At Thu, 10 May 2007 14:00:13 +0100,
Liam Girdwood wrote:
>
> Sorry for the lateness in the current merge window of this patch
> series.
Well, I guess the patches will be unlikely merge to 2.6.22 at this
moment... Anyway, I'll try to merge into ALSA tree.
> From: Graeme Gregory <gg@xxxxxxxxxxxxxxxxxxxxxxxxxxx>
>
> This patch adds AC97 support to the Samsung S3C2443 CPU.
>
> Signed-off-by: Graeme Gregory <gg@xxxxxxxxxxxxxxxxxxxxxxxxxxx>
> Signed-off-by: Liam Girdwood <lg@xxxxxxxxxxxxxxxxxxxxxxxxxxx>
> [2 s3c2443-ac97.patch <text/x-patch; UTF-8 (7bit)>]
> --- /dev/null Thu Jan 01 00:00:00 1970 +0000
> +++ b/soc/s3c24xx/s3c2443-ac97.c Wed May 09 13:54:02 2007 +0100
> @@ -0,0 +1,405 @@
(snip)
> +static struct completion ac97_completion;
Use DECLARE_COMPLETION() rather than calling init_completion() at each
time. Since this is accessed in the irq handler, it must be
initialized properly before anything happens.
> + if (addr != reg)
> + printk(KERN_ERR "s3c24xx-ac97: req addr = %02x, rep addr = %02x\n", reg, addr);
Try to keep within 80 chars.
> +static void s3c2443_ac97_warm_reset(struct snd_ac97 *ac97)
> +{
> + u32 ac_glbctrl;
> +
> + ac_glbctrl = readl(s3c24xx_ac97.regs + S3C_AC97_GLBCTRL);
> + ac_glbctrl = S3C_AC97_GLBCTRL_WARMRESET;
> + writel(ac_glbctrl, s3c24xx_ac97.regs + S3C_AC97_GLBCTRL);
> + udelay(1000);
msleep() can be better here.
> + ac_glbctrl = 0;
> + writel(ac_glbctrl, s3c24xx_ac97.regs + S3C_AC97_GLBCTRL);
> + udelay(1000);
Ditto (also in s3c2443_ac97_cold_reset and s3c2443_ac97_probe()).
> +static irqreturn_t s3c2443_ac97_irq(int irq, void *dev_id)
> +{
> + int status;
> + u32 ac_glbctrl;
> +
> + status = readl(s3c24xx_ac97.regs + S3C_AC97_GLBSTAT) & codec_ready;
> +
> + if (status) {
> + ac_glbctrl = readl(s3c24xx_ac97.regs + S3C_AC97_GLBCTRL);
Keep the indentation correctly.
thanks,
Takashi
_______________________________________________
Alsa-devel mailing list
Alsa-devel@xxxxxxxxxxxxxxxx
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
