Messages in this thread |  | | From | "Regzbot (on behalf of Thorsten Leemhuis)" <> | | Subject | Linux regressions report for mainline [2024-05-12] | | Date | Sun, 12 May 2024 14:02:58 +0000 |
| |
Hi Linus, here is the weekly report for regressions introduced during this cycle. The short story: we have quite a few regressions for which fixes are available (most of them reviewed or even in -next), but only queued for 6.10 (most of them) or not yet fully ready.
This starts with the three fixes I mentioned in my extra report on Thursday[1]. Dave send a revert[2] for the nouveau regression mentioned in that report, but he didn't copy you, so I guess this is meant to wait, too. Side note: the problem from Lyude I mentioned in that report still needs more debugging. [1] https://lore.kernel.org/all/4c7be436-8a5d-469f-b88e-bfe17bafb991@leemhuis.info/ [2] https://lore.kernel.org/all/20240512121630.23898-1-wangjinchao@xfusion.com/
Fixes for a ath11k regression fall through the cracks and will likely be merged soon[3]; the fix for another ath11k regression only reported on Friday exists, but will only be formally submitted in the next few days[4]. [3] https://lore.kernel.org/linux-wireless/20240424064019.4847-1-quic_bqiang@quicinc.com/ [4] https://lore.kernel.org/linux-wireless/D15TIIDIIESY.D1EKKJLZINMA@fairphone.com/ and https://lore.kernel.org/linux-wireless/4d60ccf3-455d-4189-9100-d35488b00236@quicinc.com/
For the record: Kalle yesterday reported a suspend regression that made stress test stalls within 30 minutes, but that one is not bisected yet[5]. [5] https://lore.kernel.org/lkml/87o79cjjik.fsf@kernel.org/
Ciao, Thorsten
---
Hi, this is regzbot, the Linux kernel regression tracking bot.
Currently I'm aware of 10 regressions in linux-mainline. Find the current status below and the latest on the web:
https://linux-regtracking.leemhuis.info/regzbot/mainline/
Bye bye, hope to see you soon for the next report. Regzbot (on behalf of Thorsten Leemhuis)
====================================================== current cycle (v6.8.. aka v6.9-rc), culprit identified ======================================================
[ *NEW* ] ath11k: connection to 6 GHz AP fails ---------------------------------------------- https://linux-regtracking.leemhuis.info/regzbot/regression/lore/87wmo0k71i.fsf@kernel.org/ https://lore.kernel.org/linux-wireless/87wmo0k71i.fsf@kernel.org/
By Kalle Valo; 1 days ago; 1 activities, latest 1 days ago. Introduced in bc8a0fac8677 (v6.9-rc1)
Fix incoming: * wifi: ath11k: move power type check to ASSOC stage when connecting to 6 GHz AP https://lore.kernel.org/regressions/0834b74a-b360-4404-994b-7f20c89d7257@leemhuis.info/
[ *NEW* ] ath11k: WCN6750 firmware crashes during initialisation ---------------------------------------------------------------- https://linux-regtracking.leemhuis.info/regzbot/regression/lore/D15TIIDIIESY.D1EKKJLZINMA@fairphone.com/ https://lore.kernel.org/linux-wireless/D15TIIDIIESY.D1EKKJLZINMA@fairphone.com/
By Luca Weiss; 2 days ago; 10 activities, latest 1 days ago. Introduced in f019f4dff2e4 (v6.9-rc1)
Recent activities from: Carl Huang (5), Luca Weiss (3), Kalle Valo (2)
2 patch postings are associated with this regression, the latest is this: * Re: [PATCH 1/2] wifi: ath11k: supports 2 station interfaces https://lore.kernel.org/linux-wireless/4d60ccf3-455d-4189-9100-d35488b00236@quicinc.com/ 2 days ago, by Carl Huang
[ *NEW* ] nouveau: init failed, no display output ------------------------------------------------- https://linux-regtracking.leemhuis.info/regzbot/regression/lore/20240506182331.8076-1-dan@danm.net/ https://lore.kernel.org/stable/20240506182331.8076-1-dan@danm.net/ https://lore.kernel.org/stable/20240506185910.17917-1-dan@danm.net/
By Dan Moulding and Dan Moulding; 5 days ago; 4 activities, latest 2 days ago. Introduced in 52a6947bf576 (v6.9-rc7)
Recent activities from: Dan Moulding (2), David Airlie (1), Linux regression tracking (Thorsten Leemhuis) (1)
x86/topology: system stopped booting ------------------------------------ https://linux-regtracking.leemhuis.info/regzbot/regression/lore/3d77cb89857ee43a9c31249f4eab7196013bc4b4.camel@redhat.com/ https://lore.kernel.org/lkml/3d77cb89857ee43a9c31249f4eab7196013bc4b4.camel@redhat.com/
By Lyude Paul; 24 days ago; 19 activities, latest 2 days ago. Introduced in f0551af02130 (v6.9-rc1)
Recent activities from: Lyude Paul (3), Thomas Gleixner (2), Mario Limonciello (1)
2 patch postings are associated with this regression, the latest is this: * Re: Early boot regression from f0551af0213 ("x86/topology: Ignore non-present APIC IDs in a present package") https://lore.kernel.org/lkml/87le59vw1y.ffs@tglx/ 22 days ago, by Thomas Gleixner
net: Bluetooth: firmware loading problems with older firmware ------------------------------------------------------------- https://linux-regtracking.leemhuis.info/regzbot/regression/lore/20240401144424.1714-1-mike@fireburn.co.uk/ https://lore.kernel.org/lkml/20240401144424.1714-1-mike@fireburn.co.uk/
By Mike Lothian; 40 days ago; 12 activities, latest 3 days ago. Introduced in 1cb63d80fff6 (v6.9-rc1)
Fix incoming: * Bluetooth: btusb: Fix the patch for MT7920 the affected to MT7921 https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?h=master&id=5e970bdf73e619935a8977f9f22b581480211164
Ryzen 7840HS CPU single core never boosts to max frequency ---------------------------------------------------------- https://linux-regtracking.leemhuis.info/regzbot/regression/bugzilla.kernel.org/218759/ https://bugzilla.kernel.org/show_bug.cgi?id=218759
By Gaha; 20 days ago; 45 activities, latest 4 days ago. Introduced in f3a052391822 (v6.9-rc1)
Fix incoming: * cpufreq: amd-pstate: fix the highest frequency issue which limit performance https://lore.kernel.org/regressions/0834b74a-b360-4404-994b-7f20c89d7257@leemhuis.info/
leds: hangs on boot ------------------- https://linux-regtracking.leemhuis.info/regzbot/regression/lore/30f757e3-73c5-5473-c1f8-328bab98fd7d@candelatech.com/ https://lore.kernel.org/lkml/30f757e3-73c5-5473-c1f8-328bab98fd7d@candelatech.com/
By Ben Greear; 39 days ago; 29 activities, latest 4 days ago. Introduced in f5c31bcf604d (v6.9-rc1)
Fix incoming: * wifi: iwlwifi: Use request_module_nowait https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?h=master&id=3d913719df14c28c4d3819e7e6d150760222bda4
kbuild: deb-pkg and bindeb-pkg: build failure (and slowness, too) ----------------------------------------------------------------- https://linux-regtracking.leemhuis.info/regzbot/regression/lore/ZjcRPelwZP34N42s@gmail.com/ https://lore.kernel.org/lkml/ZjcRPelwZP34N42s@gmail.com/
By Ingo Molnar; 7 days ago; 5 activities, latest 7 days ago. Introduced in 1d7bae8f8c85 (v6.9-rc1)
=================================================== current cycle (v6.8.. aka v6.9-rc), unknown culprit ===================================================
[ *NEW* ] x86: mitigations: suspend stress test stalls within 30 minutes ------------------------------------------------------------------------ https://linux-regtracking.leemhuis.info/regzbot/regression/lore/87o79cjjik.fsf@kernel.org/ https://lore.kernel.org/lkml/87o79cjjik.fsf@kernel.org/
By Kalle Valo; 0 days ago; 4 activities, latest 0 days ago. Introduced in v6.9-rc3..v6.9-rc6
Recent activities from: Kalle Valo (2), Borislav Petkov (2)
net: Beaglebone Ethernet Probe Failure In 6.8+ ---------------------------------------------- https://linux-regtracking.leemhuis.info/regzbot/regression/lore/Zh/tyozk1n0cFv+l@euler/ https://lore.kernel.org/netdev/Zh%2Ftyozk1n0cFv%2Bl@euler/
By Colin Foster; 24 days ago; 8 activities, latest 13 days ago. Introduced in v6.8..v6.9-rc5
One patch associated with this regression: * Re: Beaglebone Ethernet Probe Failure In 6.8+ https://lore.kernel.org/netdev/Zicyc0pj3g7%2FMemK@euler/ 19 days ago, by Colin Foster
============= End of report =============
All regressions marked '[ *NEW* ]' were added since the previous report, which can be found here: https://lore.kernel.org/r/171491873337.1284823.18146984489788770141@leemhuis.info
Thanks for your attention, have a nice day!
Regzbot, your hard working Linux kernel regression tracking robot
P.S.: Wanna know more about regzbot or how to use it to track regressions for your subsystem? Then check out the getting started guide or the reference documentation:
https://gitlab.com/knurd42/regzbot/-/blob/main/docs/getting_started.md https://gitlab.com/knurd42/regzbot/-/blob/main/docs/reference.md
The short version: if you see a regression report you want to see tracked, just send a reply to the report where you Cc regressions@lists.linux.dev with a line like this:
#regzbot introduced: v5.13..v5.14-rc1
If you want to fix a tracked regression, just do what is expected anyway: add a 'Link:' tag with the url to the report, e.g.:
Link: https://lore.kernel.org/all/30th.anniversary.repost@klaava.Helsinki.FI/
|  |