Relax the clearance calculating for breaking partial register dependency.

Summary: LLVM assumes that large clearance will hide the partial register spill penalty. But in our experiment, 16 clearance is too small. As the inserted XOR is normally fairly cheap, we should have a higher clearance threshold to aggressively insert XORs that is necessary to break partial register dependency.

Reviewers: wmi, davidxl, stoklund, zansari, myatsina, RKSimon, DavidKreitzer, mkuper, joerg, spatel

Subscribers: davidxl, llvm-commits

Differential Revision: http://reviews.llvm.org/D21560

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274068 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/X86/X86InstrInfo.cpp b/lib/Target/X86/X86InstrInfo.cpp
index 4a0a57a..6574108 100644 (file)
--- a/lib/Target/X86/X86InstrInfo.cpp
+++ b/lib/Target/X86/X86InstrInfo.cpp
@@ -58,6 +58,17 @@ static cl::opt<bool>
 ReMatPICStubLoad("remat-pic-stub-load",
                  cl::desc("Re-materialize load from stub in PIC mode"),
                  cl::init(false), cl::Hidden);
+static cl::opt<unsigned>
+PartialRegUpdateClearance("partial-reg-update-clearance",
+                          cl::desc("Clearance between two register writes "
+                                   "for inserting XOR to avoid partial "
+                                   "register update"),
+                          cl::init(64), cl::Hidden);
+static cl::opt<unsigned>
+UndefRegClearance("undef-reg-clearance",
+                  cl::desc("How many idle instructions we would like before "
+                           "certain undef register reads"),
+                  cl::init(64), cl::Hidden);
 
 enum {
   // Select which memory operand is being unfolded.
@@ -5972,10 +5983,10 @@ getPartialRegUpdateClearance(const MachineInstr *MI, unsigned OpNum,
       return 0;
   }
 
-  // If any of the preceding 16 instructions are reading Reg, insert a
-  // dependency breaking instruction.  The magic number is based on a few
-  // Nehalem experiments.
-  return 16;
+  // If any instructions in the clearance range are reading Reg, insert a
+  // dependency breaking instruction, which is inexpensive and is likely to
+  // be hidden in other instruction's cycles.
+  return PartialRegUpdateClearance;
 }
 
 // Return true for any instruction the copies the high bits of the first source
@@ -6060,8 +6071,7 @@ getUndefRegClearance(const MachineInstr *MI, unsigned &OpNum,
 
   const MachineOperand &MO = MI->getOperand(OpNum);
   if (MO.isUndef() && TargetRegisterInfo::isPhysicalRegister(MO.getReg())) {
-    // Use the same magic number as getPartialRegUpdateClearance.
-    return 16;
+    return UndefRegClearance;
   }
   return 0;
 }

diff --git a/test/CodeGen/X86/vec_int_to_fp.ll b/test/CodeGen/X86/vec_int_to_fp.ll
index 1540b27..6cb4b18 100644 (file)
--- a/test/CodeGen/X86/vec_int_to_fp.ll
+++ b/test/CodeGen/X86/vec_int_to_fp.ll
@@ -1580,6 +1580,7 @@ define <4 x float> @uitofp_4i64_to_4f32(<4 x i64> %a) {
 ; AVX1-NEXT:  .LBB45_10:
 ; AVX1-NEXT:    shrq %rax
 ; AVX1-NEXT:    orq %rax, %rcx
+; AVX1-NEXT:    vxorps %xmm0, %xmm0, %xmm0
 ; AVX1-NEXT:    vcvtsi2ssq %rcx, %xmm0, %xmm0
 ; AVX1-NEXT:    vaddss %xmm0, %xmm0, %xmm0
 ; AVX1-NEXT:    vinsertps {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[0]
@@ -1647,6 +1648,7 @@ define <4 x float> @uitofp_4i64_to_4f32(<4 x i64> %a) {
 ; AVX2-NEXT:  .LBB45_10:
 ; AVX2-NEXT:    shrq %rax
 ; AVX2-NEXT:    orq %rax, %rcx
+; AVX2-NEXT:    vxorps %xmm0, %xmm0, %xmm0
 ; AVX2-NEXT:    vcvtsi2ssq %rcx, %xmm0, %xmm0
 ; AVX2-NEXT:    vaddss %xmm0, %xmm0, %xmm0
 ; AVX2-NEXT:    vinsertps {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[0]
@@ -2773,6 +2775,7 @@ define <4 x float> @uitofp_load_4i64_to_4f32(<4 x i64> *%a) {
 ; AVX1-NEXT:  .LBB74_10:
 ; AVX1-NEXT:    shrq %rax
 ; AVX1-NEXT:    orq %rax, %rcx
+; AVX1-NEXT:    vxorps %xmm0, %xmm0, %xmm0
 ; AVX1-NEXT:    vcvtsi2ssq %rcx, %xmm0, %xmm0
 ; AVX1-NEXT:    vaddss %xmm0, %xmm0, %xmm0
 ; AVX1-NEXT:    vinsertps {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[0]
@@ -2841,6 +2844,7 @@ define <4 x float> @uitofp_load_4i64_to_4f32(<4 x i64> *%a) {
 ; AVX2-NEXT:  .LBB74_10:
 ; AVX2-NEXT:    shrq %rax
 ; AVX2-NEXT:    orq %rax, %rcx
+; AVX2-NEXT:    vxorps %xmm0, %xmm0, %xmm0
 ; AVX2-NEXT:    vcvtsi2ssq %rcx, %xmm0, %xmm0
 ; AVX2-NEXT:    vaddss %xmm0, %xmm0, %xmm0
 ; AVX2-NEXT:    vinsertps {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[0]
@@ -2993,6 +2997,7 @@ define <8 x float> @uitofp_load_8i64_to_8f32(<8 x i64> *%a) {
 ; SSE-NEXT:  .LBB78_10:
 ; SSE-NEXT:    shrq %rax
 ; SSE-NEXT:    orq %rax, %rcx
+; SSE-NEXT:    xorps %xmm5, %xmm5
 ; SSE-NEXT:    cvtsi2ssq %rcx, %xmm5
 ; SSE-NEXT:    addss %xmm5, %xmm5
 ; SSE-NEXT:  .LBB78_12:
@@ -3016,11 +3021,13 @@ define <8 x float> @uitofp_load_8i64_to_8f32(<8 x i64> *%a) {
 ; SSE-NEXT:    testq %rax, %rax
 ; SSE-NEXT:    js .LBB78_16
 ; SSE-NEXT:  # BB#17:
+; SSE-NEXT:    xorps %xmm1, %xmm1
 ; SSE-NEXT:    cvtsi2ssq %rax, %xmm1
 ; SSE-NEXT:    jmp .LBB78_18
 ; SSE-NEXT:  .LBB78_16:
 ; SSE-NEXT:    shrq %rax
 ; SSE-NEXT:    orq %rax, %rcx
+; SSE-NEXT:    xorps %xmm1, %xmm1
 ; SSE-NEXT:    cvtsi2ssq %rcx, %xmm1
 ; SSE-NEXT:    addss %xmm1, %xmm1
 ; SSE-NEXT:  .LBB78_18:
@@ -3165,11 +3172,13 @@ define <8 x float> @uitofp_load_8i64_to_8f32(<8 x i64> *%a) {
 ; AVX1-NEXT:    testq %rax, %rax
 ; AVX1-NEXT:    js .LBB78_19
 ; AVX1-NEXT:  # BB#20:
+; AVX1-NEXT:    vxorps %xmm0, %xmm0, %xmm0
 ; AVX1-NEXT:    vcvtsi2ssq %rax, %xmm0, %xmm5
 ; AVX1-NEXT:    jmp .LBB78_21
 ; AVX1-NEXT:  .LBB78_19:
 ; AVX1-NEXT:    shrq %rax
 ; AVX1-NEXT:    orq %rax, %rcx
+; AVX1-NEXT:    vxorps %xmm0, %xmm0, %xmm0
 ; AVX1-NEXT:    vcvtsi2ssq %rcx, %xmm0, %xmm0
 ; AVX1-NEXT:    vaddss %xmm0, %xmm0, %xmm5
 ; AVX1-NEXT:  .LBB78_21:
@@ -3292,11 +3301,13 @@ define <8 x float> @uitofp_load_8i64_to_8f32(<8 x i64> *%a) {
 ; AVX2-NEXT:    testq %rax, %rax
 ; AVX2-NEXT:    js .LBB78_19
 ; AVX2-NEXT:  # BB#20:
+; AVX2-NEXT:    vxorps %xmm0, %xmm0, %xmm0
 ; AVX2-NEXT:    vcvtsi2ssq %rax, %xmm0, %xmm5
 ; AVX2-NEXT:    jmp .LBB78_21
 ; AVX2-NEXT:  .LBB78_19:
 ; AVX2-NEXT:    shrq %rax
 ; AVX2-NEXT:    orq %rax, %rcx
+; AVX2-NEXT:    vxorps %xmm0, %xmm0, %xmm0
 ; AVX2-NEXT:    vcvtsi2ssq %rcx, %xmm0, %xmm0
 ; AVX2-NEXT:    vaddss %xmm0, %xmm0, %xmm5
 ; AVX2-NEXT:  .LBB78_21: