Michael Buesch wrote: > On Saturday 03 February 2007 18:50, Larry Finger wrote: >> If you are having trouble with communications after a suspend/resume cycle, >> please try the patch >> below. The patch was made against wireless-2.6, but should apply to any >> 2.6.18 or later kernel. I >> cannot test this one as my video driver refuses to suspend. > > I think this is redundant. We clear (=0xFFFF) txctl2 on resume, too, because > we call init_board, which calls the function to init the structures.
I don't think that happens on SoftMAC. The original initialization is in read_phyinfo, which is called from attach_board, and that is only called from init_one. I don't see init_board calling anything that touches txctl2. As a result, I only think the initialization happens at initial module load, not at resume. Of course, the original txctl2 initialization has a "FIXME: this is in the wrong place" attached to it. Perhaps that should be moved to init_board. Are there any other initializations that must be considered? Larry _______________________________________________ Bcm43xx-dev mailing list [email protected] https://lists.berlios.de/mailman/listinfo/bcm43xx-dev
