public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Richard Sandiford <richard.sandiford@arm.com>
To: Andrew Carlotti via Gcc-patches <gcc-patches@gcc.gnu.org>
Cc: Andrew Carlotti <andrew.carlotti@arm.com>
Subject: Re: [PATCH] aarch64: Remove architecture dependencies from intrinsics
Date: Tue, 27 Jun 2023 07:23:32 +0100	[thread overview]
Message-ID: <mpt4jmta0ez.fsf@arm.com> (raw)
In-Reply-To: <98b3efca-4c1c-7797-022c-0be09087d086@e124511.cambridge.arm.com> (Andrew Carlotti via Gcc-patches's message of "Mon, 26 Jun 2023 14:55:48 +0100")

Andrew Carlotti via Gcc-patches <gcc-patches@gcc.gnu.org> writes:
> Many intrinsics currently depend on both an architecture version and a
> feature, despite the corresponding instructions being available within
> GCC at lower architecture versions.
>
> LLVM has already removed these explicit architecture version
> dependences; this patch does the same for GCC, as well as removing an
> unecessary simd dependency for the scalar fp16 intrinsics.
>
> Binutils does not support all of these architecture+feature combinations
> yet, but this is an existing problem that is already reachable from GCC.
> For example, compiling the test gcc.target/aarch64/usadv16qi-dotprod.c
> with -O3 -march=armv8-a+dotprod has resulted in an assembler error since
> GCC 10. I intend to patch this in binutils.
>
> This patch retains explicit architecture version dependencies for
> features that do not currently have a separate feature flag.
>
> Ok for master, and backport to GCC 13?
>
> gcc/ChangeLog:
>
>  * config/aarch64/aarch64.h (TARGET_MEMTAG): Remove armv8.5
>  dependency.
>  * config/aarch64/arm_acle.h: Remove unnecessary armv8.x
>  dependencies from target pragmas.
>  * config/aarch64/arm_fp16.h (target): Likewise.

The change to this file is a bit different from the others,
since it's removing an implicit dependency on +simd, rather
than a dependency on an architecture level.  I think it'd be
worth mentioning that explicitly in the changelog.

OK with that change, thanks.

(Arguably we should add +nosimd to many of the other pragmas in
arm_acle.h, but that's logically a separate patch.)

Richard

>  * config/aarch64/arm_neon.h (target): Likewise.
>
> gcc/testsuite/ChangeLog:
>
>  * gcc.target/aarch64/feature-bf16-backport.c: New test.
>  * gcc.target/aarch64/feature-dotprod-backport.c: New test.
>  * gcc.target/aarch64/feature-fp16-backport.c: New test.
>  * gcc.target/aarch64/feature-fp16-scalar-backport.c: New test.
>  * gcc.target/aarch64/feature-fp16fml-backport.c: New test.
>  * gcc.target/aarch64/feature-i8mm-backport.c: New test.
>  * gcc.target/aarch64/feature-memtag-backport.c: New test.
>  * gcc.target/aarch64/feature-sha3-backport.c: New test.
>  * gcc.target/aarch64/feature-sm4-backport.c: New test.
>
>
> diff --git a/gcc/config/aarch64/aarch64.h b/gcc/config/aarch64/aarch64.h
> index 7129ed1ff370d597895b3f46b56b1250da7fa190..cdb664eb8f7db820b6b06b2667bfad6dc14cb7a2 100644
> --- a/gcc/config/aarch64/aarch64.h
> +++ b/gcc/config/aarch64/aarch64.h
> @@ -292,7 +292,7 @@ enum class aarch64_feature : unsigned char {
>  #define TARGET_RNG (AARCH64_ISA_RNG)
>  
>  /* Memory Tagging instructions optional to Armv8.5 enabled through +memtag.  */
> -#define TARGET_MEMTAG (AARCH64_ISA_V8_5A && AARCH64_ISA_MEMTAG)
> +#define TARGET_MEMTAG (AARCH64_ISA_MEMTAG)
>  
>  /* I8MM instructions are enabled through +i8mm.  */
>  #define TARGET_I8MM (AARCH64_ISA_I8MM)
> diff --git a/gcc/config/aarch64/arm_acle.h b/gcc/config/aarch64/arm_acle.h
> index e0ac591d2c8d6c4c4c8a074b2d9881c47b1db1ab..87fb42f47c5821adecbb0ea441e0a38c63972e77 100644
> --- a/gcc/config/aarch64/arm_acle.h
> +++ b/gcc/config/aarch64/arm_acle.h
> @@ -325,7 +325,7 @@ __rndrrs (uint64_t *__res)
>  #pragma GCC pop_options
>  
>  #pragma GCC push_options
> -#pragma GCC target ("arch=armv8.5-a+memtag")
> +#pragma GCC target ("+nothing+memtag")
>  
>  #define __arm_mte_create_random_tag(__ptr, __u64_mask) \
>    __builtin_aarch64_memtag_irg(__ptr, __u64_mask)
> diff --git a/gcc/config/aarch64/arm_fp16.h b/gcc/config/aarch64/arm_fp16.h
> index a8fa4dbbdfe1bab4aa604bb311ef66d4e1de18ac..84b2ed66f9ba19fba6ccd8be33940d7239bfa22e 100644
> --- a/gcc/config/aarch64/arm_fp16.h
> +++ b/gcc/config/aarch64/arm_fp16.h
> @@ -30,7 +30,7 @@
>  #include <stdint.h>
>  
>  #pragma GCC push_options
> -#pragma GCC target ("arch=armv8.2-a+fp16")
> +#pragma GCC target ("+nothing+fp16+nosimd")
>  
>  typedef __fp16 float16_t;
>  
> diff --git a/gcc/config/aarch64/arm_neon.h b/gcc/config/aarch64/arm_neon.h
> index eeec9f162e223df8cf7803b3227aef22e94227ac..a078674376af121c36bbebef76631c25a6815b1b 100644
> --- a/gcc/config/aarch64/arm_neon.h
> +++ b/gcc/config/aarch64/arm_neon.h
> @@ -25590,7 +25590,7 @@ __INTERLEAVE_LIST (zip)
>  #include "arm_fp16.h"
>  
>  #pragma GCC push_options
> -#pragma GCC target ("arch=armv8.2-a+fp16")
> +#pragma GCC target ("+nothing+fp16")
>  
>  /* ARMv8.2-A FP16 one operand vector intrinsics.  */
>  
> @@ -26753,7 +26753,7 @@ vminnmvq_f16 (float16x8_t __a)
>  /* AdvSIMD Dot Product intrinsics.  */
>  
>  #pragma GCC push_options
> -#pragma GCC target ("arch=armv8.2-a+dotprod")
> +#pragma GCC target ("+nothing+dotprod")
>  
>  __extension__ extern __inline uint32x2_t
>  __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> @@ -26844,7 +26844,7 @@ vdotq_laneq_s32 (int32x4_t __r, int8x16_t __a, int8x16_t __b, const int __index)
>  #pragma GCC pop_options
>  
>  #pragma GCC push_options
> -#pragma GCC target ("arch=armv8.2-a+sm4")
> +#pragma GCC target ("+nothing+sm4")
>  
>  __extension__ extern __inline uint32x4_t
>  __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> @@ -26911,7 +26911,7 @@ vsm4ekeyq_u32 (uint32x4_t __a, uint32x4_t __b)
>  #pragma GCC pop_options
>  
>  #pragma GCC push_options
> -#pragma GCC target ("arch=armv8.2-a+sha3")
> +#pragma GCC target ("+nothing+sha3")
>  
>  __extension__ extern __inline uint64x2_t
>  __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> @@ -27547,7 +27547,7 @@ vcmlaq_rot270_laneq_f32 (float32x4_t __r, float32x4_t __a, float32x4_t __b,
>  #pragma GCC pop_options
>  
>  #pragma GCC push_options
> -#pragma GCC target ("arch=armv8.2-a+fp16fml")
> +#pragma GCC target ("+nothing+fp16fml")
>  
>  __extension__ extern __inline float32x2_t
>  __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> @@ -27856,7 +27856,7 @@ vrnd64xq_f64 (float64x2_t __a)
>  #include "arm_bf16.h"
>  
>  #pragma GCC push_options
> -#pragma GCC target ("arch=armv8.2-a+bf16")
> +#pragma GCC target ("+nothing+bf16")
>  
>  __extension__ extern __inline bfloat16x4_t
>  __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> @@ -28535,7 +28535,7 @@ vst4q_lane_bf16 (bfloat16_t *__ptr, bfloat16x8x4_t __val, const int __lane)
>  /* AdvSIMD 8-bit Integer Matrix Multiply (I8MM) intrinsics.  */
>  
>  #pragma GCC push_options
> -#pragma GCC target ("arch=armv8.2-a+i8mm")
> +#pragma GCC target ("+nothing+i8mm")
>  
>  __extension__ extern __inline int32x2_t
>  __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
> diff --git a/gcc/testsuite/gcc.target/aarch64/feature-bf16-backport.c b/gcc/testsuite/gcc.target/aarch64/feature-bf16-backport.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..3a03255bbb9a3e3da45c9f2ecafaf12685278057
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/aarch64/feature-bf16-backport.c
> @@ -0,0 +1,10 @@
> +/* { dg-do compile } */
> +/* { dg-additional-options "-march=armv8-a+bf16" } */
> +
> +#include <arm_neon.h>
> +
> +float32x4_t bar (float32x4_t r, bfloat16x8_t a, bfloat16x8_t b) {
> +        return vbfmlalbq_f32 (r, a, b);
> +}
> +
> +/* { dg-final { scan-assembler {\tbfmlalb\t} } } */
> diff --git a/gcc/testsuite/gcc.target/aarch64/feature-dotprod-backport.c b/gcc/testsuite/gcc.target/aarch64/feature-dotprod-backport.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..5f66fffa5f5b09a9c04da7b861055e3aa89bccec
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/aarch64/feature-dotprod-backport.c
> @@ -0,0 +1,10 @@
> +/* { dg-do compile } */
> +/* { dg-additional-options "-march=armv8-a+dotprod" } */
> +
> +#include <arm_neon.h>
> +
> +uint32x4_t bar (uint32x4_t r, uint8x16_t a, uint8x16_t b) {
> +        return vdotq_u32(r, a, b);
> +}
> +
> +/* { dg-final { scan-assembler {\tudot\t} } } */
> diff --git a/gcc/testsuite/gcc.target/aarch64/feature-fp16-backport.c b/gcc/testsuite/gcc.target/aarch64/feature-fp16-backport.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..eb94ae060664eef4275023440c0a18b52ae27b42
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/aarch64/feature-fp16-backport.c
> @@ -0,0 +1,10 @@
> +/* { dg-do compile } */
> +/* { dg-additional-options "-march=armv8-a+fp16" } */
> +
> +#include <arm_neon.h>
> +
> +float16x8_t bar (float16x8_t a, float16x8_t b) {
> +        return vaddq_f16(a, b);
> +}
> +
> +/* { dg-final { scan-assembler {\tfadd\t} } } */
> diff --git a/gcc/testsuite/gcc.target/aarch64/feature-fp16-scalar-backport.c b/gcc/testsuite/gcc.target/aarch64/feature-fp16-scalar-backport.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..9cde6b5558c2e3699620d0d35ebc3e679fdfe5eb
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/aarch64/feature-fp16-scalar-backport.c
> @@ -0,0 +1,10 @@
> +/* { dg-do compile } */
> +/* { dg-additional-options "-march=armv8-a+fp16+nosimd" } */
> +
> +#include <arm_fp16.h>
> +
> +float16_t bar (float16_t a, float16_t b) {
> +        return vaddh_f16(a, b);
> +}
> +
> +/* { dg-final { scan-assembler {\tfadd\t} } } */
> diff --git a/gcc/testsuite/gcc.target/aarch64/feature-fp16fml-backport.c b/gcc/testsuite/gcc.target/aarch64/feature-fp16fml-backport.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..86c4748b86d9730d433cc206700d7789d996bb6d
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/aarch64/feature-fp16fml-backport.c
> @@ -0,0 +1,10 @@
> +/* { dg-do compile } */
> +/* { dg-additional-options "-march=armv8-a+fp16fml" } */
> +
> +#include <arm_neon.h>
> +
> +float32x4_t bar (float32x4_t r, float16x8_t a, float16x8_t b) {
> +        return vfmlalq_high_f16 (r, a, b);
> +}
> +
> +/* { dg-final { scan-assembler {\tfmlal2\t} } } */
> diff --git a/gcc/testsuite/gcc.target/aarch64/feature-i8mm-backport.c b/gcc/testsuite/gcc.target/aarch64/feature-i8mm-backport.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..6dd0214faab9d470a4f363252d48ac0de92bfe7b
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/aarch64/feature-i8mm-backport.c
> @@ -0,0 +1,10 @@
> +/* { dg-do compile } */
> +/* { dg-additional-options "-march=armv8-a+i8mm" } */
> +
> +#include <arm_neon.h>
> +
> +int32x4_t bar (int32x4_t r, int8x16_t a, int8x16_t b) {
> +        return vmmlaq_s32 (r, a, b);
> +}
> +
> +/* { dg-final { scan-assembler {\tsmmla\t} } } */
> diff --git a/gcc/testsuite/gcc.target/aarch64/feature-memtag-backport.c b/gcc/testsuite/gcc.target/aarch64/feature-memtag-backport.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..5cb071b77ccbd11039d4b8ee9e2cd78a708f2fad
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/aarch64/feature-memtag-backport.c
> @@ -0,0 +1,10 @@
> +/* { dg-do compile } */
> +/* { dg-additional-options "-march=armv8-a+memtag" } */
> +
> +#include <arm_acle.h>
> +
> +int *bar (int *src) {
> +        return __arm_mte_create_random_tag(src, 2<<16-1);
> +}
> +
> +/* { dg-final { scan-assembler {\tirg\t} } } */
> diff --git a/gcc/testsuite/gcc.target/aarch64/feature-sha3-backport.c b/gcc/testsuite/gcc.target/aarch64/feature-sha3-backport.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..e194f1a10ea12439195c8cf1aae08dae26607c14
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/aarch64/feature-sha3-backport.c
> @@ -0,0 +1,10 @@
> +/* { dg-do compile } */
> +/* { dg-additional-options "-march=armv8-a+sha3" } */
> +
> +#include <arm_neon.h>
> +
> +uint64x2_t bar (uint64x2_t a, uint64x2_t b, uint64x2_t c) {
> +        return vsha512hq_u64(a, b, c);
> +}
> +
> +/* { dg-final { scan-assembler {\tsha512h\t} } } */
> diff --git a/gcc/testsuite/gcc.target/aarch64/feature-sm4-backport.c b/gcc/testsuite/gcc.target/aarch64/feature-sm4-backport.c
> new file mode 100644
> index 0000000000000000000000000000000000000000..604a58bcb68e5753d2eec661a374b4983ad29088
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/aarch64/feature-sm4-backport.c
> @@ -0,0 +1,10 @@
> +/* { dg-do compile } */
> +/* { dg-additional-options "-march=armv8-a+sm4" } */
> +
> +#include <arm_neon.h>
> +
> +uint32x4_t bar (uint32x4_t a, uint32x4_t b, uint32x4_t c) {
> +        return vsm3tt1aq_u32(a, b, c, 2);
> +}
> +
> +/* { dg-final { scan-assembler {\tsm3tt1a\t} } } */

  reply	other threads:[~2023-06-27  6:23 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-06-26 13:55 Andrew Carlotti
2023-06-27  6:23 ` Richard Sandiford [this message]
2023-06-29 18:24   ` Andrew Carlotti
2023-07-19 16:44     ` [GCC 13 PATCH] " Andrew Carlotti
2023-07-19 18:35       ` Ramana Radhakrishnan
2023-07-19 23:27         ` Andrew Carlotti
2023-07-20  6:48       ` Richard Sandiford
2023-07-20  7:37         ` Richard Biener
2023-07-20  8:22           ` Andrew Carlotti

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=mpt4jmta0ez.fsf@arm.com \
    --to=richard.sandiford@arm.com \
    --cc=andrew.carlotti@arm.com \
    --cc=gcc-patches@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).