Bug 213961 - Oops while loading radeon driver
Summary: Oops while loading radeon driver
Status: NEW
Alias: None
Product: Drivers
Classification: Unclassified
Component: Video(Other) (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: platform_ppc-32
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2021-08-03 16:45 UTC by Elimar Riesebieter
Modified: 2021-08-05 11:44 UTC (History)
1 user (show)

See Also:
Kernel Version: 5.14-rc4
Subsystem:
Regression: Yes
Bisected commit-id:


Attachments
Boot proto (3.26 KB, text/plain)
2021-08-03 16:45 UTC, Elimar Riesebieter
Details
dmesg dump (36.74 KB, text/plain)
2021-08-04 13:29 UTC, Elimar Riesebieter
Details
.config (94.97 KB, text/plain)
2021-08-04 13:30 UTC, Elimar Riesebieter
Details
config with KALLSYMS enabled (95.01 KB, text/plain)
2021-08-04 19:54 UTC, Elimar Riesebieter
Details
dmesg dump with KALLSYMS enabled (37.61 KB, text/plain)
2021-08-04 19:55 UTC, Elimar Riesebieter
Details

Description Elimar Riesebieter 2021-08-03 16:45:00 UTC
Created attachment 298183 [details]
Boot proto

Please find attached boot proto. The virtual console freezes. X doesn' start. ssh login from remote gives access to the system.

Thanks in advance
Comment 1 Christophe Leroy 2021-08-04 05:44:04 UTC
There is not much that can be done with the log provided.

Could you please rebuild your kernel with CONFIG_DEBUG_INFO so that we get the name of involved functions in the BUG report.

Could you also provide your .config, and a full dump of 'dmesg'

Thanks
Comment 2 Elimar Riesebieter 2021-08-04 13:29:52 UTC
Created attachment 298195 [details]
dmesg dump
Comment 3 Elimar Riesebieter 2021-08-04 13:30:20 UTC
Created attachment 298197 [details]
.config
Comment 4 Christophe Leroy 2021-08-04 14:26:04 UTC
Thanks, unfortunately the info is still not usable because we don't have function names in the dump:

[   15.810083] Call Trace:
[   15.810090] [f2b6dc30] [c1022318] 0xc1022318 (unreliable)
[   15.810102] [f2b6dc50] [be987860] 0xbe987860
[   15.810112] [f2b6dc80] [be85074c] 0xbe85074c
[   15.810121] [f2b6dcb0] [be984484] 0xbe984484
[   15.810130] [f2b6dcd0] [c037fd8c] 0xc037fd8c
[   15.810140] [f2b6dd00] [c03e1460] 0xc03e1460
[   15.810149] [f2b6dd20] [c03e17f0] 0xc03e17f0
[   15.810158] [f2b6dd40] [c03e18f4] 0xc03e18f4
[   15.810168] [f2b6dd70] [c03e21f4] 0xc03e21f4
[   15.810177] [f2b6dd90] [c03deaf4] 0xc03deaf4
[   15.810187] [f2b6ddc0] [c03e04d8] 0xc03e04d8
[   15.810196] [f2b6ddf0] [c03e2d6c] 0xc03e2d6c
[   15.810205] [f2b6de10] [c0006c34] 0xc0006c34
[   15.810215] [f2b6de80] [c00cb0d4] 0xc00cb0d4
[   15.810224] [f2b6deb0] [c00cc2d4] 0xc00cc2d4
[   15.810233] [f2b6df30] [c0017098] 0xc0017098


Looking at the .config, it seems like CONFIG_KALLSYMS is missing.

Can you retry with CONFIG_KALLSYMS ?

Thanks
Comment 5 Elimar Riesebieter 2021-08-04 19:53:25 UTC
Configuring KALLSYMS doesn't give the function names
Comment 6 Elimar Riesebieter 2021-08-04 19:54:37 UTC
Created attachment 298203 [details]
config with KALLSYMS enabled
Comment 7 Elimar Riesebieter 2021-08-04 19:55:16 UTC
Created attachment 298205 [details]
dmesg dump with KALLSYMS enabled
Comment 8 Christophe Leroy 2021-08-05 08:04:08 UTC
Great:

[   15.246367] NIP [bea42a80] radeon_agp_head_init+0x1c/0xf8 [radeon]
[   15.246969] LR [bea39860] radeon_driver_load_kms+0x1bc/0x1f4 [radeon]
[   15.247160] Call Trace:
[   15.247168] [f2b75c30] [c0c1ec60] 0xc0c1ec60 (unreliable)
[   15.247180] [f2b75c50] [bea39860] radeon_driver_load_kms+0x1bc/0x1f4 [radeon]
[   15.247343] [f2b75c80] [be8cc74c] drm_dev_register+0x10c/0x268 [drm]
[   15.247718] [f2b75cb0] [bea36484] radeon_pci_probe+0x108/0x190 [radeon]
[   15.248001] [f2b75cd0] [c03832fc] pci_device_probe+0xf4/0x1a4

So we now know we have a NULL pointer dereference in radeon_agp_head_init().

Looks like all this code is quite recent, at least there are recent modification, so I think you should address it with RADEON people, I'm not sure the problem is a PPC32 subject.
Comment 9 Elimar Riesebieter 2021-08-05 10:52:52 UTC
Well, 5.13.8 just runs fine, though.
Comment 10 Christophe Leroy 2021-08-05 11:00:05 UTC
(In reply to Elimar Riesebieter from comment #9)
> Well, 5.13.8 just runs fine, though.

Yes, recent changes to Radeon appear for the first time in v5.14-rc1

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