jdk/test/hotspot/jtreg/compiler/vectorapi/VectorStoreMaskIdentityTest.java
erfang fca9b3e513 8370863: VectorAPI: Optimize the VectorMaskCast chain in specific patterns
`VectorMaskCastNode` is used to cast a vector mask from one type to
another type. The cast may be generated by calling the vector API `cast`
or generated by the compiler. For example, some vector mask operations
like `trueCount` require the input mask to be integer types, so for
floating point type masks, the compiler will cast the mask to the
corresponding integer type mask automatically before doing the mask
operation. This kind of cast is very common.

If the vector element size is not changed, the `VectorMaskCastNode`
don't generate code, otherwise code will be generated to extend or narrow
the mask. This IR node is not free no matter it generates code or not
because it may block some optimizations. For example:
1. `(VectorStoremask (VectorMaskCast (VectorLoadMask x)))`
The middle `VectorMaskCast` prevented the following optimization:
`(VectorStoremask (VectorLoadMask x)) => (x)`
2. `(VectorMaskToLong (VectorMaskCast (VectorLongToMask x)))`, which
blocks the optimization `(VectorMaskToLong (VectorLongToMask x)) => (x)`.

In these IR patterns, the value of the input `x` is not changed, so we
can safely do the optimization. But if the input value is changed, we
can't eliminate the cast.

The general idea of this PR is introducing an `uncast_mask` helper
function, which can be used to uncast a chain of `VectorMaskCastNode`,
like the existing `Node::uncast(bool)` function. The funtion returns
the first non `VectorMaskCastNode`.

The intended use case is when the IR pattern to be optimized may
contain one or more consecutive `VectorMaskCastNode` and this does not
affect the correctness of the optimization. Then this function can be
called to eliminate the `VectorMaskCastNode` chain.

Current optimizations related to `VectorMaskCastNode` include:
1. `(VectorMaskCast (VectorMaskCast x)) => (x)`, see JDK-8356760.
2. `(XorV (VectorMaskCast (VectorMaskCmp src1 src2 cond)) (Replicate -1))
    => (VectorMaskCast (VectorMaskCmp src1 src2 ncond))`, see JDK-8354242.

This PR does the following optimizations:
1. Extends the optimization pattern `(VectorMaskCast (VectorMaskCast x)) => (x)`
as `(VectorMaskCast (VectorMaskCast  ... (VectorMaskCast x))) => (x)`.
Because as long as types of the head and tail `VectorMaskCastNode` are
consistent, the optimization is correct.
2. Supports a new optimization pattern
`(VectorStoreMask (VectorMaskCast ... (VectorLoadMask x))) => (x)`.
Since the value before and after the pattern is a boolean vector, it
remains unchanged as long as the vector length remains the same, and
this is guranteed in the api level.

I conducted some simple research on different mask generation methods
and mask operations, and obtained the following table, which includes
some potential optimization opportunities that may use this `uncast_mask`
function.

```
mask_gen\op    toLong   anyTrue allTrue trueCount firstTrue lastTrue
compare        N/A      N/A     N/A     N/A       N/A       N/A
maskAll        TBI      TBI     TBI     TBI       TBI       TBI
fromLong       TBI      TBI     N/A     TBI       TBI       TBI

mask_gen\op    and      or      xor     andNot    not       laneIsSet
compare        N/A      N/A     N/A     N/A       TBI       N/A
maskAll        TBI      TBI     TBI     TBI       TBI       TBI
fromLong       N/A      N/A     N/A     N/A       TBI       TBI
```
`TBI` indicated that there may be potential optimizations here that
require further investigation.

Benchmarks:

On a Nvidia Grace machine with 128-bit SVE2:
```
Benchmark			Unit	Before	Error	After	Error	Uplift
microMaskLoadCastStoreByte64	ops/us	59.23	0.21	148.12	0.07	2.50
microMaskLoadCastStoreDouble128	ops/us	2.43	0.00	38.31	0.01	15.73
microMaskLoadCastStoreFloat128	ops/us	6.19	0.00	75.67	0.11	12.22
microMaskLoadCastStoreInt128	ops/us	6.19	0.00	75.67	0.03	12.22
microMaskLoadCastStoreLong128	ops/us	2.43	0.00	38.32	0.01	15.74
microMaskLoadCastStoreShort64	ops/us	28.89	0.02	75.60	0.09	2.62
```

On a Nvidia Grace machine with 128-bit NEON:
```
Benchmark			Unit	Before	Error	After	Error	Uplift
microMaskLoadCastStoreByte64	ops/us	75.75	0.19	149.74	0.08	1.98
microMaskLoadCastStoreDouble128	ops/us	8.71	0.03	38.71	0.05	4.44
microMaskLoadCastStoreFloat128	ops/us	24.05	0.03	76.49	0.05	3.18
microMaskLoadCastStoreInt128	ops/us	24.06	0.02	76.51	0.05	3.18
microMaskLoadCastStoreLong128	ops/us	8.72	0.01	38.71	0.02	4.44
microMaskLoadCastStoreShort64	ops/us	24.64	0.01	76.43	0.06	3.10
```

On an AMD EPYC 9124 16-Core Processor with AVX3:
```
Benchmark			Unit	Before	Error	After	Error	Uplift
microMaskLoadCastStoreByte64	ops/us	82.13	0.31	115.14	0.08	1.40
microMaskLoadCastStoreDouble128	ops/us	0.32	0.00	0.32	0.00	1.01
microMaskLoadCastStoreFloat128	ops/us	42.18	0.05	57.56	0.07	1.36
microMaskLoadCastStoreInt128	ops/us	42.19	0.01	57.53	0.08	1.36
microMaskLoadCastStoreLong128	ops/us	0.30	0.01	0.32	0.00	1.05
microMaskLoadCastStoreShort64	ops/us	42.18	0.05	57.59	0.01	1.37
```

On an AMD EPYC 9124 16-Core Processor with AVX2:
```
Benchmark			Unit	Before	Error	After	Error	Uplift
microMaskLoadCastStoreByte64	ops/us	73.53	0.20	114.98	0.03	1.56
microMaskLoadCastStoreDouble128	ops/us	0.29	0.01	0.30	0.00	1.00
microMaskLoadCastStoreFloat128	ops/us	30.78	0.14	57.50	0.01	1.87
microMaskLoadCastStoreInt128	ops/us	30.65	0.26	57.50	0.01	1.88
microMaskLoadCastStoreLong128	ops/us	0.30	0.00	0.30	0.00	0.99
microMaskLoadCastStoreShort64	ops/us	24.92	0.00	57.49	0.01	2.31
```

On an AMD EPYC 9124 16-Core Processor with AVX1:
```
Benchmark			Unit	Before	Error	After	Error	Uplift
microMaskLoadCastStoreByte64	ops/us	79.68	0.01	248.49	0.91	3.12
microMaskLoadCastStoreDouble128	ops/us	0.28	0.00	0.28	0.00	1.00
microMaskLoadCastStoreFloat128	ops/us	31.11	0.04	95.48	2.27	3.07
microMaskLoadCastStoreInt128	ops/us	31.10	0.03	99.94	1.87	3.21
microMaskLoadCastStoreLong128	ops/us	0.28	0.00	0.28	0.00	0.99
microMaskLoadCastStoreShort64	ops/us	31.11	0.02	94.97	2.30	3.05
```

This PR was tested on 128-bit, 256-bit, and 512-bit (QEMU) aarch64
environments, and two 512-bit x64 machines with various configurations,
including sve2, sve1, neon, avx3, avx2, avx1, sse4 and sse3, all tests
passed.
2025-11-14 01:05:55 +00:00

290 lines
13 KiB
Java

/*
* Copyright (c) 2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
*
* This code is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License version 2 only, as
* published by the Free Software Foundation.
*
* This code is distributed in the hope that it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
* version 2 for more details (a copy is included in the LICENSE file that
* accompanied this code).
*
* You should have received a copy of the GNU General Public License version
* 2 along with this work; if not, write to the Free Software Foundation,
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
*
* Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
* or visit www.oracle.com if you need additional information or have any
* questions.
*/
/*
* @test
* @bug 8370863
* @library /test/lib /
* @summary VectorStoreMaskNode Identity optimization tests
* @modules jdk.incubator.vector
*
* @run driver compiler.vectorapi.VectorStoreMaskIdentityTest
*/
package compiler.vectorapi;
import compiler.lib.ir_framework.*;
import jdk.incubator.vector.*;
import jdk.test.lib.Asserts;
public class VectorStoreMaskIdentityTest {
private static final int LENGTH = 256; // large enough
private static boolean[] mask_in;
private static boolean[] mask_out;
static {
mask_in = new boolean[LENGTH];
mask_out = new boolean[LENGTH];
for (int i = 0; i < LENGTH; i++) {
mask_in[i] = (i & 3) == 0;
}
}
@ForceInline
private static void testOneCastKernel(VectorSpecies<?> from_species,
VectorSpecies<?> to_species) {
VectorMask.fromArray(from_species, mask_in, 0)
.cast(to_species).intoArray(mask_out, 0);
}
@ForceInline
private static void testTwoCastsKernel(VectorSpecies<?> from_species,
VectorSpecies<?> to_species1,
VectorSpecies<?> to_species2) {
VectorMask.fromArray(from_species, mask_in, 0)
.cast(to_species1)
.cast(to_species2).intoArray(mask_out, 0);
}
@ForceInline
private static void testThreeCastsKernel(VectorSpecies<?> from_species,
VectorSpecies<?> to_species1,
VectorSpecies<?> to_species2,
VectorSpecies<?> to_species3) {
VectorMask.fromArray(from_species, mask_in, 0)
.cast(to_species1)
.cast(to_species2)
.cast(to_species3).intoArray(mask_out, 0);
}
@DontInline
private static void verifyResult(int vlen) {
for (int i = 0; i < vlen; i++) {
Asserts.assertEquals(mask_in[i], mask_out[i], "index " + i);
}
}
@Test
@IR(counts = { IRNode.VECTOR_LOAD_MASK, "= 0",
IRNode.VECTOR_STORE_MASK, "= 0",
IRNode.VECTOR_MASK_CAST, "= 0" },
applyIfCPUFeatureOr = { "asimd", "true", "avx", "true" })
public static void testVectorMaskStoreIdentityByte() {
testOneCastKernel(ByteVector.SPECIES_64, ShortVector.SPECIES_128);
verifyResult(ByteVector.SPECIES_64.length());
testTwoCastsKernel(ByteVector.SPECIES_64, ShortVector.SPECIES_128, ByteVector.SPECIES_64);
verifyResult(ByteVector.SPECIES_64.length());
testThreeCastsKernel(ByteVector.SPECIES_64, ShortVector.SPECIES_128, ByteVector.SPECIES_64, ShortVector.SPECIES_128);
verifyResult(ByteVector.SPECIES_64.length());
}
@Test
@IR(counts = { IRNode.VECTOR_LOAD_MASK, "= 0",
IRNode.VECTOR_STORE_MASK, "= 0",
IRNode.VECTOR_MASK_CAST, "= 0" },
applyIfCPUFeatureOr = { "asimd", "true", "avx2", "true" },
applyIf = { "MaxVectorSize", ">16" })
public static void testVectorMaskStoreIdentityByte256() {
testOneCastKernel(ByteVector.SPECIES_64, IntVector.SPECIES_256);
verifyResult(ByteVector.SPECIES_64.length());
testTwoCastsKernel(ByteVector.SPECIES_64, ShortVector.SPECIES_128, IntVector.SPECIES_256);
verifyResult(ByteVector.SPECIES_64.length());
testThreeCastsKernel(ByteVector.SPECIES_64, ShortVector.SPECIES_128, FloatVector.SPECIES_256, IntVector.SPECIES_256);
verifyResult(ByteVector.SPECIES_64.length());
}
@Test
@IR(counts = { IRNode.VECTOR_LOAD_MASK, "= 0",
IRNode.VECTOR_STORE_MASK, "= 0",
IRNode.VECTOR_MASK_CAST, "= 0" },
applyIfCPUFeatureOr = { "asimd", "true", "avx", "true" })
public static void testVectorMaskStoreIdentityShort() {
testOneCastKernel(ShortVector.SPECIES_128, ByteVector.SPECIES_64);
verifyResult(ShortVector.SPECIES_128.length());
testTwoCastsKernel(ShortVector.SPECIES_64, IntVector.SPECIES_128, ShortVector.SPECIES_64);
verifyResult(ShortVector.SPECIES_64.length());
testThreeCastsKernel(ShortVector.SPECIES_128, ByteVector.SPECIES_64, ShortVector.SPECIES_128, ByteVector.SPECIES_64);
verifyResult(ShortVector.SPECIES_128.length());
}
@Test
@IR(counts = { IRNode.VECTOR_LOAD_MASK, "= 0",
IRNode.VECTOR_STORE_MASK, "= 0",
IRNode.VECTOR_MASK_CAST, "= 0" },
applyIfCPUFeatureOr = { "asimd", "true", "avx2", "true" },
applyIf = { "MaxVectorSize", ">16" })
public static void testVectorMaskStoreIdentityShort256() {
testOneCastKernel(ShortVector.SPECIES_128, IntVector.SPECIES_256);
verifyResult(ShortVector.SPECIES_128.length());
testTwoCastsKernel(ShortVector.SPECIES_64, IntVector.SPECIES_128, LongVector.SPECIES_256);
verifyResult(ShortVector.SPECIES_64.length());
testThreeCastsKernel(ShortVector.SPECIES_128, ByteVector.SPECIES_64, FloatVector.SPECIES_256, IntVector.SPECIES_256);
verifyResult(ShortVector.SPECIES_128.length());
}
@Test
@IR(counts = { IRNode.VECTOR_LOAD_MASK, "= 0",
IRNode.VECTOR_STORE_MASK, "= 0",
IRNode.VECTOR_MASK_CAST, "= 0" },
applyIfCPUFeatureOr = { "asimd", "true", "avx", "true" })
public static void testVectorMaskStoreIdentityInt() {
testOneCastKernel(IntVector.SPECIES_MAX, FloatVector.SPECIES_MAX);
verifyResult(IntVector.SPECIES_MAX.length());
testTwoCastsKernel(IntVector.SPECIES_128, ShortVector.SPECIES_64, FloatVector.SPECIES_128);
verifyResult(IntVector.SPECIES_128.length());
testThreeCastsKernel(IntVector.SPECIES_128, ShortVector.SPECIES_64, FloatVector.SPECIES_128, ShortVector.SPECIES_64);
verifyResult(IntVector.SPECIES_128.length());
}
@Test
@IR(counts = { IRNode.VECTOR_LOAD_MASK, "= 0",
IRNode.VECTOR_STORE_MASK, "= 0",
IRNode.VECTOR_MASK_CAST, "= 0" },
applyIfCPUFeatureOr = { "asimd", "true", "avx2", "true" },
applyIf = { "MaxVectorSize", ">16" })
public static void testVectorMaskStoreIdentityInt256() {
testOneCastKernel(IntVector.SPECIES_MAX, FloatVector.SPECIES_MAX);
verifyResult(IntVector.SPECIES_MAX.length());
testTwoCastsKernel(IntVector.SPECIES_128, ShortVector.SPECIES_64, LongVector.SPECIES_256);
verifyResult(IntVector.SPECIES_128.length());
testThreeCastsKernel(IntVector.SPECIES_128, ShortVector.SPECIES_64, FloatVector.SPECIES_128, LongVector.SPECIES_256);
verifyResult(IntVector.SPECIES_128.length());
}
@Test
@IR(counts = { IRNode.VECTOR_LOAD_MASK, "= 0",
IRNode.VECTOR_STORE_MASK, "= 0",
IRNode.VECTOR_MASK_CAST, "= 0" },
applyIfCPUFeatureOr = { "asimd", "true", "avx", "true" })
public static void testVectorMaskStoreIdentityLong() {
testOneCastKernel(LongVector.SPECIES_MAX, DoubleVector.SPECIES_MAX);
verifyResult(LongVector.SPECIES_MAX.length());
testTwoCastsKernel(LongVector.SPECIES_128, IntVector.SPECIES_64, DoubleVector.SPECIES_128);
verifyResult(LongVector.SPECIES_128.length());
testThreeCastsKernel(LongVector.SPECIES_128, IntVector.SPECIES_64, DoubleVector.SPECIES_128, IntVector.SPECIES_64);
verifyResult(LongVector.SPECIES_128.length());
}
@Test
@IR(counts = { IRNode.VECTOR_LOAD_MASK, "= 0",
IRNode.VECTOR_STORE_MASK, "= 0",
IRNode.VECTOR_MASK_CAST, "= 0" },
applyIfCPUFeatureOr = { "asimd", "true", "avx2", "true" },
applyIf = { "MaxVectorSize", ">16" })
public static void testVectorMaskStoreIdentityLong256() {
testOneCastKernel(LongVector.SPECIES_MAX, DoubleVector.SPECIES_MAX);
verifyResult(LongVector.SPECIES_MAX.length());
testTwoCastsKernel(LongVector.SPECIES_256, IntVector.SPECIES_128, ShortVector.SPECIES_64);
verifyResult(LongVector.SPECIES_256.length());
testThreeCastsKernel(LongVector.SPECIES_256, IntVector.SPECIES_128, FloatVector.SPECIES_128, ShortVector.SPECIES_64);
verifyResult(LongVector.SPECIES_256.length());
}
@Test
@IR(counts = { IRNode.VECTOR_LOAD_MASK, "= 0",
IRNode.VECTOR_STORE_MASK, "= 0",
IRNode.VECTOR_MASK_CAST, "= 0" },
applyIfCPUFeatureOr = { "asimd", "true", "avx", "true" })
public static void testVectorMaskStoreIdentityFloat() {
testOneCastKernel(FloatVector.SPECIES_MAX, IntVector.SPECIES_MAX);
verifyResult(FloatVector.SPECIES_MAX.length());
testTwoCastsKernel(FloatVector.SPECIES_128, ShortVector.SPECIES_64, IntVector.SPECIES_128);
verifyResult(FloatVector.SPECIES_128.length());
testThreeCastsKernel(FloatVector.SPECIES_128, ShortVector.SPECIES_64, IntVector.SPECIES_128, ShortVector.SPECIES_64);
verifyResult(FloatVector.SPECIES_128.length());
}
@Test
@IR(counts = { IRNode.VECTOR_LOAD_MASK, "= 0",
IRNode.VECTOR_STORE_MASK, "= 0",
IRNode.VECTOR_MASK_CAST, "= 0" },
applyIfCPUFeatureOr = { "asimd", "true", "avx2", "true" },
applyIf = { "MaxVectorSize", ">16" })
public static void testVectorMaskStoreIdentityFloat256() {
testOneCastKernel(FloatVector.SPECIES_MAX, IntVector.SPECIES_MAX);
verifyResult(FloatVector.SPECIES_MAX.length());
testTwoCastsKernel(FloatVector.SPECIES_128, ShortVector.SPECIES_64, LongVector.SPECIES_256);
verifyResult(FloatVector.SPECIES_128.length());
testThreeCastsKernel(FloatVector.SPECIES_128, ShortVector.SPECIES_64, IntVector.SPECIES_128, LongVector.SPECIES_256);
verifyResult(FloatVector.SPECIES_128.length());
}
@Test
@IR(counts = { IRNode.VECTOR_LOAD_MASK, "= 0",
IRNode.VECTOR_STORE_MASK, "= 0",
IRNode.VECTOR_MASK_CAST, "= 0" },
applyIfCPUFeatureOr = { "asimd", "true", "avx", "true" })
public static void testVectorMaskStoreIdentityDouble() {
testOneCastKernel(DoubleVector.SPECIES_MAX, LongVector.SPECIES_MAX);
verifyResult(DoubleVector.SPECIES_MAX.length());
testTwoCastsKernel(DoubleVector.SPECIES_128, IntVector.SPECIES_64, LongVector.SPECIES_128);
verifyResult(DoubleVector.SPECIES_128.length());
testThreeCastsKernel(DoubleVector.SPECIES_128, IntVector.SPECIES_64, LongVector.SPECIES_128, IntVector.SPECIES_64);
verifyResult(DoubleVector.SPECIES_128.length());
}
@Test
@IR(counts = { IRNode.VECTOR_LOAD_MASK, "= 0",
IRNode.VECTOR_STORE_MASK, "= 0",
IRNode.VECTOR_MASK_CAST, "= 0" },
applyIfCPUFeatureOr = { "asimd", "true", "avx2", "true" },
applyIf = { "MaxVectorSize", ">16" })
public static void testVectorMaskStoreIdentityDouble256() {
testOneCastKernel(DoubleVector.SPECIES_MAX, LongVector.SPECIES_MAX);
verifyResult(DoubleVector.SPECIES_MAX.length());
testTwoCastsKernel(DoubleVector.SPECIES_256, ShortVector.SPECIES_64, IntVector.SPECIES_128);
verifyResult(DoubleVector.SPECIES_256.length());
testThreeCastsKernel(DoubleVector.SPECIES_256, ShortVector.SPECIES_64, LongVector.SPECIES_256, IntVector.SPECIES_128);
verifyResult(DoubleVector.SPECIES_256.length());
}
public static void main(String[] args) {
TestFramework testFramework = new TestFramework();
testFramework.setDefaultWarmup(10000)
.addFlags("--add-modules=jdk.incubator.vector")
.start();
}
}