I'll try disabling RTL8192DU and RTL8812AU on 12.0.2 next, to see if disabling them somehow causes the issue (again, i don't expect it too, but you never know).
As expected it did not. I've simply cherry-picked the four commits from the PR that disable them onto the 12.0.2 tag in case anyone wants to replicate it.
So I've bit the bullet bisected commits that touched linux kernel, drivers and LibreELEC options between 12.0.2 and 12.2.1 with
git bisect start '--' 'distributions/LibreELEC/options' 'packages/linux-drivers/' 'packages/linux'
According to bisect, this is the first bad commit
This is my bisect log
git bisect start '--' 'distributions/LibreELEC/options' 'packages/linux-drivers/' 'packages/linux'
# status: waiting for both good and bad commits
# bad: [6209f5a3b067c45fbf527286dd84ee483294b46a] Merge pull request #10633 from heitbaum/binaries-12.2
git bisect bad 6209f5a3b067c45fbf527286dd84ee483294b46a
# status: waiting for good commit(s), bad commit known
# good: [f3fdd11916f8a47dc5a11c3a4c99cb7c7ffac78b] Merge pull request #9707 from chewitt/kodi-tag-revert
git bisect good f3fdd11916f8a47dc5a11c3a4c99cb7c7ffac78b
# bad: [ba3527f093fc645566acb4a445a9f70d0573c59c] Merge pull request #10241 from heitbaum/uplift-kernel-projects-to-master
git bisect bad ba3527f093fc645566acb4a445a9f70d0573c59c
# good: [81c3943df84f052abc72b2571482f2eab5b4007f] mali-midgard: drop package
git bisect good 81c3943df84f052abc72b2571482f2eab5b4007f
# bad: [b9fbadfa84960761e9ab01c0da954cf49f41f533] linux: update to 6.15.6
git bisect bad b9fbadfa84960761e9ab01c0da954cf49f41f533
# skip: [1267bd8d3d8fa0305328bfab220e29803397375f] linux (RPi): apply rtw88 patches
git bisect skip 1267bd8d3d8fa0305328bfab220e29803397375f
# skip: [5d1274fe0966b4ce0400f10cd2c25cd35a078e0a] linux: add rtw88 patches
git bisect skip 5d1274fe0966b4ce0400f10cd2c25cd35a078e0a
# good: [1318b53abc8f210a683076270abcd0e6a97503b8] mali-utgard: drop package
git bisect good 1318b53abc8f210a683076270abcd0e6a97503b8
# bad: [014e83d91f5884dde6be828fb8effd5892487741] linux (RPi): update to 6.12.38-9c09b75
git bisect bad 014e83d91f5884dde6be828fb8effd5892487741
# good: [92c221ad0bbc12e981584e24805dc5eb40c6d7c0] linux: drop ancient patches that never went upstream
git bisect good 92c221ad0bbc12e981584e24805dc5eb40c6d7c0
# first bad commit: [014e83d91f5884dde6be828fb8effd5892487741] linux (RPi): update to 6.12.38-9c09b751267bd8 and 5d1274f do not build (which is also mentioned in https://github.com/LibreELEC/LibreELEC.tv/pull/10247) so i skipped them.
014e83d also did not compile at first due to a missing call_read_iter which is referenced in out-of-kernel driver RTL8192DU and was removed in kernel version 6.10. For my bisect I simply created this patch to get it compling.
diff --git a/os_dep/osdep_service.c b/os_dep/osdep_service.c
index fd36e3b..6a2ae13 100644
--- a/os_dep/osdep_service.c
+++ b/os_dep/osdep_service.c
@@ -56,7 +56,7 @@ static ssize_t new_sync_read(struct file *filp, void __user *buf, __kernel_size_
kiocb.ki_pos = *ppos;
iov_iter_init(&iter, READ, &iov, 1, len);
- ret = call_read_iter(filp, &kiocb, &iter);
+ ret = filp->f_op->read_iter(&kiocb, &iter);
BUG_ON(ret == -EIOCBQUEUED);
*ppos = kiocb.ki_pos;
return ret;I have also opened a issue in the driver repo in case it is helpful to someone.
I'll try the different version of the rpi-kernel (rpi-6.{7,8,9,10,11}.y) next to narrow down when it breaks. Might be a few days until I report back though as I won't be home over the weekend.
I was able to test them before leaving.
rpi-6.7.y (which uses 6.7.12) and rpi-6.8.y (which uses 6.8.12) work, rpi-6.9.y (which uses 6.9.12) breaks.
For the exact config I used see the commits on this branch.
I'll try to git bisect the relevant rpi kernel commits once I get back.
So I started bisecting between the two commits and need(ed) to test a merge base
Bisecting: a merge base must be tested
[e8f897f4afef0031fe618a8e94127a0934896aba] Linux 6.8Unfortunately the build for this commit fails with
...
[QA CHECK] [linux] [safe_remove]:
path does not exist: /build/build.LibreELEC-RPi2.arm-12.2-devel/install_pkg/linux-e8f897f4afef0031fe618a8e94127a0934896aba/usr/share/bootloader/overlay_map.dtb
cp: cannot stat 'arch/arm/boot/dts/overlays/README': No such file or directory
FAILURE: scripts/build linux:target during makeinstall_target (package.mk)
*********** FAILED COMMAND ***********
cp -p arch/${TARGET_KERNEL_ARCH}/boot/dts/overlays/README ${INSTALL}/usr/share/bootloader/overlays
**************************************
*********** FAILED COMMAND ***********
${SCRIPTS}/build "${1}" "${PARENT_PKG}"
**************************************
FAILURE: scripts/install linux:target has failed!There was a similar issue in
Rataplan626 do you remember what you did to solve it (to bisect successfully)?
GamerBene19 it would be good to see connman debug + iwd debug logs during an invalid-key failure, and then the equivalent connman debug + wpa_supplicant debug during a successful connect, with both running on an LE13 master branch base. The goal would be to throw the logs into something like Claude AI with a prompt telling it to trace the interactions and find the root cause of the invalid-key failure. It may come up with a load of garbage. It might come up with something insightful. I've had some success with it tracing complex sequences of activities accross multiple binaries to find bugs in media codecs so you never know.