[llvm] 9c596bc - [X86] AMD Zen 3: same-reg SSE XMM XORPD is a 1-cycle(!) dep-breaking zero-idiom

[Roman Lebedev via llvm-commits]

Author: Roman Lebedev
Date: 2021-05-14T11:56:07+03:00
New Revision: 9c596bc5416a16247c4dfb4a564d93c6cb97fb9f

URL: https://github.com/llvm/llvm-project/commit/9c596bc5416a16247c4dfb4a564d93c6cb97fb9f
DIFF: https://github.com/llvm/llvm-project/commit/9c596bc5416a16247c4dfb4a564d93c6cb97fb9f.diff

LOG: [X86] AMD Zen 3: same-reg SSE XMM XORPD is a 1-cycle(!) dep-breaking zero-idiom

Same as with it's float friend, unlike their AVX versions.
As confirmed by exegesis, 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 673439459a84..6d2a88c2209e 100644
--- a/llvm/lib/Target/X86/X86ScheduleZnver3.td
+++ b/llvm/lib/Target/X86/X86ScheduleZnver3.td
@@ -1534,7 +1534,7 @@ def Zn3WriteFZeroIdiom : SchedWriteVariant<[
     SchedVar<MCSchedPredicate<ZeroIdiomPredicate>, [Zn3WriteZeroLatency]>,
     SchedVar<NoSchedPred,                          [WriteFLogic]>
 ]>;
-// NOTE: XORPSrr is not zero-cycle!
+// NOTE: XORPSrr, XORPDrr are not zero-cycle!
 def : InstRW<[Zn3WriteFZeroIdiom], (instrs VXORPSrr)>;
 
 def Zn3WriteFZeroIdiomY : SchedWriteVariant<[
@@ -1551,7 +1551,7 @@ def : IsZeroIdiomFunction<[
                      SUB64rr, SUB64rr_REV ], ZeroIdiomPredicate>,
 
   // SSE XMM Zero-idioms.
-  DepBreakingClass<[ XORPSrr ], ZeroIdiomPredicate>,
+  DepBreakingClass<[ XORPSrr, XORPDrr ], ZeroIdiomPredicate>,
 
   // AVX XMM Zero-idioms.
   DepBreakingClass<[ VXORPSrr ], 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 3cc35353974b..12faa2f5baca 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
@@ -106,12 +106,12 @@ xorpd %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:
@@ -127,13 +127,13 @@ xorpd %xmm0, %xmm1
 # CHECK-NEXT:  1      1     0.25                        xorpd	%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
@@ -171,16 +171,16 @@ xorpd %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    -      -      -      -      -      -      -      -      -      -      -     xorpd	%xmm1, %xmm1
-# CHECK-NEXT:  -      -      -      -      -      -      -      -     0.50    -     0.50    -      -      -      -      -      -      -      -      -      -      -      -     xorpd	%xmm0, %xmm1
+# CHECK-NEXT:  -      -      -      -      -      -      -      -      -     0.50   0.25   0.25    -      -      -      -      -      -      -      -      -      -      -     xorpd	%xmm1, %xmm1
+# CHECK-NEXT:  -      -      -      -      -      -      -      -     0.50    -     0.25   0.25    -      -      -      -      -      -      -      -      -      -      -     xorpd	%xmm0, %xmm1
 
 # CHECK:      Timeline view:
-# CHECK-NEXT: Index     0123456
+# CHECK-NEXT: Index     01234
 
-# CHECK:      [0,0]     DeER ..   xorpd	%xmm1, %xmm1
-# CHECK-NEXT: [0,1]     D=eER..   xorpd	%xmm0, %xmm1
-# CHECK-NEXT: [1,0]     D==eER.   xorpd	%xmm1, %xmm1
-# CHECK-NEXT: [1,1]     D===eER   xorpd	%xmm0, %xmm1
+# CHECK:      [0,0]     DeER.   xorpd	%xmm1, %xmm1
+# CHECK-NEXT: [0,1]     D=eER   xorpd	%xmm0, %xmm1
+# CHECK-NEXT: [1,0]     DeE-R   xorpd	%xmm1, %xmm1
+# CHECK-NEXT: [1,1]     D=eER   xorpd	%xmm0, %xmm1
 
 # CHECK:      Average Wait times (based on the timeline view):
 # CHECK-NEXT: [0]: Executions
@@ -189,6 +189,6 @@ xorpd %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       xorpd	%xmm1, %xmm1
-# CHECK-NEXT: 1.     2     3.0    0.0    0.0       xorpd	%xmm0, %xmm1
-# CHECK-NEXT:        2     2.5    0.3    0.0       <total>
+# CHECK-NEXT: 0.     2     1.0    1.0    0.5       xorpd	%xmm1, %xmm1
+# CHECK-NEXT: 1.     2     2.0    0.0    0.0       xorpd	%xmm0, %xmm1
+# CHECK-NEXT:        2     1.5    0.5    0.3       <total>