RFR: 8294588: Auto vectorize half precision floating point conversion APIs [v7]
Vladimir Kozlov
kvn at openjdk.org
Wed Dec 7 23:57:07 UTC 2022
On Wed, 7 Dec 2022 22:12:20 GMT, Vladimir Kozlov <kvn at openjdk.org> wrote:
>> New test failed when run with `-XX:UseAVX=1`. I added output to RFE in comment.
>>
>> - counts: Graph contains wrong number of nodes:
>> * Constraint 1: "(\\d+(\\s){2}(VectorCastH2F.*)+(\\s){2}===.*)"
>> - Failed comparison: [found] 0 > 0 [given]
>> - No nodes matched!
>
>> @vnkozlov I have addressed comments from Fei Gao and Xiaohong Gong. I have limited vectorization to avx2 and higher. If the changes look good to you, could you kindly run the tests? Thanks for all your help.
>
> @smita-kamath, can you explain why it does not work with AVX1? If it really requires AVX2 then you should just disable F16C for `(AVX < 2)` instead of current `(AVX < 1)` in `vm_version_x86.cpp`. And you would not need to modify `.ad` file and test.
> @vnkozlov I have updated the test case to work with AVX=1.
Can you explain what was wrong with AVX1 and what change fixed the issue?
I see you renamed classes and addressed @fg1417 comment about `opcode`. It is not clear to me what fixed AVX1 issue.
-------------
PR: https://git.openjdk.org/jdk/pull/11471
More information about the hotspot-compiler-dev
mailing list