Re: [alsa-devel] ALSA: core: add support for compress_offload
Hello Vinod Koul,
The patch b21c60a4edd2: "ALSA: core: add support for compress_offload" from Dec 23, 2011, leads to the following warning: sound/core/compress_offload.c:85 snd_compr_open() warn: bitwise AND condition is false here
83 if (f->f_flags & O_WRONLY) 84 dirn = SND_COMPRESS_PLAYBACK; 85 else if (f->f_flags & O_RDONLY) 86 dirn = SND_COMPRESS_CAPTURE; 87 else { 88 pr_err("invalid direction\n"); 89 return -EINVAL; 90 }
O_RDONLY is 0 so we never do compressed capture. Perhaps the tests should be:
if ((file->f_flags & O_ACCMODE) == O_WRONLY) dirn = SND_COMPRESS_PLAYBACK; else if (file->f_flags & O_ACCMODE) == O_RDONLY) dirn = SND_COMPRESS_CAPTURE; else return -EINVAL;
Also it seems like a bad thing to print an error here, because can the user flood dmesg?
regards, dan carpenter
At Tue, 4 Sep 2012 19:33:00 +0300, Dan Carpenter wrote:
Hello Vinod Koul,
The patch b21c60a4edd2: "ALSA: core: add support for compress_offload" from Dec 23, 2011, leads to the following warning: sound/core/compress_offload.c:85 snd_compr_open() warn: bitwise AND condition is false here
83 if (f->f_flags & O_WRONLY) 84 dirn = SND_COMPRESS_PLAYBACK; 85 else if (f->f_flags & O_RDONLY) 86 dirn = SND_COMPRESS_CAPTURE; 87 else { 88 pr_err("invalid direction\n"); 89 return -EINVAL; 90 }
O_RDONLY is 0 so we never do compressed capture. Perhaps the tests should be:
if ((file->f_flags & O_ACCMODE) == O_WRONLY) dirn = SND_COMPRESS_PLAYBACK; else if (file->f_flags & O_ACCMODE) == O_RDONLY) dirn = SND_COMPRESS_CAPTURE; else return -EINVAL;
Also it seems like a bad thing to print an error here, because can the user flood dmesg?
Agreed on both points. Care to send a fix patch?
thanks,
Takashi
participants (2)
-
Dan Carpenter
-
Takashi Iwai