This patch exports the finup operation where available and adds
a default finup operation for ahash.  The operations final, finup
and digest also will now deal with unaligned result pointers by
copying it.  Finally export/import operations are will now be
exported too.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

We currently use GFP_ATOMIC in the unaligned setkey function
to allocate the temporary aligned buffer.  Since setkey must
be called in a sleepable context, we can use GFP_KERNEL instead.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

When we encounter an unaligned pointer we are supposed to copy
it to a temporary aligned location.  However the temporary buffer
isn't aligned properly.  This patch fixes that.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Some unaligned buffers on the stack weren't zapped properly which
may cause secret data to be leaked.  This patch fixes them by doing
a zero memset.

It is also possible for us to place random kernel stack contents
in the digest buffer if a digest operation fails.  This is fixed
by only copying if the operation succeeded.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Now that all ahash implementations have been converted to the new
ahash type, we can remove old_ahash_alg and its associated support.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch changes cryptd to use the new style ahash type.  In
particular, the instance is enlarged to encapsulate the new
ahash_alg structure.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch changes cryptd to use the template->create function
instead of alloc in anticipation for the switch to new style
ahash algorithms.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch adds a mask parameter to complement the existing type
parameter.  This is useful when instantiating algorithms that
require a mask other than the default, e.g., ahash algorithms.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch adds support for creating ahash instances and using
ahash as spawns.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch converts crypto_ahash to the new style.  The old ahash
algorithm type is retained until the existing ahash implementations
are also converted.  All ahash users will automatically get the
new crypto_ahash type.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

As the extsize and init_tfm functions belong to the frontend the
frontend argument is superfluous.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch makes cryptd use crypto_ahash_set_reqsize to avoid
accessing crypto_ahash directly.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch changes cryptd to use shash algorithms instead of the
legacy hash interface.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch exports the async functions so that they can be reused
by cryptd when it switches over to using shash.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch removes the implementation of hash and digest now that
no algorithms use them anymore.  The interface though will remain
until the users are converted across.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

Now that there are no more legacy hash implementations we can
remove the reference to crypto_hash.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch converts the xcbc algorithm to the new shash type.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch changes hmac to the new shash interface.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch changes descsize to a run-time attribute so that
implementations can change it in their init functions.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch moves the run-time null setkey check to shash_prepare_alg
just like we did for finup/digest.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch adds export/import support to sha256_generic.  The exported
type is defined by struct sha256_state, which is basically the entire
descriptor state of sha256_generic.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch replaces the two 32-bit counter code in sha256_generic
with the simpler 64-bit counter code from sha1.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch adds export/import support to sha1_generic.  The exported
type is defined by struct sha1_state, which is basically the entire
descriptor state of sha1_generic.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch moves the run-time null finup/digest checks to the
shash_prepare_alg function which is run at registration time.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch replaces the full descriptor export with an export of
the partial hash state.  This allows the use of a consistent export
format across all implementations of a given algorithm.

This is useful because a number of cases require the use of the
partial hash state, e.g., PadLock can use the SHA1 hash state
to get around the fact that it can only hash contiguous data
chunks.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch allows crypto_drop_spawn to be called on spawns that
have not been initialised or have failed initialisation.  This
fixes potential crashes during initialisation without adding
special case code.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch fixes crypto_shash_import to propagate the value returned
by reinit.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch simplifies the default digest function by using finup.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch adds shash_register_instance so that shash instances
can be registered without bypassing the shash checks applied to
normal algorithms.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch adds the helper shash_attr_alg2 which locates a shash
algorithm based on the information in the given attribute.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch adds the helper crypto_attr_alg2 which is similar to
crypto_attr_alg but takes an extra frontend argument.  This is
intended to be used by new style algorithm types such as shash.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch adds the functions needed to create and use shash
spawns, i.e., to use shash algorithms in a template.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch modifies the spawn infrastructure to support new style
algorithms like shash.  In particular, this means storing the
frontend type in the spawn and using crypto_create_tfm to allocate
the tfm.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch adds shash_instance and the associated alloc/free
functions.  This is meant to be an instance that with a shash
algorithm under it.  Note that the instance itself doesn't have
to be shash.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch adds a new argument to crypto_alloc_instance which
sets aside some space before the instance for use by algorithms
such as shash that place type-specific data before crypto_alg.

For compatibility the function has been renamed so that existing
users aren't affected.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch introduces the template->create function intended
to replace the existing alloc function.  The intention is for
create to handle the registration directly, whereas currently
the caller of alloc has to handle the registration.

This allows type-specific code to be run prior to registration.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

As reported by Eric Sesterhenn the re-allocation of the cipher in reset leads
to:
|BUG: sleeping function called from invalid context at kernel/rwsem.c:21
|in_atomic(): 1, irqs_disabled(): 0, pid: 4926, name: modprobe
|INFO: lockdep is turned off.
|Pid: 4926, comm: modprobe Tainted: G   M 2.6.31-rc1-22297-g52989765 #24
|Call Trace:
| [<c011dd93>] __might_sleep+0xf9/0x101
| [<c0777aa0>] down_read+0x16/0x68
| [<c048bf04>] crypto_alg_lookup+0x16/0x34
| [<c048bf52>] crypto_larval_lookup+0x30/0xf9
| [<c048c038>] crypto_alg_mod_lookup+0x1d/0x62
| [<c048c13e>] crypto_alloc_base+0x1e/0x64
| [<c04bf991>] reset_prng_context+0xab/0x13f
| [<c04e5cfc>] ? __spin_lock_init+0x27/0x51
| [<c04bfce1>] cprng_init+0x2a/0x42
| [<c048bb4c>] __crypto_alloc_tfm+0xfa/0x128
| [<c048c153>] crypto_alloc_base+0x33/0x64
| [<c04933c9>] alg_test_cprng+0x30/0x1f4
| [<c0493329>] alg_test+0x12f/0x19f
| [<c0177f1f>] ? __alloc_pages_nodemask+0x14d/0x481
| [<d09219e2>] do_test+0xf9d/0x163f [tcrypt]
| [<d0920de6>] do_test+0x3a1/0x163f [tcrypt]
| [<d0926035>] tcrypt_mod_init+0x35/0x7c [tcrypt]
| [<c010113c>] _stext+0x54/0x12c
| [<d0926000>] ? tcrypt_mod_init+0x0/0x7c [tcrypt]
| [<c01398a3>] ? up_read+0x16/0x2b
| [<c0139fc4>] ? __blocking_notifier_call_chain+0x40/0x4c
| [<c014ee8d>] sys_init_module+0xa9/0x1bf
| [<c010292b>] sysenter_do_call+0x12/0x32

because a spin lock is held and crypto_alloc_base() may sleep.
There is no reason to re-allocate the cipher, the state is resetted in
->setkey(). This patches makes the cipher allocation a one time thing and
moves it to init.
Reported-by: NEric Sesterhenn <eric.sesterhenn@lsexperts.de>
Signed-off-by: NSebastian Andrzej Siewior <sebastian@breakpoint.cc>
Acked-by: NNeil Horman <nhorman@tuxdriver.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

The current code uses a mix of sping_lock() & spin_lock_irqsave(). This can
lead to deadlock with the correct timming & cprng_get_random() + cprng_reset()
sequence.
I've converted them to bottom half locks since all three user grab just a BH
lock so this runs probably in softirq :)
Signed-off-by: NSebastian Andrzej Siewior <sebastian@breakpoint.cc>
Acked-by: NNeil Horman <nhorman@tuxdriver.com>
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

This patch adds the support for testing specific implementations.
This should only be used in very specific situations.  Right now
this means specific implementations of random number generators.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>

As it stands we use chainiv for sync algorithms and eseqiv for
async algorithms.  However, when there is more than one CPU
chainiv forces all processing to be serialised which is usually
not what you want.  Also, the added overhead of eseqiv isn't that
great.

Therefore this patch changes the default sync geniv on SMP machines
to eseqiv.  For the odd situation where the overhead is unacceptable
then chainiv is still available as an option.

Note that on UP machines chainiv is still preferred over eseqiv
for sync algorithms.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>