https://gcc.gnu.org/g:d50776bbd70b766f0cd9f40ba31a5167176c9a0a
commit d50776bbd70b766f0cd9f40ba31a5167176c9a0a Author: Michael Meissner <[email protected]> Date: Fri Jan 30 17:03:26 2026 -0500 PR target/117251: Improve vector andc to vector eqv 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 #33 of 45 to generate the 'XXEVAL' instruction on power10 and power11 instead of using the Altivec 'VANDC' instruction feeding into 'VEQV'. 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: vandc t,c,d veqv 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,210 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? 2026-01-30 Michael Meissner <[email protected]> gcc/ PR target/117251 * config/rs6000/fusion.md: Regenerate. * config/rs6000/genfusion.pl (gen_logical_addsubf): Add support to generate vector andc => eqv 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 a1129799c1b4..ab42abdbe1a5 100644 --- a/gcc/config/rs6000/fusion.md +++ b/gcc/config/rs6000/fusion.md @@ -2204,20 +2204,23 @@ ;; logical-logical fusion pattern generated by gen_logical_addsubf ;; vector vandc -> veqv (define_insn "*fuse_vandc_veqv" - [(set (match_operand:VM 3 "altivec_register_operand" "=&0,&1,&v,v") - (not:VM (xor:VM (and: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") + (not:VM (xor:VM (and: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)" "@ vandc %3,%1,%0\;veqv %3,%3,%2 vandc %3,%1,%0\;veqv %3,%3,%2 vandc %3,%1,%0\;veqv %3,%3,%2 + xxeval %x3,%x2,%x1,%x0,210 vandc %4,%1,%0\;veqv %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 veqv -> veqv diff --git a/gcc/config/rs6000/genfusion.pl b/gcc/config/rs6000/genfusion.pl index b9f9d274f519..998f2c7e2e12 100755 --- a/gcc/config/rs6000/genfusion.pl +++ b/gcc/config/rs6000/genfusion.pl @@ -247,6 +247,7 @@ sub gen_logical_addsubf "vorc_vxor" => 180, "vorc_vor" => 191, "vandc_vnor" => 208, + "vandc_veqv" => 210, ); KIND: foreach $kind ('scalar','vector') {
