Re: [PATCH] aarch64: fix errno address calculation in SYSCALL_ERROR_HANDLER()

[Yury Norov <ynorov at caviumnetworks dot com>]

On Tue, Feb 07, 2017 at 02:42:25PM -0200, Adhemerval Zanella wrote:
> 
> 
> On 07/02/2017 12:54, Yury Norov wrote:
> > On Tue, Feb 07, 2017 at 10:10:42AM -0200, Adhemerval Zanella wrote:
> >> If it is an user visible issue it requires a bugzilla report (although I am not
> >> sure in this situation since I think this issue should arise only for ilp32).
> > 
> > This is not a user-visible bug because arm64/ilp32 is not a user
> > visible feature at now. If you think that I should create a bug, I can do it.
> 
> Right, aarch64 lp64 does not trigger so I think we may proceed without
> opening a bug report.

 [...]
 
> > If you think there should be the specific test for this case, I can
> > write it. But I think it should be the test for pseudo syscalls
> > wrapper, not the part of mmap() or write() test.
> 
> My only concern is if and why glibc own testsuite did not trigger this
> kind of issue in any tests and in this case if it was by chance or lack
> of coverage. If glibc own testsuite does not trigger this in any case,
> I think a regression should be added.

OK, I'll write some test tomorrow. (Too late in my place)

Is the patch fine by itself? Can you (someone) push it?

Yury