src: enable libm trig functions in V8

[jasnell]

Should provide better performance on some platforms.

                       confidence improvement accuracy (*)   (**)  (***)
util/trig.js n=1000000        ***     37.41 %       ±2.11% ±2.81% ±3.67%

Be aware that when doing many comparisons the risk of a false-positive
result increases. In this case, there are 1 comparisons, you can thus
expect the following amount of false-positive results:
  0.05 false positives, when considering a   5% risk acceptance (*, **, ***),
  0.01 false positives, when considering a   1% risk acceptance (**, ***),
  0.00 false positives, when considering a 0.1% risk acceptance (***)

Context: @t3dotgg put together some benchmarks that show the trig functions running in node.js aren't as fast as they could be. We did some digging and think this may help. https://github.com/t3dotgg/cf-vs-vercel-bench/blob/main/vanilla-bench/vercel-edition/api/

There's a possibility this won't work on all platforms/archs so running some tests in CI to verify first. Appears to be fine in CI!

/cc @nodejs/v8

[nodejs-github-bot]

Review requested:

• @nodejs/gyp
• @nodejs/performance

[nodejs-github-bot]

CI: https://ci.nodejs.org/job/node-test-pull-request/69607/ 💛

[devsnek]

just a note of interest... several years ago I tried applying libm optimizations to some operations in V8 (though, not the trig functions) and we eventually found the output to be bad enough in certain cases to necessitate reverting it.

[mcollina]

lgtm

[targos]

I don't oppose to the idea but I don't see how this change can have any effect, given that the build flag is not implemented in the GYP files:

[jasnell]

@targos ... that's why I wanted to be sure to tag you here ... :-) ... it definitely seemed to have an impact locally but I couldn't quite figure out if it was completely wired in correctly. What additional changes do you think would be necessary?

[lemire]

I don't oppose to the idea but I don't see how this change can have any effect

Don't we have a 37% gain in one case ? Or do I misread what @jasnell posted?

[jasnell]

That was what I saw locally but it's possible something else is happening there. Need to be certain

[targos]

We would need to port at least:

node/deps/v8/gni/v8.gni

Lines 207 to 208 in f0aa073

	# TODO: macros for determining endian type are clang specific.
	v8_use_libm_trig_functions = is_clang

node/deps/v8/BUILD.gn

Lines 1417 to 1419 in f0aa073

	if (v8_use_libm_trig_functions) {
	defines += [ "V8_USE_LIBM_TRIG_FUNCTIONS" ]
	}

This should go in https://github.com/nodejs/node/blob/f0aa073907fc88f64ca95e4821eb7fdf49b133e2/tools/v8_gypfiles/features.gypi, similar to many other build flags with a variable declaration at the top and a new condition for the define.

And possibly:

node/deps/v8/BUILD.gn

Lines 6930 to 6932 in f0aa073

	if (v8_use_libm_trig_functions) {
	deps += [ ":libm" ]
	}

node/deps/v8/BUILD.gn

Lines 6937 to 6957 in f0aa073

	if (v8_use_libm_trig_functions) {
	source_set("libm") {
	sources = [
	"third_party/glibc/src/sysdeps/ieee754/dbl-64/branred.c",
	"third_party/glibc/src/sysdeps/ieee754/dbl-64/branred.h",
	"third_party/glibc/src/sysdeps/ieee754/dbl-64/dla.h",
	"third_party/glibc/src/sysdeps/ieee754/dbl-64/endian.h",
	"third_party/glibc/src/sysdeps/ieee754/dbl-64/mydefs.h",
	"third_party/glibc/src/sysdeps/ieee754/dbl-64/s_sin.c",
	"third_party/glibc/src/sysdeps/ieee754/dbl-64/sincostab.c",
	"third_party/glibc/src/sysdeps/ieee754/dbl-64/trig.h",
	"third_party/glibc/src/sysdeps/ieee754/dbl-64/usncs.h",
	]
	configs += [ "//build/config/compiler:no_chromium_code" ]
	configs -= [ "//build/config/compiler:chromium_code" ]
	if (!is_debug) {
	# Build code using -O3, see: crbug.com/1084371.
	configs += [ "//build/config/compiler:optimize_speed" ]
	}
	}
	}

This is less trivial. We don't have third_party/glibc in our tree and I'm not sure we want to pull it?

[jasnell]

@targos ... yeah, that's pretty much what I suspected but was taking the optimistic approach initially ;-) ... If this is going to require the third_party/glibc then this might be a blocker. I'll dig more a bit later this week.