x86: Fix an unknown vector operation in memchr-evex.S

Message ID CAAd+X4Wmo-g5sKS1h8jD5bA94pM8iUg7vUk71mKTY8MU7vvGHw@mail.gmail.com
State New
Headers show
Series
  • x86: Fix an unknown vector operation in memchr-evex.S
Related show

Commit Message

H.J. Lu via Libc-alpha May 8, 2021, 1:23 a.m.
An unknown vector operation occurred in commit 2a76821c308. Fixed it
by using "ymm{k1}{z}" but not "ymm {k1} {z}".

---
 sysdeps/x86_64/multiarch/memchr-evex.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--
2.25.1

Comments

H.J. Lu via Libc-alpha May 8, 2021, 1:32 a.m. | #1
On Fri, May 7, 2021 at 6:23 PM Alice Xu <alice.d.xu@gmail.com> wrote:
>

> An unknown vector operation occurred in commit 2a76821c308. Fixed it

> by using "ymm{k1}{z}" but not "ymm {k1} {z}".

>

> ---

>  sysdeps/x86_64/multiarch/memchr-evex.S | 2 +-

>  1 file changed, 1 insertion(+), 1 deletion(-)

>

> diff --git a/sysdeps/x86_64/multiarch/memchr-evex.S b/sysdeps/x86_64/multiarch/memchr-evex.S

> index 81d5cd6486..f3fdad4fda 100644

> --- a/sysdeps/x86_64/multiarch/memchr-evex.S

> +++ b/sysdeps/x86_64/multiarch/memchr-evex.S

> @@ -271,7 +271,7 @@ L(loop_4x_vec):

>         vpxorq  (VEC_SIZE * 6)(%rdi), %YMMMATCH, %YMM3

>         VPCMP   $0, (VEC_SIZE * 7)(%rdi), %YMMMATCH, %k3

>         /* Reduce VEC2 / VEC3 with min and VEC1 with zero mask.  */

> -       VPMINU  %YMM2, %YMM3, %YMM3 {%k1} {z}

> +       VPMINU  %YMM2, %YMM3, %YMM3{%k1}{z}

>         VPCMP   $0, %YMM3, %YMMZERO, %k2

>  # ifdef USE_AS_RAWMEMCHR

>         subq    $-(VEC_SIZE * 4), %rdi

> --

> 2.25.1


LGTM.

Reviewed-by: H.J. Lu <hjl.tools@gmail.com>


Do you want me to commit it for you?

Thanks.

-- 
H.J.
H.J. Lu via Libc-alpha May 8, 2021, 1:55 a.m. | #2
Yes, please help to commit it. Thanks.

Regards

Alice

On Sat, May 8, 2021 at 9:32 AM H.J. Lu <hjl.tools@gmail.com> wrote:

> On Fri, May 7, 2021 at 6:23 PM Alice Xu <alice.d.xu@gmail.com> wrote:

> >

> > An unknown vector operation occurred in commit 2a76821c308. Fixed it

> > by using "ymm{k1}{z}" but not "ymm {k1} {z}".

> >

> > ---

> >  sysdeps/x86_64/multiarch/memchr-evex.S | 2 +-

> >  1 file changed, 1 insertion(+), 1 deletion(-)

> >

> > diff --git a/sysdeps/x86_64/multiarch/memchr-evex.S

> b/sysdeps/x86_64/multiarch/memchr-evex.S

> > index 81d5cd6486..f3fdad4fda 100644

> > --- a/sysdeps/x86_64/multiarch/memchr-evex.S

> > +++ b/sysdeps/x86_64/multiarch/memchr-evex.S

> > @@ -271,7 +271,7 @@ L(loop_4x_vec):

> >         vpxorq  (VEC_SIZE * 6)(%rdi), %YMMMATCH, %YMM3

> >         VPCMP   $0, (VEC_SIZE * 7)(%rdi), %YMMMATCH, %k3

> >         /* Reduce VEC2 / VEC3 with min and VEC1 with zero mask.  */

> > -       VPMINU  %YMM2, %YMM3, %YMM3 {%k1} {z}

> > +       VPMINU  %YMM2, %YMM3, %YMM3{%k1}{z}

> >         VPCMP   $0, %YMM3, %YMMZERO, %k2

> >  # ifdef USE_AS_RAWMEMCHR

> >         subq    $-(VEC_SIZE * 4), %rdi

> > --

> > 2.25.1

>

> LGTM.

>

> Reviewed-by: H.J. Lu <hjl.tools@gmail.com>

>

> Do you want me to commit it for you?

>

> Thanks.

>

> --

> H.J.

>
H.J. Lu via Libc-alpha May 8, 2021, 2 a.m. | #3
On Fri, May 7, 2021 at 6:55 PM Alice Xu <alice.d.xu@gmail.com> wrote:
>

> Yes, please help to commit it. Thanks.


I will commit it for you.  Next time, please use "git format-patch" to create
patches.

> Regards

>

> Alice

>

> On Sat, May 8, 2021 at 9:32 AM H.J. Lu <hjl.tools@gmail.com> wrote:

>>

>> On Fri, May 7, 2021 at 6:23 PM Alice Xu <alice.d.xu@gmail.com> wrote:

>> >

>> > An unknown vector operation occurred in commit 2a76821c308. Fixed it

>> > by using "ymm{k1}{z}" but not "ymm {k1} {z}".

>> >

>> > ---

>> >  sysdeps/x86_64/multiarch/memchr-evex.S | 2 +-

>> >  1 file changed, 1 insertion(+), 1 deletion(-)

>> >

>> > diff --git a/sysdeps/x86_64/multiarch/memchr-evex.S b/sysdeps/x86_64/multiarch/memchr-evex.S

>> > index 81d5cd6486..f3fdad4fda 100644

>> > --- a/sysdeps/x86_64/multiarch/memchr-evex.S

>> > +++ b/sysdeps/x86_64/multiarch/memchr-evex.S

>> > @@ -271,7 +271,7 @@ L(loop_4x_vec):

>> >         vpxorq  (VEC_SIZE * 6)(%rdi), %YMMMATCH, %YMM3

>> >         VPCMP   $0, (VEC_SIZE * 7)(%rdi), %YMMMATCH, %k3

>> >         /* Reduce VEC2 / VEC3 with min and VEC1 with zero mask.  */

>> > -       VPMINU  %YMM2, %YMM3, %YMM3 {%k1} {z}

>> > +       VPMINU  %YMM2, %YMM3, %YMM3{%k1}{z}

>> >         VPCMP   $0, %YMM3, %YMMZERO, %k2

>> >  # ifdef USE_AS_RAWMEMCHR

>> >         subq    $-(VEC_SIZE * 4), %rdi

>> > --

>> > 2.25.1

>>

>> LGTM.

>>

>> Reviewed-by: H.J. Lu <hjl.tools@gmail.com>

>>

>> Do you want me to commit it for you?

>>

>> Thanks.

>>

>> --

>> H.J.




-- 
H.J.
H.J. Lu via Libc-alpha May 8, 2021, 2:34 a.m. | #4
On Fri, May 7, 2021 at 7:00 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>

> On Fri, May 7, 2021 at 6:55 PM Alice Xu <alice.d.xu@gmail.com> wrote:

> >

> > Yes, please help to commit it. Thanks.

>

> I will commit it for you.  Next time, please use "git format-patch" to create

> patches.

>


This is the patch I am checking in for you after fixing indentation and
space.

-- 
H.J.
From 6ea916adfa0ab9af6e7dc6adcf6f977dfe017835 Mon Sep 17 00:00:00 2001
From: Alice Xu <alice.d.xu@gmail.com>
Date: Fri, 7 May 2021 19:03:21 -0700
Subject: [PATCH] x86-64: Fix an unknown vector operation in memchr-evex.S

An unknown vector operation occurred in commit 2a76821c308. Fixed it
by using "ymm{k1}{z}" but not "ymm {k1} {z}".

Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
---
 sysdeps/x86_64/multiarch/memchr-evex.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sysdeps/x86_64/multiarch/memchr-evex.S b/sysdeps/x86_64/multiarch/memchr-evex.S
index 81d5cd6486..f3fdad4fda 100644
--- a/sysdeps/x86_64/multiarch/memchr-evex.S
+++ b/sysdeps/x86_64/multiarch/memchr-evex.S
@@ -271,7 +271,7 @@ L(loop_4x_vec):
 	vpxorq	(VEC_SIZE * 6)(%rdi), %YMMMATCH, %YMM3
 	VPCMP	$0, (VEC_SIZE * 7)(%rdi), %YMMMATCH, %k3
 	/* Reduce VEC2 / VEC3 with min and VEC1 with zero mask.  */
-	VPMINU	%YMM2, %YMM3, %YMM3 {%k1} {z}
+	VPMINU	%YMM2, %YMM3, %YMM3{%k1}{z}
 	VPCMP	$0, %YMM3, %YMMZERO, %k2
 # ifdef USE_AS_RAWMEMCHR
 	subq	$-(VEC_SIZE * 4), %rdi
H.J. Lu via Libc-alpha May 8, 2021, 3:17 a.m. | #5
Thanks. I used " git format-patch  ", but when I used git send-email to
send the patch, I got an SMTP server error, so I use my web Gmail to send
it. Next time I'll attach the patch file inner mail.
Thanks a lot.

Regards
Alice.

On Sat, May 8, 2021 at 10:35 AM H.J. Lu <hjl.tools@gmail.com> wrote:

> On Fri, May 7, 2021 at 7:00 PM H.J. Lu <hjl.tools@gmail.com> wrote:

> >

> > On Fri, May 7, 2021 at 6:55 PM Alice Xu <alice.d.xu@gmail.com> wrote:

> > >

> > > Yes, please help to commit it. Thanks.

> >

> > I will commit it for you.  Next time, please use "git format-patch" to

> create

> > patches.

> >

>

> This is the patch I am checking in for you after fixing indentation and

> space.

>

> --

> H.J.

>

Patch

diff --git a/sysdeps/x86_64/multiarch/memchr-evex.S
b/sysdeps/x86_64/multiarch/memchr-evex.S
index 81d5cd6486..f3fdad4fda 100644
--- a/sysdeps/x86_64/multiarch/memchr-evex.S
+++ b/sysdeps/x86_64/multiarch/memchr-evex.S
@@ -271,7 +271,7 @@  L(loop_4x_vec):
        vpxorq  (VEC_SIZE * 6)(%rdi), %YMMMATCH, %YMM3
        VPCMP   $0, (VEC_SIZE * 7)(%rdi), %YMMMATCH, %k3
        /* Reduce VEC2 / VEC3 with min and VEC1 with zero mask.  */
-       VPMINU  %YMM2, %YMM3, %YMM3 {%k1} {z}
+       VPMINU  %YMM2, %YMM3, %YMM3{%k1}{z}
        VPCMP   $0, %YMM3, %YMMZERO, %k2
 # ifdef USE_AS_RAWMEMCHR
        subq    $-(VEC_SIZE * 4), %rdi