linux-stable.git/include, branch v3.2.57

net: ip, ipv6: handle gso skbs in forwarding path

2014-04-09T01:20:45+00:00

commit fe6cc55f3a9a053482a76f5a6b2257cee51b4663 upstream.

[ use zero netdev_feature mask to avoid backport of
  netif_skb_dev_features function ]

Marcelo Ricardo Leitner reported problems when the forwarding link path
has a lower mtu than the incoming one if the inbound interface supports GRO.

Given:
Host  R1  R2

Host sends tcp stream which is routed via R1 and R2.  R1 performs GRO.

In this case, the kernel will fail to send ICMP fragmentation needed
messages (or pkt too big for ipv6), as GSO packets currently bypass dstmtu
checks in forward path. Instead, Linux tries to send out packets exceeding
the mtu.

When locking route MTU on Host (i.e., no ipv4 DF bit set), R1 does
not fragment the packets when forwarding, and again tries to send out
packets exceeding R1-R2 link mtu.

This alters the forwarding dstmtu checks to take the individual gso
segment lengths into account.

For ipv6, we send out pkt too big error for gso if the individual
segments are too big.

For ipv4, we either send icmp fragmentation needed, or, if the DF bit
is not set, perform software segmentation and let the output path
create fragments when the packet is leaving the machine.
It is not 100% correct as the error message will contain the headers of
the GRO skb instead of the original/segmented one, but it seems to
work fine in my (limited) tests.

Eric Dumazet suggested to simply shrink mss via ->gso_size to avoid
sofware segmentation.

However it turns out that skb_segment() assumes skb nr_frags is related
to mss size so we would BUG there.  I don't want to mess with it considering
Herbert and Eric disagree on what the correct behavior should be.

Hannes Frederic Sowa notes that when we would shrink gso_size
skb_segment would then also need to deal with the case where
SKB_MAX_FRAGS would be exceeded.

This uses sofware segmentation in the forward path when we hit ipv4
non-DF packets and the outgoing link mtu is too small.  Its not perfect,
but given the lack of bug reports wrt. GRO fwd being broken this is a
rare case anyway.  Also its not like this could not be improved later
once the dust settles.

Acked-by: Herbert Xu 
Reported-by: Marcelo Ricardo Leitner 
Signed-off-by: Florian Westphal 
Signed-off-by: David S. Miller 
Signed-off-by: Ben Hutchings

net: add and use skb_gso_transport_seglen()

2014-04-09T01:20:45+00:00

commit de960aa9ab4decc3304959f69533eef64d05d8e8 upstream.

[ no skb_gso_seglen helper in 3.4, leave tbf alone ]

This moves part of Eric Dumazets skb_gso_seglen helper from tbf sched to
skbuff core so it may be reused by upcoming ip forwarding path patch.

Signed-off-by: Florian Westphal 
Acked-by: Eric Dumazet 
Signed-off-by: David S. Miller 
Signed-off-by: Ben Hutchings

ext4: atomically set inode->i_flags in ext4_set_inode_flags()

2014-04-09T01:20:44+00:00

commit 00a1a053ebe5febcfc2ec498bd894f035ad2aa06 upstream.

Use cmpxchg() to atomically set i_flags instead of clearing out the
S_IMMUTABLE, S_APPEND, etc. flags and then setting them from the
EXT4_IMMUTABLE_FL, EXT4_APPEND_FL flags, since this opens up a race
where an immutable file has the immutable flag cleared for a brief
window of time.

Reported-by: John Sullivan 
Signed-off-by: "Theodore Ts'o" 
Signed-off-by: Linus Torvalds 
[bwh: Backported to 3.2: adjust context]
Signed-off-by: Ben Hutchings

jiffies: Avoid undefined behavior from signed overflow

2014-04-01T23:59:01+00:00

commit 5a581b367b5df0531265311fc681c2abd377e5e6 upstream.

According to the C standard 3.4.3p3, overflow of a signed integer results
in undefined behavior.  This commit therefore changes the definitions
of time_after(), time_after_eq(), time_after64(), and time_after_eq64()
to avoid this undefined behavior.  The trick is that the subtraction
is done using unsigned arithmetic, which according to 6.2.5p9 cannot
overflow because it is defined as modulo arithmetic.  This has the added
(though admittedly quite small) benefit of shortening four lines of code
by four characters each.

Note that the C standard considers the cast from unsigned to
signed to be implementation-defined, see 6.3.1.3p3.  However, on a
two's-complement system, an implementation that defines anything other
than a reinterpretation of the bits is free to come to me, and I will be
happy to act as a witness for its being committed to an insane asylum.
(Although I have nothing against saturating arithmetic or signals in some
cases, these things really should not be the default when compiling an
operating-system kernel.)

Signed-off-by: Paul E. McKenney 
Cc: John Stultz 
Cc: "David S. Miller" 
Cc: Arnd Bergmann 
Cc: Ingo Molnar 
Cc: Linus Torvalds 
Cc: Eric Dumazet 
Cc: Kevin Easton 
[ paulmck: Included time_after64() and time_after_eq64(), as suggested
  by Eric Dumazet, also fixed commit message.]
Reviewed-by: Josh Triplett 
Signed-off-by: Ben Hutchings

tracing: Do not add event files for modules that fail tracepoints

2014-04-01T23:58:57+00:00

commit 45ab2813d40d88fc575e753c38478de242d03f88 upstream.

If a module fails to add its tracepoints due to module tainting, do not
create the module event infrastructure in the debugfs directory. As the events
will not work and worse yet, they will silently fail, making the user wonder
why the events they enable do not display anything.

Having a warning on module load and the events not visible to the users
will make the cause of the problem much clearer.

Link: http://lkml.kernel.org/r/20140227154923.265882695@goodmis.org

Fixes: 6d723736e472 "tracing/events: add support for modules to TRACE_EVENT"
Acked-by: Mathieu Desnoyers 
Cc: Rusty Russell 
Signed-off-by: Steven Rostedt 
Signed-off-by: Ben Hutchings

compiler/gcc4: Make quirk for asm_volatile_goto() unconditional

2014-04-01T23:58:51+00:00

commit a9f180345f5378ac87d80ed0bea55ba421d83859 upstream.

I started noticing problems with KVM guest destruction on Linux
3.12+, where guest memory wasn't being cleaned up. I bisected it
down to the commit introducing the new 'asm goto'-based atomics,
and found this quirk was later applied to those.

Unfortunately, even with GCC 4.8.2 (which ostensibly fixed the
known 'asm goto' bug) I am still getting some kind of
miscompilation. If I enable the asm_volatile_goto quirk for my
compiler, KVM guests are destroyed correctly and the memory is
cleaned up.

So make the quirk unconditional for now, until bug is found
and fixed.

Suggested-by: Linus Torvalds 
Signed-off-by: Steven Noonan 
Cc: Peter Zijlstra 
Cc: Steven Rostedt 
Cc: Jakub Jelinek 
Cc: Richard Henderson 
Cc: Andrew Morton 
Cc: Oleg Nesterov 
Link: http://lkml.kernel.org/r/1392274867-15236-1-git-send-email-steven@uplinklabs.net
Link: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58670
Signed-off-by: Ingo Molnar 
Signed-off-by: Ben Hutchings

xen: install xen/gntdev.h and xen/gntalloc.h

2014-04-01T23:58:50+00:00

commit 564eb714f5f09ac733c26860d5f0831f213fbdf1 upstream.

xen/gntdev.h and xen/gntalloc.h both provide userspace ABIs so they
should be installed.

Signed-off-by: David Vrabel 
Signed-off-by: Konrad Rzeszutek Wilk 
[bwh: Backported to 3.2: no renaming is required]
Signed-off-by: Ben Hutchings

ore: Fix wrong math in allocation of per device BIO

2014-04-01T23:58:45+00:00

commit aad560b7f63b495f48a7232fd086c5913a676e6f upstream.

At IO preparation we calculate the max pages at each device and
allocate a BIO per device of that size. The calculation was wrong
on some unaligned corner cases offset/length combination and would
make prepare return with -ENOMEM. This would be bad for pnfs-objects
that would in that case IO through MDS. And fatal for exofs were it
would fail writes with EIO.

Fix it by doing the proper math, that will work in all cases. (I
ran a test with all possible offset/length combinations this time
round).

Also when reading we do not need to allocate for the parity units
since we jump over them.

Also lower the max_io_length to take into account the parity pages
so not to allocate BIOs bigger than PAGE_SIZE

Signed-off-by: Boaz Harrosh 
Signed-off-by: Ben Hutchings

fuse: fix pipe_buf_operations

2014-04-01T23:58:44+00:00

commit 28a625cbc2a14f17b83e47ef907b2658576a32aa upstream.

Having this struct in module memory could Oops when if the module is
unloaded while the buffer still persists in a pipe.

Since sock_pipe_buf_ops is essentially the same as fuse_dev_pipe_buf_steal
merge them into nosteal_pipe_buf_ops (this is the same as
default_pipe_buf_ops except stealing the page from the buffer is not
allowed).

Reported-by: Al Viro 
Signed-off-by: Miklos Szeredi 
[bwh: Backported to 3.2: adjust context]
Signed-off-by: Ben Hutchings

libata: disable LPM for some WD SATA-I devices

2014-04-01T23:58:43+00:00

commit ecd75ad514d73efc1bbcc5f10a13566c3ace5f53 upstream.

For some reason, some early WD drives spin up and down drives
erratically when the link is put into slumber mode which can reduce
the life expectancy of the device significantly.  Unfortunately, we
don't have full list of devices and given the nature of the issue it'd
be better to err on the side of false positives than the other way
around.  Let's disable LPM on all WD devices which match one of the
known problematic model prefixes and are SATA-I.

As horkage list doesn't support matching SATA capabilities, this is
implemented as two horkages - WD_BROKEN_LPM and NOLPM.  The former is
set for the known prefixes and sets the latter if the matched device
is SATA-I.

Note that this isn't optimal as this disables all LPM operations and
partial link power state reportedly works fine on these; however, the
way LPM is implemented in libata makes it difficult to precisely map
libata LPM setting to specific link power state.  Well, these devices
are already fairly outdated.  Let's just disable whole LPM for now.

Signed-off-by: Tejun Heo 
Reported-and-tested-by: Nikos Barkas 
Reported-and-tested-by: Ioannis Barkas 
References: https://bugzilla.kernel.org/show_bug.cgi?id=57211
[bwh: Backported to 3.2:
 - Adjust context
 - Use literal 76 instead of ATA_ID_SATA_CAPABILITY]
Signed-off-by: Ben Hutchings