Bug 218731
Summary: | Tri-band AMD RZ608 (MediaTek MT7921K) has 6GHz band disabled in kernel 6.7+ despite working in <=6.6 | ||
---|---|---|---|
Product: | Drivers | Reporter: | AlexDeLorenzo.dev (kernel) |
Component: | network-wireless | Assignee: | drivers_network-wireless (drivers_network-wireless) |
Status: | NEW --- | ||
Severity: | normal | CC: | pmenzel+bugzilla.kernel.org, regressions |
Priority: | P3 | ||
Hardware: | All | ||
OS: | Linux | ||
Kernel Version: | Subsystem: | ||
Regression: | No | Bisected commit-id: | |
Attachments: |
Kernel 6.6 dmesg
Kernel 6.6 iw list Kernel 6.6 iw reg get Kernel 6.8 dmesg Kernel 6.8 iw list Kernel 6.8 iw reg get |
Description
AlexDeLorenzo.dev
2024-04-15 23:38:25 UTC
Created attachment 306162 [details]
Kernel 6.6 dmesg
Created attachment 306163 [details]
Kernel 6.6 iw list
Created attachment 306164 [details]
Kernel 6.6 iw reg get
Created attachment 306165 [details]
Kernel 6.8 dmesg
Created attachment 306166 [details]
Kernel 6.8 iw list
Created attachment 306167 [details]
Kernel 6.8 iw reg get
Forwarded by mail: https://lore.kernel.org/regressions/e17e5bf2-657a-4a22-a925-94db95fe8ca1@leemhuis.info/T/#u Still an issue in 6.8.7 I am sorry, that you experience an regression. This should not happen, so thank you for reporting it.
> I can provide more logs or help debug the issue if needed.
As no developer has responded yet, bisecting the issue would in my opinion really help. As you can easily reproduce the issue, it’d be great if you could find the commit introducing the regression. If you are more experienced, to avoid rebooting you could pass the WLAN device to a QEMU virtual machine, and quickly boot the Linux kernels. In either case, it should not be more than 20 regression steps, I assume.
(In reply to Paul Menzel from comment #9) > As no developer has responded yet, Maybe waiting a day or two more if the forwarding mentioned earlier leads to some results might be wise, maybe it's known by some developer that was out of office in the past few days. > bisecting the issue would in my opinion really help. FWIW, there is a guide: https://docs.kernel.org/admin-guide/verify-bugs-and-bisect-regressions.html > If you are more experienced, to avoid rebooting you could pass the WLAN > device to a QEMU virtual machine [off topic: maybe it would be wise to add instructions about that to above document, too; not sure.] (In reply to The Linux kernel's regression tracker (Thorsten Leemhuis) from comment #10) > > > bisecting the issue would in my opinion really help. > FWIW, there is a guide: > https://docs.kernel.org/admin-guide/verify-bugs-and-bisect-regressions.html No developer responded, so this is likely needed. Knowing is latest mainline is affected (which the guide makes you check) would be a good start, as that might make some developer look into this. (In reply to Paul Menzel from comment #9) > I am sorry, that you experience an regression. This should not happen, so > thank you for reporting it. > > > I can provide more logs or help debug the issue if needed. > > As no developer has responded yet, bisecting the issue would in my opinion > really help. As you can easily reproduce the issue, it’d be great if you > could find the commit introducing the regression. If you are more > experienced, to avoid rebooting you could pass the WLAN device to a QEMU > virtual machine, and quickly boot the Linux kernels. In either case, it > should not be more than 20 regression steps, I assume. Thanks for the response, I'll look into doing this as the issue persists. (In reply to The Linux kernel's regression tracker (Thorsten Leemhuis) from comment #11) > (In reply to The Linux kernel's regression tracker (Thorsten Leemhuis) from > comment #10) > > > > > bisecting the issue would in my opinion really help. > > FWIW, there is a guide: > > https://docs.kernel.org/admin-guide/verify-bugs-and-bisect-regressions.html > > No developer responded, so this is likely needed. Knowing is latest mainline > is affected (which the guide makes you check) would be a good start, as that > might make some developer look into this. Will do, as well. (In reply to The Linux kernel's regression tracker (Thorsten Leemhuis) from comment #11) > (In reply to The Linux kernel's regression tracker (Thorsten Leemhuis) from > comment #10) > > > > > bisecting the issue would in my opinion really help. > > FWIW, there is a guide: > > https://docs.kernel.org/admin-guide/verify-bugs-and-bisect-regressions.html > > No developer responded, so this is likely needed. Knowing is latest mainline > is affected (which the guide makes you check) would be a good start, as that > might make some developer look into this. Compiled and booted into mainline 6.9-rc6 and this problem is still present. I'm unable to use the 6GHz radio. Built and ran a bunch of kernels, looks like the regression begins to appear in kernel 6.7.0 while it isn't present in 6.6.30. None of the developers responded, so they have no idea; in that case to get some results you most likely have to use a bisection to find the change that introduced the problem (developers then will be obliged to fix this): https://docs.kernel.org/admin-guide/verify-bugs-and-bisect-regressions.html |