RFR: 8333964: RISC-V: C2: Check "requires_strict_order" flag for floating-point add reduction [v2]
Gui Cao
gcao at openjdk.org
Wed Jun 12 12:58:12 UTC 2024
On Wed, 12 Jun 2024 09:33:25 GMT, Fei Yang <fyang at openjdk.org> wrote:
> Can you add a comment here? Like:
>
> ```
> // Distinguish two cases based on requires_strict_order
> // 1. Non strictly-ordered AddReductionVF/D. For example, AddReductionVF/D
> // generated by Vector API. It is more beneficial performance-wise to do
> // an unordered FP reduction sum (vfredusum.vs).
> // 2. Strictly-ordered AddReductionVF/D. For example, AddReductionVF/D
> // generated by auto-vectorization. Must do an ordered FP reduction sum
> // (vfredosum.vs).
> ```
Thanks for the review. Fixed.
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/19649#discussion_r1636415817
More information about the hotspot-compiler-dev
mailing list