mbox series

[0/2] aarch64: Fix variant PCS support [BZ #26978]

Message ID cover.1603909005.git.szabolcs.nagy@arm.com
Headers
Series aarch64: Fix variant PCS support [BZ #26978] |

Message

Szabolcs Nagy Oct. 29, 2020, 8:39 a.m. UTC 
  My original implementation for variant PCS support has a bug
that makes it ineffective.

I plan to backport this fix to the release branches after some
testing, the test case is separate because it requires configure
changes to check binutils support and thus harder to backport.

Szabolcs Nagy (2):
  aarch64: Fix DT_AARCH64_VARIANT_PCS handling [BZ #26798]
  aarch64: Add variant PCS lazy binding test [BZ #26798]

 sysdeps/aarch64/Makefile       |   7 ++
 sysdeps/aarch64/configure      |  40 ++++++++++
 sysdeps/aarch64/configure.ac   |  22 +++++
 sysdeps/aarch64/dl-machine.h   |  12 +--
 sysdeps/aarch64/tst-vpcs-mod.S | 141 +++++++++++++++++++++++++++++++++
 sysdeps/aarch64/tst-vpcs.c     |  78 ++++++++++++++++++
 6 files changed, 292 insertions(+), 8 deletions(-)
 create mode 100644 sysdeps/aarch64/tst-vpcs-mod.S
 create mode 100644 sysdeps/aarch64/tst-vpcs.c

Comments

Szabolcs Nagy Nov. 2, 2020, 9:43 a.m. UTC | #1 
The 10/29/2020 08:39, Szabolcs Nagy via Libc-alpha wrote:
> My original implementation for variant PCS support has a bug
> that makes it ineffective.
> 
> I plan to backport this fix to the release branches after some
> testing, the test case is separate because it requires configure
> changes to check binutils support and thus harder to backport.

these patches are committed to master now.

commit 558251bd8785760ad40fcbfeaaee5d27fa5b0fe4
commit e156dabc766d6f6f99ce9402999eae380a3ec1f2


> 
> Szabolcs Nagy (2):
>   aarch64: Fix DT_AARCH64_VARIANT_PCS handling [BZ #26798]
>   aarch64: Add variant PCS lazy binding test [BZ #26798]
> 
>  sysdeps/aarch64/Makefile       |   7 ++
>  sysdeps/aarch64/configure      |  40 ++++++++++
>  sysdeps/aarch64/configure.ac   |  22 +++++
>  sysdeps/aarch64/dl-machine.h   |  12 +--
>  sysdeps/aarch64/tst-vpcs-mod.S | 141 +++++++++++++++++++++++++++++++++
>  sysdeps/aarch64/tst-vpcs.c     |  78 ++++++++++++++++++
>  6 files changed, 292 insertions(+), 8 deletions(-)
>  create mode 100644 sysdeps/aarch64/tst-vpcs-mod.S
>  create mode 100644 sysdeps/aarch64/tst-vpcs.c
> 
> -- 
> 2.17.1
>
Florian Weimer Nov. 2, 2020, 12:35 p.m. UTC | #2 
* Szabolcs Nagy via Libc-alpha:

> The 10/29/2020 08:39, Szabolcs Nagy via Libc-alpha wrote:
>> My original implementation for variant PCS support has a bug
>> that makes it ineffective.
>> 
>> I plan to backport this fix to the release branches after some
>> testing, the test case is separate because it requires configure
>> changes to check binutils support and thus harder to backport.
>
> these patches are committed to master now.
>
> commit 558251bd8785760ad40fcbfeaaee5d27fa5b0fe4
> commit e156dabc766d6f6f99ce9402999eae380a3ec1f2

Do you plan to backport them to recent stable branches?

Thanks,
Florian
Szabolcs Nagy Nov. 2, 2020, 2:52 p.m. UTC | #3 
The 11/02/2020 13:35, Florian Weimer wrote:
> * Szabolcs Nagy via Libc-alpha:
> 
> > The 10/29/2020 08:39, Szabolcs Nagy via Libc-alpha wrote:
> >> My original implementation for variant PCS support has a bug
> >> that makes it ineffective.
> >> 
> >> I plan to backport this fix to the release branches after some
> >> testing, the test case is separate because it requires configure
> >> changes to check binutils support and thus harder to backport.
> >
> > these patches are committed to master now.
> >
> > commit 558251bd8785760ad40fcbfeaaee5d27fa5b0fe4
> > commit e156dabc766d6f6f99ce9402999eae380a3ec1f2
> 
> Do you plan to backport them to recent stable branches?

yes, i will wait two days and then backport these
to release branches that have broken vpcs support.