On Sat, Mar 11, 2017 at 12:29:57AM +0100, Juan Manuel Guerrero wrote:
> Certainly not the most important fix but here a small patch to pacify
> the compiler by removing the comma at the end of enumerator list.
Which compiler would that be?
--
Alan Modra
Australia Development La
On Tue, Apr 11, 2017 at 10:48:06AM +0200, Sébastien Dusuel wrote:
> I seem to have found a bug in GNU as, when using the intel syntax on x86_64,
Indeed you have. I have opened
https://sourceware.org/bugzilla/show_bug.cgi?id=21373
to track this issue.
--
Alan Modra
Australia Development
t;
> Any idea what could be causing this? Let me know if you need any more
> information from me.
Most likely the test needs to be linked against a library that your
compiler does not automatically add when linking.
--
Alan Modra
Australia Development Lab, IBM
___
s:
>
> 0018 0018 001c FDE cie=
> pc=00012520..c00f252125200400
> Augmentation data: 00
>
> DW_CFA_nop
> DW_CFA_nop
I could not reproduce anything like this using binutils-2
ze_setx being converted in
tc_gen_reloc.
That's not the only problem. The relocs emitted in synthetize_setx
will cause text relocations in a dynamic library or PIE. So it looks
to me that setx with a non-constant value when PIC is simply
unsupported. I'd encourage you to post a patch to
binut...@sourceware.org fixing the problem, or at least emitting an
error rather than bad code.
--
Alan Modra
Australia Development Lab, IBM
mbols in
order to apply relocations. If a symbol with a section of SHN_COMMON
is seen, a per-file COMMON section will be created. Hmm, I suppose
you could argue that since this is done for the linker, there is no
need to do so for simple_get_relocated_section_contents.
--
Alan Modra
Australia Development Lab, IBM
On Fri, Sep 30, 2022 at 09:52:41PM +0800, bjchan9an wrote:
> There is an uncontrolled stack recursion vulnerability in binutils-2.38
Sorry, fuzzing reports against anything but current mainline will be
ignored.
--
Alan Modra
Australia Development Lab, IBM
124013.html
--
Alan Modra
Australia Development Lab, IBM
INFOA" and similarly for
the other output sections in your extra script. By specifying the
input sections these will match your output section before orphan
sections are handled. This modification should work with older
linkers too.
2) Add INSERT BEFORE .infoA; at the end of your extra script.
--
Alan Modra
Australia Development Lab, IBM
On Tue, Feb 28, 2023 at 01:23:25AM -0800, Yuri wrote:
> On 2/28/23 01:14, Nick Clifton wrote:
> > Hi Yuri,
> >
> > > Where is ENUM_BITFIELD expected to come from?
> >
> > /usr/include/anisdecl.h
>
>
> There's no such file on FreeBSD.
includ
On Tue, Feb 28, 2023 at 09:07:17AM -0800, Yuri wrote:
> On 2/28/23 01:39, Alan Modra wrote:
> > include/ansidecl.h is in the binutils sources.
>
> Ok, I see that include/ansidecl.h is not installed by binutils.
It is installed for me. With default configure and current mainline:
This was fixed with commit 907b52f4ceb6 2022-10-03.
--
Alan Modra
Australia Development Lab, IBM
other files, but only this one trigger this thing.
See the ARM EABI documents. Thumb function symbols are marked by
setting the low bit of their value.
--
Alan Modra
Australia Development Lab, IBM
->dwarf_str_buffer);
>free (file->dwarf_ranges_buffer);
> + free (file->dwarf_rnglists_buffer);
>free (file->dwarf_line_buffer);
>free (file->dwarf_abbrev_buffer);
>free (file->dwarf_info_buffer);
Thanks, I've committed this patch to c
cgi?id=30598 which were fixed
in June 2023.
$ binutils/strings -n 4294967295 /usr/bin/strings
binutils/strings: minimum string length 4294967295 is too big
--
Alan Modra
Australia Development Lab, IBM
On Thu, Apr 11, 2024 at 04:32:31AM +0200, oset wrote:
> Why is 32-bit objdump not able to read 64-bit executable?
It can, if you configure bfd with the targets you require. If you
don't select a 64-bit target the you need to use --enable-64-bit-bfd.
--
Alan Modra
Australia Development Lab, IBM
n configure with
--disable-gdb --disable-sim --disable-readline --disable-libdecnumber
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-binutils
(long) header->e_shstrndx);
+ printf (_(" Section header string table index: %u"),
+ header->e_shstrndx);
if (filedata->section_headers != NULL
&& header->e_shstrndx == (SHN_XINDEX & 0
ease report this bug.
gas shouldn't be reporting an internal error here. Very likely you
have a wrongly quoted string in the input to gas. Please open a bug
at https://sourceware.org/bugzilla/enter_bug.cgi?product=binutils with
the input to gas (.s file) attached.
--
Alan Modra
Austra
On Tue, Feb 19, 2019 at 05:06:08PM +0100, Michael Roitzsch wrote:
> My proposed fix is attached as a patch.
Applied.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
https://lists.gnu.org/mail
ster, please open a bug report at
https://sourceware.org/bugzilla/enter_bug.cgi?product=binutils
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-binutils
fd);
return FALSE;
}
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-binutils
rdup
> /src/llvm-project/compiler-rt/lib/asan/asan_interceptors.cpp:468:3
> #1 0x6e76d0 in get_field
> /src/binutils-gdb/opcodes/microblaze-dis.c:46:11
This is PR25249, fixed with git commit 378fd436405b3.
--
Alan Modra
Australia Development Lab, IBM
e fill expression is evaluated when parsing the script for
the first time, before the symbol table is even created.
--
Alan Modra
Australia Development Lab, IBM
line with git commit
c48acf6f267 but not applied to the 2.34 branch. I'm unsure whether
changes like this should be applied to a release branch, but it seems
likely they should not.
--
Alan Modra
Australia Development Lab, IBM
ystem10error_codeE+0x83
Lose the +0x83, which isn't part of the name and it can
$ c++filt _ZN5boost4asio6detail9scheduler3runERNS_6system10error_codeE
boost::asio::detail::scheduler::run(boost::system::error_code&)
--
Alan Modra
Australia Development Lab, IBM
bfd *exec_bfd);
char * rs6000coff_core_file_failing_command (bfd *abfd);
--
Alan Modra
Australia Development Lab, IBM
> int (64 bit integer) in the host platform (Windows), however, the function
> os_time from the file sim/common/callback.c:414 (path relative to project
> root), takes a long (32 bit integer) as a parameter:
>
> static longos_time (host_callback *p, long *t){
> return wrap (p, time (t));
> }
>
> How can I fix this error?
>
>
> Stepan Klymonchuk
--
Alan Modra
Australia Development Lab, IBM
On Sun, Jul 18, 2021 at 04:57:19PM -0400, Christopher Wellons wrote:
> This fixes a defect introduced in 25162c795. The "uint" type has not
> been explicitly defined here, causing compilation to fail.
Thanks, I pushed this to the 2.37 branch, commit 999566402e.
--
Alan
ivered to anyone.
I see all of these reports are unsigned integer overflow, which is an
odd thing for ubsan to complain about. Overflow of unsigned integers
is not undefined behaviour!
--
Alan Modra
Australia Development Lab, IBM
Not a bug.
../../binutils/readelf.c:5586:28: runtime error: unsigned integer overflow:
18446744073178963944 + 536870912 cannot be represented in type 'unsigned long'
Not a bug.
../../binutils/readelf.c:9312:17: runtime error: unsigned integer overflow:
18446744073709421054 + 4294967299 cannot be represented in type 'unsigned long'
Not a bug.
I'll be committing a few fixes for the real bugs you found.
--
Alan Modra
Australia Development Lab, IBM
On Tue, Aug 17, 2021 at 12:14:46PM +0100, Sudip Mukherjee wrote:
> arch/mips/dec/prom/locore.S:29: Error: opcode not supported on this
> processor: r4600 (mips3) `rfe'
Answered in pr28241.
--
Alan Modra
Australia Development Lab, IBM
I modified
> input-file.c, and I successfully solved the problem.
The problem with this patch is that ungetc is only guaranteed to work
for *one* character. Also, since f_in might be a pipe you can't
rewind.
--
Alan Modra
Australia Development Lab, IBM
ort of binutils build problem, for example you
might have a problem caused by system installed libbfd headers or
shared library. That sort of thing has been reported by other
people in the past.
--
Alan Modra
Australia Development Lab, IBM
generates these prototypes. Thanks for the report and
patch. I've extended your patch to other files too.
https://sourceware.org/pipermail/binutils/2021-November/118441.html
--
Alan Modra
Australia Development Lab, IBM
0d64622696e02
nm now shows
-C, --demangle[=STYLE] Decode mangled/processed symbol names
STYLE can be "none", "auto", "gnu-v3", "java",
"gnat", "dlang", "rust"
--
Alan Modra
Australia Development Lab, IBM
[d|i|l|e|x|h]}
@itemx --unicode=@var{[default|invalid|locale|escape|hex|highlight]}
-Controls the display of UTF-8 encoded mulibyte characters in strings.
+Controls the display of UTF-8 encoded multibyte characters in strings.
The default (@option{--unicode=default}) is to give them no special
treatment, and instead rely upon the setting of the
@option{--encoding} option. The other values for this option
--
Alan Modra
Australia Development Lab, IBM
l the various
> forms which deal with eax on Vol 2A page 2-7.
Nope. Read over the description again. /0 is part of the opcode.
The rest of the modrm byte can select registers as well as memory.
--
Alan Modra
Australia Development Lab, IBM
_
On Mon, Jul 28, 2008 at 10:11:38AM +0400, Ineiev wrote:
> There are few things that break current CVS
> build with gcc-2.95.3 for these targets.
Patch applied. Thanks!
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing li
On Tue, Sep 09, 2008 at 01:27:09PM -0500, [EMAIL PROTECTED] wrote:
> The configure script is supposed to check for makeinfo >= 4.4, but the
> regular
> expression used will fail for makeinfo >= 4.10 and < 5.0.
Thanks. This has already been fixed in CVS sources.
--
Al
s like this has already been fixed on mainline and 2.19.
2008-02-11 Daniel Jacobowitz <[EMAIL PROTECTED]>
* cache.c (cache_bread): Set bfd_error_file_truncated if EOF
was reached.
* srec.c (srec_scan): Calculate the checksum. Complain on mismatch.
--
Alan Modra
REL)
+#define TC_FORCE_RELOCATION_SUB_ABS(FIX, SEG) \
+ ((FIX)->fx_r_type == BFD_RELOC_32_PLT_PCREL \
+ || (!md_register_arithmetic && (SEG) == reg_section))
/* Don't complain when we leave fx_subsy around. */
#undef TC_VALIDATE_FIX_SUB
--
Alan Mod
61: 0 NOTYPE WEAK DEFAULT UND __gmon_start__
$
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
On Fri, Mar 13, 2009 at 02:41:31PM +0530, kapilraj chakuri wrote:
> Can anybody give mee the suggestion using the ld command in IBM powerpc (p5).
Please use gcc to link. gcc will pass the right options to ld.
--
Alan Modra
Australia Development Lab,
p; *s != 0)
+ fatal (_("invalid integer argument %s"), argv[numeric_opt - 1] + 1);
+}
if (string_min < 1)
fatal (_("invalid minimum string length %d"), string_min);
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
p; *s != 0)
+ fatal (_("invalid integer argument %s"), argv[numeric_opt - 1] + 1);
+}
if (string_min < 1)
fatal (_("invalid minimum string length %d"), string_min);
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
ocation on the length field of the CIE.
RELOCATION RECORDS FOR [.eh_frame]:
OFFSET TYPE VALUE
R_CR16_SWITCH32 .LECIE1+0x0018
You should not have a relocation here. The length expression should
be fully resolved in gas.
--
Alan Modra
Australia
012 R_CR16_NUM32a ___gxx_personality_v0
> =
>
> But still the error message displayed by the linker:
> ===
> libgcc.a(unwind-dw2-fde.o)(.eh_frame); no .eh_frame_hdr table will be
> created.
>
>
> Do I miss some thing else?
Run ld under gdb and find out where you branch
result is:
>
> 1000 A _end
> c067f678 A _end2
> c068 A _end3
> c068 A _end4
Possibly some section with a zero vma is being placed before _end.
Generate a link map to see if this is so.
--
Alan Modra
Australia Development Lab, IBM
orphan sections.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
On Thu, Jul 09, 2009 at 02:31:53PM -0500, Edmar Wienskoski-RA8797 wrote:
> Kumar Gala wrote:
>>
>> On Jul 8, 2009, at 11:40 PM, Alan Modra wrote:
>>
>>> On Wed, Jul 08, 2009 at 10:52:59PM -0500, Kumar Gala wrote:
>>>> To further verify this if I s
een
_end3 = . ;
. = _end3;
So dot is restored after the orphan section sets it.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
otes:
>
> binutils-2.19.50.0.1 is based on CVS binutils 2008 1007
> binutils-2.19.51.0.1 is based on CVS binutils 2009 0106
Yes, I already have good reason to suspect this patch
2008-10-22 Alan Modra
* ldlang.c (lang_output_section_find_by_flags): Handle non-alloc
sections.
On Sat, Jul 11, 2009 at 09:35:03AM +0930, Alan Modra wrote:
> Did you try the patch I posted?
/me reads other email. I see you did. Applying.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
h
utils-2.19.1 did not treat an aix6 target as special. This bug has
been fixed in current sources.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
On Fri, Oct 16, 2009 at 10:38:59AM -0700, Danny Sung wrote:
> ../../gas/config/tc-arm.c:2489: warning: empty body in an if-statement
This has now been fixed.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binut
x
"symbol .equ value" but not a different .set syntax.
> Is there any option to make GNU assembler to understand the above syntax for
> powerpc target?
No.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
On Wed, Nov 18, 2009 at 01:37:55AM -0800, david levis wrote:
> ar: symbol lookup error: ar: undefined symbol: bfd_plugin_set_program_name
Please show the options you passed to configure.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binut
i should be recognized, unless you have disabled it by using
one of the assembler's -m options for a cpu without this instruction,
or done the same with a .machine directive.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing l
bjcopy -O binary on a relocatable file loses the relocs.
X86_64 uses rela style relocs where the addend is in the reloc, x86
uses rel where the addend is in the section. So in the latter case
you get the "+4" from .text+4, in the former you don't.
--
Alan Modra
Australia Deve
On Thu, Dec 10, 2009 at 04:43:15PM +, me wrote:
> Does anyone know what I should use instead of
> "objcopy -O binary a.out a.com"
> to stop losing the relocs?
If you want a binary output you must lose the relocs. So don't try to
objcopy relocatable file. Final li
at the core file, the return value from find_plt_ent() is not
> checked in several points, leading to a NULL pointer dereference in some
> situations.
I have just committed a patch that should fix this problem.
http://sourceware.org/ml/binutils/2009-12/msg00231.html
--
Alan Modra
Austral
adding --verbose to the ld command
line. You can set the start to zero by adding -Ttext=0 or
-Ttext-segment=0 to the linker command line (depending on version of ld).
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binut
to this target. See
http://sourceware.org/ml/binutils/2004-04/msg00498.html
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
m:2: error: expected idetifier or '(' before '%' token
You shouldn't be compiling that file! Broken "make"?
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
haslinno = TRUE;
+#ifdef COFF_IMAGE_WITH_PE
if ((current->flags & SEC_DEBUGGING) != 0
&& ! is_reloc_section)
hasdebug = TRUE;
+#endif
#ifdef RS6000COFF_C
#ifndef XCOFF64
--
Alan Modra
Australia Development Lab, IBM
___
rmats.
Agreed. The easy solution is to choose one of these two targets as
the default, in this case it's preferable to choose elf32-i386. So
add --target=i686-unknown-linux-gnu to your configure line. I'm
assuming here that you're o
On Sun, Jun 06, 2010 at 02:16:06AM -0400, Luciann Bennet wrote:
> 1 .__ksetup 4000 00101000 00101000 b000 2**12
> CONTENTS, READONLY
This section isn't ALLOC. The linker is correct in not making it part
of a loadable segment.
--
Alan Modr
r operands, where the source register is
%st, and destination register is %st(i).
The affected opcode map is dceX, dcfX, deeX, defX. */
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.
ation.
> So it should be bug.
What happens if you do not use an archive? ie. extract the files in
the archive then specify all of them on the ld command line?
--
Alan Modra
Australia Development Lab, IBM
__
nter_bug.cgi?product=binutils
attaching the object files/libraries and specify the ld command line
so that someone else can reproduce the problem.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://l
t get
committed due to lack of review. Oops. I've just given him the OK.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
Thanks! Applying the following to mainline.
Index: binutils/ChangeLog
===
RCS file: /cvs/src/src/binutils/ChangeLog,v
retrieving revision 1.1714
diff -u -p -r1.1714 ChangeLog
--- binutils/ChangeLog 11 Nov 2010 10:23:37 - 1.
nting at 1.
> #: dwarf.c:2970
> #, c-format
> msgid ""
> "\n"
> "%s/%s:\n"
> msgstr ""
>
> #: dwarf.c:3070
> #, c-format
> msgid "%-35s %11d %#18lx\n"
> msgstr ""
>
> #: dwarf.c:3074
> #, c
version of GNU
ld?
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
On Mon, Nov 22, 2010 at 02:34:29PM +0100, Roger Jeurninck wrote:
> the version which we use is 2.17.
That's over 4 years old! We are about to release 2.21. I recall
fixing one uninitialised memory bug in that time. There were likely
others too.
--
Alan Modra
Australia Development
production scenario at that
> time?
http://sourceware.org/bugzilla/show_bug.cgi?id=11887 Likely not
related to your problem.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
the ar -s command."
This is an entirely different symbol table. An archive containing
object files has a table that tells the linker which object file
contains a definition for each global symbol.
--
Alan Modra
Australia Development Lab, IBM
___
both like c source file and shell
> source file. can you tell me what it is? how does it used for? thank you very
> much !
It is a shell file that creates a C file. You can see how they are
used by inspecting the ld Makefile, or by simply watching what happens
when building ld.
se version 3 or (at your option) a later version.
This program has absolutely no warranty.
Is the --no-demangle option really making it to ld?
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
ing --{no-,}demangle.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
nothing uses
aout_gnu.h. See include/aout/aout64.h for the values actually used.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
ker
command line. ie. all the input to the linker.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
crt -luser32 -lkernel32 -ladvapi32 -lshell32 -lmingw32 -lgcc
-lmoldname -lmingwex -lmsvcrt crtend.o
$ echo $?
0
$ ~/build/gas/i586-mingw32/ld/ld-new -v
GNU ld (GNU Binutils) 2.21.51.20110123
--
Alan Modra
Australia Development Lab, IBM
___
bug-binut
for help on a mingw list. How you configure depends on
where your current cross-tools are.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
there a way to get the total file size here, and if it is greater
> than 4gig, to make this field wider? I realize that it will then be
> greater than 80 chars...
Patches welcome.. Alternatively, build on a 64-bit host.
--
Alan Modra
Australia Development Lab, IBM
On Wed, Feb 02, 2011 at 05:40:09PM -0800, Bruce Waggoner wrote:
> Assertion failure in ignore_rest_of_line
I believe this bug has already been fixed. Upgrade your binutils to
2.21.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binut
vide a relocatable symbol like Manufacturer_ID_String by
two (unless your target happens to have a relocation that would allow
the linker to perform the divide for you).
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing
On Thu, Mar 03, 2011 at 11:01:11AM -0800, Michael Snyder wrote:
> 2011-03-03 Michael Snyder
>
> * archive.c (_bfd_slurp_extended_name_table): Fail if bfd_seek fails.
OK.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutil
On Thu, Mar 03, 2011 at 12:12:52PM -0800, Michael Snyder wrote:
> 2011-03-03 Michael Snyder
>
> * dwarf2.c (scan_unit_for_symbols): Stop memory leak.
OK.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing
On Thu, Mar 03, 2011 at 10:25:31AM -0800, Michael Snyder wrote:
> 2011-03-03 Michael Snyder
>
> * aoutx.txt (aout_final_link): Use sizeof int not sizeof int*.
aoutx.h. OK.
--
Alan Modra
Australia Development Lab, IBM
___
bug
On Thu, Mar 03, 2011 at 10:32:18AM -0800, Michael Snyder wrote:
> * aoutx.h (aout_link_write_other_symbol): Missing break statement.
OK.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
h
On Thu, Mar 03, 2011 at 10:46:09AM -0800, Michael Snyder wrote:
> * archive.c (bsd_write_armap): Fail if stat fails.
I think it would be better to just set the timestamp to zero if stat
fails. There is precedent for this, eg. opncls.c:_maybe_make_executable.
--
Alan Modra
Austra
On Thu, Mar 03, 2011 at 11:12:34AM -0800, Michael Snyder wrote:
> * coff-x86_64.c (coff_amd64_rtype_to_howto): Fencepost error.
OK.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
h
On Thu, Mar 03, 2011 at 12:27:02PM -0800, Michael Snyder wrote:
> * elf64-x86-64.c (elf_x86_64_relocate_section): Remove dead code.
OK.
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.org
h
On Thu, Mar 03, 2011 at 12:06:12PM -0800, Michael Snyder wrote:
> * peXXigen.c (_bfd_XXi_swap_aux_in): Use sizeof in memcpy.
> (_bfd_XXi_swap_aux_out): Ditto.
OK. The less code I have to look at to verify something is correct,
the more I like it.
--
Alan Modra
Australia Devel
ocation of x.o as specified in x.a is "../x.o" and not
> "y/x.o". It doesn't seem to be a problem in the opposite direction
> -- "ar rT x.a y/x.o" generates the relative location as "y/x.o".
>
> Actually, is this a bug?
I thi
ab->elf.dynobj = abfd;
+ if (!_bfd_elf_create_ifunc_sections (htab->elf.dynobj, info))
return FALSE;
break;
}
--
Alan Modra
Australia Development Lab, IBM
___
bug-binutils mailing list
bug-binutils@gnu.o
g.c 7 Jun 2011 09:03:41 -
@@ -5885,6 +5885,9 @@ lang_one_common (struct bfd_link_hash_en
bfd_vma size;
asection *section;
+ if (h->type == bfd_link_hash_warning)
+h = h->u.i.link;
+
if (h->type != bfd_link_hash_common)
return TRUE;
--
Alan Modra
Australia Developm
an point to any other sym type.
The lang_one_common change is needed in any function called by
bfd_link_hash_traverse. When warning symbols are created, they
replace the "real" entry in the hash table, so you never get to see
the real symbol in a hash traversal.
--
Alan Modra
101 - 200 of 203 matches
Mail list logo
