- 03 Jan, 2017 1 commit
-
-
Max Kellermann authored
-
- 28 Dec, 2016 1 commit
-
-
Max Kellermann authored
-
- 16 Sep, 2016 1 commit
-
-
Max Kellermann authored
-
- 26 Feb, 2016 1 commit
-
-
Max Kellermann authored
-
- 01 Jan, 2015 1 commit
-
-
Max Kellermann authored
-
- 14 Dec, 2014 13 commits
-
-
Max Kellermann authored
SocketMonitor::Cancel() does not actually unregister the socket; it only disables the event.
-
Max Kellermann authored
-
Max Kellermann authored
-
Max Kellermann authored
-
Max Kellermann authored
When mounting had not yet finished, SocketMonitor::IsDefined() was always false, due to the workaround at the beginning of the function that calls SocketMonitor::Steal(). This commit drops the IsDefined() check because it was never necessary and breaks reconnect.
-
Max Kellermann authored
-
Max Kellermann authored
nfs_destroy_context() will invoke all pending callbacks with err==-EINTR. In CancellableCallback::Callback(), this will invoke NfsConnection::DeferClose(), which however is only designed to be called from nfs_service(). In non-debug mode, this will leak memory because nfs_close_async() is never called. Workaround: before nfs_destroy_context(), invoke nfs_close_async() on all pending file handles.
-
Max Kellermann authored
-
Max Kellermann authored
-
Max Kellermann authored
-
Max Kellermann authored
-
Max Kellermann authored
The method NfsConnection::CancellableCallback::Callback() will always invoke NfsConnection::Close() on the file handle, even if the void pointer is not a nfsfh. This can happen if the Open() was not successful, e.g. when the file does not exist.
-
Max Kellermann authored
-
- 25 Nov, 2014 2 commits
-
-
Max Kellermann authored
During the NfsLease::OnNfsConnectionFailed() call, the old (defunct) nfs_context may be used to close file handles. Such code does not yet exist, but will be added soon to fix other bugs.
-
Max Kellermann authored
Fixes potential second mount attempt after the old connection to the NFS server was shut down.
-
- 09 Oct, 2014 1 commit
-
-
Max Kellermann authored
-
- 07 Oct, 2014 2 commits
-
-
Max Kellermann authored
-
Max Kellermann authored
-
- 01 Oct, 2014 7 commits
-
-
Max Kellermann authored
If an async opertion is in progress, nfs_close_async() will make libnfs crash because the RPC callback will dereference an object that was freed by nfs_close_async().
-
Max Kellermann authored
Close the newly allocated file handle passed to the callback.
-
Max Kellermann authored
All locks are currenly held from only a single thread (the IOThread) and thus we don't need the Mutex.
-
Max Kellermann authored
-
Max Kellermann authored
-
Max Kellermann authored
This allows eliminating the indirection code from the NfsConnection destructor.
-
Max Kellermann authored
Reduce Remove() overhead because we don't have to walk the list to find an iterator by reference.
-
- 31 Aug, 2014 5 commits
-
-
Max Kellermann authored
-
Max Kellermann authored
Since BroadcastError() calls BroadcastMountError(), there's no need to check mount_finished here.
-
Max Kellermann authored
While reconnecting to the NFS server, the socket is -1, because there is no socket. Fixes a potential crash (assertion failure).
-
Max Kellermann authored
-
Max Kellermann authored
-
- 30 Jun, 2014 1 commit
-
-
Max Kellermann authored
-
- 17 Jun, 2014 1 commit
-
-
Max Kellermann authored
More robust and cancellable.
-