Ian Henriksen

I'm including the PR dropping Tilera, Itanium, MIPS, and Sparc architectures in the 1.21 release. I'm clearing this issue from the release milestone though since the 32-bit stuff is still...

Fixed in #273.

Okay, just looked into the status of this a bit more. External API functions and some internal stuff is all marked. I think the only change needed here is setting...

Fixed in https://github.com/sandialabs/qthreads/pull/320. There's more work to do for Windows DLLs if we ever get around to supporting those. The current visibility header does the dllexport/dllimport stuff wrong. I'm calling...

Debugged some more. It's not actually a deadlock. It's just that that particular test hits the thread sanitizer performance penalty really hard. Adjusting the problem size gets it down to...

Closing in favor of https://github.com/sandialabs/qthreads/issues/303

Poking at some of the memory fences for syncvars seems like a partial fix. Some additional atomics in the nemesis scheduler may be necessary to get the sequencing right, but...

Update on this one, I'm fairly sure a lot (all?) of these errors are caused by thread sanitizer not being able to properly follow dependencies through the atomic logic in...

Similar: https://github.com/sandialabs/qthreads/blob/d6ce514a70c65b74c5e04906615ec51c7f288e0f/src/syncvar.c#L268 and https://github.com/sandialabs/qthreads/blob/d6ce514a70c65b74c5e04906615ec51c7f288e0f/src/syncvar.c#L455.

Okay, it looks like the syncvar code frequently uses the idiom of copying the whole `syncvar_t` to the local stack to manipulate things there. That's not something we can fix...