linux-dev/fs/dlm, branch linus/master

dlm: use kref_put_lock in __put_lkb

2022-05-02T16:23:49Z

This patch will optimize __put_lkb() by using kref_put_lock(). The function kref_put_lock() will only take the lock if the reference is going to be zero, if not the lock will never be held. Signed-off-by: Alexander Aring Signed-off-by: David Teigland

dlm: use kref_put_lock in put_rsb

2022-05-02T16:22:56Z

This patch will optimize put_rsb() by using kref_put_lock(). The function kref_put_lock() will only take the lock if the reference is going to be zero, if not the lock will never be held. Signed-off-by: Alexander Aring Signed-off-by: David Teigland

dlm: remove unnecessary error assign

2022-05-02T16:22:15Z

This patch removes unnecessary error assigns to 0 at places we know that error is zero because it was checked on non-zero before. Signed-off-by: Alexander Aring Signed-off-by: David Teigland

dlm: fix missing lkb refcount handling

2022-05-02T16:15:59Z

We always call hold_lkb(lkb) if we increment lkb->lkb_wait_count. So, we always need to call unhold_lkb(lkb) if we decrement lkb->lkb_wait_count. This patch will add missing unhold_lkb(lkb) if we decrement lkb->lkb_wait_count. In case of setting lkb->lkb_wait_count to zero we need to countdown until reaching zero and call unhold_lkb(lkb). The waiters list unhold_lkb(lkb) can be removed because it's done for the last lkb_wait_count decrement iteration as it's done in _remove_from_waiters(). This issue was discovered by a dlm gfs2 test case which use excessively dlm_unlock(LKF_CANCEL) feature. Probably the lkb->lkb_wait_count value never reached above 1 if this feature isn't used and so it was not discovered before. The testcase ended in a rsb on the rsb keep data structure with a refcount of 1 but no lkb was associated with it, which is itself an invalid behaviour. A side effect of that was a condition in which the dlm was sending remove messages in a looping behaviour. With this patch that has not been reproduced. Cc: stable@vger.kernel.org Signed-off-by: Alexander Aring Signed-off-by: David Teigland

fs: dlm: cast resource pointer to uintptr_t

2022-04-07T14:54:45Z

This patch fixes the following warning when doing a 32 bit kernel build when pointers are 4 byte long: In file included from ./include/linux/byteorder/little_endian.h:5, from ./arch/x86/include/uapi/asm/byteorder.h:5, from ./include/asm-generic/qrwlock_types.h:6, from ./arch/x86/include/asm/spinlock_types.h:7, from ./include/linux/spinlock_types_raw.h:7, from ./include/linux/ratelimit_types.h:7, from ./include/linux/printk.h:10, from ./include/asm-generic/bug.h:22, from ./arch/x86/include/asm/bug.h:87, from ./include/linux/bug.h:5, from ./include/linux/mmdebug.h:5, from ./include/linux/gfp.h:5, from ./include/linux/slab.h:15, from fs/dlm/dlm_internal.h:19, from fs/dlm/rcom.c:12: fs/dlm/rcom.c: In function ‘dlm_send_rcom_lock’: ./include/uapi/linux/byteorder/little_endian.h:32:43: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast] #define __cpu_to_le64(x) ((__force __le64)(__u64)(x)) ^ ./include/linux/byteorder/generic.h:86:21: note: in expansion of macro ‘__cpu_to_le64’ #define cpu_to_le64 __cpu_to_le64 ^~~~~~~~~~~~~ fs/dlm/rcom.c:457:14: note: in expansion of macro ‘cpu_to_le64’ rc->rc_id = cpu_to_le64(r); The rc_id value in dlm rcom is handled as u64. The rcom implementation uses for an unique number generation the pointer value of the used dlm_rsb instance. However if the pointer value is 4 bytes long -Wpointer-to-int-cast will print a warning. We get rid of that warning to cast the pointer to uintptr_t which is either 4 or 8 bytes. There might be a very unlikely case where this number isn't unique anymore if using dlm in a mixed cluster of nodes and sizeof(uintptr_t) returns 4 and 8. However this problem was already been there and this patch should get rid of the warning. Fixes: 2f9dbeda8dc0 ("dlm: use __le types for rcom messages") Reported-by: kernel test robot Signed-off-by: Alexander Aring Signed-off-by: David Teigland

dlm: replace usage of found with dedicated list iterator variable

2022-04-06T19:03:14Z

To move the list iterator variable into the list_for_each_entry_*() macro in the future it should be avoided to use the list iterator variable after the loop body. To *never* use the list iterator variable after the loop it was concluded to use a separate iterator variable instead of a found boolean [1]. This removes the need to use a found variable and simply checking if the variable was set, can determine if the break/goto was hit. Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1] Signed-off-by: Jakob Koschel Signed-off-by: Alexander Aring Signed-off-by: David Teigland

dlm: remove usage of list iterator for list_add() after the loop body

2022-04-06T19:03:13Z

In preparation to limit the scope of a list iterator to the list traversal loop, use a dedicated pointer to point to the found element [1]. Before, the code implicitly used the head when no element was found when using &pos->list. Since the new variable is only set if an element was found, the list_add() is performed within the loop and only done after the loop if it is done on the list head directly. Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1] Signed-off-by: Jakob Koschel Signed-off-by: Alexander Aring Signed-off-by: David Teigland

dlm: fix pending remove if msg allocation fails

2022-04-06T19:03:09Z

This patch unsets ls_remove_len and ls_remove_name if a message allocation of a remove messages fails. In this case we never send a remove message out but set the per ls ls_remove_len ls_remove_name variable for a pending remove. Unset those variable should indicate possible waiters in wait_pending_remove() that no pending remove is going on at this moment. Cc: stable@vger.kernel.org Signed-off-by: Alexander Aring Signed-off-by: David Teigland

dlm: fix wake_up() calls for pending remove

2022-04-06T19:03:05Z

This patch move the wake_up() call at the point when a remove message completed. Before it was only when a remove message was going to be sent. The possible waiter in wait_pending_remove() waits until a remove is done if the resource name matches with the per ls variable ls->ls_remove_name. If this is the case we must wait until a pending remove is done which is indicated if DLM_WAIT_PENDING_COND() returns false which will always be the case when ls_remove_len and ls_remove_name are unset to indicate that a remove is not going on anymore. Fixes: 21d9ac1a5376 ("fs: dlm: use event based wait for pending remove") Cc: stable@vger.kernel.org Signed-off-by: Alexander Aring Signed-off-by: David Teigland

dlm: check required context while close

2022-04-06T19:03:01Z

This patch adds a WARN_ON() check to validate the right context while dlm_midcomms_close() is called. Even before commit 489d8e559c65 ("fs: dlm: add reliable connection if reconnect") in this context dlm_lowcomms_close() flushes all ongoing transmission triggered by dlm application stack. If we do that, it's required that no new message will be triggered by the dlm application stack. The function dlm_midcomms_close() is not called often so we can check if all lockspaces are in such context. Signed-off-by: Alexander Aring Signed-off-by: David Teigland