NetBSD-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: wired memory leak?



On Sat, 31 Oct 2026, Greg Troxel wrote:

My desktop has the wired growth problem, and

 cpu0: "Intel(R) Core(TM) i7-9700 CPU @ 3.00GHz"
 cpu0: Intel 7th or 8th gen Core (Kaby Lake, Coffee Lake) or Xeon E (Coffee Lake) (686-class), 3000.00 MHz
 cpu0: CPU base freq 3000000000 Hz
 cpu0: CPU max freq 4700000000 Hz
 cpu0: TSC freq CPUID 3000000000 Hz
 cpu0: family 0x6 model 0x9e stepping 0xd (id 0x906ed)

 pchb0 at pci0 dev 0 function 0: Intel Xeon E (S, Desktop) Host Bridge, DRAM (rev. 0x0d)
 i915drmkms0 at pci0 dev 2 function 0: Intel UHD Graphics 630 (rev. 0x02)


Is this with the `modesetting' DDX driver? Because, if I use the `intel' Xorg
driver, all wired memory is reclaimed. The downside is that the `intel' driver
is old, obsolete, and crashed the kernel when I did a kernel compile while
watching the pretty fishies[1].

A small table:

Driver			Wired mem grows?
------			----------------
DRM/modesetting		yes
DRM/intel		no
DRM/wsfb		no
genfb/wsfb		no


HTH,

-RVP

Note that to get DRMKMS to work on my HW at all, I had to apply the patch
recently appended to PR 58657. My HW is:

```
cpu0: Intel(R) Core(TM) i3-1005G1 CPU @ 1.20GHz, id 0x706e5
[...]
pchb0 at pci0 dev 0 function 0: Intel product 8a02 (rev. 0x03)
i915drmkms0 at pci0 dev 2 function 0: Intel product 8a56 (rev. 0x07)
[...]
{drm:netbsd:i915_ggtt_probe_hw+0x4b2} GGTT size = 4096M
{drm:netbsd:i915_ggtt_probe_hw+0x4d0} GMADR size = 256M
{drm:netbsd:i915_ggtt_probe_hw+0x4f9} DSM size = 0M
[...]
intelfb0 at i915drmkms0
i915 device info: pciid=0x8a56 rev=0x07 platform=ICELAKE (subplatform=0x1) gen=11
[...]
intelfb0: framebuffer at 0x4000040000, size 1920x1200, depth 32, stride 7680
```

---
[1]: https://webglsamples.org/aquarium/aquarium.html

Also: (warning: some of the demos crash even Linux!)
[2]: https://www.shadertoy.com


Home | Main Index | Thread Index | Old Index