diff options
| author | Lynn Boger <laboger@linux.vnet.ibm.com> | 2020-04-01 10:30:05 -0400 |
|---|---|---|
| committer | Lynn Boger <laboger@linux.vnet.ibm.com> | 2020-04-09 19:18:56 +0000 |
| commit | a1550d3ca3a6a90b8bbb610950d1b30649411243 (patch) | |
| tree | ab0fd924a51e78b8615baf584d63d7336fd101bd /test/codegen | |
| parent | 7c0ee1127bf41bf274b08170de3e42b171a903c0 (diff) | |
| download | go-a1550d3ca3a6a90b8bbb610950d1b30649411243.tar.gz go-a1550d3ca3a6a90b8bbb610950d1b30649411243.zip | |
cmd/compile: use isel with variable shifts on ppc64x
This changes the code generated for variable length shift
counts to use isel instead of instructions that set and
read the carry flag.
This reduces the generated code for shifts like this
by 1 instruction and avoids the use of instructions to
set and read the carry flag.
This sequence can be found in strconv with these results
on power9:
Atof64Decimal 71.6ns ± 0% 68.3ns ± 0% -4.61%
Atof64Float 95.3ns ± 0% 90.9ns ± 0% -4.62%
Atof64FloatExp 153ns ± 0% 149ns ± 0% -2.61%
Atof64Big 234ns ± 0% 232ns ± 0% -0.85%
Atof64RandomBits 348ns ± 0% 369ns ± 0% +6.03%
Atof64RandomFloats 262ns ± 0% 262ns ± 0% ~
Atof32Decimal 72.0ns ± 0% 68.2ns ± 0% -5.28%
Atof32Float 92.1ns ± 0% 87.1ns ± 0% -5.43%
Atof32FloatExp 159ns ± 0% 158ns ± 0% -0.63%
Atof32Random 194ns ± 0% 191ns ± 0% -1.55%
Some tests in codegen/shift.go are enabled to verify the
expected instructions are generated.
Change-Id: I968715d10ada405a8c46132bf19b8ed9b85796d1
Reviewed-on: https://go-review.googlesource.com/c/go/+/227337
Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Diffstat (limited to 'test/codegen')
| -rw-r--r-- | test/codegen/shift.go | 24 |
1 files changed, 24 insertions, 0 deletions
diff --git a/test/codegen/shift.go b/test/codegen/shift.go index 305c39a1d8..5e50ea6bff 100644 --- a/test/codegen/shift.go +++ b/test/codegen/shift.go @@ -12,61 +12,85 @@ package codegen func lshMask64x64(v int64, s uint64) int64 { // s390x:-".*AND",-".*MOVDGE" + // ppc64le:"ANDCC",-"ORN",-"ISEL" + // ppc64:"ANDCC",-"ORN",-"ISEL" return v << (s & 63) } func rshMask64Ux64(v uint64, s uint64) uint64 { // s390x:-".*AND",-".*MOVDGE" + // ppc64le:"ANDCC",-"ORN",-"ISEL" + // ppc64:"ANDCC",-"ORN",-"ISEL" return v >> (s & 63) } func rshMask64x64(v int64, s uint64) int64 { // s390x:-".*AND",-".*MOVDGE" + // ppc64le:"ANDCC",-ORN",-"ISEL" + // ppc64:"ANDCC",-"ORN",-"ISEL" return v >> (s & 63) } func lshMask32x64(v int32, s uint64) int32 { // s390x:-".*AND",-".*MOVDGE" + // ppc64le:"ISEL",-"ORN" + // ppc64:"ISEL",-"ORN" return v << (s & 63) } func rshMask32Ux64(v uint32, s uint64) uint32 { // s390x:-".*AND",-".*MOVDGE" + // ppc64le:"ISEL",-"ORN" + // ppc64:"ISEL",-"ORN" return v >> (s & 63) } func rshMask32x64(v int32, s uint64) int32 { // s390x:-".*AND",-".*MOVDGE" + // ppc64le:"ISEL",-"ORN" + // ppc64:"ISEL",-"ORN" return v >> (s & 63) } func lshMask64x32(v int64, s uint32) int64 { // s390x:-".*AND",-".*MOVDGE" + // ppc64le:"ANDCC",-"ORN" + // ppc64:"ANDCC",-"ORN" return v << (s & 63) } func rshMask64Ux32(v uint64, s uint32) uint64 { // s390x:-".*AND",-".*MOVDGE" + // ppc64le:"ANDCC",-"ORN" + // ppc64:"ANDCC",-"ORN" return v >> (s & 63) } func rshMask64x32(v int64, s uint32) int64 { // s390x:-".*AND",-".*MOVDGE" + // ppc64le:"ANDCC",-"ORN",-"ISEL" + // ppc64:"ANDCC",-"ORN",-"ISEL" return v >> (s & 63) } func lshMask64x32Ext(v int64, s int32) int64 { // s390x:-".*AND",-".*MOVDGE" + // ppc64le:"ANDCC",-"ORN",-"ISEL" + // ppc64:"ANDCC",-"ORN",-"ISEL" return v << uint(s&63) } func rshMask64Ux32Ext(v uint64, s int32) uint64 { // s390x:-".*AND",-".*MOVDGE" + // ppc64le:"ANDCC",-"ORN",-"ISEL" + // ppc64:"ANDCC",-"ORN",-"ISEL" return v >> uint(s&63) } func rshMask64x32Ext(v int64, s int32) int64 { // s390x:-".*AND",-".*MOVDGE" + // ppc64le:"ANDCC",-"ORN",-"ISEL" + // ppc64:"ANDCC",-"ORN",-"ISEL" return v >> uint(s&63) } |