* [PATCH] i386: Fix wrong result for AMX-TILE intrinsic when parsing expression.
@ 2021-11-04 1:18 Hongyu Wang
2021-11-04 2:24 ` Hongtao Liu
0 siblings, 1 reply; 4+ messages in thread
From: Hongyu Wang @ 2021-11-04 1:18 UTC (permalink / raw)
To: hongtao.liu; +Cc: gcc-patches
Hi,
_tile_loadd, _tile_stored, _tile_streamloadd intrinsics are defined by
macro, so the parameters should be wrapped by parentheses to accept
expressions.
Bootstraped/regtested on x86_64-pc-linux-gnu{-m32,} and sde.
OK for master and backport to GCC11 branch?
gcc/ChangeLog:
* config/i386/amxtileintrin.h (_tile_loadd_internal): Add
parentheses to base and stride.
(_tile_stream_loadd_internal): Likewise.
(_tile_stored_internal): Likewise.
---
gcc/config/i386/amxtileintrin.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/gcc/config/i386/amxtileintrin.h b/gcc/config/i386/amxtileintrin.h
index 75d784ad160..3a0a6b44c17 100644
--- a/gcc/config/i386/amxtileintrin.h
+++ b/gcc/config/i386/amxtileintrin.h
@@ -62,7 +62,7 @@ _tile_release (void)
#define _tile_loadd_internal(dst,base,stride) \
__asm__ volatile \
("{tileloadd\t(%0,%1,1), %%tmm"#dst"|tileloadd\t%%tmm"#dst", [%0+%1*1]}" \
- :: "r" ((const void*) base), "r" ((long) stride))
+ :: "r" ((const void*) (base)), "r" ((long) (stride)))
#define _tile_stream_loadd(dst,base,stride) \
_tile_stream_loadd_internal (dst, base, stride)
@@ -70,7 +70,7 @@ _tile_release (void)
#define _tile_stream_loadd_internal(dst,base,stride) \
__asm__ volatile \
("{tileloaddt1\t(%0,%1,1), %%tmm"#dst"|tileloaddt1\t%%tmm"#dst", [%0+%1*1]}" \
- :: "r" ((const void*) base), "r" ((long) stride))
+ :: "r" ((const void*) (base)), "r" ((long) (stride)))
#define _tile_stored(dst,base,stride) \
_tile_stored_internal (dst, base, stride)
@@ -78,7 +78,7 @@ _tile_release (void)
#define _tile_stored_internal(src,base,stride) \
__asm__ volatile \
("{tilestored\t%%tmm"#src", (%0,%1,1)|tilestored\t[%0+%1*1], %%tmm"#src"}" \
- :: "r" ((void*) base), "r" ((long) stride) \
+ :: "r" ((void*) (base)), "r" ((long) (stride)) \
: "memory")
#define _tile_zero(dst) \
--
2.18.1
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] i386: Fix wrong result for AMX-TILE intrinsic when parsing expression.
2021-11-04 1:18 [PATCH] i386: Fix wrong result for AMX-TILE intrinsic when parsing expression Hongyu Wang
@ 2021-11-04 2:24 ` Hongtao Liu
2021-11-04 3:25 ` Hongyu Wang
0 siblings, 1 reply; 4+ messages in thread
From: Hongtao Liu @ 2021-11-04 2:24 UTC (permalink / raw)
To: Hongyu Wang; +Cc: Liu, Hongtao, gcc-patches
On Thu, Nov 4, 2021 at 9:19 AM Hongyu Wang via Gcc-patches
<gcc-patches@gcc.gnu.org> wrote:
>
> Hi,
>
> _tile_loadd, _tile_stored, _tile_streamloadd intrinsics are defined by
> macro, so the parameters should be wrapped by parentheses to accept
> expressions.
>
> Bootstraped/regtested on x86_64-pc-linux-gnu{-m32,} and sde.
>
> OK for master and backport to GCC11 branch?
Could you add a testcase for that?
>
> gcc/ChangeLog:
>
> * config/i386/amxtileintrin.h (_tile_loadd_internal): Add
> parentheses to base and stride.
> (_tile_stream_loadd_internal): Likewise.
> (_tile_stored_internal): Likewise.
> ---
> gcc/config/i386/amxtileintrin.h | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/gcc/config/i386/amxtileintrin.h b/gcc/config/i386/amxtileintrin.h
> index 75d784ad160..3a0a6b44c17 100644
> --- a/gcc/config/i386/amxtileintrin.h
> +++ b/gcc/config/i386/amxtileintrin.h
> @@ -62,7 +62,7 @@ _tile_release (void)
> #define _tile_loadd_internal(dst,base,stride) \
> __asm__ volatile \
> ("{tileloadd\t(%0,%1,1), %%tmm"#dst"|tileloadd\t%%tmm"#dst", [%0+%1*1]}" \
> - :: "r" ((const void*) base), "r" ((long) stride))
> + :: "r" ((const void*) (base)), "r" ((long) (stride)))
>
> #define _tile_stream_loadd(dst,base,stride) \
> _tile_stream_loadd_internal (dst, base, stride)
> @@ -70,7 +70,7 @@ _tile_release (void)
> #define _tile_stream_loadd_internal(dst,base,stride) \
> __asm__ volatile \
> ("{tileloaddt1\t(%0,%1,1), %%tmm"#dst"|tileloaddt1\t%%tmm"#dst", [%0+%1*1]}" \
> - :: "r" ((const void*) base), "r" ((long) stride))
> + :: "r" ((const void*) (base)), "r" ((long) (stride)))
>
> #define _tile_stored(dst,base,stride) \
> _tile_stored_internal (dst, base, stride)
> @@ -78,7 +78,7 @@ _tile_release (void)
> #define _tile_stored_internal(src,base,stride) \
> __asm__ volatile \
> ("{tilestored\t%%tmm"#src", (%0,%1,1)|tilestored\t[%0+%1*1], %%tmm"#src"}" \
> - :: "r" ((void*) base), "r" ((long) stride) \
> + :: "r" ((void*) (base)), "r" ((long) (stride)) \
> : "memory")
>
> #define _tile_zero(dst) \
> --
> 2.18.1
>
--
BR,
Hongtao
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] i386: Fix wrong result for AMX-TILE intrinsic when parsing expression.
2021-11-04 2:24 ` Hongtao Liu
@ 2021-11-04 3:25 ` Hongyu Wang
2021-11-04 4:46 ` Hongtao Liu
0 siblings, 1 reply; 4+ messages in thread
From: Hongyu Wang @ 2021-11-04 3:25 UTC (permalink / raw)
To: Hongtao Liu; +Cc: Hongyu Wang, Liu, Hongtao, gcc-patches
[-- Attachment #1: Type: text/plain, Size: 2788 bytes --]
> Could you add a testcase for that?
Yes, updated patch.
Hongtao Liu via Gcc-patches <gcc-patches@gcc.gnu.org> 于2021年11月4日周四 上午10:25写道:
>
> On Thu, Nov 4, 2021 at 9:19 AM Hongyu Wang via Gcc-patches
> <gcc-patches@gcc.gnu.org> wrote:
> >
> > Hi,
> >
> > _tile_loadd, _tile_stored, _tile_streamloadd intrinsics are defined by
> > macro, so the parameters should be wrapped by parentheses to accept
> > expressions.
> >
> > Bootstraped/regtested on x86_64-pc-linux-gnu{-m32,} and sde.
> >
> > OK for master and backport to GCC11 branch?
> Could you add a testcase for that?
> >
> > gcc/ChangeLog:
> >
> > * config/i386/amxtileintrin.h (_tile_loadd_internal): Add
> > parentheses to base and stride.
> > (_tile_stream_loadd_internal): Likewise.
> > (_tile_stored_internal): Likewise.
> > ---
> > gcc/config/i386/amxtileintrin.h | 6 +++---
> > 1 file changed, 3 insertions(+), 3 deletions(-)
> >
> > diff --git a/gcc/config/i386/amxtileintrin.h b/gcc/config/i386/amxtileintrin.h
> > index 75d784ad160..3a0a6b44c17 100644
> > --- a/gcc/config/i386/amxtileintrin.h
> > +++ b/gcc/config/i386/amxtileintrin.h
> > @@ -62,7 +62,7 @@ _tile_release (void)
> > #define _tile_loadd_internal(dst,base,stride) \
> > __asm__ volatile \
> > ("{tileloadd\t(%0,%1,1), %%tmm"#dst"|tileloadd\t%%tmm"#dst", [%0+%1*1]}" \
> > - :: "r" ((const void*) base), "r" ((long) stride))
> > + :: "r" ((const void*) (base)), "r" ((long) (stride)))
> >
> > #define _tile_stream_loadd(dst,base,stride) \
> > _tile_stream_loadd_internal (dst, base, stride)
> > @@ -70,7 +70,7 @@ _tile_release (void)
> > #define _tile_stream_loadd_internal(dst,base,stride) \
> > __asm__ volatile \
> > ("{tileloaddt1\t(%0,%1,1), %%tmm"#dst"|tileloaddt1\t%%tmm"#dst", [%0+%1*1]}" \
> > - :: "r" ((const void*) base), "r" ((long) stride))
> > + :: "r" ((const void*) (base)), "r" ((long) (stride)))
> >
> > #define _tile_stored(dst,base,stride) \
> > _tile_stored_internal (dst, base, stride)
> > @@ -78,7 +78,7 @@ _tile_release (void)
> > #define _tile_stored_internal(src,base,stride) \
> > __asm__ volatile \
> > ("{tilestored\t%%tmm"#src", (%0,%1,1)|tilestored\t[%0+%1*1], %%tmm"#src"}" \
> > - :: "r" ((void*) base), "r" ((long) stride) \
> > + :: "r" ((void*) (base)), "r" ((long) (stride)) \
> > : "memory")
> >
> > #define _tile_zero(dst) \
> > --
> > 2.18.1
> >
>
>
> --
> BR,
> Hongtao
[-- Attachment #2: 0001-i386-Fix-wrong-result-for-AMX-TILE-intrinsic-when-pa.patch --]
[-- Type: application/octet-stream, Size: 3634 bytes --]
From 78f3d669fa6d67719c9e044e4cc581c77c7326da Mon Sep 17 00:00:00 2001
From: Hongyu Wang <hongyu.wang@intel.com>
Date: Wed, 3 Nov 2021 13:58:52 +0800
Subject: [PATCH] i386: Fix wrong result for AMX-TILE intrinsic when parsing
expression.
_tile_loadd, _tile_stored, _tile_streamloadd intrinsics are defined by
macro, so the parameters should be wrapped by parentheses to accept
expressions.
gcc/ChangeLog:
* config/i386/amxtileintrin.h (_tile_loadd_internal): Add
parentheses to base and stride.
(_tile_stream_loadd_internal): Likewise.
(_tile_stored_internal): Likewise.
gcc/testsuite/ChangeLog:
* gcc.target/i386/amxtile-3.c: New test.
---
gcc/config/i386/amxtileintrin.h | 6 ++---
gcc/testsuite/gcc.target/i386/amxtile-3.c | 28 +++++++++++++++++++++++
2 files changed, 31 insertions(+), 3 deletions(-)
create mode 100644 gcc/testsuite/gcc.target/i386/amxtile-3.c
diff --git a/gcc/config/i386/amxtileintrin.h b/gcc/config/i386/amxtileintrin.h
index 75d784ad160..3a0a6b44c17 100644
--- a/gcc/config/i386/amxtileintrin.h
+++ b/gcc/config/i386/amxtileintrin.h
@@ -62,7 +62,7 @@ _tile_release (void)
#define _tile_loadd_internal(dst,base,stride) \
__asm__ volatile \
("{tileloadd\t(%0,%1,1), %%tmm"#dst"|tileloadd\t%%tmm"#dst", [%0+%1*1]}" \
- :: "r" ((const void*) base), "r" ((long) stride))
+ :: "r" ((const void*) (base)), "r" ((long) (stride)))
#define _tile_stream_loadd(dst,base,stride) \
_tile_stream_loadd_internal (dst, base, stride)
@@ -70,7 +70,7 @@ _tile_release (void)
#define _tile_stream_loadd_internal(dst,base,stride) \
__asm__ volatile \
("{tileloaddt1\t(%0,%1,1), %%tmm"#dst"|tileloaddt1\t%%tmm"#dst", [%0+%1*1]}" \
- :: "r" ((const void*) base), "r" ((long) stride))
+ :: "r" ((const void*) (base)), "r" ((long) (stride)))
#define _tile_stored(dst,base,stride) \
_tile_stored_internal (dst, base, stride)
@@ -78,7 +78,7 @@ _tile_release (void)
#define _tile_stored_internal(src,base,stride) \
__asm__ volatile \
("{tilestored\t%%tmm"#src", (%0,%1,1)|tilestored\t[%0+%1*1], %%tmm"#src"}" \
- :: "r" ((void*) base), "r" ((long) stride) \
+ :: "r" ((void*) (base)), "r" ((long) (stride)) \
: "memory")
#define _tile_zero(dst) \
diff --git a/gcc/testsuite/gcc.target/i386/amxtile-3.c b/gcc/testsuite/gcc.target/i386/amxtile-3.c
new file mode 100644
index 00000000000..31b34d0ed15
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/amxtile-3.c
@@ -0,0 +1,28 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mamx-tile " } */
+/* { dg-final { scan-assembler "tileloadd\[ \\t]+\[^\n\]*\\(%\[a-z0-9]*\,%\[a-z0-9\]*\,\[124\]\\)+\[^\n\]*%tmm\[0-9\]" } } */
+/* { dg-final { scan-assembler "tileloaddt1\[ \\t]+\[^\n\]*\\(%\[a-z0-9]*\,%\[a-z0-9\]*\,\[124\]\\)+\[^\n\]*%tmm\[0-9\]" } } */
+/* { dg-final { scan-assembler "tilestored\[ \\t]+\[^\n\]*%tmm\[0-9\]+\[^\n\]*\\(%\[a-z0-9]*\,%\[a-z0-9\]*\,\[124\]\\)" } } */
+/* { dg-final { scan-assembler "leaq\[ \\t]+4" } } */
+/* { dg-final { scan-assembler "leaq\[ \\t]+8" } } */
+/* { dg-final { scan-assembler "addq\[ \\t]+\\\$12" } } */
+/* { dg-final { scan-assembler-not "leaq\[ \\t]+1" } } */
+/* { dg-final { scan-assembler-not "leaq\[ \\t]+2" } } */
+/* { dg-final { scan-assembler-not "addq\[ \\t]+\\\$3" } } */
+#include <immintrin.h>
+
+extern int a[];
+extern const float* base;
+extern const int stride;
+
+#define TMM0 0
+#define TMM1 1
+#define TMM2 2
+#define TMM3 3
+
+void TEST ()
+{
+ _tile_loadd (TMM3, base + 1, stride);
+ _tile_stream_loadd (TMM2, base + 2, stride);
+ _tile_stored (TMM2, base + 3, stride);
+}
--
2.18.1
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] i386: Fix wrong result for AMX-TILE intrinsic when parsing expression.
2021-11-04 3:25 ` Hongyu Wang
@ 2021-11-04 4:46 ` Hongtao Liu
0 siblings, 0 replies; 4+ messages in thread
From: Hongtao Liu @ 2021-11-04 4:46 UTC (permalink / raw)
To: Hongyu Wang; +Cc: Hongyu Wang, Liu, Hongtao, GCC Patches
On Thu, Nov 4, 2021 at 11:25 AM Hongyu Wang <wwwhhhyyy333@gmail.com> wrote:
>
> > Could you add a testcase for that?
>
> Yes, updated patch.
>
> Hongtao Liu via Gcc-patches <gcc-patches@gcc.gnu.org> 于2021年11月4日周四 上午10:25写道:
> >
> > On Thu, Nov 4, 2021 at 9:19 AM Hongyu Wang via Gcc-patches
> > <gcc-patches@gcc.gnu.org> wrote:
> > >
> > > Hi,
> > >
> > > _tile_loadd, _tile_stored, _tile_streamloadd intrinsics are defined by
> > > macro, so the parameters should be wrapped by parentheses to accept
> > > expressions.
> > >
> > > Bootstraped/regtested on x86_64-pc-linux-gnu{-m32,} and sde.
> > >
> > > OK for master and backport to GCC11 branch?
LGTM.
> > Could you add a testcase for that?
> > >
> > > gcc/ChangeLog:
> > >
> > > * config/i386/amxtileintrin.h (_tile_loadd_internal): Add
> > > parentheses to base and stride.
> > > (_tile_stream_loadd_internal): Likewise.
> > > (_tile_stored_internal): Likewise.
> > > ---
> > > gcc/config/i386/amxtileintrin.h | 6 +++---
> > > 1 file changed, 3 insertions(+), 3 deletions(-)
> > >
> > > diff --git a/gcc/config/i386/amxtileintrin.h b/gcc/config/i386/amxtileintrin.h
> > > index 75d784ad160..3a0a6b44c17 100644
> > > --- a/gcc/config/i386/amxtileintrin.h
> > > +++ b/gcc/config/i386/amxtileintrin.h
> > > @@ -62,7 +62,7 @@ _tile_release (void)
> > > #define _tile_loadd_internal(dst,base,stride) \
> > > __asm__ volatile \
> > > ("{tileloadd\t(%0,%1,1), %%tmm"#dst"|tileloadd\t%%tmm"#dst", [%0+%1*1]}" \
> > > - :: "r" ((const void*) base), "r" ((long) stride))
> > > + :: "r" ((const void*) (base)), "r" ((long) (stride)))
> > >
> > > #define _tile_stream_loadd(dst,base,stride) \
> > > _tile_stream_loadd_internal (dst, base, stride)
> > > @@ -70,7 +70,7 @@ _tile_release (void)
> > > #define _tile_stream_loadd_internal(dst,base,stride) \
> > > __asm__ volatile \
> > > ("{tileloaddt1\t(%0,%1,1), %%tmm"#dst"|tileloaddt1\t%%tmm"#dst", [%0+%1*1]}" \
> > > - :: "r" ((const void*) base), "r" ((long) stride))
> > > + :: "r" ((const void*) (base)), "r" ((long) (stride)))
> > >
> > > #define _tile_stored(dst,base,stride) \
> > > _tile_stored_internal (dst, base, stride)
> > > @@ -78,7 +78,7 @@ _tile_release (void)
> > > #define _tile_stored_internal(src,base,stride) \
> > > __asm__ volatile \
> > > ("{tilestored\t%%tmm"#src", (%0,%1,1)|tilestored\t[%0+%1*1], %%tmm"#src"}" \
> > > - :: "r" ((void*) base), "r" ((long) stride) \
> > > + :: "r" ((void*) (base)), "r" ((long) (stride)) \
> > > : "memory")
> > >
> > > #define _tile_zero(dst) \
> > > --
> > > 2.18.1
> > >
> >
> >
> > --
> > BR,
> > Hongtao
--
BR,
Hongtao
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2021-11-04 4:39 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-11-04 1:18 [PATCH] i386: Fix wrong result for AMX-TILE intrinsic when parsing expression Hongyu Wang
2021-11-04 2:24 ` Hongtao Liu
2021-11-04 3:25 ` Hongyu Wang
2021-11-04 4:46 ` Hongtao Liu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).