Hello Nick Kossifidis,
The patch 3017fcab416d: "ath5k: Extend get_default_sifs/slot_time"
from Nov 23, 2010, leads to the following warning:
drivers/net/wireless/ath/ath5k/pcu.c:223 ath5k_hw_get_default_sifs()
warn: missing break? reassigning 'sifs'
209 switch (ah->ah_bwmode) {
210 case AR5K_BWMODE_40MHZ:
211 sifs = AR5K_INIT_SIFS_TURBO;
212 break;
213 case AR5K_BWMODE_10MHZ:
214 sifs = AR5K_INIT_SIFS_HALF_RATE;
215 break;
216 case AR5K_BWMODE_5MHZ:
217 sifs = AR5K_INIT_SIFS_QUARTER_RATE;
218 break;
219 case AR5K_BWMODE_DEFAULT:
220 sifs = AR5K_INIT_SIFS_DEFAULT_BG;
221 default:
Was it intentional to fall through to the default case here? Most of
the intentional fall through sites (80-90%) have a /* fall through */
comment.
222 if (channel->band == IEEE80211_BAND_5GHZ)
223 sifs = AR5K_INIT_SIFS_DEFAULT_A;
224 break;
225 }
regards,
dan carpenter
2013/1/11 Dan Carpenter <[email protected]>:
> ath5k: Extend get_default_sifs/slot_time
Hello Dan and thanks a lot for the review !
I should probably just move
sifs = AR5K_INIT_SIFS_DEFAULT_BG;
to default: and treat AR5K_BWMODE_DEFAULT and default: the same way,
it's misleading this way + if bwmode is not set during attach (right
now it does), sifs might stay unassigned. The same problem occurs with
ath5k_hw_get_default_slottime also.
I'll post a patch as soon as i find some time to work on ath5k again,
right now I have some physics exams at uni. It works but you are right
it's ugly.
Again thanks a lot for your time :-)
--
GPG ID: 0xEE878588
As you read this post global entropy rises. Have Fun ;-)
Nick