From 5231c924ea13f71c1f9be5e722b575ceda316ac4 Mon Sep 17 00:00:00 2001 From: Georg Lehmann Date: Thu, 9 Apr 2026 17:36:45 +0200 Subject: [PATCH] intel/peephole_fma: use nir_fp_no_contract instead of exact Reviewed-by: Alyssa Rosenzweig Part-of: --- src/intel/compiler/intel_nir_opt_peephole_ffma.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/src/intel/compiler/intel_nir_opt_peephole_ffma.c b/src/intel/compiler/intel_nir_opt_peephole_ffma.c index b94bb6141e4..8fefc4e14c8 100644 --- a/src/intel/compiler/intel_nir_opt_peephole_ffma.c +++ b/src/intel/compiler/intel_nir_opt_peephole_ffma.c @@ -50,13 +50,13 @@ get_mul_for_src(nir_alu_src *src, unsigned num_components, nir_alu_instr *alu = nir_src_as_alu(src->src); /* We want to bail if any of the other ALU operations involved is labeled - * exact. One reason for this is that, while the value that is changing is + * no-contract. One reason for this is that, while the value that is changing is * actually the result of the add and not the multiply, the intention of - * the user when they specify an exact multiply is that they want *that* + * the user when they specify an no-contract multiply is that they want *that* * value and what they don't care about is the add. Another reason is that * SPIR-V explicitly requires this behaviour. */ - if (!alu || nir_alu_instr_is_exact(alu)) + if (!alu || nir_alu_instr_no_contract(alu)) return NULL; switch (alu->op) { @@ -142,7 +142,7 @@ intel_nir_opt_peephole_ffma_instr(nir_builder *b, if (add->op != nir_op_fadd) return false; - if (nir_alu_instr_is_exact(add)) + if (nir_alu_instr_no_contract(add)) return false;