ath9k is continuously resting because of beacon stuck at 5GHz. I am working on a board with AR9462 module. I checked the environment noise levels, my test environment is clean and there is no interference. While checking solution for the issue, I found that if comment the last part of “int ath9k_hw_loadnf(struct ath_hw *ah, struct ath9k_channel *chan)” which is given below, then driver works properly also. But if I remove this part, I expect that driver should not able to perform noise floor calibration properly. I found a related patch, https://patchwork.kernel.org/patch/8215351/, but it is already applied to driver. Also if I increase the minimum CCA power threshold values from the initvals (in ar9462_2p0_initvals.h), driver works properly. But these values are above the allowed values considering ETSI compliance: https://www.spinics.net/lists/linux-wireless/msg118665.html Many thanks in advance for any help. /* * Restore maxCCAPower register parameter again so that we're not capped * by the median we just loaded. This will be initial (and max) value * of next noise floor calibration the baseband does. */ ENABLE_REG_RMW_BUFFER(ah); for (i = 0; i < NUM_NF_READINGS; i++) { if (chainmask & (1 << i)) { if ((i >= AR5416_MAX_CHAINS) && !IS_CHAN_HT40(chan)) continue; REG_RMW(ah, ah->nf_regs[i], (((u32) (-50) << 1) & 0x1ff), 0x1ff); } } REG_RMW_BUFFER_FLUSH(ah); Regards, Cevahir
