linux-stable.git/drivers/crypto/nx, branch linux-4.1.y

crypto: nx - respect sg limit bounds when building sg lists for SHA

2015-09-13T16:07:45+00:00

commit d3392f41f6d3cd0a034bd0aca47fabea2b47218e upstream.

Commit 000851119e80 changed sha256/512 update functions to
pass more data to nx_build_sg_list(), which ends with
sg list overflows and usually with update functions failing
for data larger than max_sg_len * NX_PAGE_SIZE.

This happens because:
- both "total" and "to_process" are updated, which leads to
  "to_process" getting overflowed for some data lengths
  For example:
    In first iteration "total" is 50, and let's assume "to_process"
    is 30 due to sg limits. At the end of first iteration "total" is
    set to 20. At start of 2nd iteration "to_process" overflows on:
      to_process = total - to_process;
- "in_sg" is not reset to nx_ctx->in_sg after each iteration
- nx_build_sg_list() is hitting overflow because the amount of data
  passed to it would require more than sgmax elements
- as consequence of previous item, data stored in overflowed sg list
  may no longer be aligned to SHA*_BLOCK_SIZE

This patch changes sha256/512 update functions so that "to_process"
respects sg limits and never tries to pass more data to
nx_build_sg_list() to avoid overflows. "to_process" is calculated
as minimum of "total" and sg limits at start of every iteration.

Fixes: 000851119e80 ("crypto: nx - Fix SHA concurrence issue and sg
		      limit bounds")
Signed-off-by: Jan Stancek 
Cc: Leonidas Da Silva Barbosa 
Cc: Marcelo Henrique Cerri 
Cc: Fionnuala Gunter 
Cc: "David S. Miller" 
Signed-off-by: Herbert Xu 
Signed-off-by: Greg Kroah-Hartman

crypto: nx - Fix reentrancy bugs

2015-08-17T03:52:16+00:00

commit 030f4e968741d65aea9cd5f7814d1164967801ef upstream.

This patch fixes a host of reentrancy bugs in the nx driver.  The
following algorithms are affected:

* CCM
* GCM
* CTR
* XCBC
* SHA256
* SHA512

The crypto API allows a single transform to be used by multiple
threads simultaneously.  For example, IPsec will use a single tfm
to process packets for a given SA.  As packets may arrive on
multiple CPUs that tfm must be reentrant.

The nx driver does try to deal with this by using a spin lock.
Unfortunately only the basic AES/CBC/ECB algorithms do this in
the correct way.

The symptom of these bugs may range from the generation of incorrect
output to memory corruption.

Signed-off-by: Herbert Xu 
Signed-off-by: Greg Kroah-Hartman

crypto: nx - Fixing SHA update bug

2015-08-17T03:52:16+00:00

commit 10d87b730e1d9f1442cae6487bb3aef8632bed23 upstream.

Bug happens when a data size less than SHA block size is passed.
Since first attempt will be saved in buffer, second round attempt
get into two step to calculate op.inlen and op.outlen. The issue
resides in this step. A  wrong value of op.inlen and outlen was being
calculated.

This patch fix this eliminate the nx_sha_build_sg_list, that is
useless in SHA's algorithm context. Instead we call nx_build_sg_list
directly and pass a previous calculated max_sg_len to it.

Signed-off-by: Leonidas S. Barbosa 
Signed-off-by: Herbert Xu 
Signed-off-by: Greg Kroah-Hartman

crypto: nx - Fixing NX data alignment with nx_sg list

2015-08-17T03:52:16+00:00

commit c3365ce130e50176533debe1cabebcdb8e61156c upstream.

In NX we need to pass always a 16 multiple size nx_sg_list to
co processor. Trim function handle with this assuring all nx_sg_lists
are 16 multiple size, although data was not being considerated when
crop was done. It was causing an unalignment between size of the list
and data, corrupting csbcpb fields returning a -23 H_ST_PARM error, or
invalid operation.

This patch fix this recalculating how much data should be put back
in to_process variable what assures the size of sg_list will be
correct with size of the data.

Signed-off-by: Leonidas S. Barbosa 
Signed-off-by: Herbert Xu 
Signed-off-by: Greg Kroah-Hartman

crypto: replace scatterwalk_sg_next with sg_next

2015-01-26T00:34:22+00:00

Modify crypto drivers to use the generic SG helper since
both of them are equivalent and the one from crypto is redundant.

See also:
  468577abe37ff7b453a9ac613e0ea155349203ae   reverted in
  b2ab4a57b018aafbba35bff088218f5cc3d2142e

Signed-off-by: Cristian Stoica 
Signed-off-by: Herbert Xu

Merge git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6

2014-12-13T21:33:26+00:00

Pull crypto update from Herbert Xu:
 - The crypto API is now documented :)
 - Disallow arbitrary module loading through crypto API.
 - Allow get request with empty driver name through crypto_user.
 - Allow speed testing of arbitrary hash functions.
 - Add caam support for ctr(aes), gcm(aes) and their derivatives.
 - nx now supports concurrent hashing properly.
 - Add sahara support for SHA1/256.
 - Add ARM64 version of CRC32.
 - Misc fixes.

* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (77 commits)
  crypto: tcrypt - Allow speed testing of arbitrary hash functions
  crypto: af_alg - add user space interface for AEAD
  crypto: qat - fix problem with coalescing enable logic
  crypto: sahara - add support for SHA1/256
  crypto: sahara - replace tasklets with kthread
  crypto: sahara - add support for i.MX53
  crypto: sahara - fix spinlock initialization
  crypto: arm - replace memset by memzero_explicit
  crypto: powerpc - replace memset by memzero_explicit
  crypto: sha - replace memset by memzero_explicit
  crypto: sparc - replace memset by memzero_explicit
  crypto: algif_skcipher - initialize upon init request
  crypto: algif_skcipher - removed unneeded code
  crypto: algif_skcipher - Fixed blocking recvmsg
  crypto: drbg - use memzero_explicit() for clearing sensitive data
  crypto: drbg - use MODULE_ALIAS_CRYPTO
  crypto: include crypto- module prefix in template
  crypto: user - add MODULE_ALIAS
  crypto: sha-mb - remove a bogus NULL check
  crytpo: qat - Fix 64 bytes requests
  ...

of/reconfig: Always use the same structure for notifiers

2014-11-24T22:25:03+00:00

The OF_RECONFIG notifier callback uses a different structure depending
on whether it is a node change or a property change. This is silly, and
not very safe. Rework the code to use the same data structure regardless
of the type of notifier.

Signed-off-by: Grant Likely 
Cc: Benjamin Herrenschmidt 
Cc: Rob Herring 
Cc: Pantelis Antoniou 
Cc:

crypto: nx - Fixing the limit number of bytes to be processed

2014-11-06T15:15:03+00:00

The previous limits were estimated locally in a single step
basead on bound values, however it was not correct since
when given certain scatterlist the function nx_build_sg_lists
was consuming more sg entries than allocated causing a
memory corruption and crashes.

e.g.: in the worst case we could have one sg entry for a single byte.

This patch fixes it modifying the logic of the bound limit
moving it to nx_sg_build_lists and set a correct sg_max limit,
adding a trim function to ensure the bound in sg_list. Also fixing
nx_build_sg_list  NULL and untreated return in case of overflow.

Signed-off-by: Leonidas S. Barbosa 
Signed-off-by: Herbert Xu

crypto: nx - Fix SHA concurrence issue and sg limit bounds

2014-11-06T15:15:02+00:00

NX SHA algorithms stores the message digest into tfm what
cause a concurrence issue where hashes may be replaced by others.
This patch cleans up the cases where it's handling unnecessarily shared
variables in nx context and copies the current msg digest to a sctx->state
in order to safetly handle with the hashe's state.

Also fixes and does some clean ups regarding the right sg max limit
and bounds to the sg list avoind a memory crash.

Signed-off-by: Leonidas S. Barbosa 
Signed-off-by: Herbert Xu

crypto: nx - Moving NX-AES-XCBC to be processed logic

2014-11-06T15:15:02+00:00

The previous limits were estimated locally in a single step
basead on bound values, however it was not correct since
when given certain scatterlist the function nx_build_sg_lists
was consuming more sg entries than allocated causing a
memory corruption and crashes.

This patch removes the old logic and replaces it into nx_sg_build_lists
in order to build a correct nx_sg list using the correct sg_max limit
and bounds.

Signed-off-by: Leonidas S. Barbosa 
Signed-off-by: Herbert Xu