W dniu 14 sierpnia 2011 00:17 użytkownik David Woodhouse <[email protected]> napisał: > On Sun, 2011-08-14 at 00:12 +0200, Rafał Miłecki wrote: >> I've just checked wl and they don't do that anyway. The keep order >> depending on the alignment bool. > > Ah, OK. So they still write TXCTL first, if the device can't support > unaligned descriptors? > > The brcm80211 driver doesn't; it writes TXCTL last, unconditionally. > > Do you want me to send another patch which makes it conditional? It > complicates the code with something that might be unnecessary legacy > baggage... but means we don't have to be worried about regressions on > older hardware.
Here you have a log from wl from 14e4:432b: write32 0xfaafc208 <- 0x00000ff0 read32 0xfaafc208 -> 0x00000000 [that means no unaligned support] ... read32 0xfaafc200 -> 0x00000000 write32 0xfaafc200 <- 0x00000801 write32 0xfaafc208 <- 0x1b550000 write32 0xfaafc20c <- 0x80000000 [when no unaligned, first enable, then set addresses] -- Rafał _______________________________________________ b43-dev mailing list [email protected] http://lists.infradead.org/mailman/listinfo/b43-dev
