| View previous topic :: View next topic |
| Author |
Message |
freke
l33t
Joined: 23 Jan 2003
Posts: 976
Location: Somewhere in Denmark
|
 Posted: Thu Apr 12, 2018 8:48 pm Post subject: [SOLVED] Kernel panic with 4.16.1 - CCP |
 |
|
Hi,
Just tried out 4.16.1 coming from 4.15.15 with an make oldconfig
| Code: | [ 2.038474] mmc0: SDHCI controller on PCI [0000:00:14.7] using ADMA
[ 2.045218] ccp 0000:00:08.0: 5 command queues available
[ 2.051633] ccp 0000:00:08.0: ccp enabled
[ 2.055700] ccp 0000:00:08.0: enabled
[ 2.059497] general protection fault: 0000 [#1] SMP NOPTI
[ 2.063351] Modules linked in:
[ 2.066687] CPU: 1 PID: 1 Comm: swapper/0 Not tainted 4.16.1-gentoo #1
[ 2.070016] Hardware name: PC Engines APU2/APU2, BIOS 4.0.7 02/28/2017
[ 2.076705] RIP: 0010:0xffffffff81b63755
[ 2.083355] RSP: 0018:ffffc90000013e38 EFLAGS: 00010002
[ 2.090023] RAX: ffffffff81b6374e RBX: 7ab5cfc8fd2f3377 RCX: 00000000000002fe
[ 2.093345] RDX: 00000000000000ff RSI: ffff88011eca2080 RDI: ffffffff82066c74
[ 2.100024] RBP: 0000000000000246 R08: 0000000000022080 R09: ffffffff8158bc69
[ 2.110023] R10: ffffea00046850c0 R11: ffff88011a001c00 R12: 0000000000000000
[ 2.113348] R13: 0000000000000007 R14: 0000000000000030 R15: ffffffff81fa9d78
[ 2.120026] FS: 0000000000000000(0000) GS:ffff88011ec80000(0000) knlGS:0000000000000000
[ 2.133348] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 2.136682] CR2: 0000000000000000 CR3: 0000000001c0a000 CR4: 00000000000406e0
[ 2.146682] Call Trace:
[ 2.146682] ? sp_get_psp_master_device+0x54/0x70
[ 2.153347] ? psp_pci_init+0x1b/0xd0
[ 2.156712] ? sp_mod_init+0x42/0x68
[ 2.160038] ? firmware_map_add_early+0xf6/0xf6
[ 2.163375] ? do_one_initcall+0x49/0x1a0
[ 2.166681] ? kernel_init_freeable+0x187/0x294
[ 2.173423] ? rest_init+0xa0/0xa0
[ 2.176717] ? kernel_init+0x5/0xf0
[ 2.180023] ? ret_from_fork+0x22/0x40
[ 2.183348] Code: 6f 72 00 6d 65 6d 6d 61 70 00 64 72 69 76 65 72 73 2f 66 69 72 6d 77 61 72 65 2f 6d 65 6d 6d 61 70 2e 63 00 7f 2e 31 2e 30 00 6d <69> 73 73 69 6e 67 20 64 72 69 76 65 72 20 64 61 74 61 0a 00 44
[ 2.203372] RIP: 0xffffffff81b63755 RSP: ffffc90000013e38
[ 2.206688] ---[ end trace 75c97d460f806dd3 ]---
[ 2.214239] Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b
[ 2.214239]
[ 2.220055] Kernel Offset: disabled
[ 2.220055] ---[ end Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b
[ 2.220055] |
Anyone knows what could cause this?
4.15.15 config - http://bpaste.net/show/e13b9c7f8287
4.16.1 config - http://bpaste.net/show/72b6ca2e1a85
Tia 
Last edited by freke on Fri Apr 13, 2018 10:43 pm; edited 2 times in total |
|
| Back to top |
|
 |
khayyam
Watchman
Joined: 07 Jun 2012
Posts: 6227
Location: Room 101
|
| Posted: Fri Apr 13, 2018 12:56 am Post subject: Re: Kernel panic with 4.16.1 |
|
|
| freke wrote: | | Code: | [ 2.146682] Call Trace:
[ 2.146682] ? sp_get_psp_master_device+0x54/0x70
[ 2.153347] ? psp_pci_init+0x1b/0xd0
[ 2.156712] ? sp_mod_init+0x42/0x68
[ 2.160038] ? firmware_map_add_early+0xf6/0xf6
[ 2.163375] ? do_one_initcall+0x49/0x1a0
[ 2.166681] ? kernel_init_freeable+0x187/0x294
[ 2.173423] ? rest_init+0xa0/0xa0
[ 2.176717] ? kernel_init+0x5/0xf0
[ 2.180023] ? ret_from_fork+0x22/0x40 |
|
freke ... it looks to be CRYPTO_DEV_SP_PSP, Platform Security Processor (PSP) device ... which a diff of the two configs shows is =y on 14.16.1 but absent from 14.15.15. I know nothing of this particular processor/feature other than what's provided in the above link, however, as you didn't have it previously I don't expect it to be critical.
HTH & best ... khay |
|
| Back to top |
|
|
freke
l33t
Joined: 23 Jan 2003
Posts: 976
Location: Somewhere in Denmark
|
| Posted: Fri Apr 13, 2018 3:47 pm Post subject: |
|
|
Thanks for taking the time; turned out I needed to disable | Code: | | CONFIG_CRYPTO_DEV_CCP_DD |
too, to not get a kernel-panic.
Working 4.15.15 with CONFIG_CRYPTO_DEV_CCP_DD enabled (lshw): | Code: | *-generic:0
description: Encryption controller
product: Advanced Micro Devices, Inc. [AMD]
vendor: Advanced Micro Devices, Inc. [AMD]
physical id: 8
bus info: pci@0000:00:08.0
version: 00
width: 64 bits
clock: 33MHz
capabilities: msix ht pm bus_master cap_list
configuration: driver=ccp latency=0
resources: irq:0 memory:feb00000-feb1ffff memory:fe900000-fe9fffff memory:feb24000-feb24fff memory:fea00000-feafffff memory:feb20000-feb21fff |
With 4.16.1 and disabled CONFIG_CRYPTO_DEV_CCP_DD to not get a kernel-panic: | Code: | *-generic:0 UNCLAIMED
description: Encryption controller
product: Advanced Micro Devices, Inc. [AMD]
vendor: Advanced Micro Devices, Inc. [AMD]
physical id: 8
bus info: pci@0000:00:08.0
version: 00
width: 64 bits
clock: 33MHz
capabilities: msix ht pm cap_list
configuration: latency=0
resources: memory:feb00000-feb1ffff memory:fe900000-fe9fffff memory:feb24000-feb24fff memory:fea00000-feafffff memory:feb20000-feb21fff |
|
|
| Back to top |
|
|
khayyam
Watchman
Joined: 07 Jun 2012
Posts: 6227
Location: Room 101
|
| Posted: Fri Apr 13, 2018 4:16 pm Post subject: |
|
|
freke ...
makes sense as CRYPTO_DEV_CCP_DD is also a "secure processor device driver". It's obviously a bug, you shouldn't need to disable it, as the hardware is there. I can't see anyone having reported it, so I'd suggest you do if you want to get a fix anytime soon.
best ... khay |
|
| Back to top |
|
|
freke
l33t
Joined: 23 Jan 2003
Posts: 976
Location: Somewhere in Denmark
|
| Posted: Fri Apr 13, 2018 8:03 pm Post subject: |
|
|
Thanks - might need some more looking into - just tried the 4.16.2 kernel as I saw it in portage today.
make oldconfig with my 4.15.15 config; disabling any new options - works/boots fine... ie with CRYPTO_DEV_CCP_DD enabled which caused my 4.16.1-kernel to panic...
So either fixed or actually caused in conjunction with another option I might have enabled on my 4.16.11 try....
And seeing that no devices is unclaimed in lshw I guess I've no need to play with CRYPTO_DEV_SP_PSP |
|
| Back to top |
|
|
khayyam
Watchman
Joined: 07 Jun 2012
Posts: 6227
Location: Room 101
|
| Posted: Fri Apr 13, 2018 10:29 pm Post subject: |
|
|
freke ...
ok, good. It may just be the build system, given that the option was just added, or your 'oldconfig' caused some option(s) to be set/unset when they shouldn't be. I wouldn't worry about it ... now it works.
btw, you should probably mark the thread '[SOLVED]' (by editing the first post and changing the 'subject').
best ... khay |
|
| Back to top |
|
|
|
Kernel & Hardware
