[llvm] r338569 - AMDGPU: Allow fp32-denormals feature for r600 targets

[Jan Vesely via llvm-commits]

Author: jvesely
Date: Wed Aug  1 08:04:36 2018
New Revision: 338569

URL: http://llvm.org/viewvc/llvm-project?rev=338569&view=rev
Log:
AMDGPU: Allow fp32-denormals feature for r600 targets

This was accidentally removed in r335942.

Differential Revision: https://reviews.llvm.org/D49934

Modified:
    llvm/trunk/lib/Target/AMDGPU/AMDGPU.td
    llvm/trunk/lib/Target/AMDGPU/AMDGPUFeatures.td

Modified: llvm/trunk/lib/Target/AMDGPU/AMDGPU.td
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/AMDGPU.td?rev=338569&r1=338568&r2=338569&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AMDGPU/AMDGPU.td (original)
+++ llvm/trunk/lib/Target/AMDGPU/AMDGPU.td Wed Aug  1 08:04:36 2018
@@ -267,15 +267,6 @@ def FeatureD16PreservesUnusedBits : Subt
 // Subtarget Features (options and debugging)
 //===------------------------------------------------------------===//
 
-// Some instructions do not support denormals despite this flag. Using
-// fp32 denormals also causes instructions to run at the double
-// precision rate for the device.
-def FeatureFP32Denormals : SubtargetFeature<"fp32-denormals",
-  "FP32Denormals",
-  "true",
-  "Enable single precision denormal handling"
->;
-
 // Denormal handling for fp64 and fp16 is controlled by the same
 // config register when fp16 supported.
 // TODO: Do we need a separate f16 setting when not legal?

Modified: llvm/trunk/lib/Target/AMDGPU/AMDGPUFeatures.td
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/AMDGPUFeatures.td?rev=338569&r1=338568&r2=338569&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AMDGPU/AMDGPUFeatures.td (original)
+++ llvm/trunk/lib/Target/AMDGPU/AMDGPUFeatures.td Wed Aug  1 08:04:36 2018
@@ -19,6 +19,15 @@ def FeatureFMA : SubtargetFeature<"fmaf"
   "Enable single precision FMA (not as fast as mul+add, but fused)"
 >;
 
+// Some instructions do not support denormals despite this flag. Using
+// fp32 denormals also causes instructions to run at the double
+// precision rate for the device.
+def FeatureFP32Denormals : SubtargetFeature<"fp32-denormals",
+  "FP32Denormals",
+  "true",
+  "Enable single precision denormal handling"
+>;
+
 class SubtargetFeatureLocalMemorySize <int Value> : SubtargetFeature<
   "localmemorysize"#Value,
   "LocalMemorySize",