On 4/4/23 14:56, Aithal, Srikanth wrote: > Hello, > > Observing below kernel crash on AMD arch, from next-20230330 onwards till recent build [next-20230404]: > > [ 68.282610] BUG: kernel NULL pointer dereference, address: 0000000000000074 > [ 68.282611] #PF: supervisor read access in kernel mode > [ 68.282613] #PF: error_code(0x0000) - not-present page > [ 68.282614] PGD 0 P4D 0 > [ 68.282615] Oops: 0000 [#1] PREEMPT SMP NOPTI > [ 68.282618] CPU: 24 PID: 905 Comm: kworker/24:1 Not tainted 6.3.0-rc5-next-20230403next-20230403 #1 > [ 68.282620] Hardware name: Dell Inc. PowerEdge R6515/07PXPY, BIOS 2.8.5 08/18/2022 > [ 68.282621] Workqueue: events fbcon_register_existing_fbs > [ 68.282626] RIP: 0010:drm_crtc_next_vblank_start+0x2c/0x80 [drm] > [ 68.282653] Code: 1e fa 0f 1f 44 00 00 8b 87 90 00 00 00 48 8b 17 55 48 8d 0c c0 48 89 e5 41 54 53 48 8d 1c 48 48 c1 e3 04 48 03 9a 40 01 00 00 <8b> 53 74 85 d2 74 3f 8b 43 78 85 c0 74 38 31 d2 49 89 f4 e8 1c ff > [ 68.282655] RSP: 0018:ffffb30e0fbf7b00 EFLAGS: 00010246 > [ 68.282656] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 > [ 68.282657] RDX: ffff896b06f26000 RSI: ffffb30e0fbf7b20 RDI: ffff896b06f26928 > [ 68.282657] RBP: ffffb30e0fbf7b10 R08: ffff898a7fcd4000 R09: ffff896b1078f580 > [ 68.282658] R10: 0000000000000100 R11: 0000000000000000 R12: 0000000000000000 > [ 68.282659] R13: ffff896b12280600 R14: 0000000000000000 R15: ffff896b2f375330 > [ 68.282659] FS: 0000000000000000(0000) GS:ffff898a00400000(0000) knlGS:0000000000000000 > [ 68.282660] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 > [ 68.282661] CR2: 0000000000000074 CR3: 00000001513b8001 CR4: 0000000000770ee0 > [ 68.282662] PKRU: 55555554 > [ 68.282663] Call Trace: > [ 68.282664] <TASK> > [ 68.282667] drm_atomic_helper_wait_for_fences+0x87/0x1f0 [drm_kms_helper] > [ 68.282679] ? drm_gem_begin_shadow_fb_access+0x29/0x40 [drm_kms_helper] > [ 68.282691] drm_atomic_helper_commit+0x9e/0x150 [drm_kms_helper] > [ 68.282701] drm_atomic_commit+0x9d/0xd0 [drm] > [ 68.282722] ? __pfx___drm_printfn_info+0x10/0x10 [drm] > [ 68.282745] drm_client_modeset_commit_atomic+0x1e2/0x220 [drm] > [ 68.282767] drm_client_modeset_commit_locked+0x5b/0x170 [drm] > [ 68.282788] drm_client_modeset_commit+0x2a/0x50 [drm] > [ 68.282808] __drm_fb_helper_restore_fbdev_mode_unlocked+0xa1/0xd0 [drm_kms_helper] > [ 68.282819] drm_fb_helper_set_par+0x3d/0x70 [drm_kms_helper] > [ 68.282829] fbcon_init+0x27d/0x500 > [ 68.282831] visual_init+0xd5/0x140 > [ 68.282834] do_bind_con_driver+0x1a0/0x3b0 > [ 68.282836] do_take_over_console+0x112/0x1b0 > [ 68.282838] do_fbcon_takeover+0x60/0xc0 > [ 68.282840] do_fb_registered+0x1bc/0x1d0 > [ 68.282842] fbcon_register_existing_fbs+0x51/0x90 > [ 68.282843] process_one_work+0x215/0x430 > [ 68.282847] worker_thread+0x38/0x3d0 > [ 68.282848] ? __pfx_worker_thread+0x10/0x10 > [ 68.282850] kthread+0xfd/0x130 > [ 68.282852] ? __pfx_kthread+0x10/0x10 > [ 68.282853] ret_from_fork+0x2c/0x50 > [ 68.282856] </TASK> > [ 68.282857] Modules linked in: stp llc nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 ip_set nf_tables nfnetlink sunrpc nls_iso8859_1 ipmi_ssif intel_rapl_msr intel_rapl_common amd64_edac kvm_amd kvm crct10dif_pclmul crc32_pclmul ghash_clmulni_intel sha512_ssse3 aesni_intel acpi_ipmi crypto_simd cryptd ipmi_si ccp rapl sp5100_tco ipmi_devintf input_leds joydev wmi_bmof pcspkr efi_pstore i2c_piix4 k10temp ipmi_msghandler acpi_power_meter evbug mac_hid sch_fq_codel xfs libcrc32c hid_generic usbkbd usbmouse usbhid hid mgag200 drm_kms_helper syscopyarea sysfillrect sysimgblt i2c_algo_bit drm_shmem_helper drm mpt3sas tg3 raid_class scsi_transport_sas wmi msr autofs4 > [ 68.282891] CR2: 0000000000000074 > [ 68.282892] ---[ end trace 0000000000000000 ]--- > [ 68.302761] pstore: backend (erst) writing error (-28) > [ 68.302763] RIP: 0010:drm_crtc_next_vblank_start+0x2c/0x80 [drm] > > > Commit d39e48ca80c0 is causing the issue. I have verified same issue exists on https://gitlab.freedesktop.org/drm/msm [tag: dma-fence-deadline]. > Should have been fixed by [1], please test. Thanks. [1]: https://lore.kernel.org/lkml/20230403160314.1210533-1-robdclark@xxxxxxxxx/ -- An old man doll... just what I always wanted! - Clara
