Mesa (staging/20.0): ac/llvm: fix 64-bit fmed3

[GitLab Mirror]

Module: Mesa
Branch: staging/20.0
Commit: c2f6f63f7bf81a88cc4cbb36b1e0b9de25c17b5a
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=c2f6f63f7bf81a88cc4cbb36b1e0b9de25c17b5a

Author: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Date:   Wed Feb 26 15:04:38 2020 +0100

ac/llvm: fix 64-bit fmed3

Lower 64-bit fmed3 because LLVM doesn't expose an intrinsic.

Fixes dEQP-VK.spirv_assembly.instruction.amd_trinary_minmax.mid3.f64.*.

Fixes: d6a07732c9c ("ac: use llvm.amdgcn.fmed3 intrinsic for nir_op_fmed3")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas at basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3962>
(cherry picked from commit 50b8c2527464dbe18a01ab6412de4465cebf2225)

---

 .pick_status.json            |  2 +-
 src/amd/llvm/ac_llvm_build.c | 48 ++++++++++++++++++++++++++++----------------
 2 files changed, 32 insertions(+), 18 deletions(-)

diff --git a/.pick_status.json b/.pick_status.json
index eb9d9a2077d..43f623965f6 100644
--- a/.pick_status.json
+++ b/.pick_status.json
@@ -1066,7 +1066,7 @@
         "description": "ac/llvm: fix 64-bit fmed3",
         "nominated": true,
         "nomination_type": 1,
-        "resolution": 0,
+        "resolution": 1,
         "master_sha": null,
         "because_sha": "d6a07732c9c155c73f7d2cddc10faa7eab768df9"
     },
diff --git a/src/amd/llvm/ac_llvm_build.c b/src/amd/llvm/ac_llvm_build.c
index 8d706944a45..db7964d6aa9 100644
--- a/src/amd/llvm/ac_llvm_build.c
+++ b/src/amd/llvm/ac_llvm_build.c
@@ -2723,27 +2723,41 @@ LLVMValueRef ac_build_fmed3(struct ac_llvm_context *ctx, LLVMValueRef src0,
 			    LLVMValueRef src1, LLVMValueRef src2,
 			    unsigned bitsize)
 {
-	LLVMTypeRef type;
-	char *intr;
+	LLVMValueRef result;
 
-	if (bitsize == 16) {
-		intr = "llvm.amdgcn.fmed3.f16";
-		type = ctx->f16;
-	} else if (bitsize == 32) {
-		intr = "llvm.amdgcn.fmed3.f32";
-		type = ctx->f32;
+	if (bitsize == 64) {
+		/* Lower 64-bit fmed because LLVM doesn't expose an intrinsic. */
+		LLVMValueRef min1, min2, max1;
+
+		min1 = ac_build_fmin(ctx, src0, src1);
+		max1 = ac_build_fmax(ctx, src0, src1);
+		min2 = ac_build_fmin(ctx, max1, src2);
+
+		result = ac_build_fmax(ctx, min2, min1);
 	} else {
-		intr = "llvm.amdgcn.fmed3.f64";
-		type = ctx->f64;
+		LLVMTypeRef type;
+		char *intr;
+
+		if (bitsize == 16) {
+			intr = "llvm.amdgcn.fmed3.f16";
+			type = ctx->f16;
+		} else {
+			assert(bitsize == 32);
+			intr = "llvm.amdgcn.fmed3.f32";
+			type = ctx->f32;
+		}
+
+		LLVMValueRef params[] = {
+			src0,
+			src1,
+			src2,
+		};
+
+		result = ac_build_intrinsic(ctx, intr, type, params, 3,
+					    AC_FUNC_ATTR_READNONE);
 	}
 
-	LLVMValueRef params[] = {
-		src0,
-		src1,
-		src2,
-	};
-	return ac_build_intrinsic(ctx, intr, type, params, 3,
-				  AC_FUNC_ATTR_READNONE);
+	return result;
 }
 
 LLVMValueRef ac_build_fract(struct ac_llvm_context *ctx, LLVMValueRef src0,