22 Aug, 2008
1 commit
-
Authenc works in two stages for encryption, it first encrypts and
then computes an ICV. The context memory of the request is used
by both operations. The problem is that when an asynchronous
encryption completes, we will compute the ICV and then reread the
context memory of the encryption to get the original request.It just happens that we have a buffer of 16 bytes in front of the
request pointer, so ICVs of 16 bytes (such as SHA1) do not trigger
the bug. However, any attempt to uses a larger ICV instantly kills
the machine when the first asynchronous encryption is completed.This patch fixes this by saving the request pointer before we start
the ICV computation.Signed-off-by: Herbert Xu
13 Aug, 2008
2 commits
-
The changeset ca786dc738f4f583b57b1bba7a335b5e8233f4b0
crypto: hash - Fixed digest size check
missed one spot for the digest type. This patch corrects that
error.Signed-off-by: Herbert Xu
-
My changeset 4b22f0ddb6564210c9ded7ba25b2a1007733e784
crypto: tcrpyt - Remove unnecessary kmap/kunmap calls
introduced a typo that broke AEAD chunk testing. In particular,
axbuf should really be xbuf.There is also an issue with testing the last segment when encrypting.
The additional part produced by AEAD wasn't tested. Similarly, on
decryption the additional part of the AEAD input is mistaken for
corruption.Signed-off-by: Herbert Xu
06 Aug, 2008
1 commit
-
Found-by: Yuri Tikhonov
Signed-off-by: Dan Williams
24 Jul, 2008
1 commit
-
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx: (24 commits)
I/OAT: I/OAT version 3.0 support
I/OAT: tcp_dma_copybreak default value dependent on I/OAT version
I/OAT: Add watchdog/reset functionality to ioatdma
iop_adma: cleanup iop_chan_xor_slot_count
iop_adma: document how to calculate the minimum descriptor pool size
iop_adma: directly reclaim descriptors on allocation failure
async_tx: make async_tx_test_ack a boolean routine
async_tx: remove depend_tx from async_tx_sync_epilog
async_tx: export async_tx_quiesce
async_tx: fix handling of the "out of descriptor" condition in async_xor
async_tx: ensure the xor destination buffer remains dma-mapped
async_tx: list_for_each_entry_rcu() cleanup
dmaengine: Driver for the Synopsys DesignWare DMA controller
dmaengine: Add slave DMA interface
dmaengine: add DMA_COMPL_SKIP_{SRC,DEST}_UNMAP flags to control dma unmap
dmaengine: Add dma_client parameter to device_alloc_chan_resources
dmatest: Simple DMA memcpy test client
dmaengine: DMA engine driver for Marvell XOR engine
iop-adma: fix platform driver hotplug/coldplug
dmaengine: track the number of clients using a channel
...Fixed up conflict in drivers/dca/dca-sysfs.c manually
18 Jul, 2008
5 commits
-
All callers of async_tx_sync_epilog have called async_tx_quiesce on the
depend_tx, so async_tx_sync_epilog need only call the callback to
complete the operation.Signed-off-by: Dan Williams
-
Replace open coded "wait and acknowledge" instances with async_tx_quiesce.
Signed-off-by: Dan Williams
-
Ensure forward progress is made when a dmaengine driver is unable to
allocate an xor descriptor by breaking the dependency chain with
async_tx_quisce() and issue any pending descriptors.Tested with iop-adma by setting device->max_xor = 2 to force multiple
calls to device_prep_dma_xor for each call to async_xor and limiting the
descriptor slot pool to 5. Discovered that the minimum descriptor pool
size for iop-adma is 2 * iop_chan_xor_slot_cnt(device->max_xor) + 1.Signed-off-by: Dan Williams
-
When the number of source buffers for an xor operation exceeds the hardware
channel maximum async_xor creates a chain of dependent operations. The result
of one operation is reused as an input to the next to continue the xor
calculation. The destination buffer should remain mapped for the duration of
the entire chain. To provide this guarantee the code must no longer be allowed
to fallback to the synchronous path as this will preclude the buffer from being
unmapped, i.e. the dma-driver will potentially miss the descriptor with
!DMA_COMPL_SKIP_DEST_UNMAP.Cc: Neil Brown
Signed-off-by: Dan Williams -
In the rcu update side, don't use list_for_each_entry_rcu().
Signed-off-by: Li Zefan
Signed-off-by: Dan Williams
16 Jul, 2008
2 commits
-
* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
Revert crypto: prng - Deterministic CPRNG
15 Jul, 2008
1 commit
-
This patch is clearly not ready yet for prime time.
Signed-off-by: Herbert Xu
11 Jul, 2008
1 commit
-
Conflicts:
include/linux/rculist.h
kernel/rcupreempt.cSigned-off-by: Ingo Molnar
10 Jul, 2008
26 commits
-
All new crypto interfaces should go into individual files as much
as possible in order to ensure that crypto.h does not collapse under
its own weight.This patch moves the ahash code into crypto/hash.h and crypto/internal/hash.h
respectively.Signed-off-by: Herbert Xu
-
This patch reimplements crc32c using the ahash interface. This
allows one tfm to be used by an unlimited number of users provided
that they all use the same key (which all current crc32c users do).Signed-off-by: Herbert Xu
-
This patch adds the walking helpers for hash algorithms akin to
those of block ciphers. This is a necessary step before we can
reimplement existing hash algorithms using the new ahash interface.Signed-off-by: Herbert Xu
-
This patch adds a cryptographic pseudo-random number generator
based on CTR(AES-128). It is meant to be used in cases where a
deterministic CPRNG is required.One of the first applications will be as an input in the IPsec IV
generation process.Signed-off-by: Neil Horman
Signed-off-by: Herbert Xu -
The base field in ahash_tfm appears to have been cut-n-pasted from
ablkcipher. It isn't needed here at all. Similarly, the info field
in ahash_request also appears to have originated from its cipher
counter-part and is vestigial.Signed-off-by: Herbert Xu
-
The digest size check on hash algorithms is incorrect. It's
perfectly valid for hash algorithms to have a digest length
longer than their block size. For example crc32c has a block
size of 1 and a digest size of 4. Rather than having it lie
about its block size, this patch fixes the checks to do what
they really should which is to bound the digest size so that
code placing the digest on the stack continue to work.HMAC however still needs to check this as it's only defined
for such algorithms.Signed-off-by: Herbert Xu
-
Similar to the rmd128.c annotations, significantly cuts down on the
noise.Signed-off-by: Harvey Harrison
Signed-off-by: Herbert Xu -
Signed-off-by: Harvey Harrison
Signed-off-by: Herbert Xu -
Remove the private implementation of 32-bit rotation and unaligned
access with byteswapping.As a bonus, fixes sparse warnings:
crypto/camellia.c:602:2: warning: cast to restricted __be32
crypto/camellia.c:603:2: warning: cast to restricted __be32
crypto/camellia.c:604:2: warning: cast to restricted __be32
crypto/camellia.c:605:2: warning: cast to restricted __be32
crypto/camellia.c:710:2: warning: cast to restricted __be32
crypto/camellia.c:711:2: warning: cast to restricted __be32
crypto/camellia.c:712:2: warning: cast to restricted __be32
crypto/camellia.c:713:2: warning: cast to restricted __be32
crypto/camellia.c:714:2: warning: cast to restricted __be32
crypto/camellia.c:715:2: warning: cast to restricted __be32
crypto/camellia.c:716:2: warning: cast to restricted __be32
crypto/camellia.c:717:2: warning: cast to restricted __be32Signed-off-by: Harvey Harrison
Signed-off-by: Herbert Xu -
Noticed by Neil Horman: we are doing unnecessary kmap/kunmap calls
on kmalloced memory. This patch removes them. For the purposes of
testing SG construction, the underlying crypto code already does plenty
of kmap/kunmap calls anyway.Signed-off-by: Herbert Xu
-
Patch to add checking of DES3 test vectors using CBC mode. FIPS-140-2
compliance mandates that any supported mode of operation must include a self
test. This satisfies that requirement for cbc(des3_ede). The included test
vector was generated by me using openssl. Key/IV was generated with the
following command:openssl enc -des_ede_cbc -P
input and output values were generated by repeating the string "Too many
secrets" a few times over, truncating it to 128 bytes, and encrypting it with
openssl using the aformentioned key. Tested successfully by myselfSigned-off-by: Neil Horman
Acked-by: Adrian-Ken Rueegsegger
Signed-off-by: Herbert Xu -
This patch converts the relevant code in the rmd implementations to
use the pointer form of the endian swapping operations. This allows
certain architectures to generate more optimised code. For example,
on sparc64 this more than halves the CPU cycles on a typical hashing
operation.Based on a patch by David Miller.
Signed-off-by: Herbert Xu
-
This patch fixes endian issues making rmd320 work
properly on big-endian machines.Signed-off-by: Adrian-Ken Rueegsegger
Acked-by: Sebastian Siewior
Signed-off-by: Herbert Xu -
This patch fixes endian issues making rmd256 work
properly on big-endian machines.Signed-off-by: Adrian-Ken Rueegsegger
Acked-by: Sebastian Siewior
Signed-off-by: Herbert Xu -
This patch fixes endian issues making rmd160 work
properly on big-endian machines.Signed-off-by: Adrian-Ken Rueegsegger
Acked-by: Sebastian Siewior
Signed-off-by: Herbert Xu -
This patch is based on Sebastian Siewior's patch and
fixes endian issues making rmd128 work properly on
big-endian machines.Signed-off-by: Adrian-Ken Rueegsegger
Acked-by: Sebastian Siewior
Signed-off-by: Herbert Xu -
This patch changes tcrypt to use the new asynchronous hash interface
for testing hash algorithm correctness. The speed tests will continue
to use the existing interface for now.Signed-off-by: Loc Ho
Signed-off-by: Herbert Xu -
This patch adds asynchronous hash support to crypto daemon.
Signed-off-by: Loc Ho
Signed-off-by: Herbert Xu -
This patch adds asynchronous hash and digest support.
Signed-off-by: Loc Ho
Signed-off-by: Herbert Xu -
This patch adds Kconfig entries for RIPEMD-256 and RIPEMD-320.
Signed-off-by: Adrian-Ken Rueegsegger
Signed-off-by: Herbert Xu -
This patch adds test vectors for RIPEMD-256 and
RIPEMD-320 hash algorithms.The test vectors are taken from
Signed-off-by: Adrian-Ken Rueegsegger
Signed-off-by: Herbert Xu -
This patch adds support for the extended RIPEMD hash
algorithms RIPEMD-256 and RIPEMD-320.Signed-off-by: Adrian-Ken Rueegsegger
Signed-off-by: Herbert Xu -
This patch puts all common RIPEMD values in the
appropriate header file. Initial values and constants
are the same for all variants of RIPEMD.Signed-off-by: Adrian-Ken Rueegsegger
Signed-off-by: Herbert Xu -
Check whether the destination buffer is written to beyond the last
byte contained in the scatterlist.Also change IDX1 of the cross-page access offsets to a multiple of 4.
This triggers a corruption in the HIFN driver and doesn't seem to
negatively impact other testcases.Signed-off-by: Patrick McHardy
Signed-off-by: Herbert Xu -
Change logs should be kept in source control systems, not the source.
This patch removes the change log from tcrpyt to stop people from
extending it any more.Signed-off-by: Herbert Xu
-
This patch adds Kconfig entries for RIPEMD-128 and RIPEMD-160.
Signed-off-by: Adrian-Ken Rueegsegger
Signed-off-by: Herbert Xu
