Vulnerability · Published May 21, 2024 · Assigned by Linux

CVE-2023-52738drm/amdgpu/fence: Fix oops due to non-matching drm_sched init/fini

Medium 5.3

In plain English

A medium-severity security flaw in Linux from Linux. There are no confirmed reports of exploitation so far. The chance it is exploited in the next 30 days is estimated at 0.8% (higher than 57% of all known vulnerabilities). Fixed versions are available - upgrade the affected package.

Severity (CVSS 3.1)5.3Medium
Exploit odds (30 days)0.8%top 43.4% of all CVEs
Exploited in the wildNo reportsCISA SSVC: none
Attack toolingNoneno public tooling known

Official description

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/fence: Fix oops due to non-matching drm_sched init/fini Currently amdgpu calls drm_sched_fini() from the fence driver sw fini routine - such function is expected to be called only after the respective init function - drm_sched_init() - was executed successfully. Happens that we faced a driver probe failure in the Steam Deck recently, and the function drm_sched_fini() was called even without its counter-part had been previously called, causing the following oops: amdgpu: probe of 0000:04:00.0 failed with error -110 BUG: kernel NULL pointer dereference, address: 0000000000000090 PGD 0 P4D 0 Oops: 0002 [#1] PREEMPT SMP NOPTI CPU: 0 PID: 609 Comm: systemd-udevd Not tainted 6.2.0-rc3-gpiccoli #338 Hardware name: Valve Jupiter/Jupiter, BIOS F7A0113 11/04/2022 RIP: 0010:drm_sched_fini+0x84/0xa0 [gpu_sched] [...] Call Trace: <TASK> amdgpu_fence_driver_sw_fini+0xc8/0xd0 [amdgpu] amdgpu_device_fini_sw+0x2b/0x3b0 [amdgpu] amdgpu_driver_release_kms+0x16/0x30 [amdgpu] devm_drm_dev_init_release+0x49/0x70 [...] To prevent that, check if the drm_sched was properly initialized for a given ring before calling its fini counter-part. Notice ideally we'd use sched.ready for that; such field is set as the latest thing on drm_sched_init(). But amdgpu seems to "override" the meaning of such field - in the above oops for example, it was a GFX ring causing the crash, and the sched.ready field was set to true in the ring init routine, regardless of the state of the DRM scheduler. Hence, we ended-up using sched.ops as per Christian's suggestion [0], and also removed the no_scheduler check [1]. [0] https://lore.kernel.org/amd-gfx/984ee981-2906-0eaf-ccec-9f80975cb136@amd.com/ [1] https://lore.kernel.org/amd-gfx/cd0e2994-f85f-d837-609f-7056d5fb7231@amd.com/

Affected packages & fixed versions

EcosystemPackageFixed inAdvisory
Debian:12linux6.1.12-1DEBIAN-CVE-2023-52738
Debian:13linux6.1.12-1DEBIAN-CVE-2023-52738
Debian:14linux6.1.12-1DEBIAN-CVE-2023-52738
Ubuntu:22.04:LTSlinux5.15.0-72.79UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-allwinner-5.19no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-aws5.15.0-1036.40UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-aws-5.19no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-aws-6.2no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-azure5.15.0-1038.45UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-azure-5.19no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-azure-6.2no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-azure-fde5.15.0-1038.45.1UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-azure-fde-5.19no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-azure-fde-6.2no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-gcp5.15.0-1034.42UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-gcp-5.19no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-gcp-6.2no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-gke5.15.0-1033.38UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-gkeop5.15.0-1020.25UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-hwe-5.19no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-hwe-6.2no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-ibm5.15.0-1030.33UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-intel-iotg5.15.0-1030.35UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-kvm5.15.0-1033.38UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-lowlatency5.15.0-72.79UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-lowlatency-hwe-5.19no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-lowlatency-hwe-6.2no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-nvidia5.15.0-1025.25UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-nvidia-6.2no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-oem-5.17no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-oem-6.0no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-oem-6.1no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-oracle5.15.0-1035.41UBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-raspi5.15.0-1029.31UBUNTU-CVE-2023-52738
Ubuntu:24.04:LTSlinux-raspi-realtimeno fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-realtimeno fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-riscvno fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-riscv-5.19no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-starfive-5.19no fix listedUBUNTU-CVE-2023-52738
Ubuntu:22.04:LTSlinux-starfive-6.2no fix listedUBUNTU-CVE-2023-52738

Affected products (as reported)

VendorProductVersions
Linux Linux 067f44c8b4590c3f24d21a037578a478590f2175 – before 2e557c8ca2c585bdef591b8503ba83b85f5d0afd067f44c8b4590c3f24d21a037578a478590f2175 – before 2bcbbef9cace772f5b7128b11401c515982de34b067f44c8b4590c3f24d21a037578a478590f2175 – before 5ad7bbf3dba5c4a684338df1f285080f2588b5358ba968ae672b3075794c8086aa164595b0175abe5.14.10 – before 5.15
Linux Linux 5.15

References