Move ib_umad module to using cdev_alloc() and class_device_create() so
that we can handle device lifetime properly.  Now we can make sure we
keep all of our data structures around until the last way to reach
them is gone.
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Change the way IPoIB handles RX packets when it can't allocate a new
receive skbuff.  If the allocation of a new receive skb fails, we now
drop the packet we just received and repost the original receive skb.
This means that the receive ring always stays full and we don't have
to monkey around with trying to schedule a refill task for later.
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Simplify user_mad.c code in a few places, and convert from kmalloc() +
memset() to kzalloc().  This also fixes a theoretical race window by
not accessing packet->length after posting the send buffer (the send
could complete and packet could be freed before we get to the return
statement at the end of ib_umad_write()).
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

The conversion of user_mad.c to the new MAD send API was slightly off:
in a few places, we used packet->msg instead of packet->msg->mad when
referring to the actual data buffer, which ended up corrupting the
underlying data structure and crashing when we free an invalid pointer.
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Add some initial support for detecting and reporting catastrophic
errors reported by Mellanox HCAs.  We start a periodic timer which
polls the catastrophic error reporting buffer in device memory.  If an
error is detected, we dump the contents of the buffer for port-mortem
debugging, and report a fatal asynchronous error to higher levels.

In the future we can try to recover from these errors by resetting the
device, but this will require some work in higher-level code as well.
Let's get this in now, so that we at least get catastrophic errors
reported in logs.
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Change alloc_response_msg() in mad_rmpp.c to return the struct
it allocates directly (or an error code a la ERR_PTR), rather than
returning a status and passing the struct back in a pointer param.
This simplifies the code and gets rid of warnings like

    drivers/infiniband/core/mad_rmpp.c: In function nack_recv:
    drivers/infiniband/core/mad_rmpp.c:192: warning: msg may be used uninitialized in this function

with newer versions of gcc.
Signed-off-by: Roland Dreier <rolandd@cisco.com>

The UC transport does not support RDMA reads or atomic operations, so
we shouldn't require or even allow the consumer to set attributes
relating to these operations for UC QPs.
Signed-off-by: Roland Dreier <rolandd@cisco.com>

The MAD layer was violating the DMA API by touching data buffers used
for sends after the DMA mapping was done.  This causes problems on
non-cache-coherent architectures, because the device doing DMA won't
see updates to the payload buffers that exist only in the CPU cache.

Fix this by having all MAD consumers use ib_create_send_mad() to
allocate their send buffers, and moving the DMA mapping into the MAD
layer so it can be done just before calling send (and after any
modifications of the send buffer by the MAD layer).

Tested on a non-cache-coherent PowerPC 440SPe system.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Fix cm_init_qp_init_attr(), cm_init_qp_rtr_attr() and cm_init_qp_rts_attr()
so that they correctly handle the differences between UC and RC QPs.  This
fixes problems with setting up UC QPs through the CM.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Add idr_destroy() calls to the module_exit() functions of the four IB
driver modules that use idrs, so we don't leak idr_layer_cache objects
when these modules are unloaded.
Signed-off-by: Roland Dreier <rolandd@cisco.com>

We should always re-arm an event queue's interrupt in
mthca_tavor_interrupt() if the corresponding bit is set in the event cause
register (ECR), even if we didn't find any entries in the EQ.  If we don't,
then there's a window where we miss an EQ entry and then get stuck because
we don't get another EQ event.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Fix a bug which was reported and diagnosed by
Stefan Jones <stefan.jones@churchillrandoms.co.uk>

IDR trees include a cache of idr_layer objects.  There's no way to destroy
this cache, so when we discard an overall idr tree we end up leaking some
memory.

Add and use idr_destroy() for this.  v9fs and infiniband also need to use
idr_destroy() to avoid leaks.

Or, we make the cache global, like radix_tree_preload().  Which is probably
better.  Later.

Cc: Eric Van Hensbergen <ericvh@ericvh.myip.org>
Cc: Roland Dreier <rolandd@cisco.com>
Cc: Robert Love <rml@novell.com>
Cc: John McCutchan <ttb@tentacle.dhs.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

On 2005-05-01, Gerd Knorr sent in a patch to add cx22702 to cx88-dvb:

 [PATCH] dvb: cx22702 frontend driver update
http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=9990d744bea7d28e83c420e2c9d524c7a8a2d136

...but as we can see, the Kconfig portion of his patch was incorrectly
applied to saa7134-dvb instead of cx88-dvb.

On 2005-06-24, Adrian bunk fixed cx88-dvb:

 [PATCH] VIDEO_CX88_DVB must select DVB_CX22702
http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=d6988588e13616587aa879c2e0bd7cd811705e5d

...but we never removed the original patch from Gerd.

This patch sets things straight:

saa7134-dvb should not select cx22702
Signed-off-by: Michael Krufky <mkrufky@m1k.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This patch fixes error handling in sel_make_bools(), where currently we'd
get a memory leak via security_get_bools() and try to kfree() the wrong
pointer if called again.
Signed-off-by: James Morris <jmorris@namei.org>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This patch fixes a possible NULL dereference in policydb_destroy, where
p->type_attr_map can be NULL if policydb_destroy is called to clean up a
partially loaded policy upon an error during policy load.  Please apply.
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Another case of missing call to security_file_permission: aio functions
(namely, io_submit) does not check credentials with security modules.

Below is the simple patch to the problem.  It seems that it is enough to
check for rights at the request submission time.
Signed-off-by: Kostik Belousov <kostikbel@gmail.com>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Fix typos & trailing whitespace.
Add blank lines in a few places.
Remove "AM53C974=" option:  driver does not exist.
Restrict to < 80 columns in most places (but don't split formatted
  command-line arguments).
Add a few option arguments for completeness.
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

That's what we've always historically done, and bigger windows seem to
confuse some cardbus bridges. Or something.

Alan reports that this makes the ThinkPad 600x series work properly
again: the 4kB IO window for some reason made IDE DMA not work, which
makes IDE painfully slow even if it works after DMA timeouts.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Bursty timers aren't good for anybody, very much including latency for
other programs when we trigger lots of timers in interrupt context.  So
set a random limit, after which we'll handle the rest on the next timer
tick.

Noted by Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

neigh_changeaddr attempts to delete neighbour timers without setting
nud_state.  This doesn't work because the timer may have already fired
when we acquire the write lock in neigh_changeaddr.  The result is that
the timer may keep firing for quite a while until the entry reaches
NEIGH_FAILED.

It should be setting the nud_state straight away so that if the timer
has already fired it can simply exit once we relinquish the lock.

In fact, this whole function is simply duplicating the logic in
neigh_ifdown which in turn is already doing the right thing when
it comes to deleting timers and setting nud_state.

So all we have to do is take that code out and put it into a common
function and make both neigh_changeaddr and neigh_ifdown call it.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

neigh_add_timer cannot use add_timer unconditionally.  The reason is that
by the time it has obtained the write lock someone else (e.g., neigh_update)
could have already added a new timer.

So it should only use mod_timer and deal with its return value accordingly.

This bug would have led to rare neighbour cache entry leaks.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

Stack traces are very helpful in determining the exact nature of a bug.
So let's print a stack trace when the timer is added twice.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

As stated in Documentation/atomic_ops.txt, atomic functions
returning values must have the memory barriers both before and after
the operation.

Thanks to DaveM for pointing that out.
Signed-off-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

On architectures where the char type defaults to unsigned some of the
arithmetic in the AX.25 stack to fail, resulting in some packets being dropped
on receive.

Credits for tracking this down and the original patch to
Bob Brose N0QBJ <linuxhams@n0qbj-11.ampr.org>.
Signed-off-by: Ralf Baechle DL5RB <ralf@linux-mips.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

IPVS used flag NFC_IPVS_PROPERTY in nfcache but as now nfcache was removed the
new flag 'ipvs_property' still needs to be copied. This patch should be
included in 2.6.14.

Further comments from Harald Welte:

Sorry, seems like the bug was introduced by me.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>

We should always re-arm an event queue's interrupt in
mthca_tavor_interrupt() if the corresponding bit is set in the event
cause register (ECR), even if we didn't find any entries in the EQ.
If we don't, then there's a window where we miss an EQ entry and then
get stuck because we don't get another EQ event.
Signed-off-by: Roland Dreier <rolandd@cisco.com>

Not sure how it slipped by, but here's a trivial typo fix for powernow.
Signed-off-by: Chris Wright <chrisw@osdl.org>
[ It's "nurter" backwards.. Maybe we have a hillbilly The Shining fan? ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

When I originally moved exit_itimers into __exit_signal, that was the only
place where we could reliably know it was the last thread in the group
dying, without races.  Since then we've gotten the signal_struct.live
counter, and do_exit can reliably do group-wide cleanup work.

This patch moves the call to do_exit, where it's made without locks.  This
avoids the deadlock issues that the old __exit_signal code's comment talks
about, and the one that Oleg found recently with process CPU timers.

[ This replaces e03d13e9, which is why
  it was just reverted. ]
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Revert commit e03d13e9, to be replaced
by a much nicer fix from Roland.

AMD recently discovered that on some hardware, there is a race condition
possible when a C-state change request goes onto the bus at the same
time as a P-state change request.

Both requests happen, but the southbridge hardware only acknowledges the
C-state change.  The PowerNow! driver is then stuck in a loop, waiting
for the P-state change acknowledgement.  The driver eventually times
out, but can no longer perform P-state changes.

It turns out the solution is to resend the P-state change, which the
southbridge will acknowledge normally.

Thanks to Johannes Winkelmann for reporting this and testing the fix.
Signed-off-by: Mark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This fixes a stupid typo bug in the iSeries hash table code.

When we place a hash PTE in the secondary bucket, instead of setting the
SECONDARY flag bit, as we should, we (redundantly) set the VALID flag.

This was introduced with the patch abolishing bitfields from the hash
table code.  Mea culpa, oops.  It hasn't been noticed until now because
in practice we don't hit the secondary bucket terribly often.
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The wrong state emission routines were being called for G550, and
consistent maps weren't correctly mapped...
Signed-off-by: Dave Airlie <airlied@linux.ie>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

While working on 64K pages, I found this little buglet in our
update_mmu_cache() implementation.

The code calls __hash_page() passing it an "access" parameter (the type
of access that triggers the hash) containing the bits _PAGE_RW and
_PAGE_USER of the linux PTE.  The latter is useless in this case and the
former is wrong.  In fact, if we have a writeable PTE and we pass
_PAGE_RW to hash_page(), it will set _PAGE_DIRTY (since we track dirty
that way, by hash faulting !dirty) which is not what we want.

In fact, the correct fix is to always pass 0. That means that only
read-only or already dirty read write PTEs will be preloaded. The
(hopefully rare) case of a non dirty read write PTE can't be preloaded
this way, it will have to fault in hash_page on the actual access.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This fixes a typo in the div128_by_32 function used in the timekeeping
calculations on ppc64.  If you look at the code it's quite obvious
that we need (rb + c) rather than (rb + b).  The "b" is clearly just a
typo.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>