[f2fs-dev] F2FS kernel panic - latest 5.10 stable

2023-03-23 Thread Micha L via Linux-f2fs-devel
Hello everyone,

I'm maintaining a kernel for Pixel 7 android devices running the 5.10
kernel. After merging the latest updates from f2fs stable for 5.10,
users experience a kernel panic, which I believe is caused by the most
recently pushed tag there.

I never personally experienced the panic, but I got two ramoops/pstore
logs that look exactly the same since I pushed the kernel containing
the latest f2fs stable commits. One panic happened while the user was
browsing chrome, the other while watching YouTube.
Both crashes look identical in the kernel log.


Here's the snipped from the kernel log:
[  271.459684][ T9723] Unable to handle kernel NULL pointer
dereference at virtual address 
[  271.459713][ T9723] Mem abort info:
[  271.459725][ T9723]   ESR = 0x9606
[  271.459737][ T9723]   EC = 0x25: DABT (current EL), IL = 32 bits
[  271.459746][ T9723]   SET = 0, FnV = 0
[  271.459755][ T9723]   EA = 0, S1PTW = 0
[  271.459763][ T9723] Data abort info:
[  271.459772][ T9723]   ISV = 0, ISS = 0x0006
[  271.459782][ T9723]   CM = 0, WnR = 0
[  271.459793][ T9723] user pgtable: 4k pages, 39-bit VAs, pgdp=d0a97000
[  271.459803][ T9723] [] pgd=000949ddc003,
p4d=000949ddc003, pud=000949ddc003, pmd=
[  271.459831][ T9723] Internal error: Oops: 9606 [#1] PREEMPT SMP
[  271.459851][ T9723] debug-snapshot dss: core register saved(CPU:5)
[  271.459862][ T9723] debug-snapshot dss: ECC error check erridr_el1.num = 0x2
[  271.459874][ T9723] debug-snapshot dss: ERRSELR_EL1.SEL = 0, NOT
Error, ERXSTATUS_EL1 = 0x0
[  271.459884][ T9723] debug-snapshot dss: ERRSELR_EL1.SEL = 1, NOT
Error, ERXSTATUS_EL1 = 0x0
[  271.459892][ T9723] debug-snapshot dss: context saved(CPU:5)
[  271.460130][ T9723] item - log_kevents is disabled
[  271.460135][ T9723] Modules linked in: snd_soc_cs40l26(O)
input_cs40l26_i2c(O) cl_dsp(O) syna_touch(O) bcmdhd4389(O)
wlan_ptracker(O) snd_soc_cs35l45_i2c(O) snd_soc_cs35l41_i2c(O)
overheat_mitigation(O) mcps802154_region_pctt(O) max77729_charger(O)
mali_kbase(O) mali_pixel(O) lwis(O) gxp(O) google_dual_batt_gauge(O)
google_dock(O) goog_touch_interface(O) drv2624(O) bigocean(O)
audiometrics(O) coresight_etm4x coresight_replicator coresight_funnel
coresight_tmc coresight pps_gpio pps_core arm_dsu_pmu gpu_cooling
goodixfp stmvl53l1 usbc_cooling_dev usb_f_etr_miu scsi_debug
crc_t10dif slg51002_core slg51000_core sbb_mux null_blk
slg51002_regulator slg51000_regulator pixel_stat_mm pixel_stat_sysfs
bcm_dbg pixel_em vh_mm pinctrl_slg51002 pinctrl_slg51000
crct10dif_generic crct10dif_common touch_bus_negotiator(O)
touch_offload(O) st21nfc(O) snd_soc_cs35l45_spi(O) snd_soc_cs35l45(O)
snd_soc_cs35l41_spi(O) snd_soc_cs35l41(O) snd_soc_wm_adsp(O)
pca9468(O) panel_samsung_sofef01(O)
[  271.460242][ T9723]  panel_samsung_s6e3hc4(O)
panel_samsung_s6e3hc3(O) panel_samsung_s6e3hc3_c10(O)
panel_samsung_s6e3hc2(O) panel_samsung_s6e3fc3(O)
panel_samsung_s6e3fc3_p10(O) panel_samsung_emul(O)
panel_boe_nt37290(O) panel_samsung_drv(O) p9221(O) nitrous(O)
mcps802154_region_nfcc_coex(O) mcps802154_region_fira(O)
max77759_charger(O) max77729_uic(O) max77729_pmic(O) max20339(O)
max1720x_battery(O) mailbox_wc(O) janeiro(O) heatmap(O) google_cpm(O)
google_charger(O) google_battery(O) exynos_reboot(O) google_bms(O)
exynos_drm(O) st54spi(O) st33spi(O) dw3000(O) mcps802154(O)
aoc_uwb_platform_drv(O) aoc_uwb_service_dev(O) aoc_usb_driver(O)
aoc_control_dev(O) aoc_char_dev(O) aoc_channel_dev(O) aoc_alsa_dev(O)
aoc_alsa_dev_util(O) aoc_core(O) trusty_virtio trusty_test trusty_log
trusty_irq memlat_devfreq exynos_acme softdog s2mpg13_spmic_thermal
gs_thermal google_bcl odpm debug_reboot smfc exynos_mfc i2c_dev
i2c_acpm i2c_exynos5 rtc_s2mpg12 keycombo s2mpg12_key tcpci_max77759
[  271.460337][ T9723]  max77759_contaminant bc_max77759
max77759_helper usb_psy usb_f_dm1 usb_f_dm usb_f_rndis xhci_exynos
spidev sg ufs_exynos_core ufs_pixel_fips140 s2mpg1x_gpio_gs201
bcm47765 sscoredump at24 zcomp_eh zcomp_cpu zram g2d samsung_iommu
samsung_iommu_group exynos_tty s2mpg13_powermeter s2mpg13_regulator
s2mpg12_powermeter s2mpg12_regulator s2mpg12_mfd s2mpg13_mfd
pmic_class exynos_seclog dbgcore_dump sysrq_hook hardlockup_debug
s3c2410_wdt eh cp_thermal_zone cpif bts cpif_page boot_device_spi
exynos_bcm_dbg_dump gsa_gsc slc_acpm slc_pmon slc_dummy acpm_mbox_test
exynos_devfreq exynos_dm slc_pt power_stats exynos_pd_dbg exynos_pd
dwc3_exynos_usb gvotable exynos_cpuhp pixel_metrics vh_i2c vh_cgroup
vh_fs vh_thermal vh_preemptirq_long vh_sched arm_memlat_mon
governor_memlat sched_tp pixel_boot_metrics exynos_adv_tracer_s2d
keydebug kernel_top exynos_coresight_etm exynos_ecc_handler
exynos_coresight exynos_debug_test pixel_debug_test ehld
hardlockup_watchdog sjtag_driver
[  271.460460][ T9723]  gsa trusty_ipc samsung_dma_heap trusty_core
samsung_secure_iova exynos_adv_tracer debug_snapshot_debug_kinfo
clk_exynos pcie_exynos_core exynos_pm acpm_flexpmu_dbg
pcie_exynos_gs20

Re: [f2fs-dev] [Bug 217266] kernel panic on f2fs filesystem when f2fs_commit_atomic_write

2023-03-30 Thread Micha L via Linux-f2fs-devel
Am Do., 30. März 2023 um 05:31 Uhr schrieb Chao Yu :
>
> On 2023/3/30 10:48, [email protected] wrote:
> > https://bugzilla.kernel.org/show_bug.cgi?id=217266
> >
> > --- Comment #2 from shilka ([email protected]) ---
> > (In reply to michalechner92 from comment #1)
> >> That looks identical to what I reported last week here:
> >>
> >> https://sourceforge.net/p/linux-f2fs/mailman/message/37794257/
>
> Could you please have a try w/ below patch?
>
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=a46bebd502fe1a3bd1d22f64cedd93e7e7702693
>

Hello Chao,

Thanks for the answer.

The crashes were reported to me after I merged the latest f2fs-stable
patches for 5.10 from
https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs-stable.git/log/?h=linux-5.10.y

Which includes the commit you just mentioned
https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs-stable.git/commit/?h=linux-5.10.y&id=720709c3276220fdb70e9385d87c6169476f04e7


My source is here:
https://github.com/freak07/Kirisakura_Pantah/commits/pantah-tq2a.230305/fs/f2fs

The crashes were reported to me after I merged the latest tag of
f2fs-stable, if you look back through my history on March 4th and
March 11th.

I reverted the latest f2fs-stable patches on march 23rd and didn´t
receive a further report of the crash so far.


> >>
> >>
> >
> > Yep, it seems very similar, but I do not encounter this problem until 
> > update to
> > kernel 6.2.8.
> > Before I saw your reply, I guess there might exist some subtle conflicts
> > between new feature in 6.2.8 kernel and the f2fs module. With your info, it 
> > is
> > highly possible that the implementation of f2fs exists some bug, and new 
> > kernel
> > feature might frequently trigger it.
> >
>
>
> ___
> Linux-f2fs-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel


___
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel


Re: [f2fs-dev] [Bug 217266] kernel panic on f2fs filesystem when f2fs_commit_atomic_write

2023-04-03 Thread Micha L via Linux-f2fs-devel
Am Mo., 3. Apr. 2023 um 16:08 Uhr schrieb :
>
> https://bugzilla.kernel.org/show_bug.cgi?id=217266
>
> Chao Yu ([email protected]) changed:
>
>What|Removed |Added
> 
>  Status|NEW |ASSIGNED
>  CC||[email protected]
>
> --- Comment #5 from Chao Yu ([email protected]) ---
> Hi shilka, Micha L,
>
> Do you have a reproducer?
>
> --
> You may reply to this email to add a comment.
>
> You are receiving this mail because:
> You are watching the assignee of the bug.
>
> ___
> Linux-f2fs-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

Hey Chao,

I don´t have a reproducer. It just seems to happen randomly while
using the device very rarely.


Am Mo., 3. Apr. 2023 um 19:00 Uhr schrieb :
>
> https://bugzilla.kernel.org/show_bug.cgi?id=217266
>
> Jaegeuk Kim ([email protected]) changed:
>
>What|Removed |Added
> 
>  CC||[email protected]
>
> --- Comment #6 from Jaegeuk Kim ([email protected]) ---
> I think this fixes the panic.
> https://lore.kernel.org/linux-f2fs-devel/[email protected]/T/#u
>
> --
> You may reply to this email to add a comment.
>
> You are receiving this mail because:
> You are watching the assignee of the bug.
>
> ___
> Linux-f2fs-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

Hey Jaegeuk,

Thanks for the link and taking a look! I will cherry-pick the fix to
my tree and ship it on the next release on XDA.

I´ll report back in case I get more logs with the panic I shared earlier.


___
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel