https://gcc.gnu.org/g:a444dc4bd0ba208f1ac06ac67bef918f5ffbf1f6
commit a444dc4bd0ba208f1ac06ac67bef918f5ffbf1f6
Author: Michael Meissner <meiss...@linux.ibm.com>
Date: Wed Jul 9 01:38:23 2025 -0400
PR target/117251: Improve vector orc to vector and fusion
See the following post for a complete explanation of what the patches
for PR target/117251:
* https://gcc.gnu.org/pipermail/gcc-patches/2025-June/686474.html
This is patch #7 of 45 to generate the 'XXEVAL' instruction on power10
and power11 instead of using the Altivec 'VORC' instruction feeding
into 'VAND'. The 'XXEVAL' instruction can use all 64 vector registers,
instead of the 32 registers that traditional Altivec vector
instructions use. By allowing all of the vector registers to be used,
it reduces the amount of spilling that a large benchmark generated.
Currently the following code:
vector int a, b, c, d;
a = (c | ~ d) & b;
Generates:
vorc t,c,d
vand a,t,b
Now in addition with this patch, if the arguments or result is
allocated to a traditional FPR register, the GCC compiler will now
generate the following code instead of adding vector move instructions:
xxeval a,b,c,11
Since fusion using 2 Altivec instructions is slightly faster than using
the 'XXEVAL' instruction we prefer to generate the Altivec instructions
if we can. In addition, because 'XXEVAL' is a prefixed instruction, it
possibly might generate an extra NOP instruction to align the 'XXEVAL'
instruction.
I have tested these patches on both big endian and little endian
PowerPC servers, with no regressions. Can I check these patchs into
the trunk?
2025-07-09 Michael Meissner <meiss...@linux.ibm.com>
gcc/
PR target/117251
* config/rs6000/fusion.md: Regenerate.
* config/rs6000/genfusion.pl (gen_logical_addsubf): Add support
to generate vector/vector orc/and fusion if XXEVAL is
supported.
Diff:
---
gcc/config/rs6000/fusion.md | 15 +++++++++------
gcc/config/rs6000/genfusion.pl | 1 +
2 files changed, 10 insertions(+), 6 deletions(-)
diff --git a/gcc/config/rs6000/fusion.md b/gcc/config/rs6000/fusion.md
index cce179e0c974..e27f05f85f12 100644
--- a/gcc/config/rs6000/fusion.md
+++ b/gcc/config/rs6000/fusion.md
@@ -1994,20 +1994,23 @@
;; logical-logical fusion pattern generated by gen_logical_addsubf
;; vector vorc -> vand
(define_insn "*fuse_vorc_vand"
- [(set (match_operand:VM 3 "altivec_register_operand" "=&0,&1,&v,v")
- (and:VM (ior:VM (not:VM (match_operand:VM 0 "altivec_register_operand"
"v,v,v,v"))
- (match_operand:VM 1 "altivec_register_operand"
"v,v,v,v"))
- (match_operand:VM 2 "altivec_register_operand" "v,v,v,v")))
- (clobber (match_scratch:VM 4 "=X,X,X,&v"))]
+ [(set (match_operand:VM 3 "vector_fusion_operand" "=&0,&1,&v,wa,v")
+ (and:VM (ior:VM (not:VM (match_operand:VM 0 "vector_fusion_operand"
"v,v,v,wa,v"))
+ (match_operand:VM 1 "vector_fusion_operand"
"v,v,v,wa,v"))
+ (match_operand:VM 2 "vector_fusion_operand" "v,v,v,wa,v")))
+ (clobber (match_scratch:VM 4 "=X,X,X,X,&v"))]
"(TARGET_P10_FUSION)"
"@
vorc %3,%1,%0\;vand %3,%3,%2
vorc %3,%1,%0\;vand %3,%3,%2
vorc %3,%1,%0\;vand %3,%3,%2
+ xxeval %x3,%x2,%x1,%x0,11
vorc %4,%1,%0\;vand %3,%4,%2"
[(set_attr "type" "fused_vector")
(set_attr "cost" "6")
- (set_attr "length" "8")])
+ (set_attr "length" "8")
+ (set_attr "prefixed" "*,*,*,yes,*")
+ (set_attr "isa" "*,*,*,xxeval,*")])
;; logical-logical fusion pattern generated by gen_logical_addsubf
;; vector vxor -> vand
diff --git a/gcc/config/rs6000/genfusion.pl b/gcc/config/rs6000/genfusion.pl
index 268b94089484..a3cc8b121eab 100755
--- a/gcc/config/rs6000/genfusion.pl
+++ b/gcc/config/rs6000/genfusion.pl
@@ -221,6 +221,7 @@ sub gen_logical_addsubf
"vor_vand" => 7,
"vnor_vand" => 8,
"veqv_vand" => 9,
+ "vorc_vand" => 11,
);
KIND: foreach $kind ('scalar','vector') {
