【ceph研发】ceph+nfs-ganesha异步读写git历史记录
ceph+nfs-ganesha异步读写git历史记录
1. client: unlock client_lock when copying data and do more check for the client_lock
2. Client: C_Read_Async_Finisher should assume client_lock is held
4. client: handle callback when async io fails #54690 [2023年11月28日]
this refinement gets rid of the rw lock in the I/O path and provides a bit
more fairness to open/upgrade/downgrade/close and uses dec_and_lock to
better prevent spurious wakeup due to a race that is not habdled above
increment io_work.
if fd_work != 0
lock mutex
if fd_work != 0
// fd work is waiting, block until it's done
if decrement io_work == 0
signal condition var
while fd_work != 0
wait on condition var
// we are done waiting for fd work, so resume I/O
increment io_work
unlock mutex
// now we know no fd work is waiting or in progress and can't start
initiate I/O
on I/O complete (async or immediate):
-------------------------------------
if decrement_and_lock io_work
if fd_work != 0
signal condition var
unlock mutex
open/upgrade/downgrade/close:
-----------------------------
increment fd_work
lock mutex
while io_work != 0
// now fd_work is non-zero and io_work is zero, any I/O that
// tries to start will block until fd_work goes to zero
wait on cond var
// io_work was zero, and because we checked while holding the mutex, but made
// fd_work non-zero before taking the mutex any I/O threads that try to start
// after we took the mutex MUST be blocked, proceed
unlock mutex
// serialize any open/upgrade/downgrade/close
take write lock
re-open fd in new mode or close it
release write lock
// now ready to allow I/O to resume
if decrement_and_lock fd_work
if fd_work == 0
signal cond var
unlock mutex
