[llvm] r348286 - AMDGPU: Add f32 vectors to SGPR register classes
Matt Arsenault via llvm-commits
llvm-commits at lists.llvm.org
Tue Dec 4 09:51:36 PST 2018
Author: arsenm
Date: Tue Dec 4 09:51:36 2018
New Revision: 348286
URL: http://llvm.org/viewvc/llvm-project?rev=348286&view=rev
Log:
AMDGPU: Add f32 vectors to SGPR register classes
Modified:
llvm/trunk/lib/Target/AMDGPU/SIRegisterInfo.td
Modified: llvm/trunk/lib/Target/AMDGPU/SIRegisterInfo.td
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/SIRegisterInfo.td?rev=348286&r1=348285&r2=348286&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AMDGPU/SIRegisterInfo.td (original)
+++ llvm/trunk/lib/Target/AMDGPU/SIRegisterInfo.td Tue Dec 4 09:51:36 2018
@@ -435,7 +435,7 @@ def SReg_32 : RegisterClass<"AMDGPU", [i
let AllocationPriority = 7;
}
-def SGPR_64 : RegisterClass<"AMDGPU", [v2i32, i64, f64, v4i16, v4f16], 32, (add SGPR_64Regs)> {
+def SGPR_64 : RegisterClass<"AMDGPU", [v2i32, i64, v2f32, f64, v4i16, v4f16], 32, (add SGPR_64Regs)> {
let CopyCost = 1;
let AllocationPriority = 8;
}
@@ -444,13 +444,13 @@ def TTMP_64 : RegisterClass<"AMDGPU", [v
let isAllocatable = 0;
}
-def SReg_64_XEXEC : RegisterClass<"AMDGPU", [v2i32, i64, f64, i1, v4i16, v4f16], 32,
+def SReg_64_XEXEC : RegisterClass<"AMDGPU", [v2i32, i64, v2f32, f64, i1, v4i16, v4f16], 32,
(add SGPR_64, VCC, FLAT_SCR, XNACK_MASK, TTMP_64, TBA, TMA)> {
let CopyCost = 1;
let AllocationPriority = 8;
}
-def SReg_64 : RegisterClass<"AMDGPU", [v2i32, i64, f64, i1, v4i16, v4f16], 32,
+def SReg_64 : RegisterClass<"AMDGPU", [v2i32, i64, v2f32, f64, i1, v4i16, v4f16], 32,
(add SReg_64_XEXEC, EXEC)> {
let CopyCost = 1;
let AllocationPriority = 8;
@@ -459,15 +459,15 @@ def SReg_64 : RegisterClass<"AMDGPU", [v
// Requires 2 s_mov_b64 to copy
let CopyCost = 2 in {
-def SGPR_128 : RegisterClass<"AMDGPU", [v4i32, v16i8, v2i64], 32, (add SGPR_128Regs)> {
+def SGPR_128 : RegisterClass<"AMDGPU", [v4i32, v4f32, v16i8, v2i64], 32, (add SGPR_128Regs)> {
let AllocationPriority = 10;
}
-def TTMP_128 : RegisterClass<"AMDGPU", [v4i32, v16i8, v2i64], 32, (add TTMP_128Regs)> {
+def TTMP_128 : RegisterClass<"AMDGPU", [v4i32, v4f32, v16i8, v2i64], 32, (add TTMP_128Regs)> {
let isAllocatable = 0;
}
-def SReg_128 : RegisterClass<"AMDGPU", [v4i32, v16i8, v2i64, v2f64], 32,
+def SReg_128 : RegisterClass<"AMDGPU", [v4i32, v4f32, v16i8, v2i64, v2f64], 32,
(add SGPR_128, TTMP_128)> {
let AllocationPriority = 10;
}
More information about the llvm-commits
mailing list