RFR: 8312425: [vectorapi] AArch64: Optimize vector math operations with SLEEF [v3]

[Paul Sandoz]

On Wed, 15 Nov 2023 01:32:00 GMT, Xiaohong Gong <xgong at openjdk.org> wrote:

>> Currently the vector floating-point math APIs like `VectorOperators.SIN/COS/TAN...` are not intrinsified on AArch64 platform, which causes large performance gap on AArch64. Note that those APIs are optimized by C2 compiler on X86 platforms by calling Intel's SVML code [1]. To close the gap, we would like to optimize these APIs for AArch64 by calling a third-party vector library called libsleef [2], which are available in mainstream Linux distros (e.g. [3] [4]).
>> 
>> SLEEF supports multiple accuracies. To match Vector API's requirement and implement the math ops on AArch64, we 1) call 1.0 ULP accuracy with FMA instructions used stubs in libsleef for most of the operations by default, and 2) add the vector calling convention to apply with the runtime calls to stub code in libsleef. Note that for those APIs that libsleef does not support 1.0 ULP, we choose 0.5 ULP instead.
>> 
>> To help loading the expected libsleef library, this patch also adds an experimental JVM option (i.e. `-XX:UseSleefLib`) for AArch64 platforms. People can use it to denote the libsleef path/name explicitly. By default, it points to the system installed library. If the library does not exist or the dynamic loading of it in runtime fails, the math vector ops will fall-back to use the default scalar version without error. But a warning is printed out if people specifies a nonexistent library explicitly.
>> 
>> Note that this is a part of the original proposed patch in panama-dev [5], just with some initial review comments addressed. And now we'd like to get some wider feedbacks from more hotspot experts.
>> 
>> [1] https://github.com/openjdk/jdk/pull/3638
>> [2] https://sleef.org/
>> [3] https://packages.fedoraproject.org/pkgs/sleef/sleef/
>> [4] https://packages.debian.org/bookworm/libsleef3
>> [5] https://mail.openjdk.org/pipermail/panama-dev/2022-December/018172.html
>
> Xiaohong Gong has updated the pull request with a new target base due to a merge or a rebase. The incremental webrev excludes the unrelated changes brought in by the merge/rebase. The pull request contains five additional commits since the last revision:
> 
>  - Add a bundled native lib in jdk as a bridge to libsleef
>  - Merge 'jdk:master' into JDK-8312425
>  - Disable sleef by default
>  - Merge 'jdk:master' into JDK-8312425
>  - 8312425: [vectorapi] AArch64: Optimize vector math operations with SLEEF

Have you considered the possibility of copying the sleef source to the OpenJDK repository and thereby it becomes part of the build process? I don't know how straightforward that is technically and IANAL but I think it's worth exploring.

Also it may enable us to use sleef for other platforms where we have gaps (looking at Table 1.1 of https://sleef.org/). Further out it should inspire us to do a Java Vector API port to as indicated in a prior comment.  

> Yes, libsleef is used to build the binary if found. And at runtime, if the libsleef with right version is not found, `dlopen` to the libvmath.so will fail. And then all the operations will be fall-back to the java default implementation. X86_64 has also bundled the Intel's SVML binary to jdk image at build time. And we use the same way loading/opening the library at runtime.

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

PR Comment: https://git.openjdk.org/jdk/pull/16234#issuecomment-1821885433