[llvm-mca] Print the "Block RThroughput" in the SummaryView.

This patch implements the "block reciprocal throughput" computation in the
SummaryView.

The block reciprocal throughput is computed as the MAX of:
  - NumMicroOps / DispatchWidth
  - Resource Cycles / #Units   (for every resource consumed).

The block throughput is bounded from above by the hardware dispatch throughput.
That is because the DispatchWidth is an upper bound on how many opcodes can be part
of a single dispatch group.

The block throughput is also limited by the amount of hardware parallelism. The
number of available resource units affects how the resource pressure is
distributed, and also how many blocks can be delivered every cycle.

llvm-svn: 333095

1 files changed, 6 insertions, 5 deletions

diff --git a/llvm/test/tools/llvm-mca/X86/BtVer2/vec-logic-read-after-ld-1.s b/llvm/test/tools/llvm-mca/X86/BtVer2/vec-logic-read-after-ld-1.s
index 7cbc0a8c7ff..acdcdebe261 100644
--- a/llvm/test/tools/llvm-mca/X86/BtVer2/vec-logic-read-after-ld-1.s
+++ b/llvm/test/tools/llvm-mca/X86/BtVer2/vec-logic-read-after-ld-1.s
@@ -4,11 +4,12 @@
 vaddps %xmm0, %xmm0, %xmm1
 vandps (%rdi), %xmm1, %xmm2
 
-# CHECK:      Iterations:     1
-# CHECK-NEXT: Instructions:   2
-# CHECK-NEXT: Total Cycles:   9
-# CHECK-NEXT: Dispatch Width: 2
-# CHECK-NEXT: IPC:            0.22
+# CHECK:      Iterations:        1
+# CHECK-NEXT: Instructions:      2
+# CHECK-NEXT: Total Cycles:      9
+# CHECK-NEXT: Dispatch Width:    2
+# CHECK-NEXT: IPC:               0.22
+# CHECK-NEXT: Block RThroughput: 1.0
 
 # CHECK:      Instruction Info:
 # CHECK-NEXT: [1]: #uOps