From: Jaroslav Kysela <perex@perex.cz>
Date: Tue, 21 Apr 2009 13:30:31 +0000 (+0200)
Subject: [ALSA] intel8x0: another attempt to fix ac97_clock measure routine
X-Git-Url: http://git.osdn.net/view?a=commitdiff_plain;h=30fd9940eee910d847f48bd8740b2d0eaa8d2cfc;p=sagit-ice-cold%2Fkernel_xiaomi_msm8998.git

[ALSA] intel8x0: another attempt to fix ac97_clock measure routine

Appearently, a big delay ~300ms is required before hw is settled and ready
to transfer samples on some hardware variants. Also, return back
"clocking to 48000Hz" message when something fails.

Signed-off-by: Jaroslav Kysela <perex@perex.cz>
---

diff --git a/sound/pci/intel8x0.c b/sound/pci/intel8x0.c
index 5dced5b79387..c4ba486785c6 100644
--- a/sound/pci/intel8x0.c
+++ b/sound/pci/intel8x0.c
@@ -2751,11 +2751,12 @@ static void __devinit intel8x0_measure_ac97_clock(struct intel8x0 *chip)
 	if (pos == 0) {
 		snd_printk(KERN_ERR "intel8x0: measure - unreliable DMA position..\n");
 	      __retry:
-		if (attempt < 2) {
+		if (attempt < 3) {
+			msleep(300);
 			attempt++;
 			goto __again;
 		}
-		return;
+		goto __end;
 	}
 
 	pos /= 4;
@@ -2782,6 +2783,7 @@ static void __devinit intel8x0_measure_ac97_clock(struct intel8x0 *chip)
 	else if (pos < 47500 || pos > 48500)
 		/* not 48000Hz, tuning the clock.. */
 		chip->ac97_bus->clock = (chip->ac97_bus->clock * 48000) / pos;
+      __end:
 	printk(KERN_INFO "intel8x0: clocking to %d\n", chip->ac97_bus->clock);
 	snd_ac97_update_power(chip->ac97[0], AC97_PCM_FRONT_DAC_RATE, 0);
 }