From: Mike Isely Date: Sat, 7 Mar 2009 02:58:15 +0000 (-0300) Subject: V4L/DVB (11162): pvrusb2: Tie up loose ends with v4l2-subdev setup X-Git-Url: https://git.stricted.de/?a=commitdiff_plain;h=446dfdc6cc62890098a1b92a4a84019a34dce0cc;p=GitHub%2Fexynos8895%2Fandroid_kernel_samsung_universal8895.git V4L/DVB (11162): pvrusb2: Tie up loose ends with v4l2-subdev setup Tie up loose ends with v4l2-subdev setup. Set attached module's group ID to match our internal ID, emit a few useful messages when sub-devices are dealt with, implement better error legs, and fix an error in the old i2c layer (caused by changes related to the v4l2-subdev work here). Signed-off-by: Mike Isely Signed-off-by: Mauro Carvalho Chehab --- diff --git a/drivers/media/video/pvrusb2/pvrusb2-hdw.c b/drivers/media/video/pvrusb2/pvrusb2-hdw.c index 0e0d086bb278..c764f360a774 100644 --- a/drivers/media/video/pvrusb2/pvrusb2-hdw.c +++ b/drivers/media/video/pvrusb2/pvrusb2-hdw.c @@ -2020,17 +2020,26 @@ static void pvr2_hdw_load_subdev(struct pvr2_hdw *hdw, i2caddr); } + if (!sd) { + pvr2_trace(PVR2_TRACE_ERROR_LEGS, + "Module ID %u for device %s failed to load" + " (this is probably a bad thing...)", + mid, hdw->hdw_desc->description); + return; + } + + /* Tag this sub-device instance with the module ID we know about. + In other places we'll use that tag to determine if the instance + requires special handling. */ + sd->grp_id = mid; + /* If we have both old and new i2c layers enabled, make sure that old layer isn't also tracking this module. This is a debugging aid, in normal situations there's no reason for both mechanisms to be enabled. */ pvr2_i2c_untrack_subdev(hdw, sd); + pvr2_trace(PVR2_TRACE_INIT, "Attached sub-driver %s", fname); - // ????? - /* Based on module ID, we should remember subdev pointers - so that we can send certain custom commands where - needed. */ - // ????? }