[SOLVED] slow intermittent wireless on AC-8260

mcloaked · 2017-01-08 22:18:59

I have a Skylake NUC (NUC6i5SYH) with Intel AC-8260 wireless.

With everything up to date I was seeing somewhat slow and intermittent wifi performance.

Looking at the output of lspci -v:

01:00.0 Network controller: Intel Corporation Wireless 8260 (rev 3a)
        Subsystem: Intel Corporation Device 9010
        Flags: bus master, fast devsel, latency 0, IRQ 278
        Memory at df000000 (64-bit, non-prefetchable) [size=8K]
        Capabilities: <access denied>
        Kernel driver in use: iwlwifi
        Kernel modules: iwlwifi

and noting from the journal log:

Jan 08 22:04:48 home2 kernel: iwlwifi 0000:01:00.0: enabling device (0000 -> 0002)
Jan 08 22:04:48 home2 kernel: iwlwifi 0000:01:00.0: Direct firmware load for iwlwifi-8000C-24.ucode failed with error -2
Jan 08 22:04:48 home2 kernel: iwlwifi 0000:01:00.0: Direct firmware load for iwlwifi-8000C-23.ucode failed with error -2
Jan 08 22:04:48 home2 kernel: iwlwifi 0000:01:00.0: loaded firmware version 22.361476.0 op_mode iwlmvm
Jan 08 22:04:48 home2 kernel: iwlwifi 0000:01:00.0: Detected Intel(R) Dual Band Wireless AC 8260, REV=0x208

and looking at the available microcode files:

$ ls /lib/firmware/iwlwifi-8000C*
/lib/firmware/iwlwifi-8000C-13.ucode  /lib/firmware/iwlwifi-8000C-21.ucode
/lib/firmware/iwlwifi-8000C-16.ucode  /lib/firmware/iwlwifi-8000C-22.ucode

I see that the kernel module seemed to be attempting to get at revision 23 and 24 which are not yet in the firmware directory.

I have made a little progress by making a module modification via the file:

$ cat /etc/modprobe.d/iwlwifi.conf 
options iwlwifi 11n_disable=1

and after rebooting the wireless at least is not stopping for periods but it is not giving the speed I would hope for (of course disabling wireless-N will slow down the wireless speed) but I am still not getting good speeds.

I have seen a report from another distribution earlier in the year that some firmware versions are buggy but I could not find any current information on the status of support for this wireless chip.

Can anyone offer any workarounds to get an improved and consistent wifi connection with AC-8260?

Edit: The version 25 for the firmware is available at http://www.intel.co.uk/content/www/uk/e … 05511.html but isn't yet in the firmware files. Maybe there is an update that will provide the more recent version before too long? Though it is not clear if that would fix the underlying problem.

Edit2: For reference I have a laptop with N-7260 Intel wireless that gets transfer speeds of around 15MByte/s (120Mbps) whereas the AC-8260 at the same spot connected to the same AP is getting around 3MByte/s (24Mbps) with intermittent connection or around 1MByte/s (8Mbps) when 11N is disabled but less intermittent.

Last edited by mcloaked (2017-02-28 18:19:17)

jeremy31 · 2017-01-09 22:56:54

Does

iwconfig

show power management on?

I would also suggest

echo "options iwlwifi 11n_disable=8" | sudo tee -a /etc/modprobe.d/iwlwifi.conf

Reboot and see if enabling aggressive tx helps

mcloaked · 2017-01-10 16:48:02

jeremy31 wrote:

Does
iwconfig
show power management on?
I would also suggest
echo "options iwlwifi 11n_disable=8" | sudo tee -a /etc/modprobe.d/iwlwifi.conf
Reboot and see if enabling aggressive tx helps

I am away from that machine for a few days but I will try that when I next have access and report back.

xfzfflm · 2017-01-13 04:50:53

I have the same problem.

rgzfxf6bhu · 2017-01-13 09:41:23

I also have the same problem, plus the connection drops quite easily (see https://bbs.archlinux.org/viewtopic.php?id=215300).
It got a bit better though with the drops, but still not comparable to the connections of other laptops that I tested.

Did you try to disable bluetooth in the BIOS? Does the connection get better?

mcloaked · 2017-01-13 14:23:38

rgzfxf6bhu wrote:

I also have the same problem, plus the connection drops quite easily (see https://bbs.archlinux.org/viewtopic.php?id=215300).
It got a bit better though with the drops, but still not comparable to the connections of other laptops that I tested.
Did you try to disable bluetooth in the BIOS? Does the connection get better?

I am still not yet back with physical access to my machine to test further - but a comment I can add is that with netctl once the connection drops it says dropped until manually restarted. With NetworkManager it drops and recovers itself (maybe NM is a little cleverer with managing it when the connection drops out?). I know that the bluetooth is combined in the same chip but would be useful to know if disabling it helps the wifi signal management.

By the way the link to the other thread seems to imply that the firmware after version 16 is buggy?

Last edited by mcloaked (2017-01-13 14:24:55)

rgzfxf6bhu · 2017-01-13 15:12:08

mcloaked wrote:

By the way the link to the other thread seems to imply that the firmware after version 16 is buggy?

I would say no. Because when I use firmware version 16 on my current arch here then I still have the same issue.
However, when I tried it with Ubuntu live (which was using fw version 16 back then) it was working way better.
But I was never able to pin point it down to what made the differences. I went through the config files as well. And used the same config in arch, but no success.

Haven't tried it with the latest Ubuntu live though...

rgzfxf6bhu · 2017-01-13 15:30:37

mcloaked wrote:

Edit: The version 25 for the firmware is available at http://www.intel.co.uk/content/www/uk/e … 05511.html but isn't yet in the firmware files. Maybe there is an update that will provide the more recent version before too long? Though it is not clear if that would fix the underlying problem.

Doesn't seem to be a new driver though, downloaded it and checked it against the one that is in the firmware files:

diff -s /lib/firmware/iwlwifi-8000C-13.ucode ~/Downloads/iwlwifi-8000-ucode-25.30.13.0/iwlwifi-8000-ucode-25.30.13.0/iwlwifi-8000C-13.ucode
---------------------
Files /lib/firmware/iwlwifi-8000C-13.ucode and ~/Downloads/iwlwifi-8000-ucode-25.30.13.0/iwlwifi-8000-ucode-25.30.13.0/iwlwifi-8000C-13.ucode are identical

It's just the file name where one might think that there is a version 25 out there, but check the README:

The files iwlwifi-8000C-13.ucode provided in this package
must be present on your system in order for the Intel Wireless WiFi Link
AGN driver for Linux (iwlwifi) to operate on your system.

The "-13" in the filename reflects an interface/architecture version number.
It will change only when changes in new uCode releases make the new uCode
incompatible with earlier drivers.

jeremy31 · 2017-01-13 22:00:43

The 25 version of the firmware will not load in your kernel as the dmesg error shows the 24 version is the newest it will use- you could try renaming the 25 to 24 and see if it will load.

If it is true that there is an issue with 16 and newer you could rename the iwlwifi-8000C files for version 16+ and add .bak to the filename and reboot, check

modinfo iwlwifi | grep 8000

To see if the 13 version is listed in the firmware like

firmware:       iwlwifi-8000-13.ucode

As this will tell you the oldest firmware version supported

rgzfxf6bhu · 2017-01-14 08:26:13

jeremy31 wrote:

The 25 version of the firmware will not load in your kernel as the dmesg error shows the 24 version is the newest it will use- you could try renaming the 25 to 24 and see if it will load.

Where do you find the firmware iwlwifi-8000C-25.ucode?
There is no 25 on the intel website and nor one here: https://wireless.wiki.kernel.org/en/use … rs/iwlwifi
The file name iwlwifi-8000-ucode-25.30.13.0.tgz is misleading, see my previous post https://bbs.archlinux.org/viewtopic.php … 0#p1683530

jeremy31 wrote:

If it is true that there is an issue with 16 and newer you could rename the iwlwifi-8000C files for version 16+ and add .bak to the filename and reboot, check
modinfo iwlwifi | grep 8000
To see if the 13 version is listed in the firmware like
firmware:       iwlwifi-8000-13.ucode
As this will tell you the oldest firmware version supported

Did you read my previous comment?
https://bbs.archlinux.org/viewtopic.php … 7#p1683527
I don't think we can say that there is an issue with 16 or newer ...

Last edited by rgzfxf6bhu (2017-01-14 09:37:12)

mcloaked · 2017-01-14 13:50:37

I thought I would do a quick check on firmware difference:

$ diff -s /lib/firmware/iwlwifi-8000C-16.ucode /lib/firmware/iwlwifi-8000C-22.ucode 
Binary files /lib/firmware/iwlwifi-8000C-16.ucode and /lib/firmware/iwlwifi-8000C-22.ucode differ

So it does seem that version 16 which people report works, and version 22 which is the version that gets loaded in my machine are indeed different, but what the differences are would need a look at the details of the source files. Is it possible that version 25 was set to revert to the last known version that did work properly whilst the developers look at writing an updated version that does not have the regressions?

Last edited by mcloaked (2017-01-14 13:53:32)

rgzfxf6bhu · 2017-01-14 14:12:36

mcloaked wrote:

Is it possible that version 25 was set to revert to the last known version that did work properly whilst the developers look at writing an updated version that does not have the regressions?

I am still puzzled who do you know that there even is a version 25. I don't see a version 25 anywhere (there might be one in the future, but currently I cannot find a version 25).
The link in your first post just shows me that there is a file that I can download which is called iwlwifi-8000-ucode-25.30.13.0.tgz.
But this file is from June 2015, and does not contain version 25 of the firmware.

So how do you know there ever was a version 25 ??????

Last edited by rgzfxf6bhu (2017-01-14 14:13:22)

rgzfxf6bhu · 2017-01-14 15:05:56

Okay I still haven't found version 25 though, but there is since 44 hours a version 27 in iwlwifi/linux-firmware.

iwlwifi/linux-firmware that feeds the official linux-firmware tree. It contains early releases, or content that just hasn't been merged in linux-firmware yet.

CAUTION! I tried it but created a lot of other problems here:
http://git.kernel.org/cgit/linux/kernel … C-27.ucode

Yes, I tried it, but my XPS 13 was super unresponsive and I couldn't even start NetworkManager and nothing else. So I just removed it again...

EDIT:

Having a look at the logs:

[    1.590168] iwlwifi 0000:3a:00.0: capa flags index 3 larger than supported by driver
[    1.590182] iwlwifi 0000:3a:00.0: Driver unable to support your firmware API. Driver supports v24, firmware is v27.

Last edited by rgzfxf6bhu (2017-01-14 15:06:25)

mcloaked · 2017-01-14 15:57:17

It seems from http://git.kernel.org/cgit/linux/kernel … mware.git/ that firmware version 22 was updated 7 days ago - so presumably the updated file now differs from the version 22 that is in the official linux-firmware package? Maybe version 27 is a work in progress?

rgzfxf6bhu · 2017-01-14 16:39:10

mcloaked wrote:

... presumably the updated file now differs from the version 22 that is in the official linux-firmware package?

I just checked, and yes it does.
Will probably try it out later... Don't have time today.

EDIT:
Did you try the new v22?
I just tried it, but my impression is that there is no improvement regarding our issue.

Last edited by rgzfxf6bhu (2017-01-16 07:45:49)

mcloaked · 2017-01-16 16:32:36

rgzfxf6bhu wrote:

mcloaked wrote:
... presumably the updated file now differs from the version 22 that is in the official linux-firmware package?
I just checked, and yes it does.
Will probably try it out later... Don't have time today.
EDIT:
Did you try the new v22?
I just tried it, but my impression is that there is no improvement regarding our issue.

Not yet - but it does sound like there are bugs that need to be addressed in whichever version is coming beyond v27. Perhaps the developers need more diagnostics/log files etc but I don't know what the best way to get at logs that will be useful is?

rgzfxf6bhu · 2017-01-24 12:29:47

rgzfxf6bhu wrote:

I just tried it, but my impression is that there is no improvement regarding our issue.

Now I have tried it for some days and it doesn't get better with the new ucode -22.ucode version 391740.0:

[    1.635312] iwlwifi 0000:3a:00.0: loaded firmware version 22.391740.0 op_mode iwlmvm

mcloaked wrote:

... sound like there are bugs that need to be addressed in whichever version is coming beyond v27. Perhaps the developers need more diagnostics/log files etc but I don't know what the best way to get at logs that will be useful is?

Here you can find what you need for the logs:
https://wireless.wiki.kernel.org/en/use … /debugging

Oh and now they have added the -27.ucode already here:
https://wireless.wiki.kernel.org/en/use … re_release

Does anyone know how to test the new -27.ucode, i.e. how do deal with following?

[    1.590168] iwlwifi 0000:3a:00.0: capa flags index 3 larger than supported by driver
[    1.590182] iwlwifi 0000:3a:00.0: Driver unable to support your firmware API. Driver supports v24, firmware is v27.

rgzfxf6bhu · 2017-01-24 12:30:43

I was also wondering if it might be a hardware problems, similar to this one:
http://askubuntu.com/a/647575

Last edited by rgzfxf6bhu (2017-01-24 12:32:04)

javamarket · 2017-02-11 19:07:51

I've found that the AC8260c driver has issues with power management and that by disabling same, wifi experience is much improved.
My approach:

Ensure WLAN_IFACE is properly assigned for your interface and using whatever network management necessary for a hook (I use NetworkManager)

/etc/NetworkManager/dispatcher.d/wifi-powerman-off.sh

#!/usr/bin/env sh

IFACE=$1 STATUS=$2
IWCONFIG=/usr/bin/iwconfig
WLAN_IFACE=wlp1s0

if [ ! -x $IWCONFIG ]; then
    exit 1
fi

if [ "$IFACE" = $WLAN_IFACE ] && [ "$STATUS" = "up" ]; then
    echo "[SCRIPT] Disabling wifi power mgmt"
    $IWCONFIG $IFACE power off
fi

mcloaked · 2017-02-11 20:42:53

javamarket: thanks. I will try that on my system when I get a chance to tinker with it. I do note that iwconfig was deprecated in favour of iw some time ago but your shell script can be adapted for iw instead, since the command "iw --help" yields the switch for power saving:

        dev <devname> set power_save <on|off>
                Set power save state to on or off.

and also you can query the power_save status with:

        dev <devname> get power_save <param>
                Retrieve power save state.

i.e. if the interface is wlp1s0 then switching off power_save would need iw dev wlp1s0 set power_save off which is I presume the equivalent of switching off power management in iwconfig.

Edit: I will make a file /etc/NetworkManager/dispatcher.d/wifi-powersave-off.sh with:

#!/bin/bash
## This needs to be placed in /etc/NetworkManager/dispatcher.d/ and will need to be chmod 700
## and chown root:root

IFACE=$1 STATUS=$2
IW=/usr/bin/iw
WLAN_IFACE=wlp1s0

if [ ! -x $IW ]; then
    exit 1
fi

if [ "$IFACE" = $WLAN_IFACE ] && [ "$STATUS" = "up" ]; then
    echo "[SCRIPT] Disabling wifi power_save"
    $IW dev $IFACE set power_save off
fi

Hopefully that will achieve the same as in your script.

Last edited by mcloaked (2017-02-11 21:15:03)

javamarket · 2017-02-11 22:25:50

Appreciate the shared update @mcloaked
I hope it solves your issue(s)

I've tested and can confirm the update indeed performs the same action.

One way you may want to test the impact is with MTR to your AP (ordinarily your router in residential).
I could see extreme packet loss prior to disabling this setting.

Example with IP of Router / AP in my home:

 sudo mtr --gtk 192.168.1.1

Last edited by javamarket (2017-02-11 22:37:35)

mcloaked · 2017-02-12 08:47:50

Thanks javamarket, I will be away from the machine concerned till Tuesday, but once I get back I will test this next week and let you know if this resolves the issue for me also.

Edit: I have finally been able to get to the machine and test - with the additional disabling of power_save for the wireless card the wireless connection is still not stable - and disconnects periodically, particularly when rsyncing large files over the network (local) - the connection is happier with small files but is still not ideal even then. So further work is needed on this issue at least for my machine.

Last edited by mcloaked (2017-02-20 20:56:38)

mcloaked · 2017-02-20 21:17:40

With power_save on but trying the suggestion to make 11n_disable=8 does give an improvement in the maximum speed of around 50% but the connection breaks as frequently as before. In the logs I see for the last disconnect and reconnect:

Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Queue 2 stuck for 10000 ms.
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Current SW read_ptr 40 write_ptr 104
Feb 20 21:12:25 home2 kernel: iwl data: 00000000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: FH TRBs(0) = 0x8915fa87
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: FH TRBs(1) = 0xc0110007
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: FH TRBs(2) = 0xeda9b79d
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: FH TRBs(3) = 0x8030000b
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: FH TRBs(4) = 0x27cdb943
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: FH TRBs(5) = 0x417d9080
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: FH TRBs(6) = 0x44b7d97c
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: FH TRBs(7) = 0x007090d7
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 0 is active and mapped to fifo 3 ra_tid 0x0000 [12,12]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 1 is active and mapped to fifo 2 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 2 is active and mapped to fifo 1 ra_tid 0x0000 [40,104]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 3 is active and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 4 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 5 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 6 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 7 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 8 is active and mapped to fifo 3 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 9 is active and mapped to fifo 7 ra_tid 0x0000 [216,216]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 10 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 11 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 12 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 13 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 14 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 15 is active and mapped to fifo 5 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 16 is active and mapped to fifo 1 ra_tid 0x00ff [248,153]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 17 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 18 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 19 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 20 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 21 is inactive and mapped to fifo 0 ra_tid 0xff00 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 22 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 23 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 24 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 25 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 26 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 27 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 28 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 29 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Q 30 is inactive and mapped to fifo 0 ra_tid 0x0000 [0,0]
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Microcode SW error detected.  Restarting 0x2000000.
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: CSR values:
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: (2nd byte of CSR_INT_COALESCING is CSR_INT_PERIODIC_REG)
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:        CSR_HW_IF_CONFIG_REG: 0X00c89008
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:          CSR_INT_COALESCING: 0X00000040
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:                     CSR_INT: 0X00000000
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:                CSR_INT_MASK: 0X00000000
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:           CSR_FH_INT_STATUS: 0X00000000
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:                 CSR_GPIO_IN: 0X00000019
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:                   CSR_RESET: 0X00000000
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:                CSR_GP_CNTRL: 0X08040005
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:                  CSR_HW_REV: 0X00000201
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:              CSR_EEPROM_REG: 0Xd55555d5
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:               CSR_EEPROM_GP: 0Xd55555d5
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:              CSR_OTP_GP_REG: 0Xd55555d5
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:                 CSR_GIO_REG: 0X001f0042
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:            CSR_GP_UCODE_REG: 0X00000000
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:           CSR_GP_DRIVER_REG: 0X00000000
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:           CSR_UCODE_DRV_GP1: 0X00000000
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:           CSR_UCODE_DRV_GP2: 0X00000000
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:                 CSR_LED_REG: 0X00000060
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:        CSR_DRAM_INT_TBL_REG: 0X88455b8c
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:        CSR_GIO_CHICKEN_BITS: 0X07800200
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:             CSR_ANA_PLL_CFG: 0Xd55555d5
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:      CSR_MONITOR_STATUS_REG: 0Xc03803c0
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:           CSR_HW_REV_WA_REG: 0X0001001a
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:        CSR_DBG_HPET_MEM_REG: 0Xffff0010
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: FH register values:
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:         FH_RSCSR_CHNL0_STTS_WPTR_REG: 0X45ac0e00
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:        FH_RSCSR_CHNL0_RBDCB_BASE_REG: 0X045ac0d0
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:                  FH_RSCSR_CHNL0_WPTR: 0X000000e8
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:         FH_MEM_RCSR_CHNL0_CONFIG_REG: 0X00801054
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:          FH_MEM_RSSR_SHARED_CTRL_REG: 0X000000fc
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:            FH_MEM_RSSR_RX_STATUS_REG: 0X03030000
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:    FH_MEM_RSSR_RX_ENABLE_ERR_IRQ2DRV: 0X00000000
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:                FH_TSSR_TX_STATUS_REG: 0X07ff0001
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0:                 FH_TSSR_TX_ERROR_REG: 0X00000000
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Start IWL Error Log Dump:
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Status: 0x00000000, count: 6
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Loaded firmware version: 22.361476.0
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000084 | NMI_INTERRUPT_UNKNOWN       
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x000002F0 | trm_hw_status0
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000000 | trm_hw_status1
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x0000E9C8 | branchlink2
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x0002843C | interruptlink1
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000180 | interruptlink2
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000000 | data1
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000080 | data2
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x07830000 | data3
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x1E001F39 | beacon time
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x6CD140C9 | tsf low
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x0000006B | tsf hi
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000000 | time gp1
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x073DA50A | time gp2
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000000 | uCode revision type
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000016 | uCode version major
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00058404 | uCode version minor
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000201 | hw version
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00C89008 | board version
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x10F8001C | hcmd
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00022002 | isr0
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00800000 | isr1
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x0800180A | isr2
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x004174C0 | isr3
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000000 | isr4
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000110 | last cmd Id
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000000 | wait_event
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000934 | l2p_control
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000020 | l2p_duration
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000003 | l2p_mhvalid
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000010 | l2p_addr_match
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x0000000D | lmpm_pmg_sel
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x03071928 | timestamp
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x0034E8F8 | flow_handler
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Start IWL Error Log Dump:
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: Status: 0x00000000, count: 7
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000070 | ADVANCED_SYSASSERT
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000000 | umac branchlink1
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0xC008383C | umac branchlink2
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0xC008166C | umac interruptlink1
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0xC008166C | umac interruptlink2
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000800 | umac data1
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0xC008166C | umac data2
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0xDEADBEEF | umac data3
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000016 | umac major
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00058404 | umac minor
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0xC0886280 | frame pointer
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0xC0886280 | stack pointer
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x09D7014E | last host cmd
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: 0x00000000 | isr status reg
Feb 20 21:12:25 home2 kernel: ieee80211 phy0: Hardware restart was requested
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: L1 Enabled - LTR Enabled
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: L1 Enabled - LTR Enabled
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: L1 Enabled - LTR Enabled
Feb 20 21:12:25 home2 kernel: iwlwifi 0000:01:00.0: L1 Enabled - LTR Enabled

Further investigation is needed. This seems similar to the thread at http://www.spinics.net/lists/linux-wire … 48741.html

I found a relevant kernel bug which I have added a comment and log to at https://bugzilla.kernel.org/show_bug.cgi?id=187461 and there is another kernel bugzilla report that may be relevant at https://bugzilla.kernel.org/show_bug.cgi?id=172431

Last edited by mcloaked (2017-02-20 22:03:07)

mcloaked · 2017-02-21 08:21:22

I have opened a new bug report upstream at https://bugzilla.kernel.org/show_bug.cgi?id=194643 following comments from one of the kernel wireless developers.

Edit: with the release of linux-firmware-20170217.12987ca-1 along with kernel 4.9.11-1 today I will need to see if this update set improves wireless performance, and in particular note that the new linux firmware now has /usr/lib/firmware/iwlwifi-8000C-27.ucode where the previous kernels were at version .22

Last edited by mcloaked (2017-02-21 19:59:27)

mcloaked · 2017-02-23 15:46:22

The main wireless developer replied to my bug report referenced in the previous comment, and asked me to try using the driver from the backports tree to get at the debugging diagnostics - I have tried to build the iwlwifi driver from git backports tree but it fails to build - I am not experienced with builds so I wonder if anyone else who does this kind of thing regularly might know how to get the build to work - i.e. are there different parameter sets in the make command that will get this to work or is it necessary to wait for the source code to be changed so that the build works? Any help needed with this is appreciated.

Last edited by mcloaked (2017-02-23 15:46:37)

Arch Linux

#1 2017-01-08 22:18:59

[SOLVED] slow intermittent wireless on AC-8260

#2 2017-01-09 22:56:54

Re: [SOLVED] slow intermittent wireless on AC-8260

#3 2017-01-10 16:48:02

Re: [SOLVED] slow intermittent wireless on AC-8260

#4 2017-01-13 04:50:53

Re: [SOLVED] slow intermittent wireless on AC-8260

#5 2017-01-13 09:41:23

Re: [SOLVED] slow intermittent wireless on AC-8260

#6 2017-01-13 14:23:38

Re: [SOLVED] slow intermittent wireless on AC-8260

#7 2017-01-13 15:12:08

Re: [SOLVED] slow intermittent wireless on AC-8260

#8 2017-01-13 15:30:37

Re: [SOLVED] slow intermittent wireless on AC-8260

#9 2017-01-13 22:00:43

Re: [SOLVED] slow intermittent wireless on AC-8260

#10 2017-01-14 08:26:13

Re: [SOLVED] slow intermittent wireless on AC-8260

#11 2017-01-14 13:50:37

Re: [SOLVED] slow intermittent wireless on AC-8260

#12 2017-01-14 14:12:36

Re: [SOLVED] slow intermittent wireless on AC-8260

#13 2017-01-14 15:05:56

Re: [SOLVED] slow intermittent wireless on AC-8260

#14 2017-01-14 15:57:17

Re: [SOLVED] slow intermittent wireless on AC-8260

#15 2017-01-14 16:39:10

Re: [SOLVED] slow intermittent wireless on AC-8260

#16 2017-01-16 16:32:36

Re: [SOLVED] slow intermittent wireless on AC-8260

#17 2017-01-24 12:29:47

Re: [SOLVED] slow intermittent wireless on AC-8260

#18 2017-01-24 12:30:43

Re: [SOLVED] slow intermittent wireless on AC-8260

#19 2017-02-11 19:07:51

Re: [SOLVED] slow intermittent wireless on AC-8260

#20 2017-02-11 20:42:53

Re: [SOLVED] slow intermittent wireless on AC-8260

#21 2017-02-11 22:25:50

Re: [SOLVED] slow intermittent wireless on AC-8260

#22 2017-02-12 08:47:50

Re: [SOLVED] slow intermittent wireless on AC-8260

#23 2017-02-20 21:17:40

Re: [SOLVED] slow intermittent wireless on AC-8260

#24 2017-02-21 08:21:22

Re: [SOLVED] slow intermittent wireless on AC-8260

#25 2017-02-23 15:46:22

Re: [SOLVED] slow intermittent wireless on AC-8260

Board footer