Hi Andrew,
Thanks a lot.
I will check the the scripts/checkpatch.pl after the codes are modified.
About the subject, that's my carelessness and I will correct them by your recommendation next time.
In addition, should I resend the patches after they are modified? Thanks.
BRs,
Joseph Chan
-----Original Message-----
From: Andrew Morton [mailto:akpm@linux-foundation.org]
Sent: Thursday, May 08, 2008 5:04 AM
To: Krzysztof Helt
Cc: Joseph Chan; geert@linux-m68k.org; <redacted>; <redacted>
Subject: Re: [Linux-fbdev-devel] [PATCH 1/9] viafb: VIA Frame Buffer Device Driver - Resend
On Wed, 7 May 2008 18:43:33 +0200
Krzysztof Helt [off-list ref] wrote:
A general comment is that you use to many global variables.
yes, that was my first impression. The _names_ of those variables is a real problem. lcd_mode, DeviceStatus, second_size, primary_dev, etc. The driver is attempting to reserve all those names for itself, kernel-wide.
So at a minimum, going through and prepending viafb_ onto all those would be needed.
There are a number of layout glitches in there. Such as a strange combination of two styles:
void foo(void)
{
versus
void foo(void ) {
Many of these things can be pinpointed by processing the diff with scripts/checkpatch.pl.
Administrative note: all nine patches had the same title "viafb: VIA Frame Buffer Device Driver". Please choose unique and meaningful names for each patch, such as "viafb: add utility header file".
Apart from those fairly easily-fixed things, the driver looks pretty good from a quick scan, given how large it is, and given that it probably has been around for quite a long time..