On Jun 23 06:29, Radek Barton via Cygwin-patches wrote:
> Hello.
> 
> Thank you for pointing this out. I didn't know this was related to ABI
> calling conventions in this particular case. Simple removal of the
> `sub` instruction is causing tests regressions in our development
> branch [...]

Just removing the sub expression seems wrong anyway because the code
would neglect to set the sp register, isn't it?


Thanks,
Corinna


> current upstream master branch (induced by recent changes to pthread).
> It will take some time to investigate and validate this change.
> 
> Thank you for your patience,
> 
> Radek
> 
> ________________________________________
> From: Jeremy Drake <[email protected]>
> Sent: Saturday, June 21, 2025 8:47 PM
> To: Radek Barton <[email protected]>
> Cc: [email protected] <[email protected]>
> Subject: [EXTERNAL] Re: [PATCH] Cygwin: stack base initialization for AArch64
>  
> On Wed, 18 Jun 2025, Jeremy Drake via Cygwin-patches wrote:
> 
> > On Wed, 18 Jun 2025, Radek Barton via Cygwin-patches wrote:
> >
> > > -#ifdef __x86_64__
> > >            /* Set stack pointer to new address.  Set frame pointer to
> > >               stack pointer and subtract 32 bytes for shadow space. */
> > > +#if defined(__x86_64__)
> > >            __asm__ ("\n\
> > >                     movq %[ADDR], %%rsp \n\
> > >                     movq  %%rsp, %%rbp  \n\
> > >                     subq  $32,%%rsp     \n"
> > >                     : : [ADDR] "r" (stackaddr));
> > > +#elif defined(__aarch64__)
> > > +         __asm__ ("\n\
> > > +                  mov fp, %[ADDR] \n\
> > > +                  sub sp, fp, #32 \n"
> >
> > Is the 32-byte shadow space part of the aarch64 calling convention spec,
> > or is this just copying what x86_64 was doing?  My impression is that this
> > space was part of the x86_64 calling convention.
> 
> The patch for pthread stack initialization dropped the 32-byte shadow
> space, and I believe this patch should as well.

Reply via email to