RFR: 8265128: [REDO] Optimize Vector API slice and unslice operations
Paul Sandoz
psandoz at openjdk.java.net
Thu Apr 29 23:12:51 UTC 2021
On Thu, 29 Apr 2021 21:29:03 GMT, Sandhya Viswanathan <sviswanathan at openjdk.org> wrote:
> All the slice and unslice variants that take more than one argument can benefit from already intrinsic methods on similar lines as slice(origin) and unslice(origin).
>
> Changes include:
> * Rewrite Vector API slice/unslice using already intrinsic methods
> * Fix in library_call.cpp:inline_preconditions_checkIndex() to not modify control if intrinsification fails
> * Vector API conversion tests thresholds adjustment
>
> Base Performance:
> Benchmark (size) Mode Cnt Score Error Units
> TestSlice.vectorSliceOrigin 1024 thrpt 5 11763.372 ± 254.580 ops/ms
> TestSlice.vectorSliceOriginVector 1024 thrpt 5 599.286 ± 326.770 ops/ms
> TestSlice.vectorSliceUnsliceOrigin 1024 thrpt 5 6627.601 ± 22.060 ops/ms
> TestSlice.vectorSliceUnsliceOriginVector 1024 thrpt 5 401.858 ± 220.340 ops/ms
> TestSlice.vectorSliceUnsliceOriginVectorPart 1024 thrpt 5 421.993 ± 231.703 ops/ms
>
> Performance with patch:
> Benchmark (size) Mode Cnt Score Error Units
> TestSlice.vectorSliceOrigin 1024 thrpt 5 11792.091 ± 37.296 ops/ms
> TestSlice.vectorSliceOriginVector 1024 thrpt 5 8388.174 ± 115.886 ops/ms
> TestSlice.vectorSliceUnsliceOrigin 1024 thrpt 5 6662.159 ± 8.203 ops/ms
> TestSlice.vectorSliceUnsliceOriginVector 1024 thrpt 5 5206.300 ± 43.637 ops/ms
> TestSlice.vectorSliceUnsliceOriginVectorPart 1024 thrpt 5 5194.278 ± 13.376 ops/ms
src/jdk.incubator.vector/share/classes/jdk/incubator/vector/X-Vector.java.template line 2283:
> 2281: @ForceInline
> 2282: $abstractvectortype$ sliceTemplate(int origin) {
> 2283: Objects.checkIndex(origin, length());
For `Objects.checkIndex` the upper bound is exclusive, where as i think in this case it needs to be inclusive.
The original bound check was:
if ((origin < 0) || (origin >= VLENGTH)) {
throw new ArrayIndexOutOfBoundsException("Index " + origin + " out of bounds for vector length " + VLENGTH);
}
...
in accordance with the spec in the JavaDoc.
src/jdk.incubator.vector/share/classes/jdk/incubator/vector/X-Vector.java.template line 2284:
> 2282: $abstractvectortype$ sliceTemplate(int origin) {
> 2283: Objects.checkIndex(origin, length());
> 2284: VectorShuffle<$Boxtype$> Iota = iotaShuffle();
Suggestion:
VectorShuffle<$Boxtype$> iota = iotaShuffle();
use lower case first letter for variable names.
src/jdk.incubator.vector/share/classes/jdk/incubator/vector/X-Vector.java.template line 2287:
> 2285: VectorMask<$Boxtype$> BlendMask = Iota.toVector().compare(VectorOperators.LT, (broadcast(($type$)(length() - origin))));
> 2286: Iota = iotaShuffle(origin, 1, true);
> 2287: return vspecies().zero().blend(this.rearrange(Iota), BlendMask);
Observation: this is like `this.rearrange(iota, blendMask)` but we avoid the checking for exceptional shuffle indexes, since we know there are no such exceptional indexes.
And, of course,this method is like `slice(origin vspecies().zero())`. I wonder if this template should defer to the vector accepting tempting? Or the method on vector itself?
test/jdk/jdk/incubator/vector/AbstractVectorConversionTest.java line 51:
> 49: }
> 50:
> 51: static final int INVOC_COUNT = Integer.getInteger("jdk.incubator.vector.test.loop-iterations", 100);
Why did you change this?
-------------
PR: https://git.openjdk.java.net/jdk/pull/3804
More information about the core-libs-dev
mailing list