Alsa-Devel Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Sergei Shtylylov <sshtylyov@ru.mvista.com>
To: alsa-devel@lists.sourceforge.net
Subject: [PATCH] AMD Au1x00: fix DMA init/cleanup
Date: Tue, 21 Mar 2006 00:20:28 +0300	[thread overview]
Message-ID: <441F1C9C.3080504@ru.mvista.com> (raw)

[-- Attachment #1: Type: text/plain, Size: 216 bytes --]

Hello.

      AMD Au1x00 ALSA driver causes kernel oops in au1000_init() by trying to
set DMA channel to -1 in yet unallocated audio streams. Here's the patch that
staightens up DMA init/cleanup code.

WBR, Sergei



[-- Attachment #2: Au1x00-ALSA-fix-DMA-init-and-cleanup.patch --]
[-- Type: text/plain, Size: 1798 bytes --]

diff --git a/sound/mips/au1x00.c b/sound/mips/au1x00.c
index 6d8f8b3..a7a221a 100644
--- a/sound/mips/au1x00.c
+++ b/sound/mips/au1x00.c
@@ -610,14 +610,17 @@ snd_au1000_free(struct snd_card *card)
 		release_and_free_resource(au1000->ac97_res_port);
 	}
 
-	if (au1000->stream[PLAYBACK]->dma >= 0)
-		free_au1000_dma(au1000->stream[PLAYBACK]->dma);
-
-	if (au1000->stream[CAPTURE]->dma >= 0)
-		free_au1000_dma(au1000->stream[CAPTURE]->dma);
+	if (au1000->stream[PLAYBACK]) {
+	  	if (au1000->stream[PLAYBACK]->dma >= 0)
+			free_au1000_dma(au1000->stream[PLAYBACK]->dma);
+		kfree(au1000->stream[PLAYBACK]);
+	}
 
-	kfree(au1000->stream[PLAYBACK]);
-	kfree(au1000->stream[CAPTURE]);
+	if (au1000->stream[CAPTURE]) {
+		if (au1000->stream[CAPTURE]->dma >= 0)
+			free_au1000_dma(au1000->stream[CAPTURE]->dma);
+		kfree(au1000->stream[CAPTURE]);
+	}
 }
 
 
@@ -636,15 +630,19 @@ au1000_init(void)
 
 	card->private_free = snd_au1000_free;
 	au1000 = card->private_data;
-	/* so that snd_au1000_free will work as intended */
 	au1000->card = card;
-	au1000->stream[PLAYBACK]->dma = -1;
-	au1000->stream[CAPTURE]->dma = -1;
- 	au1000->ac97_res_port = NULL;
+
 	au1000->stream[PLAYBACK] = kmalloc(sizeof(struct audio_stream), GFP_KERNEL);
-	au1000->stream[CAPTURE] = kmalloc(sizeof(struct audio_stream), GFP_KERNEL);
+	au1000->stream[CAPTURE ] = kmalloc(sizeof(struct audio_stream), GFP_KERNEL);
+	/* so that snd_au1000_free will work as intended */
+ 	au1000->ac97_res_port = NULL;
+	if (au1000->stream[PLAYBACK])
+		au1000->stream[PLAYBACK]->dma = -1;
+	if (au1000->stream[CAPTURE ])
+		au1000->stream[CAPTURE ]->dma = -1;
+
 	if (au1000->stream[PLAYBACK] == NULL ||
-	    au1000->stream[CAPTURE] == NULL) {
+	    au1000->stream[CAPTURE ] == NULL) {
 		snd_card_free(card);
 		return -ENOMEM;
 	}



             reply	other threads:[~2006-03-20 21:20 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-03-20 21:20 Sergei Shtylylov [this message]
  -- strict thread matches above, loose matches on Subject: below --
2006-03-21  0:59 [PATCH] AMD Au1x00: fix DMA init/cleanup Sergei Shtylylov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=441F1C9C.3080504@ru.mvista.com \
    --to=sshtylyov@ru.mvista.com \
    --cc=alsa-devel@lists.sourceforge.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox