Skip to contentSkip to conversation

Pinctrl crash on latest linux-next/master

DISCUSSION Viresh Kumar 4 participants 5 messages Apr 16, 2012 – Apr 17, 2012

lore
Share & tools
lore
#0 lore raw

Hi Linus,

I don't know if the issue is in my patches or something else, but i was able to boot it earlier. Till my V3 for pinctrl. And nothing much changed since then, i just rebased on latest kernel.

Now, kernel crashes like:

....

pinctrl core: initialized pinctrl subsystem Registered protocol family 16
driver_bound: device b4000000.pinmux already bound Unhandled fault: alignment exception at 0xffffffe

Internal error: ARM Modules linked in

...
...

next_device+0x10/0x1c
bus_for_each_dev+0x68/0x84

driver_attach
...
...
spear310_pinctrl_init+0x14/0x1c

[Sorry if crash log doesn't look nice.
I have written this crash log manually, as i didn't got this on console. I copied it from log_buf global variable.]

Signature · 1 line
viresh
#1 lore raw
Quoted 3 lines · Viresh Kumar

I don't know if the issue is in my patches or something else, but i was able to boot it earlier. Till my V3 for pinctrl. And nothing much changed since then, i just rebased on latest kernel.

Hm I haven't tested the pinctrl tree on v3.4-rc3 yet.

Did you git bisect to figure out which patch broke it?

Yours,
Linus Walleij

#2 lore raw
Quoted 9 lines · Linus Walleij
Quoted 3 lines · Viresh Kumar

I don't know if the issue is in my patches or something else, but i was able to boot it earlier. Till my V3 for pinctrl. And nothing much changed since then, i just rebased on latest kernel.

Hm I haven't tested the pinctrl tree on v3.4-rc3 yet.

Did you git bisect to figure out which patch broke it?

No.
Don't have a lot of SPEAr patches upstream till now. And have to apply all of them, on all bisections. So skipped it :)

--
viresh

#3 lore raw
In reply to Linus Walleij
Quoted 8 lines · Linus Walleij
Quoted 3 lines · Viresh Kumar

I don't know if the issue is in my patches or something else, but i was able to boot it earlier. Till my V3 for pinctrl. And nothing much changed since then, i just rebased on latest kernel.

Hm I haven't tested the pinctrl tree on v3.4-rc3 yet.

Your for-next tree is ok since i'm working on it for pinctrl-imx driver. :-) It seems it's based on v3.4-rc2.

Hi Viresh,
Some info for your reference:

You may want to check if you call platform_set_drvdata after pinctrl_register like:
ipmx->pctl = pinctrl_register(&imx_pmx_desc, &pdev->dev, ipmx);
...
platform_set_drvdata(pdev, ipmx);

If that, you may need to change to:
platform_set_drvdata(pdev, ipmx);
..
ipmx->pctl = pinctrl_register(&imx_pmx_desc, &pdev->dev, ipmx);

Because during the registration, the get_group_count will be called which may reply on platform_set_drvdata(pdev, ipmx);

Regards
Dong Aisheng

Quoted 10 lines

Did you git bisect to figure out which patch broke it?

Yours,
Linus Walleij

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

#4 lore raw
Quoted 8 lines · Dong Aisheng
Quoted 6 lines
Quoted 6 lines
Quoted 3 lines
Quoted 3 lines

I don't know if the issue is in my patches or something else, but i was able to boot it earlier. Till my V3 for pinctrl. And nothing much changed since then, i just rebased on latest kernel.

Hm I haven't tested the pinctrl tree on v3.4-rc3 yet.

Your for-next tree is ok since i'm working on it for pinctrl-imx driver. :-) It seems it's based on v3.4-rc2.

At last, found it. It was my code :(
I was writing on more memory than was allocated to me.

Thanks for your inputs.

Signature · 1 line
viresh