This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] [BZ #18078] FAIL: elf/check-localplt with ld from binutils 2.26


On Tue, Jul 14, 2015 at 12:43 PM, H.J. Lu <hjl.tools@gmail.com> wrote:
> On Wed, Mar 11, 2015 at 10:02 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
>> On Tue, Mar 10, 2015 at 3:21 PM, Roland McGrath <roland@hack.frob.com> wrote:
>>> I don't think this is the way to do it.  For those cases it is mandatory
>>> that libc is actually using the PLT entry, so we don't want a check that
>>> would pass if libc stopped using it at all.  Perhaps instead there is a way
>>> to modify scripts/localplt.awk and/or scripts/check-localplt.awk so that
>>> they will grok the new linker's output as constituting use of a PLT entry.
>>> Maybe a GLOB_DAT reloc to a symbol that's STT_FUNC is a close enough proxy?
>>
>> Here is the patch to extend local PLT reference check to support
>> alternate relocations.  OK for master?
>>
>>
>> --
>> H.J.
>> ---
>> On x86, linker in binutils 2.26 and newer consolidates R_*_JUMP_SLOT with
>> R_*_GLOB_DAT relocation against the same symbol.  This patch extends
>> local PLT reference check to support alternate relocations.
>>
>> [BZ #18078]
>> * scripts/check-localplt.awk: Support alternate relocations.
>> * scripts/localplt.awk: Also check relocations in DT_RELA/DT_REL
>> sections.
>> * sysdeps/unix/sysv/linux/i386/localplt.data: Mark free and
>> malloc entries with + REL R_386_GLOB_DAT.
>> * sysdeps/x86_64/localplt.data: New file.
>
> PING:
>
> https://sourceware.org/ml/libc-alpha/2015-03/msg00456.html
>

I checked it in.

-- 
H.J.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]