Re: RFR: 8285633: Take better advantage of generic MethodType cache [v2]
On Tue, 26 Apr 2022 20:47:39 GMT, Claes Redestad wrote: >> The `MethodType.genericMethodType` methods provide convenience methods for >> certain common method types and also provide `@Stable` cache that allows for >> constant folding. This patch enhances the more generic `methodType` methods >> to take advantage of this cache, when possible. This allows calls like >> `MethodType.methodType(Object.class, Object.class, ...)` to constant fold, >> making them 50x+ faster and allocation-free. >> >> Baseline: >> >> BenchmarkMode Cnt Score Error Units >> MethodTypeAcquire.baselineRawavgt 15 0.673 ? 0.017 ns/op >> MethodTypeAcquire.testGenericObject avgt 15 0.579 ? 0.125 ns/op >> MethodTypeAcquire.testMultiPType avgt 15 46.671 ? 1.134 ns/op >> MethodTypeAcquire.testMultiPType_Arg avgt 15 27.019 ? 0.437 ns/op >> MethodTypeAcquire.testMultiPType_ObjectAndA avgt 15 57.217 ? 0.505 ns/op >> MethodTypeAcquire.testMultiPType_ObjectOnly avgt 15 57.114 ? 1.214 ns/op >> MethodTypeAcquire.testObjectObject avgt 15 33.380 ? 1.810 ns/op >> MethodTypeAcquire.testReturnInt avgt 15 28.043 ? 0.187 ns/op >> MethodTypeAcquire.testReturnObject avgt 15 24.313 ? 0.074 ns/op >> MethodTypeAcquire.testReturnVoid avgt 15 24.360 ? 0.155 ns/op >> MethodTypeAcquire.testSinglePTypeavgt 15 33.572 ? 1.101 ns/op >> >> >> Patched: >> >> BenchmarkMode Cnt Score Error Units >> MethodTypeAcquire.baselineRawavgt 15 0.683 ? 0.024 ns/op >> MethodTypeAcquire.testGenericObject avgt 15 0.547 ? 0.047 ns/op >> MethodTypeAcquire.testMultiPType avgt 15 48.532 ? 0.982 ns/op >> MethodTypeAcquire.testMultiPType_Arg avgt 15 31.390 ? 5.269 ns/op >> MethodTypeAcquire.testMultiPType_ObjectAndA avgt 15 60.165 ? 2.756 ns/op >> MethodTypeAcquire.testMultiPType_ObjectOnly avgt 15 0.599 ? 0.166 ns/op >> MethodTypeAcquire.testObjectObject avgt 15 0.541 ? 0.045 ns/op >> MethodTypeAcquire.testReturnInt avgt 15 28.174 ? 0.257 ns/op >> MethodTypeAcquire.testReturnObject avgt 15 0.542 ? 0.045 ns/op >> MethodTypeAcquire.testReturnVoid avgt 15 24.621 ? 0.202 ns/op >> MethodTypeAcquire.testSinglePTypeavgt 15 33.614 ? 1.109 ns/op >> >> >> The cost on method types that don't match are in the noise (0-2ns/op). Even >> on a test I constructed to act as a worst case (`testMultiPType_ObjectAndA`) >> there's barely any appreciable cost (~3ns/op, with overlapping error). > > Claes Redestad has updated the pull request incrementally with one additional > commit since the last revision: > > Add micros using non-constant arguments to assess performance in absense of > constant folding Good work. - PR: https://git.openjdk.java.net/jdk/pull/8398
Re: RFR: 8285633: Take better advantage of generic MethodType cache [v2]
On Tue, 26 Apr 2022 19:30:31 GMT, Jorn Vernee wrote: >> Right.. I did a quick experiment and there's a large speed-up in the trivial >> `methodType(obj, obj)` case: >> >> Benchmark Mode Cnt Score Error >> Units >> MethodTypeAcquire.testObjectObjectNonConstant avgt5 30.052 ? 3.440 >> ns/op # baseline >> MethodTypeAcquire.testObjectObjectNonConstant avgt5 1.171 ? 0.001 >> ns/op # patch >> >> >> I'll add a non-constant variant for the multi-arg Object variants, too. It >> should scale linearly and see a gain in the same ballpark. I think we need >> to keep the number of microbenchmarks here somewhat under control, though. > > That sounds good, thanks Added 3 micros I think carry their own weight as they assess some different major pathways through the caching fast-path. Baseline: Benchmark Mode Cnt Score Error Units MethodTypeAcquire.testMultiPType_ObjectAndA_NonConst avgt 15 63.523 ? 2.452 ns/op MethodTypeAcquire.testMultiPType_ObjectOnly_NonConst avgt 15 57.770 ? 0.501 ns/op MethodTypeAcquire.testObjectObject_NonConst avgt 15 30.090 ? 0.251 ns/op Patched: Benchmark Mode Cnt Score Error Units MethodTypeAcquire.testMultiPType_ObjectAndA_NonConst avgt 15 64.570 ? 0.410 ns/op MethodTypeAcquire.testMultiPType_ObjectOnly_NonConst avgt 15 4.735 ? 0.063 ns/op MethodTypeAcquire.testObjectObject_NonConst avgt 15 1.171 ? 0.001 ns/op The relative speed-up diminishes a bit for the loopy 6-arg `methodType` calls, but is still a healthy 12x. Overhead on the negative test might be there but is very much in the noise on my tests (tried doubling warmup time to no avail) - PR: https://git.openjdk.java.net/jdk/pull/8398
Re: RFR: 8285633: Take better advantage of generic MethodType cache [v2]
> The `MethodType.genericMethodType` methods provide convenience methods for > certain common method types and also provide `@Stable` cache that allows for > constant folding. This patch enhances the more generic `methodType` methods > to take advantage of this cache, when possible. This allows calls like > `MethodType.methodType(Object.class, Object.class, ...)` to constant fold, > making them 50x+ faster and allocation-free. > > Baseline: > > BenchmarkMode Cnt Score Error Units > MethodTypeAcquire.baselineRawavgt 15 0.673 ? 0.017 ns/op > MethodTypeAcquire.testGenericObject avgt 15 0.579 ? 0.125 ns/op > MethodTypeAcquire.testMultiPType avgt 15 46.671 ? 1.134 ns/op > MethodTypeAcquire.testMultiPType_Arg avgt 15 27.019 ? 0.437 ns/op > MethodTypeAcquire.testMultiPType_ObjectAndA avgt 15 57.217 ? 0.505 ns/op > MethodTypeAcquire.testMultiPType_ObjectOnly avgt 15 57.114 ? 1.214 ns/op > MethodTypeAcquire.testObjectObject avgt 15 33.380 ? 1.810 ns/op > MethodTypeAcquire.testReturnInt avgt 15 28.043 ? 0.187 ns/op > MethodTypeAcquire.testReturnObject avgt 15 24.313 ? 0.074 ns/op > MethodTypeAcquire.testReturnVoid avgt 15 24.360 ? 0.155 ns/op > MethodTypeAcquire.testSinglePTypeavgt 15 33.572 ? 1.101 ns/op > > > Patched: > > BenchmarkMode Cnt Score Error Units > MethodTypeAcquire.baselineRawavgt 15 0.683 ? 0.024 ns/op > MethodTypeAcquire.testGenericObject avgt 15 0.547 ? 0.047 ns/op > MethodTypeAcquire.testMultiPType avgt 15 48.532 ? 0.982 ns/op > MethodTypeAcquire.testMultiPType_Arg avgt 15 31.390 ? 5.269 ns/op > MethodTypeAcquire.testMultiPType_ObjectAndA avgt 15 60.165 ? 2.756 ns/op > MethodTypeAcquire.testMultiPType_ObjectOnly avgt 15 0.599 ? 0.166 ns/op > MethodTypeAcquire.testObjectObject avgt 15 0.541 ? 0.045 ns/op > MethodTypeAcquire.testReturnInt avgt 15 28.174 ? 0.257 ns/op > MethodTypeAcquire.testReturnObject avgt 15 0.542 ? 0.045 ns/op > MethodTypeAcquire.testReturnVoid avgt 15 24.621 ? 0.202 ns/op > MethodTypeAcquire.testSinglePTypeavgt 15 33.614 ? 1.109 ns/op > > > The cost on method types that don't match are in the noise (0-2ns/op). Even > on a test I constructed to act as a worst case (`testMultiPType_ObjectAndA`) > there's barely any appreciable cost (~3ns/op, with overlapping error). Claes Redestad has updated the pull request incrementally with one additional commit since the last revision: Add micros using non-constant arguments to assess performance in absense of constant folding - Changes: - all: https://git.openjdk.java.net/jdk/pull/8398/files - new: https://git.openjdk.java.net/jdk/pull/8398/files/6d2edc59..fcc020e6 Webrevs: - full: https://webrevs.openjdk.java.net/?repo=jdk=8398=01 - incr: https://webrevs.openjdk.java.net/?repo=jdk=8398=00-01 Stats: 18 lines in 1 file changed: 18 ins; 0 del; 0 mod Patch: https://git.openjdk.java.net/jdk/pull/8398.diff Fetch: git fetch https://git.openjdk.java.net/jdk pull/8398/head:pull/8398 PR: https://git.openjdk.java.net/jdk/pull/8398
