Re: [PATCH 00/11] mtd: change printks to pr_*

[Artem Bityutskiy <dedekind1@gmail.com>]

On Tue, 2011-07-19 at 10:06 -0700, Brian Norris wrote:
> We've had quite some discussion about whether to use printk, pr_*, or
> dev_* functions for printing info, error, and debugging messages.
> Unfortunately, MTD doesn't quite have a nice device model, so we can't
> always use dev_* to print nice device information. This patch series
> mostly transitions the NAND subsystem to use pr_* functions, which are
> a little more brief and slightly more flexible than printk.
> 
> Along the way, we've dumped a lot of old DEBUG() code that was written
> specifically for MTD in favor of dynamic debugging. I tried killing the
> corresponding Kconfig options as well. Not sure if patches 9 and 10 are
> necessary.
> 
> These patches apply on top of l2-mtd-2.6.git
> 
> Brian
> 
> Brian Norris (11):
>   mtd: nand: change KERN_DEBUG to KERN_INFO
>   mtd: nand: convert printk() to pr_*()
>   mtd: nand: style fixups in pr_* messages
>   mtd: replace DEBUG() with pr_debug()
>   mtd: cleanup style on pr_debug messages
>   mtd: pcmciamtd: remove custom DEBUG() function
>   mtd: remove CONFIG_MTD_DEBUG
>   mtd: cleanup last uses of MTD_DEBUG config macros
>   arch: remove references to CONFIG_MTD_DEBUG and
>     CONFIG_MTD_DEBUG_VERBOSE
>   arch: remove references to CONFIG_MTD_PARTITIONS
>   mtd: Kbuild: remove reference to MTD_PARTITIONS

Pushed the series to the l2-mtd-2.6.git tree, except of the 2 "arch:
blah" patches. I think for defconfigs the practice is that the owners
update them themselves. People who use the defconfigs will not have
problems anyway because a non-existing Kconfig options like
CONFIG_MTD_PARTITIONS will be ignored.

Besides, there was a big hassle around defconfigs recently. In short -
Linus did not like that ARM defconfigs are constanly changing and add a
lot of noise.

Thanks!



-- 
Best Regards,
Artem Bityutskiy