From: Timothy Madden <terminatorul@gmail.com>
To: nouveau@lists.freedesktop.org
Subject: [Nouveau] Error "kernel rejected pushbuf" for GA102 chipset OpenGL / Vulkan
Date: Sun, 8 Oct 2023 04:48:28 +0300 [thread overview]
Message-ID: <uft1pc$7mv$1@ciao.gmane.io> (raw)
Hello
I have an RTX 3090 Strix card with GA102 chipset, and I get the below
error whenever I try some GL / Vk benchmarks:
nouveau: kernel rejected pushbuf: No such device
nouveau: ch20: krec 0 pushes 1 bufs 7 relocs 0
nouveau: ch20: buf 00000000 00000014 00000004 00000004 00000000
0x7f5dc2f80000 0xc800000 0x80000
nouveau: ch20: buf 00000001 00000008 00000002 00000002 00000000 (nil)
0xce0000 0xd0000
nouveau: ch20: buf 00000002 0000000a 00000002 00000002 00000000 (nil)
0xdb0000 0x20000
nouveau: ch20: buf 00000003 00000009 00000002 00000002 00000002 (nil)
0xe00000 0xa600000
nouveau: ch20: buf 00000004 00000006 00000004 00000000 00000004
0x7f5dd004d000 0x14000 0x1000
nouveau: ch20: buf 00000005 0000001b 00000002 00000002 00000000 (nil)
0xce00000 0xa00000
nouveau: ch20: buf 00000006 00000019 00000004 00000000 00000004
0x7f5d8533f000 0x28400000 0x9c4000
nouveau: ch20: psh 00000000 00000523e8 000005240c
nouveau: 0x20048100
nouveau: 0x00000000
nouveau: 0x0ce00000
nouveau: 0x00000000
nouveau: 0x28400000
nouveau: 0x20018106
nouveau: 0x009c4000
nouveau: 0x200180c0
nouveau: 0x00000186
luxmark.bin: ../nouveau/pushbuf.c:730: nouveau_pushbuf_data: Assertion
`kref' failed.
I tried with a number of applications, Mesa versions and kernel
versions, and I always get this error.
The error was not present with my old card RTX 2080 Ti (TU102 chipset,
Mesa 23.0.0, Linux 6.2).
Apps that trigger the issue:
- Unigine Valley 1.0 benchmark (OpenGL)
- Unigine Heaven 4.0 (benchmark starts but errors out later)
(OpenGL)
- LuxMark v3.1 with rusticl (on NVK experimental from git)
The issue is not triggered with:
- glxgears
- vkgears, vkcube
- Unigine Tropics Demo 1.3 (OpenGL)
- Unigine Valley, Unigine Heave running under wine (DX11)
- FurMark 1.36 running under wine (OpenGL)
Tried with a number of kernels:
- 6.4.4
- 6.6.0-rc1
- 6.6.0-rc4
Tried with Mesa versions:
- 23.1.7
- 23.2.0
- 23.2.1
- 23.3.0-devel (git main)
Sample dmesg output from 6.6.0-rc1 kernel:
[ 8696.671977] nouveau 0000:02:00.0: fifo: fault 00 [VIRT_READ] at
000000004abc0000 engine 40 [gr] client 11 [GPC3/GCC] reason 02 [PTE] on
channel 23 [05fe930000 valley_x64[11487]]
[ 8696.671987] nouveau 0000:02:00.0:
fifo:c00000:0017:[valley_x64[11487]] rc scheduled
[ 8696.671990] nouveau 0000:02:00.0: fifo:c00000: rc scheduled
[ 8696.672019] nouveau 0000:02:00.0:
fifo:c00000:0017:0017:[valley_x64[11487]] errored - disabling channel
[ 8696.672023] nouveau 0000:02:00.0: Xwayland[2603]: channel 23 killed!
(https://gitlab.freedesktop.org/nouveau/mesa/uploads/8ea30740f8ce6c5e82a22c2dd6ccea7a/dmesg_output.txt)
Sample dmesg output from 6.6.0-rc4 kernel:
https://pastebin.com/RPvSaiPX
Can you help me please ? I can not run most of my benchmarks because of
this issue...
I do have a multi-GPU system with two additional GPUs (with no monitors):
- Radeon Vega 8 iGPU (AMD Ryzen 7 5700G CPU)
- Radeon RX 580 Pulse dGPU
Here is my inxi output:
timothy@localhost:~/packages/mesa> inxi -GSC -xx
System:
Host: localhost.localdomain Kernel: 6.6.0-rc4-1.g019d4ec-default
arch: x86_64 bits: 64 compiler: gcc v: 13.2.1 Desktop: KDE Plasma
v: 5.27.8
tk: Qt v: 5.15.10 wm: kwin_wayland dm: SDDM Distro: openSUSE Tumbleweed
20231006
CPU:
Info: 8-core model: AMD Ryzen 7 5700G with Radeon Graphics bits: 64
type: MT MCP arch: Zen 3 rev: 0 cache: L1: 512 KiB L2: 4 MiB L3: 16 MiB
Speed (MHz): avg: 888 high: 3011 min/max: 400/4673 cores: 1: 400 2: 3000
3: 400 4: 400 5: 400 6: 400 7: 400 8: 400 9: 400 10: 400 11: 3011
12: 400
13: 400 14: 400 15: 400 16: 3000 bogomips: 121653
Flags: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 sse4a ssse3 svm
Graphics:
Device-1: AMD Ellesmere [Radeon RX 470/480/570/570X/580/580X/590]
vendor: Sapphire driver: amdgpu v: kernel arch: GCN-4 pcie: speed:
2.5 GT/s
lanes: 8 ports: active: none empty: DP-5, DP-6, DVI-D-1, HDMI-A-3,
HDMI-A-4 bus-ID: 01:00.0 chip-ID: 1002:67df temp: 34.0 C
Device-2: NVIDIA GA102 [GeForce RTX 3090] vendor: ASUSTeK driver: nouveau
v: kernel arch: Ampere pcie: speed: 8 GT/s lanes: 8 ports:
active: DP-1,DP-4,DP-8 empty: DP-2, DP-3, DP-7, DP-9, HDMI-A-1,
HDMI-A-2
bus-ID: 02:00.0 chip-ID: 10de:2204
Device-3: AMD Cezanne [Radeon Vega Series / Radeon Mobile Series]
vendor: Gigabyte driver: amdgpu v: kernel arch: GCN-5 pcie: speed:
8 GT/s
lanes: 16 ports: active: none empty: HDMI-A-5 bus-ID: 0c:00.0
chip-ID: 1002:1638 temp: 38.0 C
Display: wayland server: X.org v: 1.21.1.8 with: Xwayland v: 23.2.1
compositor: kwin_wayland driver: X: loaded: modesetting,nouveau
unloaded: fbdev,vesa alternate: nv,nvidia dri: nouveau,radeonsi
gpu: amdgpu,nouveau,amdgpu d-rect: 4720x3360 display-ID: 0
Monitor-1: DP-1 pos: bottom-r res: 2560x1440 size: N/A
Monitor-2: DP-4 pos: primary,top-left res: 1080x1920 size: N/A
Monitor-3: DP-8 pos: top-center res: 1080x1920 size: N/A
API: OpenGL v: 4.3 Mesa 23.2.1 renderer: NV172 direct-render: Yes
timothy@localhost:~/packages/mesa>
reply other threads:[~2023-10-08 1:48 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='uft1pc$7mv$1@ciao.gmane.io' \
--to=terminatorul@gmail.com \
--cc=nouveau@lists.freedesktop.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.