If we are unable to offload async_mult() or async_sum_product(), then
unmap the buffers before falling through to the synchronous path.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Testing on x86_64 with NDISKS=255 yields:
do_IRQ: modprobe near stack overflow (cur:ffff88007d19c000,sp:ffff88007d19c128)
...and eventually
general protection fault: 0000 [#1]
Moving the scribble buffers off the stack allows the test to complete
successfully.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Some engines have transfer size and address alignment restrictions. Add
a per-operation alignment property to struct dma_device that the async
routines and dmatest can use to check alignment capabilities.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Channel switching is problematic for some dmaengine drivers as the
architecture precludes separating the ->prep from ->submit. In these
cases the driver can select ASYNC_TX_DISABLE_CHANNEL_SWITCH to modify
the async_tx allocator to only return channels that support all of the
required asynchronous operations.
For example MD_RAID456=y selects support for asynchronous xor, xor
validate, pq, pq validate, and memcpy. When
ASYNC_TX_DISABLE_CHANNEL_SWITCH=y any channel with all these
capabilities is marked DMA_ASYNC_TX allowing async_tx_find_channel() to
quickly locate compatible channels with the guarantee that dependency
chains will remain on one channel. When
ASYNC_TX_DISABLE_CHANNEL_SWITCH=n async_tx_find_channel() may select
channels that lead to operation chains that need to cross channel
boundaries using the async_tx channel switch capability.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Some engines optimize operation by reading ahead in the descriptor chain
such that descriptor2 may start execution before descriptor1 completes.
If descriptor2 depends on the result from descriptor1 then a fence is
required (on descriptor2) to disable this optimization. The async_tx
api could implicitly identify dependencies via the 'depend_tx'
parameter, but that would constrain cases where the dependency chain
only specifies a completion order rather than a data dependency. So,
provide an ASYNC_TX_FENCE to explicitly identify data dependencies.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Port drivers/md/raid6test/test.c to use the async raid6 recovery
routines. This is meant as a unit test for raid6 acceleration drivers. In
addition to the 16-drive test case this implements tests for the 4-disk and
5-disk special cases (dma devices can not generically handle less than 2
sources), and adds a test for the D+Q case.
Reviewed-by: Andre Noll <maan@systemlinux.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
async_raid6_2data_recov() recovers two data disk failures
async_raid6_datap_recov() recovers a data disk and the P disk
These routines are a port of the synchronous versions found in
drivers/md/raid6recov.c. The primary difference is breaking out the xor
operations into separate calls to async_xor. Two helper routines are
introduced to perform scalar multiplication where needed.
async_sum_product() multiplies two sources by scalar coefficients and
then sums (xor) the result. async_mult() simply multiplies a single
source by a scalar.
This implemention also includes, in contrast to the original
synchronous-only code, special case handling for the 4-disk and 5-disk
array cases. In these situations the default N-disk algorithm will
present 0-source or 1-source operations to dma devices. To cover for
dma devices where the minimum source count is 2 we implement 4-disk and
5-disk handling in the recovery code.
[ Impact: asynchronous raid6 recovery routines for 2data and datap cases ]
Cc: Yuri Tikhonov <yur@emcraft.com>
Cc: Ilya Yanok <yanok@emcraft.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Andre Noll <maan@systemlinux.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
[ Based on an original patch by Yuri Tikhonov ]
This adds support for doing asynchronous GF multiplication by adding
two additional functions to the async_tx API:
async_gen_syndrome() does simultaneous XOR and Galois field
multiplication of sources.
async_syndrome_val() validates the given source buffers against known P
and Q values.
When a request is made to run async_pq against more than the hardware
maximum number of supported sources we need to reuse the previous
generated P and Q values as sources into the next operation. Care must
be taken to remove Q from P' and P from Q'. For example to perform a 5
source pq op with hardware that only supports 4 sources at a time the
following approach is taken:
p, q = PQ(src0, src1, src2, src3, COEF({01}, {02}, {04}, {08}))
p', q' = PQ(p, q, q, src4, COEF({00}, {01}, {00}, {10}))
p' = p + q + q + src4 = p + src4
q' = {00}*p + {01}*q + {00}*q + {10}*src4 = q + {10}*src4
Note: 4 is the minimum acceptable maxpq otherwise we punt to
synchronous-software path.
The DMA_PREP_CONTINUE flag indicates to the driver to reuse p and q as
sources (in the above manner) and fill the remaining slots up to maxpq
with the new sources/coefficients.
Note1: Some devices have native support for P+Q continuation and can skip
this extra work. Devices with this capability can advertise it with
dma_set_maxpq. It is up to each driver how to handle the
DMA_PREP_CONTINUE flag.
Note2: The api supports disabling the generation of P when generating Q,
this is ignored by the synchronous path but is implemented by some dma
devices to save unnecessary writes. In this case the continuation
algorithm is simplified to only reuse Q as a source.
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Yuri Tikhonov <yur@emcraft.com>
Signed-off-by: Ilya Yanok <yanok@emcraft.com>
Reviewed-by: Andre Noll <maan@systemlinux.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
We currently walk the parent chain when waiting for a given tx to
complete however this walk may race with the driver cleanup routine.
The routines in async_raid6_recov.c may fall back to the synchronous
path at any point so we need to be prepared to call async_tx_quiesce()
(which calls dma_wait_for_async_tx). To remove the ->parent walk we
guarantee that every time a dependency is attached ->issue_pending() is
invoked, then we can simply poll the initial descriptor until
completion.
This also allows for a lighter weight 'issue pending' implementation as
there is no longer a requirement to iterate through all the channels'
->issue_pending() routines as long as operations have been submitted in
an ordered chain. async_tx_issue_pending() is added for this case.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
If module_init and module_exit are nops then neither need to be defined.
[ Impact: pure cleanup ]
Reviewed-by: Andre Noll <maan@systemlinux.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Replace the flat zero_sum_result with a collection of flags to contain
the P (xor) zero-sum result, and the soon to be utilized Q (raid6 reed
solomon syndrome) zero-sum result. Use the SUM_CHECK_ namespace instead
of DMA_ since these flags will be used on non-dma-zero-sum enabled
platforms.
Reviewed-by: Andre Noll <maan@systemlinux.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
On HIGHMEM64G systems dma_addr_t is known to be larger than (void *)
which precludes async_xor from performing dma address conversions by
reusing the input parameter address list. However, other parts of the
dmaengine infrastructure do not suffer this constraint, so the
HIGHMEM64G restriction can be down-levelled.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The xor tests are run on uninitialized data, because it is doesn't
really matter what the underlying data is. Annotate this false-
positive warning.
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com>
async_xor() needs space to perform dma and page address conversions. In
most cases the code can simply reuse the struct page * array because the
size of the native pointer matches the size of a dma/page address. In
order to support archs where sizeof(dma_addr_t) is larger than
sizeof(struct page *), or to preserve the input parameters, we utilize a
memory region passed in by the caller.
Since the code is now prepared to handle the case where it cannot
perform address conversions on the stack, we no longer need the
!HIGHMEM64G dependency in drivers/dma/Kconfig.
[ Impact: don't clobber input buffers for address conversions ]
Reviewed-by: Andre Noll <maan@systemlinux.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Prepare the api for the arrival of a new parameter, 'scribble'. This
will allow callers to identify scratchpad memory for dma address or page
address conversions. As this adds yet another parameter, take this
opportunity to convert the common submission parameters (flags,
dependency, callback, and callback argument) into an object that is
passed by reference.
Also, take this opportunity to fix up the kerneldoc and add notes about
the relevant ASYNC_TX_* flags for each routine.
[ Impact: moves api pass-by-value parameters to a pass-by-reference struct ]
Signed-off-by: Andre Noll <maan@systemlinux.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
In support of inter-channel chaining async_tx utilizes an ack flag to
gate whether a dependent operation can be chained to another. While the
flag is not set the chain can be considered open for appending. Setting
the ack flag closes the chain and flags the descriptor for garbage
collection. The ASYNC_TX_DEP_ACK flag essentially means "close the
chain after adding this dependency". Since each operation can only have
one child the api now implicitly sets the ack flag at dependency
submission time. This removes an unnecessary management burden from
clients of the api.
[ Impact: clean up and enforce one dependency per operation ]
Reviewed-by: Andre Noll <maan@systemlinux.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Besdies, for the old code, gcc-4.3.3 produced this warning:
"format not a string literal and no format arguments"
Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
As it stands we will each test hash vector both linearly and as
a scatter list if applicable. This means that we cannot have
vectors longer than a page, even with scatter lists.
This patch fixes this by skipping test vectors with np != 0 when
testing linearly.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
As we cannot guarantee the availability of contiguous pages at
run-time, all test vectors must either fit within a page, or use
scatter lists. In some cases vectors were not checked as to
whether they fit inside a page. This patch adds all the missing
checks.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
At present, the tcrypt module always exits with an -EAGAIN upon
successfully completing all the tests its been asked to run. In fips
mode, integrity checking is done by running all self-tests from the
initrd, and its much simpler to check the ret from modprobe for
success than to scrape dmesg and/or /proc/crypto. Simply stay
loaded, giving modprobe a retval of 0, if self-tests all pass and
we're in fips mode.
A side-effect of tracking success/failure for fips mode is that in
non-fips mode, self-test failures will return the actual failure
return codes, rather than always returning -EAGAIN, which seems more
correct anyway.
The tcrypt_test() portion of the patch is dependent on my earlier
pair of patches that skip non-fips algs in fips mode, at least to
achieve the fully intended behavior.
Nb: testing this patch against the cryptodev tree revealed a test
failure for sha384, which I have yet to look into...
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
If crypto_{,de}compress_{update,final}() succeed, return the actual number of
bytes produced instead of zero, so their users don't have to calculate that
theirselves.
Signed-off-by: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Because all fips-allowed algorithms must be self-tested before they
can be used, they will all have entries in testmgr.c's alg_test_descs[].
Skip self-tests for any algs not flagged as fips_approved and return
-EINVAL when in fips mode.
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Set the fips_allowed flag in testmgr.c's alg_test_descs[] for algs
that are allowed to be used when in fips mode.
One caveat: des isn't actually allowed anymore, but des (and thus also
ecb(des)) has to be permitted, because disallowing them results in
des3_ede being unable to properly register (see des module init func).
Also, crc32 isn't technically on the fips approved list, but I think
it gets used in various places that necessitate it being allowed.
This list is based on
http://csrc.nist.gov/groups/STM/cavp/index.html
Important note: allowed/approved here does NOT mean "validated", just
that its an alg that *could* be validated.
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Now with multi-block test vectors, all from SP800-38A, Appendix F.5.
Also added ctr(aes) to case 10 in tcrypt.
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
We currently allocate temporary memory that is used for testing
statically. This renders the testing engine non-reentrant. As
algorithms may nest, i.e., one may construct another in order to
carry out a part of its operation, this is unacceptable. For
example, it has been reported that an AEAD implementation allocates
a cipher in its setkey function, which causes it to fail during
testing as the temporary memory is overwritten.
This patch replaces the static memory with dynamically allocated
buffers. We need a maximum of 16 pages so this slightly increases
the chances of an algorithm failing due to memory shortage.
However, as testing usually occurs at registration, this shouldn't
be a big problem.
Reported-by: Shasi Pulijala <spulijala@amcc.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
According to our FIPS CAVS testing lab guru, when we're in fips mode,
we must print out notices of successful self-test completion for
every alg to be compliant.
New and improved v2, without strncmp crap. Doesn't need to touch a flag
though, due to not moving the notest label around anymore.
Applies atop '[PATCH v2] crypto: catch base cipher self-test failures
in fips mode'.
Personally, I wouldn't mind seeing this info printed out regardless of
whether or not we're in fips mode, I think its useful info, but will
stick with only in fips mode for now.
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add ANSI X9.31 Continuous Pseudo-Random Number Generator (AES mode),
aka 'ansi_cprng' test vectors, taken from Appendix B.2.9 and B.2.10
of the NIST RNGVS document, found here:
http://csrc.nist.gov/groups/STM/cavp/documents/rng/RNGVS.pdf
Successfully tested against both the cryptodev-2.6 tree and a Red
Hat Enterprise Linux 5.4 kernel, via 'modprobe tcrypt mode=150'.
The selection of 150 was semi-arbitrary, didn't seem like it should
go any place in particular, so I started a new range for rng tests.
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add some necessary infrastructure to make it possible to run
self-tests for ansi_cprng. The bits are likely very specific
to the ANSI X9.31 CPRNG in AES mode, and thus perhaps should
be named more specifically if/when we grow additional CPRNG
support...
Successfully tested against the cryptodev-2.6 tree and a
Red Hat Enterprise Linux 5.x kernel with the follow-on
patch that adds the actual test vectors.
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add an array of encryption and decryption + verification self-tests
for rfc4309(ccm(aes)).
Test vectors all come from sample FIPS CAVS files provided to
Red Hat by a testing lab. Unfortunately, all the published sample
vectors in RFC 3610 and NIST Special Publication 800-38C contain nonce
lengths that the kernel's rfc4309 implementation doesn't support, so
while using some public domain vectors would have been preferred, its
not possible at this time.
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add infrastructure to tcrypt/testmgr to support handling ccm decryption
test vectors that are expected to fail verification.
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
make C=1:
| crypto/pcompress.c:77:5: warning: symbol 'crypto_register_pcomp' was not declared. Should it be static?
| crypto/pcompress.c:89:5: warning: symbol 'crypto_unregister_pcomp' was not declared. Should it be static?
Signed-off-by: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Because kernel_fpu_begin() and kernel_fpu_end() operations are too
slow, the performance gain of general mode implementation + aes-aesni
is almost all compensated.
The AES-NI support for more modes are implemented as follow:
- Add a new AES algorithm implementation named __aes-aesni without
kernel_fpu_begin/end()
- Use fpu(<mode>(AES)) to provide kenrel_fpu_begin/end() invoking
- Add <mode>(AES) ablkcipher, which uses cryptd(fpu(<mode>(AES))) to
defer cryption to cryptd context in soft_irq context.
Now the ctr, lrw, pcbc and xts support are added.
Performance testing based on dm-crypt shows that cryption time can be
reduced to 50% of general mode implementation + aes-aesni implementation.
Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Blkcipher touching FPU need to be enclosed by kernel_fpu_begin() and
kernel_fpu_end(). If they are invoked in cipher algorithm
implementation, they will be invoked for each block, so that
performance will be hurt, because they are "slow" operations. This
patch implements "fpu" template, which makes these operations to be
invoked for each request.
Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Use crypto_alloc_base() instead of crypto_alloc_ablkcipher() to
allocate underlying tfm in cryptd_alloc_ablkcipher. Because
crypto_alloc_ablkcipher() prefer GENIV encapsulated crypto instead of
raw one, while cryptd_alloc_ablkcipher needed the raw one.
Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Use kzfree() instead of memset() + kfree().
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Applying kernel janitors todos (printk calls need KERN_*
constants on linebeginnings, reduce stack footprint where
possible) to tcrypts test_hash_speed (where stacks
memory footprint was very high (on i386 1184 bytes to
160 now).
Signed-off-by: Frank Seidel <frank@f-seidel.de>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
A quirk that we've always supported is having an sg entry that's
bigger than a page, or more generally an sg entry that crosses
page boundaries. Even though it would be better to explicitly have
to sg entries for this, we need to support it for the existing users,
in particular, IPsec.
The new ahash sg walking code did try to handle this, but there was
a bug where we didn't increment the page so kept on walking on the
first page over an dover again.
This patch fixes it.
Tested-by: Martin Willi <martin@strongswan.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
crypto: padlock - Revert aes-all alias to aes
crypto: api - Fix algorithm module auto-loading
crypto: eseqiv - Fix IV generation for sync algorithms
crypto: ixp4xx - check firmware for crypto support
The commit a760a6656e (crypto:
api - Fix module load deadlock with fallback algorithms) broke
the auto-loading of algorithms that require fallbacks. The
problem is that the fallback mask check is missing an and which
cauess bits that should be considered to interfere with the
result.
Reported-by: Chuck Ebbert <cebbert@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
If crypto_ablkcipher_encrypt() returns synchronous,
eseqiv_complete2() is called even if req->giv is already the
pointer to the generated IV. The generated IV is overwritten
with some random data in this case. This patch fixes this by
calling eseqiv_complete2() just if the generated IV has to be
copied to req->giv.
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
'zero_sum' does not properly describe the operation of generating parity
and checking that it validates against an existing buffer. Change the
name of the operation to 'val' (for 'validate'). This is in
anticipation of the p+q case where it is a requirement to identify the
target parity buffers separately from the source buffers, because the
target parity buffers will not have corresponding pq coefficients.
Reviewed-by: Andre Noll <maan@systemlinux.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx:
dma: Add SoF and EoF debugging to ipu_idmac.c, minor cleanup
dw_dmac: add cyclic API to DW DMA driver
dmaengine: Add privatecnt to revert DMA_PRIVATE property
dmatest: add dma interrupts and callbacks
dmatest: add xor test
dmaengine: allow dma support for async_tx to be toggled
async_tx: provide __async_inline for HAS_DMA=n archs
dmaengine: kill some unused headers
dmaengine: initialize tx_list in dma_async_tx_descriptor_init
dma: i.MX31 IPU DMA robustness improvements
dma: improve section assignment in i.MX31 IPU DMA driver
dma: ipu_idmac driver cosmetic clean-up
dmaengine: fail device registration if channel registration fails
* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
crypto: ixp4xx - Fix handling of chained sg buffers
crypto: shash - Fix unaligned calculation with short length
hwrng: timeriomem - Use phys address rather than virt
