Some dm-crypt testing revealed several bugs in the 256-bit unrolled
loops.

The DECRYPT_256_2() macro had two errors:

1) Missing reload of KEY registers %f60 and %f62

2) Missing "\" in penultimate line of definition.

In aes_sparc64_ecb_decrypt_256, we were storing the second half of the
encryption result from the wrong source registers.

In aes_sparc64_ctr_crypt_256 we have to be careful when we fall out of
the 32-byte-at-a-time loop and handle a trailing 16-byte chunk.  In
that case we've clobbered the final key holding registers and have to
restore them before executing the ENCRYPT_256() macro.  Inside of the
32-byte-at-a-time loop things are OK, because we do this key register
restoring during the first few rounds of the ENCRYPT_256_2() macro.
Signed-off-by: David S. Miller <davem@davemloft.net>

Put the opcode macros in a common header
Signed-off-by: David S. Miller <davem@davemloft.net>

Before:

testing speed of ctr(aes) encryption
test 0 (128 bit key, 16 byte blocks): 1 operation in 206 cycles (16 bytes)
test 1 (128 bit key, 64 byte blocks): 1 operation in 244 cycles (64 bytes)
test 2 (128 bit key, 256 byte blocks): 1 operation in 360 cycles (256 bytes)
test 3 (128 bit key, 1024 byte blocks): 1 operation in 814 cycles (1024 bytes)
test 4 (128 bit key, 8192 byte blocks): 1 operation in 5021 cycles (8192 bytes)
test 5 (192 bit key, 16 byte blocks): 1 operation in 206 cycles (16 bytes)
test 6 (192 bit key, 64 byte blocks): 1 operation in 240 cycles (64 bytes)
test 7 (192 bit key, 256 byte blocks): 1 operation in 378 cycles (256 bytes)
test 8 (192 bit key, 1024 byte blocks): 1 operation in 939 cycles (1024 bytes)
test 9 (192 bit key, 8192 byte blocks): 1 operation in 6395 cycles (8192 bytes)
test 10 (256 bit key, 16 byte blocks): 1 operation in 209 cycles (16 bytes)
test 11 (256 bit key, 64 byte blocks): 1 operation in 249 cycles (64 bytes)
test 12 (256 bit key, 256 byte blocks): 1 operation in 414 cycles (256 bytes)
test 13 (256 bit key, 1024 byte blocks): 1 operation in 1073 cycles (1024 bytes)
test 14 (256 bit key, 8192 byte blocks): 1 operation in 7110 cycles (8192 bytes)

testing speed of ctr(aes) decryption
test 0 (128 bit key, 16 byte blocks): 1 operation in 225 cycles (16 bytes)
test 1 (128 bit key, 64 byte blocks): 1 operation in 233 cycles (64 bytes)
test 2 (128 bit key, 256 byte blocks): 1 operation in 344 cycles (256 bytes)
test 3 (128 bit key, 1024 byte blocks): 1 operation in 810 cycles (1024 bytes)
test 4 (128 bit key, 8192 byte blocks): 1 operation in 5021 cycles (8192 bytes)
test 5 (192 bit key, 16 byte blocks): 1 operation in 206 cycles (16 bytes)
test 6 (192 bit key, 64 byte blocks): 1 operation in 240 cycles (64 bytes)
test 7 (192 bit key, 256 byte blocks): 1 operation in 376 cycles (256 bytes)
test 8 (192 bit key, 1024 byte blocks): 1 operation in 938 cycles (1024 bytes)
test 9 (192 bit key, 8192 byte blocks): 1 operation in 6380 cycles (8192 bytes)
test 10 (256 bit key, 16 byte blocks): 1 operation in 214 cycles (16 bytes)
test 11 (256 bit key, 64 byte blocks): 1 operation in 251 cycles (64 bytes)
test 12 (256 bit key, 256 byte blocks): 1 operation in 411 cycles (256 bytes)
test 13 (256 bit key, 1024 byte blocks): 1 operation in 1070 cycles (1024 bytes)
test 14 (256 bit key, 8192 byte blocks): 1 operation in 7114 cycles (8192 bytes)

After:

testing speed of ctr(aes) encryption
test 0 (128 bit key, 16 byte blocks): 1 operation in 211 cycles (16 bytes)
test 1 (128 bit key, 64 byte blocks): 1 operation in 246 cycles (64 bytes)
test 2 (128 bit key, 256 byte blocks): 1 operation in 344 cycles (256 bytes)
test 3 (128 bit key, 1024 byte blocks): 1 operation in 799 cycles (1024 bytes)
test 4 (128 bit key, 8192 byte blocks): 1 operation in 4975 cycles (8192 bytes)
test 5 (192 bit key, 16 byte blocks): 1 operation in 210 cycles (16 bytes)
test 6 (192 bit key, 64 byte blocks): 1 operation in 236 cycles (64 bytes)
test 7 (192 bit key, 256 byte blocks): 1 operation in 365 cycles (256 bytes)
test 8 (192 bit key, 1024 byte blocks): 1 operation in 888 cycles (1024 bytes)
test 9 (192 bit key, 8192 byte blocks): 1 operation in 6055 cycles (8192 bytes)
test 10 (256 bit key, 16 byte blocks): 1 operation in 209 cycles (16 bytes)
test 11 (256 bit key, 64 byte blocks): 1 operation in 255 cycles (64 bytes)
test 12 (256 bit key, 256 byte blocks): 1 operation in 404 cycles (256 bytes)
test 13 (256 bit key, 1024 byte blocks): 1 operation in 1010 cycles (1024 bytes)
test 14 (256 bit key, 8192 byte blocks): 1 operation in 6669 cycles (8192 bytes)

testing speed of ctr(aes) decryption
test 0 (128 bit key, 16 byte blocks): 1 operation in 210 cycles (16 bytes)
test 1 (128 bit key, 64 byte blocks): 1 operation in 233 cycles (64 bytes)
test 2 (128 bit key, 256 byte blocks): 1 operation in 340 cycles (256 bytes)
test 3 (128 bit key, 1024 byte blocks): 1 operation in 818 cycles (1024 bytes)
test 4 (128 bit key, 8192 byte blocks): 1 operation in 4956 cycles (8192 bytes)
test 5 (192 bit key, 16 byte blocks): 1 operation in 206 cycles (16 bytes)
test 6 (192 bit key, 64 byte blocks): 1 operation in 239 cycles (64 bytes)
test 7 (192 bit key, 256 byte blocks): 1 operation in 361 cycles (256 bytes)
test 8 (192 bit key, 1024 byte blocks): 1 operation in 888 cycles (1024 bytes)
test 9 (192 bit key, 8192 byte blocks): 1 operation in 5996 cycles (8192 bytes)
test 10 (256 bit key, 16 byte blocks): 1 operation in 214 cycles (16 bytes)
test 11 (256 bit key, 64 byte blocks): 1 operation in 248 cycles (64 bytes)
test 12 (256 bit key, 256 byte blocks): 1 operation in 395 cycles (256 bytes)
test 13 (256 bit key, 1024 byte blocks): 1 operation in 1010 cycles (1024 bytes)
test 14 (256 bit key, 8192 byte blocks): 1 operation in 6664 cycles (8192 bytes)
Signed-off-by: David S. Miller <davem@davemloft.net>

Before:

testing speed of ecb(aes) decryption
test 0 (128 bit key, 16 byte blocks): 1 operation in 223 cycles (16 bytes)
test 1 (128 bit key, 64 byte blocks): 1 operation in 230 cycles (64 bytes)
test 2 (128 bit key, 256 byte blocks): 1 operation in 325 cycles (256 bytes)
test 3 (128 bit key, 1024 byte blocks): 1 operation in 719 cycles (1024 bytes)
test 4 (128 bit key, 8192 byte blocks): 1 operation in 4266 cycles (8192 bytes)
test 5 (192 bit key, 16 byte blocks): 1 operation in 211 cycles (16 bytes)
test 6 (192 bit key, 64 byte blocks): 1 operation in 234 cycles (64 bytes)
test 7 (192 bit key, 256 byte blocks): 1 operation in 353 cycles (256 bytes)
test 8 (192 bit key, 1024 byte blocks): 1 operation in 808 cycles (1024 bytes)
test 9 (192 bit key, 8192 byte blocks): 1 operation in 5344 cycles (8192 bytes)
test 10 (256 bit key, 16 byte blocks): 1 operation in 214 cycles (16 bytes)
test 11 (256 bit key, 64 byte blocks): 1 operation in 243 cycles (64 bytes)
test 12 (256 bit key, 256 byte blocks): 1 operation in 393 cycles (256 bytes)
test 13 (256 bit key, 1024 byte blocks): 1 operation in 939 cycles (1024 bytes)
test 14 (256 bit key, 8192 byte blocks): 1 operation in 6039 cycles (8192 bytes)

After:

testing speed of ecb(aes) decryption
test 0 (128 bit key, 16 byte blocks): 1 operation in 226 cycles (16 bytes)
test 1 (128 bit key, 64 byte blocks): 1 operation in 231 cycles (64 bytes)
test 2 (128 bit key, 256 byte blocks): 1 operation in 313 cycles (256 bytes)
test 3 (128 bit key, 1024 byte blocks): 1 operation in 681 cycles (1024 bytes)
test 4 (128 bit key, 8192 byte blocks): 1 operation in 3964 cycles (8192 bytes)
test 5 (192 bit key, 16 byte blocks): 1 operation in 205 cycles (16 bytes)
test 6 (192 bit key, 64 byte blocks): 1 operation in 240 cycles (64 bytes)
test 7 (192 bit key, 256 byte blocks): 1 operation in 341 cycles (256 bytes)
test 8 (192 bit key, 1024 byte blocks): 1 operation in 770 cycles (1024 bytes)
test 9 (192 bit key, 8192 byte blocks): 1 operation in 5050 cycles (8192 bytes)
test 10 (256 bit key, 16 byte blocks): 1 operation in 216 cycles (16 bytes)
test 11 (256 bit key, 64 byte blocks): 1 operation in 250 cycles (64 bytes)
test 12 (256 bit key, 256 byte blocks): 1 operation in 371 cycles (256 bytes)
test 13 (256 bit key, 1024 byte blocks): 1 operation in 869 cycles (1024 bytes)
test 14 (256 bit key, 8192 byte blocks): 1 operation in 5494 cycles (8192 bytes)
Signed-off-by: David S. Miller <davem@davemloft.net>

The AES opcodes have a 3 cycle latency, so by doing 32-bytes at a
time we avoid a pipeline bubble in between every round.

For the 256-bit key case, it looks like we're doing more work in
order to reload the KEY registers during the loop to make space
for scarce temporaries.  But the load dual issues with the AES
operations so we get the KEY reloads essentially for free.

Before:

testing speed of ecb(aes) encryption
test 0 (128 bit key, 16 byte blocks): 1 operation in 264 cycles (16 bytes)
test 1 (128 bit key, 64 byte blocks): 1 operation in 231 cycles (64 bytes)
test 2 (128 bit key, 256 byte blocks): 1 operation in 329 cycles (256 bytes)
test 3 (128 bit key, 1024 byte blocks): 1 operation in 715 cycles (1024 bytes)
test 4 (128 bit key, 8192 byte blocks): 1 operation in 4248 cycles (8192 bytes)
test 5 (192 bit key, 16 byte blocks): 1 operation in 221 cycles (16 bytes)
test 6 (192 bit key, 64 byte blocks): 1 operation in 234 cycles (64 bytes)
test 7 (192 bit key, 256 byte blocks): 1 operation in 359 cycles (256 bytes)
test 8 (192 bit key, 1024 byte blocks): 1 operation in 803 cycles (1024 bytes)
test 9 (192 bit key, 8192 byte blocks): 1 operation in 5366 cycles (8192 bytes)
test 10 (256 bit key, 16 byte blocks): 1 operation in 209 cycles (16 bytes)
test 11 (256 bit key, 64 byte blocks): 1 operation in 255 cycles (64 bytes)
test 12 (256 bit key, 256 byte blocks): 1 operation in 379 cycles (256 bytes)
test 13 (256 bit key, 1024 byte blocks): 1 operation in 938 cycles (1024 bytes)
test 14 (256 bit key, 8192 byte blocks): 1 operation in 6041 cycles (8192 bytes)

After:

testing speed of ecb(aes) encryption
test 0 (128 bit key, 16 byte blocks): 1 operation in 266 cycles (16 bytes)
test 1 (128 bit key, 64 byte blocks): 1 operation in 256 cycles (64 bytes)
test 2 (128 bit key, 256 byte blocks): 1 operation in 305 cycles (256 bytes)
test 3 (128 bit key, 1024 byte blocks): 1 operation in 676 cycles (1024 bytes)
test 4 (128 bit key, 8192 byte blocks): 1 operation in 3981 cycles (8192 bytes)
test 5 (192 bit key, 16 byte blocks): 1 operation in 210 cycles (16 bytes)
test 6 (192 bit key, 64 byte blocks): 1 operation in 233 cycles (64 bytes)
test 7 (192 bit key, 256 byte blocks): 1 operation in 340 cycles (256 bytes)
test 8 (192 bit key, 1024 byte blocks): 1 operation in 766 cycles (1024 bytes)
test 9 (192 bit key, 8192 byte blocks): 1 operation in 5136 cycles (8192 bytes)
test 10 (256 bit key, 16 byte blocks): 1 operation in 206 cycles (16 bytes)
test 11 (256 bit key, 64 byte blocks): 1 operation in 268 cycles (64 bytes)
test 12 (256 bit key, 256 byte blocks): 1 operation in 368 cycles (256 bytes)
test 13 (256 bit key, 1024 byte blocks): 1 operation in 890 cycles (1024 bytes)
test 14 (256 bit key, 8192 byte blocks): 1 operation in 5718 cycles (8192 bytes)
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

Instead of testing and branching off of the key size on every
encrypt/decrypt call, use method ops assigned at key set time.

Reverse the order of float registers used for decryption to make
future changes easier.

Align all assembler routines on a 32-byte boundary.
Signed-off-by: David S. Miller <davem@davemloft.net>

On SPARC-T4 fsrc2 has 1 cycle of latency, whereas fsrc1 has 11 cycles.

True story.
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>

Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>

Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>

Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>

Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>

Describe how we support two types of PMU setups, one with a single control
register and two counters stored in a single register, and another with
one control register per counter and each counter living in it's own
register.
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

When cpuc->n_events is zero, we actually don't do anything and we just
write the cpuc->pcr[0] value as-is without any modifications.

The "pcr = 0;" assignment there was just useless and confusing.
Signed-off-by: David S. Miller <davem@davemloft.net>

Make the per-cpu pcr save area an array instead of one u64.

Describe how many PCR and PIC registers the chip has in the sparc_pmu
descriptor.
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

Now specified in sparc_pmu descriptor.
Signed-off-by: David S. Miller <davem@davemloft.net>

Starting with SPARC-T4 we have a seperate PCR control register
for each performance counter, and there are absolutely no
restrictions on what events can run on which counters.

Add flags that we can use to elide the conflict and dependency
logic used to handle older chips.
Signed-off-by: David S. Miller <davem@davemloft.net>

This is enough to get the NMIs working, more work is needed
for perf events.
Signed-off-by: David S. Miller <davem@davemloft.net>

We assumed PCR_PIC_PRIV can always be used to disable it, but that
won't be true for SPARC-T4.

This allows us also to get rid of some messy defines used in only
one location.
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

And, like for the PCR, allow indexing of different PIC register
numbers.

This also removes all of the non-__KERNEL__ bits from asm/perfctr.h,
nothing kernel side should include it any more.
Signed-off-by: David S. Miller <davem@davemloft.net>

SPARC-T4 and later have multiple PCR registers, one for each
PIC counter.
Signed-off-by: David S. Miller <davem@davemloft.net>

Unlike for previous chips, access to the perf-counter control
registers are all hyper-privileged.  Therefore, access to them must go
through a hypervisor interface.
Signed-off-by: David S. Miller <davem@davemloft.net>

Compare and branch, pause, and the various new cryptographic opcodes.

We advertise the crypto opcodes to userspace using one hwcap bit,
HWCAP_SPARC_CRYPTO.

This essentially indicates that the %cfr register can be interrograted
and used to determine exactly which crypto opcodes are available on
the current cpu.

We use the %cfr register to report all of the crypto opcodes available
in the bootup CPU caps log message, and via /proc/cpuinfo.
Signed-off-by: David S. Miller <davem@davemloft.net>

Pull ARM fixes from Russell King:
 "The largest thing in this set of changes is bringing back some of the
  ARMv3 code to fix a compile problem noticed on RiscPC, which we still
  support, even though we only support ARMv4 there.

  (The reason is that the system bus doesn't support ARMv4 half-word
  accesses, so we need the ARMv3 library code for this platform.)

  The rest are all quite minor fixes."

* 'fixes' of git://git.linaro.org/people/rmk/linux-arm:
  ARM: 7490/1: Drop duplicate select for GENERIC_IRQ_PROBE
  ARM: Bring back ARMv3 IO and user access code
  ARM: 7489/1: errata: fix workaround for erratum #720789 on UP systems
  ARM: 7488/1: mm: use 5 bits for swapfile type encoding
  ARM: 7487/1: mm: avoid setting nG bit for user mappings that aren't present
  ARM: 7486/1: sched_clock: update epoch_cyc on resume
  ARM: 7484/1: Don't enable GENERIC_LOCKBREAK with ticket spinlocks
  ARM: 7483/1: vfp: only advertise VFPv4 in hwcaps if CONFIG_VFPv3 is enabled
  ARM: 7482/1: topology: fix section mismatch warning for init_cpu_topology

Pull power management fixes from Rafael J. Wysocki:
  - Fixes for three obscure problems in the runtime PM core code found
   recently.
 - Two fixes for the new "coupled" cpuidle code from Colin Cross and Jon
   Medhurst.
 - intel_idle driver fix from Konrad Rzeszutek Wilk.

* tag 'pm-for-3.6-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  intel_idle: Check cpu_idle_get_driver() for NULL before dereferencing it.
  cpuidle: Prevent null pointer dereference in cpuidle_coupled_cpu_notify
  cpuidle: coupled: fix sleeping while atomic in cpu notifier
  PM / Runtime: Check device PM QoS setting before "no callbacks" check
  PM / Runtime: Clear power.deferred_resume on success in rpm_suspend()
  PM / Runtime: Fix rpm_resume() return value for power.no_callbacks set

Pull vfs fixes from Miklos Szeredi.

This mainly fixes some confusion about whether the open 'mode' variable
passed around should contain the full file type (S_IFREG etc)
information or just the permission mode.  In particular, the lack of
proper file type information had confused fuse.

* 'vfs-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs:
  vfs: fix propagation of atomic_open create error on negative dentry
  fuse: check create mode in atomic open
  vfs: pass right create mode to may_o_create()
  vfs: atomic_open(): fix create mode usage
  vfs: canonicalize create mode in build_open_flags()

Pull md fixes from NeilBrown:
 "2 fixes for md, tagged for -stable"

* tag 'md-3.6-fixes' of git://neil.brown.name/md:
  md/raid10: fix problem with on-stack allocation of r10bio structure.
  md: Don't truncate size at 4TB for RAID0 and Linear

A 'struct r10bio' has an array of per-copy information at the end.
This array is declared with size [0] and r10bio_pool_alloc allocates
enough extra space to store the per-copy information depending on the
number of copies needed.

So declaring a 'struct r10bio on the stack isn't going to work.  It
won't allocate enough space, and memory corruption will ensue.

So in the two places where this is done, declare a sufficiently large
structure and use that instead.

The two call-sites of this bug were introduced in 3.4 and 3.5
so this is suitable for both those kernels.  The patch will have to
be modified for 3.4 as it only has one bug.

Cc: stable@vger.kernel.org
Reported-by: Ivan Vasilyev <ivan.vasilyev@gmail.com>
Tested-by: Ivan Vasilyev <ivan.vasilyev@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>

Pull infiniband/rdma fixes from Roland Dreier:
 "Grab bag of InfiniBand/RDMA fixes:
   - IPoIB fixes for regressions introduced by path database conversion
   - mlx4 fixes for bugs with large memory systems and regressions from
     SR-IOV patches
   - RDMA CM fix for passing bad event up to userspace
   - Other minor fixes"

* tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
  IB/mlx4: Check iboe netdev pointer before dereferencing it
  mlx4_core: Clean up buddy bitmap allocation
  mlx4_core: Fix integer overflow issues around MTT table
  mlx4_core: Allow large mlx4_buddy bitmaps
  IB/srp: Fix a race condition
  IB/qib: Fix error return code in qib_init_7322_variables()
  IB: Fix typos in infiniband drivers
  IB/ipoib: Fix RCU pointer dereference of wrong object
  IB/ipoib: Add missing locking when CM object is deleted
  RDMA/ucma.c: Fix for events with wrong context on iWARP
  RDMA/ocrdma: Don't call vlan_dev_real_dev() for non-VLAN netdevs
  IB/mlx4: Fix possible deadlock on sm_lock spinlock