[PATCH,plugins,head+2.21] Handle COMDAT groups on ELF.

[Dave Korn]

On 07/05/2011 17:23, H.J. Lu wrote:
> On Sat, May 7, 2011 at 9:04 AM, Dave Korn <dave.korn.cygwin@gmail.com> wrote:
>> On 07/05/2011 16:58, H.J. Lu wrote:
>>> On Sat, May 7, 2011 at 8:54 AM, Dave Korn <dave.korn.cygwin@gmail.com> wrote:
>>>> On 07/05/2011 16:06, H.J. Lu wrote:
>>>>> On Sat, May 7, 2011 at 7:52 AM, Dave Korn wrote:
>>>>>>  Any comments on the implementation, while we're waiting?  Or OK as it is?
>>>>>>
>>>>> Do you have a testcase to show it fails without your change?
>>>>  Nope, but I guess we could try and extract one from the gcc/g++ testsuite;
>>>> pretty much every testcase that involves EH or typeinfo is busted without it.
>>> I don't see any LTO failures on Linux/x86 with the current BFD linker.
>>>
>>  I'll have to retest then.  I did initially get this many (attached)
>> failures, but it's possible that Alan's pr12365 (or related) changes resolved
>> them somehow.
> 
> I believe so.

  Yes, the retest without that patch suggests so, and I think I know why; it
was because of the differing times at which we set the SEC_EXCLUDE flag and
clear the output_sections, and the combination of Alan's earlier setting and
fixing the bug in the COFF backend has obviated the need for this patch.

  So, patch withdrawn.

    cheers,
      DaveK