diff mbox

BZ #19590: Fixed build of shared objects that use libmvec.so functions

Message ID CAMXFM3veSy9SCa=Ke15zON_h+_TD7eK3k8QNS2mrqfBKDDs+vA@mail.gmail.com
State New, archived
Headers show

Commit Message

Andrew Senkevich Feb. 17, 2016, 9:04 p.m. UTC
2016-02-17 23:06 GMT+03:00 H.J. Lu <hjl.tools@gmail.com>:
> On Wed, Feb 17, 2016 at 11:30 AM, Andrew Senkevich
> <andrew.n.senkevich@gmail.com> wrote:
>> 2016-02-17 21:44 GMT+03:00 Joseph Myers <joseph@codesourcery.com>:
>>> On Wed, 17 Feb 2016, Andrew Senkevich wrote:
>>>
>>>> Is attached version finally Ok for trunk?
>>>
>>> Please resubmit, with the coding style issues fixed and tests moved to
>>> test-skeleton.c unless there's a reason they can't use it, after the
>>> freeze is over.  This is much too risky to consider during the freeze, and
>>> would need to wait a while on master before any backports could be
>>> considered (e.g. the new tests could break the testsuite build in some
>>> configurations for unforeseen reasons).
>>
>> May be separate fix and tests to commit fix earlier?
>>
>
> This is a good idea.  Please submit a fix-only patch.

2016-02-18  Andrew Senkevich  <andrew.senkevich@intel.com>
                    H.J. Lu  <hongjiu.lu@intel.com>

        [BZ #19590]
        * sysdeps/x86_64/fpu/svml_finite_alias.S (ALIAS_IMPL): Use PIC
        relocation.



Ok to commit?


--
WBR,
Andrew

Comments

H.J. Lu Feb. 17, 2016, 9:36 p.m. UTC | #1
On Wed, Feb 17, 2016 at 1:04 PM, Andrew Senkevich
<andrew.n.senkevich@gmail.com> wrote:
> 2016-02-17 23:06 GMT+03:00 H.J. Lu <hjl.tools@gmail.com>:
>> On Wed, Feb 17, 2016 at 11:30 AM, Andrew Senkevich
>> <andrew.n.senkevich@gmail.com> wrote:
>>> 2016-02-17 21:44 GMT+03:00 Joseph Myers <joseph@codesourcery.com>:
>>>> On Wed, 17 Feb 2016, Andrew Senkevich wrote:
>>>>
>>>>> Is attached version finally Ok for trunk?
>>>>
>>>> Please resubmit, with the coding style issues fixed and tests moved to
>>>> test-skeleton.c unless there's a reason they can't use it, after the
>>>> freeze is over.  This is much too risky to consider during the freeze, and
>>>> would need to wait a while on master before any backports could be
>>>> considered (e.g. the new tests could break the testsuite build in some
>>>> configurations for unforeseen reasons).
>>>
>>> May be separate fix and tests to commit fix earlier?
>>>
>>
>> This is a good idea.  Please submit a fix-only patch.
>
> 2016-02-18  Andrew Senkevich  <andrew.senkevich@intel.com>
>                     H.J. Lu  <hongjiu.lu@intel.com>
>
>         [BZ #19590]
>         * sysdeps/x86_64/fpu/svml_finite_alias.S (ALIAS_IMPL): Use PIC
>         relocation.
>
> diff --git a/sysdeps/x86_64/fpu/svml_finite_alias.S
> b/sysdeps/x86_64/fpu/svml_finite_alias.S
> index 0062fe4..2dcfc37 100644
> --- a/sysdeps/x86_64/fpu/svml_finite_alias.S
> +++ b/sysdeps/x86_64/fpu/svml_finite_alias.S
> @@ -23,8 +23,7 @@
>
>  #define ALIAS_IMPL(alias, target) \
>  ENTRY (alias); \
> - call target; \
> - ret; \
> + jmp *target@GOTPCREL(%rip); \
>  END (alias)
>
>   .text
>
>
> Ok to commit?
>

OK.  Please leave BZ 19590 open until testcase is added.

Thanks.
Joseph Myers Feb. 17, 2016, 9:44 p.m. UTC | #2
On Wed, 17 Feb 2016, H.J. Lu wrote:

> OK.  Please leave BZ 19590 open until testcase is added.

No, when a bug is fixed it needs to be closed immediately as FIXED with 
the milestone set accordingly, otherwise the list of fixed bugs in NEWS 
will be wrong.  The matter of the testcase needs to be tracked separately.
H.J. Lu Feb. 17, 2016, 10:12 p.m. UTC | #3
On Wed, Feb 17, 2016 at 1:44 PM, Joseph Myers <joseph@codesourcery.com> wrote:
> On Wed, 17 Feb 2016, H.J. Lu wrote:
>
>> OK.  Please leave BZ 19590 open until testcase is added.
>
> No, when a bug is fixed it needs to be closed immediately as FIXED with
> the milestone set accordingly, otherwise the list of fixed bugs in NEWS
> will be wrong.  The matter of the testcase needs to be tracked separately.

Open a new bug report for testcase.
H.J. Lu Feb. 17, 2016, 10:45 p.m. UTC | #4
On Wed, Feb 17, 2016 at 2:12 PM, H.J. Lu <hjl.tools@gmail.com> wrote:
> On Wed, Feb 17, 2016 at 1:44 PM, Joseph Myers <joseph@codesourcery.com> wrote:
>> On Wed, 17 Feb 2016, H.J. Lu wrote:
>>
>>> OK.  Please leave BZ 19590 open until testcase is added.
>>
>> No, when a bug is fixed it needs to be closed immediately as FIXED with
>> the milestone set accordingly, otherwise the list of fixed bugs in NEWS
>> will be wrong.  The matter of the testcase needs to be tracked separately.
>
> Open a new bug report for testcase.

I checked in the fix for Andrew, resolved:

https://sourceware.org/bugzilla/show_bug.cgi?id=19590

and opened:

https://sourceware.org/bugzilla/show_bug.cgi?id=19654
diff mbox

Patch

diff --git a/sysdeps/x86_64/fpu/svml_finite_alias.S
b/sysdeps/x86_64/fpu/svml_finite_alias.S
index 0062fe4..2dcfc37 100644
--- a/sysdeps/x86_64/fpu/svml_finite_alias.S
+++ b/sysdeps/x86_64/fpu/svml_finite_alias.S
@@ -23,8 +23,7 @@ 

 #define ALIAS_IMPL(alias, target) \
 ENTRY (alias); \
- call target; \
- ret; \
+ jmp *target@GOTPCREL(%rip); \
 END (alias)

  .text