Optimize Math.random in CacheIR and Warp
Categories
(Core :: JavaScript Engine: JIT, task, P3)
Tracking
()
| Tracking | Status | |
|---|---|---|
| firefox79 | --- | fixed |
People
(Reporter: jandem, Assigned: jandem)
References
(Blocks 1 open bug)
Details
Attachments
(5 files)
Using Ion's RNG code for CacheIR too makes Baseline quite a lot faster on some Math.random micro-benchmarks.
|
Assignee | |
Comment 1•4 years ago
|
||
|
|
Assignee | |
Comment 2•4 years ago
|
||
Note that the math-random.js jit-test is a good correctness test for this. It
overrides the RNG's seed, generates a few thousand numbers, and then checks some
of them. I verified this test fails if I add a bug to the Math.random JIT code.
Depends on D77800
|
|
Assignee | |
Comment 3•4 years ago
|
||
Depends on D77801
|
|
Assignee | |
Comment 4•4 years ago
|
||
Drive-by performance improvement. convertUInt64ToDouble is slower because it has
a different code path for negative inputs but that can't happen here.
Depends on D77802
|
|
Assignee | |
Comment 5•4 years ago
|
||
Based on the existing convertUInt64ToDouble.
Depends on D77802
|
||
Updated•4 years ago
|
|
||
Comment 7•4 years ago
|
||
| bugherder | ||
https://hg.mozilla.org/mozilla-central/rev/034c28800409
https://hg.mozilla.org/mozilla-central/rev/a58f7048e7f4
https://hg.mozilla.org/mozilla-central/rev/e519a1393ea3
https://hg.mozilla.org/mozilla-central/rev/897c90b9b4c1
https://hg.mozilla.org/mozilla-central/rev/0aea458b79e4
Description
•