linux.git/fs/ceph/addr.c, branch v5.12

ceph: allow queueing cap/snap handling after putting cap references

2021-02-16T11:09:51+00:00

Testing with the fscache overhaul has triggered some lockdep warnings
about circular lock dependencies involving page_mkwrite and the
mmap_lock. It'd be better to do the "real work" without the mmap lock
being held.

Change the skip_checking_caps parameter in __ceph_put_cap_refs to an
enum, and use that to determine whether to queue check_caps, do it
synchronously or not at all. Change ceph_page_mkwrite to do a
ceph_put_cap_refs_async().

Signed-off-by: Jeff Layton 
Reviewed-by: Ilya Dryomov 
Signed-off-by: Ilya Dryomov

ceph: pass down the flags to grab_cache_page_write_begin

2020-12-14T22:21:47+00:00

write_begin operations are passed a flags parameter that we need to
mirror here, so that we don't (e.g.) recurse back into filesystem code
inappropriately.

Signed-off-by: Jeff Layton 
Reviewed-by: Ilya Dryomov 
Signed-off-by: Ilya Dryomov

ceph: add new RECOVER mount_state when recovering session

2020-12-14T22:21:46+00:00

When recovering a session (a'la recover_session=clean), we want to do
all of the operations that we do on a forced umount, but changing the
mount state to SHUTDOWN is can cause queued MDS requests to fail when
the session comes back. Most of those can idle until the session is
recovered in this situation.

Reserve SHUTDOWN state for forced umount, and make a new RECOVER state
for the forced reconnect situation. Change several tests for equality with
SHUTDOWN to test for that or RECOVER.

Signed-off-by: Jeff Layton 
Reviewed-by: Xiubo Li 
Reviewed-by: "Yan, Zheng" 
Signed-off-by: Ilya Dryomov

ceph: promote to unsigned long long before shifting

2020-10-12T13:29:27+00:00

On 32-bit systems, this shift will overflow for files larger than 4GB.

Cc: stable@vger.kernel.org
Fixes: 61f68816211e ("ceph: check caps in filemap_fault and page_mkwrite")
Signed-off-by: Matthew Wilcox (Oracle) 
Reviewed-by: Jeff Layton 
Signed-off-by: Ilya Dryomov

ceph: don't SetPageError on readpage errors

2020-10-12T13:29:27+00:00

PageError really only has meaning within a particular subsystem. Nothing
looks at this bit in the core kernel code, and ceph itself doesn't care
about it. Don't bother setting the PageError bit on error.

Signed-off-by: Jeff Layton 
Reviewed-by: Matthew Wilcox (Oracle) 
Signed-off-by: Ilya Dryomov

ceph: fold ceph_update_writeable_page into ceph_write_begin

2020-10-12T13:29:27+00:00

...and reorganize the loop for better clarity.

Signed-off-by: Jeff Layton 
Signed-off-by: Ilya Dryomov

ceph: fold ceph_sync_writepages into writepage_nounlock

2020-10-12T13:29:27+00:00

Signed-off-by: Jeff Layton 
Signed-off-by: Ilya Dryomov

ceph: fold ceph_sync_readpages into ceph_readpage

2020-10-12T13:29:27+00:00

Signed-off-by: Jeff Layton 
Signed-off-by: Ilya Dryomov

ceph: don't call ceph_update_writeable_page from page_mkwrite

2020-10-12T13:29:27+00:00

page_mkwrite should only be called with Uptodate pages, so we should
only need to flush incompatible snap contexts.

Signed-off-by: Jeff Layton 
Signed-off-by: Ilya Dryomov

ceph: break out writeback of incompatible snap context to separate function

2020-10-12T13:29:27+00:00

When dirtying a page, we have to flush incompatible contexts. Move the
search for an incompatible context into a separate function, and fix up
the caller to wait and retry if there is one.

Signed-off-by: Jeff Layton 
Signed-off-by: Ilya Dryomov