Bug 197363 - invalid opcode: 0000 [#1] SMP [aesni_intel]
Summary: invalid opcode: 0000 [#1] SMP [aesni_intel]
Status: NEW
Alias: None
Product: Other
Classification: Unclassified
Component: Modules (show other bugs)
Hardware: x86-64 Linux
: P1 normal
Assignee: other_modules
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2017-10-23 15:49 UTC by SviMik
Modified: 2017-10-23 15:49 UTC (History)
0 users

See Also:
Kernel Version: 4.8.13-1.el6.elrepo.x86_64
Subsystem:
Regression: No
Bisected commit-id:


Attachments
oops (24.44 KB, text/plain)
2017-10-23 15:49 UTC, SviMik
Details

Description SviMik 2017-10-23 15:49:39 UTC
Created attachment 260345 [details]
oops

Hi!

1) Got the following kernel panic:

invalid opcode: 0000 [#1] SMP
CPU: 0 PID: 1449 Comm: openvpn Not tainted 4.8.13-1.el6.elrepo.x86_64 #1
----cut----
Call Trace:
<IRQ>
[<ffffffff810be15e>] ? enqueue_entity+0x45e/0x6f0
[<ffffffffa0104f25>] ? aesni_gcm_enc_avx+0x95/0xd0 [aesni_intel]
[<ffffffffa0105417>] helper_rfc4106_encrypt+0x167/0x2f0 [aesni_intel]
[<ffffffffa0104bfb>] rfc4106_encrypt+0x5b/0x90 [aesni_intel]
----cut----
(See full dump in attachment!)


2) `modinfo aesni_intel` says:

filename:
/lib/modules/4.8.13-1.el6.elrepo.x86_64/kernel/arch/x86/crypto/aesni-intel.ko
alias:          crypto-aes
alias:          aes
license:        GPL
description:    Rijndael (AES) Cipher Algorithm, Intel AES-NI
instructions optimized
alias:          crypto-fpu
alias:          fpu
srcversion:     E063372EF971D4EA858A602
alias:          cpu:type:x86,ven*fam*mod*:feature:*0099*
depends:        glue_helper,aes-x86_64,lrw,cryptd,ablk_helper
intree:         Y
vermagic:       4.8.13-1.el6.elrepo.x86_64 SMP mod_unload modversions


3) `cat /proc/cpuinfo` says it does have AES-NI:

model name      : Intel(R) Xeon(R) CPU E5-2620 0 @ 2.00GHz
flags           : fpu de tsc msr pae cx8 apic sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall nx lm constant_tsc rep_good nopl nonstop_tsc eagerfpu pni pclmulqdq ssse3 cx16 pcid sse4_1 sse4_2 popcnt tsc_deadline_timer aes avx hypervisor lahf_lm


4) Have tested with `openssl speed -elapsed -evp aes-256-gcm`

Results: AES-NI works fine when doing manual tests, could not trigger this bug intentionally.


5) A full dump of three oops occasions can be found in attachment. Dumps were received using netconsole.

Note You need to log in before you can comment on or make changes to this bug.