Skip to content
Jump to bottom

Add Linux 5.4.10 #1574

Merged
merged 3 commits into from Jan 10, 2020
Merged

Add Linux 5.4.10 #1574

merged 3 commits into from Jan 10, 2020

Commits on Jan 10, 2020

  1. linux: Add LTS version 5.4.10 

    *   https://lkml.org/lkml/2019/12/21/66
*   https://lkml.org/lkml/2019/12/31/202
*   https://lkml.org/lkml/2020/1/5/140
*   https://lkml.org/lkml/2020/1/9/920
*   https://lkml.org/lkml/2020/1/9/924
    @pmenzel
    pmenzel committed Jan 10, 2020
    Copy the full SHA
    e6865ef View commit details
    Browse the repository at this point in the history

  2. linux-5.4.10: Patch i915 to fix hangs 

    The monitor doesn’t come back up after it is turned off, and Linux shows
the messages below.

```
2020-01-06T07:12:21+01:00 amaru automount[498]: expired /project/miso
2020-01-06T07:20:15.939399+01:00 amaru kernel: [333722.724352] i915 0000:00:02.0: GPU HANG: ecode 9:1:0x00000000, hang on rcs0
2020-01-06T07:20:15.939406+01:00 amaru kernel: [333722.724352] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
2020-01-06T07:20:15.939406+01:00 amaru kernel: [333722.724353] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel
2020-01-06T07:20:15.939407+01:00 amaru kernel: [333722.724353] drm/i915 developers can then reassign to the right component if it's not a kernel issue.
2020-01-06T07:20:15.939408+01:00 amaru kernel: [333722.724353] The GPU crash dump is required to analyze GPU hangs, so please always attach it.
2020-01-06T07:20:15.939409+01:00 amaru kernel: [333722.724353] GPU crash dump saved to /sys/class/drm/card0/error
2020-01-06T07:20:15.940403+01:00 amaru kernel: [333722.725362] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:20:15.940409+01:00 amaru kernel: [333722.726079] [drm:gen8_reset_engines [i915]] *ERROR* rcs0 reset request timed out: {request: 00000001, RESET_CTL: 00000001}
2020-01-06T07:20:15.941346+01:00 amaru kernel: [333722.726313] i915 0000:00:02.0: Resetting chip for hang on rcs0
2020-01-06T07:20:15.942398+01:00 amaru kernel: [333722.728031] [drm:gen8_reset_engines [i915]] *ERROR* rcs0 reset request timed out: {request: 00000001, RESET_CTL: 00000001}
2020-01-06T07:20:15.943430+01:00 amaru kernel: [333722.728741] [drm:gen8_reset_engines [i915]] *ERROR* rcs0 reset request timed out: {request: 00000001, RESET_CTL: 00000001}
2020-01-06T07:20:21.955508+01:00 amaru kernel: [333728.740356] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:20:29.955470+01:00 amaru kernel: [333736.740352] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:20:30.873401+01:00 amaru kernel: [333737.658366] GpuWatchdog[7339]: segfault at 0 ip 0000562464c95bdd sp 00007fc578ea07c0 error 6 in vivaldi-bin[562460c9c000+72c3000]
2020-01-06T07:20:30.873407+01:00 amaru kernel: [333737.658369] Code: 48 c1 c9 03 48 81 f9 af 00 00 00 0f 87 c9 00 00 00 48 8d 15 18 3a 91 fb f6 04 11 80 0f 84 b8 00 00 00 be 01 00 00 00 ff 50 30 <c7> 04 25 00 00 00 00 37 13 00 00 c6 05 41 d7 a3 03 01 80 7d 8f 00
2020-01-06T07:21:30+01:00 amaru automount[498]: attempting to mount entry /project/miso
2020-01-06T07:21:30+01:00 amaru automount[498]: mounted /project/miso
2020-01-06T07:21:37.923474+01:00 amaru kernel: [333804.708221] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:21:39.971410+01:00 amaru kernel: [333806.756227] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:22:07.939475+01:00 amaru kernel: [333834.724208] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:22:09.923481+01:00 amaru kernel: [333836.708210] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:22:43.971484+01:00 amaru kernel: [333870.756149] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:22:45.955484+01:00 amaru kernel: [333872.740135] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:22:47.939483+01:00 amaru kernel: [333874.724136] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:22:57.923477+01:00 amaru kernel: [333884.708106] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:22:59.971411+01:00 amaru kernel: [333886.756071] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:23:17.955476+01:00 amaru kernel: [333904.740070] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:23:19.939470+01:00 amaru kernel: [333906.724077] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:24:59.971485+01:00 amaru kernel: [334006.755948] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:25:01.955471+01:00 amaru kernel: [334008.739944] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:25:03.939495+01:00 amaru kernel: [334010.723944] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:25:25.955505+01:00 amaru kernel: [334032.739910] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:26:25.923488+01:00 amaru kernel: [334092.707821] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:26:27.971503+01:00 amaru kernel: [334094.755805] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:27:05.923475+01:00 amaru kernel: [334132.707753] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0
2020-01-06T07:27:21+01:00 amaru automount[498]: expiring path /project/miso
[…]
```

The [patch][1] below might fix it.

> Subject: [PATCH] drm/i915/gt: Detect if we miss WaIdleLiteRestore
> Date: Mon, 30 Dec 2019 11:15:30 +0000
> Message-ID: <20191230111530.3750048-1-chris@chris-wilson.co.uk> (raw)
>
> In order to avoid confusing the HW, we must never submit an empty ring
> during lite-restore, that is we should always advance the RING_TAIL
> before submitting to stay ahead of the RING_HEAD.
>
> Normally this is prevented by keeping a couple of spare NOPs in the
> request->wa_tail so that on resubmission we can advance the tail. This
> relies on the request only being resubmitted once, which is the normal
> condition as it is seen once for ELSP[1] and then later in ELSP[0]. On
> preemption, the requests are unwound and the tail reset back to the
> normal end point (as we know the request is incomplete and therefore its
> RING_HEAD is even earlier).
>
> However, if this w/a should fail we would try and resubmit the request
> with the RING_TAIL already set to the location of this request's wa_tail
> potentially causing a GPU hang. We can spot when we do try and
> incorrectly resubmit without advancing the RING_TAIL and spare any
> embarrassment by forcing the context restore.
>
> In the case of preempt-to-busy, we leave the requests running on the HW
> while we unwind. As the ring is still live, we cannot rewind our
> rq->tail without forcing a reload so leave it set to rq->wa_tail and
> only force a reload if we resubmit after a lite-restore. (Normally, the
> forced reload will be a part of the preemption event.)
>
> Fixes: 22b7a426bbe1 ("drm/i915/execlists: Preempt-to-busy")
> Closes: https://gitlab.freedesktop.org/drm/intel/issues/673
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Cc: stable@vger.kernel.org
> Link: https://patchwork.freedesktop.org/patch/msgid/20191209023215.3519970-1-chris@chris-wilson.co.uk
> (cherry picked from commit 82c69bf58650e644c61aa2bf5100b63a1070fd2f)

[1]: https://lore.kernel.org/stable/20191230111530.3750048-1-chris@chris-wilson.co.uk/
    @pmenzel
    pmenzel committed Jan 10, 2020
    Copy the full SHA
    f928617 View commit details
    Browse the repository at this point in the history

  3. nvidia_linux: Build 440.44 for Linux 5.4.10

    @pmenzel
    pmenzel committed Jan 10, 2020
    Copy the full SHA
    c138f11 View commit details
    Browse the repository at this point in the history