r240176 caused a regression in reading MIPS elf files.

Rafael Espíndola rafael.espindola at gmail.com
Mon Jul 6 05:19:38 PDT 2015


Fixed in r241440. Thanks a lot for reporting the problem.

On 6 July 2015 at 04:29, Aboud, Amjad <amjad.aboud at intel.com> wrote:
> Hi,
>
> This is how you can reproduce the issue I saw.
>
> Run the following command line:
> clang.exe -cc1  -o - -triple mips64 -g -emit-obj -O0 < foo.c | llvm-dwarfdump -debug-dump=info -
>
> Where foo.c contains:
> int foo(void) {
>   return 0;
> }
>
> You will notice that " DW_AT_name" for all entries have the same string (which suits the first entry in the string section).
>
> Regards,
> Amjad
>
> -----Original Message-----
> From: Daniel Sanders [mailto:Daniel.Sanders at imgtec.com]
> Sent: Friday, July 03, 2015 16:32
> To: Rafael Espíndola
> Cc: Aboud, Amjad; llvm-commits at cs.uiuc.edu
> Subject: RE: r240176 caused a regression in reading MIPS elf files.
>
> Not without more information from Amjad. I don't yet know what the problem is. I have been able to produce a testcase that doesn't resolve an R_MIPS_32 reference to $section_abbrev+4 correctly on the N64 ABI but this failed before your change too.
>
>> -----Original Message-----
>> From: Rafael Espíndola [mailto:rafael.espindola at gmail.com]
>> Sent: 02 July 2015 16:12
>> To: Daniel Sanders
>> Cc: Aboud, Amjad; llvm-commits at cs.uiuc.edu
>> Subject: Re: r240176 caused a regression in reading MIPS elf files.
>>
>> Interesting. Can you provide a testcase (.s) showing the problem?
>> llvm-dwarfdump seems to be the best way to trigger this.
>>
>> On 2 July 2015 at 10:48, Daniel Sanders <Daniel.Sanders at imgtec.com> wrote:
>> >> Sorry, but what issues did it cause? 32 bits mips uses REL instead of RELA.
>> >
>> > I don't know about the issues caused (yet?) but I should mention
>> > that the
>> visitELF_MIPS_32() mentioned in Amjad's patch handles the R_MIPS_32
>> reloc for both the REL and RELA cases. It's not referring to 32-bit Mips.
>> >
>> >> -----Original Message-----
>> >> From: llvm-commits-bounces at cs.uiuc.edu [mailto:llvm-commits-
>> >> bounces at cs.uiuc.edu] On Behalf Of Rafael Espíndola
>> >> Sent: 02 July 2015 13:31
>> >> To: Aboud, Amjad
>> >> Cc: llvm-commits at cs.uiuc.edu
>> >> Subject: Re: r240176 caused a regression in reading MIPS elf files.
>> >>
>> >> On 1 July 2015 at 15:02, Aboud, Amjad <amjad.aboud at intel.com> wrote:
>> >> > Hi Rafael,
>> >> >
>> >> > I just noticed that your commit in revision 240176 caused a
>> >> > regression in
>> >> reading elf file for MIPS architecture.
>> >> > It seems a copy paste issue, please see attach patch with the
>> >> > proposed
>> fix.
>> >> >
>> >> > If you agree, please can you take care of the issue?
>> >>
>> >> Sorry, but what issues did it cause? 32 bits mips uses REL instead of RELA.
>> >>
>> >> Cheers,
>> >> Rafael
>> >> _______________________________________________
>> >> llvm-commits mailing list
>> >> llvm-commits at cs.uiuc.edu
>> >> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
> ---------------------------------------------------------------------
> Intel Israel (74) Limited
>
> This e-mail and any attachments may contain confidential material for
> the sole use of the intended recipient(s). Any review or distribution
> by others is strictly prohibited. If you are not the intended
> recipient, please contact the sender and delete all copies.




More information about the llvm-commits mailing list