[From nobody Tue Apr  1 23:21:01 2008
Date: Tue, 01 Apr 2008 15:23:58 -0400
From: Tyson Whitehead &lt;twhitehe@uwo.ca&gt;
Subject: Re: [alsa-devel] [regression] 2.6.25-rc4 snd-es18xx broken on Alpha
In-reply-to: &lt;20080329063523.979EEDBA2@gherkin.frus.com&gt;
To: Bob Tracy &lt;rct@frus.com&gt;
Cc: Takashi Iwai &lt;tiwai@suse.de&gt;
Message-id: &lt;47F28BCE.9090108@uwo.ca&gt;
MIME-version: 1.0
Content-type: text/plain; charset=ISO-8859-1; format=flowed
Content-transfer-encoding: 7BIT
References: &lt;20080329063523.979EEDBA2@gherkin.frus.com&gt;
User-Agent: Mozilla-Thunderbird 2.0.0.9 (X11/20080109)

Bob Tracy wrote:
&gt; Takashi Iwai wrote:
&gt;&gt; I vaguely remember about the patch...  The patch below was on my local
&gt;&gt; tree but never pushed because of lack of testing.  Does it work for
&gt;&gt; you?
&gt;&gt;
&gt;&gt; ---
&gt;&gt;
&gt;&gt; diff -r 82e6201fc907 sound/isa/es18xx.c
&gt;&gt; --- a/sound/isa/es18xx.c	Mon Mar 17 14:36:24 2008 +0100
&gt;&gt; +++ b/sound/isa/es18xx.c	Mon Mar 17 17:32:59 2008 +0100
&gt;&gt; @@ -765,9 +765,10 @@ static irqreturn_t snd_es18xx_interrupt(
&gt;&gt;  		/* Read Interrupt status */
&gt;&gt;  		status = snd_es18xx_mixer_read(chip, 0x7f) &gt;&gt; 4;
&gt;&gt;  	}
&gt;&gt; -#if 0
&gt;&gt; -	else {
&gt;&gt; -		status = 0;
&gt;&gt; +
&gt;&gt; +#ifdef CONFIG_ALPHA
&gt;&gt; +	if (!(status &amp; (AUDIO1_IRQ | AUDIO2_IRQ))) {
&gt;&gt; +		/* status = 0; */
&gt;&gt;  		if (inb(chip-&gt;port + 0x0C) &amp; 0x01)
&gt;&gt;  			status |= AUDIO1_IRQ;
&gt;&gt;  		if (snd_es18xx_mixer_read(chip, 0x7A) &amp; 0x80)
&gt;&gt; @@ -777,7 +778,6 @@ static irqreturn_t snd_es18xx_interrupt(
&gt;&gt;  			status |= HWV_IRQ;
&gt;&gt;  	}
&gt;&gt;  #endif
&gt;&gt; -
&gt;&gt;  	/* Audio 1 &amp; Audio 2 */
&gt;&gt;          if (status &amp; AUDIO2_IRQ) {
&gt;&gt;                  if (chip-&gt;active &amp; DAC2)
&gt;&gt; --
&gt; 
&gt; The above patch doesn't make any difference for me as far as how the
&gt; ES1888 works.  I get the same broken behavior as previously described.

I wonder if the difference between Bob and I (i.e., that I can get sound 
working with patching while he cannot) might be that all along I've been 
using CONFIG_ALPHA_GENERIC, while he has (possibly -- he will have to 
confirm this) been using CONFIG_ALPHA_MIATA.

As I mentioned in the email I just sent out, I just finished discovering 
that if I compile with CONFIG_ALPHA_MIATA, I can't get sound to work no 
matter what I do (patched or unpatched es18xx or sb8).

When I compile with CONFIG_ALPHA_GENERIC I can get sound working (modulo 
this new &quot;Bad page state in process 'mpg321'&quot; thing that started 
sometime after 2.6.14) with either the patched es18xx or standard sb8.

Cheers!  -Tyson

]