Bugzilla – Bug 1219258
add binder kernel module
Last modified: 2024-01-29 09:24:49 UTC
For waydroid and anbox, as well as similar programs to work, the binder kernel module is required, which is not in the standard opensuse kernel. is it possible to add it as kmp? > [werwolf@home] ~ > ❯ zgrep BINDER /proc/config.gz > # CONFIG_ANDROID_BINDER_IPC is not set > > [werwolf@home] ~ > ❯ rpm -qf /boot/vmlinuz-6.7.1-2-default > kernel-default-6.7.1-2.1.x86_64 > > [werwolf@home] ~ > ❯ LANG=c zypper info kernel-default > Loading repository data... > Reading installed packages... > > > Information for package kernel-default: > --------------------------------------- > Repository : repo-oss > Name : kernel-default > Version : 6.7.1-2.1 > Arch : x86_64 > Vendor : openSUSE > Installed Size : 240.3 MiB > Installed : Yes > Status : up-to-date > Source package : kernel-default-6.7.1-2.1.nosrc > Upstream URL : https://www.kernel.org/ > Summary : The Standard Kernel > Description : > The standard kernel for both uniprocessor and multiprocessor systems. > > > Source Timestamp: 2024-01-26 06:32:58 +0000 > GIT Revision: 6aca25404404e4f72fa252563503b29fe5828cca > GIT Branch: stable
> [werwolf@home] ~ > ❯ sudo waydroid init > [22:48:55] Failed to load binder driver > [22:48:55] modprobe: FATAL: Module binder_linux not found in directory /usr/lib/modules/6.7.1-2-default > [22:48:55] ERROR: Binder node "binder" for waydroid not found > [22:48:55] See also: <https://github.com/waydroid> > Run 'waydroid log' for details. > > [werwolf@home] ~ > ❯ LANG=c zypper info waydroid > Loading repository data... > Reading installed packages... > > > Information for package waydroid: > --------------------------------- > Repository : home:Dead_Mozay:waydroid > Name : waydroid > Version : 1.4.1-2.4 > Arch : noarch > Vendor : obs://build.opensuse.org/home:Dead_Mozay > Installed Size : 393.8 KiB > Installed : Yes > Status : up-to-date > Source package : waydroid-1.4.1-2.4.src > Upstream URL : https://github.com/waydroid/waydroid > Summary : Uses a container-based approach to boot a full Android system > Description : > Waydroid uses Linux namespaces (user, pid, uts, net, mount, ipc) to run a full Android system > in a container and provide Android applications on any GNU/Linux-based platform. > The Android system inside the container has direct access to any needed hardware. > The Android runtime environment ships with a minimal customized Android system image based on LineageOS. > The image is currently based on Android 10.
. *** This bug has been marked as a duplicate of bug 1189456 ***
(In reply to hui from comment #2) > . > > *** This bug has been marked as a duplicate of bug 1189456 *** Sorry, but Bug 1219258 is closed as WONTFIX, but as you can see from the message above, binder was never enabled in the kernel build config. as dead_mozay@opensuse.org writes, no third-party patches or assembly from third-party repositories are required, it’s enough just to correct a few lines in the existing config. Should this be regarded as a refusal or an impossibility? I reopened the report to get a final answer like "this will never be done because..." or "this will be done in such and such a release"
(In reply to Dmitry Markov from comment #3) > Should this be regarded as a refusal or an impossibility? > I reopened the report to get a final answer like "this will never be done > because..." or "this will be done in such and such a release" Please read comment#1 in bsc#1189456 for an explanation.
(In reply to Ivan Ivanov from comment #4) > (In reply to Dmitry Markov from comment #3) > > > Should this be regarded as a refusal or an impossibility? > > I reopened the report to get a final answer like "this will never be done > > because..." or "this will be done in such and such a release" > > Please read comment#1 in bsc#1189456 for an explanation. does this mean that I should have reopened not this request but bsc#1189456 with the same question? Sorry, I really didn't understand.
(In reply to Dmitry Markov from comment #5) > (In reply to Ivan Ivanov from comment #4) > > (In reply to Dmitry Markov from comment #3) > > > > > Should this be regarded as a refusal or an impossibility? > > > I reopened the report to get a final answer like "this will never be done > > > because..." or "this will be done in such and such a release" > > > > Please read comment#1 in bsc#1189456 for an explanation. > > does this mean that I should have reopened not this request but bsc#1189456 > with the same question? > Sorry, I really didn't understand. To be considerd for enabling "Android IPC" should be able to be build as "loadable modules". This is not currently possible. This should be "fixed" upstream first (like in the Linus Torvalds tree). CONFIG_ANDROID_BINDER_IPC=m
(In reply to Ivan Ivanov from comment #6) > (In reply to Dmitry Markov from comment #5) > > (In reply to Ivan Ivanov from comment #4) > > > (In reply to Dmitry Markov from comment #3) > > > > > > > Should this be regarded as a refusal or an impossibility? > > > > I reopened the report to get a final answer like "this will never be done > > > > because..." or "this will be done in such and such a release" > > > > > > Please read comment#1 in bsc#1189456 for an explanation. > > > > does this mean that I should have reopened not this request but bsc#1189456 > > with the same question? > > Sorry, I really didn't understand. > > To be considerd for enabling "Android IPC" should be able to be build as > "loadable modules". This is not currently possible. This should be > "fixed" upstream first (like in the Linus Torvalds tree). > > CONFIG_ANDROID_BINDER_IPC=m oh, now I understand. required kernel modules cannot be built as loadable ones. Thanks for clarifying. *** This bug has been marked as a duplicate of bug 1189456 ***