[PATCH, rs6000] Merge two vector shift when their sources are the same

[PATCH, rs6000] Merge two vector shift when their sources are the same

[HAO CHEN GUI]

Hi,
  This patch merges two "vsldoi" insns when their sources are the
same. Particularly, it is simplified to be one move if the total
shift is multiples of 16 bytes.

  Bootstrapped and tested on powerpc64-linux BE and LE with no
regressions.

Thanks
Gui Haochen


ChangeLog
2023-02-20  Haochen Gui <guihaoc@linux.ibm.com>

gcc/
	* config/rs6000/altivec.md (*altivec_vsldoi_dup_<mode>): New
	insn_and_split to merge two vsldoi.

gcc/testsuite/
	* gcc.target/powerpc/vsldoi_merge.c: New.


patch.diff
diff --git a/gcc/config/rs6000/altivec.md b/gcc/config/rs6000/altivec.md
index 84660073f32..22e9c4c1fc5 100644
--- a/gcc/config/rs6000/altivec.md
+++ b/gcc/config/rs6000/altivec.md
@@ -2529,6 +2529,35 @@ (define_insn "altivec_vsldoi_<mode>"
   "vsldoi %0,%1,%2,%3"
   [(set_attr "type" "vecperm")])

+(define_insn_and_split "*altivec_vsldoi_dup_<mode>"
+  [(set (match_operand:VM 0 "register_operand" "=v")
+	(unspec:VM [(unspec:VM [(match_operand:VM 1 "register_operand" "v")
+				(match_operand:VM 2 "register_operand" "v")
+				(match_operand:QI 3 "immediate_operand" "i")]
+			       UNSPEC_VSLDOI)
+		    (unspec:VM [(match_dup 1)
+				(match_dup 2)
+				(match_dup 3)]
+			       UNSPEC_VSLDOI)
+		    (match_operand:QI 4 "immediate_operand" "i")]
+		   UNSPEC_VSLDOI))]
+  "TARGET_ALTIVEC"
+  "#"
+  "&& 1"
+  [(const_int 0)]
+{
+  unsigned int shift1 = UINTVAL (operands[3]);
+  unsigned int shift2 = UINTVAL (operands[4]);
+
+  unsigned int shift = (shift1 + shift2) % 16;
+  if (shift)
+    emit_insn (gen_altivec_vsldoi_<mode> (operands[0], operands[1],
+					  operands[1], GEN_INT (shift)));
+  else
+    emit_move_insn (operands[0], operands[1]);
+  DONE;
+})
+
 (define_insn "altivec_vupkhs<VU_char>"
   [(set (match_operand:VP 0 "register_operand" "=v")
 	(unspec:VP [(match_operand:<VP_small> 1 "register_operand" "v")]
diff --git a/gcc/testsuite/gcc.target/powerpc/vsldoi_merge.c b/gcc/testsuite/gcc.target/powerpc/vsldoi_merge.c
new file mode 100644
index 00000000000..4ea72561282
--- /dev/null
+++ b/gcc/testsuite/gcc.target/powerpc/vsldoi_merge.c
@@ -0,0 +1,21 @@
+/* { dg-do compile } */
+/* { dg-require-effective-target powerpc_vsx_ok } */
+/* { dg-options "-O2 -mvsx" } */
+
+#include "altivec.h"
+
+vector signed int test1 (vector signed int a, vector signed int b)
+{
+  a = vec_sld (a, b, 2);
+  a = vec_sld (a, a, 4);
+  return a;
+}
+
+vector signed int test2 (vector signed int a, vector signed int b)
+{
+  a = vec_sld (a, b, 14);
+  a = vec_sld (a, a, 2);
+  return a;
+}
+
+/* { dg-final { scan-assembler-times {\mvsldoi\M} 1 } } */

Re: [PATCH, rs6000] Merge two vector shift when their sources are the same

[Kewen.Lin]

Hi Haochen,

on 2023/2/20 10:04, HAO CHEN GUI wrote:
> Hi,
>   This patch merges two "vsldoi" insns when their sources are the
> same. Particularly, it is simplified to be one move if the total
> shift is multiples of 16 bytes.
> 
>   Bootstrapped and tested on powerpc64-linux BE and LE with no
> regressions.
> 
> Thanks
> Gui Haochen
> 
> 
> ChangeLog
> 2023-02-20  Haochen Gui <guihaoc@linux.ibm.com>
> 
> gcc/
> 	* config/rs6000/altivec.md (*altivec_vsldoi_dup_<mode>): New
> 	insn_and_split to merge two vsldoi.
> 
> gcc/testsuite/
> 	* gcc.target/powerpc/vsldoi_merge.c: New.
> 
> 
> patch.diff
> diff --git a/gcc/config/rs6000/altivec.md b/gcc/config/rs6000/altivec.md
> index 84660073f32..22e9c4c1fc5 100644
> --- a/gcc/config/rs6000/altivec.md
> +++ b/gcc/config/rs6000/altivec.md
> @@ -2529,6 +2529,35 @@ (define_insn "altivec_vsldoi_<mode>"
>    "vsldoi %0,%1,%2,%3"
>    [(set_attr "type" "vecperm")])
> 
> +(define_insn_and_split "*altivec_vsldoi_dup_<mode>"
> +  [(set (match_operand:VM 0 "register_operand" "=v")
> +	(unspec:VM [(unspec:VM [(match_operand:VM 1 "register_operand" "v")
> +				(match_operand:VM 2 "register_operand" "v")
> +				(match_operand:QI 3 "immediate_operand" "i")]
> +			       UNSPEC_VSLDOI)
> +		    (unspec:VM [(match_dup 1)
> +				(match_dup 2)
> +				(match_dup 3)]
> +			       UNSPEC_VSLDOI)
> +		    (match_operand:QI 4 "immediate_operand" "i")]
> +		   UNSPEC_VSLDOI))]
> +  "TARGET_ALTIVEC"
> +  "#"
> +  "&& 1"
> +  [(const_int 0)]
> +{
> +  unsigned int shift1 = UINTVAL (operands[3]);
> +  unsigned int shift2 = UINTVAL (operands[4]);
> +
> +  unsigned int shift = (shift1 + shift2) % 16;
> +  if (shift)
> +    emit_insn (gen_altivec_vsldoi_<mode> (operands[0], operands[1],
> +					  operands[1], GEN_INT (shift)));
> +  else
> +    emit_move_insn (operands[0], operands[1]);
> +  DONE;
> +})

This patch looks wrong, I think we need to ensure operand 1 and operand 2
are the same (dup 1 for 2)? one simple counter example for this proposed
fix is that two given vector a {A0, A1} and {B0, B1} (all A0/A1/B0/B1 are
doublewords) on BE:

a = vec_sld (a, b, 8);  // (1) res a = {A1, B0}
a = vec_sld (a, a, 8);  // (2) res a = {B0, A1}

it would get the unexpected result a {A0, A1} with this patch.

Since this patch got bootstrapped and regress-tested, I think we don't
have enough coverage on this part, it's a good thing to add one dg-do run
test case as well. :)

BR,
Kewen