[PATCH] aarch64: fix cfi directives around __libc_arm_za_disable

[Yury Khrustalev]

On Wed, Oct 29, 2025 at 03:44:54PM -0300, Thiago Jung Bauermann wrote:
> Hello Yury,
> 
> Sorry for the delay, it was a holiday here yesterday.

No worries, thanks for the help with this regression!

> Yury Khrustalev <yury.khrustalev@arm.com> writes:
> 
> > On Tue, Oct 28, 2025 at 09:47:04AM -0300, Adhemerval Zanella Netto wrote:
> >> 
> >> On 28/10/25 08:50, Yury Khrustalev wrote:
> >> > Incorrect CFI directive corrupted call stack information
> >> > and prevented debuggers from correctly displaying call
> >> > stack information.
> >> > 
> >> > ---
> >> > 
> >> > Passes regression on AArch64. OK for trunk?
> >> > Base commit: 013f5167b9
> >> 
> >> LGTM, thanks.  I take this fixes the gdb regression triggered by Linaro CI, right?
> >
> > Yes, it fixes this regression https://linaro.atlassian.net/browse/GNU-1706.
> >
> > Thiago, would you be able to check if the patch fixes all these
> > regressions? I couldn't reproduce failures in the GDB testsuite.
> 
> Yes, the patch fixed the regressions. Thanks!

Thank you, I've pushed the patch to master.

Kind regards,
Yury