[llvm-bugs] [Bug 40282] New: tsan: false positives on shmat

via llvm-bugs llvm-bugs at lists.llvm.org
Thu Jan 10 02:51:18 PST 2019


            Bug ID: 40282
           Summary: tsan: false positives on shmat
           Product: compiler-rt
           Version: unspecified
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: tsan
          Assignee: unassignedbugs at nondot.org
          Reporter: dvyukov at google.com
                CC: llvm-bugs at lists.llvm.org

Reported internally, don't have a public repro.
The crux: shmat/shmdt allocate/deallocate memory (like mmap/munmap). We don't
do anything in shmat/shmdt interceptors, so if the memory becomes reused across
shmat calls in different threads, tsan may report a false positive. Accesses to
the new mapping of the memory "race" with accesses to the old mapping at the
same address.
We need shmat/shmdt interceptors to do the same as mmap/munmap interceptors.
Probably the common code to "map" and "unmap" a range can be factored out.

You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20190110/f43865ef/attachment.html>

More information about the llvm-bugs mailing list