linux-kernel - Re: OSS audio compatibility layer regression - Linux-3.x

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <s5h1usaw1ut.wl%tiwai@suse.de>
Date:	Mon, 12 Dec 2011 10:55:38 +0100
From:	Takashi Iwai <tiwai@...e.de>
To:	wallak@...e.fr
Cc:	linux-kernel@...r.kernel.org
Subject: Re: OSS audio compatibility layer regression - Linux-3.x - unable to register OSS PCM device 0:0

At Sat, 10 Dec 2011 20:32:54 +0100,
wallak@...e.fr wrote:
> 
> I've an issue, when I use the SND_OSSEMUL compatibility layer with the
> snd_intel8x0 module. The devices can't be registered. Indeed, theses devices are
> registered at the oss_init() stage, and are not available later. A patch to fix
> this issue is available below. The patch uses the previous linux behavior.

I guess you set CONFIG_SOUND_OSS=y?  This conflicts with the OSS
emulation.  It looked as if both can co-exist in the former kernels,
but it's just because we didn't check it strictly.  Both try to
register the very same devices -- especially OSS does this at init
even before needed.
As a workaround, just disable CONFIG_SOUND_OSS unless you really need
it.  Or, if this isn't the case, give your .config file.

If this is the case, we can add a Kconfig rule to exclude the
conflict in future...


thanks,

Takashi

> 
> 
> Best Regards,
> Wallak.
> 
> 
> #log output
> snd_intel8x0 0000:00:1f.5: PCI INT B -> Link[LNKB] -> GSI 10 (level, low) -> IRQ
> 10
> intel8x0_measure_ac97_clock: measured 54899 usecs (2645 samples)
> intel8x0: clocking to 48000
> unable to register OSS PCM device 0:0
> Bridge firewalling registered
> 
> #patch fixing this issue: (commit 848669da3a92fa6ab815e0517af3294afb3ea928 adds
> this issue)
> --- linux-3.1.5-mdf/sound/oss/soundcard.c.orig    2011-12-09 17:57:05.000000000
> +0100
> +++ linux-3.1.5-mdf/sound/oss/soundcard.c    2011-12-10 15:18:02.000000000 +0100
> @@ -526,21 +526,31 @@
>  }
> 
> 
> +/* These device names follow the official Linux device list,
> + * Documentation/devices.txt.  Let us know if there are other
> + * common names we should support for compatibility.
> + * Only those devices not created by the generic code in sound_core.c are
> + * registered here.
> + */
> +static const struct {
> +    unsigned short minor;
> +    char *name;
> +    umode_t mode;
> +    int *num;
> +} dev_list[] = { /* list of minor devices */
> +/* seems to be some confusion here -- this device is not in the device list */
> +    {SND_DEV_DSP16,     "dspW",     S_IWUGO | S_IRUSR | S_IRGRP,
> +     &num_audiodevs},
> +    {SND_DEV_AUDIO,     "audio",     S_IWUGO | S_IRUSR | S_IRGRP,
> +     &num_audiodevs},
> +};
> +
>  static int dmabuf;
>  static int dmabug;
> 
>  module_param(dmabuf, int, 0444);
>  module_param(dmabug, int, 0444);
> 
> -/* additional minors for compatibility */
> -struct oss_minor_dev {
> -    unsigned short minor;
> -    unsigned int enabled;
> -} dev_list[] = {
> -    { SND_DEV_DSP16 },
> -    { SND_DEV_AUDIO },
> -};
> -
>  static int __init oss_init(void)
>  {
>      int             err;
> @@ -561,12 +571,18 @@
>      sound_dmap_flag = (dmabuf > 0 ? 1 : 0);
> 
>      for (i = 0; i < ARRAY_SIZE(dev_list); i++) {
> -        j = 0;
> -        do {
> -            unsigned short minor = dev_list[i].minor + j * 0x10;
> -            if (!register_sound_special(&oss_sound_fops, minor))
> -                dev_list[i].enabled = (1 << j);
> -        } while (++j < num_audiodevs);
> +        device_create(sound_class, NULL,
> +                  MKDEV(SOUND_MAJOR, dev_list[i].minor), NULL,
> +                  "%s", dev_list[i].name);
> +
> +        if (!dev_list[i].num)
> +            continue;
> +
> +        for (j = 1; j < *dev_list[i].num; j++)
> +            device_create(sound_class, NULL,
> +                      MKDEV(SOUND_MAJOR,
> +                        dev_list[i].minor + (j*0x10)),
> +                      NULL, "%s%d", dev_list[i].name, j);
>      }
> 
>      if (sound_nblocks >= MAX_MEM_BLOCKS - 1)
> @@ -580,11 +596,11 @@
>      int i, j;
> 
>      for (i = 0; i < ARRAY_SIZE(dev_list); i++) {
> -        j = 0;
> -        do {
> -            if (dev_list[i].enabled & (1 << j))
> -                unregister_sound_special(dev_list[i].minor);
> -        } while (++j < num_audiodevs);
> +        device_destroy(sound_class, MKDEV(SOUND_MAJOR, dev_list[i].minor));
> +        if (!dev_list[i].num)
> +            continue;
> +        for (j = 1; j < *dev_list[i].num; j++)
> +            device_destroy(sound_class, MKDEV(SOUND_MAJOR, dev_list[i].minor +
> (j*0x10)));
>      }
> 
>      unregister_sound_special(1);
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/