Hi!
While *minmax<mode>3_1 correctly uses
if (MEM_P (operands[1]))
operands[1] = force_reg (<MODE>mode, operands[1]);
to ensure operands[1] is not a MEM, *minmax<mode>3_2 does it wrongly
by calling force_reg but ignoring its return value.
The following borderingly obvious patch fixes that, bootstrapped/regtested
on x86_64-linux and i686-linux, ok for trunk?
Didn't find similar other errors in the backend with force_reg calls.
2024-10-04 Jakub Jelinek <ja...@redhat.com>
PR target/116925
* config/i386/sse.md (*minmax<mode>3_2): Assign force_reg result
back to operands[2] instead of throwing it away.
* g++.target/i386/avx-pr116925.C: New test.
--- gcc/config/i386/sse.md.jj 2024-10-01 09:38:57.000000000 +0200
+++ gcc/config/i386/sse.md 2024-10-03 17:33:12.071507421 +0200
@@ -3269,7 +3269,7 @@ (define_insn_and_split "*minmax<mode>3_2
u = UNSPEC_IEEE_MAX;
if (MEM_P (operands[2]))
- force_reg (<MODE>mode, operands[2]);
+ operands[2] = force_reg (<MODE>mode, operands[2]);
rtvec v = gen_rtvec (2, operands[2], operands[1]);
rtx tmp = gen_rtx_UNSPEC (<MODE>mode, v, u);
emit_move_insn (operands[0], tmp);
--- gcc/testsuite/g++.target/i386/avx-pr116925.C.jj 2024-10-03
17:36:10.124061111 +0200
+++ gcc/testsuite/g++.target/i386/avx-pr116925.C 2024-10-03
17:35:26.805656671 +0200
@@ -0,0 +1,12 @@
+// PR target/116925
+// { dg-do compile }
+// { dg-options "-O2 -mavx -ffloat-store" }
+
+typedef float V __attribute__((vector_size (16)));
+V a, b, c;
+
+void
+foo ()
+{
+ c = a > b ? a : b;
+}
Jakub
