RFR: 8308340: C2: Idealize Fma nodes [v3]

[Fei Yang]

On Tue, 27 Jun 2023 16:05:54 GMT, Vladimir Kozlov <kvn at openjdk.org> wrote:

>> Fei Gao 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 four additional commits since the last revision:
>> 
>>  - Merge branch 'master' into fg8308340
>>  - Move check for UseFMA from c2compiler.cpp to Matcher::match_rule_supported in .ad files
>>  - Merge branch 'master' into fg8308340
>>  - 8308340: C2: Idealize Fma nodes
>>    
>>    Some platforms, like aarch64, ppc, and riscv, support fusing
>>    `Math.fma(-a, b, c)` or `Math.fma(a, -b, c)` by generating
>>    partially symmetric match rules like:
>>    
>>    ```
>>      match(Set dst (FmaF src3 (Binary (NegF src1) src2)));
>>      match(Set dst (FmaF src3 (Binary src1 (NegF src2))));
>>    ```
>>    
>>    Since `Fma` is partially communitive, the patch is to convert
>>    `Math.fma(-a, b, c)` to `Math.fma(b, -a, c)` in gvn phase,
>>    making node patterns canonical. Then we can remove redundant
>>    rules.
>>    
>>    Also, we should guarantee that C2 generates `Fma` nodes only on
>>    platforms supporting `Fma` instructions before matcher, so we
>>    can remove all `predicate(UseFMA)` for all `Fma` rules.
>>    
>>    After the patch, the code size of libjvm.so on aarch64 platform
>>    decreased by 63.4k.
>>    
>>    The patch passed all tier 1 - 3 on aarch64 and x86 platforms.
>
> Looks good to me.
> You need second review.

> Thanks for your review @vnkozlov .
> 
> I would appreciate it very much if some expert on ppc or riscv could help review it! Perhaps @RealFYang @reinrich

Hello, the RISC-V part looks fine from what this PR is supposed to do. And this has passed tier1-3 tests on linux-riscv64 platform. Note that I didn't check the shared code changes.

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

PR Comment: https://git.openjdk.org/jdk/pull/14576#issuecomment-1643608041