RE: [PATCH 3/4][AArch32]: Add support for sign differing dot-product usdot for NEON.

[Kyrylo Tkachov <Kyrylo.Tkachov@arm.com>]

> -----Original Message-----
> From: Tamar Christina <Tamar.Christina@arm.com>
> Sent: 25 May 2021 16:02
> To: gcc-patches@gcc.gnu.org
> Cc: Richard Earnshaw <Richard.Earnshaw@arm.com>; nd <nd@arm.com>;
> Ramana Radhakrishnan <Ramana.Radhakrishnan@arm.com>; Kyrylo
> Tkachov <Kyrylo.Tkachov@arm.com>
> Subject: RE: [PATCH 3/4][AArch32]: Add support for sign differing dot-
> product usdot for NEON.
> 
> Forgot to include the list
> 
> > -----Original Message-----
> > From: Tamar Christina
> > Sent: Tuesday, May 25, 2021 3:57 PM
> > To: Tamar Christina <Tamar.Christina@arm.com>
> > Cc: Richard Earnshaw <Richard.Earnshaw@arm.com>; nd <nd@arm.com>;
> > Ramana Radhakrishnan <Ramana.Radhakrishnan@arm.com>; Kyrylo
> Tkachov
> > <Kyrylo.Tkachov@arm.com>
> > Subject: RE: [PATCH 3/4][AArch32]: Add support for sign differing dot-
> > product usdot for NEON.
> >
> > Hi All,
> >
> > This is a respin based on the feedback gotten from the AArch64 review.
> >
> > Bootstrapped Regtested on aarch64-none-linux-gnu and no issues.
> >
> > Ok for master?
> >

Ok.
Thanks,
Kyrill

> > Thanks,
> > Tamar
> >
> > gcc/ChangeLog:
> >
> > 	* config/arm/neon.md (usdot_prod<vsi2qi>): New.
> >
> > gcc/testsuite/ChangeLog:
> >
> > 	* gcc.target/arm/simd/vusdot-autovec.c: New test.
> >
> > > -----Original Message-----
> > > From: Gcc-patches <gcc-patches-bounces@gcc.gnu.org> On Behalf Of
> > Tamar
> > > Christina via Gcc-patches
> > > Sent: Wednesday, May 5, 2021 6:42 PM
> > > To: gcc Patches <gcc-patches@gcc.gnu.org>
> > > Cc: Richard Earnshaw <Richard.Earnshaw@arm.com>; nd
> <nd@arm.com>;
> > > Ramana Radhakrishnan <Ramana.Radhakrishnan@arm.com>
> > > Subject: FW: [PATCH 3/4][AArch32]: Add support for sign differing dot-
> > > product usdot for NEON.
> > >
> > > Forgot to CC maintainers..
> > >
> > > -----Original Message-----
> > > From: Tamar Christina <tamar.christina@arm.com>
> > > Sent: Wednesday, May 5, 2021 6:39 PM
> > > To: gcc-patches@gcc.gnu.org
> > > Cc: nd <nd@arm.com>
> > > Subject: [PATCH 3/4][AArch32]: Add support for sign differing
> > > dot-product usdot for NEON.
> > >
> > > Hi All,
> > >
> > > This adds optabs implementing usdot_prod.
> > >
> > > The following testcase:
> > >
> > > #define N 480
> > > #define SIGNEDNESS_1 unsigned
> > > #define SIGNEDNESS_2 signed
> > > #define SIGNEDNESS_3 signed
> > > #define SIGNEDNESS_4 unsigned
> > >
> > > SIGNEDNESS_1 int __attribute__ ((noipa)) f (SIGNEDNESS_1 int res,
> > > SIGNEDNESS_3 char *restrict a,
> > >    SIGNEDNESS_4 char *restrict b)
> > > {
> > >   for (__INTPTR_TYPE__ i = 0; i < N; ++i)
> > >     {
> > >       int av = a[i];
> > >       int bv = b[i];
> > >       SIGNEDNESS_2 short mult = av * bv;
> > >       res += mult;
> > >     }
> > >   return res;
> > > }
> > >
> > > Generates
> > >
> > > f:
> > >         vmov.i32        q8, #0  @ v4si
> > >         add     r3, r2, #480
> > > .L2:
> > >         vld1.8  {q10}, [r2]!
> > >         vld1.8  {q9}, [r1]!
> > >         vusdot.s8       q8, q9, q10
> > >         cmp     r3, r2
> > >         bne     .L2
> > >         vadd.i32        d16, d16, d17
> > >         vpadd.i32       d16, d16, d16
> > >         vmov.32 r3, d16[0]
> > >         add     r0, r0, r3
> > >         bx      lr
> > >
> > > instead of
> > >
> > > f:
> > >         vmov.i32        q8, #0  @ v4si
> > >         add     r3, r2, #480
> > > .L2:
> > >         vld1.8  {q9}, [r2]!
> > >         vld1.8  {q11}, [r1]!
> > >         cmp     r3, r2
> > >         vmull.s8 q10, d18, d22
> > >         vmull.s8 q9, d19, d23
> > >         vaddw.s16       q8, q8, d20
> > >         vaddw.s16       q8, q8, d21
> > >         vaddw.s16       q8, q8, d18
> > >         vaddw.s16       q8, q8, d19
> > >         bne     .L2
> > >         vadd.i32        d16, d16, d17
> > >         vpadd.i32       d16, d16, d16
> > >         vmov.32 r3, d16[0]
> > >         add     r0, r0, r3
> > >         bx      lr
> > >
> > > For NEON.  I couldn't figure out if the MVE instruction vmlaldav.s16
> > > could be used to emulate this.  Because it would require additional
> > > widening to work I left MVE out of this patch set but perhaps someone
> > should take a look.
> > >
> > > Bootstrapped Regtested on aarch64-none-linux-gnu and no issues.
> > >
> > > Ok for master?
> > >
> > > Thanks,
> > > Tamar
> > >
> > > gcc/ChangeLog:
> > >
> > > 	* config/arm/neon.md (usdot_prod<vsi2qi>): New.
> > >
> > > gcc/testsuite/ChangeLog:
> > >
> > > 	* gcc.target/arm/simd/vusdot-autovec.c: New test.
> > >
> > > --- inline copy of patch --
> > > diff --git a/gcc/config/arm/neon.md b/gcc/config/arm/neon.md index
> > >
> >
> fec2cc91d24b6eff7b6fc8fdd54f39b3d646c468..23ad411178db77c5d19bee74
> 52
> > > bc1070331c1aa0 100644
> > > --- a/gcc/config/arm/neon.md
> > > +++ b/gcc/config/arm/neon.md
> > > @@ -3075,6 +3075,24 @@ (define_expand "<sup>dot_prod<vsi2qi>"
> > >    DONE;
> > >  })
> > >
> > > +;; Auto-vectorizer pattern for usdot
> > > +(define_expand "usdot_prod<vsi2qi>"
> > > +  [(set (match_operand:VCVTI 0 "register_operand")
> > > +	(plus:VCVTI (unspec:VCVTI [(match_operand:<VSI2QI> 1
> > > +							"register_operand")
> > > +				   (match_operand:<VSI2QI> 2
> > > +							"register_operand")]
> > > +		     UNSPEC_DOT_US)
> > > +		    (match_operand:VCVTI 3 "register_operand")))]
> > > +  "TARGET_I8MM"
> > > +{
> > > +  emit_insn (
> > > +    gen_neon_usdot<vsi2qi> (operands[3], operands[3], operands[1],
> > > +			    operands[2]));
> > > +  emit_insn (gen_rtx_SET (operands[0], operands[3]));
> > > +  DONE;
> > > +})
> > > +
> > >  (define_expand "neon_copysignf<mode>"
> > >    [(match_operand:VCVTF 0 "register_operand")
> > >     (match_operand:VCVTF 1 "register_operand") diff --git
> > > a/gcc/testsuite/gcc.target/arm/simd/vusdot-autovec.c
> > > b/gcc/testsuite/gcc.target/arm/simd/vusdot-autovec.c
> > > new file mode 100644
> > > index
> > >
> >
> 0000000000000000000000000000000000000000..7cc56f68817d77d6950df0
> ab37
> > > 2d6fbaad6b3813
> > > --- /dev/null
> > > +++ b/gcc/testsuite/gcc.target/arm/simd/vusdot-autovec.c
> > > @@ -0,0 +1,38 @@
> > > +/* { dg-do compile } */
> > > +/* { dg-options "-O3 -march=armv8.2-a+i8mm" } */
> > > +
> > > +#define N 480
> > > +#define SIGNEDNESS_1 unsigned
> > > +#define SIGNEDNESS_2 signed
> > > +#define SIGNEDNESS_3 signed
> > > +#define SIGNEDNESS_4 unsigned
> > > +
> > > +SIGNEDNESS_1 int __attribute__ ((noipa)) f (SIGNEDNESS_1 int res,
> > > +SIGNEDNESS_3 char *restrict a,
> > > +   SIGNEDNESS_4 char *restrict b)
> > > +{
> > > +  for (__INTPTR_TYPE__ i = 0; i < N; ++i)
> > > +    {
> > > +      int av = a[i];
> > > +      int bv = b[i];
> > > +      SIGNEDNESS_2 short mult = av * bv;
> > > +      res += mult;
> > > +    }
> > > +  return res;
> > > +}
> > > +
> > > +SIGNEDNESS_1 int __attribute__ ((noipa)) g (SIGNEDNESS_1 int res,
> > > +SIGNEDNESS_3 char *restrict b,
> > > +   SIGNEDNESS_4 char *restrict a)
> > > +{
> > > +  for (__INTPTR_TYPE__ i = 0; i < N; ++i)
> > > +    {
> > > +      int av = a[i];
> > > +      int bv = b[i];
> > > +      SIGNEDNESS_2 short mult = av * bv;
> > > +      res += mult;
> > > +    }
> > > +  return res;
> > > +}
> > > +
> > > +/* { dg-final { scan-assembler-times {vusdot.s8} 2 { target {
> > > +arm-*-*-gnueabihf } } } } */
> > >
> > >
> > > --