Kenel Ooops with: "drm/amdgpu: poll ras_controller_irq and err_event_athub_irq status" [bisected].

[Przemek Socha <soprwa-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>]

[-- Attachment #1.1: Type: text/plain, Size: 12821 bytes --]

Hello everyone,

after today sync with amd-staging-drm-next repo my machine was hit by Ooops 
bug.
Maybe my google-foo is weak, but I could not find any fix on patchwork for this 
that will/was implemented or planned.

Machine is a Lenovo netbook with a6-6310 APU, R4 (CIK).

I have done bisection and here are the results:


1.  dmesg output from pstore after kernel panic:

<6>[   13.133880] [drm] amdgpu kernel modesetting enabled.
<6>[   13.133923] amdgpu 0000:00:01.0: remove_conflicting_pci_framebuffers: bar 
0: 0xe0000000 -> 0xefffffff
<6>[   13.133927] amdgpu 0000:00:01.0: remove_conflicting_pci_framebuffers: bar 
2: 0xf0000000 -> 0xf07fffff
<6>[   13.133930] amdgpu 0000:00:01.0: remove_conflicting_pci_framebuffers: bar 
5: 0xf0c00000 -> 0xf0c3ffff
<7>[   13.133933] checking generic (e0000000 420000) vs hw (e0000000 10000000)
<6>[   13.133935] fb0: switching to amdgpudrmfb from EFI VGA
<6>[   13.133999] Console: switching to colour dummy device 80x25
<6>[   13.136463] [drm] initializing kernel modesetting (MULLINS 0x1002:0x9851 
0x17AA:0x3801 0x00).
<6>[   13.136826] [drm] register mmio base: 0xF0C00000
<6>[   13.136827] [drm] register mmio size: 262144
<6>[   13.136837] [drm] add ip block number 0 <cik_common>
<6>[   13.136839] [drm] add ip block number 1 <gmc_v7_0>
<6>[   13.136840] [drm] add ip block number 2 <cik_ih>
<6>[   13.136842] [drm] add ip block number 3 <gfx_v7_0>
<6>[   13.136844] [drm] add ip block number 4 <cik_sdma>
<6>[   13.136845] [drm] add ip block number 5 <kv_dpm>
<6>[   13.136847] [drm] add ip block number 6 <dm>
<6>[   13.136849] [drm] add ip block number 7 <uvd_v4_2>
<6>[   13.136850] [drm] add ip block number 8 <vce_v2_0>
<6>[   13.136857] amdgpu 0000:00:01.0: kfd not supported on this ASIC
<6>[   13.136916] ATOM BIOS: BR45787.ts5
<6>[   13.137031] [drm] vm size is 64 GB, 2 levels, block size is 10-bit, 
fragment size is 9-bit
<6>[   13.137042] amdgpu 0000:00:01.0: VRAM: 1024M 0x000000F400000000 - 
0x000000F43FFFFFFF (1024M used)
<6>[   13.137046] amdgpu 0000:00:01.0: GART: 1024M 0x000000FF00000000 - 
0x000000FF3FFFFFFF
<6>[   13.137056] [drm] Detected VRAM RAM=1024M, BAR=1024M
<6>[   13.137057] [drm] RAM width 64bits UNKNOWN
<6>[   13.138102] sdhci: Secure Digital Host Controller Interface driver
<6>[   13.138105] sdhci: Copyright(c) Pierre Ossman
<6>[   13.138741] [TTM] Zone  kernel: Available graphics memory: 3541568 KiB
<6>[   13.138744] [TTM] Zone   dma32: Available graphics memory: 2097152 KiB
<6>[   13.138745] [TTM] Initializing pool allocator
<6>[   13.138754] [TTM] Initializing DMA pool allocator
<6>[   13.138882] [drm] amdgpu: 1024M of VRAM memory ready
<6>[   13.138891] [drm] amdgpu: 3072M of GTT memory ready.
<6>[   13.138932] [drm] GART: num cpu pages 262144, num gpu pages 262144
<6>[   13.138970] [drm] PCIE GART of 1024M enabled (table at 
0x000000F400401000).
<6>[   13.176861] [drm] Internal thermal controller without fan control
<6>[   13.176865] [drm] amdgpu: dpm initialized
<6>[   13.176872] [drm] Found UVD firmware Version: 1.64 Family ID: 9
<6>[   13.178133] sdhci-pci 0000:00:14.7: SDHCI controller found [1022:7813] 
(rev 1)
<6>[   13.180552] [drm] Found VCE firmware Version: 50.10 Binary ID: 2
<6>[   13.186202] kvm: Nested Virtualization enabled
<6>[   13.186205] kvm: Nested Paging enabled
<6>[   13.191378] mmc0: SDHCI controller on PCI [0000:00:14.7] using ADMA
<3>[   13.196258] [drm:dm_pp_get_static_clocks [amdgpu]] *ERROR* DM_PPLIB: 
invalid powerlevel state: 0!
<4>[   13.196308] [drm] Unsupported Connector type:5!
<6>[   13.213496] [drm] Display Core initialized with v3.2.48!
<6>[   13.221850] [drm] SADs count is: -2, don't need to read it
<6>[   13.230392] ath: phy0: WB335 2-ANT card detected
<6>[   13.230395] ath: phy0: Set BT/WLAN RX diversity capability
<6>[   13.247472] ath: phy0: Enable LNA combining
<6>[   13.248570] ath: phy0: ASPM enabled: 0x43
<7>[   13.248574] ath: EEPROM regdomain: 0x6a
<7>[   13.248575] ath: EEPROM indicates we should expect a direct regpair map
<7>[   13.248579] ath: Country alpha2 being used: 00
<7>[   13.248580] ath: Regpair used: 0x6a
<7>[   13.261552] ieee80211 phy0: Selected rate control algorithm 
'minstrel_ht'
<6>[   13.261857] ieee80211 phy0: Atheros AR9565 Rev:1 mem=0xffffa9f1c0400000, 
irq=43
<6>[   13.296215] ath9k 0000:01:00.0 wlp1s0: renamed from wlan0
<6>[   13.304323] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
<6>[   13.304325] [drm] Driver supports precise vblank timestamp query.
<6>[   13.321092] [drm] UVD initialized successfully.
<6>[   13.373473] usb 1-1: new high-speed USB device number 2 using ehci-pci
<6>[   13.386794] usb 4-1: new high-speed USB device number 2 using ehci-pci
<6>[   13.442287] [drm] VCE initialized successfully.
<1>[   13.444174] BUG: kernel NULL pointer dereference, address: 
00000000000000a8
<1>[   13.444191] #PF: supervisor read access in kernel mode
<1>[   13.444197] #PF: error_code(0x0000) - not-present page
<6>[   13.444202] PGD 0 P4D 0 
<4>[   13.444210] Oops: 0000 [#1] PREEMPT SMP
<4>[   13.444218] CPU: 1 PID: 3311 Comm: laptop_mode Not tainted 5.2.0-rc1+ 
#94
<4>[   13.444224] Hardware name: LENOVO 80E3/Lancer 5B2, BIOS A2CN45WW(V2.13) 
08/04/2016
<4>[   13.444392] RIP: 0010:amdgpu_irq_handler+0x28/0x78 [amdgpu]
<4>[   13.444401] Code: 00 00 41 54 55 53 48 8b 6e 28 48 89 f3 48 89 ef 48 8d 
b5 88 5f 00 00 e8 0e 0a 00 00 41 89 c4 ff c8 74 3e 48 8b 85 d0 70 00 00 <48> 8b 
90 a8 00 00 00 48 85 d2 74 0f 48 89 ef e8 1c 75 f7 cb 48 8b
<4>[   13.444414] RSP: 0000:ffffa9f1c00ecf00 EFLAGS: 00010012
<4>[   13.444420] RAX: 0000000000000000 RBX: ffff947b96a700b0 RCX: 
0000000000000018
<4>[   13.444427] RDX: 00000000008e7d30 RSI: 001a351391f4b553 RDI: 
ffffffff8ca17720
<4>[   13.444433] RBP: ffff947b8fb80000 R08: ffffffff8c6077e0 R09: ffff947b97ba4af8
<4>[   13.444440] R10: ffff947b969cd2b8 R11: ffff947b969cd2a8 R12: 
0000000000000001
<4>[   13.444446] R13: 0000000000000000 R14: ffffa9f1c00ecf64 R15: 
0000000000000000
<4>[   13.444453] FS:  00007f40b84aa740(0000) GS:ffff947b97a80000(0000) knlGS:
0000000000000000
<4>[   13.444461] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
<4>[   13.444466] CR2: 00000000000000a8 CR3: 000000020e627000 CR4: 
00000000000406e0
<4>[   13.444472] Call Trace:
<4>[   13.444481]  <IRQ>
<4>[   13.444492]  __handle_irq_event_percpu+0x3d/0x1a0
<4>[   13.444501]  handle_irq_event_percpu+0x2c/0x78
<4>[   13.444508]  handle_irq_event+0x2f/0x4c
<4>[   13.444515]  handle_edge_irq+0x95/0x1c0
<4>[   13.444523]  handle_irq+0x17/0x20
<4>[   13.444531]  do_IRQ+0x4a/0xe0
<4>[   13.444539]  common_interrupt+0xf/0xf
<4>[   13.444545]  </IRQ>
<4>[   13.444550] RIP: 0033:0x56277e2c6830
<4>[   13.444556] Code: 68 7d 00 00 00 e9 10 f8 ff ff ff 25 22 52 0d 00 68 7e 00 
00 00 e9 00 f8 ff ff ff 25 1a 52 0d 00 68 7f 00 00 00 e9 f0 f7 ff ff <ff> 25 12 52 
0d 00 68 80 00 00 00 e9 e0 f7 ff ff ff 25 0a 52 0d 00
<4>[   13.444568] RSP: 002b:00007ffddc457328 EFLAGS: 00000246 ORIG_RAX: 
ffffffffffffffda
<4>[   13.444576] RAX: 0000562780370044 RBX: 0000000000000056 RCX: 
0000000000000045
<4>[   13.444582] RDX: 0000000000000001 RSI: 000056278037f670 RDI: 
00005627804390c0
<4>[   13.444588] RBP: 0000562780443150 R08: 0000000000000000 R09: 
0000000000003cff
<4>[   13.444595] R10: 0000000000100000 R11: 0000000000000098 R12: 
00005627804390c0
<4>[   13.444601] R13: 0000562780440044 R14: 0000562780457a00 R15: 
00000000000000bc
<4>[   13.444609] Modules linked in: ath9k ath9k_common ath9k_hw kvm_amd 
sdhci_pci iosf_mbi mac80211 cqhci kvm sdhci irqbypass crc32_pclmul 
ghash_clmulni_intel serio_raw mmc_core ath amdgpu(+) cfg80211 gpu_sched 
mfd_core ttm xhci_pci ehci_pci xhci_hcd sp5100_tco ehci_hcd
<4>[   13.444645] CR2: 00000000000000a8
<4>[   13.444654] ---[ end trace cd97c823583992aa ]---
<6>[   13.446294] [drm] fb mappable at 0xA07ED000
<6>[   13.446305] [drm] vram apper at 0xA0000000
<6>[   13.446310] [drm] size 5767168
<6>[   13.446315] [drm] fb depth is 24
<6>[   13.446319] [drm]    pitch is 5632
<6>[   13.446480] fbcon: amdgpudrmfb (fb0) is primary device
<4>[   13.486378] hpet1: lost 1 rtc interrupts
<4>[   13.531123] hpet1: lost 1 rtc interrupts
<4>[   13.572920] hpet1: lost 1 rtc interrupts
<6>[   13.573579] usb 1-1: New USB device found, idVendor=0438, 
idProduct=7900, bcdDevice= 0.18
<6>[   13.573583] usb 1-1: New USB device strings: Mfr=0, Product=0, 
SerialNumber=0
<6>[   13.573689] usb 4-1: New USB device found, idVendor=0438, 
idProduct=7900, bcdDevice= 0.18
<6>[   13.573692] usb 4-1: New USB device strings: Mfr=0, Product=0, 
SerialNumber=0
<6>[   13.573994] hub 4-1:1.0: USB hub found
<6>[   13.574107] hub 1-1:1.0: USB hub found
<6>[   13.574120] hub 4-1:1.0: 4 ports detected
<6>[   13.574182] hub 1-1:1.0: 4 ports detected
<4>[   13.612313] hpet1: lost 1 rtc interrupts
<4>[   13.651976] hpet1: lost 1 rtc interrupts
<4>[   13.690645] hpet1: lost 1 rtc interrupts
<4>[   13.732985] hpet1: lost 1 rtc interrupts
<4>[   13.773804] hpet1: lost 1 rtc interrupts
<4>[   13.815399] hpet1: lost 1 rtc interrupts
<4>[   13.857053] hpet1: lost 1 rtc interrupts
<6>[   13.943198] usb 4-1.2: new high-speed USB device number 3 using ehci-pci
<6>[   13.943227] usb 1-1.3: new high-speed USB device number 3 using ehci-pci
<4>[   14.017533] RIP: 0010:amdgpu_irq_handler+0x28/0x78 [amdgpu]
<4>[   14.017538] Code: 00 00 41 54 55 53 48 8b 6e 28 48 89 f3 48 89 ef 48 8d 
b5 88 5f 00 00 e8 0e 0a 00 00 41 89 c4 ff c8 74 3e 48 8b 85 d0 70 00 00 <48> 8b 
90 a8 00 00 00 48 85 d2 74 0f 48 89 ef e8 1c 75 f7 cb 48 8b
<4>[   14.017540] RSP: 0000:ffffa9f1c00ecf00 EFLAGS: 00010012
<4>[   14.017544] RAX: 0000000000000000 RBX: ffff947b96a700b0 RCX: 
0000000000000018
<4>[   14.017546] RDX: 00000000008e7d30 RSI: 001a351391f4b553 RDI: 
ffffffff8ca17720
<4>[   14.017547] RBP: ffff947b8fb80000 R08: ffffffff8c6077e0 R09: ffff947b97ba4af8
<4>[   14.017549] R10: ffff947b969cd2b8 R11: ffff947b969cd2a8 R12: 
0000000000000001
<4>[   14.017551] R13: 0000000000000000 R14: ffffa9f1c00ecf64 R15: 
0000000000000000
<4>[   14.017553] FS:  00007f40b84aa740(0000) GS:ffff947b97a80000(0000) knlGS:
0000000000000000
<4>[   14.017555] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
<4>[   14.017557] CR2: 00000000000000a8 CR3: 000000020e627000 CR4: 
00000000000406e0
<0>[   14.017559] Kernel panic - not syncing: Fatal exception in interrupt
<0>[   14.017575] Kernel Offset: 0xa800000 from 0xffffffff81000000 (relocation 
range: 0xffffffff80000000-0xffffffffbfffffff)


2. full git bisect log:


git bisect start
# good: [f1f7ad1b3b98a22229e71d51a1b983049e8bae6b] drm/amd/display: fix 
calc_pll_max_vco_construct
git bisect good f1f7ad1b3b98a22229e71d51a1b983049e8bae6b
# bad: [3913cc8cdcf3e27d5ffd31b70779f189e61e6c71] drm/amdgpu: Move null pointer 
dereference check
git bisect bad 3913cc8cdcf3e27d5ffd31b70779f189e61e6c71
# good: [f7ffd234bc4acc41612fd6aac83408a1aceffceb] drm/amd/display: Add hubp 
block for Renoir (v2)
git bisect good f7ffd234bc4acc41612fd6aac83408a1aceffceb
# good: [0460fba0adac1c0e6211ec5308cfb58941cf26b8] drm/amdgpu: Handle job is 
NULL use case in amdgpu_device_gpu_recover
git bisect good 0460fba0adac1c0e6211ec5308cfb58941cf26b8
# bad: [62c64055ab6d618b1afb28dd4b119cfc1e5d59cb] drm/amdgpu: switch to 
amdgpu_ras_late_init for gfx v9 block (v2)
git bisect bad 62c64055ab6d618b1afb28dd4b119cfc1e5d59cb
# good: [1b64dd1871d952c3f999aac8176ba2afbd5ff661] drm/amdgpu: add nbif v7_4 
irq source header for vega20
git bisect good 1b64dd1871d952c3f999aac8176ba2afbd5ff661
# good: [82e6cc2843fc844e5164c0618e6ec133f405a25f] drm/amdgpu: add 
ras_controller and err_event_athub interrupt support
git bisect good 82e6cc2843fc844e5164c0618e6ec133f405a25f
# bad: [598de6e65a1c1cbd36decb09d190071c99f100f8] drm/amdgpu: add helper 
function to do common ras_late_init/fini (v3)
git bisect bad 598de6e65a1c1cbd36decb09d190071c99f100f8
# bad: [ab2d6f7463d1f6eaf0529c163754feadc353469b] drm/amdgpu: poll 
ras_controller_irq and err_event_athub_irq status
git bisect bad ab2d6f7463d1f6eaf0529c163754feadc353469b
# first bad commit: [ab2d6f7463d1f6eaf0529c163754feadc353469b] drm/amdgpu: poll 
ras_controller_irq and err_event_athub_irq status



commit ab2d6f7463d1f6eaf0529c163754feadc353469b
Author: Hawking Zhang <Hawking.Zhang-5C7GfCeVMHo@public.gmane.org>
Date:   Wed Jun 5 14:40:57 2019 +0800

    drm/amdgpu: poll ras_controller_irq and err_event_athub_irq status
    
    For the hardware that can not enable BIF ring for IH cookies for both
    ras_controller_irq and err_event_athub_irq, the driver has to poll the
    status register in irq handling and ack the hardware properly when there
    is interrupt triggered
    
    Signed-off-by: Hawking Zhang <Hawking.Zhang-5C7GfCeVMHo@public.gmane.org>
    Reviewed-by: Alex Deucher <alexander.deucher-5C7GfCeVMHo@public.gmane.org>

Any help is appreciated.

Thanks,
Przemek.

[-- Attachment #1.2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

[-- Attachment #2: Type: text/plain, Size: 153 bytes --]

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx