i915drmkms heartbeat failure, reset failure and hard lockup

Paul Ripke <[email protected]>
Newsgroups gmane.os.netbsd.devel.kernel
Message-ID <[email protected]>
I've had a few of these of late on amd64 netbsd-10 from ~2024-10-22.
It seems to be triggered by high system load, likely memory pressure
and high paging (pkgsrc builds, I blame rust, mostly).

I get a i915drmkms heartbeat failure, followed by weird screen artifacts
(horizontal tearing and redraws taking a second or two). Some time later
(minutes, hours, days even), I get a hard lock (no response to console,
ping, etc, watchdogs on swwdog0 & tco0 also didn't fire).

Anyone seen anything similar? Any hints? I'm planning on upgrading this
old machine soon, but figured I can't be the only one running into this.

fwiw, i915 is specifically:
i915drmkms0 at pci0 dev 2 function 0: Intel Sandy Bridge (desktop) GI1 Integrated Graphics Device (rev. 0x09)

X reports:
[    87.659] (II) intel(0): Using Kernel Mode Setting driver: i915, version 1.6.0 20200114
[    87.661] (WW) Falling back to old probe method for modesetting
[    87.661] (WW) Falling back to old probe method for wsfb
[    87.661] (WW) VGA arbiter: cannot open kernel arbiter, no multi-card support
[    87.662] (--) intel(0): Integrated Graphics Chipset: Intel(R) HD Graphics 2000
[    87.662] (--) intel(0): CPU: x86-64, sse2, sse3, ssse3, sse4.1, sse4.2, avx; using a maximum of 4 threads

Heartbeat error:
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973] bcs0 heartbeat {prio:-2147483645} not ticking
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]        Awake? 4
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]        Barriers?: no
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]        Latency: 95us
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]        Heartbeat: 3000 ms ago
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]        Reset count: 0 (global 0)
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]        Requests:
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                active  4:2d9f7b*  @ 5571ms: X[1511]
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                ring->start:  0x7fff3000
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                ring->head:   0x00002dd8
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                ring->tail:   0x000034c8
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                ring->emit:   0x000034c8
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                ring->space:  0x00002858
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                ring->hwsp:   0x7fff7100
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973] [head 2de8, postfix 2e60, tail 2e70, batch 0x00000000_07bc6000]:
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973] warning: /home/netbsd/netbsd-10/src/sys/external/bsd/drm2/dist/drm/i915/gt/intel_engine_cs.c:1234: WARN_ON_ONCE(hex_dump_to_buffer(buf + pos, len - pos, rowsize, sizeof(u32), line, sizeof(line), 0) >= sizeof(line))
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973] [0000] 01402413 04020000 00000000 00000000 01402013 04020000 00000000 00000000
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973] 01000011 20220200 ffffffff 01000011 28220200 0000df7f 0
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973] [0020] 01000011 20220200 ffffffff 01000011 28220200 0000df7f 01004012 28220200
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973] 00f0ff7f 01000011 c0200200 00020002 01402013 04020000 0
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973] bcs0 heartbeat {prio:-2147483645} not ticking[ 120338.1753973]         IPEHR: 0x16060720
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973] b      E  4:2d9f7b*  @ 5571ms: X[1511]
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                E  4:2d9f7c  @ 5570ms: X[1511]
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                E  4:2d9f7d  @ 5570ms: X[1511]
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                E  4:2d9f7e  @ 5570ms: X[1511]
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                E  4:2d9f7f  @ 5569ms: X[1511]
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                E  4:2d9f80  @ 5569ms: X[1511]
Nov 14 09:18:29 slave /netbsd: [ 120338.1753973]                E  4:2d9f81  @ 5569ms: X[1511]
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973]                ...skipping 5 executing requests...
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973]                E  4:2d9f87  @ 3001ms: [i915]
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] HWSP:
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] HWSP:[ 120338.1763973] 00000000 00000000 00000000 00000000 00000000 00000000 0
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] [0020] 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] 00000000 00000000 00000000 00000000 00000000 00000000 0
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] *
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] [0100] 7a9f2d00 00000000 00000000 00000000 00000000 00000000 00000000 00000000
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] 00000000 00000000 00000000 00000000 00000000 00000000 0
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] [0120] 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] 00000000 00000000 00000000 00000000 00000000 00000000 0
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] *
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] Idle? no
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] Signals:
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973]        [4:2d9f86] @ 5568ms
Nov 14 09:18:29 slave /netbsd: [ 120338.1763973] i915drmkms0: notice: Resetting chip for stopped heartbeat on bcs0
Nov 14 09:18:29 slave /netbsd: [ 120338.2423960] i915drmkms0: autoconfiguration error: error: Failed to reset chip

-- 
Paul Ripke
"Great minds discuss ideas, average minds discuss events, small minds
 discuss people."
-- Disputed: Often attributed to Eleanor Roosevelt. 1948.
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.