The ablkcipher structure in ablk_ecb_alg is lacking an ivsize setting.
This causes the algorithm to fail it's self-test when it's registered
with the error message:
[ 0.806920] alg: skcipher: Failed to load transform for ecb-aes-aesni: -2
crypto_lookup_skcipher doesn't like a missing ivsize. Adding the ivsize
allows the self-test lookup to pass.
Signed-off-by: Josh Boyer <[email protected]>
---
arch/x86/crypto/aesni-intel_glue.c | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c
index feee8ff..711cc3c 100644
--- a/arch/x86/crypto/aesni-intel_glue.c
+++ b/arch/x86/crypto/aesni-intel_glue.c
@@ -587,6 +587,7 @@ static struct crypto_alg ablk_ecb_alg = {
.ablkcipher = {
.min_keysize = AES_MIN_KEY_SIZE,
.max_keysize = AES_MAX_KEY_SIZE,
+ .ivsize = AES_BLOCK_SIZE,
.setkey = ablk_set_key,
.encrypt = ablk_encrypt,
.decrypt = ablk_decrypt,
--
1.7.6
On Thu, Aug 11, 2011 at 12:40:41PM -0400, Josh Boyer wrote:
> The ablkcipher structure in ablk_ecb_alg is lacking an ivsize setting.
> This causes the algorithm to fail it's self-test when it's registered
> with the error message:
>
> [ 0.806920] alg: skcipher: Failed to load transform for ecb-aes-aesni: -2
>
> crypto_lookup_skcipher doesn't like a missing ivsize. Adding the ivsize
> allows the self-test lookup to pass.
That's not right. ECB does not require an IV and therefore it
shouldn't have a non-zero IV size.
I'm not quite sure what you mean by crypto_lookup_skcipher not
liking a missing IV size. It appears to handle that case just
fine by returning the algorithm as is.
Can you show me what your /proc/crypto contains after you get the
aformentioned error?
Thanks,
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
On Mon, Aug 15, 2011 at 03:15:28PM +0800, Herbert Xu wrote:
> On Thu, Aug 11, 2011 at 12:40:41PM -0400, Josh Boyer wrote:
> > The ablkcipher structure in ablk_ecb_alg is lacking an ivsize setting.
> > This causes the algorithm to fail it's self-test when it's registered
> > with the error message:
> >
> > [ 0.806920] alg: skcipher: Failed to load transform for ecb-aes-aesni: -2
> >
> > crypto_lookup_skcipher doesn't like a missing ivsize. Adding the ivsize
> > allows the self-test lookup to pass.
>
> That's not right. ECB does not require an IV and therefore it
> shouldn't have a non-zero IV size.
Hm. OK.
> I'm not quite sure what you mean by crypto_lookup_skcipher not
> liking a missing IV size. It appears to handle that case just
> fine by returning the algorithm as is.
Well, yes, it does in this code:
if (!((alg->cra_flags & CRYPTO_ALG_TYPE_MASK) ==
CRYPTO_ALG_TYPE_BLKCIPHER ? alg->cra_blkcipher.ivsize :
alg->cra_ablkcipher.ivsize)) {
printk(KERN_ERR "%s: %d\n", __func__, __LINE__);
return alg;
}
But the lookup seems to fail after it returns (see traces below).
> Can you show me what your /proc/crypto contains after you get the
> aformentioned error?
Sure.
Without the patch:
[jwboyer@zod ~]$ cat /proc/crypto
name : arc4
driver : arc4-generic
module : arc4
priority : 0
refcnt : 3
selftest : passed
type : cipher
blocksize : 1
min keysize : 1
max keysize : 256
name : fpu(xts(__aes-aesni))
driver : cryptd(fpu(xts(__driver-aes-aesni)))
module : kernel
priority : 50
refcnt : 9
selftest : passed
type : ablkcipher
async : yes
blocksize : 16
min keysize : 32
max keysize : 64
ivsize : 16
geniv : <default>
name : fpu(xts(__aes-aesni))
driver : fpu(xts(__driver-aes-aesni))
module : kernel
priority : 0
refcnt : 9
selftest : passed
type : blkcipher
blocksize : 16
min keysize : 32
max keysize : 64
ivsize : 16
geniv : <default>
name : xts(__aes-aesni)
driver : xts(__driver-aes-aesni)
module : xts
priority : 0
refcnt : 9
selftest : passed
type : blkcipher
blocksize : 16
min keysize : 32
max keysize : 64
ivsize : 16
geniv : <default>
name : xts(aes)
driver : xts-aes-aesni
module : kernel
priority : 400
refcnt : 9
selftest : passed
type : givcipher
async : yes
blocksize : 16
min keysize : 32
max keysize : 64
ivsize : 16
geniv : eseqiv
name : stdrng
driver : krng
module : kernel
priority : 200
refcnt : 2
selftest : passed
type : rng
seedsize : 0
name : crc32c
driver : crc32c-generic
module : kernel
priority : 100
refcnt : 1
selftest : passed
type : shash
blocksize : 1
digestsize : 4
name : aes
driver : aes-generic
module : kernel
priority : 100
refcnt : 1
selftest : passed
type : cipher
blocksize : 16
min keysize : 16
max keysize : 32
name : sha1
driver : sha1-generic
module : kernel
priority : 0
refcnt : 1
selftest : passed
type : shash
blocksize : 64
digestsize : 20
name : md5
driver : md5-generic
module : kernel
priority : 0
refcnt : 1
selftest : passed
type : shash
blocksize : 64
digestsize : 16
name : xts(aes)
driver : xts-aes-aesni
module : kernel
priority : 400
refcnt : 9
selftest : passed
type : ablkcipher
async : yes
blocksize : 16
min keysize : 32
max keysize : 64
ivsize : 16
geniv : <default>
name : pcbc(aes)
driver : pcbc-aes-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : ablkcipher
async : yes
blocksize : 16
min keysize : 16
max keysize : 32
ivsize : 16
geniv : <default>
name : lrw(aes)
driver : lrw-aes-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : ablkcipher
async : yes
blocksize : 16
min keysize : 32
max keysize : 48
ivsize : 16
geniv : <default>
name : rfc3686(ctr(aes))
driver : rfc3686-ctr-aes-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : ablkcipher
async : yes
blocksize : 1
min keysize : 20
max keysize : 36
ivsize : 8
geniv : seqiv
name : rfc4106(gcm(aes))
driver : rfc4106-gcm-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : nivaead
async : yes
blocksize : 1
ivsize : 8
maxauthsize : 16
geniv : seqiv
name : __gcm-aes-aesni
driver : __driver-gcm-aes-aesni
module : kernel
priority : 0
refcnt : 1
selftest : passed
type : aead
async : no
blocksize : 1
ivsize : 0
maxauthsize : 0
geniv : <built-in>
name : ctr(aes)
driver : ctr-aes-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : ablkcipher
async : yes
blocksize : 1
min keysize : 16
max keysize : 32
ivsize : 16
geniv : chainiv
name : __ctr-aes-aesni
driver : __driver-ctr-aes-aesni
module : kernel
priority : 0
refcnt : 1
selftest : passed
type : blkcipher
blocksize : 1
min keysize : 16
max keysize : 32
ivsize : 16
geniv : <default>
name : cbc(aes)
driver : cbc-aes-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : ablkcipher
async : yes
blocksize : 16
min keysize : 16
max keysize : 32
ivsize : 16
geniv : <default>
name : ecb(aes)
driver : ecb-aes-aesni
module : kernel
priority : 400
refcnt : 1
selftest : unknown
type : ablkcipher
async : yes
blocksize : 16
min keysize : 16
max keysize : 32
ivsize : 0
geniv : <default>
name : __cbc-aes-aesni
driver : __driver-cbc-aes-aesni
module : kernel
priority : 0
refcnt : 1
selftest : passed
type : blkcipher
blocksize : 16
min keysize : 16
max keysize : 32
ivsize : 0
geniv : <default>
name : __ecb-aes-aesni
driver : __driver-ecb-aes-aesni
module : kernel
priority : 0
refcnt : 1
selftest : passed
type : blkcipher
blocksize : 16
min keysize : 16
max keysize : 32
ivsize : 0
geniv : <default>
name : __aes-aesni
driver : __driver-aes-aesni
module : kernel
priority : 0
refcnt : 17
selftest : passed
type : cipher
blocksize : 16
min keysize : 16
max keysize : 32
name : aes
driver : aes-aesni
module : kernel
priority : 300
refcnt : 1
selftest : passed
type : cipher
blocksize : 16
min keysize : 16
max keysize : 32
name : aes
driver : aes-asm
module : kernel
priority : 200
refcnt : 1
selftest : passed
type : cipher
blocksize : 16
min keysize : 16
max keysize : 32
With the patch:
[jwboyer@zod ~]$ cat /proc/crypto
name : arc4
driver : arc4-generic
module : arc4
priority : 0
refcnt : 3
selftest : passed
type : cipher
blocksize : 1
min keysize : 1
max keysize : 256
name : fpu(xts(__aes-aesni))
driver : cryptd(fpu(xts(__driver-aes-aesni)))
module : kernel
priority : 50
refcnt : 9
selftest : passed
type : ablkcipher
async : yes
blocksize : 16
min keysize : 32
max keysize : 64
ivsize : 16
geniv : <default>
name : fpu(xts(__aes-aesni))
driver : fpu(xts(__driver-aes-aesni))
module : kernel
priority : 0
refcnt : 9
selftest : passed
type : blkcipher
blocksize : 16
min keysize : 32
max keysize : 64
ivsize : 16
geniv : <default>
name : xts(__aes-aesni)
driver : xts(__driver-aes-aesni)
module : xts
priority : 0
refcnt : 9
selftest : passed
type : blkcipher
blocksize : 16
min keysize : 32
max keysize : 64
ivsize : 16
geniv : <default>
name : xts(aes)
driver : xts-aes-aesni
module : kernel
priority : 400
refcnt : 9
selftest : passed
type : givcipher
async : yes
blocksize : 16
min keysize : 32
max keysize : 64
ivsize : 16
geniv : eseqiv
name : stdrng
driver : krng
module : kernel
priority : 200
refcnt : 2
selftest : passed
type : rng
seedsize : 0
name : crc32c
driver : crc32c-generic
module : kernel
priority : 100
refcnt : 1
selftest : passed
type : shash
blocksize : 1
digestsize : 4
name : aes
driver : aes-generic
module : kernel
priority : 100
refcnt : 1
selftest : passed
type : cipher
blocksize : 16
min keysize : 16
max keysize : 32
name : sha1
driver : sha1-generic
module : kernel
priority : 0
refcnt : 1
selftest : passed
type : shash
blocksize : 64
digestsize : 20
name : md5
driver : md5-generic
module : kernel
priority : 0
refcnt : 1
selftest : passed
type : shash
blocksize : 64
digestsize : 16
name : xts(aes)
driver : xts-aes-aesni
module : kernel
priority : 400
refcnt : 9
selftest : passed
type : ablkcipher
async : yes
blocksize : 16
min keysize : 32
max keysize : 64
ivsize : 16
geniv : <default>
name : pcbc(aes)
driver : pcbc-aes-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : ablkcipher
async : yes
blocksize : 16
min keysize : 16
max keysize : 32
ivsize : 16
geniv : <default>
name : lrw(aes)
driver : lrw-aes-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : ablkcipher
async : yes
blocksize : 16
min keysize : 32
max keysize : 48
ivsize : 16
geniv : <default>
name : rfc3686(ctr(aes))
driver : rfc3686-ctr-aes-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : ablkcipher
async : yes
blocksize : 1
min keysize : 20
max keysize : 36
ivsize : 8
geniv : seqiv
name : rfc4106(gcm(aes))
driver : rfc4106-gcm-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : nivaead
async : yes
blocksize : 1
ivsize : 8
maxauthsize : 16
geniv : seqiv
name : __gcm-aes-aesni
driver : __driver-gcm-aes-aesni
module : kernel
priority : 0
refcnt : 1
selftest : passed
type : aead
async : no
blocksize : 1
ivsize : 0
maxauthsize : 0
geniv : <built-in>
name : ctr(aes)
driver : ctr-aes-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : ablkcipher
async : yes
blocksize : 1
min keysize : 16
max keysize : 32
ivsize : 16
geniv : chainiv
name : __ctr-aes-aesni
driver : __driver-ctr-aes-aesni
module : kernel
priority : 0
refcnt : 1
selftest : passed
type : blkcipher
blocksize : 1
min keysize : 16
max keysize : 32
ivsize : 16
geniv : <default>
name : cbc(aes)
driver : cbc-aes-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : ablkcipher
async : yes
blocksize : 16
min keysize : 16
max keysize : 32
ivsize : 16
geniv : <default>
name : ecb(aes)
driver : ecb-aes-aesni
module : kernel
priority : 400
refcnt : 1
selftest : passed
type : ablkcipher
async : yes
blocksize : 16
min keysize : 16
max keysize : 32
ivsize : 16
geniv : <default>
name : __cbc-aes-aesni
driver : __driver-cbc-aes-aesni
module : kernel
priority : 0
refcnt : 1
selftest : passed
type : blkcipher
blocksize : 16
min keysize : 16
max keysize : 32
ivsize : 0
geniv : <default>
name : __ecb-aes-aesni
driver : __driver-ecb-aes-aesni
module : kernel
priority : 0
refcnt : 1
selftest : passed
type : blkcipher
blocksize : 16
min keysize : 16
max keysize : 32
ivsize : 0
geniv : <default>
name : __aes-aesni
driver : __driver-aes-aesni
module : kernel
priority : 0
refcnt : 17
selftest : passed
type : cipher
blocksize : 16
min keysize : 16
max keysize : 32
name : aes
driver : aes-aesni
module : kernel
priority : 300
refcnt : 1
selftest : passed
type : cipher
blocksize : 16
min keysize : 16
max keysize : 32
name : aes
driver : aes-asm
module : kernel
priority : 200
refcnt : 1
selftest : passed
type : cipher
blocksize : 16
min keysize : 16
max keysize : 32
And below is an instrumented kernel with some printks leading up to the
failure.
josh
Aug 11 11:31:34 zod kernel: [ 0.799743] aesni_init: 1275 registering
ablk_ecb_alg
Aug 11 11:31:34 zod kernel: [ 0.799792] __crypto_register_alg: 235
ecb(aes)
Aug 11 11:31:34 zod kernel: [ 0.799862] crypto_alg_mod_lookup: 278
ecb-aes-aesni
Aug 11 11:31:34 zod kernel: [ 0.799910] crypto_lookup_skcipher: 576
Aug 11 11:31:34 zod kernel: [ 0.801540] crypto_alg_mod_lookup: 286
cryptd(__driver-ecb-aes-aesni)
Aug 11 11:31:34 zod kernel: [ 0.802260] crypto_larval_wait: 192
Aug 11 11:31:34 zod kernel: [ 0.802307] crypto_alloc_base: 452
Aug 11 11:31:34 zod kernel: [ 0.802352] cryptd_alloc_ablkcipher: 828
Aug 11 11:31:34 zod kernel: [ 0.802398] __crypto_alloc_tfm: 400
Aug 11 11:31:34 zod kernel: [ 0.802444] __crypto_alloc_tfm:
(null) ffffffff8103ba69 -2
Aug 11 11:31:34 zod kernel: [ 0.802494] alg: skcipher: Failed to load
transform for ecb-aes-aesni: -2
Aug 11 11:31:34 zod kernel: [ 0.802546] Pid: 36, comm: cryptomgr_test
Not tainted 2.6.40-4.fc15.x86_64 #14
Aug 11 11:31:34 zod kernel: [ 0.802548] Call Trace:
Aug 11 11:31:34 zod kernel: [ 0.802554] [<ffffffff81216042>]
alg_test_skcipher+0x48/0xa3
Aug 11 11:31:34 zod kernel: [ 0.802557] [<ffffffff812162f5>] ?
alg_find_test+0x3a/0x5d
Aug 11 11:31:34 zod kernel: [ 0.802560] [<ffffffff812164d8>]
alg_test+0x1c0/0x277
Aug 11 11:31:34 zod kernel: [ 0.802563] [<ffffffff814b5b33>] ?
schedule+0x690/0x6be
Aug 11 11:31:34 zod kernel: [ 0.802566] [<ffffffff81213fd2>] ?
cryptomgr_probe+0xca/0xca
Aug 11 11:31:34 zod kernel: [ 0.802569] [<ffffffff81213ffb>]
cryptomgr_test+0x29/0x44
Aug 11 11:31:34 zod kernel: [ 0.802571] [<ffffffff8106fd2b>]
kthread+0x84/0x8c
Aug 11 11:31:34 zod kernel: [ 0.802575] [<ffffffff814beb64>]
kernel_thread_helper+0x4/0x10
Aug 11 11:31:34 zod kernel: [ 0.802577] [<ffffffff8106fca7>] ?
kthread_worker_fn+0x148/0x148
Aug 11 11:31:34 zod kernel: [ 0.802580] [<ffffffff814beb60>] ?
gs_change+0x13/0x13
Aug 11 11:31:34 zod kernel: [ 0.802586] aesni_init: 1278 err: 0
On Mon, Aug 15, 2011 at 07:50:39AM -0400, Josh Boyer wrote:
>
> Well, yes, it does in this code:
>
> if (!((alg->cra_flags & CRYPTO_ALG_TYPE_MASK) ==
> CRYPTO_ALG_TYPE_BLKCIPHER ? alg->cra_blkcipher.ivsize :
> alg->cra_ablkcipher.ivsize)) {
>
> printk(KERN_ERR "%s: %d\n", __func__, __LINE__);
> return alg;
> }
>
> But the lookup seems to fail after it returns (see traces below).
OK something's clearly screwed up. I'll try to reproduce this
here.
Thanks!
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
On Tue, Aug 16, 2011 at 08:32:27PM +0800, Herbert Xu wrote:
> On Mon, Aug 15, 2011 at 07:50:39AM -0400, Josh Boyer wrote:
> >
> > Well, yes, it does in this code:
> >
> > if (!((alg->cra_flags & CRYPTO_ALG_TYPE_MASK) ==
> > CRYPTO_ALG_TYPE_BLKCIPHER ? alg->cra_blkcipher.ivsize :
> > alg->cra_ablkcipher.ivsize)) {
> >
> > printk(KERN_ERR "%s: %d\n", __func__, __LINE__);
> > return alg;
> > }
> >
> > But the lookup seems to fail after it returns (see traces below).
>
> OK something's clearly screwed up. I'll try to reproduce this
> here.
Any luck on this? If there's anything further I can try or if you need
some data from me, just let me know.
josh
On Thu, Aug 18, 2011 at 01:44:25PM -0400, Josh Boyer wrote:
>
> Any luck on this? If there's anything further I can try or if you need
> some data from me, just let me know.
Can you send me your .config file?
Thanks!
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
On Fri, Aug 19, 2011 at 08:02:29AM +0800, Herbert Xu wrote:
> On Thu, Aug 18, 2011 at 01:44:25PM -0400, Josh Boyer wrote:
> >
> > Any luck on this? If there's anything further I can try or if you need
> > some data from me, just let me know.
>
> Can you send me your .config file?
Sure. It's simply the default Fedora config for F15, F16, etc.
F15 is using "2.6.40" which is just a rebranded 3.0. Attached.
josh
On Thu, Aug 18, 2011 at 10:15:17PM -0400, Josh Boyer wrote:
> On Fri, Aug 19, 2011 at 08:02:29AM +0800, Herbert Xu wrote:
> > On Thu, Aug 18, 2011 at 01:44:25PM -0400, Josh Boyer wrote:
> > >
> > > Any luck on this? If there's anything further I can try or if you need
> > > some data from me, just let me know.
> >
> > Can you send me your .config file?
>
> Sure. It's simply the default Fedora config for F15, F16, etc.
>
> F15 is using "2.6.40" which is just a rebranded 3.0. Attached.
Thanks, this looks like a registration ordering problem between
aesni and cryptd.
Can you try building aesni as a module and see if that fixes the
issue?
Meanwhile I'll make a patch that ensures cryptd is registered
before aesni.
Thanks,
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
On Fri, Aug 19, 2011 at 03:32:24PM +0800, Herbert Xu wrote:
>
> Meanwhile I'll make a patch that ensures cryptd is registered
> before aesni.
Please try this patch. Thanks!
commit 9d4c522a113f6caa8b792ae829a25490fa87b1a2
Author: Herbert Xu <[email protected]>
Date: Fri Aug 19 16:11:23 2011 +0800
crypto: cryptd - Use subsys_initcall to prevent races with aesni
As cryptd is depeneded on by other algorithms such as aesni-intel,
it needs to be registered before them. When everything is built
as modules, this occurs naturally. However, for this to work when
they are built-in, we need to use subsys_initcall in cryptd.
Signed-off-by: Herbert Xu <[email protected]>
diff --git a/crypto/cryptd.c b/crypto/cryptd.c
index e46d21a..671d4d6 100644
--- a/crypto/cryptd.c
+++ b/crypto/cryptd.c
@@ -945,7 +945,7 @@ static void __exit cryptd_exit(void)
crypto_unregister_template(&cryptd_tmpl);
}
-module_init(cryptd_init);
+subsys_initcall(cryptd_init);
module_exit(cryptd_exit);
MODULE_LICENSE("GPL");
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
On Fri, Aug 19, 2011 at 04:12:08PM +0800, Herbert Xu wrote:
> On Fri, Aug 19, 2011 at 03:32:24PM +0800, Herbert Xu wrote:
> >
> > Meanwhile I'll make a patch that ensures cryptd is registered
> > before aesni.
>
> Please try this patch. Thanks!
This worked great. I'm confused why adding ivsize to the ablk_ecb_alg
"worked" as well, but looking at the traces again I agree this seems
like the actual problem.
Thanks!
> commit 9d4c522a113f6caa8b792ae829a25490fa87b1a2
> Author: Herbert Xu <[email protected]>
> Date: Fri Aug 19 16:11:23 2011 +0800
>
> crypto: cryptd - Use subsys_initcall to prevent races with aesni
>
> As cryptd is depeneded on by other algorithms such as aesni-intel,
> it needs to be registered before them. When everything is built
> as modules, this occurs naturally. However, for this to work when
> they are built-in, we need to use subsys_initcall in cryptd.
>
> Signed-off-by: Herbert Xu <[email protected]>
Tested-by: Josh Boyer <[email protected]>
> diff --git a/crypto/cryptd.c b/crypto/cryptd.c
> index e46d21a..671d4d6 100644
> --- a/crypto/cryptd.c
> +++ b/crypto/cryptd.c
> @@ -945,7 +945,7 @@ static void __exit cryptd_exit(void)
> crypto_unregister_template(&cryptd_tmpl);
> }
>
> -module_init(cryptd_init);
> +subsys_initcall(cryptd_init);
> module_exit(cryptd_exit);
>
> MODULE_LICENSE("GPL");
> --
> Email: Herbert Xu <[email protected]>
> Home Page: http://gondor.apana.org.au/~herbert/
> PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
On Fri, Aug 19, 2011 at 08:15:16AM -0400, Josh Boyer wrote:
> On Fri, Aug 19, 2011 at 04:12:08PM +0800, Herbert Xu wrote:
> > On Fri, Aug 19, 2011 at 03:32:24PM +0800, Herbert Xu wrote:
> > >
> > > Meanwhile I'll make a patch that ensures cryptd is registered
> > > before aesni.
> >
> > Please try this patch. Thanks!
>
> This worked great. I'm confused why adding ivsize to the ablk_ecb_alg
> "worked" as well, but looking at the traces again I agree this seems
> like the actual problem.
Thanks for testing! Patch applied and pushed.
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt