts with __, it's symbol will be something
like .__<function name>. If the demangler is able to derive a valid
argument list from <function name>, it will find a demangled "function"
called . with some random argument list. This is happening with several
functions in glibc:
.__frexpf -> .(float, long double,...)(long long, float *)
.__flbf -> .(float, long, bool, float)
.__ffs -> .(float, float, short)
.__ffsll -> .(float, float, short, long, long)
.__sleep -> .(short, long,...)(...)( *)
.__execve -> .(...)(long long,...)(char, void,...)
.__dup -> .(double, *__restrict)
.__fixdfdi -> .(float, int, long long, double, float, double, int)
.__fixsfdi -> .(float, int, long long, short, float, double, int)
This is a problem in GDB, for instance when showing a backtrace which
goes through these functions in glibc:
(gdb) bt
#0 0x000004000015edf8 in .__libc_nanosleep () from /lib64/tls/libc.so.6
#1 0x000004000015eb74 in . () from /lib64/tls/libc.so.6
#2 0x0000000010000784 in sleeping_thread (p=0x0) at libc-nanosleep.c:8
(gdb) info symbol 0x000004000015eb74
.(short, long,...)(...)( *) + 148 in section .text
With the patch applied:
(gdb) bt
#0 0x000004000015edf8 in .__libc_nanosleep () from /lib64/tls/libc.so.6
#1 0x000004000015eb74 in .__sleep () from /lib64/tls/libc.so.6
#2 0x0000000010000784 in sleeping_thread (p=0x0) at libc-nanosleep.c:8
(gdb) info symbol 0x000004000015eb74
.__sleep + 148 in section .text
All the libiberty and GDB tests related to demangling continue to pass.
Is this ok?
--
[]'s
Thiago Jung Bauermann
Software Engineer
IBM Linux Technology Center
[-- Attachment #2: demangle-fix.diff --]
[-- Type: text/x-patch, Size: 2703 bytes --]
2008-01-09 Thiago Jung Bauermann <bauerman@br.ibm.com>
* cplus-dem.c (demangle_function_name): Changed to return value
indicating if a name was correctly demangled.
(iterate_demangle_function): Use demangle_function_name return value.
Index: src-git.demangle-fix/libiberty/cplus-dem.c
===================================================================
--- src-git.demangle-fix.orig/libiberty/cplus-dem.c 2008-01-08 15:48:12.000000000 -0200
+++ src-git.demangle-fix/libiberty/cplus-dem.c 2008-01-08 15:48:54.000000000 -0200
@@ -414,7 +414,7 @@ static int do_type (struct work_stuff *,
static int do_arg (struct work_stuff *, const char **, string *);
-static void
+static int
demangle_function_name (struct work_stuff *, const char **, string *,
const char *);
@@ -2493,10 +2493,7 @@ iterate_demangle_function (struct work_s
"__"-sequence. This is the normal case. */
if (ARM_DEMANGLING || LUCID_DEMANGLING || HP_DEMANGLING || EDG_DEMANGLING
|| strstr (scan + 2, "__") == NULL)
- {
- demangle_function_name (work, mangled, declp, scan);
- return 1;
- }
+ return demangle_function_name (work, mangled, declp, scan);
/* Save state so we can restart if the guess at the correct "__" was
wrong. */
@@ -2513,10 +2510,12 @@ iterate_demangle_function (struct work_s
while (scan[2])
{
- demangle_function_name (work, mangled, declp, scan);
- success = demangle_signature (work, mangled, declp);
- if (success)
- break;
+ if (demangle_function_name (work, mangled, declp, scan))
+ {
+ success = demangle_signature (work, mangled, declp);
+ if (success)
+ break;
+ }
/* Reset demangle state for the next round. */
*mangled = mangle_init;
@@ -4421,7 +4420,9 @@ demangle_nested_args (struct work_stuff
return result;
}
-static void
+/* Returns 1 if a valid function name was found or 0 otherwise. */
+
+static int
demangle_function_name (struct work_stuff *work, const char **mangled,
string *declp, const char *scan)
{
@@ -4461,13 +4462,13 @@ demangle_function_name (struct work_stuf
{
work -> constructor += 1;
string_clear (declp);
- return;
+ return 1;
}
else if (strcmp (declp -> b, "__dt") == 0)
{
work -> destructor += 1;
string_clear (declp);
- return;
+ return 1;
}
}
@@ -4575,6 +4576,13 @@ demangle_function_name (struct work_stuf
}
}
}
+
+ /* If a function name was obtained but it's not valid, we were not
+ successful. */
+ if (LEN_STRING (declp) == 1 && declp->b[0] == '.')
+ return 0;
+ else
+ return 1;
}
/* a mini string-handling package */
next reply other threads:[~2008-01-09 17:15 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-09 17:15 Thiago Jung Bauermann [this message]
2008-01-09 19:26 ` Jim Blandy
2008-01-09 19:55 ` Thiago Jung Bauermann
2008-01-19 3:02 ` Ian Lance Taylor
2008-01-23 1:05 ` Thiago Jung Bauermann
2008-01-23 5:03 ` Ben Elliston
2008-01-23 16:58 ` Thiago Jung Bauermann
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=1199898911.15225.48.camel@localhost.localdomain \
--to=bauerman@br.ibm.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=gdb-patches@sourceware.org \
/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