In the Linux kernel, the following vulnerability has been resolved:
drm/amdgpu: Fix kernel panic during driver load failure
Avoid kernel panic if MES init fails during driver load. The KIQ ring is falsely marked as ready as ASICs that use MES, KIQ is owned by MES.
BUG: kernel NULL pointer dereference, address: 0000000000000000 RIP: 0010:gfxv121waitregmem+0x5a/0x1f0 [amdgpu] Call Trace: gfxv121ringemitregwriteregwait+0x1f/0x30 [amdgpu] amdgpugmcfwregwriteregwait+0xb2/0x190 [amdgpu] amdgpugmcflushgputlb+0x1cc/0x230 [amdgpu] amdgpugartinvalidatetlb+0x81/0xa0 [amdgpu] amdgpugartunbind+0x72/0x90 [amdgpu] amdgputtmbackendunbind+0xa4/0xb0 [amdgpu] amdgputtmttunpopulate+0x13/0xd0 [amdgpu] amdttmttunpopulate+0x29/0x70 [amdttm] ttmboput+0x1eb/0x360 [amdttm] amdgpubofreekernel+0xf9/0x1f0 [amdgpu] amdgpuihringfini+0x5a/0x90 [amdgpu] amdgpuirqfinihw+0x58/0x80 [amdgpu] amdgpudevicefinihw+0x4e0/0x5b0 [amdgpu] amdgpudriverloadkms+0x60/0xa0 [amdgpu] amdgpupciprobe+0x28e/0x6d0 [amdgpu] pcideviceprobe+0x19f/0x220 reallyprobe+0x1ed/0x340 driverprobedevice+0x1e/0x80 __driverattach+0xd3/0x1a0 busfor_eachdev+0x68/0xa0 busadddriver+0x19f/0x270 driverregister+0x5d/0xf0 dooneinitcall+0xac/0x200 doinitmodule+0x1ec/0x280 _sesysfinitmodule+0x2de/0x310 dosyscall64+0x6a/0x250 entrySYSCALL64afterhwframe+0x4b/0x53
(cherry picked from commit 4623b958dd6da0f4c3026afdf330626a09ecb0f0)
{
"osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/68xxx/CVE-2026-68105.json",
"cna_assigner": "Linux"
}