On Sat, Feb 1, 2020 at 6:37 PM Thomas Koenig <tkoe...@netcologne.de> wrote: > > Hi Janne, > > > Simplify IO locking in libgfortran. The new IO implementation avoids > > accessing units without locks, as seen in PR 92836. It also avoids > > lock inversion (except for a corner case wrt namelist query when > > reading from stdin and outputting to stdout), making it easier to > > verify correctness with tools like valgrind or threadsanitizer. It is > > also simplified as the waiting and closed variables are not needed > > anymore, making it easier to understand and analyze. > > > > Regtested on x86_64-pc-linux-gnu, Ok for master? > > I'll look into it, this might take a bit of time.
Thanks for looking into this. > What are you planning to use as a test case? You can put > multighreading programs into libgomp/testsuite/libgomp.fortran > where they will be executed. In this case I just ran the test program in comment #14 in PR 92836 under valgrind --tool=helgrind (there is still a remaining problem with the locking of unit_cache, but that can be handled separately, and I think it's benign). I'm not sure how one would run such a test as part of the testsuite. But if anyone has tests that reliably deadlock within a relatively short time, then yeah, such testcases would be awesome. -- Janne Blomqvist
