ALSA: bebob: dismiss sleep after breaking connections
authorTakashi Sakamoto <o-takashi@sakamocchi.jp>
Fri, 11 Jun 2021 03:50:01 +0000 (12:50 +0900)
committerTakashi Iwai <tiwai@suse.de>
Fri, 11 Jun 2021 08:29:45 +0000 (10:29 +0200)
In a commit d3eabe939aee ("ALSA: bebob: expand sleep just after breaking
connections for protocol version 1"), a workaround was added for a quirk
of freeze in BeBoB protocol version 1. As long as seeing with sequence
replay for media clock recovery, the quirk disappears.

This commit removes the workaround.

Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Link: https://lore.kernel.org/r/20210611035003.26852-2-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Iwai <tiwai@suse.de>
sound/firewire/bebob/bebob_stream.c

index 470c2b70cbfa152294e4aafda1658876f1119e10..6d47c25654e6675298169a5e46dcaccd9a5e8974 100644 (file)
@@ -401,12 +401,6 @@ static void break_both_connections(struct snd_bebob *bebob)
 {
        cmp_connection_break(&bebob->in_conn);
        cmp_connection_break(&bebob->out_conn);
-
-       // These models seem to be in transition state for a longer time. When
-       // accessing in the state, any transactions is corrupted. In the worst
-       // case, the device is going to reboot.
-       if (bebob->version < 2)
-               msleep(600);
 }
 
 static int start_stream(struct snd_bebob *bebob, struct amdtp_stream *stream)