Closed
Bug 1753115
Opened 2 years ago
Closed 2 years ago
x64/x86 SIMD: Optimize moveSimd128XXX usage for VEX encoding
Categories
(Core :: JavaScript: WebAssembly, enhancement, P3)
Tracking
()
RESOLVED
FIXED
99 Branch
| Tracking | Status | |
|---|---|---|
| firefox99 | --- | fixed |
People
(Reporter: yury, Assigned: yury)
References
Details
Attachments
(1 file)
While working on bug 1752231, I noticed lots of inefficient moveSimd128, moveSimd128Int, moveSimd128Float, especially when AVX is enabled. Since we can use powers of the register allocator and VEX encoding, most of moves can be replaced by moveSimd128XXXIfNotAVX that were introduced in bug 1752231.
|
Assignee | |
Comment 1•2 years ago
|
||
- Replace with moveSimd128XXXIfNotAVX, when possible
- Fix I16x8ExtaddPairwiseI8x16S lowering
- Add postMoveSimd128IntIfNotAVX utility
Depends on D137080
|
||
Updated•2 years ago
|
Attachment #9261787 -
Attachment description: WIP: Bug 1753115 - Refactor usage of moveSimd128. → Bug 1753115 - Refactor usage of moveSimd128. r?jseward
Pushed by ydelendik@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/46048399bf0f Refactor usage of moveSimd128. r=jseward
|
||
Comment 3•2 years ago
|
||
| bugherder | ||
Status: ASSIGNED → RESOLVED
Closed: 2 years ago
status-firefox99:
--- → fixed
Resolution: --- → FIXED
Target Milestone: --- → 99 Branch
You need to log in
before you can comment on or make changes to this bug.
Description
•