linux-toradex.git/fs, branch v3.0.30

tcp: allow splice() to build full TSO packets

2012-04-27T16:51:18+00:00

[ This combines upstream commit
  2f53384424251c06038ae612e56231b96ab610ee and the follow-on bug fix
  commit 35f9c09fe9c72eb8ca2b8e89a593e1c151f28fc2 ]

vmsplice()/splice(pipe, socket) call do_tcp_sendpages() one page at a
time, adding at most 4096 bytes to an skb. (assuming PAGE_SIZE=4096)

The call to tcp_push() at the end of do_tcp_sendpages() forces an
immediate xmit when pipe is not already filled, and tso_fragment() try
to split these skb to MSS multiples.

4096 bytes are usually split in a skb with 2 MSS, and a remaining
sub-mss skb (assuming MTU=1500)

This makes slow start suboptimal because many small frames are sent to
qdisc/driver layers instead of big ones (constrained by cwnd and packets
in flight of course)

In fact, applications using sendmsg() (adding an additional memory copy)
instead of vmsplice()/splice()/sendfile() are a bit faster because of
this anomaly, especially if serving small files in environments with
large initial [c]wnd.

Call tcp_push() only if MSG_MORE is not set in the flags parameter.

This bit is automatically provided by splice() internals but for the
last page, or on all pages if user specified SPLICE_F_MORE splice()
flag.

In some workloads, this can reduce number of sent logical packets by an
order of magnitude, making zero-copy TCP actually faster than
one-copy :)

Reported-by: Tom Herbert 
Cc: Nandita Dukkipati 
Cc: Neal Cardwell 
Cc: Tom Herbert 
Cc: Yuchung Cheng 
Cc: H.K. Jerry Chu 
Cc: Maciej Żenczykowski 
Cc: Mahesh Bandewar 
Cc: Ilpo Järvinen 
Signed-off-by: Eric Dumazet 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

lockd: fix the endianness bug

2012-04-27T16:51:18+00:00

commit e847469bf77a1d339274074ed068d461f0c872bc upstream.

comparing be32 values for < is not doing the right thing...

Signed-off-by: Al Viro 
Cc: "J. Bruce Fields" 
Cc: Trond Myklebust 
Signed-off-by: Greg Kroah-Hartman

ocfs2: ->e_leaf_clusters endianness breakage

2012-04-27T16:51:18+00:00

commit 72094e43e3af5020510f920321d71f1798fa896d upstream.

le16, not le32...

Signed-off-by: Al Viro 
Cc: Mark Fasheh 
Cc: Joel Becker 
Signed-off-by: Greg Kroah-Hartman

ocfs2: ->rl_count endianness breakage

2012-04-27T16:51:18+00:00

commit 28748b325dc2d730ccc312830a91c4ae0c0d9379 upstream.

le16, not le32...

Signed-off-by: Al Viro 
Cc: Mark Fasheh 
Cc: Joel Becker 
Signed-off-by: Greg Kroah-Hartman

ocfs: ->rl_used breakage on big-endian

2012-04-27T16:51:17+00:00

commit e1bf4cc620fd143766ddfcee3b004a1d1bb34fd0 upstream.

it's le16, not le32 or le64...

Signed-off-by: Al Viro 
Cc: Mark Fasheh 
Cc: Joel Becker 
Signed-off-by: Greg Kroah-Hartman

ocfs2: ->l_next_free_req breakage on big-endian

2012-04-27T16:51:17+00:00

commit 3a251f04fe97c3d335b745c98e4b377e3c3899f2 upstream.

It's le16, not le32...

Signed-off-by: Al Viro 
Cc: Mark Fasheh 
Cc: Joel Becker 
Signed-off-by: Greg Kroah-Hartman

btrfs: btrfs_root_readonly() broken on big-endian

2012-04-27T16:51:17+00:00

commit 6ed3cf2cdfce4c9f1d73171bd3f27d9cb77b734e upstream.

->root_flags is __le64 and all accesses to it go through the helpers
that do proper conversions.  Except for btrfs_root_readonly(), which
checks bit 0 as in host-endian...

Signed-off-by: Al Viro 
Cc: Chris Mason 
Signed-off-by: Greg Kroah-Hartman

nfsd: fix compose_entry_fh() failure exits

2012-04-27T16:51:17+00:00

commit efe39651f08813180f37dc508d950fc7d92b29a8 upstream.

Restore the original logics ("fail on mountpoints, negatives and in
case of fh_compose() failures").  Since commit 8177e (nfsd: clean up
readdirplus encoding) that got broken -
	rv = fh_compose(fhp, exp, dchild, &cd->fh);
	if (rv)
	       goto out;
	if (!dchild->d_inode)
		goto out;
	rv = 0;
out:
is equivalent to
	rv = fh_compose(fhp, exp, dchild, &cd->fh);
out:
and the second check has no effect whatsoever...

Signed-off-by: Al Viro 
Cc: "J. Bruce Fields" 
Signed-off-by: Greg Kroah-Hartman

Don't limit non-nested epoll paths

2012-04-27T16:51:09+00:00

commit 93dc6107a76daed81c07f50215fa6ae77691634f upstream.

Commit 28d82dc1c4ed ("epoll: limit paths") that I did to limit the
number of possible wakeup paths in epoll is causing a few applications
to longer work (dovecot for one).

The original patch is really about limiting the amount of epoll nesting
(since epoll fds can be attached to other fds). Thus, we probably can
allow an unlimited number of paths of depth 1. My current patch limits
it at 1000. And enforce the limits on paths that have a greater depth.

This is captured in: https://bugzilla.redhat.com/show_bug.cgi?id=681578

Signed-off-by: Jason Baron 
Cc: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

ext4: fix endianness breakage in ext4_split_extent_at()

2012-04-27T16:51:09+00:00

commit af1584f570b19b0285e4402a0b54731495d31784 upstream.

->ee_len is __le16, so assigning cpu_to_le32() to it is going to do
Bad Things(tm) on big-endian hosts...

Signed-off-by: Al Viro 
Cc: Ted Ts'o 
Signed-off-by: Greg Kroah-Hartman