Bug 216420 - Unable to handle kernel NULL pointer dereference at virtual address 00000008
Summary: Unable to handle kernel NULL pointer dereference at virtual address 00000008
Status: NEW
Alias: None
Product: Drivers
Classification: Unclassified
Component: MMC/SD (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: drivers_mmc-sd
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2022-08-27 09:36 UTC by frc.gabriel
Modified: 2022-09-06 20:53 UTC (History)
1 user (show)

See Also:
Kernel Version: 6.0.0-rc2
Tree: Mainline
Regression: No


Attachments
full dmesg output (19.19 KB, text/plain)
2022-08-27 09:36 UTC, frc.gabriel
Details
good dmesg 5.19 (23.77 KB, text/plain)
2022-08-28 11:15 UTC, frc.gabriel
Details
6.0.0-rc3-next-20220901-dmesg.txt (31.15 KB, text/plain)
2022-09-03 22:13 UTC, frc.gabriel
Details
good-6.0-rc2-plus-Zhen-Lei-patch-plus-Isaac-J-Manjarres-patch.txt (24.33 KB, text/plain)
2022-09-04 11:32 UTC, frc.gabriel
Details
6.0.0-rc4-dmesg.txt (23.88 KB, text/plain)
2022-09-06 20:53 UTC, frc.gabriel
Details

Description frc.gabriel 2022-08-27 09:36:06 UTC
Created attachment 301682 [details]
full dmesg output

Hi, good morning!

my device is currently crashing during boot with kernel 6.0.0-rc2. I also tried next-20220826 and the outcome is the same.

The full dmesg log can be found attached on crash-boot.txt file. 

Thank you and have a good day,
Gabriel Francisco
Comment 1 The Linux kernel's regression tracker (Thorsten Leemhuis) 2022-08-28 10:40:51 UTC
(In reply to frc.gabriel from comment #0)

> my device is currently crashing during boot with kernel 6.0.0-rc2. I also
> tried next-20220826 and the outcome is the same.

Just to understand this properly: I assume 5.19.y worked fine I suppose (and this is thus a regression)?
Comment 2 frc.gabriel 2022-08-28 11:15:41 UTC
Created attachment 301689 [details]
good dmesg 5.19
Comment 3 frc.gabriel 2022-08-28 11:16:12 UTC
Hi,

Yes, your assumption is correct! I quickly checked and it works ok with v5.19. 

The dmesg log for v5.19 can be found at "good-dmesg-5.19.txt" file.

Thank you!
Gabriel Francisco
Comment 4 frc.gabriel 2022-09-03 22:13:13 UTC
Created attachment 301738 [details]
6.0.0-rc3-next-20220901-dmesg.txt

Hi,

with 6.0.0-rc3-next-20220901 it reaches the initramfs prompt and ls /dev shows no mmcblk device. 

Sharing in case it helps,
Thank you
Comment 5 The Linux kernel's regression tracker (Thorsten Leemhuis) 2022-09-04 09:51:36 UTC
You might want to test this patch and add your "Tested-by" to speed things up:
https://lore.kernel.org/lkml/20220818172852.3548-1-isaacmanjarres@google.com/

(that's the one mentioned in https://lore.kernel.org/lkml/CAGETcx9YvpYFf1mxHi1iA--sE5G2MZLS1x8vHnTu6Wj30ScaOA@mail.gmail.com/ which you iirc are CCed on)
Comment 6 frc.gabriel 2022-09-04 11:32:26 UTC
Created attachment 301739 [details]
good-6.0-rc2-plus-Zhen-Lei-patch-plus-Isaac-J-Manjarres-patch.txt

Done, I added the Tested-by in that patch (but it only works when combining with Zhen-Lei's patch)

I'm sharing the dmesg in case might be helpful at "good-6.0-rc2-plus-Zhen-Lei-patch-plus-Isaac-J-Manjarres-patch.txt" file.

My device boots successfully and I compared (one of) the addresses from

[   21.017179] amba fc33c000.etm: deferred probe pending
[   21.017327] amba fc33d000.etm: deferred probe pending
[   21.021356] amba fc33e000.etm: deferred probe pending
[   21.026352] amba fc33f000.etm: deferred probe pending
[   21.031414] amba fc345000.funnel: deferred probe pending
[   21.036419] amba fc307000.etf: deferred probe pending
[   21.041829] amba fc318000.tpiu: deferred probe pending
[   21.046748] amba fc31a000.funnel: deferred probe pending
[   21.051819] amba fc31b000.funnel: deferred probe pending
[   21.057256] amba fc31c000.replicator: deferred probe pending
[   21.062585] amba fc322000.etr: deferred probe pending

with the "qcom-msm8974.dtsi" file and looks like I have a amba device (never heard about it) :D

Thank you!
Gabriel Francisco
Comment 7 frc.gabriel 2022-09-06 20:53:41 UTC
Created attachment 301756 [details]
6.0.0-rc4-dmesg.txt

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