Closed
Bug 1693500
Opened 5 years ago
Closed 4 years ago
SIMD optimization x64/x86: improve i64x2.mul by constant
Categories
(Core :: JavaScript: WebAssembly, enhancement, P3)
Tracking
()
RESOLVED
FIXED
102 Branch
| Tracking | Status | |
|---|---|---|
| firefox102 | --- | fixed |
People
(Reporter: lth, Assigned: yury)
References
(Blocks 1 open bug)
Details
Attachments
(1 file)
The general i64x2.mul is 10 instructions -- SIMD code, but still. When one operand is a constant, especially when all lanes are the same constant, we should be able to do better (multiply by 2 is just an add or a shift, etc).
It is possible the front-end would already implement this optimization so we should check LLVM/emscripten output before prioritizing this work item.
|
Assignee | |
Comment 1•4 years ago
|
||
|
||
Updated•4 years ago
|
Assignee: nobody → ydelendik
Attachment #9271820 -
Attachment description: WIP: Bug 1693500 - [x64/x86] improve i64x2.mul by constant → Bug 1693500 - [x64/x86] improve i64x2.mul by constant. r?jseward
Status: NEW → ASSIGNED
Pushed by ydelendik@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/fa37a185b219
[x64/x86] improve i64x2.mul by constant. r=jseward
|
||
Comment 3•4 years ago
|
||
| bugherder | ||
Status: ASSIGNED → RESOLVED
Closed: 4 years ago
status-firefox102:
--- → fixed
Resolution: --- → FIXED
Target Milestone: --- → 102 Branch
You need to log in
before you can comment on or make changes to this bug.
Description
•