[X86] AMD Zen 3: same-reg AVX YMM VANDNPD is a zero-cycle(!) dep-breaking zero-idiom

As confirmed by exegesis measurements, and ref docs.

diff --git a/llvm/lib/Target/X86/X86ScheduleZnver3.td b/llvm/lib/Target/X86/X86ScheduleZnver3.td
index 5ab351a..72c7d41 100644 (file)
--- a/llvm/lib/Target/X86/X86ScheduleZnver3.td
+++ b/llvm/lib/Target/X86/X86ScheduleZnver3.td
@@ -1543,7 +1543,7 @@ def Zn3WriteFZeroIdiomY : SchedWriteVariant<[
     SchedVar<NoSchedPred,                          [WriteFLogicY]>
 ]>;
 def : InstRW<[Zn3WriteFZeroIdiomY], (instrs VXORPSYrr, VXORPDYrr,
-                                            VANDNPSYrr)>;
+                                            VANDNPSYrr, VANDNPDYrr)>;
 
 def : IsZeroIdiomFunction<[
   // GPR Zero-idioms.
@@ -1562,7 +1562,7 @@ def : IsZeroIdiomFunction<[
 
   // AVX YMM Zero-idioms.
   DepBreakingClass<[ VXORPSYrr, VXORPDYrr,
-                     VANDNPSYrr ], ZeroIdiomPredicate>,
+                     VANDNPSYrr, VANDNPDYrr ], ZeroIdiomPredicate>,
 ]>;
 
 def : IsDepBreakingFunction<[

diff --git a/llvm/test/tools/llvm-mca/X86/Znver3/zero-idioms-avx-ymm.s b/llvm/test/tools/llvm-mca/X86/Znver3/zero-idioms-avx-ymm.s
index a427a48..fd60e3e 100644 (file)
--- a/llvm/test/tools/llvm-mca/X86/Znver3/zero-idioms-avx-ymm.s
+++ b/llvm/test/tools/llvm-mca/X86/Znver3/zero-idioms-avx-ymm.s
@@ -298,13 +298,13 @@ vandnpd %ymm1, %ymm0, %ymm0
 
 # CHECK:      Iterations:        10000
 # CHECK-NEXT: Instructions:      20000
-# CHECK-NEXT: Total Cycles:      20003
+# CHECK-NEXT: Total Cycles:      3337
 # CHECK-NEXT: Total uOps:        20000
 
 # CHECK:      Dispatch Width:    6
-# CHECK-NEXT: uOps Per Cycle:    1.00
-# CHECK-NEXT: IPC:               1.00
-# CHECK-NEXT: Block RThroughput: 0.5
+# CHECK-NEXT: uOps Per Cycle:    5.99
+# CHECK-NEXT: IPC:               5.99
+# CHECK-NEXT: Block RThroughput: 0.3
 
 # CHECK:      Instruction Info:
 # CHECK-NEXT: [1]: #uOps
@@ -315,17 +315,17 @@ vandnpd %ymm1, %ymm0, %ymm0
 # CHECK-NEXT: [6]: HasSideEffects (U)
 
 # CHECK:      [1]    [2]    [3]    [4]    [5]    [6]    Instructions:
-# CHECK-NEXT:  1      1     0.25                        vandnpd        %ymm0, %ymm0, %ymm0
+# CHECK-NEXT:  1      0     0.17                        vandnpd        %ymm0, %ymm0, %ymm0
 # CHECK-NEXT:  1      1     0.25                        vandnpd        %ymm1, %ymm0, %ymm0
 
 # 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:         9
 
 # 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:      9
 
 # CHECK:      *  Register File #2 -- Zn3IntegerPRF:
 # CHECK-NEXT:    Number of physical registers:     192
@@ -359,20 +359,20 @@ vandnpd %ymm1, %ymm0, %ymm0
 
 # CHECK:      Resource pressure per iteration:
 # 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]
-# CHECK-NEXT:  -      -      -      -      -      -      -      -     0.50   0.50   0.50   0.50    -      -      -      -      -      -      -      -      -      -      -
+# CHECK-NEXT:  -      -      -      -      -      -      -      -     0.25   0.25   0.25   0.25    -      -      -      -      -      -      -      -      -      -      -
 
 # 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    -      -      -      -      -      -      -      -      -      -      -     vandnpd %ymm0, %ymm0, %ymm0
-# CHECK-NEXT:  -      -      -      -      -      -      -      -     0.50    -     0.50    -      -      -      -      -      -      -      -      -      -      -      -     vandnpd %ymm1, %ymm0, %ymm0
+# CHECK-NEXT:  -      -      -      -      -      -      -      -      -      -      -      -      -      -      -      -      -      -      -      -      -      -      -     vandnpd %ymm0, %ymm0, %ymm0
+# CHECK-NEXT:  -      -      -      -      -      -      -      -     0.25   0.25   0.25   0.25    -      -      -      -      -      -      -      -      -      -      -     vandnpd %ymm1, %ymm0, %ymm0
 
 # CHECK:      Timeline view:
-# CHECK-NEXT: Index     0123456
+# CHECK-NEXT: Index     0123
 
-# CHECK:      [0,0]     DeER ..   vandnpd      %ymm0, %ymm0, %ymm0
-# CHECK-NEXT: [0,1]     D=eER..   vandnpd      %ymm1, %ymm0, %ymm0
-# CHECK-NEXT: [1,0]     D==eER.   vandnpd      %ymm0, %ymm0, %ymm0
-# CHECK-NEXT: [1,1]     D===eER   vandnpd      %ymm1, %ymm0, %ymm0
+# CHECK:      [0,0]     DR .   vandnpd %ymm0, %ymm0, %ymm0
+# CHECK-NEXT: [0,1]     DeER   vandnpd %ymm1, %ymm0, %ymm0
+# CHECK-NEXT: [1,0]     D--R   vandnpd %ymm0, %ymm0, %ymm0
+# CHECK-NEXT: [1,1]     DeER   vandnpd %ymm1, %ymm0, %ymm0
 
 # CHECK:      Average Wait times (based on the timeline view):
 # CHECK-NEXT: [0]: Executions
@@ -381,6 +381,6 @@ vandnpd %ymm1, %ymm0, %ymm0
 # 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       vandnpd     %ymm0, %ymm0, %ymm0
-# CHECK-NEXT: 1.     2     3.0    0.0    0.0       vandnpd     %ymm1, %ymm0, %ymm0
-# CHECK-NEXT:        2     2.5    0.3    0.0       <total>
+# CHECK-NEXT: 0.     2     0.0    0.0    1.0       vandnpd     %ymm0, %ymm0, %ymm0
+# CHECK-NEXT: 1.     2     1.0    1.0    0.0       vandnpd     %ymm1, %ymm0, %ymm0
+# CHECK-NEXT:        2     0.5    0.5    0.5       <total>