Johannes Berg <johannes@xxxxxxxxxxxxxxxx> writes: > On Thu, 2014-04-24 at 09:49 +0200, Bjørn Mork wrote: > >> I realize that icmp echos don't necessarily tell the whole truth, but >> the same latency difference is noticable on e.g interactive ssh sessions >> as well. So I believe pinging illustrates the difference pretty well. >> The *only* thing changed between these sessions is the iwl >> firmware. Position, channel, driver, hardware is the same. This was >> captured on 2.4GHz. > > There might be differences in how the driver treats the firmware - I > think it probably turns off powersave for the older one. Have you tried > turning it off manually? Thanks for the tip. Yes, it seems to be related to that. Using modprobe iwlmvm power_scheme=1 avoids the problem. I also tried using modprobe iwlwifi power_save=N with iwlmvm.power_scheme=2 (default), but that made no difference. The problem was still present with this setting. Bjørn -- To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
