yes, of course. i only was not able to trace it before, because i greped to the wrong message. i searched for the message that 3.0.X kernels give, but the message changed, the problem is still there, but it's just the messages. check the logs i sent you about 3.1.1 kernel, you ll find these message(s) there as well. so there we have always the same problem with intel drivers: 1.) on iwlwifi/iwlagn the option aggregation_auto=off is missing 2.) on i915 the option intel_backlight=off is missing and the change between 3.0 and 3.1 kernels was confusing, because it just changed the way how the driver manifests itself. i can not believe that all these patches were only about that. On Mon, 14 Nov 2011 15:31:11 -0800 wwguy <wey-yi.w.guy@xxxxxxxxx> wrote: > On Mon, 2011-11-14 at 15:30 -0800, Lars Schotte wrote: > > kernel 3.1.0 and kernel 3.1.1 say this: > > > > [ 74.798240] iwlagn 0000:02:00.0: Tx aggregation enabled on ra = > > 20:4e:7f:5b:2d:f5 tid = 0 [ 190.506907] iwlagn 0000:02:00.0: Tx > > aggregation enabled on ra = 20:4e:7f:5b:2d:f5 tid = 0 [ 242.249951] > > iwlagn 0000:02:00.0: Tx aggregation enabled on ra = > > 20:4e:7f:5b:2d:f5 tid = 0 [ 531.725659] iwlagn 0000:02:00.0: Tx > > aggregation enabled on ra = 20:4e:7f:5b:2d:f5 tid = 0 > > [ 700.443255] iwlagn 0000:02:00.0: Tx aggregation enabled on ra = > > 20:4e:7f:5b:2d:f5 tid = 0 [ 794.438484] iwlagn 0000:02:00.0: Tx > > aggregation enabled on ra = 20:4e:7f:5b:2d:f5 tid = 0 > > [ 1228.443709] iwlagn 0000:02:00.0: Tx aggregation enabled on ra = > > 20:4e:7f:5b:2d:f5 tid = 0 [ 1427.104132] iwlagn 0000:02:00.0: Tx > > aggregation enabled on ra = 20:4e:7f:5b:2d:f5 tid = 0 > > [ 1669.849989] iwlagn 0000:02:00.0: Tx aggregation enabled on ra = > > 20:4e:7f:5b:2d:f5 tid = 0 [ 1846.380470] iwlagn 0000:02:00.0: Tx > > aggregation enabled on ra = 20:4e:7f:5b:2d:f5 tid = 0 > > [ 1919.721844] iwlagn 0000:02:00.0: Tx aggregation enabled on ra = > > 20:4e:7f:5b:2d:f5 tid = 0 [ 2042.692520] iwlagn 0000:02:00.0: Tx > > aggregation enabled on ra = 20:4e:7f:5b:2d:f5 tid = 6 > > [ 2143.464146] iwlagn 0000:02:00.0: Tx aggregation enabled on ra = > > 20:4e:7f:5b:2d:f5 tid = 0 [ 2157.828495] iwlagn 0000:02:00.0: Tx > > aggregation enabled on ra = 20:4e:7f:5b:2d:f5 tid = 0 > > > > that looks like a success, however the speed doesnt change, because > > the AP is set to only use HT20, so 1x20 MHz channel. > > now you confuse me, it is a success? the problem is showing up, right? > > Wey > > > > On Mon, 14 Nov 2011 15:17:08 -0800 > > wwguy <wey-yi.w.guy@xxxxxxxxx> wrote: > > > > > On Mon, 2011-11-14 at 15:16 -0800, Lars Schotte wrote: > > > > i looked over some changelogs .. from the linux kernel 3.1 > > > > branch. > > > > > > > > i saw that you ve introduced some patches to this, or you at > > > > least signed them off. > > > > > > > > the changelog also says that you've changed the logging level > > > > for that aggregation problem, so even if it would be there, it > > > > doesnt pop up on kernel 3.1's. > > > > > > > > i also did read there that there was like waiting for a real > > > > reason to move to aggregation. so it was there the whole time, > > > > only it did not write itself out on the log. > > > > > > > > so i am not sure if we were not just hunting ghosts here. > > > > > > > you are correct, I do have patch to modify the log level and make > > > sur ewe are not generate the un-necessary messages in the log and > > > cause confusion. But I don't think it is related to what the > > > behavior you are seeing (at least I hope not); but I will double > > > check. > > > > > > Thanks > > > Wey > > > > > > > On Mon, 14 Nov 2011 14:59:33 -0800 > > > > wwguy <wey-yi.w.guy@xxxxxxxxx> wrote: > > > > > > > > > ok, thanks > > > > > > > > > > please do let us know if it appear on 3.1.1 > > > > > > > > > > at the meantime, we will look into how to backport to 3.0.x > > > > > > > > > > Regards > > > > > Wey > > > > > > > > > > On Mon, 2011-11-14 at 14:56 -0800, Lars Schotte wrote: > > > > > > i definitely see it in 3.0.8 and 3.0.9. > > > > > > > > > > > > on the fedora fc16 linux kernel 3.1.1 it did not yet > > > > > > appear, but that may change. > > > > > > > > > > > > so maybe it would be worth a look if it is not fixed by > > > > > > 3.1.1, or if something changed. however ... if it is a FIX, > > > > > > it should definitely be fixed also in 3.0.X branch. > > > > > > > > > > > > so that message still is there on 3.0.9 and 3.0.8, we will > > > > > > see about the more current ones. > > > > > > > > > > > > On Mon, 14 Nov 2011 14:41:42 -0800 > > > > > > wwguy <wey-yi.w.guy@xxxxxxxxx> wrote: > > > > > > > > > > > > > so youno longer seeing the issue? > > > > > > > > > > > > > > Wey > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- Lars Schotte @ Hana (F16) -- 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
