Puzzling crash and strange reporting Re: ATI video card not recognized

[Reinoud Zandijk <reinoud%gorilla.13thmonkey.org@localhost>]

Hi,

On Wed, Dec 06, 2023 at 09:52:54AM +0100, Reinoud Zandijk wrote:
> On Mon, Dec 04, 2023 at 04:15:39PM +0100, Reinoud Zandijk wrote:
> > On Tue, Jun 23, 2020 at 01:26:21PM +0200, Reinoud Zandijk wrote:
> > > my old videocard died and I replaced it with a slightly newer one but it isn't
> > > recognized and nothing other than vga0 attaches. Its an Gigabyte Radeon RX460
> > > with 2 GB ram.
> > > 
> > > 002:00:0: ATI Technologies Radeon RX460 (VGA display, revision 0xcf)
> > > 002:00:1: ATI Technologies Radeon RX 460/550/640SP, RX 560/560X HD Audio
> > > Controller (mixed mode multimedia)
> > > 
> > 
> > Back again :) I tried out the videocard again in 10.0 i(beta) and got a lot
> > further. However I still stumble on a panic when starting X :

A puzzling report and a worrysome crash occured while resizing a Firefox
window:

...
[     1.000000] NetBSD 10.99.10 (GENERIC) #0: Mon Dec  4 16:01:51 CET 2023
[     1.000000] 	reinoud%gorilla.13thmonkey.org@localhost:/usr/sources/cvs.netbsd.org/src-clean/sys/arch/amd64/compile/obj/GENERIC
[     1.000000] total memory = 65456 MB
[     1.000000] avail memory = 63301 MB
...
[     4.627885] kern.module.path=/stand/amd64/10.99.10/modules
[     4.640006] [drm] initializing kernel modesetting (POLARIS11 0x1002:0x67EF 0x1458:0x22D6 0xCF).
[     4.640006] [drm] register mmio base: 0xFCE00000
[     4.640006] [drm] register mmio size: 262144
[     4.640006] [drm] PCIE atomic ops is not supported
[     4.640006] [drm] add ip block number 0 <vi_common>
[     4.640006] [drm] add ip block number 1 <gmc_v8_0>
[     4.640006] [drm] add ip block number 2 <tonga_ih>
[     4.640006] [drm] add ip block number 3 <gfx_v8_0>
[     4.640006] [drm] add ip block number 4 <sdma_v3_0>
[     4.640006] [drm] add ip block number 5 <powerplay>
[     4.640006] [drm] add ip block number 6 <dm>
[     4.648106] [drm] add ip block number 7 <uvd_v6_0>
[     4.648106] [drm] add ip block number 8 <vce_v3_0>
[     4.807888] ATOM BIOS: 113-TIC15322-X01
[     4.807888] [drm] UVD is enabled in VM mode
[     4.807888] [drm] UVD ENC is enabled in VM mode
[     4.807888] [drm] VCE enabled in VM mode
[     4.807888] [drm] vm size is 256 GB, 2 levels, block size is 10-bit, fragment size is 9-bit
[     4.818504] amdgpu0: VRAM: 2048M 0x000000F400000000 - 0x000000F47FFFFFFF (2048M used)
[     4.818504] amdgpu0: GART: 256M 0x000000FF00000000 - 0x000000FF0FFFFFFF
[     4.818504] [drm] Detected VRAM RAM=2048M, BAR=256M
[     4.818504] [drm] RAM width 128bits GDDR5
[     4.818504] Zone  kernel: Available graphics memory: 9007199252279140 KiB
?????

[     4.818504] Zone   dma32: Available graphics memory: 2097152 KiB
[     4.818504] [drm] amdgpu: 2048M of VRAM memory ready
[     4.818504] [drm] amdgpu: 3072M of GTT memory ready.
[     4.818504] [drm] GART: num cpu pages 65536, num gpu pages 65536
[     4.827886] [drm] PCIE GART of 256M enabled (table at 0x000000F400300000).
[     4.827886] amdgpu0: interrupting at msi10 vec 0 (amdgpu0)
[     4.827886] [drm] Chained IB support enabled!
[     4.827886] hwmgr_sw_init smu backed is polaris10_smu
[     4.827886] powerplay sw init successfully
[     4.827886] [drm] Found UVD firmware Version: 1.130 Family ID: 16
[     4.837885] [drm] Found VCE firmware Version: 35.1a Binary ID: 3
[     4.897884] Can't find requested voltage id in vdd_dep_on_sclk table
[     4.907972] message 308 was not supported
[     4.907972] last message was not supported
....

[   188.737147] [SCALER]:resource_build_scaling_params: Viewport:
[   188.737147] height:2160 width:3840 x:0 y:0
[   188.737147]  dst_rect:
[   188.737147] height:2160 width:3840 x:0 y:0
[   188.737147] [BANDWIDTH_CALCS]:dce112_validate_bandwidth: start[SURFACE]:Pipe:0 0xffffe647107765c8: addr hi:0xf4, addr low:0x7ba20000, src: 0, 0, 3840, 2160; dst: 0, 0, 3840, 2160;clip: 0, 0, 3840, 2160
[   188.737147] [SURFACE]:Pipe 0: width, height, x, y
[   188.737147] viewport:3840, 2160, 0, 0
[   188.737147] recout:  3840, 2160, 0, 0
[   231.876974] fatal protection fault in supervisor mode
[   231.876974] trap type 4 code 0 rip 0xffffffff80e1b829 cs 0x8 rflags 0x10286 cr2 0x708193687000 ilevel 0x8 rsp 0xffffaa90bb6ca8d0
[   231.876974] curlwp 0xffffe646e0ba5c00 pid 4574.5275 lowest kstack 0xffffaa90bb6c62c0
kernel: protection fault[ 231.8769741]  trap, code=0
Stopped in pid 4574.5275 (firefox) at	netbsd:phtree_SPLAY+0x37:	cmpq	%rdx,20(%rcx)
phtree_SPLAY() at netbsd:phtree_SPLAY+0x37
phtree_SPLAY_INSERT() at netbsd:phtree_SPLAY_INSERT+0xf
pool_grow() at netbsd:pool_grow+0x393
pool_get() at netbsd:pool_get+0xa8
pool_cache_get_slow() at netbsd:pool_cache_get_slow+0x139
pool_cache_get_paddr() at netbsd:pool_cache_get_paddr+0x263
kmem_intr_alloc() at netbsd:kmem_intr_alloc+0xcb
amdgpu_bo_list_create() at amdgpu:amdgpu_bo_list_create+0x3a
amdgpu_cs_ioctl() at amdgpu:amdgpu_cs_ioctl+0xf14
drm_ioctl() at netbsd:drm_ioctl+0x23d
drm_ioctl_shim() at netbsd:drm_ioctl_shim+0x34
sys_ioctl() at netbsd:sys_ioctl+0x56d
syscall() at netbsd:syscall+0x217
[   231.876974] --- syscall (number 54) ---
netbsd:syscall+0x217[ 231.8769741] :

[   231.876974] dumping to dev 168,3
--------

What also struck me is that if I enable WebGL in Firefox it plays well in one
tab. However if I open a 2nd tab with WebGL it either sometimes doesn't work
but mostly just crashes X. I suspect the GL context gets mixed up. This might
also play with the compositor though that mostly just screws up the display.

Any thoughs about this?

With regards,

Reinoud