No wireless support for Realtek RTL8192EE [10ec:818b]

[Adam Lee]

Open this bug to trace device driver RTL8192EE.

We have seen this wireless card on some laptops, id is [10ec:818b], driver is RTL8192EE.

This bug is only for the initial support of this WiFi adapter, it's not working 100% well yet, please report new "not working bug" or contact upstream but not keep posting comments, thanks.

Status:

pull-request for trusty was NAKed because next minor Trusty release will use Utopic LTS kernel, which had already fixed this issue.

please wait it or upgrade your whole distro to Utopic.

[Adam Lee (adam8157)]

Set to incomplete, waiting the driver shows up in mainline.

[Adam Lee (adam8157)]

still not in wireless-next by now.

[Sascha Grossenbacher (berdir)]

I have this card in a new Thinkpad T440s.

[Elliott Conant (elliottcon)]

Similarly, I have this in a Thinkpad T540p.

[Yvon Xiao (yvonxiao)]

I also have this card in Thinkpad T440p

[William Oprandi (woprandi)]

Same for T440s

[K Xu (kx-anticss)]

Also have this problem.

[Naohiro hayashi (n-hayashi1134)]

I have this problem in Thinkpad T440p

[caobenzhi (caobenzhi0915)]

I have the same problem in Thinkpad T440.

[German (gergilar)]

I have the same problem in Thinkpad L440

[Xprogrammer (programmer-x)]

I've same problem with my t440s.

[K Xu (kx-anticss)]

Any thoughts on when this will be fixed?

[Adam Lee (adam8157)]

we are working on a solution now.

[Xprogrammer (programmer-x)]

After continous search for RTL8192EE driver, I found it here.

http://netbook-remix.archive.canonical.com/updates/pool/public/o/oem-wireless-rtl-92ce-92se-92de-8723ae-88ee-8723be-92ee-dkms/

I'm able to compile and load rtl8192ee on my lenovo t440s laptop.

However, transmitt speed very slow. For couple of times I got too many Invalid misc: and practically the connection is useless.

After few reboots and it seems to be fine. I'm sending this note from t440s loaded with 8192ee internet connection. Main drawback right now, I'm able to get only at max 60kB/s.

~$ iwconfig
wlan1 IEEE 802.11bgn ESSID:"myap"
          Mode:Managed Frequency:2.462 GHz Access Point: 58:6D:8F:64:22:FF
          Bit Rate=144.4 Mb/s Tx-Power=20 dBm
          Retry long limit:7 RTS thr=2347 B Fragment thr:off
          Power Management:off
          Link Quality=70/70 Signal level=-34 dBm
          Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
          Tx excessive retries:0 Invalid misc:247 Missed beacon:0

$ uptime
 23:28:08 up 15 min, 3 users, load average: 0.39, 0.61, 0.45

$ ifconfig wlan1
wlan1 Link encap:Ethernet HWaddr XXXXXXXXXX
          inet addr:192.168.1.92 Bcast:192.168.1.127 Mask:255.255.255.128
          inet6 addr: XXXXXXXXXXXX/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:35328 errors:0 dropped:0 overruns:0 frame:0
          TX packets:26022 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:49315226 (49.3 MB) TX bytes:3186066 (3.1 MB)

[Jody McIntyre (scjody)]

I was able to build and install the driver Xprogrammer found, but it does not work for me at all. It can get an address via DHCP but after that I can't even ping the default gateway.

There were several warnings produced during the build, lots of dmesg output, and a stack dump triggered in net/mac80211/rate.c:518 (a comment in rate.c indicates "RC is busted".

I can send along all of that if it's useful to anyone, but unless you have access to hardware documentation I doubt you'll be able to fix the driver. Based on the volume of the dmesg output, it appears to be an internal or alpha driver that escaped rather than something that's intended for production. I will contact the driver authors to see if they have a newer version (the version Xprogrammer found dates from October 22).

[Xprogrammer (programmer-x)]

http://www.spinics.net/lists/linux-wireless/msg118429.html

https://kernel.googlesource.com/pub/scm/linux/kernel/git/firmware/linux-firmware/+/4164c23d2e3b4863ca42d72f40dc7cc5d9fb8c6c

This commit is made in to firmware mainstream branch. Would that help ? However I don't see source code, it has put only the firmware files.

[William Oprandi (woprandi)]

https://git.kernel.org/cgit/linux/kernel/git/firmware/linux-firmware.git/commit/?id=4164c23d2e3b4863ca42d72f40dc7cc5d9fb8c6c

[Adam Lee (adam8157)]

Realtek hasn't officially released the driver in website or kernel mainline yet.

[Abe Handler (akh2103)]

I am also having trouble with this wireless card on 12.04

[wapingguo (wapingguo)]

My laptop is Lenovo ThinkPad T440p,I have the same problem on Ubuntu 12.04.
I found the device is 2x2 11b/g/n wireless M.2 adapter.

[Xiaojun Chen (xiaojunch)]

Yes, I have the same problem.

My laptop is T440s, and the wireless card is the same one, Realtek RTL8192EE [10ec:818b].

Look forward to solutions!

[[e-mail] (mingkedu)]

I've been waiting this for almost a month, really need the driver

[Siu On Chan (sochan)]

The problem may go away starting kernel 3.14. See comments #16 in
https://bbs.archlinux.org/viewtopic.php?id=174082

Haven't tested it myself, but I'm optimistic.

[William Oprandi (woprandi)]

Does someone has succeeded to build and run the staging driver ?

[Vipin (v-rathor)]

Guys,
Just sharing my experience with this issue.
I've got Lenovo Thinkpad L440 recently which has this realtek network device. Installed 12.04 LTS and all I could see was "no network device available". So no wired & wireless connection could be made. Then I read various solutions over web and then found this bug filed.
Later I tried installing 3.14-rc7 and rebooted. Now my wired connection (ethX) is up but still no wireless support. Now I see only one UNCLAIMED (earlier there were 2) h/w in 'lshw' output.
$ sudo lshw -C network -numeric
  *-network
       description: Ethernet interface
       product: Intel Corporation [8086:153B]
       vendor: Intel Corporation [8086]
       physical id: 19
       bus info: pci@0000:00:19.0
       logical name: eth9
       version: 05
       serial: 3c:97:0e:e8:0a:25
       size: 1Gbit/s
       capacity: 1Gbit/s
       width: 32 bits
       clock: 33MHz
       capabilities: pm msi bus_master cap_list ethernet physical tp 10bt 10bt-fd 100bt 100bt-fd 1000bt-fd autonegotiation
       configuration: autonegotiation=on broadcast=yes driver=e1000e driverversion=2.3.2-k duplex=full firmware=0.13-3 ip=192.168.1.126 latency=0 link=yes multicast=yes port=twisted pair speed=1Gbit/s
       resources: irq:42 memory:f2500000-f251ffff memory:f253f000-f253ffff ioport:6080(size=32)
  *-network UNCLAIMED
       description: Network controller
       product: Realtek Semiconductor Co., Ltd. [10EC:818B]
       vendor: Realtek Semiconductor Co., Ltd. [10EC]
       physical id: 0
       bus info: pci@0000:02:00.0
       version: 00
       width: 64 bits
       clock: 33MHz
       capabilities: pm msi pciexpress bus_master cap_list
       configuration: latency=0
       resources: ioport:5000(size=256) memory:f2400000-f2403fff

I just hope this might be of use to someone. Thanks.

[Adam Lee (adam8157)]

For the record, I analyzed the code, that RTL8821AE driver only supports 8812 and 8821, pci ids in headfile means nothing. RTL8192EE has not been upstreamed yet.

rtl8821ae/sw.c
420: {RTL_PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8812, rtl8821ae_hal_cfg)},
421: {RTL_PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8821, rtl8821ae_hal_cfg)},
426: {RTL_PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8812, rtl8821ae_hal_cfg)},
427: {RTL_PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8821, rtl8821ae_hal_cfg)},

[Bruce D'Arcus (bdarcus)]

So we're waiting on Realtek to release the driver, with no ETA?

Just got a Thinkpad T440s, and bummed to see no wireless.

[Mike McNally (emmecinque)]

Also built the oem-rtl-etc. driver on 13.10 for a T440p; module loads and looks like it connects to a hub, and *sometimes* some traffic makes it through, but it basically doesn't work.

[Josh Arenson (josharenson)]

Why is this part Ubuntu certified (http://www.ubuntu.com/certification/catalog/component/pci/10ec%3A818b/) if it doesn't work at all?

[Adam Lee (adam8157)]

@Josh, our preload image contains out-of-kernel-tree driver, which is provided by realtek.

[Vipin (v-rathor)]

@Adam, may I know where can I get the driver you are talking about? preload images?
Pardon my orientataion, can you please point me in right direction? Thanks.

[Bruce D'Arcus (bdarcus)]

FWIW, I'm using (x)ubuntu.
On Mar 31, 2014 1:35 PM, "Adam Lee" <email address hidden> wrote:

> @Josh, our preload image contains out-of-kernel-tree driver, which is
> provided by realtek.
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> Status in HWE Next Project:
> Incomplete
> Status in HWE Next trusty series:
> Confirmed
> Status in “linux” package in Ubuntu:
> Confirmed
>
> Bug description:
> Open this bug to trace device driver RTL8192EE.
>
> We have seen this wireless card on some laptops, id is [10ec:818b],
> driver is RTL8192EE.
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions
>

[Sharon Talbot (sharon-talbot)]

As a point of policy, it may not make sense to utilize images that are not available to the general public, or that are older than the most recent LTE release.

[Adam Lee (adam8157)]

This is the driver package we used in OEM preload image.

Please notice, the source is from RealTek, canonical has no ability to provide support(debugging or fixing). If you are not using OEM preload image and this driver is not working, please contact RealTek. Thanks.

[Bruce D'Arcus (bdarcus)]

@Adam - thanks. I get an error when trying to install the package, so
am contacting Realtek.

[Adam Lee (adam8157)]

This is the original driver source tarball from RealTek. You can try this one if there are some issues while installing that deb package.

Same situation, please notice that Ubuntu couldn't do anything with this tarball's issues before it's upstreamed by RealTek. Thanks.

[Kenneth Martin (marjin)]

Trying Adam Lee 8157

Building only for 3.11.0-18-generic
Building for architecture x86_64
Building initial module for 3.11.0-18-generic
Error! The dkms.conf for this module includes a BUILD_EXCLUSIVE directive which
does not match this kernel/arch. This indicates that it should not be built.

[K Xu (kx-anticss)]

Is there any word on this driver? T440s here. I'd really appreciate it.

[gisboxwinner (gisboxwinner)]

According to the tar writting at #40 by Adam Lee (adam8157) , i have try and find it ok.

My basic info:
    x240
    ubuntu-12.04-LTS
    3.11.0-19-generic
The following is my steps:
    $ wget https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1239578/+attachment/4057550/+files/rtl_92ce_92se_92de_8723ae_88ee_8723be_92ee_linux_mac80211_0017.1016v2.2013.tar.gz
    $ tar zxf rtl_92ce_92se_92de_8723ae_88ee_8723be_92ee_linux_mac80211_0017.1016v2.2013.tar.gz
    $ cd rtl_92ce_92se_92de_8723ae_88ee_8723be_92ee_linux_mac80211_0017.1016v2.2013
    $ sudo make && make install
    $ mobprobe rtl8192ee

[Xiaojun Chen (xiaojunch)]

Thanks Adam for your help!

Has anyone tried to adapt this tar file on the newest Ubuntu 14.04, which is 3.13 kernal? I tried but failed with error when "make install".

[hyra (hying-t)]

According to #40 and #44 i installed the driver. My card was detected and the wirless-networks are listed. But i'm not able to receive any data or ping my gateway.

any ideas?

Info: T440s with ubuntu 14.04

#lspci
03:00.0 Network controller [0280]: Realtek Semiconductor Co., Ltd. RTL8192EE PCIe Wireless Network Adapter [10ec:818b]
 Subsystem: Realtek Semiconductor Co., Ltd. Device [10ec:001b]

# iwconfig
wlan0 IEEE 802.11bgn ESSID:"my-essid"
          Mode:Managed Frequency:2.442 GHz Access Point: xxxxxx
          Bit Rate=1 Mb/s Tx-Power=20 dBm
          Retry long limit:7 RTS thr=2347 B Fragment thr:off
          Encryption key:off
          Power Management:off
          Link Quality=64/70 Signal level=-46 dBm
          Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
          Tx excessive retries:0 Invalid misc:1 Missed beacon:0

-> look the Bit Rate: 1 Mb/s

[Xiaojun Chen (xiaojunch)]

Update: the install works now after downloading #40.

Same issue with #46, the wireless is connected, yet no data is received. Also Ubuntu 14.04 on T440s.

[Jade Shan (evokeralucard)]

I have the same problem like #46. My computer is Thinkpad T440p

I make and install from:

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1239578/+attachment/4057550/+files/rtl_92ce_92se_92de_8723ae_88ee_8723be_92ee_linux_mac80211_0017.1016v2.2013.tar.gz

lspci -nn shows :

04:00.0 Network controller [0280]: Realtek Semiconductor Co., Ltd. Device [10ec:818b]

nw-manager founds the wifi SID and can connect the wireless network, but no data receive and cannot ping gateway

[Bruce D'Arcus (bdarcus)]

I can now get it installed per #40 on my T440s, and it works ... sort
of; connection is just highly unreliable/intermittent.

[Strsljen (mario-pranjic)]

Hi!

Having the same issue with Lenovo L540 and RTL8192EE wlan adapter.

Any estimations when this bug will be corrected?

[shptg (shptg)]

as per # 46, rtl8192ee wireless still can't work as usual. Model:Thinkpad X244 Ubutu 14.04.
looking forward a solution.

[Siu On Chan (sochan)]

Larry Finger (a Linux kernel contributor) has received code for the RTL8192EE driver, which is to be included in kernel 3.16.
http://marc.info/?l=linux-wireless&m=139648497105184
Hopefully this piece of information is more reliable than my earlier one in #23 :)

[Hua Zhang (zhhuabj)]

My computer is Thinkpad T440p
1, I have the same problem as #46 said.
2, and also re-complie kernel 3.14 according to #52 (that patch has been in kernel 3.4 by checking the file), but can' work

[Frances Hocutt (frances-hocutt)]

Thinkpad T440s. Installed as in #48, which got me to the same place as #46. When I tried the commands in 44 make install failed until I ran it with sudo, but then make install and modprobe worked fine--except I can connect to the network but not download data.

Off to learn how to patch and compile a kernel, I guess. I'll be tracking this bug.

[vorcigernix (vorcigernix)]

Installed module from deb (#38), stuck in same situation as #46.

[Frances Hocutt (frances-hocutt)]

Fix is available now: http://www.spinics.net/lists/linux-wireless/msg122327.html

Kernel binary is available for download here: http://www.lwfinger.com/realtek_drivers/linux-image-3.13.9_1.0_amd64.deb

[Matt Ramir (matthew-g)]

Please note the above fix requires kernel 3.13.9 afaik available at http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.13.9-trusty/

[Raymond Liu (raymondbl)]

The fix is a modified kernel itself; there seems to be no need to separately install the kernel mentioned by Matt Ramir, as my wireless card is now fixed just by installing the kernel binary mentioned by Frances Hocutt. Thanks for the fix!

[Josh Arenson (josharenson)]

Can confirm that the kernel in #56 works! Thanks.

[Riyas (kp-riyas4u)]

#56 is working for lenova L440. Thanks

[Josh Arenson (josharenson)]

Disregard my comment in #59. The card is now recognized, but drops connection frequent and seems to have a lot of trouble with certain authentication methods. I will try and get some debug information for you later this weekend, but for now, its back to my wi-fi dongle circa 2002 :-/

[Raymond Liu (raymondbl)]

Josh, what model of computer do you have? Similarily to Riyas, I have the Thinkpad L440, and the card has been working well for me for two days.

[Josh Arenson (josharenson)]

I have a Thinkpad T440s and lshw reports that my nic is
product: RTL8192EE PCIe Wireless Network Adapter

As stated, I haven't tried on another network yet, but I'll keep trying. Its nice to hear that you are having positive results.

[William Oprandi (woprandi)]

I installed the kernel from #56.
The driver works correctly but sometimes, the flow rate seems to be reduced to 0 for several seconds.
Another problem, I'm unable to reconnect to network after waking from standby and I am forced to reboot.

[Josh Arenson (josharenson)]

I have tried on another network, and confirm very similar issues to #64.
Here is something of possible interest from dmesg:
http://pastebin.ubuntu.com/7489463/

[Larry Finger (larry-finger-lwfinger)]

Realtek sent me a patch to fix the "skb->end - skb->tail > len" problems. A version with this fix is at

http://www.lwfinger.com/realtek_drivers/linux-headers-3.13.9_1.1_amd64.deb
http://www.lwfinger.com/realtek_drivers/linux-image-3.13.9_1.1_amd64.deb

I do not know if this fix will also handle the problems of #65. Please test as I have never seen those messages.

[Josh Arenson (josharenson)]

@larry-finger-lwfinger the kernel in #66 seems to be working!

I've maintained a connection for over 20 minutes (much longer than before), and speeds are very good.
I will continue to use this kernel for a few days and report back. I'm maintaining cautions optimism...

[Larry Finger (larry-finger-lwfinger)]

Good to hear. That version was submitted to the staging tree of the kernel. It should be a standard part of kernel 3.16.

[William Oprandi (woprandi)]

I don't agree with #67, I still have same problems that I mentionned in #64 (with updated kernel 3.13.9 1.1 in #66)

[Bruce D'Arcus (bdarcus)]

The kernel and driver generally works for me, but like #69, I
sometimes have connection problems.

On Thu, May 22, 2014 at 1:05 PM, William Oprandi
<email address hidden> wrote:
> I don't agree with #67, I still have same problems that I mentionned in
> #64 (with updated kernel 3.13.9 1.1 in #66)
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> Status in HWE Next Project:
> Incomplete
> Status in HWE Next trusty series:
> Confirmed
> Status in “linux” package in Ubuntu:
> Confirmed
>
> Bug description:
> Open this bug to trace device driver RTL8192EE.
>
> We have seen this wireless card on some laptops, id is [10ec:818b],
> driver is RTL8192EE.
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions

[Larry Finger (larry-finger-lwfinger)]

I do not know about the hibernate problem. My laptop has never gone into hibernation, and I have not had any desire to debug the reason for that.

Any other problems that occur should be reported as follows:

sudo modprobe -rv r8192ee
sudo modprobe -v r8192ee debug=3

Post the dmesg output corresponding to the time when the problem occurs.

For what its worth, my system has been up for almost 27 hours. I have had a few disconnect/reconnect events, but the connection recovered without any problems.

[William Oprandi (woprandi)]

It's not a hibernate, but just a suspend (when i just fold down the screen).
This is a extract of dmesg when system is trying to reconnect me.

[Siu On Chan (sochan)]

I have encountered similar issues as #69, #70, #72. For now I'll focus on the frequent disconnection problem and not the resume-from-suspend problem.

The new driver from #66 (as well as the old one in #48) works well 2/3 of the time, but the remaining 1/3 of the time, the connection is slow, disconnection is very frequent, or I cannot connect to the wifi network at all.

I've attached the portion of dmesg output (using the kernel from #66). It corresponds to running the commands in #71, and attempting but failing to connect to the wifi network. Hope it helps.

[Josh Arenson (josharenson)]

Sorry, been traveling. I can confirm the issues others are reporting (closing my screen breaks connection). In the meantime, I've loaded the debug module.

[Siu On Chan (sochan)]

Some symptoms:

My T440s laptop usually refuses to connect right after the first 5-6 cold boots
in the morning.

Sometimes the Network Manager applet says "connected", but once I try to
connect to any site (or just do a ping), the connection breaks immediately.

After a disconnect event, my home network often disappears from the list of
Wi-Fi networks.

To get connected after a failure, I have to reboot my laptop. Simply removing
the r8192ee module and reloading it does not help.

If I successfully connect, the network sometimes feel slow, and this usually
happens when I try to go to a new website. Loading another page from the same
website seems normal.

[Larry Finger (larry-finger-lwfinger)]

I too see the slow downs at times; however, I do not think it is due to changing website. I do not know why.

The driver sometimes takes several tries to make the initial connection. Again, the cause is unknown. Is it possible that your timeout counters are set too small and that is causing the total failure? My openSUSE system using NetworkManager with the KDE applet is always able to connect to my WPA2 network.

[Siu On Chan (sochan)]

It's possible that slaggish changing website is just my biasd impression and not really there.

Another symptom: Even after a sucessful connection to my wifi network, fairly often the instant messenages I sent on Facebook fail to send (possibly 1 out of 100), and the messages sometimes arrive out of order. It may be another indication of frequente packet drops with the current driver.

On the other hand, I don't think it's the timeout counter issue, because six months ago I had almost the same setup on a different laptop with a different wifi card/driver , and I never had this frequent disconnection issue. I was also using Network Manager on Ubuntu (13.10 back then), and the connection was perfect on my WPA2 network. Unless the default networking configuration has changed drastically in Ubuntu 14.04 (which I'm using), it's must be the driver issue. (And I have perfect connection on Windows with my T440s, so my new network card is not defective).

[Lee Zamparo (zamparo)]

I've installed and am now running the patched kernel provided by @lwfinger in #66, but still can't get the driver loaded successfully:

As requested, dmesg output is here: http://pastebin.com/download.php?i=8jLrHXgL.

This is a fresh install of Debian testing on a T440s. Since I seem to be alone in how my card is non-functional, are there other packages I'm missing?

Thanks in advance,

[Lee Zamparo (zamparo)]

Forget #78. I can now see other wireless networks (after installing the firmware .bin file served by @larry-finger-lfinger (same directory as the kernel image and header .deb files in #66), but I cannot connect to any networks.

dmesg output: http://pastebin.com/3XjbKXMd

Thanks for any help you can provide

[Siu On Chan (sochan)]

Some observations that may help others:

1. About a week ago it was really difficult for me to connect to my wifi network. The error was the infamous

"wlan0: deauthenticating from xx:xx:xx:xx:xx:xx by local choice (reason=3)".

It turns out that by updating "managed=true" in /etc/NetworkManager/NetworkManager.conf , I no longer see this error message. (thanks to http://ubuntuforums.org/showthread.php?t=2070648&page=2 ). Instead I sometimes see

"wlan0: authentication with xx:xx:xx:xx:xx:xx timed out".

This error message doesn't appear as often, so now I have a much better chance of establishing the initial connection (typically only 1-4 trials).

2. The dreadful "local choice (reason=3)" error will come back if I remove and then insert the r8192ee module as suggested in #71. My solution is to either reboot, or kill the wpa_supplicant and NetworkManager processes.

3. I have reached a state where I can reliably connect to the wifi network after a few trials, and the driver feels usable (albeit the package drop rate sometimes seems higher, especial). I'm content with the current driver in #66, and I really appreciate Larry Finger's effort.

On the other hand, I think the current driver is a bit unfriendly to new (T440s and other laptops) users, because the initial connection attempts require repeatedly and manually disabling and re-enabling Wi-Fi in the Network Manager applet. I look forward to seeing reliable initial connection that is automatic.

[Lee Zamparo (zamparo)]

Thanks for the NetworkManager tip in #80 @sochan, this was the issue I was dealing with. Now my connection seems more stable, though I haven't had a lot of time to test the reliability on this laptop yet. Thanks also to @lwfinger for his timely patched kernel, so that we don't have to wait until 3.16 is released.

[Lee Zamparo (zamparo)]

So far I've seen pretty good performance, though I did get a dropped connection after about an hour's use. Occasionally I'll have to reset either NetworkManager or wpa_supplicant, but it seems stable enough for novice users like myself. Hope that your patch gets folded into a kernel asap.

[imker (imker)]

Hello together,

I got my Lenovo T440s yesterday. I use Larrys Kernel from #66. I also have configured the Network Manager like shown in #80 (thank you sochan). And well, it works just fine fro me. I did not have any WLAN connection fail, and the package lost rate is OK.

So hopefully this patch finds the way in Ubuntu 14.10, because from the security point of view I would love to be on the Ubuntu distributed kernel. But a laptop without WLAN cant be used as a laptop, so it's not a laptop. I think most users (me too) will use the Larry Kernel even with know security issues.

Thanks Larry so far.
Imker

[Abdullah Khalid (abdullahkhalids)]

Any chance the driver can be built for i386? #66 unfortunately is only amd64.

I broke my dongle today and am tethered to a wire.

Thanks

[Nick Gerner (nick-gerner)]

just FYI, we installed the kernel from #66 and couldn't get wifi working. We could occasionally scan, but never connect. Debug logging from #71 resulted in an endless spew of log messages into syslog (or dmesg). Also, we got occasional system crashes.

we've since re-installed vanilla 14.04, waiting for a proper fix get backported.

[Michael Jeanson (mjeanson)]

For anyone interrested, I've made a dkms package from the sources in the linux-next tree, it works on my T440. You can grab it here https://launchpad.net/~mjeanson/+archive/ppa/+files/rtl8192ee-dkms_1.0-1~ppa1_all.deb

[Sylvain Corlay (sylvain-corlay)]

Hi Larry, would you recommend using your patched kernel of are the latest release candidate for the linux kernel 3.16 which has your patch merged?

[Larry Finger (larry-finger-lwfinger)]

I would recommend using kernel 3.16-rcX. I do not plan on updating the deb on my web site, and I do not plan on generating a 32-bit version.

[Sylvain Corlay (sylvain-corlay)]

Thanks, 3.16-rc2 works nicely for me.

[imker (imker)]

Hello all,

I removed Larrys Kernel and installed Micheals DKMS Pakage from #86. And it works as good as before on my T440s.

Thanks to Michael, for this nice package, and also to Larry for adding this patch to the main Kernel tree.
Good work Guys.

[Lee Zamparo (zamparo)]

Has anyone else had success having the driver work after suspending to RAM? I find that on both my home network and at another I tested that I have to unload & reload the driver to have a chance to connect.

[imker (imker)]

Hi Lee,

yes WLAN worked for me after suspend to RAM. But not always. I would say 3 of 4 times, WLAN worked for me with Larrs Kernel after Suspend to RAM. If I was not connected I could just switch off and on in the network mananger to get it work again.

Since I instelld the DKMS Packaged I did not realy use Suspend to RAM. But on a short test WLAN worked after wake up (a.k. open the display). But it was a short test with only 2 Minutes in Suspend.

I will tell you how it works in a view weeks again...
Imker

[Friedrich Kretschmer (friedrich-k)]

Worked for me under Debian testing amd64. Installed kernel (3.13.9) and headers from #66 and downloaded firmware from http://www.lwfinger.com/realtek_drivers/rtl8192eefw.bin and moved the file to /lib/firmware/rtlwifi/. I initially had problems connecting to networks but #80 solved the problem! I also tried the dkms module but it did not compile for the standard testing 3.14 kernel. Thanks everybody for posting all this info!

[Sylvain Corlay (sylvain-corlay)]

Hello, I now have issues with the firmware failing to load (using kernel image 3.16rc2):
dmesg output:
[ 2.290812] r8192ee: module is from the staging directory, the quality is unknown, you have been warned.
[ 2.321112] r8192ee 0000:03:00.0: Direct firmware load failed with error -2
[ 2.321117] r8192ee 0000:03:00.0: Falling back to user helper

[Sylvain Corlay (sylvain-corlay)]

The problem (#95) appeared after an update of Ubuntu. Is anyone else experiencing this?

[Larry Finger (larry-finger-lwfinger)]

The necessary firmware in in the linux-firmware git repo. You will need to clone git://git.kernel.org/pub/scm/linux/kernel/git/dwmw2/linux-firmware.git and copy rtlwifi/rtl8192eefw.bin into /lib/firmware/rtlwifi/rtl8192eefw.bin.

[Sylvain Corlay (sylvain-corlay)]

Hi Larry, it works very well. Thanks!

[Siu On Chan (sochan)]

Hi Larry,

Here's something weird about the driver in 3.16-rc3 (and the same for 3.16-rc2). I often see the following lines from dmesg:

[ 156.538888] cfg80211: Calling CRDA for country: 1B
[ 181.518110] AP off, try to reconnect now
[ 181.518142] wlan0: Connection to AP xx:xx:xx:xx:xx:xx lost
[ 181.574190] cfg80211: Calling CRDA to update world regulatory domain
[ 181.576554] cfg80211: World regulatory domain updated:
[ 181.576557] cfg80211: DFS Master region: unset

The country code is the number "1" followed by the letter "B". This always happens with the 3.16-rc3 kernel (from Ubuntu kernel PPA), but it never shows up in the kernel from #66 (I've tested a few times). Even after I run "sudo iw reg set US" on the 3.16-rc3 kernel, I may see the following:

[ 181.576581] cfg80211: Calling CRDA for country: US
[ 181.578828] cfg80211: Regulatory domain changed to country: US
[ 181.578831] cfg80211: DFS Master region: unset

That is, a strange message says that the DFS Master region is unset. This message never appears with the #66 kernel.

Not sure whether this error is related to the new r8192ee driver or other parts of the 3.16-rc3 kernel. I don't think this error is fatal, but I'm pointing it out in case it uncovers an unknown bug.

BTW the new driver is very good, apart from the initial connection issue.

[Colin Heinzmann (cheinzmann3)]

Trying to get up and and running with #86 DKMS module and am running into a compilation issue. Below is my make.log. I tried building in kernel 3.16-rc3 and the same error occurred. Could someone shed light on this issue?

DKMS make.log for rtl8192ee-1.0 for kernel 3.15.2-1-ARCH (x86_64)
Tue Jul 1 00:47:24 PDT 2014
make: Entering directory '/usr/lib/modules/3.15.2-1-ARCH/build'
  LD /var/lib/dkms/rtl8192ee/1.0/build/built-in.o
  CC [M] /var/lib/dkms/rtl8192ee/1.0/build/base.o
  CC [M] /var/lib/dkms/rtl8192ee/1.0/build/cam.o
  CC [M] /var/lib/dkms/rtl8192ee/1.0/build/core.o
/var/lib/dkms/rtl8192ee/1.0/build/core.c:1599:2: warning: initialization from incompatible pointer type
  .flush = rtl_op_flush,
  ^
/var/lib/dkms/rtl8192ee/1.0/build/core.c:1599:2: warning: (near initialization for 'rtl92e_ops.flush')
  CC [M] /var/lib/dkms/rtl8192ee/1.0/build/debug.o
  CC [M] /var/lib/dkms/rtl8192ee/1.0/build/efuse.o
  CC [M] /var/lib/dkms/rtl8192ee/1.0/build/pci.o
  CC [M] /var/lib/dkms/rtl8192ee/1.0/build/ps.o
  CC [M] /var/lib/dkms/rtl8192ee/1.0/build/rc.o
  CC [M] /var/lib/dkms/rtl8192ee/1.0/build/regd.o
/var/lib/dkms/rtl8192ee/1.0/build/regd.c: In function '_rtl_reg_apply_beaconing_flags':
/var/lib/dkms/rtl8192ee/1.0/build/regd.c:185:20: error: 'IEEE80211_CHAN_NO_IBSS' undeclared (first use in this function)
      ch->flags &= ~IEEE80211_CHAN_NO_IBSS;
                    ^
/var/lib/dkms/rtl8192ee/1.0/build/regd.c:185:20: note: each undeclared identifier is reported only once for each function it appears in
/var/lib/dkms/rtl8192ee/1.0/build/regd.c:189:11: error: 'IEEE80211_CHAN_PASSIVE_SCAN' undeclared (first use in this function)
          ~IEEE80211_CHAN_PASSIVE_SCAN;
           ^
/var/lib/dkms/rtl8192ee/1.0/build/regd.c: In function '_rtl_reg_apply_active_scan_flags':
/var/lib/dkms/rtl8192ee/1.0/build/regd.c:218:19: error: 'IEEE80211_CHAN_PASSIVE_SCAN' undeclared (first use in this function)
   if (ch->flags & IEEE80211_CHAN_PASSIVE_SCAN)
                   ^
/var/lib/dkms/rtl8192ee/1.0/build/regd.c: In function '_rtl_reg_apply_radar_flags':
/var/lib/dkms/rtl8192ee/1.0/build/regd.c:283:8: error: 'IEEE80211_CHAN_NO_IBSS' undeclared (first use in this function)
        IEEE80211_CHAN_NO_IBSS |
        ^
/var/lib/dkms/rtl8192ee/1.0/build/regd.c:284:8: error: 'IEEE80211_CHAN_PASSIVE_SCAN' undeclared (first use in this function)
        IEEE80211_CHAN_PASSIVE_SCAN;
        ^
/var/lib/dkms/rtl8192ee/1.0/build/regd.c: In function '_rtl92e_regd_init_wiphy':
/var/lib/dkms/rtl8192ee/1.0/build/regd.c:373:18: error: 'WIPHY_FLAG_CUSTOM_REGULATORY' undeclared (first use in this function)
  wiphy->flags |= WIPHY_FLAG_CUSTOM_REGULATORY;
                  ^
/var/lib/dkms/rtl8192ee/1.0/build/regd.c:374:19: error: 'WIPHY_FLAG_STRICT_REGULATORY' undeclared (first use in this function)
  wiphy->flags &= ~WIPHY_FLAG_STRICT_REGULATORY;
                   ^
/var/lib/dkms/rtl8192ee/1.0/build/regd.c:375:19: error: 'WIPHY_FLAG_DISABLE_BEACON_HINTS' undeclared (first use in this function)
  wiphy->flags &= ~WIPHY_FLAG_DISABLE_BEACON_HINTS;
                   ^
scripts/Makefile.build:318: recipe for target '/var/lib/dkms/rtl8192ee/1.0/build/regd.o' failed
mak...

[Larry Finger (larry-finger-lwfinger)]

Comment #99: Those new messages are due to changes in cfg80211. As long as the driver works correctly, you may ignore them.

Comment #100: Those messages are due to trying to build old driver code on a kernel that is new. If you really want to build that code, you will need to research the changes in the API for kernels 3.15 and 3.16 and make the appropriate changes to the code. I added the driver to the kernel source so that you would not have to do these changes. Why are you still building an out-of-kernel version?

[Colin Heinzmann (cheinzmann3)]

Larry,

I'm trying to build the dkms because I'm not sure how to get the driver up and running (I'm kind of a linux driver noob). I was under the impression that installing 3.16-rcX would enable the driver but this is not the case with my build. How do I enable the wireless driver on my computer? Tried mod probe and it doesn't see the driver. I have that rtl8192eefw.bin in my lib folder which was installed there on the stock kernel. When I lspci, my kernel does report that it has identified the card but there is no network interface going along with it.

[Larry Finger (larry-finger-lwfinger)]

You can get the driver in 3.16 by using "make xconfig" or "make menuconfig", and enabling the driver. Once that is done, a new kernel built and installed, and the firmware is installed, the driver will automatically start upon bootup.

[Colin Heinzmann (cheinzmann3)]

Ahh my understanding of how kernel drivers is much clearer now. I previously installed 3.16-rc3 using the Arch Linux User repository and its prebuilt customization option didn't have my driver selected. A fully custom build worked. Thanks and much <3!

[Matt (hlmtre)]

Larry's debs from #66 worked for me. Seems to take a little longer to connect than stabler drivers, but works well, and speeds are good. No drops so far.

[Lee Zamparo (zamparo)]

Larry, I'm still getting cycles of association, authentication and de-association when trying to connect to any network (see this paste: )

The gist of details are here:

[126550.533856] cfg80211: Calling CRDA to update world regulatory domain
[126550.536822] cfg80211: World regulatory domain updated:
[126550.536825] cfg80211: DFS Master region: unset
[126550.536827] cfg80211: (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp), (dfs_cac_time)
[126550.536829] cfg80211: (2402000 KHz - 2472000 KHz @ 40000 KHz), (N/A, 2000 mBm), (N/A)
[126550.536830] cfg80211: (2457000 KHz - 2482000 KHz @ 40000 KHz), (N/A, 2000 mBm), (N/A)
[126550.536831] cfg80211: (2474000 KHz - 2494000 KHz @ 20000 KHz), (N/A, 2000 mBm), (N/A)
[126550.536833] cfg80211: (5170000 KHz - 5250000 KHz @ 80000 KHz), (N/A, 2000 mBm), (N/A)
[126550.536834] cfg80211: (5735000 KHz - 5835000 KHz @ 80000 KHz), (N/A, 2000 mBm), (N/A)
[126550.536835] cfg80211: (57240000 KHz - 63720000 KHz @ 2160000 KHz), (N/A, 0 mBm), (N/A)
[126562.807607] wlan0: authenticate with 00:1f:33:d2:a6:f8
[126562.843482] wlan0: send auth to 00:1f:33:d2:a6:f8 (try 1/3)
[126562.843897] wlan0: authenticated
[126562.844210] r8192ee 0000:03:00.0 wlan0: disabling HT/VHT due to WEP/TKIP use
[126562.846510] wlan0: associate with 00:1f:33:d2:a6:f8 (try 1/3)
[126562.950444] wlan0: associate with 00:1f:33:d2:a6:f8 (try 2/3)
[126563.054359] wlan0: associate with 00:1f:33:d2:a6:f8 (try 3/3)
[126563.082038] wlan0: RX AssocResp from 00:1f:33:d2:a6:f8 (capab=0x411 status=0 aid=1)
[126563.120764] wlan0: associated
[126573.115713] wlan0: disassociating from 00:1f:33:d2:a6:f8 by local choice (Reason: 3=DEAUTH_LEAVING)

This block repeats every time I try to connect. Is you think this is a configuration issue, I'd appreciate any ideas for resolving it. How is it that it's 2014 and connecting to wireless networks in Linux is still so frustrating?

[Lee Zamparo (zamparo)]

oops, pastebin link is here: http://pastebin.com/XaqL36Vx

[Lee Zamparo (zamparo)]

BTW, I'm running: $ uname -a
Linux bentobox 3.16.0-rc2 #1 SMP Tue Jun 24 22:43:14 EDT 2014 x86_64 GNU/Linux

with the firmware offered under @lwfinger's website.

[Siu On Chan (sochan)]

Lee #106: Have you tried the fix in #80?

I should say that once I switch to 3.16-rc3 (from Ubuntu kernel PPA), I no longer need this fix (and to be honest I don't understand how/why it worked in the past).

Now I still have initial connection issue (wlan0: authentication with xx;xx;xx;xx;xx;xx timed out), and this error can also repeat indefinitely. But the problem usually go away if I do the following:

sudo modprobe -r r8192ee
sudo modprobe r8192ee
sudo killall wpa_supplicant NetworkManager

[Lee Zamparo (zamparo)]

@sochan: yes, I did make that change, it helped initially. But now I'm experiencing the same message (though not the same cause).

I suspect the problem is between NetworkManager and wpa_supplicant, but I'm trying a few things suggested by others who have had this problem:

1) disabling ipv6 in the grub boot line : GRUB_CMDLINE_LINUX_DEFAULT="quiet ipv6.disable=1"

2) disabling 11_n in the r8192ee module configuration: sudo echo "options r8192ee 11n_disable=1" > /etc/modprobe.d/r1892ee.conf

[Larry Finger (larry-finger-lwfinger)]

Whatever is happening, your wireless system is dropping the connection as shown by

[126573.115713] wlan0: disassociating from 00:1f:33:d2:a6:f8 by local choice (Reason: 3=DEAUTH_LEAVING)

Please check your NetworkManager and wpa_supplicant logs to see if they offer any clues.

What is the encryption of your AP? I suspect it must be WPA1 given

[126562.844210] r8192ee 0000:03:00.0 wlan0: disabling HT/VHT due to WEP/TKIP use

[Dan Gillmor (dan-gillmor)]

Is there any ETA for this driver? I've had to buy a wifi dongle, which is annoying...

[Larry Finger (larry-finger-lwfinger)]

As stated above, it is in the staging directory for kernel 3.16. As I do not speak for Ubuntu, I have no idea when it will be available for you.

For some reason, this bug report jumps from comment #40 to #73. In #66, I posted a link for a 64-bit kernel that had this driver built in. For completeness, the link is http://www.lwfinger.com/realtek_drivers/linux-image-3.13.9_1.1_amd64.deb. The firmware is at http://www.lwfinger.com/realtek_drivers/rtl8192eefw.bin.

FYI, kernels implementing this driver are available for several other distros!

[chuxinxin (chuxinxin)]

use firmware http://www.lwfinger.com/realtek_drivers/rtl8192eefw.bin.
it Seems to work but do not connect!
i can see wifi but do not connect!

[Larry Finger (larry-finger-lwfinger)]

Please add some logs. In particular, what does the dmesg output have to say? What connection manager are you using? What encryption type? If WPA or WPA2, please check the wpa_supplicant logs.

Just saying "It does not work" is not sufficient, and is just wasting everyone's time!

[huyingqian (hdhyq)]

@Larry your files on #113 successfully work!!Thank you!
First i follow the #44 ,then my card was recognized by the system, but after i connected the wifi i found that the rate was extremely slow. Then i installed linux-image-3.13.9_1.0_amd64.deb linux-image-3.13.9_1.1_amd64.deb linux- headers-3.13.9_1.1_amd64.deb ,and download the rtl8192eefw.bin on #94 .Finally i restart my system and the card works well now.

[chuxinxin (chuxinxin)]

@larry

use dmesg show:Init MAC failed

use network-manager-gnome

[timarmstrong (tim-g-armstrong)]

I've been able to reproduce a similar problem to chuxinxin and collected logs.

Distro: Ubuntu 14.04 with Ubuntu mainline 3.16rc4
uname -a: Linux Cockatoo 3.16.0-031600rc4-generic #201407061635 SMP Sun Jul 6 20:36:26 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

Step-by-step procedure to reproduce:
1. Boot system with r8192ee module blacklisted
2. modprobe r8192ee
3. Connect to wireless network with NetworkManager
4. Connection works well for a while
5. Connection suddenly disconnects
6. The wireless network is no longer visible in NetworkManager, but others remain visible.
7. Kill NetworkManager and wpa_supplicant
8. NetworkManager no longer shows device
9. Attempt to modprobe -r r8192ee renders system unresponsive

The attached tarball include lsmod, dmesg and syslog. I split the logs into before/after killing NetworkManager & wpa_supplicant since there was a lot of spam in the logs after that event.

It may be relevant that this was done in an environment with a lot of radio interference and dropped packets.

[timarmstrong (tim-g-armstrong)]

One thing I forgot to mention: I've been able to reproduce the problem every time for several times with the procedure described.

[chuxinxin (chuxinxin)]

@timarmstrong
now i use kernel 3.16 ,it works well!

[chuxinxin (chuxinxin)]

@timarmstrong
sorry,it just lasts 50 minutes
now.......

[Lee Zamparo (zamparo)]

@lwfinger

Thanks for your work on this driver.
I find that the driver now is mostly functional, but I still experience random disconnections, and whenever reconnecting after standby I encounter cycles of scanning->authenticating->associating->disconnecting.

Jul 15 20:14:14 bentobox NetworkManager[2278]: <info> (wlan0): supplicant interface state: disconnected -> scanning
Jul 15 20:14:15 bentobox NetworkManager[2278]: <info> (wlan0): supplicant interface state: scanning -> authenticating
Jul 15 20:14:15 bentobox NetworkManager[2278]: <info> (wlan0): supplicant interface state: authenticating -> associating
Jul 15 20:14:15 bentobox NetworkManager[2278]: <info> (wlan0): supplicant interface state: associating -> disconnected
Jul 15 20:14:15 bentobox NetworkManager[2278]: <info> (wlan0): supplicant interface state: disconnected -> scanning
Jul 15 20:14:28 bentobox NetworkManager[2278]: <info> (wlan0): supplicant interface state: scanning -> authenticating
Jul 15 20:14:29 bentobox NetworkManager[2278]: <info> (wlan0): supplicant interface state: authenticating -> disconnected

(very much line the behavoiur in https://bugzilla.redhat.com/show_bug.cgi?id=759931#c6).

Have you looked into how the driver performs after waking from standby?
Does the driver behave properly WRT powersave?

When I look at modinfo r8192ee:

[snip]
parm: swlps:bool
parm: swenc:using hardware crypto (default 0 [hardware])
 (bool)
parm: ips:using no link power save (default 1 is open)
 (bool)
parm: fwlps:using linked fw control power save (default 1 is open)
 (bool)
[/snip]

Does this mean by default, the driver does not interact with link power save, nor linked firmware control power save?

Thanks,

Lee

[Larry Finger (larry-finger-lwfinger)]

Realtek has never been good at writing clear module parameter descriptions. My current working version has the following:

MODULE_PARM_DESC(swenc, "Set to 1 for software crypto (default 0)\n");
MODULE_PARM_DESC(ips, "Set to 0 to not use link power save (default 1)\n");
MODULE_PARM_DESC(swlps, "Set to 1 to use SW control power save (default 0)\n");
MODULE_PARM_DESC(fwlps, "Set to 1 to use FW control power save (default 1)\n");
MODULE_PARM_DESC(msi, "Set to 1 to use MSI interrupts mode (default 1)\n");
MODULE_PARM_DESC(debug, "Set debug level (0-5) (default 0)");

In other words, link power save is ON by default, firmware power control is also ON, and software power control is off.

My test computer refuses to hibernate or suspend, thus I have not tested "standby", whatever that means. In fact, in the days preceeding the closing of changes for 3.16, I was so busy trying to provide help for the Lenovo T440s users, and getting the code into the kernel that I barely had time for testing.

I am afraid that someone else will need to test and debug these problems.

[Damian Nadales (dnadales)]

Thanks a lot for all the work on the driver, and for getting the code into the kernel.

I am using the kernel 3.16.0-031600rc6-generic, and the wireless works but partially. It takes a long time to connect. And after resuming from suspend it does not connect anymore.

What is the best way in which I can provide useful debug information?

Thanks.

[Larry Finger (larry-finger-lwfinger)]

I have received a newer version from Realtek that I am working at getting into the kernel. At this point, I do not even know what the differences are. While this conversion is in progress, I have very little time to try to debug the current code.

I would say that the best thing to do is wait for the new code that I hope will be in 3.18.

[timarmstrong (tim-g-armstrong)]

Thanks for all the hard work. I'll keep an eye out for the new code in 3.18. If there's any way I can provide useful debug info at any point I'm happy to help.

[Chetan (chetan-saranmehra)]

I am using T440s and having problems like everybody else here. I have had a variation of every problem mentioned above withI have tried upgrading to 3.16 and that freezes my machine very often, so I removed it...I am using the dkms suggested by Jeanson

https://launchpad.net/~mjeanson/+archive/ppa/+files/rtl8192ee-dkms_1.0-1~ppa1_all.deb

I hope someone will fix this soon, else I will have to repalce this wireless card.

[Sam Dodrill (shadow-h511)]

Upgrading to Utopic's kernel fixes the issue, but every so often the wifi will drop out and then come back. What debug output should I be getting?

[Sam Dodrill (shadow-h511)]

update: Using these module parameters fixes it:

ips=0 fwlps=0 swenc=1

[[e-mail] (vworld4u-g)]

Hi

Can somebody (who got this driver for Realtek RTL8192EE working) post a summary of steps to follow to fix it (even temporarily)? I am not a deep level expert in ubuntu kernel building or changing the parameters of a module etc. Please help me fixing this issue.

I am seeing that even ethernet internet (wired connection) too drops intermittently on my L440 Laptop with Ubuntu 14.04 freshly installed (almost within 20mins of connecting). So I need to get this Wifi fixed atleast.

[Anuj (delicpsyche)]

Is Utopic compatible with Linux Mint? I will test it and let you know.
I will post the same on Mint forums as Mint users face the same issue.

[imker (imker)]

Hello vworld,

I think using the DKMS Package from Michael (Post #86) is the most easy way to get a vanilla Ubuntu work with this WLAN Adapter. To do so, you might want to use the following sequence in a terminal.

sudo apt-add-repository ppa:mjeanson/ppa
sudo apt-get update
sudo apt-get upgrade
sudo apt-get install rtl8192ee

You will add a new repository to your system. Update the system and then install the drive package. This worked fine for me on me on my Lenovo T440s

Good luck.
imker

[[e-mail] (vworld4u-g)]

Hi imker,

I just followed all the steps you mentioned, but the console log says it is unable to find the module rtl8192ee. Do you mean rtl8192ee-dkms ?

If I go for rtl8192ee-dkms I get the following log :
ERROR (dkms apport): kernel package linux-headers-3.16.0-031600rc6-lowlatency is not supported

I think I downloaded the latest debs from the URL above. I installed both llinux-headers-<version>-owlatency and normal linux-headers-<version> in my system. This is causing the issue.

Please let me know correct package and I will try to remove lowlatency and try out again.

[timarmstrong (tim-g-armstrong)]

rtl8192ee-dkms is the built name of the rtl8192ee package, so you're ok there: https://launchpad.net/~mjeanson/+archive/ubuntu/ppa/+sourcepub/4248780/+listing-archive-extra. You don't need the lowlatency headers unless you are running the special low latency kernel.

Also, there's no need to install mjeanson's package *and* the 3.16 kernel. The 3.16 kernel comes with the rtl8192ee driver already, so if you're running 3.16 then mjeanson's package is not relevant. I'm guessing that what you want is to remove the 3.16 kernel, reboot into the stable Ubuntu-provided kernel, then install mjeanson's pacakge for that kernel.

[imker (imker)]

Hi vworld,

tim is right. If you use the 3.16 kernel the rtl8192ee package may not works fine. So if you wan to try this package you should remove the 3.16 Kernel and reboot. So you will use the vanila 3.13 kernel again.

An other hint I can give you is to check if you got the firmeware file for this adapter: There should be a file /lib/firmware/rtl8192eefw.bin

If there is no such file, you can download it from http://www.lwfinger.com/realtek_drivers/rtl8192eefw.bin

Imker

[[e-mail] (vworld4u-g)]

Hi Imker/Tim,

Thanks a lot for the information. I will try it today (I am outside) and update you the success story (hopefully :) )

[Kelvin (iamkelvinxu)]

Hi Larry/everyone,

I am having the same problems that Chetan is having that my T440s ubuntu frequently freezes when using wireless.

I am using 12.04 and kernel 3.13. If anyone had a fix/any info on how to help that would really helpful.

Kelvin

[Juan (hyhh)]

Hi Larry/Michael/everyone,
I just bought a Lenovo T540p with the same Realtek 8192EE wifi adapter.
I tried both the 2.13.9 patched kernel with the firmware and the newest 2.16rc7 kernel.
Both have the same problem:
My wifi connects OK but it drops intermittently
When I try to ping "www.google.com" in the terminal about 7% pakets gets dropped
Just to make sure it's not the internet I ping the router address
192.168.1.1 and same problem. So it's definitely the wifi connection unstable.
I have a Cisco M10 router. It may be the router so I played around with the router settings a bit and changing it to "wireless N only" mode seems to help lower the timeout rate a bit but still high compare to the windows 7 results (around 1%)

Where can I find the debug logs and give you more information?

Also, how do I tweak the paramters? When I execute the modinfo command
ERROR: modinfo: could not find module rtl8192ee

Thanks in advance!

Juan

[timarmstrong (tim-g-armstrong)]

Juan/Kelvin,
 Based on Larry's earlier responses, I think we're going to have to wait until the 3.18 kernel for the new and improved driver. Realtek has apparently made a lot of fixes to its version of the driver, but porting it to the kernel is a lot of work on Larry's part.

It sounds like some people have had found the current version of the driver reasonably stable. That hasn't been my experience, but it's hard to know what variables are making the difference. Personally, I bought a cheap USB wifi adapter to use until the Linux driver for this chip is more stable - if you can find a USB adapter with good linux drivers, that may be the best option for you.

Juan: the kernel module is r8192ee, not rtl8192ee.

- Tim

[Sascha Grossenbacher (berdir)]

The driver is working fairly stable for me, I also have problems with suspend/wakeup but two things usually help me solve that:

* It is usually only a problem if I want to re-connect to the same WIFi. If I switch from one to another, it works without a problem in most cases. In case of the portable WiFi hotspot on my phone, it is even enough to just disable and enable it again. No chance to connect without that, works immediately after that. That might help to find the problem here?

* If that is not an option, it often helps to (repeatedly) kill the NetworkManager, then it usually works again.

[Juan (hyhh)]

I realized that I probably picked the wrong card when ordering my laptop.
I was offered either the Intel Wireless N 7260 or this Realtek card, same price
I am not so keen with this realtek card anyway because even in windows 7 my signal only shows 3 bars sometimes as my previous laptop it's 5 bar all-the-time
Instead of buying a usb dongle and wait for the driver I was considering to just buy the intel card for $20 and swap it in
Some of you guys may consider this option as the other card is certified by ubuntu and I assume it has the driver built-in? Can anyone with the T440/540 with the Intel card confirm that?

Thanks,
Juan

[Matt (hlmtre)]

As far as I know installing the card is a PITA. Otherwise I would have done
that already. But, if it isn't too taxing (taking the motherboard out or
taking apart the screen would be too taxing), I would love to know.

On Thu, Jul 31, 2014 at 11:08 AM, Juan <email address hidden> wrote:

> I realized that I probably picked the wrong card when ordering my laptop.
> I was offered either the Intel Wireless N 7260 or this Realtek card, same
> price
> I am not so keen with this realtek card anyway because even in windows 7
> my signal only shows 3 bars sometimes as my previous laptop it's 5 bar
> all-the-time
> Instead of buying a usb dongle and wait for the driver I was considering
> to just buy the intel card for $20 and swap it in
> Some of you guys may consider this option as the other card is certified
> by ubuntu and I assume it has the driver built-in? Can anyone with the
> T440/540 with the Intel card confirm that?
>
> Thanks,
> Juan
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions
>

[Larry Finger (larry-finger-lwfinger)]

If Lenovo does not have an access door on the bottom that provides access to the wireless card, I am doubly glad that I chose Toshiba over Lenovo when I recently purchased a new laptop. My decision was based on the bad reviews for the new keyboard and touchpad design.

[Juan (hyhh)]

I just took it apart and tried it. Easy. On my 540p there are 2 service panels at the bottom 1 for HDD and memory the other smaller one is the wifi card. I already swapped the HDD with SSD by taking apart the larger one. Just make sure you get the Intel 7260 NGW form factor for the half-mini size pci slot. I'll probably even upgrade to the Dual band version for $30, even better!

http://www.ebay.com/itm/INTEL-7260NGW-M-2-NGFF-Wireless-N-Card-802-11AN-Bluetooth-4-0-/111197326988?pt=US_Internal_Network_Cards&hash=item19e3e08e8c

[Matt (hlmtre)]

Sadly, the T440s does not have any panels at the bottom. I bought the
laptop and was very excited to have a nice Thinkpad with all the power I
needed, but have been disappointed with the thoroughly awful trackpad and
this Realtek wireless card's issues on Linux.

http://yournewsticker.com/wp-content/uploads/2014/01/Lenovo-ThinkPad-T440s-6.jpg

On Thu, Jul 31, 2014 at 12:18 PM, Juan <email address hidden> wrote:

> I just took it apart and tried it. Easy. On my 540p there are 2 service
> panels at the bottom 1 for HDD and memory the other smaller one is the
> wifi card. I already swapped the HDD with SSD by taking apart the larger
> one. Just make sure you get the Intel 7260 NGW form factor for the half-
> mini size pci slot. I'll probably even upgrade to the Dual band version
> for $30, even better!
>
> http://www.ebay.com/itm/INTEL-7260NGW-M-2-NGFF-Wireless-N-Card-802-11AN-
>
> Bluetooth-4-0-/111197326988?pt=US_Internal_Network_Cards&hash=item19e3e08e8c
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions
>

[Juan (hyhh)]

Matt,
Really sorry to hear that. I googled T440s bottom pictures and I see almost all of them have services panels? It's maybe not teh model but what particular chassis you get? Have you tried the sides? There should be at least a way to swap the hard drive out without taking the whole thing apart?

[Lee Zamparo (zamparo)]

@Matt, @Juan

This page suggests you have to remove the base cover on the T440s to access the wireless card: http://support.lenovo.com/en_IN/detail.page?DocID=PD027945

To remove the base cover you need only unclip the external battery pack, then unscrew 8 screws. No messing with the motherboard or monitor. Plus, the Intel card @Juan posted a link to sells for around $30, which is comparable to any external USB interface.

I might try the same, as the drivers for Intel cards are quite mature IIRC.

[Matt (hlmtre)]

I went back and looked and it's clear the bottom comes off entirely. I had
an old X60 that served me extremely well and it had a slot for removing the
harddrive and a little panel with the RAM exposed right below it... I was
hoping for a little panel to expose it as opposed to needing a spudger to
pull the damn whole bottom off.

But I've done it before the replace the HDD, and it isn't very hard. For my
model, I found this service guide by Lenovo:
http://download.lenovo.com/ibmdl/pub/pc/pccbbs/mobiles_pdf/t440s_hmm_en_sp40a25360_03.pdf

One thing I did notice when looking at replacement cards was the issue with
the motherboard whitelist -- is the card listed earlier in this thread good
for our machines?

On Thu, Jul 31, 2014 at 12:58 PM, Juan <email address hidden> wrote:

> Matt,
> Really sorry to hear that. I googled T440s bottom pictures and I see
> almost all of them have services panels? It's maybe not teh model but what
> particular chassis you get? Have you tried the sides? There should be at
> least a way to swap the hard drive out without taking the whole thing apart?
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions
>

[William Oprandi (woprandi)]

I also have a T440s and i'm also disappointed with the 2 problems. The trackpad is broken for the 2nd time and the wireless card. ...
My x200 has no Linux compatibility problem so I didn't care about wireless card model when I bought my T440s...

 I can compare this 2 generations of thinkpad : the new keyboard is not bad, but I think it's worse than the old. But I have backlight with the new. The T440s seems less solid than the x200 (but it's a ultrabook). And ofc the awful trackpad.....

I would like to buy and install Intel wireless card, I understand we cannot wait a decent driver by Realtek.....

[Juan (hyhh)]

Hi Matt,
The Intel 7260 card should not be an issue because some of T440/540 ship with exactly that.
http://shop.lenovo.com/us/en/laptops/
Try configure a T540p/T440p yourself I'm sure the Intel Wireless 7260 card (both single band and dual band variants) is an option.

That's why I'll only choose that model instead of any newer and better cards which may not work:)

[Juan (hyhh)]

William,
There are a few bizarre things about the new 40 series. I agree. They even removed the HDD light, no physical switches for wifi and no charging indicator when computer is off... I use it with an external screen and keyboard at home so keyboard isn't an issue for me.
The T430 I use at work has all of them (charging light, HDD light and physcial wifi switch). I like the T430, that's why I devoted to a Lenovo and only to find out after opening the box about those bizzare things.
They are no big issues, though and after a while you get used to it. It's a decent laptop. The most I liked about is Lenovo's heat dissipation much better that the HP I had before! Fan barely comes on even with heavy use!

[Larry Finger (larry-finger-lwfinger)]

One thing to note. The firmware that is currently available from the linux-firmware repo for the 7260 gets many firmware restarts. Unfortunately, this repo is the source for all distros. There is experimental firmware available for testing that seems to fix the problem. Search the linux-wireless mailing list archives to find that firmware.

[Juan (hyhh)]

Larry,
Is the firmware you mentioned the same as the ones on the official Intel website?

http://www.intel.com/support/wireless/wlan/sb/CS-034398.htm

[Larry Finger (larry-finger-lwfinger)]

No. There is a naming problem that prevents it from being "officially" released by Intel. You need to get it from https://git.kernel.org/cgit/linux/kernel/git/egrumbach/linux-firmware.git/plain/iwlwifi-7260-9.ucode?h=Core6.

[Yu Leng (tctctcly)]

tried #132. Unfortunately, it doesn't work for me.

:(

[Larry Finger (larry-finger-lwfinger)]

I was unable to get the latest Realtek drivers into a form for submission to the kernel in time to make kernel 3.17. As a result, I created a git repo at http://github.com/lwfinger/rtlwifi_new.git. I have no idea if it will work for youm but it is the latest code,'

[Yu Leng (tctctcly)]

Thank you Larry for posting the latest driver code. I am no where close to the expertise most people have on this forum. This is may well be a beginner question. Do I just git clone your repo, do "make", "make install" and then reboot?

[Larry Finger (larry-finger)]

On 08/07/2014 12:42 AM, Yu Leng wrote:
> Thank you Larry for posting the latest driver code. I am no where close
> to the expertise most people have on this forum. This is may well be a
> beginner question. Do I just git clone your repo, do "make", "make
> install" and then reboot?

You are close. Building code should always be done as a regular user, not as
root. Thus the correct commands are "make" and "sudo make install".

Larry

[[e-mail] (vworld4u-g)]

Hi,

I tried approach of (132):

sudo apt-add-repository ppa:mjeanson/ppa
sudo apt-get update
sudo apt-get upgrade
sudo apt-get install rtl8192ee-dkms

Problem faced: Now Wifi connects briefly and disconnects after 2-3 mins. Even when connected, 1-2 internet requests succeed and then every request fails.

I tried to remove the rtl8192ee-dkms module and cloned the repo as in 158, but make fails with 2 errors. Error log is shown below:

make -C /lib/modules/3.13.9/build M=/home/{username}/git/rtlwifi_new modules
make: *** /lib/modules/3.13.9/build: No such file or directory. Stop.
make: *** [all] Error 2

I have no idea how to fix this issue. I am feeling I made a mistake going for L440 laptop now Or 14.04 is such a unstable version yet! I am not sure at all! Any help appreciated!

[Perry Yan (perry-l)]

Hi all,

I can confirm that with kernel 3.16.0-031600-generic the connection is more or less stable. However, the download speed is extremely slow (maxes out at around 1 Mbps). Does anyone have a workaround to increase the throughput?

Regards
PY

[Larry Finger (larry-finger-lwfinger)]

The only thing I can suggest is to try the GitHub repo in comment #156.

[Bruce D'Arcus (bdarcus)]

Larry - just to confirm, simply doing the standard make/make install that
you mentioned previously here should correctly utilize that github code?
There's nothing more to be done?

Because I haven't found that particularly effective. I still can't get
reliable connections.

On Wed, Aug 13, 2014 at 10:21 AM, Larry Finger <email address hidden>
wrote:

> The only thing I can suggest is to try the GitHub repo in comment #156.
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> Status in HWE Next Project:
> Confirmed
> Status in HWE Next trusty series:
> Confirmed
> Status in “linux” package in Ubuntu:
> Confirmed
>
> Bug description:
> Open this bug to trace device driver RTL8192EE.
>
> We have seen this wireless card on some laptops, id is [10ec:818b],
> driver is RTL8192EE.
>
> Status: waiting for this driver being moved out of staging directory.
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions
>

[Larry Finger (larry-finger-lwfinger)]

Yes, "git pull && make && sudo make install" gets you the latest Realtek code. If that still does not work, then you will need to document what is happening on your system.

At the moment, I am working full time trying to get all the rtlwifi drivers into a stage that the new versions can be committed to the kernel. I do not have time to try to analyze what is wrong in your case. When I tested rtl8192ee, I got reasonable throughput.

Try loading the module with "ips=0" to disable power save. Perhaps your AP is getting that part wrong.

[Siu On Chan (sochan)]

@vworld4u-g #159: I believe you face the same problem as I do, namely initial authentication error. To confirm, run "dmesg | tail -20" at the terminal, and you'll see "wlan0: deauthenticating from xx:xx:xx:xx:xx:xx by local choice (Reason: 3...)". (Sometimes you'll see Reason 2).

@Bruce #162: My limited Linux knowledge suggests that you may want to run "sudo depmod" after "make" and "sudo make install", but I could be wrong.

Finally, let me say a few words about my experience with Larry's git in #156. The first time I tried the new driver, the connection felt very fast. But then it went back to normal (which is a little slow) the next day.

Also, the initial authentication error still keeps coming back (the problem I mentioned in the 1st paragraph). Almost every time I boot my machine, it takes 2-15 mins of repeatedly removing and inserting the rtl8192ee module, plus repeatedly killing wpa_supplicant and NetworkManager. I also tried "ips=0" as suggested in #163 with no luck.

In case you wonder, I'm using the 3.16 kernel from Ubuntu PPA, and I've blacklisted the old r8192ee module from the staging directory.

[Siu On Chan (sochan)]

Alright, I didn't read Makefile, which automatically runs "depmod" for you. Please ignore what I said about "depmod".

[[e-mail] (vworld4u-g)]

Hi Siu On Chan,

Thanks for your response. I am too not a big Linux dev kind of person, but I am using Ubuntu for last 5 years without any dissatisfaction. In fact I love Ubuntu for its usefulness. With respect to error I am getting

make -C /lib/modules/3.13.9/build M=/home/venkatesha/git/rtlwifi_new modules
make: *** /lib/modules/3.13.9/build: No such file or directory. Stop.
make: *** [all] Error 2

I see that I don't have build file inside /lib/modules/3.13.9 folder in my computer. Is it assuming this file to be present? If so, what packages bring this build file into the system?
Somebody please advice me on this.

[Larry Finger (larry-finger-lwfinger)]

You need the kernel headers. I have no idea what Ubuntu calls the package, but someone else will.

[Siu On Chan (sochan)]

The package in your case is "linux-headers-3.13.9-generic". Here's how you find out the package that a file belongs to http://superuser.com/questions/10997/find-what-package-a-file-belongs-to-in-ubuntu-debian .

[[e-mail] (vworld4u-g)]

Hi Siu On Chan,

Thanks for the information. With the linux-headers-3.13.9-generic for my amd64 architecture I downloaded necessary headers package and I am able to build (make and sudo make install) in my computer. Now I can see my Wifi networks, but it still not able to connect to WIFI network.

It tries to connect for long time, but fails finally. dmesg tail displays the following error:

[ 504.653769] rtl8192ee:rtl92ee_get_hw_reg():<0-0> switch case not process 5c
[ 505.836788] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready
[ 505.859274] rtl8192ee:rtl92ee_get_hw_reg():<0-0> switch case not process 5c
[ 508.773363] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready
[ 508.794403] rtl8192ee:rtl92ee_get_hw_reg():<0-0> switch case not process 5c
[ 512.450822] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready
[ 512.466441] rtl8192ee:rtl92ee_get_hw_reg():<0-0> switch case not process 5c
[ 513.163302] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready
[ 515.069647] wlan0: authenticate with 80:a1:d7:8d:62:48
[ 515.086303] wlan0: direct probe to 80:a1:d7:8d:62:48 (try 1/3)
[ 515.289423] wlan0: direct probe to 80:a1:d7:8d:62:48 (try 2/3)
[ 515.493655] wlan0: direct probe to 80:a1:d7:8d:62:48 (try 3/3)
[ 515.697869] wlan0: authentication with 80:a1:d7:8d:62:48 timed out

So it is timing out while authenticating! Any idea where it is failing!?

[Siu On Chan (sochan)]

vword4u-g:

I wish I knew how to fix it. I'm facing exactly the same problem as you. Since I started using Larry's git driver, I need to try reconnecting for 15-30 times (ie removing and inserting modules, killing wpa_supplicant etc), until I get a reliable connection. That usually takes 10+ minutes. The initial connection/authentication error has become frustratingly annoying with the git driver.

And the "managed=true" fix I suggested in #80 no longer helps. The fix changes the error message, but doesn't reduce the number of retrials.

Another observation: I travelled with my laptop last week. The same initial connection error happened on other networks (public airport Wifi, etc). It disappeared faster than on my home network (similar to what Sascha said in #140). So the error to a large extent isn't specific to my home network setup. I was using the staging driver in kernel 3.16 at that time.

[Bruce D'Arcus (bdarcus)]

Not sure if this is helpful, but it shows the dmesg output that corresponds
with the connection loss (that deauthorization bit) and then reconnect that
I'm seeing:

[17101.068960] wlan0: authenticate with 48:f8:b3:6b:12:a8
[17101.087954] wlan0: send auth to 48:f8:b3:6b:12:a8 (try 1/3)
[17101.110706] wlan0: authenticated
[17101.110976] wlan0: associate with 48:f8:b3:6b:12:a8 (try 1/3)
[17101.135135] wlan0: RX AssocResp from 48:f8:b3:6b:12:a8 (capab=0x411
status=0 aid=3)
[17101.175212] wlan0: associated
[17101.175221] IPv6: ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
[17285.889932] wlan0: AP 48:f8:b3:6b:12:a8 changed bandwidth, new config is
2447 MHz, width 1 (2447/0 MHz)
[17495.311240] wlan0: deauthenticated from 48:f8:b3:6b:12:a8 (Reason:
4=DISASSOC_DUE_TO_INACTIVITY)
[17495.407874] cfg80211: Calling CRDA to update world regulatory domain
[17495.408109] wlan0: authenticate with 48:f8:b3:6b:12:a8
[17495.427226] wlan0: send auth to 48:f8:b3:6b:12:a8 (try 1/3)
[17495.427729] cfg80211: World regulatory domain updated:
[17495.427732] cfg80211: DFS Master region: unset
[17495.427733] cfg80211: (start_freq - end_freq @ bandwidth),
(max_antenna_gain, max_eirp), (dfs_cac_time)
[17495.427737] cfg80211: (2402000 KHz - 2472000 KHz @ 40000 KHz), (300
mBi, 2000 mBm), (N/A)
[17495.427739] cfg80211: (2457000 KHz - 2482000 KHz @ 40000 KHz), (300
mBi, 2000 mBm), (N/A)
[17495.427741] cfg80211: (2474000 KHz - 2494000 KHz @ 20000 KHz), (300
mBi, 2000 mBm), (N/A)
[17495.427743] cfg80211: (5170000 KHz - 5250000 KHz @ 40000 KHz), (300
mBi, 2000 mBm), (N/A)
[17495.427745] cfg80211: (5735000 KHz - 5835000 KHz @ 40000 KHz), (300
mBi, 2000 mBm), (N/A)
[17495.450378] wlan0: authenticated
[17495.451696] wlan0: associate with 48:f8:b3:6b:12:a8 (try 1/3)
[17495.475764] wlan0: RX AssocResp from 48:f8:b3:6b:12:a8 (capab=0x411
status=0 aid=3)
[17495.515424] wlan0: associated

[Sharon Talbot (sharon-talbot)]

I've been keeping an eye on this bug for a while, and decided to give the drivers another shot today. I was very happy to see wifi come up after pulling the source, compiling, and installing.

An hour later, my wifi disconnected and would not reconnect. Nothing particularly helpful in dmesg. I decided to uninstall the drivers and it hung.

I rebooted and sudo make uninstall.

Wish I would have bought a different wireless card for this laptop. I bought a mini USB wifi dongle but it is also realtek and acts up.

[bulldozer2003 (bulldozer2003)]

I just installed 14.04.1 on my T440s. Here is how I installed the 3.16 kernel and got wifi working by adding utopic (14.10) sources to apt.

BE WARNED: This may be dangerous.

Edit /etc/apt/sources.list and add the lines:
deb http://archive.ubuntu.com/ubuntu/ utopic main restricted
deb http://archive.ubuntu.com/ubuntu/ utopic-updates main restricted
deb http://archive.ubuntu.com/ubuntu/ utopic-security main restricted

apt-get update
apt-get install linux-generic linux-firmware
DO NOT UPGRADE or DIST-UPGRADE!!!
Check that you'll be installing 3.16-x before hitting Y.

Edit /etc/apt/sources.list and COMMENT OUT the three lines from above:
#deb http://archive.ubuntu.com/ubuntu/ utopic main restricted
#deb http://archive.ubuntu.com/ubuntu/ utopic-updates main restricted
#deb http://archive.ubuntu.com/ubuntu/ utopic-security main restricted

apt-get update
^^^ THIS IS IMPORTANT TO NOT RUIN YOUR SYSTEM^^^

Reboot and enjoy wireless. Occasionally repeat the procedure above to check for kernel upgrades.

[Siu On Chan (sochan)]

@bulldozer2003 #173:

What wireless card do you have? (You can find out by running "lspci".) Some models of T440s (especially older ones) seem to have other network cards.

@Bruce #171:

Your dmesg looks almost the same as mine. I also looked into /var/log/syslog which provides more detailed info, but again I don't know what's wrong, given my limited knowledge about wireless on Linux (and I wish there were more useful pointers online).

Larry:

I even thought of modifying Larry's source code to print more useful debug messages. But I don't know how to find useful documentations about wireless drivers on Linux. So I'm unable to help debug. I hope Larry (or someone else) could point me to the right references.

Let me add that I also tried wicd in place of NetworkManager, but it shows pretty similar errors. So this problem isn't specific to NetworkManager. I've also set IPv6 to "Ignore" in NetworkManager, if that matters.

[bulldozer2003 (bulldozer2003)]

@sochan
I wish I had a different card. I would have preferred a dual band card.

Also, this driver in 3.16 seems stable, I've been using it since my earlier post without any problems despite the dmesg warning:
[ 3.751315] r8192ee: module is from the staging directory, the quality is unknown, you have been warned.

[Siu On Chan (sochan)]

Thanks bulldozer2003 for the suggestion. Having used the Utopic linux 3.16.0-8 driver for a few days, the driver feels similar to the Ubuntu PPA 3.16 kernels I was using before trying the git driver. Now I need 0-4 connection attempts to get a reliable connection after each reboot, which is acceptable. I'll stick to this driver for now.

[bulldozer2003 (bulldozer2003)]

After using the r8192ee driver in the 3.16 kernel, I found SSH sessions to
be extremely unreliable.

Web browsing was tolerable but slow and required occasional retries.

I've now purchased a WiFi USB dongle and will test the r8192ee driver as
new 3.16 kernels are released for utopic.

[Laksh Gupta (lakshgupta)]

I am also facing the same problem with my wireless card; no driver available.

02:00.0 Network controller [0280]: Realtek Semiconductor Co., Ltd. Device [10ec:818b]
08:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd. Device [10ec:5227] (rev 01)

[Nick Gerner (nick-gerner)]

We just applied the fix from #132 (with the correct rtl8192ee-dkms package name) on one of the T440s laptops in our office. Seems to have worked, but only 15 or 20 minutes have gone by so far :)

We'll try rolling this out to another one of our laptops. I'll post back if we run into any other issues.

[[e-mail] (vworld4u-g)]

Any updates about this issue fixture status? I am yet to migrate my kernel to 3.16 as Siu On Chan (sochan) told earlier! But I am yet to get some working solution on my L440 Laptop.

After today's updates, I am seeing problems in my ethernet connection too! I am clueless on how to use this laptop for official purposes!

Please keep posting the updates to the bug so that I can follow the patch or decide to purchase a wifi usb adapter for working.

[Clive Cox (clivejcox)]

I am using the r8192ee module compiled from source on my t440p and it works most of the time though does disconnect and fail periodically. I am using linux mint 17. Unfortunately I am getting periodic system crashes: keyboard stops working or starts emitting endless carriage returns. Sometime the "network manager" becomes inoperative brings up an error window. The only solution is to power off and on again.

[Larry Finger (larry-finger-lwfinger)]

What source are you using?

The latest version of the driver is at http://github.com/lwfinger/rtlwifi_new.git. That version will build on any kernel 3.0 or later, as long as the kernel headers are available.

[Clive Cox (clivejcox)]

Hi Larry. I used the source from mjeanson in #86. I will use your github and report back.

[Clive Cox (clivejcox)]

I built and installed from http://github.com/lwfinger/rtlwifi_new.git and its working fine on my T440p so far.

[Clive Cox (clivejcox)]

Unfortunately I am still getting connection locks where wifi stops working and I can't reconnect and also I can't shut down. Only way is to press external power button. Also, I still get the situation where keyboard becomes unresponsive and keeps emitting a single key, and same thing I can only hard power cycle. Obviously it could be something else causing these issues. Is there any particular things I should look for in logs - I don't see anything of interest in dmesg or kern.log but mabe I am missing something.

Just to repeat this is for lenovo T440p with drivers built from github acct from #182 - my kernel is 3.13.0-24-generic

[Damian Nadales (dnadales)]

I'm using the 3.16.0-031600rc6-generic kernel with the latest drivers fetched from github. I can connect to my router, however the speed is quite slow. It could take a couple of minutes till I get the results of a google search.

I tried to look at the debug information (using the guidelines in 0) but I cannot tell what is wrong. I'll attach the debug log, in case this can help. I don't know if anybody besides Larry is working on this.

[Damian Nadales (dnadales)]

I see a lot of scans for other networks while connected. Is this normal? (BSSID mismatch)

I really cannot draw any conclusions about this log. Does anybody can pinpoint a specific problem?

[Sylvain Corlay (sylvain-corlay)]

I am experiencing some general freezes and crashes with the kernel 3.16-2 and the firmware for RTL8192EE. Any idea on how to troubleshoot this.

[Larry Finger (larry-finger-lwfinger)]

@Sylvain: Are you using the driver from #155? If not, please do.

[David Spies (dreamcontrol8000)]

I'm on a Lenovo T440p; Ubuntu 14.04 AMD64; kernel 3.13.0-35-generic
(I had to downgrade my bios to 1.14 as suggested here: http://askubuntu.com/a/502065/63688 )

I tried installing the drivers from the git repository in 182, but as soon as I boot, it crashes. (for anyone else who runs into this problem, you can boot into recovery mode to uninstall the driver: sudo make uninstall). Here's the last part of the error message (please forgive any typos; I had to copy this by hand):

[ 5.107060] Code: c9 80 41 89 4e 10 f6 43 3a 20 74 7e 48 8d 4c 37 18 48 01 ca 0f 84 b9 07 00 00 83 7a 68 18 0f 86 b5 03 00 00 48 8b ba d8 00 00 00 <0f> b7 37 89 f1 66 81 e1 fc 00 66 81 f9 80 00 74 45 66 81 f9 c0
[ 5.107186] RIP [<ffffffffa0f9ee82>] rtl92ee_rx_query_desc+0x172/0x980 [rtl8192ee]
[ 5.107223] RSP <ffff88023e243c28>
[ 5.113573] Kernel panic - not syncing: Fatal exception in interrupt
[ 5.113621] drm_kms_helper: panic occurred, switching back to text console

After which I have to forcibly cut power to turn it off.

[Clive Cox (clivejcox)]

I just had a kernel panic when running my laptop under high memory load it was:

BUG: unable to handle kernel NULL pointer dereference
at
rtl92ee_rx_query_desc

Later on there was a stack trace with

rtl_pci_init_one_rxdesc
rtl_pci_rx_interrupt
raw_spin_unlock
ieee80211_rx_handlers
put_page
rtl_pci_interrupt

Hope that helps

running on lenovo t440p using 3.13.0-24-generic

[Larry Finger (larry-finger-lwfinger)]

The traceback is a lot more useful when you provide the numbers that follow the names of the routines. Those are of the form rtl_pci_init_one_rxdesc+0xyyyy/0xzzzz. That way I can identify exactly the statement that failed.

I have nearly reached the point where I will be testing rtl8192ee hard.

[David Spies (dreamcontrol8000)]

I upgraded to 3.16 using the method suggested in 173 and it seems to work.

However, I still can't connect to my school's wifi network (WPA2; PEAP authentication (version 0); MSCHAPv2 inner authentication).

Here's the output of dmesg | tail -20

[ 69.524510] wlan1: authenticate with d0:c2:82:7e:d6:61
[ 69.557053] wlan1: direct probe to d0:c2:82:7e:d6:61 (try 1/3)
[ 69.759838] wlan1: direct probe to d0:c2:82:7e:d6:61 (try 2/3)
[ 69.964168] wlan1: direct probe to d0:c2:82:7e:d6:61 (try 3/3)
[ 70.168492] wlan1: authentication with d0:c2:82:7e:d6:61 timed out
[ 71.915835] wlan1: authenticate with f8:4f:57:06:ea:91
[ 71.948819] wlan1: send auth to f8:4f:57:06:ea:91 (try 1/3)
[ 72.051502] wlan1: send auth to f8:4f:57:06:ea:91 (try 2/3)
[ 72.155668] wlan1: send auth to f8:4f:57:06:ea:91 (try 3/3)
[ 72.259832] wlan1: authentication with f8:4f:57:06:ea:91 timed out
[ 72.809322] wlan1: authenticate with d0:c2:82:7e:d6:61
[ 72.826038] wlan1: send auth to d0:c2:82:7e:d6:61 (try 1/3)
[ 72.928889] wlan1: send auth to d0:c2:82:7e:d6:61 (try 2/3)
[ 73.033070] wlan1: send auth to d0:c2:82:7e:d6:61 (try 3/3)
[ 73.137235] wlan1: authentication with d0:c2:82:7e:d6:61 timed out
[ 80.546231] wlan1: authenticate with 00:08:30:a6:3b:51
[ 80.578648] wlan1: direct probe to 00:08:30:a6:3b:51 (try 1/3)
[ 80.781390] wlan1: direct probe to 00:08:30:a6:3b:51 (try 2/3)
[ 80.985725] wlan1: direct probe to 00:08:30:a6:3b:51 (try 3/3)
[ 81.190102] wlan1: authentication with 00:08:30:a6:3b:51 timed out

[Amyth (aroras-official)]

I exactly face the same problem as @Clive Cox in #181. Following is what I am using

Machine: Lenovo L440
OS: Ubuntu 14.04 LTS
OS type: 64bit
Kernel Version: 3.13.0-35-generic
Drivers: rtlwifi_linux_mac80211_0019.0320.2014V62

Most of the times it works and connects properly but periodically it fails to connect and as @Clive said in #181 the machine freezes periodically. Also at times the NetworkManager goes unresponsive and the only way to get back is to restart using the power button on the machine.

[Sylvain Corlay (sylvain-corlay)]

+1, same issues. Experiencing freezes as well with the latest firmware. Must REISUB.

[Larry Finger (larry-finger-lwfinger)]

Troy, the Realtek engineer, found a locking problem today that may be the cause. Pull a new copy from GitHub.

[Clive Cox (clivejcox)]

I updated my kernel with the latest from GitHub and it work all of today and then locked up again as before unfortunately.

Is there anyway to get some debug back to you Larry? I don't see anything suspicious in the logs.

[Larry Finger (larry-finger-lwfinger)]

If a freeze never produces any logged entries, then it will be difficult to find.

Was the system more stable after the previous fix? If that helped, I just pushed a new commit that might help some more.

[Clive Cox (clivejcox)]

I'm not sure its more stable yet Larry, but seems to be I think, though I have had rare days with only 1 lockup before.

The only time I had a complete kernel panic and errors printed to screen as shown by my incomplete stack trace from #191 was when I had another process using all of memory and lots of swap and I then tried to use browser at same time.

I see you added a new branch kernel_version with recent commits, but if I switch to that and try to compile it fails for me:

In file included from /root/wifi/rtlwifi_new/regd.c:31:0:
/root/wifi/rtlwifi_new/regd.c: In function ‘_rtl_reg_apply_beaconing_flags’:
/root/wifi/rtlwifi_new/regd.h:34:32: error: ‘IEEE80211_CHAN_NO_IR’ undeclared (first use in this function)
 #define IEEE80211_CHAN_NO_IBSS IEEE80211_CHAN_NO_IR
                                ^
/root/wifi/rtlwifi_new/regd.c:180:20: note: in expansion of macro ‘IEEE80211_CHAN_NO_IBSS’
      ch->flags &= ~IEEE80211_CHAN_NO_IBSS;
                    ^

[Larry Finger (larry-finger-lwfinger)]

I had not applied that latest fix to the master branch as I thought you were on kernel_version. That latter one is only for recent kernels - I think 3.14 or newer is required.

In any case, I pushed that latest change to the master branch.

[Clive Cox (clivejcox)]

So far seem more stable Larry. I still get periods where the network connection stops for up to a minute, not sure if this is related, but no crashes today. Fingers crossed, but things seem to be moving in the right direction.

[Larry Finger (larry-finger-lwfinger)]

Good to hear that the system is more stable. Locking problems can be difficult to find as they frequently are affected by subtle differences in the speed and number of CPUs, etc. The other locking problem showed up on my system as an attempt to free DMA memory that had never been assigned. The traceback showed that the routines were being called out of order. No system freezes.

We will get to the poor performance later. Nearly all of the drivers show this problem, so it seems to be generic. Troy reproduced it yesterday, and I expect a fix soon.

[Graham (neubig)]

Hello, I attempted to install the driver on my Lenovo L540 but am consistently getting Kernel Panic when I reboot after installation. I'm using 3.13.0-36-generic and simply cloned the repository, and did "make; sudo make install".

I've attached the make log (with some warnings) and dmesg, but it looks like dmesg stopped writing once the kernel panicked, so I also attached a screenshot with what was displayed on the screen before the computer died. It'd be really nice to get wifi working, so I hope this helps!

[Larry Finger (larry-finger-lwfinger)]

I am beginning to understand this problem, but I do not have a general fix.

In kernel 3.14, the authors of mac80211 changed the API (applications interface) for routine ieee80211_is_robust_mgmt_frame() to be able to intercept a certain kind of error. For those drivers that could not fit the new model, which includes the Realtek codes, they created a new entry point that used the old method. It was named _ieee80211_is_robust_mgmt_frame(), i.e. it has an underscore. For out-of-kernel drivers such as those in the rtlwifi_new repo, there must be code that determines at compile time which API needs to be used. Such code is in these drivers. If the kernel version is 3.14 or newer, use the underscored version. If older use the one without.

Why does it fail here? The reason is that Ubuntu applied the API change to their 3.13 kernels! Now the decision gets harder. If the kernel is 3.14, or if this is Ubuntu with version 3.13.11 + some unknown patch level, use the new API. That gets quite difficult; however, if I take the simple step of having all 3.13 versions use the new API, then I break every user of 3.13 that does not run Ubuntu. I refuse to do that for the following reason. Ubuntu should not have changed the API in mid 3.13!! In doing so, they make this kind of problem for everyone trying to provide out-of-kernel drivers. For me to fix the problem for Ubuntu users by breaking every other distro's users would be effectively to reward Ubuntu for what I consider to be their bad behavior.

How do I know this is the problem. First of all, the make log that was posted shows "warning: passing argument 1 of ‘ieee80211_is_robust_mgmt_frame’ from incompatible pointer type [enabled by default]". It is true that warnings can usually be ignored; however, when arguments have an incompatible pointer type, the end result will usually be a kernel crash. When that warning occurs for a routine that has changed API, it really sticks out. Secondly, I see this change in Ubuntu's 3.11.0-36 patch file.

What can Ubuntu users do? To my thinking, switching to a 3.14 or newer kernel would be best. If you really want to stay with Ubuntu's 3.13.0-36 kernel, then run the following command in the rtlwifi_new directory:

sed -i 's/ieee80211_is_robust_mgmt_frame/_ieee80211_is_robust_mgmt_frame/g' rtl8*/trx.c

That will change the code to the new API for every kernel. Of course, that will break every kernel older than 3.13, and the command should be run *ONLY* if you see warnings of the kind noted above. To revert that change, you will need to run 'git checkout -f'.

Sorry that it has taken so long to sort this out, but I really never expected the API to change in the middle of kernel 3.13 for any distro. If the tone seems to be too bitchy, remember that I'm really angry about what has happened, and all the time I have wasted on sorting it out. My comments here are a lot milder than the way I really want to say them.

[[e-mail] (vworld4u-g)]

Hi Larry,

Thanks for a detailed post with your findings. We understand your
frustration here. But I have a strange issue in my L440 which has R8192EE
Wifi Adapter and 3.17.0-031700rc6_3.17.0-031700rc6.201409211935 kernel
which is the latest available kernel. I installed this kernel few days back
and tried with new driver from github repo. I don't find any errors in the
new kernel make log (as you were mentioning in the above post - which I saw
in default 3.13 kernel). But after installing this kernel and new driver, I
am still not able to connect to Wifi networks - It tries several times and
fails to connect. I can see following messages repeatedly on dmesg while it
tries connecting.

[ 41.078474] wlan0: authenticate with b0:a8:6e:40:38:02
[ 41.111422] wlan0: direct probe to b0:a8:6e:40:38:02 (try 1/3)
[ 41.314257] wlan0: direct probe to b0:a8:6e:40:38:02 (try 2/3)
[ 41.518483] wlan0: direct probe to b0:a8:6e:40:38:02 (try 3/3)
[ 41.722704] wlan0: authentication with b0:a8:6e:40:38:02 timed out
[ 41.875292] wlan0: authenticate with ac:4b:c8:67:da:82
[ 41.891899] wlan0: send auth to ac:4b:c8:67:da:82 (try 1/3)
[ 41.995006] wlan0: send auth to ac:4b:c8:67:da:82 (try 2/3)
[ 42.099117] wlan0: send auth to ac:4b:c8:67:da:82 (try 3/3)
[ 42.203231] wlan0: authentication with ac:4b:c8:67:da:82 timed out

I can see the following device driver details using lshw -C "Network"
command:

 *-network
       description: Wireless interface
       product: RTL8192EE PCIe Wireless Network Adapter
       vendor: Realtek Semiconductor Co., Ltd.
       physical id: 0
       bus info: pci@0000:02:00.0
       logical name: wlan0
       version: 00
       serial: 34:23:87:24:72:87
       width: 64 bits
       clock: 33MHz
       capabilities: bus_master cap_list ethernet physical wireless
       configuration: broadcast=yes driver=r8192ee
driverversion=3.17.0-031700rc6-generic firmware=N/A latency=0 link=no
multicast=yes wireless=IEEE 802.11bgn
       resources: irq:30 ioport:5000(size=256) memory:f2400000-f2403fff

Do you or anybody facing similar issue? If yes, can you please help me out
in figuring cause of this issue. I am clueless and missing out something
here. I already feeling like I chose a wrong OS Ubuntu.

For curiosity I have one more question - Does this new hardware RTL8192
works on older version of Ubuntu or any other Linux distro (probably
desktop/lite version) for that matter? Which one you safely bet? I am ready
to migrate to any other version of Linux which can fix this issue for ever
!!

Thanks in advance,
Venkatesha T R

On Thu, Sep 25, 2014 at 9:45 PM, Larry Finger <email address hidden>
wrote:

> I am beginning to understand this problem, but I do not have a general
> fix.
>
> In kernel 3.14, the authors of mac80211 changed the API (applications
> interface) for routine ieee80211_is_robust_mgmt_frame() to be able to
> intercept a certain kind of error. For those drivers that could not fit
> the new model, which includes the Realtek codes, they created a new
> entry point that used the old method. It was named
> _ieee80211_is_robust_mgmt_frame(), i.e. it has an underscore. For out-
> of-kernel drivers such as those in...

[Secrect A Gent (mrmik2000-ubuntu)]

The RTL8192 does not work in the Uberstudent 4.0 distro, either.

[Graham (neubig)]

Hi Larry,

Thanks for the reply and all your hard work. Sorry to hear you have to deal with frustrating corner cases like this.
I upgraded to the latest non-RC kernel (3.16.3-031603-generic), and am no longer getting compilation warnings or kernel panic.
I am, however experiencing exactly the same problem as #205. When I reboot the computer it looks like it connected once (the wifi icon on the top right of the screen is showing "connected"), but then when I try to access a web page it times out, loses the connection, tries to re-connect several times, then gives up.

I've attached dmesg, if that will be any help.

Thanks again!
Graham

[varunendra (varunendra)]

Hello Graham,

I've been following this bug for quite sometime, just out of curiosity. However, many of the comment posts here look like common issues that may or may not be related to this particular bug. Yours sounds similar. Oh, but not really same as #205 (Venkatesha). His was not connecting at all (dmesg suggests a possible Network Manager bug/confusion where it hops between different APs of same SSID before a connection is finally established), while yours at least seems to connect initially. I can't find your attached dmesg (forgot to attach?).

I am actively involved in wifi troubleshooting at ubuntu forums, and this kind of problem, where network (wifi or ethernet) seems to be connected but browsing internet fails. I'm not yet sure about the exact reason but it seems to be like some sort of issue with 'resolvconf' package that is supposed to take care of DNS configuration. In many of such issues, the connection works as soon as 'resolvconf' package is removed or NM (IP, Subnet, Gateway, DNS, or atleast DNS) is manually configured instead of leaving it to DHCP.

Not saying yours must be same, but it *sounds* like above, so the DNS issue is worth checking. Just try to ping first your gateway (AP/router), if successful, try pinging any internet address, for example Google's DNS servers 8.8.8.8 and 8.8.4.4. If you get ping reply, it is definitely a DNS issue.

As for the disconnection issue, I frequently need to set driver parameters "fwlps=N" and "ips=N" with almost all realtek drivers these days. I know these options are there to save power on wireless card, thus disabling them may cause unnecessary power drain issue, but disabling them with values "N" frequently seems to solve disconnection issues.

Then there is another issue where the "Power Management" on wireless kicks in automatically and above two options are either not available or don't help. Common with Intel driver, but I think I have seen that with one of the Realtek drivers as well (recently, but don't remember which one). There are other ways to keep it turned off.

I'm posting this just because I think people posting here should be aware that there are also some common issues out there that may be solved by some other ways. For detailed steps on trying the above possible fixes, just check out the "Networking & Wireless" section of Ubuntu Forums. One may find a similar problem with a solution there. Of course they can also post their own thread asking for help, with reference to this bug if they have the same card.

Good Luck,
Varun

[Graham (neubig)]

Hi Varun,

Thank you for the pointer! Because my problem is now potentially not related to this bug I'll move my request to the networking forum instead. (I tried to set "ips=0 fwlps=0" but adding a conf file to modprobe.d doesn't seem to change anything, and "modprobe rtl8192ee" gives an "Exec format error." Grr.)

Just for reference, my dmesg was actually attached, but it's a bit difficult to see it with the launchpad interface. It's in the upper right of the page as "dmesg.log".

Graham

[varunendra (varunendra)]

> It's in the upper right of the page as "dmesg.log".

Yup, found that now, a place I never noticed before.

Also, your problem may indeed be related to this bug, I'm just not sure and I know there are some other reasons too leading to same or similar behaviour. So "potentially" is the keyword in your post above. :)

I couldn't notice any obvious culprit in the dmesg log. But then I'm just a hobbyist, not a pro.

Regards,
Varun

[Larry Finger (larry-finger-lwfinger)]

The dmesg says that authentication is failing. Are you sure the wifi "secret" is correct?

[[e-mail] (vworld4u-g)]

Hi Larry/Varunendra,

As the other member was mentioning, I see same behavior in my Laptop too.
If I reboot the system once, first time I can see that it connects to the
Wifi network - but browsing internet fails. It briefly stays in connected
state for few mins and then disconnects. I see same message like
"Authentication time out" afterwards. It never connects again.

I am seeing the same behavior on various kernels I tried 3.15, 3.16 and now
on 3.17 latest. I am not sure what might be the problem - but I am very
sure about the Wifi credentials - as I work on my personal laptop which is
E430 on the same network without any interruptions.

On Mon, Sep 29, 2014 at 9:10 PM, Larry Finger <email address hidden>
wrote:

> The dmesg says that authentication is failing. Are you sure the wifi
> "secret" is correct?
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> Status in HWE Next Project:
> Confirmed
> Status in HWE Next trusty series:
> Confirmed
> Status in “linux” package in Ubuntu:
> Confirmed
>
> Bug description:
> Open this bug to trace device driver RTL8192EE.
>
> We have seen this wireless card on some laptops, id is [10ec:818b],
> driver is RTL8192EE.
>
> Status: waiting for this driver being moved out of staging directory.
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions
>

--

*Thanks and Regards*
*--------------------------------------------------------------------------------------------------------------------*
*Venkatesh T R*
*9945040858*
A Man with Many Dimensions...

[Larry Finger (larry-finger-lwfinger)]

There will be a new RTL8192EE driver in kernel 3.18. If you want to try it earlier, you need to use the code mentioned in comment #184. If you are using kernel 3.13.0-36 from Ubuntu, you will have a build problem. See Comment #204 to see why I cannot fix that problem.

If you are using this code already, then post your issue at GitHub.com.

[Jan Van Hoecke (jan-vanhoecke)]

To report on the good news as well.
I tested the new RTL8192EE driver linked in comment #184 on the 20th Sept (after commit 5edf3e917dd11dbace503f25dde96a273ce4ca59) and had the same connection problems reported in comments #212 (and others before).

However I tested again yesterday (after commit 73fea9d157e08f9ea8d3736061a3b42a7cf6a219) and now everything works fine. The connection is stable, the speed as to be expected. Used it for 2 days now against different access points, all WPA secured.

T440s Lenovo laptop
Ubuntu 14.04.1 all updates applied
Kernel 3.16.2-031602-generic

Module loaded without specific config settings.

Not sure if any of the commits are relevant, but I figured I would report this anyway.

[[e-mail] (vworld4u-g)]

Hi Larry,

I am using 3.17 latest Kernel from rupotic. I am ready to use 3.18 kernel
but don't know really how to build it from source code. I will give it a
try sometime and let you know the status.

But I am not sure why the authentication times out after first successful
connect to Wifi. Anyway I am very thankful to your contribution here.

On Mon, Sep 29, 2014 at 10:01 PM, Larry Finger <email address hidden>
wrote:

> There will be a new RTL8192EE driver in kernel 3.18. If you want to try
> it earlier, you need to use the code mentioned in comment #184. If you
> are using kernel 3.13.0-36 from Ubuntu, you will have a build problem.
> See Comment #204 to see why I cannot fix that problem.
>
> If you are using this code already, then post your issue at GitHub.com.
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> Status in HWE Next Project:
> Confirmed
> Status in HWE Next trusty series:
> Confirmed
> Status in “linux” package in Ubuntu:
> Confirmed
>
> Bug description:
> Open this bug to trace device driver RTL8192EE.
>
> We have seen this wireless card on some laptops, id is [10ec:818b],
> driver is RTL8192EE.
>
> Status: waiting for this driver being moved out of staging directory.
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions
>

--

*Thanks and Regards*
*--------------------------------------------------------------------------------------------------------------------*
*Venkatesh T R*
*9945040858*
A Man with Many Dimensions...

[Larry Finger (larry-finger-lwfinger)]

Since 3.17 is not out yet, you will not find 3.18 anywhere - in about one month, 3.18-rc1 will be available at vger.kernel.org.

[Clive Cox (clivejcox)]

Just to report back larry, I'm still getting lockups which require a manual power cycle on my t440p with kernel 3.13.0-24-generic and built from your github.

The crashes are certainly happening less but still happening. Sometimes more intensive network usage seems to cause one, like using skype or streaming video.

[Xprogrammer (programmer-x)]

Step 1) I upgraded to 14.04.1 from 13.10 using apt-get upgrade, update-manager -d.

No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 14.04.1 LTS
Release: 14.04
Codename: trusty

Step 2) Pulled driver code from GIT and built. Loaded via modprobe and added to /etc/modules too.

Step 3) After reboot, everything seems to be fine except intermediate stuck (network connectivity drop) and becomes available itself.

Linux 440s-laptop 3.13.9 #2 SMP Tue May 20 09:57:03 CDT 2014 x86_64 x86_64 x86_64 GNU/Linux

15:13:58 up 22 min, 3 users, load average: 1.33, 1.49, 1.22

also, RTT delay is quite a high for pinging directly connected router IP.

wlan2 Link encap:Ethernet HWaddr
          inet addr:10.16.0.33 Bcast:10.16.255.255 Mask:255.255.0.0
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:104458 errors:0 dropped:0 overruns:0 frame:0
          TX packets:60287 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:132596751 (132.5 MB) TX bytes:8451011 (8.4 MB)

default via 10.16.0.1 dev wlan2 proto static
10.16.0.0/16 dev wlan2 proto kernel scope link src 10.16.0.33 metric 9

64 bytes from 10.16.0.33: icmp_seq=69 ttl=64 time=5.28 ms
64 bytes from 10.16.0.33: icmp_seq=70 ttl=64 time=5.25 ms
64 bytes from 10.16.0.33: icmp_seq=71 ttl=64 time=6.28 ms
64 bytes from 10.16.0.33: icmp_seq=75 ttl=64 time=6860 ms
64 bytes from 10.16.0.33: icmp_seq=76 ttl=64 time=5863 ms
64 bytes from 10.16.0.33: icmp_seq=77 ttl=64 time=4864 ms
64 bytes from 10.16.0.33: icmp_seq=78 ttl=64 time=3865 ms
64 bytes from 10.16.0.33: icmp_seq=79 ttl=64 time=2865 ms
64 bytes from 10.16.0.33: icmp_seq=80 ttl=64 time=1868 ms
64 bytes from 10.16.0.33: icmp_seq=81 ttl=64 time=870 ms
64 bytes from 10.16.0.33: icmp_seq=82 ttl=64 time=4988 ms
64 bytes from 10.16.0.33: icmp_seq=83 ttl=64 time=3986 ms
64 bytes from 10.16.0.33: icmp_seq=84 ttl=64 time=2979 ms
64 bytes from 10.16.0.33: icmp_seq=85 ttl=64 time=1972 ms
64 bytes from 10.16.0.33: icmp_seq=86 ttl=64 time=964 ms
64 bytes from 10.16.0.33: icmp_seq=87 ttl=64 time=5074 ms
64 bytes from 10.16.0.33: icmp_seq=88 ttl=64 time=4106 ms
64 bytes from 10.16.0.33: icmp_seq=89 ttl=64 time=3099 ms
64 bytes from 10.16.0.33: icmp_seq=90 ttl=64 time=2093 ms
64 bytes from 10.16.0.33: icmp_seq=91 ttl=64 time=1085 ms
64 bytes from 10.16.0.33: icmp_seq=92 ttl=64 time=78.0 ms

Is it due to arp getting timed out ?. I see ARP timeouts.

How do I debug futher. Any help is appreciated.

[varunendra (varunendra)]

Xprogrammer,

Could you give us the pastebin link to you full wireless_script report? : http://ubuntuforums.org/showpost.php?p=13024222

In case it is not directly related to the bug, an Ubuntu Forums thread (or IRC channel #ubuntu) may be a better place for further troubleshooting, if possible at all.

Regards,
Varun

[Xprogrammer (programmer-x)]

With 3.13.9 kernel, system freezes completely after 30minutes of uptime with 8192ee module loaded.

[Xprogrammer (programmer-x)]

@Varun,

Its very much related to this driver. Unfortunately I'm not able to get Call Trace from kernel as its completely frozen. Looks like, gets getting into deadlock. I've build kernel with lock debugs

-rw------- 1 root root 0 Oct 8 23:06 lock_stat
-r--r--r-- 1 root root 0 Oct 8 23:06 locks
-r-------- 1 root root 0 Oct 8 23:06 lockdep_stats
-r-------- 1 root root 0 Oct 8 23:06 lockdep_chains
-r-------- 1 root root 0 Oct 8 23:06 lockdep

But, I don't know how do i get get where is the possible problem is ?

[Giles Carré (giles-carre)]

Just FYI

For testing, I have upgraded my Lenovo T440s under Ubuntu 14.10 beta 2 one week ago. The native kernel is 3.16 (today 3.16.0-21).

Since the upgrade, WiFi networks are seen. But it's fully impossible to connect to anyone, even if it's a fully open network, ie without authentication protocol or dialog.

And sometime, the system freezes when RTL8192ee device is on, and never when off.

Now, I expect the release candidate (october 16), then the official release (october 23).

[stano (stanislav-simko)]

I've been having lot of freezes with fedora 20/21 kernels 3.15, 3.16, 3.17 (maybe sooner). In my opinion it is related to bug I (and others) have reported https://bugzilla.redhat.com/show_bug.cgi?id=1149632 . Yesterday I reinstalled fedora back to 3.11.10-301.fc20.x86_64 and it looks like when I disable bluetooth in bios then it is going without freezes (but network-manager + nm-applet are connecting very bad to even non secured wifi connections). With enabled bluetooth I had BUG: soft lockp - CPU#0 stuck for 22s! - this message and some other info was written to terminal but keyboard/touchpoint/trackpad were not functional, had to do hard reboot. My system is thinkpad t540p with latest bios.

[Larry Finger (larry-finger-lwfinger)]

If you are using the driver from staging, then you need to know that it has been deleted from kernel 3.18 and replaced by the code that is found at http://github.com/lwfinger/rtlwifi_new.git.

When the system has a stuck CPU, and "some other info was written to the terminal", debugging is impossible unless that other info is written down and posted with the problem.

[Xprogrammer (programmer-x)]

Oct 8 23:39:46 laptop-t440s kernel: [ 2990.480801] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1c
Oct 8 23:39:47 laptop-t440s kernel: [ 2990.787430] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1d
Oct 8 23:39:47 laptop-t440s kernel: [ 2990.787785] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1c
Oct 8 23:43:02 laptop-t440s kernel: [ 3186.366719] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1d
Oct 8 23:43:02 laptop-t440s kernel: [ 3186.367267] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1c
Oct 8 23:43:05 laptop-t440s kernel: [ 3188.719317] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1d
Oct 8 23:43:05 laptop-t440s kernel: [ 3188.719839] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1c
Oct 8 23:45:03 laptop-t440s kernel: [ 3307.453517] rtlwifi:rtl_watchdog_wq_callback():<0-0> AP off, try to reconnect now
Oct 8 23:45:03 laptop-t440s kernel: [ 3307.453885] wlan2: Connection to AP 5a:6d:8f:64:22:f0 lost
Oct 8 23:45:04 laptop-t440s kernel: [ 3307.527413] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1d
Oct 8 23:45:04 laptop-t440s kernel: [ 3307.527430] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1c
Oct 8 23:45:04 laptop-t440s kernel: [ 3307.558010] cfg80211: Calling CRDA to update world regulatory domain
Oct 8 23:45:04 laptop-t440s kernel: [ 3307.561806] cfg80211: World regulatory domain updated:
Oct 8 23:45:04 laptop-t440s kernel: [ 3307.561809] cfg80211: DFS Master region: unset
Oct 8 23:45:04 laptop-t440s kernel: [ 3307.561810] cfg80211: (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp), (dfs_cac_time)
Oct 8 23:45:04 laptop-t440s kernel: [ 3307.561812] cfg80211: (2402000 KHz - 2472000 KHz @ 40000 KHz), (300 mBi, 2000 mBm), (N/A)
Oct 8 23:45:04 laptop-t440s kernel: [ 3307.561814] cfg80211: (2457000 KHz - 2482000 KHz @ 40000 KHz), (300 mBi, 2000 mBm), (N/A)
Oct 8 23:45:04 laptop-t440s kernel: [ 3307.561815] cfg80211: (2474000 KHz - 2494000 KHz @ 20000 KHz), (300 mBi, 2000 mBm), (N/A)
Oct 8 23:45:04 laptop-t440s kernel: [ 3307.561817] cfg80211: (5170000 KHz - 5250000 KHz @ 40000 KHz), (300 mBi, 2000 mBm), (N/A)
Oct 8 23:45:04 laptop-t440s kernel: [ 3307.561818] cfg80211: (5735000 KHz - 5835000 KHz @ 40000 KHz), (300 mBi, 2000 mBm), (N/A)
Oct 8 23:45:16 laptop-t440s kernel: [ 3319.846784] Watchdog[3270]: segfault at 0 ip 00007f9148a0e1ab sp 00007f913860d7d0 error 6 in chrome[7f9144c74000+4ffd000]
Oct 8 23:45:26 laptop-t440s kernel: [ 3330.252419] Watchdog[5737]: segfault at 0 ip 00007f3aa1fc11ab sp 00007f3a91bc07d0 error 6 in chrome[7f3a9e227000+4ffd000]
Oct 8 23:50:24 laptop-t440s kernel: [ 0.000000] Initializing cgroup subsys cpuset
Oct 8 23:50:24 laptop-t440s kernel: [ 0.000000] Initializing cgroup subsys cpu
Oct 8 23:50:24 laptop-t440s kernel: [ 0.000000] Initializing cgroup subsys cpuacct
Oct 8 23:50:24 laptop-t440s kernel: [ 0.000000] Linux version 3.16.0-tecmintkernel (root@laptop-t440s) (gcc version 4.8.2 (Ubuntu 4.8.2-19ubuntu1) ) #2 SMP Wed Oct 8 20:09:58 PDT 2014
Oct 8 23:50:24 laptop-t440s kernel: [ 0.000000] Command line: BOO...

[Larry Finger (larry-finger-lwfinger)]

There is nothing wrong in the log that you posted. Those switch case not processed warnings are not a problem.

I think you are using the Git repo. Which branch are you using? What does 'git log' say is the latest commit.

The information I need will appear on the console log. It can possible be seen by using CTRL-ALT-F10; however, that info will not be logged, and that screen must be showing when the condition happens.

I have run the new driver for at least a week with no crashes. As I am unable to duplicate the problem with my hardware, someone with the problem will need to supply additional information.

[Xprogrammer (programmer-x)]

[laptop-t440s]09:27:25-user~/Downloads/rtlwifi_new$ git status .
On branch master
Your branch is up-to-date with 'origin/master'.

Untracked files:
  (use "git add <file>..." to include in what will be committed)

        backup_drivers.tar

nothing added to commit but untracked files present (use "git add" to track)
[laptop-t440s]09:27:27-user~/Downloads/rtlwifi_new$ git pull
Already up-to-date.
[laptop-t440s]09:27:37-user~/Downloads/rtlwifi_new$

[Xprogrammer (programmer-x)]

commit 5068371ab982090576c863204b96adc1d03ed462
Author: Larry Finger <email address hidden>
Date: Thu Oct 2 17:11:29 2014 -0500

    rtlwifi_new: Modify Makefile to save drivers before installing new ones

    Whe uninstalling the drivers, the saved versions are restored. Firmware is
    not touched when uninstalling.

    Signed-off-by: Larry Finger <email address hidden>

commit 150455d556ba6f5167e437b72fcd88693f8086a0
Author: Larry Finger <email address hidden>
Date: Thu Oct 2 15:13:17 2014 -0500

    rtlwifi_new: Fix all drivers to build correctly on Ubuntu 3.13

    In kernel 3.13.0-XX where XX is >= 32, Ubuntu has implemented the API
    change that was introduced in mainline V3.15 by commit d8ca16db6bb2 ("mac80211:
    add length check in ieee80211_is_robust_mgmt_frame()"). The result is that the
    code in this repo gets it wrong for Ubuntu kernels. The result is
    a kernel that crashes when using the corresponging driver. Thanks to user
    "lenonk" at GitHub, there is a fix. I have not tested it as I do not use one
    of the affected kernels; however, it looks correct.

    Signed-off-by: Larry Finger <email address hidden>

[Xprogrammer (programmer-x)]

I've turned off bluetooth and loaded kernel build out of 3.16 release as per link (http://www.tecmint.com/install-kernel-3-16-in-debian/) with some more lock debugging.

System is up for 10 hours and i'm using it. So far, It has got 200MB of traffic (in + out).

[laptop-t440s]10:24:36-user~$
[laptop-t440s]10:24:36-user~$ uptime
 10:24:37 up 10:34, 10 users, load average: 0.77, 0.86, 0.65
[laptop-t440s]10:24:37-user~$
[laptop-t440s]10:24:39-user~$
[laptop-t440s]10:24:40-user~$ uname -a
Linux laptop-t440s 3.16.0-tecmintkernel #2 SMP Wed Oct 8 20:09:58 PDT 2014 x86_64 x86_64 x86_64 GNU/Linux
[laptop-t440s]10:25:49-user~$ ifconfig wlan2
wlan2 Link encap:Ethernet HWaddr 40:f0:2f:7b:66:b7
          inet addr:10.16.0.33 Bcast:10.16.255.255 Mask:255.255.0.0
          inet6 addr: fe80::42f0:2fff:fe7b:66b7/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:168311 errors:0 dropped:0 overruns:0 frame:0
          TX packets:116645 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:162567775 (162.5 MB) TX bytes:40257105 (40.2 MB)

[laptop-t440s]10:25:52-user~$

[Xprogrammer (programmer-x)]

Unfortunately, right after previous post, my system went to hang. I can't even go to any terminal. Ctrl + Alt + F1 to F12, nothing worked. Its completely unresponsive. Only option I had to power off manually and poweron in hard way by pressing power button.

Is there place where I can look for logs? Whats going on ? I dont have console/serial port on my laptop. Is there way to emulate soft serial port and make sure kernel responsive.

[Larry Finger (larry-finger-lwfinger)]

Unfortunately, when you need to power off the system in this way, information being written to the logs is lost.

Whenever there is a particular sequence that causes a crash, I usually cause a terminal to sleep for 5 secs, and then have it do the sequence - all in a single line separated by a semicolon. Once the sleep is started, I switch to the logging console. On my system, that is selected with CTRL-ALT-F10. The graphics console is at CTRL-ALT-F7. Once the crash happens, I either write down the crucial info, or I take a photo of the screen.

If you have a second system, you might also look at the network console method of capturing dump info.

[Xprogrammer (programmer-x)]

Larry,

Is there way to test this driver under virtualbox or EXSI based VM? I could give it a try in that case.

[Xprogrammer (programmer-x)]

I could bringup Ubuntu 14.04 under virtualbox and add PCIe (realtek) to VM using pciattach. Its visiable under Guest OS and after adding rtl8192ee driver, its able to find PCIe and brings up wlan0 interface. Its says, Wlan0 link is not ready.

[Xprogrammer (programmer-x)]

wlan0 Link encap:Ethernet HWaddr 40:f0:2f:7b:66:b7
          UP BROADCAST MULTICAST MTU:1500 Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)

[Xprogrammer (programmer-x)]

01:05.0 Network controller: Realtek Semiconductor Co., Ltd. RTL8192EE PCIe Wireless Network Adapter
 Subsystem: Realtek Semiconductor Co., Ltd. Device 001b
 Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
 Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort+ <TAbort- <MAbort- >SERR- <PERR- INTx-
 Interrupt: pin A routed to IRQ 17
 Region 0: I/O ports at e000 [size=256]
 Region 2: Memory at f0900000 (64-bit, non-prefetchable) [size=16K]
 Capabilities: [40] Power Management version 3
  Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=375mA PME(D0+,D1+,D2+,D3hot+,D3cold+)
  Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=0 PME-
 Capabilities: [70] Express (v2) Endpoint, MSI 00
  DevCap: MaxPayload 128 bytes, PhantFunc 0, Latency L0s <4us, L1 <64us
   ExtTag- AttnBtn- AttnInd- PwrInd- RBE+ FLReset-
  DevCtl: Report errors: Correctable- Non-Fatal- Fatal- Unsupported-
   RlxdOrd+ ExtTag- PhantFunc- AuxPwr- NoSnoop-
   MaxPayload 128 bytes, MaxReadReq 512 bytes
  DevSta: CorrErr+ UncorrErr- FatalErr- UnsuppReq+ AuxPwr+ TransPend-
  LnkCap: Port #0, Speed 2.5GT/s, Width x1, ASPM L0s L1, Exit Latency L0s <512ns, L1 <64us
   ClockPM+ Surprise- LLActRep- BwNot-
  LnkCtl: ASPM L0s L1 Enabled; RCB 64 bytes Disabled- CommClk+
   ExtSynch- ClockPM+ AutWidDis- BWInt- AutBWInt-
  LnkSta: Speed 2.5GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
  DevCap2: Completion Timeout: Not Supported, TimeoutDis+, LTR+, OBFF Via message/WAKE#
  DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis+, LTR+, OBFF Disabled
  LnkSta2: Current De-emphasis Level: -6dB, EqualizationComplete-, EqualizationPhase1-
    EqualizationPhase2-, EqualizationPhase3-, LinkEqualizationRequest-
 Capabilities: [100 v2] Advanced Error Reporting
  UESta: DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol-
  UEMsk: DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol-
  UESvrt: DLP+ SDES+ TLP- FCP+ CmpltTO- CmpltAbrt- UnxCmplt- RxOF+ MalfTLP+ ECRC- UnsupReq- ACSViol-
  CESta: RxErr+ BadTLP- BadDLLP- Rollover- Timeout+ NonFatalErr+
  CEMsk: RxErr- BadTLP- BadDLLP- Rollover- Timeout- NonFatalErr+
  AERCap: First Error Pointer: 00, GenCap+ CGenEn- ChkCap+ ChkEn-
 Capabilities: [140 v1] Device Serial Number 01-91-81-fe-ff-4c-e0-00
 Capabilities: [150 v1] Latency Tolerance Reporting
  Max snoop latency: 3145728ns
  Max no snoop latency: 3145728ns
 Capabilities: [158 v1] L1 PM Substates
  L1SubCap: PCI-PM_L1.2+ PCI-PM_L1.1+ ASPM_L1.2+ ASPM_L1.1+ L1_PM_Substates+
     PortCommonModeRestoreTime=150us PortTPowerOnTime=150us
 Kernel driver in use: rtl8192ee

skannan@skannan-vb:~$ iwconfig
eth0 no wireless extensions.

lo no wireless extensions.

wlan0 IEEE 802.11bgn ESSID:off/any
          Mode:Managed Access Point: Not-Associated Tx-Power=20 dBm
          Retry long limit:7 RTS thr=2347 B Fragment thr:off
          Power Management:off

skannan@skannan-vb:~$ ifconfig
eth0 Link encap:Ethernet HWaddr 08:00:27:16:75:fd
          inet addr:10.0.2.15 Bcast:10.0.2.255 Mask:255.255.255.0
          inet6 addr: fe80::...

[Xprogrammer (programmer-x)]

Larry,

I'm trying to see why link state is not coming up in virtualbox loaded kernel. I don't see call to netif_carrier_on upon "sudo modprobe rtl8192ee". Can you help how is this flow works? For RTL8192EE driver, how is netif_carrier_on being called and stack trace of that would help me to see why its happening. ?

[Swagat Kumar (swagat-kumar)]

System: Lenovo X240 Thinkpad
OS: Ubuntu 14.04.1 64 bit LTS
Kernel: 3.13.0-37-generic #64-Ubuntu SMP (after update)

When I installed Ubuntu 14.04.1, the network was not configured. So I was left with a system without wifi and ethernet. I downloaded the rtlwifi_new driver from the git repository posted in comment #184.

This compiles properly without any error. I install it by running "sudo make install" inside the driver folder.

$ sudo modprobe rtl8192ee
$ iwconfig
$ lshw -C "Network"

 sudo lshw -C "Network"
[sudo] password for swagat:
  *-network
       description: Ethernet interface
       product: Ethernet Connection I218-LM
       vendor: Intel Corporation
       physical id: 19
       bus info: pci@0000:00:19.0
       logical name: eth0
       version: 04
       serial: 28:d2:44:97:ec:b7
       capacity: 1Gbit/s
       width: 32 bits
       clock: 33MHz
       capabilities: pm msi bus_master cap_list ethernet physical tp 10bt 10bt-fd 100bt 100bt-fd 1000bt-fd autonegotiation
       configuration: autonegotiation=on broadcast=yes driver=e1000e driverversion=2.3.2-k firmware=0.6-3 latency=0 link=no multicast=yes port=twisted pair
       resources: irq:59 memory:f0600000-f061ffff memory:f063e000-f063efff ioport:4080(size=32)
  *-network
       description: Wireless interface
       product: RTL8192EE PCIe Wireless Network Adapter
       vendor: Realtek Semiconductor Co., Ltd.
       physical id: 0
       bus info: pci@0000:03:00.0
       logical name: wlan1
       version: 00
       serial: 18:cf:5e:bb:8c:43
       width: 64 bits
       clock: 33MHz
       capabilities: pm msi pciexpress bus_master cap_list ethernet physical wireless
       configuration: broadcast=yes driver=rtl8192ee driverversion=3.13.0-37-generic firmware=N/A ip=192.168.1.105 latency=0 link=yes multicast=yes wireless=IEEE 802.11bgn
       resources: irq:62 ioport:3000(size=256) memory:f0400000-f0403fff

$ sudo ifconfig -a

sudo ifconfig -a
eth0 Link encap:Ethernet HWaddr 28:d2:44:97:ec:b7
          UP BROADCAST MULTICAST MTU:1500 Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
          Interrupt:20 Memory:f0600000-f0620000

eth0:avahi Link encap:Ethernet HWaddr 28:d2:44:97:ec:b7
          inet addr:169.254.11.99 Bcast:169.254.255.255 Mask:255.255.0.0
          UP BROADCAST MULTICAST MTU:1500 Metric:1
          Interrupt:20 Memory:f0600000-f0620000

lo Link encap:Local Loopback
          inet addr:127.0.0.1 Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING MTU:65536 Metric:1
          RX packets:2265 errors:0 dropped:0 overruns:0 frame:0
          TX packets:2265 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:171501 (171.5 KB) TX bytes:171501 (171.5 KB)

wlan1 Link encap:Ethernet HWaddr 18:cf:5e:bb:8c:43
          inet addr:192.168.1.125 Bcast:192.168.1.255 Mask:255.255.255.0
          inet6 addr: fe80::1acf:5eff:febb:8c43/64 Scope:Link
          UP BROA...

[Xprogrammer (programmer-x)]

I tried with kernel 3.17.0-031700-generic Ubuntu 14.04.1

Things seems to be OK so far.

user1@user-laptop:~$ uname -r
3.17.0-031700-generic
user1@user-laptop:~$ uptime
 10:35:15 up 1:51, 3 users, load average: 0.52, 0.41, 0.34
user1@user-laptop:~$ ifconfig wlan2
wlan2 Link encap:Ethernet HWaddr 40:f0:2f:7b:66:b7
   inet addr:10.16.0.33 Bcast:10.16.255.255 Mask:255.255.0.0
   inet6 addr: fe80::42f0:2fff:fe7b:66b7/64 Scope:Link
   UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
   RX packets:353191 errors:0 dropped:0 overruns:0 frame:0
   TX packets:250879 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:1000
   RX bytes:417449156 (417.4 MB) TX bytes:154375572 (154.3 MB)

user1@user-laptop:~$

[Xprogrammer (programmer-x)]

System frozen after sometime.

here is the log...

Oct 13 13:13:28 skannan-dv2k kernel: [ 941.277351] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1d
Oct 13 13:13:28 skannan-dv2k kernel: [ 941.277713] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1c
Oct 13 13:13:28 skannan-dv2k kernel: [ 941.296409] cfg80211: Calling CRDA to update world regulatory domain
Oct 13 13:13:28 skannan-dv2k kernel: [ 941.298434] cfg80211: World regulatory domain updated:
Oct 13 13:13:28 skannan-dv2k kernel: [ 941.298437] cfg80211: DFS Master region: unset
Oct 13 13:13:28 skannan-dv2k kernel: [ 941.298438] cfg80211: (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp), (dfs_cac_time)
Oct 13 13:13:28 skannan-dv2k kernel: [ 941.298440] cfg80211: (2402000 KHz - 2472000 KHz @ 40000 KHz), (300 mBi, 2000 mBm), (N/A)
Oct 13 13:13:28 skannan-dv2k kernel: [ 941.298441] cfg80211: (2457000 KHz - 2482000 KHz @ 40000 KHz), (300 mBi, 2000 mBm), (N/A)
Oct 13 13:13:28 skannan-dv2k kernel: [ 941.298442] cfg80211: (2474000 KHz - 2494000 KHz @ 20000 KHz), (300 mBi, 2000 mBm), (N/A)
Oct 13 13:13:28 skannan-dv2k kernel: [ 941.298443] cfg80211: (5170000 KHz - 5250000 KHz @ 40000 KHz), (300 mBi, 2000 mBm), (N/A)
Oct 13 13:13:28 skannan-dv2k kernel: [ 941.298445] cfg80211: (5735000 KHz - 5835000 KHz @ 40000 KHz), (300 mBi, 2000 mBm), (N/A)
Oct 13 13:13:29 skannan-dv2k kernel: [ 942.333855] wlan2: authenticate with c8:d7:19:1e:ff:2f
Oct 13 13:13:29 skannan-dv2k kernel: [ 942.370407] wlan2: send auth to c8:d7:19:1e:ff:2f (try 1/3)
Oct 13 13:13:29 skannan-dv2k kernel: [ 942.472813] wlan2: send auth to c8:d7:19:1e:ff:2f (try 2/3)
Oct 13 13:13:29 skannan-dv2k kernel: [ 942.498062] wlan2: authenticated
Oct 13 13:13:29 skannan-dv2k kernel: [ 942.500797] wlan2: associate with c8:d7:19:1e:ff:2f (try 1/3)
Oct 13 13:13:29 skannan-dv2k kernel: [ 942.604736] wlan2: associate with c8:d7:19:1e:ff:2f (try 2/3)
Oct 13 13:13:30 skannan-dv2k kernel: [ 942.708626] wlan2: associate with c8:d7:19:1e:ff:2f (try 3/3)
Oct 13 13:13:30 skannan-dv2k kernel: [ 942.812629] wlan2: association with c8:d7:19:1e:ff:2f timed out
Oct 13 13:13:39 skannan-dv2k kernel: [ 952.599059] wlan2: authenticate with c8:d7:19:1e:ff:2f
Oct 13 13:13:39 skannan-dv2k kernel: [ 952.618713] wlan2: send auth to c8:d7:19:1e:ff:2f (try 1/3)
Oct 13 13:13:39 skannan-dv2k kernel: [ 952.624047] wlan2: authenticated
Oct 13 13:13:39 skannan-dv2k kernel: [ 952.627093] wlan2: associate with c8:d7:19:1e:ff:2f (try 1/3)
Oct 13 13:13:40 skannan-dv2k kernel: [ 952.651128] wlan2: RX AssocResp from c8:d7:19:1e:ff:2f (capab=0x411 status=0 aid=60)
Oct 13 13:13:40 skannan-dv2k kernel: [ 952.904540] wlan2: associated
Oct 13 13:16:50 skannan-dv2k kernel: [ 1143.366569] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1d
Oct 13 13:16:50 skannan-dv2k kernel: [ 1143.366916] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1c
Oct 13 13:17:17 skannan-dv2k kernel: [ 1169.718307] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1d
Oct 13 13:17:17 skannan-dv2k kernel: [ 1169.718640] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1c
Oct 13 1...

[Xprogrammer (programmer-x)]

Frozen again..

Portion of kern.log attached.

[Larry Finger (larry-finger-lwfinger)]

Your log makes it look as if there is an error in the pairing of rcu_read_lock()/rcu_read_unlock() calls, but everything looks OK. It won't help your problem, but I just pushed a change that gets rid of an error log splat on my system.

[Xprogrammer (programmer-x)]

Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384776] ------------[ cut here ]------------
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384809] WARNING: CPU: 0 PID: 3994 at /home/apw/COD/linux/drivers/gpu/drm/i915/intel_display.c:3361 intel_crtc_wait_for_pending_flips+0x11a/0x150 [i915]()
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384811] Modules linked in: ppp_deflate bsd_comp ppp_async crc_ccitt ctr ccm arc4 rtl8192ee(OE) btcoexist(OE) rtl_pci(OE) rtlwifi(OE) mac80211 cfg80211 pci_stub vboxpci(OE) vboxnetadp(OE) vboxnetflt(OE) vboxdrv(OE) rfcomm bnep binfmt_misc nls_iso8859_1 snd_hda_codec_hdmi intel_rapl x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_core v4l2_common videodev media crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel aes_x86_64 btusb lrw bluetooth gf128mul glue_helper snd_hda_codec_realtek i915 snd_hda_codec_generic ablk_helper snd_hda_intel snd_hda_controller cryptd snd_hda_codec snd_hwdep snd_pcm drm_kms_helper snd_seq_midi drm snd_seq_midi_event snd_rawmidi joydev snd_seq thinkpad_acpi serio_raw snd_seq_device nvram i2c_algo_bit rtsx_pci_ms rtc_efi snd_timer memstick lpc_ich mei_me snd mei soundcore video mac_hid wmi intel_smartconnect parport_pc ppdev lp parport rtsx_pci_sdmmc psmouse ahci libahci rtsx_pci e1000e ptp pps_core
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384853] CPU: 0 PID: 3994 Comm: Xorg Tainted: G OE 3.17.0-031700-generic #201410060605
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384854] Hardware name: LENOVO 20AQCTO1WW/20AQCTO1WW, BIOS GJET67WW (2.17 ) 12/10/2013
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384856] 0000000000000d21 ffff880212b13b28 ffffffff82796bd7 0000000000000007
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384858] 0000000000000000 ffff880212b13b68 ffffffff82074a3c 0000000000000202
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384860] 0000000000000000 ffff880212009000 ffff880212ae82c0 ffff8800d66c4800
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384862] Call Trace:
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384868] [<ffffffff82796bd7>] dump_stack+0x46/0x58
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384871] [<ffffffff82074a3c>] warn_slowpath_common+0x8c/0xc0
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384874] [<ffffffff82074a8a>] warn_slowpath_null+0x1a/0x20
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384889] [<ffffffffc06a91ca>] intel_crtc_wait_for_pending_flips+0x11a/0x150 [i915]
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384894] [<ffffffff820b3eb0>] ? add_wait_queue+0x60/0x60
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384908] [<ffffffffc06ad1d3>] intel_crtc_disable_planes+0x33/0x160 [i915]
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384910] [<ffffffff820ab13c>] ? check_preempt_wakeup+0x12c/0x210
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384923] [<ffffffffc06ad865>] haswell_crtc_disable+0x55/0x220 [i915]
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384936] [<ffffffffc06aa2f7>] intel_crtc_control+0x57/0x130 [i915]
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384948] [<ffffffffc06aa437>] intel_crtc_update_dpms+0x67/0x80 [i915]
Oct 13 22:56:27 skannan-dv2k kernel: [ 759.384960] [<ffffff...

[Xprogrammer (programmer-x)]

Oct 13 22:49:44 skannan-dv2k kernel: [ 355.976753] wlan2: RX AssocResp from 58:6d:8f:64:22:ff (capab=0x411 status=0 aid=2)
Oct 13 22:49:44 skannan-dv2k kernel: [ 356.016768] wlan2: associated
Oct 13 22:49:44 skannan-dv2k kernel: [ 356.016782] IPv6: ADDRCONF(NETDEV_CHANGE): wlan2: link becomes ready
Oct 13 22:49:44 skannan-dv2k kernel: [ 356.075909] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1d
Oct 13 22:49:44 skannan-dv2k kernel: [ 356.076265] rtl8192ee:rtl92ee_set_hw_reg():<0-0> switch case not process 1c
Oct 13 22:51:10 skannan-dv2k kernel: [ 442.669043] PPP BSD Compression module registered
Oct 13 22:51:10 skannan-dv2k kernel: [ 442.676552] PPP Deflate Compression module registered
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839328] INFO: rcu_sched self-detected stall on CPU { 2} (t=15000 jiffies g=26277 c=26276 q=0)
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839336] Task dump for CPU 2:
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839339] swapper/2 R running task 0 0 1 0x00000008
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839341] ffffffff82c54240 ffff88021f283bb0 ffffffff820a05b8 0000000000000002
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839344] ffffffff82c54240 ffff88021f283bd0 ffffffff820a2edf ffff88021f283c10
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839346] ffffffff82c54240 ffff88021f283c00 ffffffff820d307b ffffffff82c54240
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839348] Call Trace:
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839350] <IRQ> [<ffffffff820a05b8>] sched_show_task+0xa8/0x110
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839359] [<ffffffff820a2edf>] dump_cpu_task+0x3f/0x50
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839362] [<ffffffff820d307b>] rcu_dump_cpu_stacks+0x8b/0xc0
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839364] [<ffffffff820d4d47>] print_cpu_stall+0xd7/0x140
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839367] [<ffffffff820d4e55>] check_cpu_stall.isra.50+0xa5/0xb0
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839369] [<ffffffff820d4e98>] __rcu_pending+0x38/0x1d0
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839371] [<ffffffff820d6d0d>] rcu_check_callbacks+0xed/0x1a0
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839373] [<ffffffff820dc278>] update_process_times+0x48/0x80
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839377] [<ffffffff820ebf53>] tick_sched_handle.isra.14+0x33/0x70
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839380] [<ffffffff820ec07c>] tick_sched_timer+0x4c/0x80
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839382] [<ffffffff820dcd36>] __run_hrtimer+0x76/0x230
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839384] [<ffffffff820ec030>] ? tick_nohz_handler+0xa0/0xa0
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839386] [<ffffffff820dd53f>] hrtimer_interrupt+0xff/0x260
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839390] [<ffffffff8204b2eb>] local_apic_timer_interrupt+0x3b/0x60
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839394] [<ffffffff827a7635>] smp_apic_timer_interrupt+0x45/0x60
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839397] [<ffffffff827a571d>] apic_timer_interrupt+0x6d/0x80
Oct 13 22:54:29 skannan-dv2k kernel: [ 641.839400] [<...

[Xprogrammer (programmer-x)]

INFO: task kworker/u16:1:3631 blocked for more than 120 seconds.
      Tainted: G OE 3.17.0-031700-generic #201410060605
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
kworker/u16:1 D 0000000000000001 0 3631 2 0x00000000
Workqueue: phy0 ieee80211_ba_session_work [mac80211]
 ffff8801c901fb98 0000000000000046 ffff8801c901fb48 ffff88021f254580
 ffff8801c901ffd8 0000000000014580 ffff8801c9295080 0000000000014580
 ffff880213a51400 ffff880210fd1e00 0000000a00000007 ffff8801c901fcd8
Call Trace:
 [<ffffffff827a0279>] schedule+0x29/0x70
 [<ffffffff827a3275>] schedule_timeout+0x1b5/0x210
 [<ffffffffc0ab265c>] ? rtl92ee_phy_set_bb_reg+0x9c/0x160 [rtl8192ee]
 [<ffffffffc0ab2969>] ? rtl92ee_phy_set_rf_reg+0xf9/0x220 [rtl8192ee]
 [<ffffffff827a15b7>] wait_for_completion+0xa7/0x160
 [<ffffffff820a1030>] ? try_to_wake_up+0x290/0x290
 [<ffffffff820d59d0>] ? call_rcu_bh+0x20/0x20
 [<ffffffff820d1b2d>] wait_rcu_gp+0x4d/0x60
 [<ffffffff820d1b40>] ? wait_rcu_gp+0x60/0x60
 [<ffffffff820d475a>] synchronize_sched+0x3a/0x50
 [<ffffffff8268409c>] synchronize_net+0x1c/0x40
 [<ffffffffc0ad657e>] ___ieee80211_stop_tx_ba_session+0x10e/0x510 [mac80211]
 [<ffffffffc0ad5070>] ieee80211_ba_session_work+0x190/0x1a0 [mac80211]
 [<ffffffff8208ce2e>] process_one_work+0x14e/0x460
 [<ffffffff8208d7ab>] worker_thread+0x11b/0x3f0
 [<ffffffff8208d690>] ? create_worker+0x1e0/0x1e0
 [<ffffffff820932b9>] kthread+0xc9/0xe0
 [<ffffffff820931f0>] ? flush_kthread_worker+0x90/0x90
 [<ffffffff827a467c>] ret_from_fork+0x7c/0xb0
 [<ffffffff820931f0>] ? flush_kthread_worker+0x90/0x90
------------[ cut here ]------------

[Xprogrammer (programmer-x)]

For some reason, RTL8192EE loses its connection to AP even though things looks normal from network point of it. After consecutive attempt to reconnect causing, system freeze. Can it be ?.

IPv6: ADDRCONF(NETDEV_CHANGE): wlan2: link becomes ready
show_signal_msg: 159 callbacks suppressed
Watchdog[3538]: segfault at 0 ip 00007f4e7cace1ab sp 00007f4e6c6cd7d0 error 6 in chrome[7f4e78d34000+4ffd000]
rtlwifi:rtl_watchdog_wq_callback():<0-0> AP off, try to reconnect now
wlan2: Connection to AP c8:d7:19:1e:ff:2f lost
rtlwifi:rtl_watchdog_wq_callback():<0-0> AP off, try to reconnect now
Watchdog[4724]: segfault at 0 ip 00007f058e70d1ab sp 00007f057e30c7d0 error 6 in chrome[7f058a973000+4ffd000]
rtlwifi:rtl_watchdog_wq_callback():<0-0> AP off, try to reconnect now
rtlwifi:rtl_watchdog_wq_callback():<0-0> AP off, try to reconnect now
rtlwifi:rtl_watchdog_wq_callback():<0-0> AP off, try to reconnect now
rtlwifi:rtl_watchdog_wq_callback():<0-0> AP off, try to reconnect now

Thats where my systems goes to freeze. Hard power off/on is only way to get it back.

Initializing cgroup subsys cpuset
Initializing cgroup subsys cpu
Initializing cgroup subsys cpuacct
Linux version 3.17.0-031700-generic (apw@gomeisa) (gcc version 4.6.3 (Ubuntu/Linaro 4.6.3-1ubuntu5) ) #201410060605 SMP

[stano (stanislav-simko)]

I've been running fedora 20 with kernel 3.18.0-0.rc0.git4.1 without problem when today... 4-5 crashes :( (I only installed wget, then suspended and after that I had several freezes). Now I'm running 3.11.10-301.fc20.x86_64 with driver from git (not sure which version, I tried several versions non of which compiled - I had to install newer kernel to be able to compile them and then install. It could be one of oem-wireless-rtl-92ce-92se-92de-8723ae-88ee-8723be-92ee-dkms-0017.1016.2013~sutton1 or rtlwifi_new-master, probably the first one), unfortunately the driver is at least on some networks unusable - around 80% package loss.
during freezes I observed in xfce4-power-manager 1.2.0 that there are ~33000 wakeups on IRQ31 with command=interrupt, details=pci-msi-edge rtl_pci... Turning off bluetooth, ultrabay... didn't help.
If I can help somehow I would be very glad, because this makes my network card pretty useless and if there won't be solution soon I will have to throw it away probably...

[Xprogrammer (programmer-x)]

@ stano,

I was trying to debug in my system but complete freeze takes away entire system to be only power off/on.

If you have system which supports IOMMU, you can boot this version of kernel and PCI passthrough to see where is the freeze. that will be really useful to find out.

[Xprogrammer (programmer-x)]

@ stano,

I meant, boot this version of kernel in Virtualbox, and using IOMMU , PCI Passthrough see if things goes normal and freeze after a while.

[stano (stanislav-simko)]

unfortunatelly, I have i7 4700mq which does not support VT-d :(

[stano (stanislav-simko)]

Ok, IRQ interrupts also happening on 3.11.10-301.fc20.x86_64 with new versions of driver downloaded from https://github.com/lwfinger/rtlwifi_new (master & troy, didnt try kernel version explycitly, but it should be probably at 3.18 kernel which also had the problem). Version downloaded (probably) from here http://netbook-remix.archive.canonical.com/updates/pool/public/o/oem-wireless-rtl-92ce-92se-92de-8723ae-88ee-8723be-92ee-dkms/ does not causes IRQ explosion, but wifi drops 80% and more packets, making it pretty unusable.
If you have any suggestions that I can try I will be happy to help. @Xprogrammer - if you can navigate me how to run virtualbox without VT-d (if it is possible) I will try it (i'm lets say little advanced beginer in linux...).
Also, I tried to limit only 1 cpu core to respond to IRQs sent by realtek driver by
echo 0 > /proc/irq/48/smp_affinity
didn't help, system also froze with more than 33000 IRQ interrupts per second.
If it is usefull, I can provide log from /var/log/messages but I couldn't find any info that seemed to be important...
this is really .... situation as probably every wifi card usable (whitelisted) in this laptop has some kind of problems...

also, in git versions of wifi.h should be changed line (~2455)
#if (LINUX_VERSION_CODE <= KERNEL_VERSION(3,11,0))
to maybe 3,12 (in 3.11.10-301.fc20.x86_64 aren't those variables defined, I had to change it to .12 to be able to compile drivers)

[Xprogrammer (programmer-x)]

@Stano,

I tried in my system(which doesnt have IOMMU & VT-D), but wifi itself wont be up in virtualbox. Its not much useful.

[Xprogrammer (programmer-x)]

I assigned IRQ 48 (RTL_PCI) to CPU 3 via "echo 8 > /proc/irq/48/smp_affinity"

Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800829] NMI watchdog: BUG: soft lockup - CPU#3 stuck for 23s! [swapper/3:0]
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800833] Modules linked in: ctr ccm arc4 rtl8192ee(OE) btcoexist(OE) rtl_pci(OE) rtlwifi(OE) mac80211 xt_addrtype xt_conntrack ipt_MASQUERADE iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack bridge stp llc dm_thin_pool dm_persistent_data dm_bio_prison dm_bufio libcrc32c iptable_filter ip_tables x_tables cfg80211 8812au(OE) pci_stub vboxpci(OE) vboxnetadp(OE) vboxnetflt(OE) vboxdrv(OE) bnep rfcomm binfmt_misc nls_iso8859_1 snd_hda_codec_hdmi intel_rapl x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm btusb uvcvideo videobuf2_vmalloc videobuf2_memops bluetooth videobuf2_core v4l2_common videodev media snd_hda_codec_realtek snd_hda_codec_generic crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel aes_x86_64 lrw gf128mul glue_helper ablk_helper i915 cryptd snd_hda_intel snd_hda_controller ndiswrapper(OE) joydev snd_hda_codec drm_kms_helper serio_raw snd_hwdep mei_me snd_pcm drm rtc_efi snd_seq_midi snd_seq_midi_event snd_rawmidi rtsx_pci_ms thinkpad_acpi mei snd_seq memstick snd_seq_device nvram wmi snd_timer snd i2c_algo_bit lpc_ich video soundcore intel_smartconnect mac_hid parport_pc ppdev lp parport rtsx_pci_sdmmc psmouse e1000e ahci libahci rtsx_pci ptp pps_core
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800899] CPU: 3 PID: 0 Comm: swapper/3 Tainted: P W OEL 3.17.0-031700-generic #201410060605
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800900] Hardware name: LENOVO 20AQCTO1WW/20AQCTO1WW, BIOS GJET67WW (2.17 ) 12/10/2013
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800901] task: ffff880213a52800 ti: ffff880213a68000 task.ti: ffff880213a68000
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800914] RIP: 0010:[<ffffffff8207864b>] [<ffffffff8207864b>] __do_softirq+0x8b/0x330
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800919] RSP: 0018:ffff88021f2c3f08 EFLAGS: 00000206
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800921] RAX: 0000000000202040 RBX: ffff88021f2c3e98 RCX: 00000000000006e0
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800922] RDX: ffff880213a6bfd8 RSI: 00000000b1af924b RDI: ffff880213a52800
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800923] RBP: ffff88021f2c3f68 R08: ffff88006cd3f8e8 R09: 0000000000000001
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800924] R10: 0000000000000002 R11: 0000000000000000 R12: ffff88021f2c3e78
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800925] R13: ffffffff827a571d R14: ffff88021f2c3f68 R15: 0000000000000000
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800926] FS: 0000000000000000(0000) GS:ffff88021f2c0000(0000) knlGS:0000000000000000
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800928] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800929] CR2: 0000080a27068760 CR3: 0000000002c16000 CR4: 00000000001407e0
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800930] Stack:
Oct 21 21:00:04 skannan-dv2k kernel: [ 2630.800931] 00202040f968ae00 ffff880213a6bfd8 000000010006ddf2...

[Larry Finger (larry-finger-lwfinger)]

Why do you have ndiswrapper loaded?

[Xprogrammer (programmer-x)]

Good catch. I didn't notice that. I had tried via ndiswrapper before this driver came along. I have removed ndiswrapper completely now. Let me try again.

[stano (stanislav-simko)]

Today I managed to build (my very first) latest kernel, 3.18.rc1. Unfortunatelly, it didn't work. But I have got some logs that will be hopefully usefull... Also, if it could help i also can build another kernel if you (Larry/Xprogrammer/anybody else) will tell me which features have to be set...
 attached is dmesg output from the first message about wifi until few secconds before freeze. after freeze i was able to type into console (i was in tty2), but commands probably didn't get executed. After around 2 minutes, there started to be printed messages (atached pictures)...After about 10 minutes i "managed" to switch back to gui where I could see in power manager that there is again a lot of irq interrups from rtl_pci. I couldn't move cursor, type on keyboard... just force shutdown

[stano (stanislav-simko)]

here is first photo

[stano (stanislav-simko)]

and second photo

[Xprogrammer (programmer-x)]

@Larry,

I retried after removing ndiswrapper. Same result. System frozen after some time.

Oct 22 14:45:38 skannan-dv2k kernel: [21876.694487] Watchdog[3774]: segfault at 0 ip 00007f50cd9651ab sp 00007f50bd5
647d0 error 6 in chrome[7f50c9bcb000+4ffd000]
Oct 22 14:46:08 skannan-dv2k kernel: [21906.349910] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... render ring idle
Oct 22 14:46:20 skannan-dv2k kernel: [21918.366125] Watchdog[22566]: segfault at 0 ip 00007fc8c14911ab sp 00007fc8b10907d0 error 6 in chrome[7fc8bd6f7000+4ffd000]
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059383] INFO: rcu_sched self-detected stall on CPU { 0} (t=15000 jiffies g=616738 c=616737 q=0)
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059395] Task dump for CPU 0:
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059398] swapper/0 R running task 0 0 0 0x00000008
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059402] ffffffff82c54240 ffff88021f203bb0 ffffffff820a05b8 0000000000000000
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059406] ffffffff82c54240 ffff88021f203bd0 ffffffff820a2edf ffff88021f203c10
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059409] ffffffff82c54240 ffff88021f203c00 ffffffff820d307b ffffffff82c54240
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059412] Call Trace:
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059414] <IRQ> [<ffffffff820a05b8>] sched_show_task+0xa8/0x110
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059426] [<ffffffff820a2edf>] dump_cpu_task+0x3f/0x50
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059430] [<ffffffff820d307b>] rcu_dump_cpu_stacks+0x8b/0xc0
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059433] [<ffffffff820d4d47>] print_cpu_stall+0xd7/0x140
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059437] [<ffffffff820d4e55>] check_cpu_stall.isra.50+0xa5/0xb0
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059440] [<ffffffff820d4e98>] __rcu_pending+0x38/0x1d0
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059444] [<ffffffff820d6d0d>] rcu_check_callbacks+0xed/0x1a0
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059447] [<ffffffff820dc278>] update_process_times+0x48/0x80
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059452] [<ffffffff820ebf53>] tick_sched_handle.isra.14+0x33/0x70
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059455] [<ffffffff820ec07c>] tick_sched_timer+0x4c/0x80
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059458] [<ffffffff820dcd36>] __run_hrtimer+0x76/0x230
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059461] [<ffffffff820ec030>] ? tick_nohz_handler+0xa0/0xa0
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059464] [<ffffffff820dd53f>] hrtimer_interrupt+0xff/0x260
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059470] [<ffffffff8204b2eb>] local_apic_timer_interrupt+0x3b/0x60
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059475] [<ffffffff827a7635>] smp_apic_timer_interrupt+0x45/0x60
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059478] [<ffffffff827a571d>] apic_timer_interrupt+0x6d/0x80
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059483] [<ffffffff8207864b>] ? __do_softirq+0x8b/0x330
Oct 22 14:46:26 skannan-dv2k kernel: [21925.059487] [<ffffffff820785e0>] ? __do_softirq+0x20/0x330
Oct 22 14:46:26 sk...

[Xprogrammer (programmer-x)]

Output of /proc/interrupts and mpstat.

[Xprogrammer (programmer-x)]

Output /proc/interrupts.

===============================================

   CPU0 CPU1 CPU2 CPU3
  0: 34 0 0 0 IO-APIC-edge timer
  1: 657 4850 412 509 IO-APIC-edge i8042
  8: 1 0 0 0 IO-APIC-edge rtc0
  9: 582 198 38 86 IO-APIC-fasteoi acpi
 12: 200161 1319639 127277 121362 IO-APIC-edge i8042
 23: 0 21 2 14 IO-APIC 23-fasteoi ehci_hcd:usb1
 40: 759 6854 374468 1022237 PCI-MSI-edge xhci_hcd
 41: 1 14 10 10512 PCI-MSI-edge eth0
 42: 4 22 3 1 PCI-MSI-edge rtsx_pci
 43: 29672 81399 17076 52056 PCI-MSI-edge ahci
 44: 0 0 0 14 PCI-MSI-edge mei_me
 45: 86 142 5 7 PCI-MSI-edge snd_hda_intel
 46: 134989 967894 82142 106017 PCI-MSI-edge i915
 47: 418 96 175 273 PCI-MSI-edge snd_hda_intel
 48: 1274975 549 1263 598 PCI-MSI-edge rtl_pci
NMI: 0 0 0 0 Non-maskable interrupts
LOC: 2630287 1750798 3226887 2344785 Local timer interrupts
SPU: 0 0 0 0 Spurious interrupts
PMI: 0 0 0 0 Performance monitoring interrupts
IWI: 0 7 49 609 IRQ work interrupts
RTR: 2 0 0 0 APIC ICR read retries
RES: 454318 389561 516546 464945 Rescheduling interrupts
CAL: 1122 1277 1408 1468 Function call interrupts
TLB: 71053 63279 68739 61938 TLB shootdowns
TRM: 0 0 0 0 Thermal event interrupts
THR: 0 0 0 0 Threshold APIC interrupts
MCE: 0 0 0 0 Machine check exceptions
MCP: 72 72 72 72 Machine check polls
THR: 0 0 0 0 Hypervisor callback interrupts
ERR: 0
MIS: 0

Output of mpstat -I SUM -P 1 1
===============================

02:32:13 PM CPU intr/s
02:32:14 PM 1 249.00
02:32:15 PM 1 263.00
02:32:16 PM 1 306.00
02:32:17 PM 1 249.00
02:32:18 PM 1 306.00
02:32:19 PM 1 315.00
02:32:20 PM 1 291.00
02:32:21 PM 1 274.00
02:32:22 PM 1 310.00
02:32:23 PM 1 241.00
02:32:24 PM 1 259.00
02:32:25 PM 1 222.00
02:32:26 PM 1 207.00

[Bernardo Ávila Pires (obernardo)]

I am not sure this is relevant, but there are no issues with the card according to this thread ( http://ubuntuforums.org/showthread.php?t=2198221&page=2 ). My fiancée and I spent the afternoon configuring our T440, and the wireless card this bug refers to was detecting wireless networks in the installer.

[Josh Arenson (josharenson)]

@Bernado, my card has no problem detecting networks, but it has a < 10 % chance of actually connecting to one. I wish I could return my laptop.

[Damian Nadales (dnadales)]

I've tried the latest version of the driver from Github and I also experienced a system freeze. On the other hand, it is difficult to connect to any network, and the traffic is really slow. @Larry: which kernel version do you recommend for testing the drivers? I'm currently using 3.16.2-031602-generic.

[David Spies (dreamcontrol8000)]

I often have to restart a couple of times before I can connect to wi-fi. When I finally do, there can be anywhere from 10% - 40% packet loss.
I tried installing rtl-wifi-new and I wasn't able to connect at all (although maybe I didn't restart enough times) so I tried uninstalling it and the uninstaller destroyed my /boot/efi partition. I couldn't repair it and had to re-install Ubuntu altogether.

I'm on Ubuntu 14.10, kernel 3.16.0-23-generic. I have a Lenovo ThinkPad T440p laptop.
ThinkPad Wireless 2 x 2 BGN

[Sylvain Corlay (sylvain-corlay)]

Hello, what is the status of this?
I am using the 3.16.2 kernel and the firmware from #97 - but I am still experiencing those disconnections and kernel oops. Is the 3.18rc2 including the latest version supposed to be more stable?
Will there be significant changes before the final release of 3.18?

[stano (stanislav-simko)]

I've been trying 3.18rc1 without luck... so you can try newer one and let us know if it works :)

[charlie russ (charlieruss2128)]

I was able to get this to work on my new t440s.
I'm using kernel version 3.16.0-24-generic
I cloned from github and did the sudo make install from #184.
Lastly, I did:
 sudo modprobe r8192ee

and that seemed to do the trick because it wasnt working before that despite a restart.

I tried:

sudo modprobe rtl8192ee

and got an exec format error..

Thank you Larry for the drivers! you have saved me a lot of hassle!

[charlie russ (charlieruss2128)]

Well, that was short lived. I restarted my computer and now it has the same wifi issues as before.. connects for a second and then disconnects.

[brayan bautista (braybaut)]

I have this problem in Thinkpad x240

[Xiaojun Chen (xiaojunch)]

God, I wish I could return my T440s... I do not want to deal with this shit!

[Adam Lee (adam8157)]

Status:

Although this driver was moved out of staging dir, but because of the dependency hell, still decide to backport the staging driver.(the developed staging driver is already in Utopic, this bug is only on Trusty now.)

About the issues of this driver:

Please test the latest mainline kernel and report issues to bugzilla.kernel.org but not here.

[Adam Lee (adam8157)]

pull-request has been sent to kernel-team@

Utopic had already fixed this bug, you guys could upgrade distro too.

[Adam Lee (adam8157)]

pull-request for trusty was NAKed because next minor Trusty release will use Utopic LTS kernel, which had already fixed this issue.

please wait it or upgrade your whole distro to Utopic.

[Joseph (4-joseph)]

I upgraded to Utopic (from Ubuntu Gnome 14.04) on my Thinkpad T440 and wifi worked without any problems. However, after about a day and a half the system started freezing after using wifi for a few minutes (if I turn the wifi off, then it doesn't freeze), which is the same thing that had happened to me when I installed Larry's driver on 14.04.

I should also mention that the Ubuntu 14.04 from which I upgraded was a clean install without Larry's driver, so the wifi didn't work, but everything else worked fine.

Let me know if you want logs or anything.

[Vipin (v-rathor)]

Yesterday I too upgraded to Utopic (so now at 3.16.0-24) from Ubuntu Gnome 14.04. And for the first time, my L440 could scan and connect to wifi networks. Though connection was successful in the first try, but there was no data transfer, I could not ping any IP address. 'ifconfig' did show the correct IP address assigned to wlan0.
Let me know how can I help with debugging this. Thanks.

[Luren (lurenwang)]

I've noticed comments about system freezes. Do those system freezes have the following characteristics: 1. Mouse can move however it cannot click anything. 2. ctrl-alt-f1/f2 does not do anything. 3.REISUB does not work. 4. Random frequency. 5. No correlation with CPU load?

I've been running into this problem and the wifi issue. I never would have guessed these two could have been linked somehow.

[Joseph (4-joseph)]

Luren: when my system freezes, I cannot move the mouse or do anything else. It happens whenever I leave my wifi on for a few minutes. Otherwise, it never happens to me.

[Xiaojun Chen (xiaojunch)]

Hi all,

Just wondering if your wifi works on Utopic. In my case, I can connect, yet the speed is very slow. I suspect it only works on the 2.4G band, not the 5G band. Let me know if you have a similar problem.

[Vipin (v-rathor)]

After some days of using Utopic on my L440, I'm able to use wifi even for longer duration (~3 hours). The connection was okay-ish mostly (ping showed ~25% packet loss). Later, it failed to do a 'git push', that's when I switched to wired connection. :)
But I reckon that it has come a long way.. from not detecting the wifi card at all to somewhat 'surf-able' connection. Hoping that it'll be more stable soon. A BIG thanks to everyone who have contributed !!

[Ben (g-ben)]

I'm having the same issues with system freezes. Sometimes I can still move my mouse and issue some terminal-commands, sometimes it completely freezes and the only thing I can do is to press the power-button until it turns off.

What I noticed is that the CPU Usage in my CPU Graph Indicator (XUbuntu) shows 1 CPU constantly on 100%, while at the same time top doesn't show any high CPU Usage. I cannot relate this to any specific process. If I can provide any diagnostics while in this high-CPU-usage-state, please let me know how I can help, as it's really annoying.

This only happens if I'm connected to a wireless network. For now, I just use the ethernet cable when I'm at my desk and turn off the Wifi connection manually.

[Clive Cox (clivejcox)]

I have a Lenovo T440p. Just a recap of my experiences. I installed from http://github.com/lwfinger/rtlwifi_new.git.
My personal experiences:
 - it works ok but sometimes crashes as people have discussed requiring a hard reset
 - it crashes more often if I use skype or stream video
 - it crashes sometimes on web browser going to a new site, new DNS lookup?
 - If I click on NetworkManager (ubuntu) and restart the wifi it can repair itself. If I leave it then normally it will need a hard reset at somepoint.
 - The serious random crashes seemed to happen less after using Larry's latest rtlwifi install for me
 - ethernet connection is also an issue for me and takes several attempts to connect (I use e1000e driver), not sure if there is some interplay between the two drivers?

Personally I have a laptop I can use but need to tred carefully... Thanks to larry for doing so much. Be great if the final core issues were found with this hardware driver. I read somewhere what BIOS version you have can make a difference.

[stano (stanislav-simko)]

@Ben, Xprogrammer had some idea with running linux under virtual box with direct access to wifi card, can you check at intel's page if your CPU supports VT-d virtualization?

[Ben (g-ben)]

@stano it's a Core i5-4300U (running on a ThinkPad T440s) which supports VT-d according to Intel. I guess you're refering to comment #248?

[stano (stanislav-simko)]

yep, exactly. I think, one of the problems is that because whole system freezes there are no logs, etc... this way one could try to save something, but... so if you can run virtual box and try it.... maybe it will be of some help to those who know what next

[Christos (christos-dimitrakakis)]

This does not seem to have solved the problem on this L440. I have installed 13.10 - I can see networks, but once I connect, the connection stays up only for a couple of seconds. This is my report:

02:00.0 Network controller: Realtek Semiconductor Co., Ltd. RTL8192EE PCIe Wirel
ess Network Adapter
 Subsystem: Realtek Semiconductor Co., Ltd. Device 001b
 Flags: bus master, fast devsel, latency 0, IRQ 48
 I/O ports at 5000 [size=256]
 Memory at f2400000 (64-bit, non-prefetchable) [size=16K]
 Capabilities: [40] Power Management version 3
 Capabilities: [50] MSI: Enable+ Count=1/1 Maskable- 64bit+
 Capabilities: [70] Express Endpoint, MSI 00
 Capabilities: [100] Advanced Error Reporting
 Capabilities: [140] Device Serial Number 01-91-81-fe-ff-4c-e0-00
 Capabilities: [150] Latency Tolerance Reporting
 Capabilities: [158] L1 PM Substates
 Kernel driver in use: r8192ee

[Larry Finger (larry-finger-lwfinger)]

That report does not have any crucial info. When the above happens, what is shown at the end of the dmesg output. That will have the reason the line disconnected.

[Christos (christos-dimitrakakis)]

@Larry Finger

More info about the realtek8192
* Ubuntu 14.10
* Kernel: 3.16.0-24-generic
* dmesg output:

Initially, when I connect, I see:

IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready
wlan0: authenticate with ec:43...
wlan0: send auth to ec:43 ... (try 1/3)
wlan0: authenticated
r8192ee 0000:02:00.0 wlan0: disabling HT/VHT due to WEP/TKIP use
wlan0: associate with ec:43 .. (try 1/3)
wlan0: RX AssocResp from ec:43 (capab=0xc11 status=0 aid=1)
wlan0: associated
IPv6: ADDRCONF(NETDEV_CGANGE): wlan0: link becomes ready
cfg8021: Calling CRDA for country: SE
cfg8021: Regulatory domain changed to country: SE
cfg8021: (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp), (dfs_cac_time)
cfg8021: ... (data) ...
wlan0: deauthenticating from ec:43 ... by local choice (Reason: PREV_AUTH_NOT_VALID)
wlan0: sending auth
wlan0: authenticated
wlan0: associate with ...
wlan0: associated

* and then suddeny *
AP off, try to reconnect now
wlan0: Connection to AP ec:43 lost
cfg0211: World regulatory domain updated:
cfg8021: DFS Master region: unset
cfg8021: (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp), (dfs_cac_time)
cfg8021: ... (data) ...
...
wlan0: authenicate with ec:43 ....
wlan: send auth (try 1/3..)
wlan: send auth (try 2/3)
wlan: send auth (try 2/3)
wlan: authentication timed out.

AP off, try to reconnect
wlan0: Cnonection to AP lost
World regulatory domain updated:
cfg8021: DFS Master region: unset
cfg8021: (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp), (dfs_cac_time)
cfg8021: ... (data) ...
...
wlan0: authenicate with ec:43 ....
wlan: send auth (try 1/3..)
wlan: send auth (try 2/3)
wlan: send auth (try 2/3)
wlan: authentication timed out.

Not sure what happens exactly. I also tried an upstream kernel (3.17.3) to no avail. In fact, in the upstream, my wifi adapter doesn't seem to be active at all.

[Christos (christos-dimitrakakis)]

PS. Just to check, I set up the wireless with no authentication, with the same results, apart from the line about WEP. If I turn off/on wireless networking, I can connect to the wifi again briefly. However, afterwards it does not even appear in the menu (probably thinks it's out of range) unless I turn off/on wifi again.

[Christos (christos-dimitrakakis)]

Here is the debug=3 output (sorry, missed it in the comments)

[ 1644.574729] cfg80211: Calling CRDA to update world regulatory domain
[ 1644.603114] cfg80211: World regulatory domain updated:
[ 1644.603119] cfg80211: DFS Master region: unset
[ 1644.603122] cfg80211: (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp), (dfs_cac_time)
[ 1644.603126] cfg80211: (2402000 KHz - 2472000 KHz @ 40000 KHz), (300 mBi, 2000 mBm), (N/A)
[ 1644.603129] cfg80211: (2457000 KHz - 2482000 KHz @ 40000 KHz), (300 mBi, 2000 mBm), (N/A)
[ 1644.603132] cfg80211: (2474000 KHz - 2494000 KHz @ 20000 KHz), (300 mBi, 2000 mBm), (N/A)
[ 1644.603135] cfg80211: (5170000 KHz - 5250000 KHz @ 40000 KHz), (300 mBi, 2000 mBm), (N/A)
[ 1644.603137] cfg80211: (5735000 KHz - 5835000 KHz @ 40000 KHz), (300 mBi, 2000 mBm), (N/A)
[ 1644.605676] r8192ee: module is from the staging directory, the quality is unknown, you have been warned.
[ 1644.607487] mem mapped space: start: 0xf2400000 len:00004000 flags:00140204, after map:0xffffc900047b8000
[ 1644.607502] Pci Bridge Vendor is found index: 0
[ 1644.607505] pcidev busnumber:devnumber:funcnumber:vendor:link_ctl 2:0:0:10ec:0
[ 1644.607507] pci_bridge busnumber:devnumber:funcnumber:vendor:pcie_cap:link_ctl_reg:amd 0:28:1:8086:40:40:0
[ 1644.607531] Boot from EFUSE
[ 1644.614959] dev_addr: 38:b1:db:5e:e2:79
[ 1644.614967] RT Customized ID: 0x12
[ 1644.615172] 1T2R or 2T2R
[ 1644.616014] ieee80211 phy0: Selected rate control algorithm 'minstrel_ht'
[ 1644.616325] cfg80211: Calling CRDA for country: EC
[ 1644.616373] rtlwifi: wireless switch is on
[ 1644.616512] r8192ee 0000:02:00.0: irq 47 for MSI/MSI-X
[ 1644.616622] MSI Interrupt Mode!
[ 1644.619462] rtl_pci_start
[ 1644.619469] rtl92e_btc_init_hal_vars, antNum is 2
[ 1644.619471] rtl92e_btc_init_hal_vars, bt_exist is 1
[ 1644.619472] rtl92e_btc_init_hal_vars, bt_type is 9
[ 1644.625293] normal Firmware SIZE 32754
[ 1644.625298] Firmware Version(14), Signature(0x92e1), Size(32)
[ 1645.137265] PairwiseEncAlgorithm = 0 GroupEncAlgorithm = 0
[ 1645.137273] The SECR-value cc
[ 1645.296617] rtl_pci_start OK
[ 1645.313448] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready
[ 1645.313467] cfg80211: Regulatory domain changed to country: EC
[ 1645.313470] cfg80211: DFS Master region: unset
[ 1645.313471] cfg80211: (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp), (dfs_cac_time)
[ 1645.313473] cfg80211: (2402000 KHz - 2482000 KHz @ 40000 KHz), (N/A, 2000 mBm), (N/A)
[ 1645.313475] cfg80211: (5170000 KHz - 5250000 KHz @ 20000 KHz), (300 mBi, 1700 mBm), (N/A)
[ 1645.313476] cfg80211: (5250000 KHz - 5330000 KHz @ 20000 KHz), (300 mBi, 2300 mBm), (0 s)
[ 1645.313478] cfg80211: (5735000 KHz - 5835000 KHz @ 20000 KHz), (300 mBi, 3000 mBm), (N/A)
[ 1646.403607] IPS Set eRf nic disable
[ 1655.404882] wlan0: authenticate with ec:43:f6:37:33:a4
[ 1655.404900] IPS Set eRf nic enable
[ 1655.410710] normal Firmware SIZE 32754
[ 1655.410714] Firmware Version(14), Signature(0x92e1), Size(32)
[ 1655.922811] PairwiseEncAlgorithm = 0 GroupEncAlgorithm = 0
[ 1655.922818] The SECR-value cc
[ 1656.053663] !!!write point is 0x0, reg 0x3B4 value is 0x10000
[ ...

[K Xu (kx-anticss)]

Hi Adam,

I am using a version of this driver I got from Larry Finger, but it seems
to be buggy as my internet often drops. I am using it on a T440s on Ubuntu
12.04.

Should I update to Trusty, expecting to help solve this issue?

Thanks in advance,
Kelvin

On Tue, Mar 18, 2014 at 8:58 AM, Adam Lee <email address hidden> wrote:
>
> It's already in Trusty(14.04) daily build, you guys can try it. I will
> test it on my side and close this bug.
>
> commit 25e7ce81e86a4c47ea50e4d55567704d9ef957c5
> Author: Greg Kroah-Hartman <email address hidden>
> Date: Tue Jan 21 11:20:45 2014 -0800
>
> Staging: rtl8812ae: Add Realtek 8821 PCI WIFI driver
>
> BugLink: http://bugs.launchpad.net/bugs/1287298
>
> This comes directly from the Realtek tarball, filename:
> wifi_driver_8821ae_0018.1129.2013.tar.gz
>
> I mushed the three modules (btcoexist, rtlwifi and rtl8821ae) together
> into one, in order to make it all build as one stand-alone module.
> After the btcoexist driver gets merged upstream, I'll pull it out of
> here, and will continue to work on removing this version of rtlwifi in
> order to use the in-kernel one.
>
> Cc: Larry Finger <email address hidden>
> Signed-off-by: Greg Kroah-Hartman <email address hidden>
> (cherry picked from commit 3c05bedb5fef7a6ada63c5dbf61e1258b9019f05)
> Signed-off-by: Paolo Pisati <email address hidden>
> Signed-off-by: Tim Gardner <email address hidden>
>
> commit 9888aa51ff6d62ed1b8a57c540bb5b1dc2e5d25c
> Author: Tim Gardner <email address hidden>
> Date: Mon Mar 3 12:14:05 2014 -0700
>
> UBUNTU: [Config] CONFIG_R8821AE=m
>
> Signed-off-by: Tim Gardner <email address hidden>
>
> ** Description changed:
>
> Open this bug to trace device driver RTL8192EE.
>
> We have seen this wireless card on some laptops, id is [10ec:818b],
> - driver is RTL8192EE. The support doesn't show up in mainline or
> - wireless-next yet.
> + driver is RTL8192EE(or RTL8821AE, depends on RealTek).
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> Status in HWE Next Project:
> Incomplete
> Status in HWE Next trusty series:
> Confirmed
> Status in “linux” package in Ubuntu:
> Confirmed
>
> Bug description:
> Open this bug to trace device driver RTL8192EE.
>
> We have seen this wireless card on some laptops, id is [10ec:818b],
> driver is RTL8192EE(or RTL8821AE, depends on RealTek).
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions
>

[Giles Carré (giles-carre)]

Hi,

Le 11/12/2014 23:06, K Xu a écrit :
> I am using a version of this driver I got from Larry Finger, but it seems
> to be buggy as my internet often drops. I am using it on a T440s on Ubuntu
> 12.04.
>
> Should I update to Trusty, expecting to help solve this issue?

I have used Ubuntu 14.04 without success, because the driver wasn't
integrated in it kernel version (3.14).

I am now on Ubuntu 14.10 since october (today with the last main kernel,
3.16.0-28). I am able to see the wifi networks, so better than with
14.04, but steel unable to connect.

I don't know if this "Ubuntu" kernel integrates the last revision of RTL
8192EE driver.

Regards,

    Giles Carré

--

*Giles Carré* — *RSSI*
Téléphone : [phone]
Mobile : [phone]
<email address hidden> <mailto:<email address hidden>>

INSA Toulouse
Centre des Services Numériques
135 avenue de Rangueil
31077 Toulouse CEDEX 04

INSA Toulouse
*www.insa-toulouse.fr <http://www.insa-toulouse.fr>***

[yaxu (alex-slab)]

Is anyone successfully running with this driver?

I'd rather avoid occasional system freezes, as I use my laptop in live music performance.

[Vipin (v-rathor)]

Recently my L440 running Ubuntu Gnome 14.10 received update to 3.16.0-28. And yesterday I was able to use wifi for ~4-5 hrs. without any problem. No problem in searching, connecting to wifi network.

[stano (stanislav-simko)]

i've been running the newest drivers from troy for last 4 days on 3.17.4-200.fc20.x86_64. Had one freeze, but i'm not sure if it was because of the driver.

[Ben (g-ben)]

Can confirm running 3.16.0-28 on XUbuntu looks good, no freezes for 6h yesterday and 3h today. Connections over Wifi still seem a little slow and unreliable, but at least the crashes seem to be fixed.

[ssteinhauer (t-sebastian)]

I am running Debian on a T440s with Kernel 3.16.0-4-amd64 and have been able to stay connected with not hiccups for 24h and counting. I can also confirm the limited link speed.

iwconfig | grep Rate

returns:

Bit Rate=1 Mb/s Tx-Power=20 dBm

[Robert (rahiggs)]

Can confirm this bug still present in 3.18 kernel.

[Ben (g-ben)]

Can confirm the lower link speed, was 6.x Mb/s for me.

After manually setting it to 54 Mb/s using `iwconfig wlan0 rate 54M` the connection is much more reliable, although not perfect.

[Frances Hocutt (frances-hocutt)]

I just upgraded to the latest version of 14.10. The first time I started up after upgrade, I couldn't see any networks. After a restart, I was briefly able to see and connect to my local wireless network. When I started to load webpages it disconnected, and I was no longer able to see the network I had been connected to in the network list. I have been trying disabling and enabling wi-fi from the drop-down menu, but even when "Enable Wi-Fi" is checked I still see a greyed-out "Wi-Fi is disabled" line.

I will try restarting again and will see if letting the connection go for longer before I start trying to move data across it helps.

[Frances Hocutt (frances-hocutt)]

On restart, it tried to connect to the network automatically a few times and failed, but it still saw the network. I tried to connect manually several times; nothing. I tried disabling and enabling wifi and ran into the same problem I described in #299.

This driver appears to still not be usable for me. T440s, 3.16.0-28 kernel.

[Ben (g-ben)]

With 3.16.0-28 there's still very rare freezes for me, like once per week (running 8h/day). But already much better than before, when it froze about once per hour.

[BR (rocks-brendan)]

Just to add to the clamor, I'm in exactly the same position as Frances (T440, tried both 14.04LTS and 14.10).

[Larry Finger (larry-finger-lwfinger)]

If you want a working driver for the RTL8192EE, you will need to wait for kernel 3.19, or download the drivers from http://github.com/lwfinger/rtlwifi_new.git. The latest Realtek code is in the "troy" branch of that repo. That should fix the freezes and the wireless stalls.

The changes are so invasive that they are unlikely to be packported to any kernels older than 3.19.

[BR (rocks-brendan)]

Thanks Larry.

My first attempt with those drivers didn't work, but I'm going to try again on a fresh install. Should I be installing them on 14.04LTS or 14.10 (does it even matter)? I'm guessing I just clone the repo, checkout the troy branch, and then run 'make', 'make install', and that's it?

[Larry Finger (larry-finger-lwfinger)]

It does not matter what Ubuntu release you are using as long as the kernel is 3.2 or later.

You need to "sudo make install", and then reboot.

[Sylvain Corlay (sylvain-corlay)]

The build fails with the "troy" branch.
error: (implicit declaration of function pci_zalloc_consistent line 1273 in pci.c)
There is a bunch of warning besides this error.

[Larry Finger (larry-finger-lwfinger)]

What kernel are you using?

[Sylvain Corlay (sylvain-corlay)]

Linux 3.16.2. Shouls this work out of the box with the latest 3.19 rc?

[Christian Pinto (chrpinto)]

Hello,

 I Have tried, as suggested, to boot a 3.19 rc4 but the problem of speed drop is still there. Tested on a Thinkpad T440s + Ubuntu 14.04 + RTL8192ee.

I have also tried to compile the troy branch, compilation/installation went well.

Then at boot time i get the following:

[ 2.143768] rtlwifi: module verification failed: signature and/or required key missing - tainting kernel
[ 2.144295] rtlwifi: exports duplicate symbol rtl_evm_dbm_jaguar (owned by r8192ee)

Is there a way to sign the module without re-compiling the kernel??

Thanks,
Christian

Christian

[Christian Pinto (chrpinto)]

Sorry,

the first of the two is not an error, since the kernel on my machine has not be compiled with

CONFIG_MODULE_SIG_FORCE=y

loading an untrusted module is just tainting the kernel, and the module is loaded in any case.

Given this, the second warning remains and even installing this new driver i see the same behavior of speed drop after some time.

As a test i am trying to download a large file from my workstation at my office (University), where for sure there are no bandwidth problems.

On which card/kernel have you tried the code???

Christian

[Larry Finger (larry-finger-lwfinger)]

The code in 3.19-rcX is essentially the same as in the troy branch; however, there are several bugs that have been fixed in troy that are still pending for mainline.

I am using 3.19-rc4 with some local modifications. My card is the same as yours. I have not done much testing for speed, and have concentrated on removing the kernel oops and failures to recover when the connection drops.

My only suggestion is to load the module with the "ips=0" option.

[Christian Pinto (chrpinto)]

Thanks a lot Larry

i see something strange:

there is an r8192ee and an rtl8192ee module in my system, the former comes from the staging drivers in the kernel and it is the only actually being loaded in the system (kernel 3.17). So i am not actually trying you code.

How can i prevent the staging drivers from being mounted in the system, and only let the others to be used?

Christian

[Larry Finger (larry-finger-lwfinger)]

To keep r8192ee from loading, add a blacklist line for r8192ee in /etc/modprobe.d/50-blacklist.conf. We were careful in the naming so that the staging driver has a different name that the regular kernel one does.

[Christian Pinto (chrpinto)]

Thanks again Larry

installing the new driver got better on one sense, but worse on another.

Now testing with iperf i do not see any other speed drop, the connection is quite stable. However, i have programmed iperf to go for 500 seconds, but after about 300 the connection died at all and in dmesg i read:

[ 779.196064] rtlwifi:rtl_cam_del_entry():<0-0> cam_bitmap: 0x2
[ 779.196440] rtlwifi:rtl_cam_get_free_entry():<0-0> key_index=0,cam_bitmap: 0x12 entry_idx=4
[ 779.201434] rtlwifi:rtl_cam_del_entry():<0-0> cam_bitmap: 0x10
[ 779.201806] rtlwifi:rtl_cam_get_free_entry():<0-0> key_index=1,cam_bitmap: 0x12 entry_idx=1

Then i had either do disconnect and reconnect to the network, or disable/enable the network interface.

In modinfo i also see that the new driver is still using the old firmware rtlwifi/rtl8192eefw.bin, while in the code i see it should load rtlwifi/rtl8192eefw_new.bin. Is it just a broken message printed, or is it really not loading the new firmware?

I hope my messages help you in getting a stable version of the driver. I am bothering so much because, as most of us, i use my laptop for work and i need to have a stable wireless connection.

Christian

[Larry Finger (larry-finger-lwfinger)]

The troy branch loads rtl8192eefw_new.bin; however, after some testing and consultation with Realtek, I determined that there was no need to keep separate names. For that reason and to minimize the disk usage, the kernel loads rtl8192eefw.bin. You can copy the one with the _new in its name to the one without, if you want to use the later version. The updated firmware is already in the linux-firmware repo, and will be propagated through the distros in time.

The rtl_cam logged items indicate the changing of the wifi keys. The second one tells you that the connection has been dropped. It is NOT the cause of the drop.

I have been using the driver for over 24 hours. There have been several disconnects in that time; however, the recovery has been automatic. One thing to note is that the transmit is somewhat fragile. Operations like continuous iperf or flood pings will likely cause a failure and are discouraged at least for the moment. Realtek knows of the problem and is working on it.

[Sylvain Corlay (sylvain-corlay)]

@larry, is the troy branch supposed to build with linux 3.16?

[Larry Finger (larry-finger-lwfinger)]

It will now. Posting the first error would have been helpful.

[Sylvain Corlay (sylvain-corlay)]

Great! (You should not have hesitated to ask, I just did not want to bloat the thread and would have been happy to help. )

[Mike Michalak (michalakx)]

The drivers from comment 303,https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1239578/comments/303, worked for me.
I am running Linux Mint 17 Qiana KDE on a Lenovo T440p.

[Minjae David (minjaedavid)]

The driver in the troy branch (comment #303) works well for me. The computer no longer freezes every 15 minutes due to the driver. I am also using the new firmware by renaming it to rtl8192eefw.bin (comment #315). I see no problem with down/up speed, and stability is reasonable (drops ping packets occasionally). My setup is

Ubuntu 14.10 server
Linux 3.19 rc6
Lenovo T540p

Thanks for the excellent work.

[Mike Michalak (michalakx)]

I have an update.

First couple of days were good.

Currently the connection needs to be reconnected after coming back from sleep mode.

Connection also drops, but so far so good.

I am running Linux Mint 17 Qiana KDE on a Lenovo T440p.

[Damian Nadales (dnadales)]

Ok, I'm giving the kernel 3.19 a try. Here is a list of the steps I'm
following according to the information gathered in this thread. Please
tell me if I'm missing something:
  - Install the kernel 3.19 from
http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.19-rc7-vivid/
  - Clone the GIT repository: git clone
https://github.com/lwfinger/rtlwifi_new.git
  - Compile and install the drivers: make && sudo make install
  - Blacklist the r8192ee module

Any remarks?

Thanks.

On Thu, Jan 22, 2015 at 3:17 AM, Larry Finger <email address hidden> wrote:
> It will now. Posting the first error would have been helpful.
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> Status in HWE Next Project:
> Fix Released
> Status in linux package in Ubuntu:
> Fix Released
> Status in linux source package in Trusty:
> Won't Fix
> Status in linux source package in Utopic:
> Fix Released
>
> Bug description:
> Open this bug to trace device driver RTL8192EE.
>
> We have seen this wireless card on some laptops, id is [10ec:818b],
> driver is RTL8192EE.
>
> This bug is only for the initial support of this WiFi adapter, it's
> not working 100% well yet, please report new "not working bug" or
> contact upstream but not keep posting comments, thanks.
>
>
> Status:
>
> pull-request for trusty was NAKed because next minor Trusty release
> will use Utopic LTS kernel, which had already fixed this issue.
>
> please wait it or upgrade your whole distro to Utopic.
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions

[bulldozer2003 (bulldozer2003)]

Larry,

Thank you for your work on this wireless driver. I, however, gave up and bought an Intel 7260 dual-band for my thinkpad on ebay for $20 to stop dealing with this issue.

Would it help for me to mail you, or someone else working on development, the realtek Mini PCI Express card I no longer need?

[Larry Finger (larry-finger-lwfinger)]

Thanks for the offer, but I have two of those already.

[Nick Gerner (nick-gerner)]

To follow up on buldozer2003's offer, would it help if we bought you a beer? or in some other way contributed to your work that I think many of us find extremely valuable? :)

[Kelvin (iamkelvinxu)]

I have been trying kernel 3.19-rc4 as suggested by #311. I however, get frequent crashes especially when streaming any videos/sound. Does anyone else have this issue? I'm using a T440s with Ubuntu 12.04.

Anyone with a similar system who has something working well, please let me know!

[Christian Pinto (chrpinto)]

I am using kernel 3.19.0 on a T440s with no luck, the systems hangs and the connection speed drops, so at the moment I am using a 10EUR USB wireless card.
I think I am close to give up and change the wireless card. buldozer2003, have you solved the problem by using the Intel card? If yes, can you please provide the exact model number so that i can buy the same?

Thanks in advance,
Christian

[Ivan Alejandro (ivanalejandro0)]

I have no luck either.
Not using the 3.19 kernel, neither adding the module from lwfinger repo.
I have a L440.

[Larry Finger (larry-finger-lwfinger)]

Kernel 3.19.1 will have some additional fixes needed for the RTL8192EE that should fix the freezes. That hardware uses a completely new method for RX and TX descriptors, and it has taken a while to get it right.

The troy branch of the rtlwifi_new repo at GitHub has corrected code.

[Ivan Alejandro (ivanalejandro0)]

I'm using the troy branch... I did the same as #322 (except for blacklisting r8192 since I don't have that module here)
The connection kind of works but very often is dropped (it appears to be connected but not works).
Is there any feedback that I can give you to help?
Thanks

[Larry Finger (larry-finger-lwfinger)]

Have you loaded with the "ips=0" option?

[Christian Pinto (chrpinto)]

I am now trying the latest changes in the troy branch, and experiencing the same behavior as ivanalejandro0. The connection works, but it drops from time to time. Definitely better than previous versions. This is anyway annoying because i cannot have a skype call without interruptions.

Regarding the ips option, I think that it is set as zero by default (even if modinfo says the opposite), or at least if I execute:

cat /sys/module/rtl8192ee/parameters/ips

it returns N, which I suppose to be 0.

Christian

[Larry Finger (larry-finger-lwfinger)]

You are correct that ips is false (or 0 or N) by default. In fact, it seems that the driver never even uses that one. It does enable and use firmware-controlled power save. Load withe "fwlps=0" to completely disable power saving.

[Ivan Alejandro (ivanalejandro0)]

Thanks for the tip Larry!
As Christian said, ips is 0 by default. using "fwlps=0" seems to do the trick :D ... I'm doing some extra tests with downloads, audio conversations and so on, and it works so far!

Some extra notes in case of this is useful to someone else:

# change to load the module with "fwlps=0"
$ cat /etc/modprobe.d/rtl8192ee.conf
options rtl8192ee fwlps=0

I'm using tlp for power management optimizations and I had also some extra settings to disable:

$ grep -i wifi /etc/default/tlp
# WiFi power saving mode: 1=disable, 5=enable; not supported by all adapters.
WIFI_PWR_ON_AC=1
WIFI_PWR_ON_BAT=1 # This is set to 5 by default

I imagine that since the driver does not uses "ips" these last changes don't matter.

I was having problems on resuming after suspend (no connection), and removing/adding the module causes the system to freeze. I've fixed with this:
cat /etc/pm/sleep.d/10_reset_wifi
#! /bin/sh

# This script reload the wifi module on resume since most of the times it won't
# work as espected otherwise.

case $1 in
     suspend|suspend_hybrid|hibernate)
         # do nothing
         ;;
     resume|thaw)
         sudo rmmod rtl8192ee && sudo modprobe rtl8192ee
         ;;
esac

[Ivan Alejandro (ivanalejandro0)]

I'm having problems again, this fix kind of worked the first two days (maybe I didn't use my laptop intensively).

I'm having on and off connection.

These are the settings for the driver right now:

$ for f in /sys/module/rtl8192ee/parameters/*; do echo $f - `cat $f`; done;
/sys/module/rtl8192ee/parameters/debug - 1
/sys/module/rtl8192ee/parameters/disable_watchdog - N
/sys/module/rtl8192ee/parameters/fwlps - N
/sys/module/rtl8192ee/parameters/ips - N
/sys/module/rtl8192ee/parameters/msi - Y
/sys/module/rtl8192ee/parameters/swenc - N
/sys/module/rtl8192ee/parameters/swlps - N
i

Here is a ping output (with a wifi dongle all of these are below 100ms):

# [...] are the ones with "normal" ping times, 20~45 ms
$ ping www.google.com
PING www.google.com (173.194.42.113) 56(84) bytes of data.
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=1 ttl=53 time=64.4 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=2 ttl=53 time=23.7 ms
[...]
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=26 ttl=53 time=22.4 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=27 ttl=53 time=19.9 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=28 ttl=53 time=150 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=29 ttl=53 time=21.7 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=30 ttl=53 time=35.5 ms
[...]
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=41 ttl=53 time=54.5 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=42 ttl=53 time=4535 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=43 ttl=53 time=3535 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=44 ttl=53 time=2535 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=45 ttl=53 time=1538 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=46 ttl=53 time=538 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=48 ttl=53 time=19.5 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=49 ttl=53 time=21.9 ms
[...]
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=94 ttl=53 time=21.2 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=95 ttl=53 time=131 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=96 ttl=53 time=18.9 ms
[...]
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=157 ttl=53 time=33.3 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=158 ttl=53 time=163 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=159 ttl=53 time=4766 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=160 ttl=53 time=3760 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=161 ttl=53 time=2756 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=162 ttl=53 time=1748 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=163 ttl=53 time=740 ms
64 bytes from eze03s16-in-f17.1e100.net (173.194.42.113): icmp_seq=164 ttl=53 time=50.7 ms
[...]
64 bytes from eze03s16-in-f17.1...

[Christian Pinto (chrpinto)]

I confirm what Ivan is experiencing, I am still not able to make a simple voice call. Hope this will be solved soon.

I saw in the repo two different versions of the firmware, should we try to test rtl8192eefw_new.bin ?

Chrsitian

[Ivan Alejandro (ivanalejandro0)]

Also, when I reload the module (neccesary since otherwise I don't get connection after suspend/resume), most of the times the system freezes forcing me to restart. I don't know if this is a problem of the driver itself though.

[Larry Finger (larry-finger-lwfinger)]

Unless you post the console traceback when the system freezes, how can we tell if it is the driver?

If you can reliable force the freeze with a particular command, put a "sleep 5; " before that command. During that 5 seconds delay, switch to the logging console (CTRL/ALT/F10 on my system). When the system freezes, there should be a display on the screen. Take a picture and post it.

[Larry Finger (larry-finger-lwfinger)]

This morning, a bug was found that could cause an oops on the RTL8192EE. All branches of the repo have been fixed.

[Ivan Alejandro (ivanalejandro0)]

cool, I've updated the troy branch and installed the new kernel module.

I'm still having the connection on&off issue, some pings over 2000ms as well like before.
The system freeze happens on module reload so I'll make some time to try to reproduce the problem and post an image.

[Ivan Alejandro (ivanalejandro0)]

Here is a dmesg log that has a traceback.
Is the result of a clean power on, suspend and resume.
I didn't have a freeze this time but maybe the problem is related to this.

[ 39.996883] ACPI: Preparing to enter system sleep state S3
[ 40.028699] PM: Saving platform NVS memory
[ 40.028705] Disabling non-boot CPUs ...
[ 40.028734] intel_pstate CPU 1 exiting
[ 40.029899] kvm: disabling virtualization on CPU1
[ 40.029907] smpboot: CPU 1 is now offline
[ 40.030237] intel_pstate CPU 2 exiting
[ 40.031404] kvm: disabling virtualization on CPU2
[ 40.031411] smpboot: CPU 2 is now offline
[ 40.031688] intel_pstate CPU 3 exiting
[ 40.031813] Broke affinity for irq 16
[ 40.031821] Broke affinity for irq 23
[ 40.031822] Broke affinity for irq 24
[ 40.031823] Broke affinity for irq 27
[ 40.031824] Broke affinity for irq 32
[ 40.032826] kvm: disabling virtualization on CPU3
[ 40.032835] smpboot: CPU 3 is now offline
[ 40.034143] ACPI: Low-level resume complete
[ 40.034179] PM: Restoring platform NVS memory
[ 40.034510] Enabling non-boot CPUs ...
[ 40.034544] x86: Booting SMP configuration:
[ 40.034544] smpboot: Booting Node 0 Processor 1 APIC 0x1
[ 40.046530] kvm: enabling virtualization on CPU1
[ 40.048794] CPU1 is up
[ 40.048813] smpboot: Booting Node 0 Processor 2 APIC 0x2
[ 40.060800] kvm: enabling virtualization on CPU2
[ 40.063061] CPU2 is up
[ 40.063081] smpboot: Booting Node 0 Processor 3 APIC 0x3
[ 40.075084] kvm: enabling virtualization on CPU3
[ 40.077362] CPU3 is up
[ 40.080842] ACPI: Waking up from system sleep state S3
[ 40.190171] acpi LNXPOWER:00: Turning OFF
[ 40.234221] xhci_hcd 0000:00:14.0: System wakeup disabled by ACPI
[ 40.254241] ehci-pci 0000:00:1a.0: System wakeup disabled by ACPI
[ 40.274259] ehci-pci 0000:00:1d.0: System wakeup disabled by ACPI
[ 40.274303] PM: noirq resume of devices complete after 64.031 msecs
[ 40.442614] PM: early resume of devices complete after 168.104 msecs
[ 40.442780] e1000e 0000:00:19.0: System wakeup disabled by ACPI
[ 40.442926] rtlwifi: wireless switch is on
[ 40.443022] rtc_cmos 00:02: System wakeup disabled by ACPI
[ 40.446194] sd 4:0:0:0: [sda] Starting disk
[ 40.718975] usb 1-7: reset full-speed USB device number 2 using xhci_hcd
[ 40.762816] ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
[ 40.763412] ata5.00: ACPI cmd ef/02:00:00:00:00:a0 (SET FEATURES) succeeded
[ 40.763414] ata5.00: ACPI cmd f5/00:00:00:00:00:a0 (SECURITY FREEZE LOCK) filtered out
[ 40.764971] ata5.00: ACPI cmd ef/02:00:00:00:00:a0 (SET FEATURES) succeeded
[ 40.764973] ata5.00: ACPI cmd f5/00:00:00:00:00:a0 (SECURITY FREEZE LOCK) filtered out
[ 40.765900] ata5.00: configured for UDMA/133
[ 40.770826] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[ 40.772866] ata1.00: ACPI cmd e3/00:00:00:00:00:a0 (IDLE) succeeded
[ 40.773877] ata1.00: ACPI cmd e3/00:00:00:00:00:a0 (IDLE) succeeded
[ 40.779535] ata1.00: ACPI cmd e3/00:00:00:00:00:a0 (IDLE) succeeded
[ 40.780542] ata1.00: ACPI cmd e3/00:00:00:00:00:a0 (IDLE) succeeded
[ 40.782414] ata1.00: configured for UDMA/133
[ 40.849102] xhci_hcd 0000:00:14.0: xHCI...

[Hanson.diao (changlelele)]

dmesg information

[ 1189.347194] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
[ 1219.856463] rtlwifi-0:rtl_pci_start():<0-0> rtl_pci_start
[ 1219.862483] rtl8192ee-0:rtl92ee_download_fw():<0-0> normal Firmware SIZE 32754
[ 1219.862488] rtl8192ee-0:rtl92ee_download_fw():<0-0> Firmware Version(14), Signature(0x92e1),Size(32)
[ 1220.370714] rtl8192ee-0:rtl92ee_enable_hw_security_config():<0-0> PairwiseEncAlgorithm = 0 GroupEncAlgorithm = 0
[ 1220.370727] rtl8192ee-0:rtl92ee_enable_hw_security_config():<0-0> The SECR-value cc
[ 1220.505206] rtlwifi-0:rtl_pci_start():<0-0> rtl_pci_start OK
[ 1220.509288] !!!write point is 0x0, reg 0x3B4 value is 0x50000
[ 1220.523932] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready

root@Hanson:~/Downloads/rtlwifi_new-master# uname -a
Linux Hanson 3.13.0-24-generic #46-Ubuntu SMP Thu Apr 10 19:11:08 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

root@Hanson:~/Downloads/rtlwifi_new-master# for f in /sys/module/rtl8192ee/parameters/*; do echo $f - `cat $f`; done;
/sys/module/rtl8192ee/parameters/fwlps - Y
/sys/module/rtl8192ee/parameters/ips - N
/sys/module/rtl8192ee/parameters/swenc - N
/sys/module/rtl8192ee/parameters/swlps - N
root@Hanson:~/Downloads/rtlwifi_new-master#

root@Hanson:~/Downloads/rtlwifi_new-master# sudo lshw -C network -numeric
  *-network
       description: Ethernet interface
       product: Ethernet Connection I218-LM [8086:155A]
       vendor: Intel Corporation [8086]
       physical id: 19
       bus info: pci@0000:00:19.0
       logical name: eth0
       version: 04
       serial: 28:d2:44:65:07:dc
       size: 100Mbit/s
       capacity: 1Gbit/s
       width: 32 bits
       clock: 33MHz
       capabilities: pm msi bus_master cap_list ethernet physical tp 10bt 10bt-fd 100bt 100bt-fd 1000bt-fd autonegotiation
       configuration: autonegotiation=on broadcast=yes driver=e1000e driverversion=2.3.2-k duplex=full firmware=0.6-3 ip=192.168.1.102 latency=0 link=yes multicast=yes port=twisted pair speed=100Mbit/s
       resources: irq:61 memory:f0600000-f061ffff memory:f063e000-f063efff ioport:4080(size=32)
  *-network DISABLED
       description: Wireless interface
       product: RTL8192EE PCIe Wireless Network Adapter [10EC:818B]
       vendor: Realtek Semiconductor Co., Ltd. [10EC]
       physical id: 0
       bus info: pci@0000:03:00.0
       logical name: wlan0
       version: 00
       serial: 48:5a:b6:93:a0:71
       width: 64 bits
       clock: 33MHz
       capabilities: pm msi pciexpress bus_master cap_list ethernet physical wireless
       configuration: broadcast=yes driver=rtl8192ee driverversion=3.13.0-24-generic firmware=N/A latency=0 link=no multicast=yes wireless=IEEE 802.11bgn
       resources: irq:66 ioport:3000(size=256) memory:f0400000-f0403fff

[Larry Finger (larry-finger-lwfinger)]

Something is messed up. Those dmesg outputs are not coming from the driver in the rtlwifi_new repo. What other drivers have you installed?

[Krzysztof Janowicz (janowicz)]

Still the same, despite a new kernel (3.19) running on 14.10. Essentially, Lenovo T440(s) and T540 don't have wifi on Linux.

[Larry Finger (larry-finger-lwfinger)]

That is the early adopter's problem. For me, any branch of the rtlwifi_new git repo, or the kernel work. Are you sure it is not a problem with Ubuntu user space?

[Simon Déziel (sdeziel)]

Users of Trusty should give a try to the HWE kernel (3.16 ATM) that is officially supported.

For details on how to enable this: https://wiki.ubuntu.com/Kernel/LTSEnablementStack

[William Oprandi (woprandi)]

I'm using Fedora with kernel 3.19.3 (which use the latest rtlwifi_new if I well understood) but I'm unable to use wifi, I can connect to an access point (sometimes with difficulty) but unable to surf without wait a (very) long time to load a web page. I was forced to use a ethernet wire just to go here to write this comment.... Since end of 2013 we're waiting a decent driver (just stable to surf correctly) and still not...

[Damian Nadales (dnadales)]

@Simon: I tried with kernel 3.16 in Utopic with the drivers shipped with the kernel, and the wireless works quite bad (very low speeds, pages take several minutes to load). Won't Trusty users experience the same?

[Adam Lee (adam8157)]

@Damian, please report a new bug.

This bug is only for the initial support of this WiFi adapter, it's not working 100% well yet, please report new "not working bug" or contact upstream but not keep posting comments, thanks.

BTW, if anyone's case is Larry's github version works well but Ubuntu built-in doesn't, please report here with the realtek section in the output of `lspci -vvv`, thanks.

[Ivan Alejandro (ivanalejandro0)]

Hello again, I just installed the linux kernel 4.0 in my Thinkpad L440 and I'm still having the delayed pings and on&off connection problem. Should I install the driver from the rtl_wifi repo?
@larry, you asked about the weird dmesg to me or to Hanson.diao?

How can I figure out if the problem is the Ubuntu user space?
Is there any other information I can give you or any other test that I can do to figure out where the problem may be?

[Damian Nadales (dnadales)]

@Adam: you're right. I didn't realize that. Sorry for the spam. I'll take the issues to the places you proposed.

[freeware2000 (freeware2000)]

Work well:
+Linux Mint 17.1
+ Kernel 3.13.0-37

sudo apt-get install git
sudo git clone http://github.com/lwfinger/rtlwifi_new.git
cd ./rtlwifi_new
sudo make
sudo make install
sudo modprobe -v rtl8192ee

[Lee Zamparo (zamparo)]

A few errors in the previous message:

git repo is at https://github.com/lwfinger/rtlwifi_new.git
module is named r8192ee, not rtl8192ee

Lee.

On Mon, May 18, 2015 at 4:23 AM, freeware2000 <email address hidden> wrote:
> Work well:
> +Linux Mint 17.1
> + Kernel 3.13.0-37
>
> sudo apt-get install git
> sudo git clone http://github.com/lwfinger/rtlwifi_new.git
> cd ./rtlwifi_new
> sudo make
> sudo make install
> sudo modprobe -v rtl8192ee
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> Status in HWE Next Project:
> Fix Released
> Status in linux package in Ubuntu:
> Fix Released
> Status in linux source package in Trusty:
> Won't Fix
> Status in linux source package in Utopic:
> Fix Released
>
> Bug description:
> Open this bug to trace device driver RTL8192EE.
>
> We have seen this wireless card on some laptops, id is [10ec:818b],
> driver is RTL8192EE.
>
> This bug is only for the initial support of this WiFi adapter, it's
> not working 100% well yet, please report new "not working bug" or
> contact upstream but not keep posting comments, thanks.
>
>
> Status:
>
> pull-request for trusty was NAKed because next minor Trusty release
> will use Utopic LTS kernel, which had already fixed this issue.
>
> please wait it or upgrade your whole distro to Utopic.
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions

[Larry Finger (larry-finger-lwfinger)]

Driver r8192ee is from the staging directory for kernels older than 3.18. It is *not* the one in rtlwifi_new. You need to blacklist r8192ee!

[Lee Zamparo (zamparo)]

Indeed you’re right; my apologies for confusing the issue.

--
Lee

On May 26, 2015 at 12:00:59 AM, Larry Finger (<email address hidden>) wrote:

Driver r8192ee is from the staging directory for kernels older than
3.18. It is *not* the one in rtlwifi_new. You need to blacklist r8192ee!

--
You received this bug notification because you are subscribed to the bug
report.
https://bugs.launchpad.net/bugs/1239578

Title:
No wireless support for Realtek RTL8192EE [10ec:818b]

Status in HWE Next Project:
Fix Released
Status in linux package in Ubuntu:
Fix Released
Status in linux source package in Trusty:
Won't Fix
Status in linux source package in Utopic:
Fix Released

Bug description:
Open this bug to trace device driver RTL8192EE.

We have seen this wireless card on some laptops, id is [10ec:818b],
driver is RTL8192EE.

This bug is only for the initial support of this WiFi adapter, it's
not working 100% well yet, please report new "not working bug" or
contact upstream but not keep posting comments, thanks.

Status:

pull-request for trusty was NAKed because next minor Trusty release
will use Utopic LTS kernel, which had already fixed this issue.

please wait it or upgrade your whole distro to Utopic.

To manage notifications about this bug go to:
https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions

[Christos (christos-dimitrakakis)]

Dear Larry and other troubled souls,

Thank you very much for your efforts in getting this working. I can happily report that finally this seems to be working smoothly on an L440, with Trusty and 3.18 Kernel, r8192 blacklisted. Here are some log files for info.

wlan0 IEEE 802.11bgn ESSID:"AirMac0856442"
          Mode:Managed Frequency:2.462 GHz Access Point: 80:EA:96:EE:83:7E
          Bit Rate=144.4 Mb/s Tx-Power=20 dBm
          Retry short limit:7 RTS thr=2347 B Fragment thr:off
          Power Management:off
          Link Quality=70/70 Signal level=-32 dBm
          Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
          Tx excessive retries:0 Invalid misc:284 Missed beacon:0

And the initial set-up

[ 0.000000] PERCPU: Embedded 30 pages/cpu @ffff88011f200000 s82880 r8192 d31808 u262144
[ 0.000000] pcpu-alloc: s82880 r8192 d31808 u262144 alloc=1*2097152
[ 0.045733] Mount-cache hash table entries: 8192 (order: 4, 65536 bytes)
[ 0.045738] Mountpoint-cache hash table entries: 8192 (order: 4, 65536 bytes)
[ 5.884637] rtlwifi: module verification failed: signature and/or required key missing - tainting kernel
[ 5.959874] Using firmware rtlwifi/rtl8192eefw_new.bin
[ 5.980581] ieee80211 phy0: Selected rate control algorithm 'rtl_rc'
[ 5.980859] rtlwifi: wireless switch is on
[ 5.980951] rtl8192ee 0000:02:00.0: irq 30 for MSI/MSI-X

[Lee Zamparo (zamparo)]

I'd like to echo #356, the QOS of this card is pretty good now over WPA2 connections, thanks so much @lwfinger and the Realtek engineers for all your hard work.

I must also echo the concerns of #193 (@dreamcontrol8000), and state that I'm not able to connect to 802.11x networks that use PEAP/MSCHAPv2 authentication. Maybe this is asking too much, since most users don't have to authenticate under this protocol. I imagine it's difficult to test, given the substantial work to create a test environment that operates under PEAP. But I'll log this anyhow, since it's an issue for me.

[Lee Zamparo (zamparo)]

Sigh. Scratch #357, it was down to a detection issue by network-manager trying Tunneled TLS rather than PEAP. After correcting that, I'm able to connect first try just fine. So, thanks once again for all your diligent work.

[Larry Finger (larry-finger-lwfinger)]

I certainly do not have the test environment to test 802.11x networks. I tried once to implement a radius server to check WPA2 Enterprise, and was never able to get it going.

That said, from the driver's standpoint, it is just passing packets and should not care how the AP is getting its authentication. If you are able to capture the over-the-air packets during the authentication process, or detailed logs from the AP or authentication server, then we might be able to find out what is going wrong

[Feer (r-feer)]

Good evening

I am an owner of a T440 with a rtl card. I had to wait a long time for
Wifi as a simple user. I was a listener to your blog. With your
publication how to install I have now a working cableless internet
connection, stable.
I wish to express you my thanks.

kind regards

Rudolf Feer, MD
Switzerland

On 25.06.2015 19:02, Larry Finger wrote:
> I certainly do not have the test environment to test 802.11x networks. I
> tried once to implement a radius server to check WPA2 Enterprise, and
> was never able to get it going.
>
> That said, from the driver's standpoint, it is just passing packets and
> should not care how the AP is getting its authentication. If you are
> able to capture the over-the-air packets during the authentication
> process, or detailed logs from the AP or authentication server, then we
> might be able to find out what is going wrong
>

[Aaron Landesman (aaronlandesman)]

Hi,

After running the commands listed in #352, I still have no wifi connectivity. (Network Manager shows SSIDs but won't connect to any of them.)

There is relevant output from dmesg:

[ 6.651450] rtl8192ee:_rtl92ee_fw_free_to_go():<0-0> chksum report faill ! REG_MCUFWDL:0x00070600 .
[ 6.651454] rtl8192ee:rtl92ee_download_fw():<0-0> Firmware is not ready to run!

Does anyone have any advice how to fix what's causing these errors?

Thanks.

[Larry Finger (larry-finger-lwfinger)]

Is the firmware installed? What is the output of the command 'ls /lib/firmware/rtlwifi/*'?

[Aaron Landesman (aaronlandesman)]

I'm not certain how to tell whether it's installed, but the output of the command is

/lib/firmware/rtlwifi/Realtek-Firmware-License.txt /lib/firmware/rtlwifi/rtl8192cufw_TMSC.bin /lib/firmware/rtlwifi/rtl8723aufw_B.bin
/lib/firmware/rtlwifi/rtl8188efw.bin /lib/firmware/rtlwifi/rtl8192defw_12.bin /lib/firmware/rtlwifi/rtl8723aufw_B_NoBT.bin
/lib/firmware/rtlwifi/rtl8188eufw.bin /lib/firmware/rtlwifi/rtl8192defw.bin /lib/firmware/rtlwifi/rtl8723befw.bin
/lib/firmware/rtlwifi/rtl8192cfw.bin /lib/firmware/rtlwifi/rtl8192eefw.bin /lib/firmware/rtlwifi/rtl8723fw_B.bin
/lib/firmware/rtlwifi/rtl8192cfwU_B.bin /lib/firmware/rtlwifi/rtl8192eefw_new.bin /lib/firmware/rtlwifi/rtl8723fw.bin
/lib/firmware/rtlwifi/rtl8192cfwU.bin /lib/firmware/rtlwifi/rtl8192sefw.bin /lib/firmware/rtlwifi/rtl8812aefw.bin
/lib/firmware/rtlwifi/rtl8192cufw_A.bin /lib/firmware/rtlwifi/rtl8192sefw.old.bin /lib/firmware/rtlwifi/rtl8812aefw_wowlan.bin
/lib/firmware/rtlwifi/rtl8192cufw_B.bin /lib/firmware/rtlwifi/rtl8712u.bin /lib/firmware/rtlwifi/rtl8821aefw.bin
/lib/firmware/rtlwifi/rtl8192cufw.bin /lib/firmware/rtlwifi/rtl8723aufw_A.bin /lib/firmware/rtlwifi/rtl8821aefw_wowlan.bin

[Larry Finger (larry-finger-lwfinger)]

You have the firmware. Perhaps you have wireless disabled with some switch.

What does 'rfkill list all' show?

[Aaron Landesman (aaronlandesman)]

rfkill list all shows:

1: tpacpi_bluetooth_sw: Bluetooth
 Soft blocked: no
 Hard blocked: no
2: phy0: Wireless LAN
 Soft blocked: no
 Hard blocked: no
44: hci0: Bluetooth
 Soft blocked: no
 Hard blocked: no

[Larry Finger (larry-finger-lwfinger)]

I have no idea what your problem is. When I reboot with the RTL8192EE installed, it connects quickly to one of my APs.

The output of 'dmesg | grep rtl' is

[ 10.146821] rtl8192ee: Using firmware rtlwifi/rtl8192eefw.bin
[ 10.242108] ieee80211 phy1: Selected rate control algorithm 'rtl_rc'
[ 10.283731] rtlwifi: rtlwifi: wireless switch is on
[ 10.297365] rtl8192ee 0000:02:00.0 wlp2s0: renamed from wlan0
[ 20.079071] rtl8192ee 0000:02:00.0 wlp2s0: disabling HT/VHT due to WEP/TKIP use

The last line is because it connected with an 802.11g AP.. To check your firmware, the command 'md5sum /lib/firmware/rtlwifi/rtl8192eefw.bin' should return

c8ced945fe922aa0cdff91f40a77bf12 /lib/firmware/rtlwifi/rtl8192eefw.bin

I get the same results for either the 4.5-rc5 kernel from wireless-drivers-next or the drivers from the rtlwifi_new GitHub repo.

[Aaron Landesman (aaronlandesman)]

When I run your second command md5sum /lib/firmware/rtlwifi/rtl8192eefw.bin, I get what you say. But, when I run your first command, I get a really long string of errors which all look the same:

[247589.091374] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.147927] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.205146] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.261660] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.318961] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.375469] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.432705] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.489215] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.546472] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.603012] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.660306] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.716834] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.774080] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.830580] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.887764] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247589.944263] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.001544] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.058067] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.115331] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.171845] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.229082] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.285600] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.342842] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.399350] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.456888] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.513408] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.570984] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.627507] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.685079] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.741614] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.799105] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.855616] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.913115] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247590.969632] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247591.027117] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247591.083647] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247591.141120] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247591.197651] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247591.255155] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247591.311712] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247591.369410] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247591.425953] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247591.483186] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247591.539686] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247591.596980] rtl8192ee:rtl92ee_hw_init():<0-0> Init MAC failed
[247591.653498] rtl8192ee:rtl92ee_...

[Claudio Pacchierotti (cpacchierotti)]

I am experiencing the same problem as aaronlandesman on a Lenovo x240 running Ubuntu 16.04, kernel 4.4.0-24-generic.

[Ventusfahrer (ventusfahrer)]

Still the same problem on Lenovo T440 on Ubuntu 18.04 AND Ubuntu 18.10.

'Init MAC failed' all the time

[Larry Finger (larry-finger-lwfinger)]

None of my hardware can duplicate this problem. Is the firmware installed? Look at the dmesg output to look for firmware failures.

If you do have the correct firmware, there are two options that might help. Try the following:

sudo modprobe -rv rtl8192ee
sudo modprobe -v rtl8192ee aspm=0

If that does not help, then try

sudo modprobe -rv rtl8192ee
sudo modprobe -v rtl8192ee msi=0

If any other options show up, then your /etc/modprobe.d/ directory is screwed up. Delete the offending files and try again.

If one of those options helps, then create /etc/modprobe.d/50-rtl8192ee.conf (as root), and add the single line containing "options rtl8192ee xxxx=0" where xxxx is the option you want to set.

[Ventusfahrer (ventusfahrer)]

well, meanwhile I installed Fedora 29 on this box in order to find out whether the rtl8192ee will work there - no chance. I am posting the output of various commands based on fedora 29 in the attachment 8192.logs.
Well I know it is Fedora 29, if you need an output based on a current Ubuntu or Kubuntu installation, please let me know, I will retest and document the logs on one of these versions.

Thank you very much for any help!

[Larry Finger (larry-finger-lwfinger)]

I reached my contact at Realtek. He analyzed the output and thinks that this particular chip has a faulty EFUSE programming, or that it possible was not programmed in the factory. Either case should have been caught in testing, but mistakes happen.

Has this card been tested on Windows 10?

[Josh Arenson (josharenson)]

Ah, sorry. I swapped this card for an Intel one years ago. I had never
tried the Realtek card on windows either.

Thanks anyway for following up,
- Josh

On Thu, Jul 25, 2019 at 8:34 AM Larry Finger <email address hidden>
wrote:

> I reached my contact at Realtek. He analyzed the output and thinks that
> this particular chip has a faulty EFUSE programming, or that it possible
> was not programmed in the factory. Either case should have been caught
> in testing, but mistakes happen.
>
> Has this card been tested on Windows 10?
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions
>

--
:wq

[Aaron Landesman (aaronlandesman)]

I made a Windows 7 partition a few years ago and the Realtek card did not work on Windows 7. I'm not sure about Windows 10, though.

[timarmstrong (tim-g-armstrong)]

The card worked reliably for me on windows 8. I solved my problem by buying
a new laptop with an Intel card and giving the old laptop to my
mother-in-law to use for windows.

On Thu., 25 Jul. 2019, 10:45 Aaron Landesman, <email address hidden>
wrote:

> I made a Windows 7 partition a few years ago and the Realtek card did
> not work on Windows 7. I'm not sure about Windows 10, though.
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1239578
>
> Title:
> No wireless support for Realtek RTL8192EE [10ec:818b]
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/hwe-next/+bug/1239578/+subscriptions
>

[Larry Finger (larry-finger-lwfinger)]

If anyone still has this problem, and the card, please run the command 'lspci'. Find the line that describes the RTL8192EE and note the first column, which will be something like "04:00.0". This is the bus and number for the device. Then run 'sudo lspci -s nn:mm -xxxx', where hh:mm is replaced by the bus and number string noted above. Post the output of the last command.

[Gray (graybird)]

~ > sudo lspci -s 03:00 -xxxx
03:00.0 Network controller: Realtek Semiconductor Co., Ltd. RTL8192EE PCIe Wireless Network Adapter (rev ff)
00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
...
ff0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff

---
Just that, ff everywhere.

Kernel: 5.0.0-23

Also, I tried the alternative driver on Larry's github, it works, that is I have connectivity at least but it has some weird issues with bettercap and the monitor mode.