On Friday 18 July 2008 20:19:06 Nick Kossifidis wrote: > 2008/7/17 Bruno Randolf <[EMAIL PROTECTED]>: > > hi! > > > > this is just a preview and a RFC. i would like to know what you think > > about the general approach and will send a cleaned up version for final > > review later. > > I fully agree, i wanted to clean this stuff for a long time (that's > why i marked it as a TODO).
good :) i'll complete and send in the patch ASAP then. > > what i left out is control_rates. it would be good if we could get that > > information from mac80211 somehow but if that is not possible we can set > > up a small table for them. they are the same for a given rate in any > > mode. > > Control_rates is not difficult to implement, it's used together with > ack rate stuff... yes, it's no big deal. > > also i think i figured a way for reverse conversion from hw code to rate > > index. > > Have you tested it ? yes. i think it works. i noticed only one strange thing: sometimes the HW gives us rate values which we don't know, for example 28. but maybe that's a ATH5K_RATE_CODE_11M (24) plus AR5K_SET_SHORT_PREAMBLE 0x04. i didn't know about the short preamble flag until now. > > /* adding this flag to rate_code enables short preamble, see > > ar5212_reg.h */ #define AR5K_SET_SHORT_PREAMBLE 0x04 > > Can we please make use of this flag ? (see below) sure. i missed that. > I think it's better if we have them in hex (it's a hw value)... allright, changed. bruno _______________________________________________ ath5k-devel mailing list ath5k-devel@lists.ath5k.org https://lists.ath5k.org/mailman/listinfo/ath5k-devel
