RFR: 8349452: Fix performance regression for Arrays.fill() with AVX512 [v11]

[Emanuel Peter]

On Wed, 24 Dec 2025 22:06:43 GMT, Srinivas Vamsi Parasa <sparasa at openjdk.org> wrote:

>> The goal of this PR is to fix the performance regression in Arrays.fill() x86 stubs caused by masked AVX stores. The fix is to replace the masked AVX stores with store instructions without masks (i.e. unmasked stores). `fill32_masked()` and `fill64_masked()` stubs are replaced with `fill32_unmasked()` and `fill64_unmasked()` respectively.
>> 
>> To speedup unmasked stores, array fills for sizes < 64 bytes are broken down into sequences of 32B, 16B, 8B, 4B, 2B and 1B stores, depending on the size.
>> 
>> 
>> ### **Performance comparison for byte array fills in a loop for 1 million times**
>> 
>> 
>> UseAVX=3   ByteArray Size | +OptimizeFill    (Masked store   stub)     [secs] | -OptimizeFill   (No stub)   [secs] | --->This PR: +OptimizeFill   (Unmasked store   stub)   [secs]
>> -- | -- | -- | --
>> 1 | 0.46 | 0.14 | 0.189
>> 2 | 0.46 | 0.16 | 0.191
>> 3 | 0.46 | 0.176 | 0.199
>> 4 | 0.46 | 0.244 | 0.212
>> 5 | 0.46 | 0.29 | 0.364
>> 10 | 0.46 | 0.58 | 0.354
>> 15 | 0.46 | 0.42 | 0.325
>> 16 | 0.46 | 0.46 | 0.281
>> 17 | 0.21 | 0.5 | 0.365
>> 20 | 0.21 | 0.37 | 0.326
>> 25 | 0.21 | 0.59 | 0.343
>> 31 | 0.21 | 0.53 | 0.317
>> 32 | 0.21 | 0.58 | 0.249
>> 35 | 0.5 | 0.77 | 0.303
>> 40 | 0.5 | 0.61 | 0.312
>> 45 | 0.5 | 0.52 | 0.364
>> 48 | 0.5 | 0.66 | 0.283
>> 49 | 0.22 | 0.69 | 0.367
>> 50 | 0.22 | 0.78 | 0.344
>> 55 | 0.22 | 0.67 | 0.332
>> 60 | 0.22 | 0.67 | 0.312
>> 64 | 0.22 | 0.82 | 0.253
>> 70 | 0.51 | 1.1 | 0.394
>> 80 | 0.49 | 0.89 | 0.346
>> 90 | 0.225 | 0.68 | 0.385
>> 100 | 0.54 | 1.09 | 0.364
>> 110 | 0.6 | 0.98 | 0.416
>> 120 | 0.26 | 0.75 | 0.367
>> 128 | 0.266 | 1.1 | 0.342
>
> Srinivas Vamsi Parasa has updated the pull request incrementally with one additional commit since the last revision:
> 
>   cleanup code related to labels

Also:
- You only show data for byte arrays, right? What about other element types?
- Do you control for alignment of the array, i.e. what if the intrinsic does not start at the beginning of the array but at some arbitrary offset?
- It would be nice to see a plot for all sizes from 0-300, and maybe some select larger sizes. That would give us a clearer picture of the behavior of your intrinsics.

You could also use some selected benchmarks from my recent PR https://github.com/openjdk/jdk/pull/27315. You could use the series `fill_var_*_arrays_fill`, and compare before and after your PR.

And since this is a regression ... why not also plot the performance from before [JDK-8275047](https://bugs.openjdk.org/browse/JDK-8275047), so we have a more complete picture?

-------------

PR Comment: https://git.openjdk.org/jdk/pull/28442#issuecomment-3713732680
PR Comment: https://git.openjdk.org/jdk/pull/28442#issuecomment-3713739854