Regression on linux-next (next-20240829)

Fri Aug 30 08:51:58 UTC 2024

Hello Thomas,

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-20240829 [2], we are seeing the following regression

`````````````````````````````````````````````````````````````````````````````````
<4> [28.903830] ------------[ cut here ]------------
<4> [28.903835] WARNING: CPU: 10 PID: 855 at drivers/video/fbdev/core/fbmem.c:467 unregister_framebuffer+0x1f/0xd0
<4> [28.903840] Modules linked in: i915(+) prime_numbers i2c_algo_bit ttm drm_display_helper drm_buddy cdc_ncm cdc_ether usbnet btusb btrtl btintel btbcm bluetooth x86_pkg_temp_thermal mei_hdcp coretemp mei_pxp wmi_bmof kvm_intel r8152 mii r8169 kvm video crct10dif_pclmul i2c_i801 crc32_pclmul ghash_clmulni_intel realtek i2c_mux mei_me i2c_smbus mei intel_lpss_pci wmi
<4> [28.903870] CPU: 10 UID: 0 PID: 855 Comm: i915_module_loa Not tainted 6.11.0-rc5-next-20240829-next-20240829-gb18bbfc14a38+ #1
<4> [28.903872] Hardware name: ASUS System Product Name/PRIME Z790-P WIFI, BIOS 0812 02/24/2023
<4> [28.903873] RIP: 0010:unregister_framebuffer+0x1f/0xd0
<4> [28.903875] Code: 90 90 90 90 90 90 90 90 90 90 90 f3 0f 1e fa 55 31 f6 48 89 fd 48 c7 c7 60 2d 7c 82 e8 da aa 5d 00 48 63 45 04 83 f8 1f 76 69 <0f> 0b 48 8b bd b0 02 00 00 48 85 ff 74 09 f6 85 cc 02 00 00 01 75
<4> [28.903876] RSP: 0018:ffffc9000250f840 EFLAGS: 00010286
<4> [28.903878] RAX: 0000000000000000 RBX: ffff8881066c4740 RCX: 0000000000000000
<4> [28.903880] RDX: 0000000000000001 RSI: ffffffff82440d5a RDI: 0000000000000000
<4> [28.903881] RBP: ffff8881066a8800 R08: 0000000000000000 R09: 0000000000000000
<4> [28.903881] R10: ffffc9000250f840 R11: ffff888114940040 R12: ffffc9000250f878
<4> [28.903883] R13: ffff8881066a6810 R14: ffff8881066a6c20 R15: ffff8881066a6be0
<4> [28.903884] FS:  00007f8a936adc40(0000) GS:ffff88885f280000(0000) knlGS:0000000000000000
<4> [28.903885] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
<4> [28.903886] CR2: 00007f8a95eb3f70 CR3: 0000000117808000 CR4: 0000000000f50ef0
<4> [28.903887] PKRU: 55555554
<4> [28.903888] Call Trace:
<4> [28.903889]  <TASK>
<4> [28.903890]  ? __warn+0x91/0x1a0
<4> [28.903894]  ? unregister_framebuffer+0x1f/0xd0
<4> [28.903896]  ? report_bug+0x1f8/0x200
<4> [28.903899]  ? handle_bug+0x53/0x90
<4> [28.903902]  ? exc_invalid_op+0x18/0x70
<4> [28.903903]  ? asm_exc_invalid_op+0x1a/0x20
<4> [28.903907]  ? unregister_framebuffer+0x1f/0xd0
<4> [28.903909]  release_nodes+0x2e/0xf0
<4> [28.903912]  devres_release_all+0x8a/0xc0
<4> [28.903915]  device_unbind_cleanup+0x9/0x70
<4> [28.903917]  device_release_driver_internal+0xce/0x140
<4> [28.903920]  bus_remove_device+0xcf/0x140
<4> [28.903922]  device_del+0x179/0x3e0
<4> [28.903925]  ? pci_conf1_read+0xa3/0x100
<4> [28.903928]  platform_device_del.part.0+0xe/0x70
<4> [28.903930]  platform_device_unregister+0x1b/0x30
<4> [28.903932]  sysfb_disable+0x4c/0x90
<4> [28.903934]  aperture_remove_conflicting_pci_devices+0x25/0x80
<4> [28.903937]  i915_driver_probe+0x6f3/0xdc0 [i915]
<4> [28.904017]  ? lockdep_hardirqs_on+0x7b/0x100
<4> [28.904019]  ? _raw_spin_unlock_irqrestore+0x41/0x70
<4> [28.904022]  ? device_id_in_list+0xaf/0x120 [i915]
<4> [28.904098]  i915_pci_probe+0xd5/0x200 [i915]
<4> [28.904171]  pci_device_probe+0x95/0x120
<4> [28.904174]  really_probe+0xd9/0x370
<4> [28.904177]  ? __pfx___driver_attach+0x10/0x10
<4> [28.904179]  __driver_probe_device+0x73/0x150
<4> [28.904181]  driver_probe_device+0x19/0xa0
<4> [28.904184]  __driver_attach+0xb6/0x180
<4> [28.904186]  ? __pfx___driver_attach+0x10/0x10
<4> [28.904188]  bus_for_each_dev+0x77/0xd0
<4> [28.904191]  bus_add_driver+0x110/0x240
<4> [28.904193]  driver_register+0x5b/0x110
<4> [28.904195]  i915_init+0x23/0xc0 [i915]
<4> [28.904268]  ? __pfx_i915_init+0x10/0x10 [i915]
<4> [28.904338]  do_one_initcall+0x5c/0x2b0
<4> [28.904341]  ? __kmalloc_cache_noprof+0x21e/0x270
<4> [28.904344]  ? do_init_module+0x1e/0x210
<4> [28.904348]  do_init_module+0x5f/0x210
<4> [28.904350]  load_module+0x1d4e/0x1fd0
<4> [28.904356]  ? init_module_from_file+0x86/0xd0
<4> [28.904358]  init_module_from_file+0x86/0xd0
<4> [28.904362]  idempotent_init_module+0x181/0x280
<4> [28.904366]  __x64_sys_finit_module+0x56/0xb0
<4> [28.904369]  do_syscall_64+0x66/0x140
<4> [28.904371]  entry_SYSCALL_64_after_hwframe+0x76/0x7e 
`````````````````````````````````````````````````````````````````````````````````
Details log can be found in [3].

After bisecting the tree, the following patch [4] seems to be the first "bad"
commit

`````````````````````````````````````````````````````````````````````````````````````````````````````````
commit c2fe0480cd77526411707da46d72a44d9671c9ad
Author: Thomas Weißschuh mailto:linux at weissschuh.net
Date:   Tue Aug 27 17:25:15 2024 +0200

    fbdev/efifb: Use devm_register_framebuffer()

    This simplifies the error handling.
    Also the drvdata slot is now unused and can be used for other usecases.

    Signed-off-by: Thomas Weißschuh mailto:linux at weissschuh.net
    Signed-off-by: Helge Deller mailto:deller at gmx.de
`````````````````````````````````````````````````````````````````````````````````````````````````````````

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?

Thank you.

Regards

Chaitanya

[1] https://intel-gfx-ci.01.org/tree/linux-next/combined-alt.html?
[2] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?h=next-20240829	
[3] https://gfx-ci.igk.intel.com/tree/linux-next/next-20240829/bat-rpls-4/igt@i915_module_load@load.html
[4] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?h=next-20240829&id=c2fe0480cd77526411707da46d72a44d9671c9ad 