From 37c3a2c8e6345514522e7583b89f655a24d1c6ea Mon Sep 17 00:00:00 2001 From: Thomas Schwinge Date: Fri, 30 Jul 2010 15:47:24 +0200 Subject: open_issues/glibc_libpthread_robust_mutexes: More information. --- open_issues/glibc_libpthread_robust_mutexes.mdwn | 39 ++++++++++++++++++++++++ 1 file changed, 39 insertions(+) (limited to 'open_issues') diff --git a/open_issues/glibc_libpthread_robust_mutexes.mdwn b/open_issues/glibc_libpthread_robust_mutexes.mdwn index 74b060e5..a92c984d 100644 --- a/open_issues/glibc_libpthread_robust_mutexes.mdwn +++ b/open_issues/glibc_libpthread_robust_mutexes.mdwn @@ -13,3 +13,42 @@ License|/fdl]]."]]"""]] libpthread: glibc 44e2ad5ab8c21dbfed3e384ba2ed31d7a8fc4744 998e5fc14595229101561d76282036839e5b66ab -- The robust mutex functions are in POSIX 2008. + +--- + +IRC, #hurd, unknown date. + + neal: bad news: you remember the PTHREAD_RECURSIVE_MUTEX_INITIALIZER that points to a global __pthread_recursive_mutexattr? + that doesn't work + because some libraries like libstdc++ do not link against libpthread, while still using pthread_mutex_lock/unlock (counting on them being provided by either libc or libpthread-stubs) + sthibaul-guest * r626 pkg-hurd/hurd/trunk/debian/ (changelog patches/series): + * debian/patches/libpthread_rwlock_initializer.patch: Disable patch for now: + our initializer does not work when the application does not link against + libpthread. + + sthibaul-guest * r629 pkg-hurd/hurd/trunk/debian/ (changelog patches/series): do not disable adding PTHREAD_RWLOCK_INITIALIZER, that's not the one that poses problems + sthibaul-guest * r630 pkg-hurd/hurd/trunk/debian/ (3 files in 2 dirs): + * debian/patches/libpthread_no_recursive_mutex_initializer.patch: New patch + to drop undefined references to __pthread_recursive_mutexattr. + + I'm thinking about how to fix the PTHREAD_RECURSIVE_MUTEX_INITIALIZER + instead of a pointer to a static attribute variable, which posed problem + could we perhaps consider that page 0 is never mapped + and thus not only pointer 0 but also 1 2, etc. are invalid + I think that is a good solution + and use them as special values + alternatively, we could assume that -PAGESIZE is never valid + that makes us test it in all pthread_mutex_* functions, but it's not so bad + I'm not sure which is better + why isn't it? + because the kernel is mapped there normally + the kernel could be elsewhere + true + in a 64bit adressing space for instance + I think your solution is a good one + ok + + sthibault * r633 pkg-hurd/hurd/trunk/debian/ (3 files in 2 dirs): + * debian/patches/libpthread_recursive_mutex_initializer.patch: New patch + to fix the recursive mutex initializers usage in libraries not linking + against libpthread. -- cgit v1.2.3