This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: Trunk frozen starting tomorrow! No new features!
On 06/19/2012 11:46 AM, Carlos O'Donell wrote:
> On Tue, Jun 19, 2012 at 8:06 AM, Joseph S. Myers
> <joseph@codesourcery.com> wrote:
>> On Fri, 15 Jun 2012, Ryan S. Arnold wrote:
>>
>>> I just finished verifying that for PowerPC[32|64] the master branch
>>> currently has no new regressions.
>> What does "no new regressions" mean? That there are regressions (tests
>> that fail but used to pass), but they are regressions in 2.15 rather than
>> new since then?
>>
>> Do you see a check-abi failure on powerpc32 for _mcount@GLIBC_2.0 (bug
>> 14042)? (I see that failure on powerpc-nofpu.) If you see such a
>> failure, it should be documented on the wiki.
> $0.02.
>
> I expect "no new regressions" to mean "No new regressions versus a
> baseline of failures from 2.15."
>
> Cheers,
> Carlos.
>
Ryan is on vacation and he asked to answer the question that may arise for 2.16 release. And
I too believe his "no new regressions" means "No new regressions versus a baseline of
failures from 2.15." since in my PPC32 and PPC64 for POWER4 tests I do see the 'check-abi-libc'
errors with the following difference:
- _mcount F
- _nl_default_dirname D 0x12
+ _nl_default_dirname D 0xf
Besides that I do saw the 'rt/tst-cpuclock2' intermittent failures which afaik is due a kernel
issue (I'm testing on an old 2.6.32). For POWER7 PPC32 I also see the
'stdlib/tst-setcontext', 'stdlib/tst-makecontext3', and 'tst-setcontext-fpscr' which are all
known kernel issues.
My tests were done on Linux 2.6.32, GCC 4.6.4, Binutils 2.22.52.
--
Adhemerval Zanella Netto
Software Engineer
Linux Technology Center Brazil
Toolchain / GLIBC on Power Architecture
azanella@linux.vnet.ibm.com / azanella@br.ibm.com
+55 61 8642-9890
is