]> git.itanic.dy.fi Git - linux-stable/commit
afs: Use the operation issue time instead of the reply time for callbacks
authorDavid Howells <dhowells@redhat.com>
Wed, 31 Aug 2022 12:16:42 +0000 (13:16 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Thu, 15 Sep 2022 09:32:05 +0000 (11:32 +0200)
commit63677a09238a0891521132f59d14668e3722e73f
tree215f3877bfaac6a4ea906ba2d0951d19efd8d8e1
parentfbbd5d05ea63cabff8c521e645dfcc8f4f7f78be
afs: Use the operation issue time instead of the reply time for callbacks

[ Upstream commit 7903192c4b4a82d792cb0dc5e2779a2efe60d45b ]

rxrpc and kafs between them try to use the receive timestamp on the first
data packet (ie. the one with sequence number 1) as a base from which to
calculate the time at which callback promise and lock expiration occurs.

However, we don't know how long it took for the server to send us the reply
from it having completed the basic part of the operation - it might then,
for instance, have to send a bunch of a callback breaks, depending on the
particular operation.

Fix this by using the time at which the operation is issued on the client
as a base instead.  That should never be longer than the server's idea of
the expiry time.

Fixes: 781070551c26 ("afs: Fix calculation of callback expiry time")
Fixes: 2070a3e44962 ("rxrpc: Allow the reply time to be obtained on a client call")
Suggested-by: Jeffrey E Altman <jaltman@auristor.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
fs/afs/flock.c
fs/afs/fsclient.c
fs/afs/internal.h
fs/afs/rxrpc.c
fs/afs/yfsclient.c