diff options
| author | Carlos Eduardo Seo <cseo@linux.vnet.ibm.com> | 2019-02-08 16:18:12 -0200 |
|---|---|---|
| committer | Carlos Eduardo Seo <cseo@linux.vnet.ibm.com> | 2019-03-20 20:27:00 +0000 |
| commit | 3023d7da49cad1a6fae4684d1b9313c51a4085d4 (patch) | |
| tree | 5360e2a9e92ca4c2b632054b986bd5562c8bfc24 /test/codegen/mathbits.go | |
| parent | 23b476a3c8b5d915c4b50957a0f3935a40261eac (diff) | |
| download | go-3023d7da49cad1a6fae4684d1b9313c51a4085d4.tar.xz | |
cmd/compile/internal, cmd/internal/obj/ppc64: generate new count trailing zeros instructions on POWER9
This change adds new POWER9 instructions for counting trailing zeros (CNTTZW/CNTTZD)
to the assembler and generates them in SSA when GOPPC64=power9.
name old time/op new time/op delta
TrailingZeros-160 1.59ns ±20% 1.45ns ±10% -8.81% (p=0.000 n=14+13)
TrailingZeros8-160 1.55ns ±23% 1.62ns ±44% ~ (p=0.593 n=13+15)
TrailingZeros16-160 1.78ns ±23% 1.62ns ±38% -9.31% (p=0.003 n=14+14)
TrailingZeros32-160 1.64ns ±10% 1.49ns ± 9% -9.15% (p=0.000 n=13+14)
TrailingZeros64-160 1.53ns ± 6% 1.45ns ± 5% -5.38% (p=0.000 n=15+13)
Change-Id: I365e6ff79f3ce4d8ebe089a6a86b1771853eb596
Reviewed-on: https://go-review.googlesource.com/c/go/+/167517
Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
Diffstat (limited to 'test/codegen/mathbits.go')
| -rw-r--r-- | test/codegen/mathbits.go | 24 |
1 files changed, 16 insertions, 8 deletions
diff --git a/test/codegen/mathbits.go b/test/codegen/mathbits.go index 9a3b00cab7..5c541bfd29 100644 --- a/test/codegen/mathbits.go +++ b/test/codegen/mathbits.go @@ -261,8 +261,10 @@ func TrailingZeros(n uint) int { // arm:"CLZ" // arm64:"RBIT","CLZ" // s390x:"FLOGR" - // ppc64:"ANDN","POPCNTD" - // ppc64le:"ANDN","POPCNTD" + // ppc64/power8:"ANDN","POPCNTD" + // ppc64le/power8:"ANDN","POPCNTD" + // ppc64/power9: "CNTTZD" + // ppc64le/power9: "CNTTZD" // wasm:"I64Ctz" return bits.TrailingZeros(n) } @@ -271,8 +273,10 @@ func TrailingZeros64(n uint64) int { // amd64:"BSFQ","MOVL\t\\$64","CMOVQEQ" // arm64:"RBIT","CLZ" // s390x:"FLOGR" - // ppc64:"ANDN","POPCNTD" - // ppc64le:"ANDN","POPCNTD" + // ppc64/power8:"ANDN","POPCNTD" + // ppc64le/power8:"ANDN","POPCNTD" + // ppc64/power9: "CNTTZD" + // ppc64le/power9: "CNTTZD" // wasm:"I64Ctz" return bits.TrailingZeros64(n) } @@ -282,8 +286,10 @@ func TrailingZeros32(n uint32) int { // arm:"CLZ" // arm64:"RBITW","CLZW" // s390x:"FLOGR","MOVWZ" - // ppc64:"ANDN","POPCNTW" - // ppc64le:"ANDN","POPCNTW" + // ppc64/power8:"ANDN","POPCNTW" + // ppc64le/power8:"ANDN","POPCNTW" + // ppc64/power9: "CNTTZW" + // ppc64le/power9: "CNTTZW" // wasm:"I64Ctz" return bits.TrailingZeros32(n) } @@ -293,8 +299,10 @@ func TrailingZeros16(n uint16) int { // arm:"ORR\t\\$65536","CLZ",-"MOVHU\tR" // arm64:"ORR\t\\$65536","RBITW","CLZW",-"MOVHU\tR",-"RBIT\t",-"CLZ\t" // s390x:"FLOGR","OR\t\\$65536" - // ppc64:"POPCNTD","OR\\t\\$65536" - // ppc64le:"POPCNTD","OR\\t\\$65536" + // ppc64/power8:"POPCNTD","OR\\t\\$65536" + // ppc64le/power8:"POPCNTD","OR\\t\\$65536" + // ppc64/power9:"CNTTZD","OR\\t\\$65536" + // ppc64le/power9:"CNTTZD","OR\\t\\$65536" // wasm:"I64Ctz" return bits.TrailingZeros16(n) } |