Re: Regression on linux-next (next-20250225)

[Bartosz Golaszewski <bartosz.golaszewski@xxxxxxxxxx>]

On Wed, 26 Feb 2025 at 21:29, Borah, Chaitanya Kumar
<chaitanya.kumar.borah@xxxxxxxxx> wrote:
>
> Hello Bartosz,
>
> Hope you are doing well. I am Chaitanya from the linux graphics team in Intel.
>
> This mail is regarding a regression we are seeing in our CI runs[1] on linux-next repository.
>
> Since the version next-20250225 [2], we are seeing the following regression
>
> `````````````````````````````````````````````````````````````````````````````````
> <4>[    8.483421] gpio gpiochip1: gpiochip_add_data_with_key: get_direction failed: -22
> <4>[    8.483427] ------------[ cut here ]------------
> <4>[    8.483428] WARNING: CPU: 20 PID: 444 at drivers/gpio/gpiolib.c:349 gpiochip_get_direction+0x63/0x90
> <4>[    8.483430] Modules linked in: intel_ish_ipc(+) e1000e(+) spi_intel(+) i2c_smbus idma64(+) mei intel_ishtp realtek(+) processor_thermal_device_pci(+) processor_thermal_device processor_thermal_wt_hint video processor_thermal_rfim int3403_thermal intel_pmc_core(+) processor_thermal_rapl intel_rapl_common intel_vpu processor_thermal_wt_req ucsi_acpi(+) processor_thermal_power_floor drm_shmem_helper pmt_telemetry processor_thermal_mbox pmt_class typec_ucsi int3400_thermal drm_kms_helper acpi_tad intel_hid int340x_thermal_zone thunderbolt(+) acpi_thermal_rel intel_vsec typec pinctrl_meteorpoint(+) sparse_keymap wmi pinctrl_meteorlake acpi_pad dm_multipath msr nvme_fabrics fuse efi_pstore nfnetlink ip_tables x_tables autofs4
> <4>[    8.483465] CPU: 20 UID: 0 PID: 444 Comm: (udev-worker) Tainted: G        W          6.14.0-rc4-next-20250225-next-20250225-g0226d0ce98a4+ #1
> <4>[    8.483467] Tainted: [W]=WARN
> <4>[    8.483467] Hardware name: Intel Corporation Arrow Lake Client Platform/MTL-S UDIMM 2DPC EVCRB, BIOS MTLSFWI1.R00.4400.D85.2410100007 10/10/2024
> <4>[    8.483468] RIP: 0010:gpiochip_get_direction+0x63/0x90
> <4>[    8.483470] Code: f8 02 5d 0f 4d c2 31 d2 31 f6 31 ff c3 cc cc cc cc 48 8b 47 08 be ff ff ff ff 48 8d b8 c0 06 00 00 e8 31 2d 95 00 85 c0 75 b9 <0f> 0b 48 8b 43 38 48 85 c0 75 b7 0f 0b b8 a1 ff ff ff 5b 41 5c 5d
> <4>[    8.483472] RSP: 0018:ffffc9000206f590 EFLAGS: 00010246
> <4>[    8.483473] RAX: 0000000000000000 RBX: ffff8881087964d0 RCX: 0000000000000000
> <4>[    8.483474] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
> <4>[    8.483476] RBP: ffffc9000206f5a0 R08: 0000000000000000 R09: 0000000000000000
> <4>[    8.483477] R10: 0000000000000000 R11: 0000000000000000 R12: 00000000000001a4
> <4>[    8.483478] R13: ffffffff87f39780 R14: ffff8881087964d0 R15: 00000000000001a4
> <4>[    8.483479] FS:  00007f4df7e6f8c0(0000) GS:ffff8888dbf09000(0000) knlGS:0000000000000000
> <4>[    8.483480] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> <4>[    8.483481] CR2: 00007f4df8003966 CR3: 0000000118522005 CR4: 0000000000f70ef0
> <4>[    8.483482] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> <4>[    8.483483] DR3: 0000000000000000 DR6: 00000000ffff07f0 DR7: 0000000000000400
> <4>[    8.483484] PKRU: 55555554
> <4>[    8.483485] Call Trace:
> <4>[    8.483486]  <TASK>
> <4>[    8.483487]  ? show_regs+0x6c/0x80
> <4>[    8.483490]  ? __warn+0x93/0x1c0
> <4>[    8.483492]  ? gpiochip_get_direction+0x63/0x90
> `````````````````````````````````````````````````````````````````````````````````
> Detailed log can be found in [3].
>
> After bisecting the tree, the following patch [4] seems to be the first "bad"
> commit
>
> `````````````````````````````````````````````````````````````````````````````````````````````````````````
> commit e623c4303ed112a1fc20aec8427ba8407e2842e6
> Author: Bartosz Golaszewski mailto:bartosz.golaszewski@xxxxxxxxxx
> Date:   Mon Feb 10 11:52:02 2025 +0100
>
>     gpiolib: sanitize the return value of gpio_chip::get_direction()
> `````````````````````````````````````````````````````````````````````````````````````````````````````````
>
> We also verified that if we revert the patch the issue is not seen.
>
> Could you please check why the patch causes this regression and provide a fix if necessary?
>

Hi!

This is fixed in my tree and should be in next tomorrow, sorry for the trouble.

Bartosz