i get similar results with 1.18 (inline slower than noinline) but different results with 1.16, 1.17, and 1.19rc2 (inline faster than noinline)
goos: linux goarch: amd64 cpu: AMD Ryzen 5 5600X 6-Core Processor ======== 1.16.15 BenchmarkNoInline-12 125717362 9.607 ns/op BenchmarkInline-12 150066394 8.721 ns/op BenchmarkNoInline-12 125476344 9.710 ns/op BenchmarkInline-12 133781608 8.851 ns/op ======== 1.17.10 BenchmarkNoInline-12 100000000 10.14 ns/op BenchmarkInline-12 135818722 8.646 ns/op BenchmarkNoInline-12 123817206 10.61 ns/op BenchmarkInline-12 137691572 8.754 ns/op ======== 1.18.4 BenchmarkNoInline-12 121646458 10.13 ns/op BenchmarkInline-12 81420973 14.65 ns/op BenchmarkNoInline-12 123927972 10.05 ns/op BenchmarkInline-12 81371038 14.64 ns/op ======== 1.19rc2 BenchmarkNoInline-12 120799062 9.864 ns/op BenchmarkInline-12 147306990 8.579 ns/op BenchmarkNoInline-12 120426837 10.17 ns/op BenchmarkInline-12 129029052 8.621 ns/op On Friday, 22 July 2022 at 18:56:54 UTC-7 Kevin Chowski wrote: > Datapoint: same with windows/amd64 on Intel (running 1.19beta1): > > goos: windows > goarch: amd64 > pkg: common/sandbox > cpu: Intel(R) Core(TM) i7-6650U CPU @ 2.20GHz > BenchmarkNoInline-4 77425848 14.34 ns/op > BenchmarkInline-4 59108932 20.58 ns/op > PASS > ok common/sandbox 2.645s > > Looking at the disassembly, I noticed that in the Inline case there was a > 7-byte `lea 0xXXXXXX(%rip),%rbx` in the tight inner loop due to some > really proactive constant propagation (I hypothesize). If you manually > defeat the propagation by storing the string in a global and manually > copying it into the stack, the inlined becomes faster than NoInline again: > https://go.dev/play/p/VRgJP2y7joS > > goos: windows > goarch: amd64 > pkg: common/sandbox > cpu: Intel(R) Core(TM) i7-6650U CPU @ 2.20GHz > BenchmarkNoInline-4 81436539 14.08 ns/op > BenchmarkInline-4 59255162 21.32 ns/op > BenchmarkInlineDefeatConstProp-4 97524828 12.57 ns/op > PASS > ok common/sandbox 5.111s > > On Friday, July 22, 2022 at 11:01:00 AM UTC-6 mpr...@google.com wrote: > >> I can reproduce similar behavior on linux-amd64: >> >> $ perf stat ./example.com.test -test.bench=BenchmarkInline >> -test.benchtime=100000000x >> goos: linux >> goarch: amd64 >> pkg: example.com >> cpu: Intel(R) Xeon(R) W-2135 CPU @ 3.70GHz >> BenchmarkInline-12 100000000 16.78 ns/op >> >> PASS >> >> Performance counter stats for './example.com.test >> -test.bench=BenchmarkInline -test.benchtime=100000000x': >> >> 1,691.95 msec task-clock:u # 1.004 CPUs >> utilized >> 0 context-switches:u # 0.000 /sec >> >> 0 cpu-migrations:u # 0.000 /sec >> >> 352 page-faults:u # 208.044 /sec >> >> 6,732,752,072 cycles:u # 3.979 GHz >> >> 22,405,823,428 instructions:u # 3.33 insn per >> cycle >> 6,501,294,164 branches:u # 3.842 G/sec >> >> 149,596 branch-misses:u # 0.00% of all >> branches >> >> 1.684677260 seconds time elapsed >> >> 1.692474000 seconds user >> 0.004020000 seconds sys >> >> >> >> $ perf stat ./example.com.test -test.bench=BenchmarkNoInline >> -test.benchtime=100000000x >> goos: linux >> goarch: amd64 >> pkg: example.com >> cpu: Intel(R) Xeon(R) W-2135 CPU @ 3.70GHz >> BenchmarkNoInline-12 100000000 10.79 ns/op >> PASS >> >> Performance counter stats for './example.com.test >> -test.bench=BenchmarkNoInline -test.benchtime=100000000x': >> >> 1,091.71 msec task-clock:u # 1.005 CPUs >> utilized >> 0 context-switches:u # 0.000 /sec >> >> 0 cpu-migrations:u # 0.000 /sec >> >> 363 page-faults:u # 332.505 /sec >> >> 4,490,159,750 cycles:u # 4.113 GHz >> >> 20,205,764,499 instructions:u # 4.50 insn per >> cycle >> 6,701,281,015 branches:u # 6.138 G/sec >> >> 586,073 branch-misses:u # 0.01% of all >> branches >> >> 1.086302272 seconds time elapsed >> >> 1.087710000 seconds user >> 0.008027000 seconds sys >> >> The non-inlined version is actually fewer instructions to run the same >> benchmark, which surprises me because naively looking at the disassembly it >> seems that the inlined version is much more compact. >> >> >> On Fri, Jul 22, 2022 at 5:52 AM eric...@arm.com <eric...@arm.com> wrote: >> >>> For this piece of code, two test functions are the same, but one is >>> inlined, the other is not. However the inlined version is about 25% slower >>> than the no inlined version on apple m1 chip. Why is it? >>> >>> The code is here https://go.dev/play/p/0NkLMtTZtv4 >>> >>> -- >>> You received this message because you are subscribed to the Google >>> Groups "golang-nuts" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to golang-nuts...@googlegroups.com. >>> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/golang-nuts/527264d7-7cc1-4278-9a29-c04eb3ec4e86n%40googlegroups.com >>> >>> <https://groups.google.com/d/msgid/golang-nuts/527264d7-7cc1-4278-9a29-c04eb3ec4e86n%40googlegroups.com?utm_medium=email&utm_source=footer> >>> . >>> >> -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/golang-nuts/0bed431c-d1f0-4faf-a384-7ac9755d8807n%40googlegroups.com.
