| Message ID | 1344195337-27461-6-git-send-email-mickflemm@gmail.com (mailing list archive) |
|---|---|
| State | Not Applicable, archived |
| Headers | show |
Hi Nick, hi Felix, This patch does break the operation of all DCMA82 (AR5413), as this chips have only 8 valid piers, but with the "continue" all 10 got read. So the former version was correct as it break after 8 piers. The patch was initiated by my observation and after revisiting what I did, blame Thomas is correct here... The power variation is properly working and I miss measured based on an uncleaned patch mess in my env. Please revert this patch to get proper pier readings back in ath5k. Greetings Thomas Nick Kossifidis schrieb: > If we have a zeroed frequency on the calibration piers it means that we > shouldn't use that pier, not stop reading the EEPROM and break out from > the loop. By doing that we return the wrong offset and the whole dataset > gets corrupted. > > Signed-off-by: Nick Kossifidis <mickflemm@gmail.com> > Tested-by: Thomas Huehn <thomas@net.t-labs.tu-berlin.de> > --- > drivers/net/wireless/ath/ath5k/eeprom.c | 4 ++-- > 1 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/drivers/net/wireless/ath/ath5k/eeprom.c b/drivers/net/wireless/ath/ath5k/eeprom.c > index 4026c90..10a4396 100644 > --- a/drivers/net/wireless/ath/ath5k/eeprom.c > +++ b/drivers/net/wireless/ath/ath5k/eeprom.c > @@ -522,7 +522,7 @@ ath5k_eeprom_read_freq_list(struct ath5k_hw *ah, int *offset, int max, > > freq1 = val & 0xff; > if (!freq1) > - break; > + continue; > > pc[i++].freq = ath5k_eeprom_bin2freq(ee, > freq1, mode); > @@ -530,7 +530,7 @@ ath5k_eeprom_read_freq_list(struct ath5k_hw *ah, int *offset, int max, > > freq2 = (val >> 8) & 0xff; > if (!freq2) > - break; > + continue; > > pc[i++].freq = ath5k_eeprom_bin2freq(ee, > freq2, mode); -- To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
2012/8/8 Thomas Huehn <thomas@net.t-labs.tu-berlin.de>: > Hi Nick, hi Felix, > > This patch does break the operation of all DCMA82 (AR5413), as this > chips have only 8 valid piers, but with the "continue" all 10 got read. > So the former version was correct as it break after 8 piers. > The patch was initiated by my observation and after revisiting what I > did, blame Thomas is correct here... The power variation is properly > working and I miss measured based on an uncleaned patch mess in my env. > > Please revert this patch to get proper pier readings back in ath5k. > > > Greetings Thomas > ACK so documentation is once again misleading because it says that we always have 10 frequencies to read and if some are zero we just ignore the matching data. It doesn't say that if we get a zeroed frequency the rest are missing from the EEPROM and we should move on to the next section. I'll keep that in mind ;-) John please ignore this one and sorry for the mess !
diff --git a/drivers/net/wireless/ath/ath5k/eeprom.c b/drivers/net/wireless/ath/ath5k/eeprom.c index 4026c90..10a4396 100644 --- a/drivers/net/wireless/ath/ath5k/eeprom.c +++ b/drivers/net/wireless/ath/ath5k/eeprom.c @@ -522,7 +522,7 @@ ath5k_eeprom_read_freq_list(struct ath5k_hw *ah, int *offset, int max, freq1 = val & 0xff; if (!freq1) - break; + continue; pc[i++].freq = ath5k_eeprom_bin2freq(ee, freq1, mode); @@ -530,7 +530,7 @@ ath5k_eeprom_read_freq_list(struct ath5k_hw *ah, int *offset, int max, freq2 = (val >> 8) & 0xff; if (!freq2) - break; + continue; pc[i++].freq = ath5k_eeprom_bin2freq(ee, freq2, mode);