Bug 25189

Thanks Tony for the alert.
Not sure where to assign this: kernel or MageiaTools. Go for the latter to start with because of MCC hardware detection; it can be re-assigned.

Assignee: bugsquad => mageiatools

Time to reassign, nothing happened.

CC: (none) => yves.brungard_mageia
Assignee: mageiatools => kernel

I first thought this was supposed to be handled by the kernel after reading https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=5407fd7de69f3352aed659244d4bef18e3cabf5c 

But this is actually inside #ifdef CONFIG_RTL8XXXU_UNTESTED which doesn't give much hope.

I see we have dkms-rtl8192eu, can you try installing it?

CC: (none) => pterjan

Hello,

I  don't have the hardware for testing, buttried to install the package dkms-rtl8192eu under Cauldron and the dkms module doesn't compile.

Here is the result of the make.log

"DKMS make.log for rtl8192eu-4.3.1.1_11320.20140505-5.mga8 for kernel 5.5.7-desktop-1.mga8 (x86_64)
sam. 29 févr. 2020 23:48:33 CET
make ARCH=x86_64 CROSS_COMPILE= -C /lib/modules/5.5.7-desktop-1.mga8/build M=/var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-5.mga8/build modules
make[1] : on entre dans le répertoire « /usr/src/kernel-5.5.7-desktop-1.mga8 »
CC [M] /var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-5.mga8/build/core/rtw_cmd.o
Dans le fichier inclus depuis /var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-5.mga8/build/include/osdep_service.h:41,
depuis /var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-5.mga8/build/include/drv_types.h:32,
depuis /var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-5.mga8/build/core/rtw_cmd.c:22:
/var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-5.mga8/build/include/osdep_service_linux.h: Dans la fonction « _init_timer »:
/var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-5.mga8/build/include/osdep_service_linux.h:253:8: erreur: « _timer » {alias « struct timer_list »} n'a pas de membre nommé « data »
253 | ptimer->data = (unsigned long)cntx;
| ^~
/var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-5.mga8/build/include/osdep_service_linux.h:254:2: erreur: déclaration implicite de la fonction « init_timer »; vouliez-vous utiliser « _init_timer » ? [-Werror=implicit-function-declaration]
254 | init_timer(ptimer);
| ^~~~~~~~~~
| _init_timer
cc1 : certains avertissements sont traités comme des erreurs
make[2]: *** [scripts/Makefile.build:266 : /var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-5.mga8/build/core/rtw_cmd.o] Erreur 1
make[1]: *** [Makefile:1685 : /var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-5.mga8/build] Erreur 2
make[1] : on quitte le répertoire « /usr/src/kernel-5.5.7-desktop-1.mga8 »
make: *** [Makefile:1323 : modules] Erreur 2"

I also tested under Mageia 7 and get the same error

"DKMS make.log for rtl8192eu-4.3.1.1_11320.20140505-4.mga7 for kernel 5.5.4-desktop-1.mga7 (x86_64)
dim. mars  1 09:00:13 CET 2020
make ARCH=x86_64 CROSS_COMPILE= -C /lib/modules/5.5.4-desktop-1.mga7/build M=/var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-4.mga7/build  modules
make[1] : on entre dans le répertoire « /usr/src/kernel-5.5.4-desktop-1.mga7 »
  CC [M]  /var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-4.mga7/build/core/rtw_cmd.o
In file included from /var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-4.mga7/build/include/osdep_service.h:41,
                 from /var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-4.mga7/build/include/drv_types.h:32,
                 from /var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-4.mga7/build/core/rtw_cmd.c:22:
/var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-4.mga7/build/include/osdep_service_linux.h: Dans la fonction « _init_timer »:
/var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-4.mga7/build/include/osdep_service_linux.h:253:8: error: « _timer » {alias « struct timer_list »} n'a pas de membre nommé « data »
  ptimer->data = (unsigned long)cntx;
        ^~
/var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-4.mga7/build/include/osdep_service_linux.h:254:2: error: déclaration implicite de la fonction « init_timer »; vouliez-vous utiliser « _init_timer » ? [-Werror=implicit-function-declaration]
  init_timer(ptimer);
  ^~~~~~~~~~
  _init_timer
cc1 : certains avertissements sont traités comme des erreurs
make[2]: *** [scripts/Makefile.build:266: /var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-4.mga7/build/core/rtw_cmd.o] Error 1
make[1]: *** [Makefile:1685: /var/lib/dkms/rtl8192eu/4.3.1.1_11320.20140505-4.mga7/build] Error 2
make[1] : on quitte le répertoire « /usr/src/kernel-5.5.4-desktop-1.mga7 »
make: *** [Makefile:1323: modules] Error 2"

Note: As the 8192eu is also handled by the rtl8xxxu module ( but seems not to allow connexion to wireless access point; see the in french topic on mlo forum https://www.mageialinux-online.org/forum/topic-27292-2+mageia-8-wifi.php), i think this package should blacklist the rtl8xxxu ( i have not checked if it is the case by now).

CC: (none) => joe_c_moi

I have switched the package to https://github.com/Mange/rtl8192eu-linux-driver which contains fixes for recent kernels and it now builds

(In cauldron)

Hello,

It seems that using the last commit of the github, some box are not visible ( so not possible to connect on it).

The lasts commits of the Mange git are based on a 5.6.4 version. I think if we use this git, we need to limit to a commit before the 5.6.4 version introduced in the repository (for example the 13b9c39980a14bebe5178c044cd24ea8aeb845da ( that seems to still working on 5.5.8 kernel), i have not tested the 2 or 3 after this one that are not using a 5.6.4 version).

During the search, i find a repository at https://github.com/clnhub/rtl8192eu-linux which look to have a more recent Realtek driver (but don't know if it works).

On build instruction ( either "4.4.1" or 5.2.19.1), it is specified to blacklist the rtl8xxxu module ( this module seems to also support 8192eu chipset, but it seems that connexion not working with this one)).

I think that this bug is a duplicate of 
https://bugs.mageia.org/show_bug.cgi?id=22599 and (perhaps) https://bugs.mageia.org/show_bug.cgi?id=20508

CC: (none) => joe_c_moi

Hmmm, this hasn't bitten me recently.  In all honesty I don't recall which bit of our forest of hardware contained the cardlo
Probably best to colse it; can always re=-open if issue arises again.
Tony

Status: NEW => RESOLVED
Resolution: (none) => OLD