Bug 30856

Summary: add 8188eu driver, replace r8188eu
Product: Sisyphus Reporter: Ivan Zakharyaschev <imz>
Component: kernel-image-un-defAssignee: Vitaly Chikunov <vt>
Status: NEW --- QA Contact: qa-sisyphus
Severity: normal    
Priority: P3 CC: kernelbot, placeholder, vt
Version: unstable   
Hardware: all   
OS: Linux   
URL: http://www.altlinux.org/Сборка_модулей_ядра
Bug Depends on:    
Bug Blocks: 29888    

Description Ivan Zakharyaschev 2015-03-24 21:07:19 MSK
kernel-image-un-def-3.19.2-alt1

I have read that there alternative versions of the driver -- https://github.com/lwfinger/rtl8188eu/issues/105 .

The one currently included in the package is:

filename:       /lib/modules/3.19.2-un-def-alt1/kernel/drivers/staging/rtl8188eu/r8188eu.ko
version:        v4.1.4_6773.20130222
author:         Realtek Semiconductor Corp.

lwfinger has 4.1.8 and other versions -- https://github.com/lwfinger/rtl8188eu/issues/100 .

He ships his versions under the name 8188eu to avoid a conflict.

The current r8188eu works very bad in my case. I'm not sure why, but the lwfinger's one might be better.

Why not include it?
Comment 1 Ivan Zakharyaschev 2015-03-24 21:58:29 MSK
What does "works badky" mean in my case?

It means that 50% of pings are lost.

It means that downloading from ftp.altlinux.org goes at speed about 50kB/s whereas another nearby computer connected to the same WiFi network does this at 1100kB/s.

And it means that there is a long waiting for establishing connections, which makes it almost impossible to use.

Any log messages?.. Perhaps, from dmesg:

[   94.088489] R8188EU: ERROR indicate disassoc
[   94.090635] IPv6: ADDRCONF(NETDEV_UP): wlp0s26u1u1: link is not ready
[   94.226706] NET: Registered protocol family 17
[   96.884970] R8188EU: ERROR assoc success
[   96.919094] R8188EU: ERROR indicate disassoc
[   96.989283] R8188EU: ERROR assoc success
[  134.243616] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
[  134.253956] R8188EU: ERROR ap recv disassoc reason code(7) sta:5c:d9:98:66:55:72
[  134.254695] R8188EU: ERROR ap recv disassoc reason code(7) sta:5c:d9:98:66:55:72
[  134.255442] R8188EU: ERROR ap recv disassoc reason code(7) sta:5c:d9:98:66:55:72
[  134.256320] R8188EU: ERROR ap recv disassoc reason code(7) sta:5c:d9:98:66:55:72
[  136.235031] R8188EU: ERROR assoc success
[  150.290275] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
[  152.324974] R8188EU: ERROR assoc success
[  174.352399] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
[  180.768989] R8188EU: ERROR indicate disassoc
[  183.740954] R8188EU: ERROR indicate disassoc
[  186.917033] R8188EU: ERROR indicate disassoc
[  190.196169] R8188EU: ERROR indicate disassoc
[  199.253810] R8188EU: ERROR assoc success
[  234.376214] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
[  234.386445] R8188EU: ERROR ap recv disassoc reason code(7) sta:5c:d9:98:66:55:72
[  234.387195] R8188EU: ERROR ap recv disassoc reason code(7) sta:5c:d9:98:66:55:72
[  234.387945] R8188EU: ERROR ap recv disassoc reason code(7) sta:5c:d9:98:66:55:72
[  236.550274] R8188EU: ERROR assoc success
[  256.437811] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
[  256.447901] R8188EU: ERROR ap recv disassoc reason code(7) sta:5c:d9:98:66:55:72
[  256.448647] R8188EU: ERROR ap recv disassoc reason code(7) sta:5c:d9:98:66:55:72
[  258.489133] R8188EU: ERROR assoc success
[  258.528910] R8188EU: ERROR sta recv deauth reason code(14) sta:5c:d9:98:66:55:72
[  258.535371] R8188EU: ERROR indicate disassoc
[  258.544028] R8188EU: ERROR sta recv deauth reason code(7) sta:5c:d9:98:66:55:72
[  258.544762] R8188EU: ERROR sta recv deauth reason code(7) sta:5c:d9:98:66:55:72
[  258.545528] R8188EU: ERROR sta recv deauth reason code(7) sta:5c:d9:98:66:55:72
[  258.546414] R8188EU: ERROR sta recv deauth reason code(7) sta:5c:d9:98:66:55:72
[  258.547139] R8188EU: ERROR sta recv deauth reason code(7) sta:5c:d9:98:66:55:72
[  260.124056] R8188EU: ERROR indicate disassoc
[  261.958252] R8188EU: ERROR indicate disassoc
[  263.505266] R8188EU: ERROR indicate disassoc
[  277.347358] R8188EU: ERROR indicate disassoc
[  293.118729] R8188EU: ERROR indicate disassoc
[  302.343752] R8188EU: ERROR indicate disassoc
[  318.018032] R8188EU: ERROR indicate disassoc
[  328.460918] R8188EU: ERROR assoc success
[  363.256763] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
[  365.250693] R8188EU: ERROR assoc success
[  403.366329] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
[  405.418268] R8188EU: ERROR assoc success
[  441.468172] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
[  443.535392] R8188EU: ERROR assoc success
[  475.554920] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
[  477.553093] R8188EU: ERROR assoc success
[  525.687348] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
[  527.762811] R8188EU: ERROR assoc success
[  561.786150] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
[  563.830345] R8188EU: ERROR assoc success
[  599.884132] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
[  601.947717] R8188EU: ERROR assoc success
[  627.959475] R8188EU: ERROR linked_status_chk(wlp0s26u1u1) disconnect or roaming
Comment 2 Anton Farygin 2015-03-24 22:22:28 MSK
Ivan, it's need to be building in separated package. 
Will you  be testing new driver before packaging ?
it's easy to build for installed kernel.
Comment 3 Ivan Zakharyaschev 2015-03-25 21:10:29 MSK
(В ответ на комментарий №2)
> Ivan, it's need to be building in separated package. 
> Will you  be testing new driver before packaging ?
> it's easy to build for installed kernel.

Yes, I should try it.

Testing before packaging?.. Hmm, I haven't thought about that because I like building in hasher, i.e., with gear-hsh .

But perhaps in the case of a kernel module, installing the build deps (kernel-headers and gcc) is not a big problem.

Also I had the idea that the package could be useful for other people to try it out if they have a similar hardware, even if my problems are not solved with the alternative driver.

That's not so strange (that someone would want to try it even if my problems are not solved) because there are people (on the web) who are using it more or less successfully.
Comment 4 Anton Farygin 2015-03-25 21:54:14 MSK
I think,that not a problem build 8188eu.ko in hasher environment without rpm.

Try to install all deps (kernel-headers, gcc) into hasher root via hsh-initroot and hsh-install and then copy tarball with driver sources to hasher's chroot ".in/" directory.

Next, you will have to build driver from this source like in usual system - untar, and follow instruction from tarball.
Comment 5 Anton Farygin 2015-03-25 21:58:02 MSK
Also, why did need to test driver before packaging ?
This is philosophic  question. if you have access to hardware, then why not?
Comment 6 Michael Shigorin 2015-03-27 00:21:30 MSK
(В ответ на комментарий №3)
> Also I had the idea that the package could be useful for other people to try it
> out if they have a similar hardware, even if my problems are not solved with
> the alternative driver.
Во-первых, у тебя железо -- тебе и карты в руки (как минимум собрать сбоку модуль для проверки, если он вообще собирается, обычно недолго) ;-)

Во-вторых, не стоит размножать баги раньше времени, а затем дублировать вновь появившиеся соображения по ним: _если_ в итоге будет сделано для un-def в сизифе, _тогда_ будет смысл далее проверять на std-def там же, а затем, возможно, и в бранче.  А от такого дублирования больше вреда, чем пользы.