Abhinav Anil Sharma

> nor is zero-masking handled in scatter-gather expansion. Looking more closely for the AVX512 gather (https://www.felixcloutier.com/x86/vpgatherdd:vpgatherdq), looks like it doesn't support zeroing-masking, and performs merging-masking always. So the scatter/gather expansion...

Did PR #2872 fix this issue? From a quick reading it seems there's still more work left for reducing lock contention.

The failure that this PR tries to reproduce seems to happen only on some GA runners (https://github.com/DynamoRIO/dynamorio/issues/5329#issuecomment-1037255523). This makes it hard to debug. It'd be good to keep this PR...

This came into my radar while adding support for some new Linux system calls for #5131, of which `rt_sigtimedwait_time64` is one. `sigwaitinfo` and `sigtimedwait` wait for one of the given...

Saw a segfault on #5581 for this test: http://139.178.82.61:8080/job/DynamoRIO-AArch64-Precommit/1867/console

Looks like `Sending SIGUSR1` is missing from the actual output.

I ran the 32-bit test 1000 times on my machine, and couldn't reproduce the failure: ``` $ ctest -VV -R 'tool.drcachesim.scattergather' --repeat-until-fail 1000 ... 1/1 Test #284: code_api|tool.drcachesim.scattergather ... Passed...

I ran this 1000 times on a GA runner, no failure encountered. Maybe it reproduces on only some runners? If that is so, I'll need to try this multiple times....

Looks like the following tests are failing deterministically on my machine (and Derek's) now: ``` 221 - code_api|client.drx-scattergather (Failed) 248 - code_api|sample.memval_simple_scattergather (Failed) 307 - code_api|tool.drcachesim.scattergather (Failed) ``` Would make...

Right. The fixes made in PR #5364 reduce stack space requirement by a lot (~147KB for `dump_local_stats`, ~127KB for `dump_global_stats`), but we still had to increase stack provisioning to 64K...