public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
* build-many-glibcs.py failures for powerpc
@ 2017-01-05 13:51 Joseph Myers
  2017-01-05 18:32 ` Tulio Magno Quites Machado Filho
  0 siblings, 1 reply; 3+ messages in thread
From: Joseph Myers @ 2017-01-05 13:51 UTC (permalink / raw)
  To: libc-alpha

In tests with mainline GCC and binutils I see:

FAIL: glibcs-powerpc-linux-gnu check
FAIL: glibcs-powerpc-linux-gnu-power4 check
FAIL: glibcs-powerpc64le-linux-gnu check

https://sourceware.org/ml/libc-testresults/2017-q1/msg00026.html

The first two are elf/check-textrel failing for libm.so.  The third is 
elf/check-localplt failing ("Extra PLT reference: libc.so: strncmp").  
There are no such failures for powerpc64 big endian or for soft float / 
e500.  I suppose the first two might be GCC / binutils bugs, given that I 
don't see them with the bot using GCC 6 branch and binutils 2.28 branch.  
I don't know how long any of these failures have been around, since 
previously they'd have been hidden by the warnings causing failures for 
all architectures.

-- 
Joseph S. Myers
joseph@codesourcery.com

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: build-many-glibcs.py failures for powerpc
  2017-01-05 13:51 build-many-glibcs.py failures for powerpc Joseph Myers
@ 2017-01-05 18:32 ` Tulio Magno Quites Machado Filho
  2017-01-12  2:11   ` Joseph Myers
  0 siblings, 1 reply; 3+ messages in thread
From: Tulio Magno Quites Machado Filho @ 2017-01-05 18:32 UTC (permalink / raw)
  To: Joseph Myers, libc-alpha

Joseph Myers <joseph@codesourcery.com> writes:

> In tests with mainline GCC and binutils I see:
>
> FAIL: glibcs-powerpc-linux-gnu check
> FAIL: glibcs-powerpc-linux-gnu-power4 check
> FAIL: glibcs-powerpc64le-linux-gnu check
>
> https://sourceware.org/ml/libc-testresults/2017-q1/msg00026.html
>
> The first two are elf/check-textrel failing for libm.so.  The third is 
> elf/check-localplt failing ("Extra PLT reference: libc.so: strncmp").  
> There are no such failures for powerpc64 big endian or for soft float / 
> e500.  I suppose the first two might be GCC / binutils bugs, given that I 
> don't see them with the bot using GCC 6 branch and binutils 2.28 branch.  
> I don't know how long any of these failures have been around, since 
> previously they'd have been hidden by the warnings causing failures for 
> all architectures.

I've never seen them on my tests.  The only elf testcase failing here is
elf/tst-tls-manydynamic (BZ #20915).

I tested using:
 - A few other older combinations of Binutils and GCC
 - Binutils 2.26 + GCC 5.4
 - Binutils 2.25 + GCC 5.3 
 - Binutils 2.27 + GCC 6.2 (rev 242657)
 - Binutils 2.27 + GCC 7.0 (rev 243648)

I'm trying build-many-glibcs.py to see if I can reproduce them here.

-- 
Tulio Magno

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: build-many-glibcs.py failures for powerpc
  2017-01-05 18:32 ` Tulio Magno Quites Machado Filho
@ 2017-01-12  2:11   ` Joseph Myers
  0 siblings, 0 replies; 3+ messages in thread
From: Joseph Myers @ 2017-01-12  2:11 UTC (permalink / raw)
  To: Tulio Magno Quites Machado Filho; +Cc: libc-alpha

On Thu, 5 Jan 2017, Tulio Magno Quites Machado Filho wrote:

> Joseph Myers <joseph@codesourcery.com> writes:
> 
> > In tests with mainline GCC and binutils I see:
> >
> > FAIL: glibcs-powerpc-linux-gnu check
> > FAIL: glibcs-powerpc-linux-gnu-power4 check
> > FAIL: glibcs-powerpc64le-linux-gnu check
> >
> > https://sourceware.org/ml/libc-testresults/2017-q1/msg00026.html
> >
> > The first two are elf/check-textrel failing for libm.so.  The third is 
> > elf/check-localplt failing ("Extra PLT reference: libc.so: strncmp").  
> > There are no such failures for powerpc64 big endian or for soft float / 
> > e500.  I suppose the first two might be GCC / binutils bugs, given that I 
> > don't see them with the bot using GCC 6 branch and binutils 2.28 branch.  
> > I don't know how long any of these failures have been around, since 
> > previously they'd have been hidden by the warnings causing failures for 
> > all architectures.
> 
> I've never seen them on my tests.  The only elf testcase failing here is
> elf/tst-tls-manydynamic (BZ #20915).

I've traced the 32-bit failures to a GCC bug (however, the revision 
implicated by bisection looks like it probably just exposed a latent bug 
elsewhere).

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79066

-- 
Joseph S. Myers
joseph@codesourcery.com

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2017-01-12  2:11 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-01-05 13:51 build-many-glibcs.py failures for powerpc Joseph Myers
2017-01-05 18:32 ` Tulio Magno Quites Machado Filho
2017-01-12  2:11   ` Joseph Myers

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).