On Thu, Jul 4, 2024 at 11:00 AM Hongtao Liu wrote:
>
> On Tue, Jul 2, 2024 at 11:24 AM Hongyu Wang wrote:
> >
> > Hi,
> >
> > According to APX spec, the pushp/popp pairs should be matched,
> > otherwise the PPX hint cannot take effect and cause performance loss.
> >
> > In the ix86_expand_epilogu
On Tue, Jul 2, 2024 at 11:24 AM Hongyu Wang wrote:
>
> Hi,
>
> According to APX spec, the pushp/popp pairs should be matched,
> otherwise the PPX hint cannot take effect and cause performance loss.
>
> In the ix86_expand_epilogue, there are several optimizations that may
> cause the epilogue using
> Isn't it true in general that mismatched push/pop confuses a CPUs stack
> engine? Does the distance of the push/pop make any difference here
> or in general?
We are not sure about the impact for distance, but the PPX hint was
just a performance hint. As mentioned in the apx spec, the mismatched
On Tue, Jul 2, 2024 at 5:24 AM Hongyu Wang wrote:
>
> Hi,
>
> According to APX spec, the pushp/popp pairs should be matched,
> otherwise the PPX hint cannot take effect and cause performance loss.
>
> In the ix86_expand_epilogue, there are several optimizations that may
> cause the epilogue using
Hi,
According to APX spec, the pushp/popp pairs should be matched,
otherwise the PPX hint cannot take effect and cause performance loss.
In the ix86_expand_epilogue, there are several optimizations that may
cause the epilogue using mov to restore the regs. Check if PPX applied
and prevent usage o
