public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Kyrylo Tkachov <Kyrylo.Tkachov@arm.com>
To: Andrea Corallo <Andrea.Corallo@arm.com>,
	"gcc-patches@gcc.gnu.org" <gcc-patches@gcc.gnu.org>
Cc: Richard Earnshaw <Richard.Earnshaw@arm.com>,
	Andrea Corallo <Andrea.Corallo@arm.com>
Subject: RE: [PATCH 26/35] arm: improve tests for vmlasq*
Date: Tue, 22 Nov 2022 16:56:36 +0000	[thread overview]
Message-ID: <PAXPR08MB69263B4E0DDCC8A796E4EB7E930D9@PAXPR08MB6926.eurprd08.prod.outlook.com> (raw)
In-Reply-To: <20221117163809.1009526-27-andrea.corallo@arm.com>



> -----Original Message-----
> From: Andrea Corallo <andrea.corallo@arm.com>
> Sent: Thursday, November 17, 2022 4:38 PM
> To: gcc-patches@gcc.gnu.org
> Cc: Kyrylo Tkachov <Kyrylo.Tkachov@arm.com>; Richard Earnshaw
> <Richard.Earnshaw@arm.com>; Andrea Corallo <Andrea.Corallo@arm.com>
> Subject: [PATCH 26/35] arm: improve tests for vmlasq*
> 
> gcc/testsuite/ChangeLog:
> 
> 	* gcc.target/arm/mve/intrinsics/vmlasq_m_n_s16.c: Improve test.
> 	* gcc.target/arm/mve/intrinsics/vmlasq_m_n_s32.c: Likewise.
> 	* gcc.target/arm/mve/intrinsics/vmlasq_m_n_s8.c: Likewise.
> 	* gcc.target/arm/mve/intrinsics/vmlasq_m_n_u16.c: Likewise.
> 	* gcc.target/arm/mve/intrinsics/vmlasq_m_n_u32.c: Likewise.
> 	* gcc.target/arm/mve/intrinsics/vmlasq_m_n_u8.c: Likewise.
> 	* gcc.target/arm/mve/intrinsics/vmlasq_n_s16.c: Likewise.
> 	* gcc.target/arm/mve/intrinsics/vmlasq_n_s32.c: Likewise.
> 	* gcc.target/arm/mve/intrinsics/vmlasq_n_s8.c: Likewise.
> 	* gcc.target/arm/mve/intrinsics/vmlasq_n_u16.c: Likewise.
> 	* gcc.target/arm/mve/intrinsics/vmlasq_n_u32.c: Likewise.
> 	* gcc.target/arm/mve/intrinsics/vmlasq_n_u8.c: Likewise.

Ok.
Thanks,
Kyrill

> ---
>  .../arm/mve/intrinsics/vmlasq_m_n_s16.c       | 34 ++++++++++---
>  .../arm/mve/intrinsics/vmlasq_m_n_s32.c       | 34 ++++++++++---
>  .../arm/mve/intrinsics/vmlasq_m_n_s8.c        | 34 ++++++++++---
>  .../arm/mve/intrinsics/vmlasq_m_n_u16.c       | 50 ++++++++++++++++---
>  .../arm/mve/intrinsics/vmlasq_m_n_u32.c       | 50 ++++++++++++++++---
>  .../arm/mve/intrinsics/vmlasq_m_n_u8.c        | 50 ++++++++++++++++---
>  .../arm/mve/intrinsics/vmlasq_n_s16.c         | 24 ++++++---
>  .../arm/mve/intrinsics/vmlasq_n_s32.c         | 24 ++++++---
>  .../arm/mve/intrinsics/vmlasq_n_s8.c          | 24 ++++++---
>  .../arm/mve/intrinsics/vmlasq_n_u16.c         | 36 ++++++++++---
>  .../arm/mve/intrinsics/vmlasq_n_u32.c         | 36 ++++++++++---
>  .../arm/mve/intrinsics/vmlasq_n_u8.c          | 36 ++++++++++---
>  12 files changed, 348 insertions(+), 84 deletions(-)
> 
> diff --git a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_s16.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_s16.c
> index bf66e616ec7..af6e588adad 100644
> --- a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_s16.c
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_s16.c
> @@ -1,23 +1,41 @@
>  /* { dg-require-effective-target arm_v8_1m_mve_ok } */
>  /* { dg-add-options arm_v8_1m_mve } */
>  /* { dg-additional-options "-O2" } */
> +/* { dg-final { check-function-bodies "**" "" } } */
> 
>  #include "arm_mve.h"
> 
> +/*
> +**foo:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.s16	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  int16x8_t
> -foo (int16x8_t a, int16x8_t b, int16_t c, mve_pred16_t p)
> +foo (int16x8_t m1, int16x8_t m2, int16_t add, mve_pred16_t p)
>  {
> -  return vmlasq_m_n_s16 (a, b, c, p);
> +  return vmlasq_m_n_s16 (m1, m2, add, p);
>  }
> 
> -/* { dg-final { scan-assembler "vpst" } } */
> -/* { dg-final { scan-assembler "vmlast.s16"  }  } */
> 
> +/*
> +**foo1:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.s16	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  int16x8_t
> -foo1 (int16x8_t a, int16x8_t b, int16_t c, mve_pred16_t p)
> +foo1 (int16x8_t m1, int16x8_t m2, int16_t add, mve_pred16_t p)
>  {
> -  return vmlasq_m (a, b, c, p);
> +  return vmlasq_m (m1, m2, add, p);
>  }
> 
> -/* { dg-final { scan-assembler "vpst" } } */
> -/* { dg-final { scan-assembler "vmlast.s16"  }  } */
> +/* { dg-final { scan-assembler-not "__ARM_undef" } } */
> \ No newline at end of file
> diff --git a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_s32.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_s32.c
> index 53c21e2e5b6..9d0cc3076d9 100644
> --- a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_s32.c
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_s32.c
> @@ -1,23 +1,41 @@
>  /* { dg-require-effective-target arm_v8_1m_mve_ok } */
>  /* { dg-add-options arm_v8_1m_mve } */
>  /* { dg-additional-options "-O2" } */
> +/* { dg-final { check-function-bodies "**" "" } } */
> 
>  #include "arm_mve.h"
> 
> +/*
> +**foo:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.s32	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  int32x4_t
> -foo (int32x4_t a, int32x4_t b, int32_t c, mve_pred16_t p)
> +foo (int32x4_t m1, int32x4_t m2, int32_t add, mve_pred16_t p)
>  {
> -  return vmlasq_m_n_s32 (a, b, c, p);
> +  return vmlasq_m_n_s32 (m1, m2, add, p);
>  }
> 
> -/* { dg-final { scan-assembler "vpst" } } */
> -/* { dg-final { scan-assembler "vmlast.s32"  }  } */
> 
> +/*
> +**foo1:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.s32	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  int32x4_t
> -foo1 (int32x4_t a, int32x4_t b, int32_t c, mve_pred16_t p)
> +foo1 (int32x4_t m1, int32x4_t m2, int32_t add, mve_pred16_t p)
>  {
> -  return vmlasq_m (a, b, c, p);
> +  return vmlasq_m (m1, m2, add, p);
>  }
> 
> -/* { dg-final { scan-assembler "vpst" } } */
> -/* { dg-final { scan-assembler "vmlast.s32"  }  } */
> +/* { dg-final { scan-assembler-not "__ARM_undef" } } */
> \ No newline at end of file
> diff --git a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_s8.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_s8.c
> index ac08b15fdbe..772ad8b1e76 100644
> --- a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_s8.c
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_s8.c
> @@ -1,23 +1,41 @@
>  /* { dg-require-effective-target arm_v8_1m_mve_ok } */
>  /* { dg-add-options arm_v8_1m_mve } */
>  /* { dg-additional-options "-O2" } */
> +/* { dg-final { check-function-bodies "**" "" } } */
> 
>  #include "arm_mve.h"
> 
> +/*
> +**foo:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.s8	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  int8x16_t
> -foo (int8x16_t a, int8x16_t b, int8_t c, mve_pred16_t p)
> +foo (int8x16_t m1, int8x16_t m2, int8_t add, mve_pred16_t p)
>  {
> -  return vmlasq_m_n_s8 (a, b, c, p);
> +  return vmlasq_m_n_s8 (m1, m2, add, p);
>  }
> 
> -/* { dg-final { scan-assembler "vpst" } } */
> -/* { dg-final { scan-assembler "vmlast.s8"  }  } */
> 
> +/*
> +**foo1:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.s8	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  int8x16_t
> -foo1 (int8x16_t a, int8x16_t b, int8_t c, mve_pred16_t p)
> +foo1 (int8x16_t m1, int8x16_t m2, int8_t add, mve_pred16_t p)
>  {
> -  return vmlasq_m (a, b, c, p);
> +  return vmlasq_m (m1, m2, add, p);
>  }
> 
> -/* { dg-final { scan-assembler "vpst" } } */
> -/* { dg-final { scan-assembler "vmlast.s8"  }  } */
> +/* { dg-final { scan-assembler-not "__ARM_undef" } } */
> \ No newline at end of file
> diff --git a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_u16.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_u16.c
> index 99f1e28c7d5..b02dc64a31b 100644
> --- a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_u16.c
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_u16.c
> @@ -1,23 +1,57 @@
>  /* { dg-require-effective-target arm_v8_1m_mve_ok } */
>  /* { dg-add-options arm_v8_1m_mve } */
>  /* { dg-additional-options "-O2" } */
> +/* { dg-final { check-function-bodies "**" "" } } */
> 
>  #include "arm_mve.h"
> 
> +/*
> +**foo:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.u16	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  uint16x8_t
> -foo (uint16x8_t a, uint16x8_t b, uint16_t c, mve_pred16_t p)
> +foo (uint16x8_t m1, uint16x8_t m2, uint16_t add, mve_pred16_t p)
>  {
> -  return vmlasq_m_n_u16 (a, b, c, p);
> +  return vmlasq_m_n_u16 (m1, m2, add, p);
>  }
> 
> -/* { dg-final { scan-assembler "vpst" } } */
> -/* { dg-final { scan-assembler "vmlast.u16"  }  } */
> 
> +/*
> +**foo1:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.u16	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  uint16x8_t
> -foo1 (uint16x8_t a, uint16x8_t b, uint16_t c, mve_pred16_t p)
> +foo1 (uint16x8_t m1, uint16x8_t m2, uint16_t add, mve_pred16_t p)
>  {
> -  return vmlasq_m (a, b, c, p);
> +  return vmlasq_m (m1, m2, add, p);
>  }
> 
> -/* { dg-final { scan-assembler "vpst" } } */
> -/* { dg-final { scan-assembler "vmlast.u16"  }  } */
> +/*
> +**foo2:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.u16	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
> +uint16x8_t
> +foo2 (uint16x8_t m1, uint16x8_t m2, mve_pred16_t p)
> +{
> +  return vmlasq_m (m1, m2, 1, p);
> +}
> +
> +/* { dg-final { scan-assembler-not "__ARM_undef" } } */
> \ No newline at end of file
> diff --git a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_u32.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_u32.c
> index 8d8edca6024..0214cf2136e 100644
> --- a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_u32.c
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_u32.c
> @@ -1,23 +1,57 @@
>  /* { dg-require-effective-target arm_v8_1m_mve_ok } */
>  /* { dg-add-options arm_v8_1m_mve } */
>  /* { dg-additional-options "-O2" } */
> +/* { dg-final { check-function-bodies "**" "" } } */
> 
>  #include "arm_mve.h"
> 
> +/*
> +**foo:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.u32	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  uint32x4_t
> -foo (uint32x4_t a, uint32x4_t b, uint32_t c, mve_pred16_t p)
> +foo (uint32x4_t m1, uint32x4_t m2, uint32_t add, mve_pred16_t p)
>  {
> -  return vmlasq_m_n_u32 (a, b, c, p);
> +  return vmlasq_m_n_u32 (m1, m2, add, p);
>  }
> 
> -/* { dg-final { scan-assembler "vpst" } } */
> -/* { dg-final { scan-assembler "vmlast.u32"  }  } */
> 
> +/*
> +**foo1:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.u32	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  uint32x4_t
> -foo1 (uint32x4_t a, uint32x4_t b, uint32_t c, mve_pred16_t p)
> +foo1 (uint32x4_t m1, uint32x4_t m2, uint32_t add, mve_pred16_t p)
>  {
> -  return vmlasq_m (a, b, c, p);
> +  return vmlasq_m (m1, m2, add, p);
>  }
> 
> -/* { dg-final { scan-assembler "vpst" } } */
> -/* { dg-final { scan-assembler "vmlast.u32"  }  } */
> +/*
> +**foo2:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.u32	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
> +uint32x4_t
> +foo2 (uint32x4_t m1, uint32x4_t m2, mve_pred16_t p)
> +{
> +  return vmlasq_m (m1, m2, 1, p);
> +}
> +
> +/* { dg-final { scan-assembler-not "__ARM_undef" } } */
> \ No newline at end of file
> diff --git a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_u8.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_u8.c
> index e7f685bbcaa..c9824e332f7 100644
> --- a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_u8.c
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_m_n_u8.c
> @@ -1,23 +1,57 @@
>  /* { dg-require-effective-target arm_v8_1m_mve_ok } */
>  /* { dg-add-options arm_v8_1m_mve } */
>  /* { dg-additional-options "-O2" } */
> +/* { dg-final { check-function-bodies "**" "" } } */
> 
>  #include "arm_mve.h"
> 
> +/*
> +**foo:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.u8	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  uint8x16_t
> -foo (uint8x16_t a, uint8x16_t b, uint8_t c, mve_pred16_t p)
> +foo (uint8x16_t m1, uint8x16_t m2, uint8_t add, mve_pred16_t p)
>  {
> -  return vmlasq_m_n_u8 (a, b, c, p);
> +  return vmlasq_m_n_u8 (m1, m2, add, p);
>  }
> 
> -/* { dg-final { scan-assembler "vpst" } } */
> -/* { dg-final { scan-assembler "vmlast.u8"  }  } */
> 
> +/*
> +**foo1:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.u8	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  uint8x16_t
> -foo1 (uint8x16_t a, uint8x16_t b, uint8_t c, mve_pred16_t p)
> +foo1 (uint8x16_t m1, uint8x16_t m2, uint8_t add, mve_pred16_t p)
>  {
> -  return vmlasq_m (a, b, c, p);
> +  return vmlasq_m (m1, m2, add, p);
>  }
> 
> -/* { dg-final { scan-assembler "vpst" } } */
> -/* { dg-final { scan-assembler "vmlast.u8"  }  } */
> +/*
> +**foo2:
> +**	...
> +**	vmsr	p0, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +**	vpst(?:	@.*|)
> +**	...
> +**	vmlast.u8	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
> +uint8x16_t
> +foo2 (uint8x16_t m1, uint8x16_t m2, mve_pred16_t p)
> +{
> +  return vmlasq_m (m1, m2, 1, p);
> +}
> +
> +/* { dg-final { scan-assembler-not "__ARM_undef" } } */
> \ No newline at end of file
> diff --git a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_s16.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_s16.c
> index 8bfe3c31096..6708a741790 100644
> --- a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_s16.c
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_s16.c
> @@ -1,21 +1,33 @@
>  /* { dg-require-effective-target arm_v8_1m_mve_ok } */
>  /* { dg-add-options arm_v8_1m_mve } */
>  /* { dg-additional-options "-O2" } */
> +/* { dg-final { check-function-bodies "**" "" } } */
> 
>  #include "arm_mve.h"
> 
> +/*
> +**foo:
> +**	...
> +**	vmlas.s16	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  int16x8_t
> -foo (int16x8_t a, int16x8_t b, int16_t c)
> +foo (int16x8_t m1, int16x8_t m2, int16_t add)
>  {
> -  return vmlasq_n_s16 (a, b, c);
> +  return vmlasq_n_s16 (m1, m2, add);
>  }
> 
> -/* { dg-final { scan-assembler "vmlas.s16"  }  } */
> 
> +/*
> +**foo1:
> +**	...
> +**	vmlas.s16	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  int16x8_t
> -foo1 (int16x8_t a, int16x8_t b, int16_t c)
> +foo1 (int16x8_t m1, int16x8_t m2, int16_t add)
>  {
> -  return vmlasq (a, b, c);
> +  return vmlasq (m1, m2, add);
>  }
> 
> -/* { dg-final { scan-assembler "vmlas.s16"  }  } */
> +/* { dg-final { scan-assembler-not "__ARM_undef" } } */
> \ No newline at end of file
> diff --git a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_s32.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_s32.c
> index db06182abec..4e8bf32e016 100644
> --- a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_s32.c
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_s32.c
> @@ -1,21 +1,33 @@
>  /* { dg-require-effective-target arm_v8_1m_mve_ok } */
>  /* { dg-add-options arm_v8_1m_mve } */
>  /* { dg-additional-options "-O2" } */
> +/* { dg-final { check-function-bodies "**" "" } } */
> 
>  #include "arm_mve.h"
> 
> +/*
> +**foo:
> +**	...
> +**	vmlas.s32	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  int32x4_t
> -foo (int32x4_t a, int32x4_t b, int32_t c)
> +foo (int32x4_t m1, int32x4_t m2, int32_t add)
>  {
> -  return vmlasq_n_s32 (a, b, c);
> +  return vmlasq_n_s32 (m1, m2, add);
>  }
> 
> -/* { dg-final { scan-assembler "vmlas.s32"  }  } */
> 
> +/*
> +**foo1:
> +**	...
> +**	vmlas.s32	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  int32x4_t
> -foo1 (int32x4_t a, int32x4_t b, int32_t c)
> +foo1 (int32x4_t m1, int32x4_t m2, int32_t add)
>  {
> -  return vmlasq (a, b, c);
> +  return vmlasq (m1, m2, add);
>  }
> 
> -/* { dg-final { scan-assembler "vmlas.s32"  }  } */
> +/* { dg-final { scan-assembler-not "__ARM_undef" } } */
> \ No newline at end of file
> diff --git a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_s8.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_s8.c
> index 3a151650ef4..1cb1a31459c 100644
> --- a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_s8.c
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_s8.c
> @@ -1,21 +1,33 @@
>  /* { dg-require-effective-target arm_v8_1m_mve_ok } */
>  /* { dg-add-options arm_v8_1m_mve } */
>  /* { dg-additional-options "-O2" } */
> +/* { dg-final { check-function-bodies "**" "" } } */
> 
>  #include "arm_mve.h"
> 
> +/*
> +**foo:
> +**	...
> +**	vmlas.s8	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  int8x16_t
> -foo (int8x16_t a, int8x16_t b, int8_t c)
> +foo (int8x16_t m1, int8x16_t m2, int8_t add)
>  {
> -  return vmlasq_n_s8 (a, b, c);
> +  return vmlasq_n_s8 (m1, m2, add);
>  }
> 
> -/* { dg-final { scan-assembler "vmlas.s8"  }  } */
> 
> +/*
> +**foo1:
> +**	...
> +**	vmlas.s8	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  int8x16_t
> -foo1 (int8x16_t a, int8x16_t b, int8_t c)
> +foo1 (int8x16_t m1, int8x16_t m2, int8_t add)
>  {
> -  return vmlasq (a, b, c);
> +  return vmlasq (m1, m2, add);
>  }
> 
> -/* { dg-final { scan-assembler "vmlas.s8"  }  } */
> +/* { dg-final { scan-assembler-not "__ARM_undef" } } */
> \ No newline at end of file
> diff --git a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_u16.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_u16.c
> index b9444f2f6a3..e03c91ef298 100644
> --- a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_u16.c
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_u16.c
> @@ -1,21 +1,45 @@
>  /* { dg-require-effective-target arm_v8_1m_mve_ok } */
>  /* { dg-add-options arm_v8_1m_mve } */
>  /* { dg-additional-options "-O2" } */
> +/* { dg-final { check-function-bodies "**" "" } } */
> 
>  #include "arm_mve.h"
> 
> +/*
> +**foo:
> +**	...
> +**	vmlas.u16	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  uint16x8_t
> -foo (uint16x8_t a, uint16x8_t b, uint16_t c)
> +foo (uint16x8_t m1, uint16x8_t m2, uint16_t add)
>  {
> -  return vmlasq_n_u16 (a, b, c);
> +  return vmlasq_n_u16 (m1, m2, add);
>  }
> 
> -/* { dg-final { scan-assembler "vmlas.u16"  }  } */
> 
> +/*
> +**foo1:
> +**	...
> +**	vmlas.u16	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  uint16x8_t
> -foo1 (uint16x8_t a, uint16x8_t b, uint16_t c)
> +foo1 (uint16x8_t m1, uint16x8_t m2, uint16_t add)
>  {
> -  return vmlasq (a, b, c);
> +  return vmlasq (m1, m2, add);
>  }
> 
> -/* { dg-final { scan-assembler "vmlas.u16"  }  } */
> +/*
> +**foo2:
> +**	...
> +**	vmlas.u16	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
> +uint16x8_t
> +foo2 (uint16x8_t m1, uint16x8_t m2)
> +{
> +  return vmlasq (m1, m2, 1);
> +}
> +
> +/* { dg-final { scan-assembler-not "__ARM_undef" } } */
> \ No newline at end of file
> diff --git a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_u32.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_u32.c
> index 5708a0658a6..b80c3c7631f 100644
> --- a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_u32.c
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_u32.c
> @@ -1,21 +1,45 @@
>  /* { dg-require-effective-target arm_v8_1m_mve_ok } */
>  /* { dg-add-options arm_v8_1m_mve } */
>  /* { dg-additional-options "-O2" } */
> +/* { dg-final { check-function-bodies "**" "" } } */
> 
>  #include "arm_mve.h"
> 
> +/*
> +**foo:
> +**	...
> +**	vmlas.u32	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  uint32x4_t
> -foo (uint32x4_t a, uint32x4_t b, uint32_t c)
> +foo (uint32x4_t m1, uint32x4_t m2, uint32_t add)
>  {
> -  return vmlasq_n_u32 (a, b, c);
> +  return vmlasq_n_u32 (m1, m2, add);
>  }
> 
> -/* { dg-final { scan-assembler "vmlas.u32"  }  } */
> 
> +/*
> +**foo1:
> +**	...
> +**	vmlas.u32	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  uint32x4_t
> -foo1 (uint32x4_t a, uint32x4_t b, uint32_t c)
> +foo1 (uint32x4_t m1, uint32x4_t m2, uint32_t add)
>  {
> -  return vmlasq (a, b, c);
> +  return vmlasq (m1, m2, add);
>  }
> 
> -/* { dg-final { scan-assembler "vmlas.u32"  }  } */
> +/*
> +**foo2:
> +**	...
> +**	vmlas.u32	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
> +uint32x4_t
> +foo2 (uint32x4_t m1, uint32x4_t m2)
> +{
> +  return vmlasq (m1, m2, 1);
> +}
> +
> +/* { dg-final { scan-assembler-not "__ARM_undef" } } */
> \ No newline at end of file
> diff --git a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_u8.c
> b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_u8.c
> index d83940c7232..0f37550160e 100644
> --- a/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_u8.c
> +++ b/gcc/testsuite/gcc.target/arm/mve/intrinsics/vmlasq_n_u8.c
> @@ -1,21 +1,45 @@
>  /* { dg-require-effective-target arm_v8_1m_mve_ok } */
>  /* { dg-add-options arm_v8_1m_mve } */
>  /* { dg-additional-options "-O2" } */
> +/* { dg-final { check-function-bodies "**" "" } } */
> 
>  #include "arm_mve.h"
> 
> +/*
> +**foo:
> +**	...
> +**	vmlas.u8	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  uint8x16_t
> -foo (uint8x16_t a, uint8x16_t b, uint8_t c)
> +foo (uint8x16_t m1, uint8x16_t m2, uint8_t add)
>  {
> -  return vmlasq_n_u8 (a, b, c);
> +  return vmlasq_n_u8 (m1, m2, add);
>  }
> 
> -/* { dg-final { scan-assembler "vmlas.u8"  }  } */
> 
> +/*
> +**foo1:
> +**	...
> +**	vmlas.u8	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
>  uint8x16_t
> -foo1 (uint8x16_t a, uint8x16_t b, uint8_t c)
> +foo1 (uint8x16_t m1, uint8x16_t m2, uint8_t add)
>  {
> -  return vmlasq (a, b, c);
> +  return vmlasq (m1, m2, add);
>  }
> 
> -/* { dg-final { scan-assembler "vmlas.u8"  }  } */
> +/*
> +**foo2:
> +**	...
> +**	vmlas.u8	q[0-9]+, q[0-9]+, (?:ip|fp|r[0-9]+)(?:	@.*|)
> +**	...
> +*/
> +uint8x16_t
> +foo2 (uint8x16_t m1, uint8x16_t m2)
> +{
> +  return vmlasq (m1, m2, 1);
> +}
> +
> +/* { dg-final { scan-assembler-not "__ARM_undef" } } */
> \ No newline at end of file
> --
> 2.25.1


  reply	other threads:[~2022-11-22 16:56 UTC|newest]

Thread overview: 82+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-17 16:37 [PATCH 00/35] arm: rework MVE testsuite and rework backend where necessary (1st chunk) Andrea Corallo
2022-11-17 16:37 ` [PATCH 01/35] arm: improve vcreateq* tests Andrea Corallo
2022-11-18  9:47   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 02/35] arm: fix 'vmsr' spacing and register capitalization Andrea Corallo
2022-11-18 16:33   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 03/35] arm: improve tests and fix vddupq* Andrea Corallo
2022-11-18 16:34   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 04/35] arm: improve tests and fix vdwdupq* Andrea Corallo
2022-11-18 16:35   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 05/35] arm: improve vidupq* tests Andrea Corallo
2022-11-18 16:36   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 06/35] arm: improve tests and fix vdupq* Andrea Corallo
2022-11-18 16:37   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 07/35] arm: improve tests and fix vcmp* Andrea Corallo
2022-11-18 16:40   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 08/35] arm: improve tests for vmin* Andrea Corallo
2022-11-18 16:41   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 09/35] arm: improve tests for vmax* Andrea Corallo
2022-11-18 16:42   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 10/35] arm: improve tests for vabavq* Andrea Corallo
2022-11-18 16:43   ` Kyrylo Tkachov
2022-11-21 14:49     ` Andrea Corallo
2022-11-17 16:37 ` [PATCH 11/35] arm: improve tests for vabdq* Andrea Corallo
2022-11-18 16:44   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 12/35] arm: improve tests and fix vabsq* Andrea Corallo
2022-11-18 16:45   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 13/35] arm: further fix overloading of MVE vaddq[_m]_n intrinsic Andrea Corallo
2022-11-18 16:49   ` Kyrylo Tkachov
2022-11-21 10:45     ` Stam Markianos-Wright
2022-11-17 16:37 ` [PATCH 14/35] arm: propagate fixed overloading of MVE intrinsic scalar parameters Andrea Corallo
2022-11-18 16:51   ` Kyrylo Tkachov
2022-11-21 10:46     ` Stam Markianos-Wright
2022-11-17 16:37 ` [PATCH 15/35] arm: Explicitly specify other float types for _Generic overloading [PR107515] Andrea Corallo
2022-11-18 16:58   ` Kyrylo Tkachov
2022-11-20 22:49     ` Ramana Radhakrishnan
2022-11-21 14:11       ` Stam Markianos-Wright
2022-11-21 10:45     ` Stam Markianos-Wright
2022-11-17 16:37 ` [PATCH 16/35] arm: Add integer vector overloading of vsubq_x instrinsic Andrea Corallo
2022-11-22 10:00   ` Christophe Lyon
2022-11-22 10:54     ` Andrea Corallo
2022-11-22 16:48   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 17/35] arm: improve tests and fix vadd* Andrea Corallo
2022-11-22 16:49   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 18/35] arm: improve tests for vmulq* Andrea Corallo
2022-11-22 16:51   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 19/35] arm: improve tests and fix vsubq* Andrea Corallo
2022-11-22 16:51   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 20/35] arm: improve tests for vfmasq_m* Andrea Corallo
2022-11-22 16:52   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 21/35] arm: improve tests for vhaddq_m* Andrea Corallo
2022-11-22 16:53   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 22/35] arm: improve tests for vhsubq_m* Andrea Corallo
2022-11-22 16:53   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 23/35] arm: improve tests for viwdupq* Andrea Corallo
2022-11-22 16:54   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 24/35] arm: improve tests for vmladavaq* Andrea Corallo
2022-11-22 16:54   ` Kyrylo Tkachov
2022-11-17 16:37 ` [PATCH 25/35] arm: improve tests and fix vmlaldavaxq* Andrea Corallo
2022-11-22 16:56   ` Kyrylo Tkachov
2022-11-17 16:38 ` [PATCH 26/35] arm: improve tests for vmlasq* Andrea Corallo
2022-11-22 16:56   ` Kyrylo Tkachov [this message]
2022-11-17 16:38 ` [PATCH 27/35] arm: improve tests for vqaddq_m* Andrea Corallo
2022-11-22 16:57   ` Kyrylo Tkachov
2022-11-17 16:38 ` [PATCH 28/35] arm: improve tests for vqdmlahq_m* Andrea Corallo
2022-11-22 16:57   ` Kyrylo Tkachov
2022-11-17 16:38 ` [PATCH 29/35] arm: improve tests for vqdmul* Andrea Corallo
2022-11-22 16:58   ` Kyrylo Tkachov
2022-11-17 16:38 ` [PATCH 30/35] arm: improve tests for vqrdmlahq* Andrea Corallo
2022-11-22 17:01   ` Kyrylo Tkachov
2022-11-17 16:38 ` [PATCH 31/35] arm: improve tests for vqrdmlashq_m* Andrea Corallo
2022-11-22 17:02   ` Kyrylo Tkachov
2022-11-17 16:38 ` [PATCH 32/35] arm: improve tests for vqsubq* Andrea Corallo
2022-11-22 17:03   ` Kyrylo Tkachov
2022-11-17 16:38 ` [PATCH 33/35] arm: improve tests and fix vrmlaldavhaq* Andrea Corallo
2022-11-22 17:03   ` Kyrylo Tkachov
2022-11-17 16:38 ` [PATCH 34/35] arm: improve tests for vrshlq* Andrea Corallo
2022-11-22 17:04   ` Kyrylo Tkachov
2022-11-17 16:38 ` [PATCH 35/35] arm: improve tests for vsetq_lane* Andrea Corallo
2022-11-22 17:06   ` Kyrylo Tkachov
2022-11-24 14:43     ` [PATCH 35/35 V2] " Andrea Corallo
2022-11-24 15:28       ` Kyrylo Tkachov
2022-11-28  9:20 ` [PATCH 00/35] arm: rework MVE testsuite and rework backend where necessary (1st chunk) Andrea Corallo

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=PAXPR08MB69263B4E0DDCC8A796E4EB7E930D9@PAXPR08MB6926.eurprd08.prod.outlook.com \
    --to=kyrylo.tkachov@arm.com \
    --cc=Andrea.Corallo@arm.com \
    --cc=Richard.Earnshaw@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).