Hi David,
read_rfc1002_done:
+ /* SMBDirect will read it all or nothing */
+ msg->msg_iter.count = 0;
And this iov_iter_truncate(0);
Actually, it should probably have been iov_iter_advance().
While I'm wondering why we had this at all.
It seems all callers of cifs_read_iter_from_socket()
don't care and the code path via sock_recvmsg() doesn't
truncate it just calls copy_to_iter() via this chain:
->inet_recvmsg->tcp_recvmsg->skb_copy_datagram_msg->skb_copy_datagram_iter
->simple_copy_to_iter->copy_to_iter()
I think the old code should have called
iov_iter_advance(rc) instead of msg->msg_iter.count = 0.
But the new code doesn't need it as copy_to_iter()
calls iterate_and_advance().
Yeah, it should. I seem to remember that there were situations in which it
didn't, but it's possible I managed to get rid of them.
- default:
- /* It's a bug in upper layer to get there */
- cifs_dbg(VFS, "Invalid msg type %d\n",
- iov_iter_type(&msg->msg_iter));
- rc = -EINVAL;
- }
I guess this is actually a real fix as I just saw
CIFS: VFS: Invalid msg type 4
in logs while running the cifs/001 test.
And 4 is ITER_FOLIOQ.
Ah... Were you using "-o seal"? The encrypted data is held in a buffer formed
from a folioq with a series of folios in it.
In local.config I have this:
[smb3-1-rdma]
FSTYP=cifs
TEST_DEV=//172.31.9.1/TEST
TEST_DIR=/mnt/test
TEST_FS_MOUNT_OPTS='-ousername=administrator,password=...,rdma,noperm,vers=3.0,mfsymlinks,actimeo=0'
export MOUNT_OPTIONS='-ousername=administrator,password=...,rdma,noperm,mfsymlinks,actimeo=0'
export SCRATCH_DEV=//172.31.9.1/SCRATCH
export SCRATCH_MNT=/mnt/scratch
And called:
./check -s smb3-1-rdma -E tests/cifs/exclude.incompatible-smb3.txt -E tests/cifs/exclude.very-slow.txt cifs/001
So I don't think it used seal, I'm also not seeing encrypted stuff in the capture,
so maybe the folioq iter comes from a higher layer (via cifs_readv_receive()) instead
of receive_encrypted_read().
metze
