* Crash in crypto mcryptd
@ 2016-12-02 0:00 Mikulas Patocka
2016-12-02 1:47 ` Tim Chen
2016-12-03 0:16 ` Tim Chen
0 siblings, 2 replies; 5+ messages in thread
From: Mikulas Patocka @ 2016-12-02 0:00 UTC (permalink / raw)
To: Herbert Xu, David S. Miller, Tim Chen; +Cc: linux-crypto, dm-devel, Milan Broz
Hi
There is a bug in mcryptd initialization.
This is a test module that tries various hash algorithms. When you load
the module with "insmod test.ko 'alg=mcryptd(md5)'", the machine crashes.
Mikulas
#include <linux/module.h>
#include <linux/init.h>
#include <crypto/hash.h>
static char *alg = "md5";
module_param_named(alg, alg, charp, 0444);
MODULE_PARM_DESC(alg, "the algorith to test");
static bool sync = true;
module_param_named(sync, sync, bool, 0444);
MODULE_PARM_DESC(alg, "sync flag");
static int __init dump_init(void)
{
struct crypto_shash *h;
char key[4];
int r;
printk("testing algorithm '%s'\n", alg);
h = crypto_alloc_shash(alg, 0, sync ? CRYPTO_ALG_ASYNC : 0);
if (IS_ERR(h)) {
printk("error %d\n", (int)PTR_ERR(h));
return PTR_ERR(h);
}
printk("setting key\n");
r = crypto_shash_setkey(h, key, sizeof key);
if (r)
printk("setkey: %d\n", r);
crypto_free_shash(h);
printk("module loaded\n");
return 0;
}
static void __exit dump_exit(void)
{
printk("dump exit\n");
}
module_init(dump_init)
module_exit(dump_exit)
MODULE_LICENSE("GPL");
[898029.802035] BUG: unable to handle kernel NULL pointer dereference at
(null)
[898029.806060] IP: [<ffffffffa022014d>] md5_final+0xad/0x210 [md5]
[898029.808156] PGD 11a5d8067 [898029.809051] PUD 11a491067
PMD 0 [898029.810280]
[898029.810904] Oops: 0002 [#1] PREEMPT SMP
[898029.812239] Modules linked in: md5 testdump(O+) mcryptd uvesafb
cfbfillrect cfbimgblt cn cfbcopyarea fbcon bitblit fbcon_rotate fbcon_ccw
fbcon_ud fbcon_cw softcursor fb fbdev font ipv6 binfmt_misc mousedev
af_packet psmouse pcspkr virtio_net virtio_balloon button ext4 crc16 jbd2
mbcache dm_mod virtio_blk evdev virtio_pci virtio_ring virtio
[898029.817178] CPU: 9 PID: 187 Comm: kworker/9:1 Tainted: G O
4.9.0-rc7+ #6
[898029.818066] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[898029.818732] Workqueue: crypto mcryptd_queue_worker [mcryptd]
[898029.819394] task: ffff88011aa2bd80 task.stack: ffff880118480000
[898029.820077] RIP: 0010:[<ffffffffa022014d>] [<ffffffffa022014d>]
md5_final+0xad/0x210 [md5]
[898029.821050] RSP: 0018:ffff880118483d48 EFLAGS: 00010286
[898029.821661] RAX: 04b2008fd98c1dd4 RBX: ffff880119cd7f28 RCX:
00000000980980e9
[898029.822464] RDX: 7e42f8ec980980e9 RSI: 00000000ef1c4f74 RDI:
ffff880119cd7f30
[898029.823293] RBP: ffff880118483d68 R08: 000000001b99d513 R09:
0000000000000000
[898029.824117] R10: 0000000000000000 R11: 00000000b8b56373 R12:
ffff880119cd7f18
[898029.824944] R13: 0000000000000000 R14: ffff880119cd7f38 R15:
ffffffffa01ee43c
[898029.825776] FS: 0000000000000000(0000) GS:ffff88011fd20000(0000)
knlGS:0000000000000000
[898029.826712] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[898029.827376] CR2: 0000000000000000 CR3: 000000011a6c9000 CR4:
00000000000006a0
[898029.828204] Stack:
[898029.828452] ffff880119cd7f18 ffff88011fd3bb00 0000000000000000
ffff880119cd7e00
[898029.829351] ffff880118483da0 ffffffff8119f281 ffff880119cd7f18
ffff88011fd3bb00
[898029.830242] ffff88011fd3bae0 ffff880119cd7e00 ffffffffa01ee43c
ffff880119cd7ec8
[898029.831141] Call Trace:
[898029.831460] [<ffffffff8119f281>] ? crypto_shash_final+0x31/0xb0
[898029.832151] [<ffffffffa01ee43c>] ? mcryptd_queue_worker+0x1c/0x190
[mcryptd]
[898029.832980] [<ffffffff8119f743>] ? shash_ahash_finup+0x73/0x80
[898029.833672] [<ffffffff81016a5f>] ? __switch_to+0x27f/0x460
[898029.834305] [<ffffffffa01eeabf>] ? mcryptd_hash_digest+0x4f/0x80
[mcryptd]
[898029.835125] [<ffffffffa01ee467>] ? mcryptd_queue_worker+0x47/0x190
[mcryptd]
[898029.835963] [<ffffffff8105974f>] ? process_one_work+0x1bf/0x3f0
[898029.836681] [<ffffffff810599c2>] ? worker_thread+0x42/0x4c0
[898029.837362] [<ffffffff81059980>] ? process_one_work+0x3f0/0x3f0
[898029.838045] [<ffffffff81059980>] ? process_one_work+0x3f0/0x3f0
[898029.838739] [<ffffffff8105ea49>] ? kthread+0xb9/0xd0
[898029.839318] [<ffffffff8105e990>] ? kthread_park+0x70/0x70
[898029.839959] [<ffffffff8131b965>] ? ret_from_fork+0x25/0x30
[898029.840594] Code: 14 c5 00 00 00 00 48 c1 e8 1d 41 89 44 24 5c 41 89
54 24 58 e8 45 ea 0e e1 49 8b 44 24 10 49 8b 54 24 18 48 8d 7b 08 48 83 e7
f8 <49> 89 45 00 49 89 55 08 31 c0 49 c7 44 24 10 00 00 00 00 48 c7
[898029.843633] RIP [<ffffffffa022014d>] md5_final+0xad/0x210 [md5]
[898029.844354] RSP <ffff880118483d48>
[898029.844769] CR2: 0000000000000000
[898029.845166] ---[ end trace 2ecde0bf66717337 ]---
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: Crash in crypto mcryptd
2016-12-02 0:00 Crash in crypto mcryptd Mikulas Patocka
@ 2016-12-02 1:47 ` Tim Chen
2016-12-02 7:06 ` Eric Biggers
2016-12-03 0:16 ` Tim Chen
1 sibling, 1 reply; 5+ messages in thread
From: Tim Chen @ 2016-12-02 1:47 UTC (permalink / raw)
To: Mikulas Patocka, Herbert Xu, David S. Miller
Cc: linux-crypto, dm-devel, Milan Broz, Megha Dey
On Thu, 2016-12-01 at 19:00 -0500, Mikulas Patocka wrote:
> Hi
>
> There is a bug in mcryptd initialization.
>
> This is a test module that tries various hash algorithms. When you load
> the module with "insmod test.ko 'alg=mcryptd(md5)'", the machine crashes.
I don't think your test setup is right. The mcryptd supports only multi-buffer
algorithm. I don't think there is such an implementation for md5.
Please refer to arch/x86/crypto/sha1-mb
multi-buffer implementation of sha1 to see the proper
setup and usage with mcryptd. You can also run tcrypt test to
exercise this code.
Tim
>
> Mikulas
>
>
> #include <linux/module.h>
> #include <linux/init.h>
> #include <crypto/hash.h>
>
> static char *alg = "md5";
>
> module_param_named(alg, alg, charp, 0444);
> MODULE_PARM_DESC(alg, "the algorith to test");
>
> static bool sync = true;
>
> module_param_named(sync, sync, bool, 0444);
> MODULE_PARM_DESC(alg, "sync flag");
>
> static int __init dump_init(void)
> {
> struct crypto_shash *h;
> char key[4];
> int r;
> printk("testing algorithm '%s'\n", alg);
> h = crypto_alloc_shash(alg, 0, sync ? CRYPTO_ALG_ASYNC : 0);
> if (IS_ERR(h)) {
> printk("error %d\n", (int)PTR_ERR(h));
> return PTR_ERR(h);
> }
> printk("setting key\n");
> r = crypto_shash_setkey(h, key, sizeof key);
> if (r)
> printk("setkey: %d\n", r);
> crypto_free_shash(h);
> printk("module loaded\n");
> return 0;
> }
>
> static void __exit dump_exit(void)
> {
> printk("dump exit\n");
> }
>
> module_init(dump_init)
> module_exit(dump_exit)
> MODULE_LICENSE("GPL");
>
>
> [898029.802035] BUG: unable to handle kernel NULL pointer dereference at
> (null)
> [898029.806060] IP: [<ffffffffa022014d>] md5_final+0xad/0x210 [md5]
> [898029.808156] PGD 11a5d8067 [898029.809051] PUD 11a491067
> PMD 0 [898029.810280]
> [898029.810904] Oops: 0002 [#1] PREEMPT SMP
> [898029.812239] Modules linked in: md5 testdump(O+) mcryptd uvesafb
> cfbfillrect cfbimgblt cn cfbcopyarea fbcon bitblit fbcon_rotate fbcon_ccw
> fbcon_ud fbcon_cw softcursor fb fbdev font ipv6 binfmt_misc mousedev
> af_packet psmouse pcspkr virtio_net virtio_balloon button ext4 crc16 jbd2
> mbcache dm_mod virtio_blk evdev virtio_pci virtio_ring virtio
> [898029.817178] CPU: 9 PID: 187 Comm: kworker/9:1 Tainted: G O
> 4.9.0-rc7+ #6
> [898029.818066] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
> [898029.818732] Workqueue: crypto mcryptd_queue_worker [mcryptd]
> [898029.819394] task: ffff88011aa2bd80 task.stack: ffff880118480000
> [898029.820077] RIP: 0010:[<ffffffffa022014d>] [<ffffffffa022014d>]
> md5_final+0xad/0x210 [md5]
> [898029.821050] RSP: 0018:ffff880118483d48 EFLAGS: 00010286
> [898029.821661] RAX: 04b2008fd98c1dd4 RBX: ffff880119cd7f28 RCX:
> 00000000980980e9
> [898029.822464] RDX: 7e42f8ec980980e9 RSI: 00000000ef1c4f74 RDI:
> ffff880119cd7f30
> [898029.823293] RBP: ffff880118483d68 R08: 000000001b99d513 R09:
> 0000000000000000
> [898029.824117] R10: 0000000000000000 R11: 00000000b8b56373 R12:
> ffff880119cd7f18
> [898029.824944] R13: 0000000000000000 R14: ffff880119cd7f38 R15:
> ffffffffa01ee43c
> [898029.825776] FS: 0000000000000000(0000) GS:ffff88011fd20000(0000)
> knlGS:0000000000000000
> [898029.826712] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [898029.827376] CR2: 0000000000000000 CR3: 000000011a6c9000 CR4:
> 00000000000006a0
> [898029.828204] Stack:
> [898029.828452] ffff880119cd7f18 ffff88011fd3bb00 0000000000000000
> ffff880119cd7e00
> [898029.829351] ffff880118483da0 ffffffff8119f281 ffff880119cd7f18
> ffff88011fd3bb00
> [898029.830242] ffff88011fd3bae0 ffff880119cd7e00 ffffffffa01ee43c
> ffff880119cd7ec8
> [898029.831141] Call Trace:
> [898029.831460] [<ffffffff8119f281>] ? crypto_shash_final+0x31/0xb0
> [898029.832151] [<ffffffffa01ee43c>] ? mcryptd_queue_worker+0x1c/0x190
> [mcryptd]
> [898029.832980] [<ffffffff8119f743>] ? shash_ahash_finup+0x73/0x80
> [898029.833672] [<ffffffff81016a5f>] ? __switch_to+0x27f/0x460
> [898029.834305] [<ffffffffa01eeabf>] ? mcryptd_hash_digest+0x4f/0x80
> [mcryptd]
> [898029.835125] [<ffffffffa01ee467>] ? mcryptd_queue_worker+0x47/0x190
> [mcryptd]
> [898029.835963] [<ffffffff8105974f>] ? process_one_work+0x1bf/0x3f0
> [898029.836681] [<ffffffff810599c2>] ? worker_thread+0x42/0x4c0
> [898029.837362] [<ffffffff81059980>] ? process_one_work+0x3f0/0x3f0
> [898029.838045] [<ffffffff81059980>] ? process_one_work+0x3f0/0x3f0
> [898029.838739] [<ffffffff8105ea49>] ? kthread+0xb9/0xd0
> [898029.839318] [<ffffffff8105e990>] ? kthread_park+0x70/0x70
> [898029.839959] [<ffffffff8131b965>] ? ret_from_fork+0x25/0x30
> [898029.840594] Code: 14 c5 00 00 00 00 48 c1 e8 1d 41 89 44 24 5c 41 89
> 54 24 58 e8 45 ea 0e e1 49 8b 44 24 10 49 8b 54 24 18 48 8d 7b 08 48 83 e7
> f8 <49> 89 45 00 49 89 55 08 31 c0 49 c7 44 24 10 00 00 00 00 48 c7
> [898029.843633] RIP [<ffffffffa022014d>] md5_final+0xad/0x210 [md5]
> [898029.844354] RSP <ffff880118483d48>
> [898029.844769] CR2: 0000000000000000
> [898029.845166] ---[ end trace 2ecde0bf66717337 ]---
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: Crash in crypto mcryptd
2016-12-02 1:47 ` Tim Chen
@ 2016-12-02 7:06 ` Eric Biggers
2016-12-02 18:30 ` Tim Chen
0 siblings, 1 reply; 5+ messages in thread
From: Eric Biggers @ 2016-12-02 7:06 UTC (permalink / raw)
To: Tim Chen
Cc: Mikulas Patocka, Herbert Xu, David S. Miller, linux-crypto,
dm-devel, Milan Broz, Megha Dey
On Thu, Dec 01, 2016 at 05:47:02PM -0800, Tim Chen wrote:
> On Thu, 2016-12-01 at 19:00 -0500, Mikulas Patocka wrote:
> > Hi
> >
> > There is a bug in mcryptd initialization.
> >
> > This is a test module that tries various hash algorithms. When you load
> > the module with "insmod test.ko 'alg=mcryptd(md5)'", the machine crashes.
>
> I don't think your test setup is right. The mcryptd supports only multi-buffer
> algorithm. I don't think there is such an implementation for md5.
>
> Please refer to arch/x86/crypto/sha1-mb
> multi-buffer implementation of sha1 to see the proper
> setup and usage with mcryptd. You can also run tcrypt test to
> exercise this code.
>
> Tim
No, mcryptd must not crash the kernel if it's passed the wrong algorithm.
Users can try to instantiate it with any algorithm using AF_ALG, for example:
struct sockaddr_alg addr = {
.salg_type = "hash",
.salg_name = "mcryptd(md5)",
};
int fd = socket(AF_ALG, SOCK_SEQPACKET, 0);
bind(fd, (struct sockaddr *)&addr, sizeof(addr));
Currently, this instantly crashes the kernel.
Eric
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: Crash in crypto mcryptd
2016-12-02 7:06 ` Eric Biggers
@ 2016-12-02 18:30 ` Tim Chen
0 siblings, 0 replies; 5+ messages in thread
From: Tim Chen @ 2016-12-02 18:30 UTC (permalink / raw)
To: Eric Biggers
Cc: Mikulas Patocka, Herbert Xu, David S. Miller, linux-crypto,
dm-devel, Milan Broz, Megha Dey
On Thu, 2016-12-01 at 23:06 -0800, Eric Biggers wrote:
> On Thu, Dec 01, 2016 at 05:47:02PM -0800, Tim Chen wrote:
> >
> > On Thu, 2016-12-01 at 19:00 -0500, Mikulas Patocka wrote:
> > >
> > > Hi
> > >
> > > There is a bug in mcryptd initialization.
> > >
> > > This is a test module that tries various hash algorithms. When you load
> > > the module with "insmod test.ko 'alg=mcryptd(md5)'", the machine crashes.
> > I don't think your test setup is right. The mcryptd supports only multi-buffer
> > algorithm. I don't think there is such an implementation for md5.
> >
> > Please refer to arch/x86/crypto/sha1-mb
> > multi-buffer implementation of sha1 to see the proper
> > setup and usage with mcryptd. You can also run tcrypt test to
> > exercise this code.
> >
> > Tim
> No, mcryptd must not crash the kernel if it's passed the wrong algorithm.
That's true. We should detect the case when mcryptd is being passed invalid
argument and not crash. We'll take a look at it.
Tim
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: Crash in crypto mcryptd
2016-12-02 0:00 Crash in crypto mcryptd Mikulas Patocka
2016-12-02 1:47 ` Tim Chen
@ 2016-12-03 0:16 ` Tim Chen
1 sibling, 0 replies; 5+ messages in thread
From: Tim Chen @ 2016-12-03 0:16 UTC (permalink / raw)
To: Mikulas Patocka, Herbert Xu, David S. Miller
Cc: linux-crypto, dm-devel, Milan Broz
On Thu, 2016-12-01 at 19:00 -0500, Mikulas Patocka wrote:
> Hi
>
> There is a bug in mcryptd initialization.
>
> This is a test module that tries various hash algorithms. When you load
> the module with "insmod test.ko 'alg=mcryptd(md5)'", the machine crashes.
>
> Mikulas
>
>
Mikulas,
Can you try out the patch that I've sent out in a separate mail?
Thanks.
Tim
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2016-12-03 0:17 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-12-02 0:00 Crash in crypto mcryptd Mikulas Patocka
2016-12-02 1:47 ` Tim Chen
2016-12-02 7:06 ` Eric Biggers
2016-12-02 18:30 ` Tim Chen
2016-12-03 0:16 ` Tim Chen
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.