[llvm] f38dcbe - [X86] AMD Zen 3: same-reg SSE XMM ANDNPS is a 1-cycle(!) dep-breaking zero-idiom
Roman Lebedev via llvm-commits
llvm-commits at lists.llvm.org
Fri May 14 04:06:56 PDT 2021
Author: Roman Lebedev
Date: 2021-05-14T14:06:23+03:00
New Revision: f38dcbecb643e30931b56bbcf37254477eac3977
URL: https://github.com/llvm/llvm-project/commit/f38dcbecb643e30931b56bbcf37254477eac3977
DIFF: https://github.com/llvm/llvm-project/commit/f38dcbecb643e30931b56bbcf37254477eac3977.diff
LOG: [X86] AMD Zen 3: same-reg SSE XMM ANDNPS is a 1-cycle(!) dep-breaking zero-idiom
Same as SSE XMM XORPS/XORPD, it is not zero-cycle, even though it breaks the deps.
As confirmed by the exegesis measurements, and ref docs.
Added:
Modified:
llvm/lib/Target/X86/X86ScheduleZnver3.td
llvm/test/tools/llvm-mca/X86/Znver3/zero-idioms-sse-xmm.s
Removed:
################################################################################
diff --git a/llvm/lib/Target/X86/X86ScheduleZnver3.td b/llvm/lib/Target/X86/X86ScheduleZnver3.td
index 0116163ec354..96234e067b1f 100644
--- a/llvm/lib/Target/X86/X86ScheduleZnver3.td
+++ b/llvm/lib/Target/X86/X86ScheduleZnver3.td
@@ -1551,7 +1551,8 @@ def : IsZeroIdiomFunction<[
SUB64rr, SUB64rr_REV ], ZeroIdiomPredicate>,
// SSE XMM Zero-idioms.
- DepBreakingClass<[ XORPSrr, XORPDrr ], ZeroIdiomPredicate>,
+ DepBreakingClass<[ XORPSrr, XORPDrr,
+ ANDNPSrr ], ZeroIdiomPredicate>,
// AVX XMM Zero-idioms.
DepBreakingClass<[ VXORPSrr, VXORPDrr ], ZeroIdiomPredicate>,
diff --git a/llvm/test/tools/llvm-mca/X86/Znver3/zero-idioms-sse-xmm.s b/llvm/test/tools/llvm-mca/X86/Znver3/zero-idioms-sse-xmm.s
index 835e6c2bfac0..cf20c9224243 100644
--- a/llvm/test/tools/llvm-mca/X86/Znver3/zero-idioms-sse-xmm.s
+++ b/llvm/test/tools/llvm-mca/X86/Znver3/zero-idioms-sse-xmm.s
@@ -202,12 +202,12 @@ andnps %xmm0, %xmm1
# CHECK: Iterations: 10000
# CHECK-NEXT: Instructions: 20000
-# CHECK-NEXT: Total Cycles: 20003
+# CHECK-NEXT: Total Cycles: 5004
# CHECK-NEXT: Total uOps: 20000
# CHECK: Dispatch Width: 6
-# CHECK-NEXT: uOps Per Cycle: 1.00
-# CHECK-NEXT: IPC: 1.00
+# CHECK-NEXT: uOps Per Cycle: 4.00
+# CHECK-NEXT: IPC: 4.00
# CHECK-NEXT: Block RThroughput: 0.5
# CHECK: Instruction Info:
@@ -223,13 +223,13 @@ andnps %xmm0, %xmm1
# CHECK-NEXT: 1 1 0.25 andnps %xmm0, %xmm1
# CHECK: Register File statistics:
-# CHECK-NEXT: Total number of mappings created: 20000
-# CHECK-NEXT: Max number of mappings used: 66
+# CHECK-NEXT: Total number of mappings created: 10000
+# CHECK-NEXT: Max number of mappings used: 37
# CHECK: * Register File #1 -- Zn3FpPRF:
# CHECK-NEXT: Number of physical registers: 160
-# CHECK-NEXT: Total number of mappings created: 20000
-# CHECK-NEXT: Max number of mappings used: 66
+# CHECK-NEXT: Total number of mappings created: 10000
+# CHECK-NEXT: Max number of mappings used: 37
# CHECK: * Register File #2 -- Zn3IntegerPRF:
# CHECK-NEXT: Number of physical registers: 192
@@ -267,16 +267,16 @@ andnps %xmm0, %xmm1
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12.0] [12.1] [13] [14.0] [14.1] [14.2] [15.0] [15.1] [15.2] [16.0] [16.1] Instructions:
-# CHECK-NEXT: - - - - - - - - - 0.50 - 0.50 - - - - - - - - - - - andnps %xmm1, %xmm1
-# CHECK-NEXT: - - - - - - - - 0.50 - 0.50 - - - - - - - - - - - - andnps %xmm0, %xmm1
+# CHECK-NEXT: - - - - - - - - - 0.50 0.25 0.25 - - - - - - - - - - - andnps %xmm1, %xmm1
+# CHECK-NEXT: - - - - - - - - 0.50 - 0.25 0.25 - - - - - - - - - - - andnps %xmm0, %xmm1
# CHECK: Timeline view:
-# CHECK-NEXT: Index 0123456
+# CHECK-NEXT: Index 01234
-# CHECK: [0,0] DeER .. andnps %xmm1, %xmm1
-# CHECK-NEXT: [0,1] D=eER.. andnps %xmm0, %xmm1
-# CHECK-NEXT: [1,0] D==eER. andnps %xmm1, %xmm1
-# CHECK-NEXT: [1,1] D===eER andnps %xmm0, %xmm1
+# CHECK: [0,0] DeER. andnps %xmm1, %xmm1
+# CHECK-NEXT: [0,1] D=eER andnps %xmm0, %xmm1
+# CHECK-NEXT: [1,0] DeE-R andnps %xmm1, %xmm1
+# CHECK-NEXT: [1,1] D=eER andnps %xmm0, %xmm1
# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
@@ -285,6 +285,6 @@ andnps %xmm0, %xmm1
# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
# CHECK: [0] [1] [2] [3]
-# CHECK-NEXT: 0. 2 2.0 0.5 0.0 andnps %xmm1, %xmm1
-# CHECK-NEXT: 1. 2 3.0 0.0 0.0 andnps %xmm0, %xmm1
-# CHECK-NEXT: 2 2.5 0.3 0.0 <total>
+# CHECK-NEXT: 0. 2 1.0 1.0 0.5 andnps %xmm1, %xmm1
+# CHECK-NEXT: 1. 2 2.0 0.0 0.0 andnps %xmm0, %xmm1
+# CHECK-NEXT: 2 1.5 0.5 0.3 <total>
More information about the llvm-commits
mailing list