linux-toradex.git/include/linux/crypto.h, branch v5.10-rc3

crypto: algapi - introduce the flag CRYPTO_ALG_ALLOCATES_MEMORY

2020-07-16T11:49:09+00:00

Introduce a new algorithm flag CRYPTO_ALG_ALLOCATES_MEMORY.  If this
flag is set, then the driver allocates memory in its request routine.
Such drivers are not suitable for disk encryption because GFP_ATOMIC
allocation can fail anytime (causing random I/O errors) and GFP_KERNEL
allocation can recurse into the block layer, causing a deadlock.

For now, this flag is only implemented for some algorithm types.  We
also assume some usage constraints for it to be meaningful, since there
are lots of edge cases the crypto API allows (e.g., misaligned or
fragmented scatterlists) that mean that nearly any crypto algorithm can
allocate memory in some case.  See the comment for details.

Also add this flag to CRYPTO_ALG_INHERITED_FLAGS so that when a template
is instantiated, this flag is set on the template instance if it is set
on any algorithm the instance uses.

Based on a patch by Mikulas Patocka 
(https://lore.kernel.org/r/alpine.LRH.2.02.2006301414580.30526@file01.intranet.prod.int.rdu2.redhat.com).

Signed-off-by: Eric Biggers 
Signed-off-by: Herbert Xu

crypto: algapi - add NEED_FALLBACK to INHERITED_FLAGS

2020-07-16T11:49:08+00:00

CRYPTO_ALG_NEED_FALLBACK is handled inconsistently.  When it's requested
to be clear, some templates propagate that request to child algorithms,
while others don't.

It's apparently desired for NEED_FALLBACK to be propagated, to avoid
deadlocks where a module tries to load itself while it's being
initialized, and to avoid unnecessarily complex fallback chains where we
have e.g. cbc-aes-$driver falling back to cbc(aes-$driver) where
aes-$driver itself falls back to aes-generic, instead of cbc-aes-$driver
simply falling back to cbc(aes-generic).  There have been a number of
fixes to this effect:

commit 89027579bc6c ("crypto: xts - Propagate NEED_FALLBACK bit")
commit d2c2a85cfe82 ("crypto: ctr - Propagate NEED_FALLBACK bit")
commit e6c2e65c70a6 ("crypto: cbc - Propagate NEED_FALLBACK bit")

But it seems that other templates can have the same problems too.

To avoid this whack-a-mole, just add NEED_FALLBACK to INHERITED_FLAGS so
that it's always inherited.

Signed-off-by: Eric Biggers 
Signed-off-by: Herbert Xu

crypto: api - permit users to specify numa node of acomp hardware

2020-07-09T08:25:23+00:00

For a Linux server with NUMA, there are possibly multiple (de)compressors
which are either local or remote to some NUMA node. Some drivers will
automatically use the (de)compressor near the CPU calling acomp_alloc().
However, it is not necessarily correct because users who send acomp_req
could be from different NUMA node with the CPU which allocates acomp.

Just like kernel has kmalloc() and kmalloc_node(), here crypto can have
same support.

Cc: Seth Jennings 
Cc: Dan Streetman 
Cc: Vitaly Wool 
Cc: Andrew Morton 
Cc: Jonathan Cameron 
Signed-off-by: Barry Song 
Signed-off-by: Herbert Xu

crypto: api - Prune inclusions in crypto.h

2020-06-26T04:46:43+00:00

We haven't used string.h since the memcpy calls were removed so
this patch removes its inclusion.  The file uaccess.h isn't needed
at all.  However, removing it reveals that we do need to add an
inclusion for refcount.h.

Signed-off-by: Herbert Xu

crypto: remove propagation of CRYPTO_TFM_RES_* flags

2020-01-09T03:30:53+00:00

The CRYPTO_TFM_RES_* flags were apparently meant as a way to make the
->setkey() functions provide more information about errors.  But these
flags weren't actually being used or tested, and in many cases they
weren't being set correctly anyway.  So they've now been removed.

Also, if someone ever actually needs to start better distinguishing
->setkey() errors (which is somewhat unlikely, as this has been unneeded
for a long time), we'd be much better off just defining different return
values, like -EINVAL if the key is invalid for the algorithm vs.
-EKEYREJECTED if the key was rejected by a policy like "no weak keys".
That would be much simpler, less error-prone, and easier to test.

So just remove CRYPTO_TFM_RES_MASK and all the unneeded logic that
propagates these flags around.

Signed-off-by: Eric Biggers 
Signed-off-by: Herbert Xu

crypto: remove CRYPTO_TFM_RES_WEAK_KEY

2020-01-09T03:30:53+00:00

The CRYPTO_TFM_RES_WEAK_KEY flag was apparently meant as a way to make
the ->setkey() functions provide more information about errors.

However, no one actually checks for this flag, which makes it pointless.
There are also no tests that verify that all algorithms actually set (or
don't set) it correctly.

This is also the last remaining CRYPTO_TFM_RES_* flag, which means that
it's the only thing still needing all the boilerplate code which
propagates these flags around from child => parent tfms.

And if someone ever needs to distinguish this error in the future (which
is somewhat unlikely, as it's been unneeded for a long time), it would
be much better to just define a new return value like -EKEYREJECTED.
That would be much simpler, less error-prone, and easier to test.

So just remove this flag.

Signed-off-by: Eric Biggers 
Signed-off-by: Herbert Xu

crypto: remove CRYPTO_TFM_RES_BAD_KEY_LEN

2020-01-09T03:30:53+00:00

The CRYPTO_TFM_RES_BAD_KEY_LEN flag was apparently meant as a way to
make the ->setkey() functions provide more information about errors.

However, no one actually checks for this flag, which makes it pointless.

Also, many algorithms fail to set this flag when given a bad length key.
Reviewing just the generic implementations, this is the case for
aes-fixed-time, cbcmac, echainiv, nhpoly1305, pcrypt, rfc3686, rfc4309,
rfc7539, rfc7539esp, salsa20, seqiv, and xcbc.  But there are probably
many more in arch/*/crypto/ and drivers/crypto/.

Some algorithms can even set this flag when the key is the correct
length.  For example, authenc and authencesn set it when the key payload
is malformed in any way (not just a bad length), the atmel-sha and ccree
drivers can set it if a memory allocation fails, and the chelsio driver
sets it for bad auth tag lengths, not just bad key lengths.

So even if someone actually wanted to start checking this flag (which
seems unlikely, since it's been unused for a long time), there would be
a lot of work needed to get it working correctly.  But it would probably
be much better to go back to the drawing board and just define different
return values, like -EINVAL if the key is invalid for the algorithm vs.
-EKEYREJECTED if the key was rejected by a policy like "no weak keys".
That would be much simpler, less error-prone, and easier to test.

So just remove this flag.

Signed-off-by: Eric Biggers 
Reviewed-by: Horia Geantă 
Signed-off-by: Herbert Xu

crypto: remove CRYPTO_TFM_RES_BAD_BLOCK_LEN

2020-01-09T03:30:52+00:00

The flag CRYPTO_TFM_RES_BAD_BLOCK_LEN is never checked for, and it's
only set by one driver.  And even that single driver's use is wrong
because the driver is setting the flag from ->encrypt() and ->decrypt()
with no locking, which is unsafe because ->encrypt() and ->decrypt() can
be executed by many threads in parallel on the same tfm.

Just remove this flag.

Signed-off-by: Eric Biggers 
Signed-off-by: Herbert Xu

crypto: remove unused tfm result flags

2020-01-09T03:30:52+00:00

The tfm result flags CRYPTO_TFM_RES_BAD_KEY_SCHED and
CRYPTO_TFM_RES_BAD_FLAGS are never used, so remove them.

Signed-off-by: Eric Biggers 
Signed-off-by: Herbert Xu

crypto: api - remove unneeded semicolon

2019-12-27T10:18:02+00:00

Fixes coccicheck warning:

./include/linux/crypto.h:573:2-3: Unneeded semicolon

Signed-off-by: Chen Zhou 
Signed-off-by: Herbert Xu