RE: [PATCH] ep93xx_eth.c: general cleanup

From: H Hartley Sweeten
Date: Wed Dec 16 2009 - 12:44:18 EST


On Wednesday, December 16, 2009 10:32 AM, Joe Perches wrote:

Added netdev to the Cc: list, and removed Nick M. (unintended)

> On Wed, 2009-12-16 at 12:09 -0500, H Hartley Sweeten wrote:
>> +#define pr_fmt(fmt) KBUILD_MODNAME ": %s: " fmt, __func__
>
> I think KBUILD_MODNAME ":%s: " fmt, __func__
> or its equivalent direct use is more common.
> It's possibly useful to standardize.

I wasn't sure about that and could not find anything "standard" in mainline.
The biggest user of pr_fmt that I could find is in drivers/s390 and that
uses a local KMSG_COMPONENT define instead of KBUILD_MODNAME.

>> -#define DRV_MODULE_NAME "ep93xx-eth"
>> -#define DRV_MODULE_VERSION "0.1"
>> +#include <mach/hardware.h>
>
> I think these sorts of defines are common and useful.

I can add those back if needed. Just seemed a waste to have the two defines
and only use them in one place.

>> static int __init ep93xx_eth_init_module(void)
>> {
>> - printk(KERN_INFO DRV_MODULE_NAME " version " DRV_MODULE_VERSION " loading\n");
>
> Why remove this printk completely?

In my dmesg is just looks like noise.

ep93xx-eth version 0.1 loading
eth0: ep93xx on-chip ethernet, IRQ 39, 00:21:f5:00:00:17

If the drive loads correctly you will get the second line. If it error's for
some reason you will get one of the dev_err messages.

But, if the defines go back in I can also add this message back.

Regards,
Hartley


N‹§²æìr¸›yúèšØb²X¬¶ÇvØ^–)Þ{.nÇ+‰·¥Š{±‘êçzX§¶›¡Ü}©ž²ÆzÚ&j:+v‰¨¾«‘êçzZ+€Ê+zf£¢·hšˆ§~†­†Ûiÿûàz¹®w¥¢¸?™¨è­Ú&¢)ßf”ù^jÇy§m…á@A«a¶Úÿ 0¶ìh®å’i