Re: [PATCH] hwmon: (gpio-fan) Allow usage of gpio operations that may sleep

[Guenter Roeck]

On Thu, Dec 04, 2014 at 10:58:47AM -0600, Nishanth Menon wrote:
> Certain I2C based GPIO expanders could be used in sleepable context,
> this results in:
> [  115.890569] ------------[ cut here ]------------
> [  115.895422] WARNING: CPU: 0 PID: 1115 at drivers/gpio/gpiolib.c:1370 gpiod_set_raw_value+0x40/0x4c()
> [  115.905024] Modules linked in:
> [  115.908229] CPU: 0 PID: 1115 Comm: sh Tainted: G        W      3.18.0-rc7-next-20141203-dirty #1
> [  115.917461] Hardware name: Generic DRA74X (Flattened Device Tree)
> [  115.923876] [<c0015368>] (unwind_backtrace) from [<c00119f4>] (show_stack+0x10/0x14)
> [  115.932013] [<c00119f4>] (show_stack) from [<c05b78e8>] (dump_stack+0x78/0x94)
> [  115.939594] [<c05b78e8>] (dump_stack) from [<c003de28>] (warn_slowpath_common+0x7c/0xb4)
> [  115.948094] [<c003de28>] (warn_slowpath_common) from [<c003de7c>] (warn_slowpath_null+0x1c/0x24)
> [  115.957315] [<c003de7c>] (warn_slowpath_null) from [<c03461e8>] (gpiod_set_raw_value+0x40/0x4c)
> [  115.966457] [<c03461e8>] (gpiod_set_raw_value) from [<c04866f4>] (set_fan_speed+0x4c/0x64)
> [  115.975145] [<c04866f4>] (set_fan_speed) from [<c04868a8>] (set_rpm+0x98/0xac)
> [  115.982742] [<c04868a8>] (set_rpm) from [<c039fb4c>] (dev_attr_store+0x18/0x24)
> [  115.990426] [<c039fb4c>] (dev_attr_store) from [<c01b0a28>] (sysfs_kf_write+0x4c/0x50)
> [  115.998742] [<c01b0a28>] (sysfs_kf_write) from [<c01afe1c>] (kernfs_fop_write+0xbc/0x19c)
> [  116.007333] [<c01afe1c>] (kernfs_fop_write) from [<c0148cc4>] (vfs_write+0xb0/0x1a0)
> [  116.015461] [<c0148cc4>] (vfs_write) from [<c0148fbc>] (SyS_write+0x44/0x84)
> [  116.022881] [<c0148fbc>] (SyS_write) from [<c000e5c0>] (ret_fast_syscall+0x0/0x48)
> [  116.030833] ---[ end trace 3a0b636123acab82 ]---
> 
> So, switch over to sleepable GPIO operations as there is no mandatory
> need for non-sleepable gpio operations in the fan driver.
> 
> This allows the fan driver to be used with i2c based gpio expanders such
> as palmas_gpio.
> 
> Signed-off-by: Nishanth Menon <nm@xxxxxx>

Applied to hwmon-next. Do we need this in older kernels ?

Thanks,
Guenter
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/