Re: next-20250623: arm64 devices kernel panic Internal error Oops at pidfs_free_pid (fs/pidfs.c:162)

[Christian Brauner <brauner@kernel.org>]

On Mon, Jun 23, 2025 at 08:26:32PM +0530, Naresh Kamboju wrote:
> On Mon, 23 Jun 2025 at 18:26, Christian Brauner <brauner@kernel.org> wrote:
> >
> > On Mon, Jun 23, 2025 at 05:29:38PM +0530, Naresh Kamboju wrote:
> > > Regressions on arm64 devices and qemu-arm64 while running LTP controllers
> > > and selftests cgroup test cases the following kernel Panic Internal error oops
> > > found on the Linux next-20250623 tag.
> > >
> > > Regressions found on arm64 device
> > >   - Kernel Panic Internal oops @ LTP controllers
> > >   - Kernel Panic Internal oops @ selftest cgroups
> > >
> > > Test environments:
> > >  - Dragonboard-410c
> > >  - e850-96
> > >  - FVP
> > >  - Juno-r2
> > >  - rk3399-rock-pi-4b
> > >  - qemu-arm64
> > >
> > > Regression Analysis:
> > >  - New regression? Yes
> > >  - Reproducibility? Yes
> > >
> > > Boot regression: arm64 devices kernel panic Internal error Oops at
> > > pidfs_free_pid (fs/pidfs.c:162)
> > >
> > > Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> > >
> > > ## Test log
> > > [   67.087303] Internal error: Oops: 0000000096000004 [#1]  SMP
> > > [   67.094021] Modules linked in: snd_soc_hdmi_codec venus_enc
> > > venus_dec videobuf2_dma_contig pm8916_wdt qcom_wcnss_pil
> > > coresight_cpu_debug coresight_tmc coresight_replicator qcom_camss
> > > coresight_stm snd_soc_lpass_apq8016 msm qrtr coresight_funnel
> > > snd_soc_msm8916_digital snd_soc_lpass_cpu coresight_tpiu
> > > snd_soc_msm8916_analog videobuf2_dma_sg stm_core coresight_cti
> > > snd_soc_lpass_platform snd_soc_apq8016_sbc venus_core
> > > snd_soc_qcom_common qcom_q6v5_mss v4l2_fwnode coresight snd_soc_core
> > > qcom_pil_info v4l2_async snd_compress llcc_qcom snd_pcm_dmaengine
> > > ocmem qcom_q6v5 v4l2_mem2mem videobuf2_memops snd_pcm qcom_sysmon
> > > drm_exec adv7511 snd_timer videobuf2_v4l2 gpu_sched qcom_common snd
> > > videodev drm_dp_aux_bus qcom_glink_smem soundcore qcom_spmi_vadc
> > > mdt_loader drm_display_helper qnoc_msm8916 qmi_helpers
> > > videobuf2_common qcom_vadc_common qcom_spmi_temp_alarm rtc_pm8xxx
> > > qcom_pon qcom_stats mc cec drm_client_lib qcom_rng rpmsg_ctrl
> > > display_connector rpmsg_char phy_qcom_usb_hs socinfo drm_kms_helper
> > > rmtfs_mem ramoops
> > > [   67.094437]  reed_solomon fuse drm backlight ip_tables x_tables
> > > [   67.189084] CPU: 3 UID: 0 PID: 0 Comm: swapper/3 Not tainted
> > > 6.16.0-rc3-next-20250623 #1 PREEMPT
> > > [   67.194810] Hardware name: Qualcomm Technologies, Inc. APQ 8016 SBC (DT)
> > > [   67.234078] pc : pidfs_free_pid (fs/pidfs.c:162)
> >
> > Thanks, I see the issue. I'm pushing out a fix. Please let me know if
> > that reproduces in the next few days.
> 
> Thanks. Please share the proposed fix patches.
> I would like to build and test in LKFT test framework.

It's in vfs-6.17.pidfs. Syzbot has a reproducer for the bug.