aarch64: Disable A64FX memcpy/memmove BTI unconditionally
Checks
Context |
Check |
Description |
dj/TryBot-apply_patch |
success
|
Patch applied to master at the time it was sent
|
dj/TryBot-32bit |
success
|
Build for i686
|
Commit Message
This patch disables A64FX memcpy/memmove BTI instruction insertion
unconditionally such as A64FX memset patch [1] for performance.
[1] commit 07b427296b8d59f439144029d9a948f6c1ce0a31
---
sysdeps/aarch64/multiarch/memcpy_a64fx.S | 3 +++
1 file changed, 3 insertions(+)
Comments
The 09/24/2021 07:49, Naohiro Tamura wrote:
> This patch disables A64FX memcpy/memmove BTI instruction insertion
> unconditionally such as A64FX memset patch [1] for performance.
>
> [1] commit 07b427296b8d59f439144029d9a948f6c1ce0a31
this is ok to commit, i assume you tested that it does
not unfavourably misalign some critical loop. i can
commit this for you if you want.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
> ---
> sysdeps/aarch64/multiarch/memcpy_a64fx.S | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/sysdeps/aarch64/multiarch/memcpy_a64fx.S b/sysdeps/aarch64/multiarch/memcpy_a64fx.S
> index 65528405bb12..ae7464e09f42 100644
> --- a/sysdeps/aarch64/multiarch/memcpy_a64fx.S
> +++ b/sysdeps/aarch64/multiarch/memcpy_a64fx.S
> @@ -19,6 +19,9 @@
>
> #include <sysdep.h>
>
> +#undef BTI_C
> +#define BTI_C
> +
> /* Assumptions:
> *
> * ARMv8.2-a, AArch64, unaligned accesses, sve
> --
> 2.17.1
>
Hi Szabolcs,
> The 09/24/2021 07:49, Naohiro Tamura wrote:
> > This patch disables A64FX memcpy/memmove BTI instruction insertion
> > unconditionally such as A64FX memset patch [1] for performance.
> >
> > [1] commit 07b427296b8d59f439144029d9a948f6c1ce0a31
>
> this is ok to commit, i assume you tested that it does
> not unfavourably misalign some critical loop. i can
> commit this for you if you want.
Thank you for the review.
Yes, I've tested and confirmed no problem.
Please merge it for me.
Thanks.
Naohiro
@@ -19,6 +19,9 @@
#include <sysdep.h>
+#undef BTI_C
+#define BTI_C
+
/* Assumptions:
*
* ARMv8.2-a, AArch64, unaligned accesses, sve