testsuite results, binutils-2.29 branch

classic Classic list List threaded Threaded
12 messages Options
Reply | Threaded
Open this post in threaded view
|

testsuite results, binutils-2.29 branch

Matthias Klose-6
Some updated test results for binutils builds from the 2.29 branch.  I'm
preparing an archive test rebuild using the new binutils, but will only be able
to post results starting with Jul 17. Also found that glibc-2.24 fails to build
in the check target on aarch64 and arm32 with the new binutils (see PR21725).

Matthias


amd64:
no failures

i686:
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-i386/i386.exp ...
FAIL: Run pr19031
FAIL: Run got1
FAIL: Undefined weak symbol (-fPIE -no-pie)
FAIL: Undefined weak symbol (-fPIE -pie)
FAIL: Build ifunc-1a with PIE -z ibtplt
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-i386/no-plt.exp ...
FAIL: No PLT (dynamic 1b)
FAIL: No PLT (PIE 1e)
FAIL: No PLT (PIE 1f)
FAIL: No PLT (PIE 1g)

arm64:
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-aarch64/aarch64-elf.exp ...
FAIL: ld-aarch64/dt_textrel
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-ifunc/ifunc.exp ...
FAIL: Could not link a dynamic executable using local ifunc

armhf:
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-elf/indirect.exp ...
FAIL: Run indirect5 1
FAIL: Run indirect5 2
FAIL: indirect5a dynsym
FAIL: indirect5b dynsym
FAIL: Run indirect5 3
FAIL: Run indirect5 4
FAIL: indirect5c dynsym
FAIL: indirect5d dynsym
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-elf/shared.exp ...
FAIL: PR ld/21233 dynamic symbols with section GC (--undefined)
FAIL: PR ld/21233 dynamic symbols with section GC (--require-defined)
FAIL: PR ld/21233 dynamic symbols with section GC (EXTERN)
XPASS: Run pr19719 fun undefined
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-elfcomm/elfcomm.exp ...
FAIL: size/aligment change of common symbols (warning 1)

ppc64el:
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-ifunc/ifunc.exp ...
FAIL: Could not link a dynamic executable using local ifunc
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-powerpc/powerpc.exp ...
FAIL: ld-powerpc/ppc476-shared
FAIL: ld-powerpc/ppc476-shared2

s390x:
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-elf/indirect.exp ...
FAIL: Run indirect5 1
FAIL: Run indirect5 2
FAIL: indirect5a dynsym
FAIL: indirect5b dynsym
FAIL: Run indirect5 3
FAIL: Run indirect5 4
FAIL: indirect5c dynsym
FAIL: indirect5d dynsym
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-plugin/plugin.exp ...
FAIL: plugin claimfile replace symbol with source
FAIL: plugin claimfile resolve symbol with source
FAIL: plugin claimfile replace file with source
FAIL: plugin set symbol visibility with source
FAIL: plugin ignore lib with source
FAIL: plugin claimfile replace lib with source
FAIL: plugin 2 with source lib
FAIL: load plugin 2 with source
FAIL: plugin 3 with source lib
FAIL: load plugin 3 with source
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-s390/s390.exp ...
FAIL: TLS -fpic -shared transitions

mips:
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-elf/binutils.exp ...
FAIL: objcopy -shared -z relro (tbss1)
FAIL: objcopy -shared -z relro (tbss2)
FAIL: objcopy -shared -z relro (tbss3)
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-elf/indirect.exp ...
FAIL: Run indirect5 1
FAIL: Run indirect5 2
FAIL: Run indirect6 1
FAIL: Run indirect6 2
FAIL: indirect5a dynsym
FAIL: indirect5b dynsym
FAIL: indirect6a dynsym
FAIL: indirect6b dynsym
FAIL: Run indirect5 3
FAIL: Run indirect5 4
FAIL: Run indirect6 3
FAIL: Run indirect6 4
FAIL: indirect5c dynsym
FAIL: indirect5d dynsym
FAIL: indirect6c dynsym
FAIL: indirect6d dynsym
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-elf/shared.exp ...
FAIL: Build libpr16496b.so
FAIL: Run pr2404
FAIL: Run pr2404n
FAIL: Run pr2404 with PIE
FAIL: Run pr2404 with PIE (-z now)
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-elfvers/vers.exp ...
ERROR:  tmpdir/vers1.s: assembly failed
ERROR:  tmpdir/vers2.s: assembly failed
ERROR:  tmpdir/vers3.s: assembly failed
ERROR:  tmpdir/vers4.s: assembly failed
ERROR:  tmpdir/vers4a.s: assembly failed
ERROR:  tmpdir/vers4b.s: assembly failed
ERROR:  tmpdir/vers5.s: assembly failed
ERROR:  tmpdir/vers6.s: assembly failed
ERROR:  tmpdir/vers8.s: assembly failed
ERROR:  tmpdir/vers9.s: assembly failed
ERROR:  tmpdir/vers10.s: assembly failed
ERROR: tcl error sourcing /<<PKGBUILDDIR>>/ld/testsuite/ld-elfvers/vers.exp.
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-gc/gc.exp ...
FAIL: --gc-sections with --defsym
FAIL: --gc-sections with KEEP
FAIL: --gc-sections with __start_SECTIONNAME
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-plugin/lto.exp ...
FAIL: ld-plugin/lto-3r
FAIL: ld-plugin/lto-5r
FAIL: PR ld/19317 (2)
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-plugin/plugin.exp ...
FAIL: plugin claimfile lost symbol
FAIL: plugin claimfile replace symbol
FAIL: plugin claimfile resolve symbol
FAIL: plugin claimfile lost symbol with source
FAIL: plugin claimfile replace symbol with source
FAIL: plugin claimfile resolve symbol with source
FAIL: plugin 2 with source lib
FAIL: load plugin 2 with source
FAIL: plugin 3 with source lib
FAIL: load plugin 3 with source
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-shared/shared.exp ...
FAIL: shared (non PIC)
FAIL: shared (PIC main, non PIC so)
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-unique/unique.exp ...
FAIL: ld-unique/pr21529

hppa:
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-elfcomm/elfcomm.exp ...
FAIL: size/aligment change of common symbols (warning 1)

powerpc:
Running /«PKGBUILDDIR»/ld/testsuite/ld-elf/shared.exp ...
FAIL: Run with pr11138-2.c libpr11138-1.so
FAIL: Run with libpr11138-1.so pr11138-2.c
Running /«PKGBUILDDIR»/ld/testsuite/ld-powerpc/powerpc.exp ...
FAIL: ld-powerpc/ppc476-shared
FAIL: ld-powerpc/ppc476-shared2

ppc64:
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-powerpc/powerpc.exp ...
FAIL: ld-powerpc/ppc476-shared
FAIL: ld-powerpc/ppc476-shared2

sparc64:
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-elfvsb/elfvsb.exp ...
FAIL: visibility (hidden_normal) (PIC main)
FAIL: visibility (hidden_weak)
FAIL: visibility (hidden_weak) (PIC main)
Running /<<PKGBUILDDIR>>/ld/testsuite/ld-ifunc/ifunc.exp ...
FAIL: Could not link a dynamic executable using local ifunc

complete build logs at
https://launchpad.net/ubuntu/+source/binutils/2.28.90.20170704-0ubuntu1
https://buildd.debian.org/status/package.php?p=binutils&suite=experimental
Reply | Threaded
Open this post in threaded view
|

Re: testsuite results, binutils-2.29 branch

H.J. Lu-30
On Thu, Jul 6, 2017 at 5:50 AM, Matthias Klose <[hidden email]> wrote:

> Some updated test results for binutils builds from the 2.29 branch.  I'm
> preparing an archive test rebuild using the new binutils, but will only be able
> to post results starting with Jul 17. Also found that glibc-2.24 fails to build
> in the check target on aarch64 and arm32 with the new binutils (see PR21725).
>
> Matthias
>
>
> amd64:
> no failures
>
> i686:
> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-i386/i386.exp ...
> FAIL: Run pr19031
> FAIL: Run got1
> FAIL: Undefined weak symbol (-fPIE -no-pie)
> FAIL: Undefined weak symbol (-fPIE -pie)
> FAIL: Build ifunc-1a with PIE -z ibtplt
> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-i386/no-plt.exp ...
> FAIL: No PLT (dynamic 1b)
> FAIL: No PLT (PIE 1e)
> FAIL: No PLT (PIE 1f)
> FAIL: No PLT (PIE 1g)
>

Please try master branch.  It is clean for me on i686 with GCC 7 default
to PIE.


--
H.J.
Reply | Threaded
Open this post in threaded view
|

Re: testsuite results, binutils-2.29 branch

Matthias Klose-6
On 06.07.2017 15:57, H.J. Lu wrote:

> On Thu, Jul 6, 2017 at 5:50 AM, Matthias Klose <[hidden email]> wrote:
>> Some updated test results for binutils builds from the 2.29 branch.  I'm
>> preparing an archive test rebuild using the new binutils, but will only be able
>> to post results starting with Jul 17. Also found that glibc-2.24 fails to build
>> in the check target on aarch64 and arm32 with the new binutils (see PR21725).
>>
>> Matthias
>>
>>
>> amd64:
>> no failures
>>
>> i686:
>> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-i386/i386.exp ...
>> FAIL: Run pr19031
>> FAIL: Run got1
>> FAIL: Undefined weak symbol (-fPIE -no-pie)
>> FAIL: Undefined weak symbol (-fPIE -pie)
>> FAIL: Build ifunc-1a with PIE -z ibtplt
>> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-i386/no-plt.exp ...
>> FAIL: No PLT (dynamic 1b)
>> FAIL: No PLT (PIE 1e)
>> FAIL: No PLT (PIE 1f)
>> FAIL: No PLT (PIE 1g)
>>
>
> Please try master branch.  It is clean for me on i686 with GCC 7 default
> to PIE.

yes, no failures, using GCC 6. Please could you backport the fixes to the 2.29
branch?

Matthias
Reply | Threaded
Open this post in threaded view
|

Re: testsuite results, binutils-2.29 branch

H.J. Lu-30
On Thu, Jul 6, 2017 at 8:01 AM, Matthias Klose <[hidden email]> wrote:

> On 06.07.2017 15:57, H.J. Lu wrote:
>> On Thu, Jul 6, 2017 at 5:50 AM, Matthias Klose <[hidden email]> wrote:
>>> Some updated test results for binutils builds from the 2.29 branch.  I'm
>>> preparing an archive test rebuild using the new binutils, but will only be able
>>> to post results starting with Jul 17. Also found that glibc-2.24 fails to build
>>> in the check target on aarch64 and arm32 with the new binutils (see PR21725).
>>>
>>> Matthias
>>>
>>>
>>> amd64:
>>> no failures
>>>
>>> i686:
>>> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-i386/i386.exp ...
>>> FAIL: Run pr19031
>>> FAIL: Run got1
>>> FAIL: Undefined weak symbol (-fPIE -no-pie)
>>> FAIL: Undefined weak symbol (-fPIE -pie)
>>> FAIL: Build ifunc-1a with PIE -z ibtplt
>>> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-i386/no-plt.exp ...
>>> FAIL: No PLT (dynamic 1b)
>>> FAIL: No PLT (PIE 1e)
>>> FAIL: No PLT (PIE 1f)
>>> FAIL: No PLT (PIE 1g)
>>>
>>
>> Please try master branch.  It is clean for me on i686 with GCC 7 default
>> to PIE.
>
> yes, no failures, using GCC 6. Please could you backport the fixes to the 2.29
> branch?

Done.   Please test binutils-2_29-branch.

--
H.J.
Reply | Threaded
Open this post in threaded view
|

Re: testsuite results, binutils-2.29 branch

Alan Modra-3
In reply to this post by Matthias Klose-6
On Thu, Jul 06, 2017 at 02:50:55PM +0200, Matthias Klose wrote:
> ppc64el:
> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-ifunc/ifunc.exp ...
> FAIL: Could not link a dynamic executable using local ifunc

ld.log shows ld emitting an error: "text relocations and GNU indirect
functions will result in a segfault at runtime".
This is the --enable-default-pie compiler problem fixed by
https://gcc.gnu.org/ml/gcc-patches/2017-06/msg01678.html

> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-powerpc/powerpc.exp ...
> FAIL: ld-powerpc/ppc476-shared
> FAIL: ld-powerpc/ppc476-shared2

I can't reproduce the above.  Can I see your ld.log?

--
Alan Modra
Australia Development Lab, IBM
Reply | Threaded
Open this post in threaded view
|

Re: testsuite results, binutils-2.29 branch

Alan Modra-3
On Fri, Jul 07, 2017 at 11:55:37AM +0200, Matthias Klose wrote:

> On 07.07.2017 09:19, Alan Modra wrote:
> > On Thu, Jul 06, 2017 at 02:50:55PM +0200, Matthias Klose wrote:
> >> ppc64el:
> >> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-ifunc/ifunc.exp ...
> >> FAIL: Could not link a dynamic executable using local ifunc
> >
> > ld.log shows ld emitting an error: "text relocations and GNU indirect
> > functions will result in a segfault at runtime".
> > This is the --enable-default-pie compiler problem fixed by
> > https://gcc.gnu.org/ml/gcc-patches/2017-06/msg01678.html
>
> hmm, my impression was that you retracted that patch ...

I hope that's not what potential reviewers think!

> PR 80044 is closed as won't fix.

Yes, that's because I don't intend to make -pie -static work together
(with the traditional meaning of -static).

> However PR 81170 is still open.

and your PR 81295.

> >> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-powerpc/powerpc.exp ...
> >> FAIL: ld-powerpc/ppc476-shared
> >> FAIL: ld-powerpc/ppc476-shared2
> >
> > I can't reproduce the above.  Can I see your ld.log?
>
> attached.

The fails are due to Ubuntu patches, and a testcase that is too fussy
over expected output.  Problem is that .dynamic and other sections sit
between .text and .bss, so anything like changing hash style makes
.bss move.  This patch should make the tests more robust.

        * testsuite/ld-powerpc/ppc476-shared.lnk: Align .bss.
        * testsuite/ld-powerpc/ppc476-shared.d: Adjust.
        * testsuite/ld-powerpc/ppc476-shared2.d: Adjust.

diff --git a/ld/testsuite/ld-powerpc/ppc476-shared.d b/ld/testsuite/ld-powerpc/ppc476-shared.d
index fe438f4..72f8a3d 100644
--- a/ld/testsuite/ld-powerpc/ppc476-shared.d
+++ b/ld/testsuite/ld-powerpc/ppc476-shared.d
@@ -23,7 +23,7 @@ Disassembly of section \.text:
  \.\.\.
    3fff0: (42 9f 00 05|05 00 9f 42) bcl     .*
    3fff4: (7d 28 02 a6|a6 02 28 7d) mflr    r9
-   3fff8: (3d 29 00 00|00 00 29 3d) addis   r9,r9,0
+   3fff8: (3d 29 00 01|01 00 29 3d) addis   r9,r9,1
  3fff[8a]: R_PPC_REL16_HA \.bss\+0x[46]
    3fffc: (48 00 00 34|34 00 00 48) b       40030 .*
    40000: (3c 60 00 00|00 00 60 3c) lis     r3,0
@@ -41,7 +41,7 @@ Disassembly of section \.text:
    40024: (4b fe ff dc|dc ff fe 4b) b       30000 .*
    40028: (48 00 00 02|02 00 00 48) ba      0 .*
    4002c: (48 00 00 02|02 00 00 48) ba      0 .*
-   40030: (39 29 01 50|50 01 29 39) addi    r9,r9,336
+   40030: (39 29 00 0c|0c 00 29 39) addi    r9,r9,12
  4003[02]: R_PPC_REL16_LO \.bss\+0x3[ce]
    40034: (4b ff ff cc|cc ff ff 4b) b       40000 .*
    40038: (48 00 00 02|02 00 00 48) ba      0 .*
diff --git a/ld/testsuite/ld-powerpc/ppc476-shared.lnk b/ld/testsuite/ld-powerpc/ppc476-shared.lnk
index 5339358..03e66b3 100644
--- a/ld/testsuite/ld-powerpc/ppc476-shared.lnk
+++ b/ld/testsuite/ld-powerpc/ppc476-shared.lnk
@@ -2,5 +2,5 @@ SECTIONS
 {
   . = 0xfffc;
   .text : { *(.text) }
-  .bss : { *(.bss) }
+  .bss : ALIGN (0x10000) { *(.bss) }
 }
diff --git a/ld/testsuite/ld-powerpc/ppc476-shared2.d b/ld/testsuite/ld-powerpc/ppc476-shared2.d
index 813ea2e..5bf0a03 100644
--- a/ld/testsuite/ld-powerpc/ppc476-shared2.d
+++ b/ld/testsuite/ld-powerpc/ppc476-shared2.d
@@ -8,9 +8,9 @@
 
 DYNAMIC RELOCATION RECORDS
 OFFSET   TYPE              VALUE
-0001000[02] R_PPC_ADDR16_LO   \.text\+0x00040144
-0002000[02] R_PPC_ADDR16_LO   \.text\+0x00040144
-0003000[02] R_PPC_ADDR16_LO   \.text\+0x00040144
-0004000[02] R_PPC_ADDR16_HA   \.text\+0x00040144
-0004001[02] R_PPC_ADDR16_HA   \.text\+0x00040144
-0004002[02] R_PPC_ADDR16_HA   \.text\+0x00040144
+0001000[02] R_PPC_ADDR16_LO   \.text\+0x00050000
+0002000[02] R_PPC_ADDR16_LO   \.text\+0x00050000
+0003000[02] R_PPC_ADDR16_LO   \.text\+0x00050000
+0004000[02] R_PPC_ADDR16_HA   \.text\+0x00050000
+0004001[02] R_PPC_ADDR16_HA   \.text\+0x00050000
+0004002[02] R_PPC_ADDR16_HA   \.text\+0x00050000

--
Alan Modra
Australia Development Lab, IBM
Reply | Threaded
Open this post in threaded view
|

Re: testsuite results, binutils-2.29 branch

Jiong Wang-4
In reply to this post by Matthias Klose-6
On 06/07/17 13:50, Matthias Klose wrote:

> Also found that glibc-2.24 fails to build
> in the check target on aarch64 and arm32 with the new binutils (see PR21725).

Both reproduced, I have updated the bugzilla about some initial investigation on this.

For aarch64, it should be the same issue ashttps://sourceware.org/ml/binutils/2017-06/msg00226.html, glibc needs backport the fix.

For arm32, from backtrace, it seems to be a different issue which I will have a look.

Reply | Threaded
Open this post in threaded view
|

Re: testsuite results, binutils-2.29 branch

Jiong Wang-4
In reply to this post by Matthias Klose-6
On 06/07/17 13:50, Matthias Klose wrote:
> arm64:
> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-aarch64/aarch64-elf.exp ...
> FAIL: ld-aarch64/dt_textrel

The tested Binutils was configured with --enable-new-dtags, so it generated extra
dtag.  "dt_textrel.d" was using ".*" to filter remaining unrelated results while
".*" seems only works on single line, we should use #pass there to be more
robust.

I will commit attached patch as obivious.

> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-ifunc/ifunc.exp ...
> FAIL: Could not link a dynamic executable using local ifunc

Can't reproduce using latest GCC, I guess it's what's Alan explained at:

https://sourceware.org/ml/binutils/2017-07/msg00086.html

Thanks for reporting this.

ld/
2017-07-11  Jiong Wang  <[hidden email]>

        * testsuite/ld-aarch64/dt_textrel.d: Use "#pass" instead of ".*" to
         filter out remaining lines.


1.patch (344 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: testsuite results, binutils-2.29 branch

Matthias Klose-6
On 11.07.2017 12:27, Jiong Wang wrote:

> On 06/07/17 13:50, Matthias Klose wrote:
>> arm64:
>> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-aarch64/aarch64-elf.exp ...
>> FAIL: ld-aarch64/dt_textrel
>
> The tested Binutils was configured with --enable-new-dtags, so it generated extra
> dtag.  "dt_textrel.d" was using ".*" to filter remaining unrelated results while
> ".*" seems only works on single line, we should use #pass there to be more
> robust.
>
> I will commit attached patch as obivious.
>
>> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-ifunc/ifunc.exp ...
>> FAIL: Could not link a dynamic executable using local ifunc

thanks! Please could you commit this to the 2.29 branch as well, if Tristan
approves?

Matthias
Reply | Threaded
Open this post in threaded view
|

Re: testsuite results, binutils-2.29 branch

Jiong Wang-4
On 11/07/17 11:33, Matthias Klose wrote:

> On 11.07.2017 12:27, Jiong Wang wrote:
>> On 06/07/17 13:50, Matthias Klose wrote:
>>> arm64:
>>> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-aarch64/aarch64-elf.exp ...
>>> FAIL: ld-aarch64/dt_textrel
>> The tested Binutils was configured with --enable-new-dtags, so it generated extra
>> dtag.  "dt_textrel.d" was using ".*" to filter remaining unrelated results while
>> ".*" seems only works on single line, we should use #pass there to be more
>> robust.
>>
>> I will commit attached patch as obivious.
>>
>>> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-ifunc/ifunc.exp ...
>>> FAIL: Could not link a dynamic executable using local ifunc
> thanks! Please could you commit this to the 2.29 branch as well, if Tristan
> approves?

Sure and I have committed to 2.29 branch as I feel it's quite obvious to backport.

Reply | Threaded
Open this post in threaded view
|

Re: testsuite results, binutils-2.29 branch

Tristan Gingold-2
In reply to this post by Matthias Klose-6
On 11/07/2017 12:33, Matthias Klose wrote:

> On 11.07.2017 12:27, Jiong Wang wrote:
>> On 06/07/17 13:50, Matthias Klose wrote:
>>> arm64:
>>> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-aarch64/aarch64-elf.exp ...
>>> FAIL: ld-aarch64/dt_textrel
>>
>> The tested Binutils was configured with --enable-new-dtags, so it generated extra
>> dtag.  "dt_textrel.d" was using ".*" to filter remaining unrelated results while
>> ".*" seems only works on single line, we should use #pass there to be more
>> robust.
>>
>> I will commit attached patch as obivious.
>>
>>> Running /<<PKGBUILDDIR>>/ld/testsuite/ld-ifunc/ifunc.exp ...
>>> FAIL: Could not link a dynamic executable using local ifunc
>
> thanks! Please could you commit this to the 2.29 branch as well, if Tristan
> approves?

Ok for me.


Reply | Threaded
Open this post in threaded view
|

Re: testsuite results, binutils-2.29 branch

Jiong Wang-4
In reply to this post by Jiong Wang-4
On 07/07/17 15:27, Jiong Wang wrote:

> On 06/07/17 13:50, Matthias Klose wrote:
>
>> Also found that glibc-2.24 fails to build
>> in the check target on aarch64 and arm32 with the new binutils (see
>> PR21725).
>
> Both reproduced, I have updated the bugzilla about some initial
> investigation on this.
>
> For aarch64, it should be the same issue
> ashttps://sourceware.org/ml/binutils/2017-06/msg00226.html, glibc
> needs backport the fix.
>
> For arm32, from backtrace, it seems to be a different issue which I
> will have a look.

Confirmed GLIBC ARM backend ought to be fixed, details please see

   https://sourceware.org/bugzilla/show_bug.cgi?id=21725

I have also confirmed GLIBC master may suffer from the same issue.

GLIBC fix posted at:

   https://sourceware.org/ml/libc-alpha/2017-07/msg00518.html