Bug 164472
| Summary: | REGRESSION (r192721): GC timers stop to work when Heap claims it grew | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | VaL <valkov> |
| Component: | JavaScriptCore | Assignee: | Nobody <webkit-unassigned> |
| Status: | NEW | ||
| Severity: | Normal | CC: | ap, fpizlo, ivlev.igor, kling, mark.lam |
| Priority: | P2 | ||
| Version: | WebKit Nightly Build | ||
| Hardware: | Unspecified | ||
| OS: | Linux | ||
| See Also: | https://bugs.webkit.org/show_bug.cgi?id=151521 | ||
VaL
Due to https://bugs.webkit.org/show_bug.cgi?id=151521
when deathRate() now returns 0 (when heap reported that e.g. m_sizeAfterLastEdenCollect > m_sizeBeforeLastEdenCollect or m_sizeAfterLastFullCollect > m_sizeBeforeLastFullCollect after garbage collecting)
GCActivityCallback::didAllocate uses
double bytesExpectedToReclaim = static_cast<double>(bytes) * deathRate();
double newDelay = lastGCLength() / gcTimeSlice(bytesExpectedToReclaim);
scheduleTimer(newDelay);
which causes newDelay to be positive inf.
GCActivityCallback::scheduleTimer configures the timer to std::chrono::microseconds::max();
and the timer never invoked.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |