Description:

Since upgrading from 5.4 lts to 5.10 lts, I'm experiencing frequent serious stability problems.
After the screen blocks up I have no recourse but to reboot.

Finally I booted up an old laptop to ssh into the machine and
I notice the following kernel log:
févr. 22 18:45:33 sarchx64 kernel: general protection fault, probably for non-canonical address 0xeb6f95a0d500b36f: 0000 [#1] SMP NOPTI
févr. 22 18:45:33 sarchx64 kernel: CPU: 9 PID: 6240 Comm: Xorg Tainted: P OE 5.10.17-1-lts #1
févr. 22 18:45:33 sarchx64 kernel: Hardware name: Supermicro H8SGL/H8SGL, BIOS 3.5b 03/18/2016
févr. 22 18:45:33 sarchx64 kernel: RIP: 0010:kmem_cache_alloc_trace+0xdb/0x270
févr. 22 18:45:33 sarchx64 kernel: Code: 05 e2 bd 76 6c 49 8b 00 49 83 78 10 00 48 89 04 24 0f 84 57 01 00 00 48 85 c0 0f 84 4e 01 00 00 8b 4d 28 48 8b 7d 00 48 01 c1 <48> 8b 19 48 89 ce 48 33 9d b8 00 00 00 48 0f ce 48 31 f3 40 f6 c7
févr. 22 18:45:33 sarchx64 kernel: RSP: 0018:ffffb31e22d23cd0 EFLAGS: 00010286
févr. 22 18:45:33 sarchx64 kernel: RAX: eb6f95a0d500b33f RBX: 0000000000000cc0 RCX: eb6f95a0d500b36f
févr. 22 18:45:33 sarchx64 kernel: RDX: 000000000003a2c5 RSI: 0000000000000cc0 RDI: 00000000000300c0
févr. 22 18:45:33 sarchx64 kernel: RBP: ffff9ede40043a00 R08: ffff9ee55fcf00c0 R09: 0000000000000000
févr. 22 18:45:33 sarchx64 kernel: R10: ffff9ee1d49c84b8 R11: ffffb31e22d23d40 R12: 0000000000000cc0
févr. 22 18:45:33 sarchx64 kernel: R13: 0000000000000048 R14: ffffffffc147a629 R15: 0000000000000000
févr. 22 18:45:33 sarchx64 kernel: FS: 00007efc19f45940(0000) GS:ffff9ee55fcc0000(0000) knlGS:0000000000000000
févr. 22 18:45:33 sarchx64 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
févr. 22 18:45:33 sarchx64 kernel: CR2: 00007fd79b9fa000 CR3: 00000001448b4000 CR4: 00000000000406e0
févr. 22 18:45:33 sarchx64 kernel: Call Trace:
févr. 22 18:45:33 sarchx64 kernel: nouveau_gem_object_close+0x119/0x1f0 [nouveau]
févr. 22 18:45:33 sarchx64 kernel: drm_gem_object_release_handle+0x30/0x90 [drm]
févr. 22 18:45:33 sarchx64 kernel: drm_gem_handle_delete+0x58/0x90 [drm]
févr. 22 18:45:33 sarchx64 kernel: ? drm_gem_handle_create+0x40/0x40 [drm]
févr. 22 18:45:33 sarchx64 kernel: drm_ioctl_kernel+0xb2/0x100 [drm]
févr. 22 18:45:33 sarchx64 kernel: drm_ioctl+0x215/0x390 [drm]
févr. 22 18:45:33 sarchx64 kernel: ? drm_gem_handle_create+0x40/0x40 [drm]
févr. 22 18:45:33 sarchx64 kernel: nouveau_drm_ioctl+0x55/0xa0 [nouveau]
févr. 22 18:45:33 sarchx64 kernel: __x64_sys_ioctl+0x83/0xb0
févr. 22 18:45:33 sarchx64 kernel: do_syscall_64+0x33/0x40
févr. 22 18:45:33 sarchx64 kernel: entry_SYSCALL_64_after_hwframe+0x44/0xa9

The system is not reliable enough now to stay in production!

Was working great under 5.4.

Additional info:
* package version(s)
* config and/or log files etc.
* link to upstream bug report, if any

Mainboard is a Supermicro H8SGL/H8SGL, BIOS 3.5b 03/18/2016
with an AMD Opteron(tm) Processor 6338P
and a NVIDIA Corporation GK208B [GeForce GT 730] (rev a1)

Linux sarchx64 5.10.17-1-lts #1 SMP Wed, 17 Feb 2021 11:11:31 +0000 x86_64 GNU/Linux

nouveau-fw 340.32-1
xf86-video-nouveau 1.0.17-1
xorg-server 1.20.10-3

Steps to reproduce:
upgrade from 5.4 lts to latest 5.10

I seem to always get hit when using firefox after about a 15-30 minutes.
It happens frequently but I can't say I can reproduce after a serious a keystrokes.