PT-2026-47348 · Linux · Linux

Published

2026-06-08

·

Updated

2026-06-08

·

CVE-2026-46276

None

No severity ratings or metrics are available. When they are, we'll update the corresponding info on the page.
In the Linux kernel, the following vulnerability has been resolved:
drm/amdgpu: fix zero-size GDS range init on RDNA4
RDNA4 (GFX 12) hardware removes the GDS, GWS, and OA on-chip memory resources. The gfx v12 0 initialisation code correctly leaves adev->gds.gds size, adev->gds.gws size, and adev->gds.oa size at zero to reflect this.
amdgpu ttm init() unconditionally calls amdgpu ttm init on chip() for each of these resources regardless of size. When the size is zero, amdgpu ttm init on chip() forwards the call to ttm range man init(), which calls drm mm init(mm, 0, 0). drm mm init() immediately fires DRM MM BUG ON(start + size <= start) -- trivially true when size is zero -- crashing the kernel during modprobe of amdgpu on an RX 9070 XT.
Guard against this by returning 0 early from amdgpu ttm init on chip() when size in page is zero. This skips TTM resource manager registration for hardware resources that are absent, without affecting any other GPU type.
DRM MM BUG ON() only asserts if CONFIG DRM DEBUG MM is enabled in the kernel config. This is apparently rarely enabled as these chips have been in the market for over a year and this issue was only reported now.
Oops-Analysis: http://oops.fenrus.org/reports/bugzilla.korg/221376/report.html (cherry picked from commit 5719ce5865279cad4fd5f01011fe037168503f2d)

Related Identifiers

CVE-2026-46276

Affected Products

Linux