summaryrefslogtreecommitdiffstats
path: root/crypto/lskcipher.c (follow)
Commit message (Collapse)AuthorAgeFilesLines
* crypto: skcipher - Pass statesize for simple lskcipher instancesHerbert Xu2023-12-221-0/+1
| | | | | | | | | | | When ecb is used to wrap an lskcipher, the statesize isn't set correctly. Fix this by making the simple instance creator set the statesize. Reported-by: syzbot+8ffb0839a24e9c6bfa76@syzkaller.appspotmail.com Reported-by: Edward Adam Davis <eadavis@qq.com> Fixes: 662ea18d089b ("crypto: skcipher - Make use of internal state") Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: skcipher - Make use of internal stateHerbert Xu2023-12-081-6/+28
| | | | | | | | | This patch adds code to the skcipher/lskcipher API to make use of the internal state if present. In particular, the skcipher lskcipher wrapper will allocate a buffer for the IV/state and feed that to the underlying lskcipher algorithm. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: skcipher - Add internal state supportHerbert Xu2023-12-081-6/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Unlike chaining modes such as CBC, stream ciphers other than CTR usually hold an internal state that must be preserved if the operation is to be done piecemeal. This has not been represented in the API, resulting in the inability to split up stream cipher operations. This patch adds the basic representation of an internal state to skcipher and lskcipher. In the interest of backwards compatibility, the default has been set such that existing users are assumed to be operating in one go as opposed to piecemeal. With the new API, each lskcipher/skcipher algorithm has a new attribute called statesize. For skcipher, this is the size of the buffer that can be exported or imported similar to ahash. For lskcipher, instead of providing a buffer of ivsize, the user now has to provide a buffer of ivsize + statesize. Each skcipher operation is assumed to be final as they are now, but this may be overridden with a request flag. When the override occurs, the user may then export the partial state and reimport it later. For lskcipher operations this is reversed. All operations are not final and the state will be exported unless the FINAL bit is set. However, the CONT bit still has to be set for the state to be used. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: skcipher - fix weak key check for lskciphersEric Biggers2023-10-201-8/+0
| | | | | | | | | | | | When an algorithm of the new "lskcipher" type is exposed through the "skcipher" API, calls to crypto_skcipher_setkey() don't pass on the CRYPTO_TFM_REQ_FORBID_WEAK_KEYS flag to the lskcipher. This causes self-test failures for ecb(des), as weak keys are not rejected anymore. Fix this. Fixes: 31865c4c4db2 ("crypto: skcipher - Add lskcipher") Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: lskcipher - Return EINVAL when ecb_name fails sanity checksHerbert Xu2023-10-201-0/+1
| | | | | | | | | | | Set the error value to -EINVAL instead of zero when the underlying name (within "ecb()") fails basic sanity checks. Fixes: 8aee5d4ebd11 ("crypto: lskcipher - Add compatibility wrapper around ECB") Reported-by: kernel test robot <lkp@intel.com> Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Closes: https://lore.kernel.org/r/202310111323.ZjK7bzjw-lkp@intel.com/ Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: lskcipher - Add compatibility wrapper around ECBHerbert Xu2023-09-201-5/+52
| | | | | | | | | | | | As an aid to the transition from cipher algorithm implementations to lskcipher, add a temporary wrapper when creating simple lskcipher templates by using ecb(X) instead of X if an lskcipher implementation of X cannot be found. This can be reverted once all cipher implementations have switched over to lskcipher. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* crypto: skcipher - Add lskcipherHerbert Xu2023-09-201-0/+594
Add a new API type lskcipher designed for taking straight kernel pointers instead of SG lists. Its relationship to skcipher will be analogous to that between shash and ahash. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>