Automotive Vulnerabilities. Hunted worldwide every 24 hours, across 🇺🇸🇨🇳🇯🇵🇰🇷🇹🇼🇩🇪🇫🇷🇨🇭🇦🇹🇳🇱🇧🇪🇵🇱🇨🇦🇱🇺🇪🇺. Stay ahead.
A flaw in the Linux kernel's ecryptfs component allows a malformed tag 70 packet with a body smaller than the fixed metadata fields to underflow a size calculation, which could let an attacker cause a denial of service or other undefined behavior. The advisory does not state a severity rating. This could be relevant where the Linux kernel is deployed in automotive systems; confirm applicability through the product SBOM or dependency inventory.
A flaw in the Linux kernel's ecryptfs component can leave a message context stuck on an allocated list when sending to a userspace daemon fails, preventing that context from being reused. An attacker able to trigger this failure could cause a resource leak that degrades or exhausts the affected system's ability to process ecryptfs messages. The advisory does not state a severity rating. This could be relevant where the Linux kernel is deployed in automotive systems, so confirm applicability through the product SBOM or dependency inventory.
The advisory describes a rate-limiting change to the efivarfs statfs() handler, a Linux kernel filesystem component, but it does not state a specific vulnerability, attacker capability, or severity. This could be relevant where efivarfs is deployed in automotive systems, since it relates to EFI variable handling in embedded Linux environments. Confirm applicability through the product SBOM or dependency inventory.
The Linux kernel's erofs filesystem has a flaw in how it resizes its global buffer pool: if a resize partially succeeds and then fails, a retry can either fail with an out-of-memory error or overflow a temporary pointer array by copying more entries than it can hold. This could allow memory corruption or a denial-of-service condition, though the advisory does not state a severity rating. This could be relevant where the Linux kernel's erofs filesystem is deployed in automotive systems. Confirm applicability through the product SBOM or dependency inventory.
The Linux kernel ext2 filesystem had a flaw where inode updates for IS_SYNC inodes could be lost because sync_inode_metadata() only acts on already-dirty inodes, so changes made by ext2_setsize() and ext2_xattr_set2() might not be written back. An attacker or faulty operation could therefore cause inode metadata changes to be silently dropped, though the advisory does not state a severity rating. This could be relevant where the Linux kernel with ext2 is deployed in automotive systems; confirm applicability through the product SBOM or dependency inventory.
The Linux kernel's omapfb panel-dsi-cm display driver registers its display before initializing a mutex lock, so another consumer could reach a callback that uses the uninitialized lock. This could allow an attacker to trigger a race condition leading to undefined behavior in the affected driver. The advisory does not state a severity rating. This could be relevant where the omapfb panel-dsi-cm driver is deployed in automotive systems. Confirm applicability through the product SBOM or dependency inventory.
The advisory describes a Linux kernel fbdev driver change for ssd1307fb that defers I2C transfers out of damage callbacks, affecting systems using that display driver. The advisory does not state an attacker capability or a severity rating, so no exploit impact or severity level can be reported from the supplied text. This could be relevant where the ssd1307fb display driver is deployed in automotive systems, but the advisory does not confirm automotive use. Confirm applicability through the product SBOM or dependency inventory.
The Linux kernel's uvesafb framebuffer driver fails to unregister its v86d connector callback when platform driver registration fails during initialization, leaving the callback registered. This is a cleanup flaw rather than an attacker-exploitable condition, and the advisory states no severity rating. This could be relevant where the Linux kernel is deployed in automotive systems; confirm applicability through the product SBOM or dependency inventory.
The Linux kernel's forcedeth Ethernet driver has an off-by-one error in its suspend and resume code that causes an out-of-bounds memory access and a read/write one dword past the mapped hardware register window. An attacker generally cannot exploit this directly; the stated impact is a kernel trap when CONFIG_UBSAN_TRAP is enabled, aborting running kernel code, and the flaw was caught by UBSAN on an Apple Macmini3,1 (MCP79) system. This could be relevant where the forcedeth driver is deployed in automotive embedded systems; confirm applicability through the product SBOM or dependency inventory.
A race condition in the Linux kernel's fsnotify subsystem can leave an object's event mask stale after concurrent mark updates, causing fanotify and inotify to miss events such as FAN_MODIFY or IN_MODIFY. An attacker able to trigger this interleaving could cause writes to fail to generate expected notifications for the affected group. The advisory does not state a severity rating. This could be relevant where the Linux kernel is deployed in automotive systems; confirm applicability through the product SBOM or dependency inventory.
The Linux kernel has a vulnerability in its hugetlb memory management code where unmapping a memory range can incorrectly adjust the reservation count, causing it to underflow when a page is mapped in both a parent and child process. An attacker or faulty operation could trigger this underflow, though the advisory does not state a severity rating. This could be relevant where the Linux kernel is deployed in automotive systems; confirm applicability through the product SBOM or dependency inventory.
The Linux kernel's OMAP SSI driver fails to initialize a DMA mask for its synthetic HSI controller device, which can cause crashes or warnings during DMA mapping operations after recent DMA API hardening changes. The issue is resolved by explicitly setting a 32-bit DMA mask matching the hardware capabilities. This could be relevant where the Linux kernel is deployed in automotive systems, particularly on OMAP-based platforms. Confirm applicability through the product SBOM or dependency inventory.
The Linux kernel's accel/rocket driver has a flaw where an error during job submission can cause cleanup code to use an uninitialized domain pointer, and a related issue where a freed task array could be freed again. An attacker able to trigger these error paths could cause memory corruption or a system crash. The advisory does not state a severity rating. This could be relevant where the Linux kernel accel/rocket driver is deployed in automotive systems. Confirm applicability through the product SBOM or dependency inventory.
The Linux kernel has a flaw in its ACPI/APEI GHES error-handling code for CXL CPER work locks, where a lock held with interrupts enabled can deadlock against the same lock acquired from a hardware interrupt handler. An attacker able to trigger the relevant error notification path could cause the affected CPU to hang, resulting in a denial of service. The advisory does not state a severity rating. This could be relevant where the Linux kernel is deployed in automotive systems, but the advisory does not confirm any automotive deployment. Confirm applicability through the product SBOM or dependency inventory.
The Linux kernel's "rocket" accelerator driver has error-path bugs in rocket_job_run() that can leak a fence reference and a runtime power-management reference, potentially preventing the NPU from suspending, and can return an unsignaled fence that triggers a kernel warning. The advisory does not state an automotive connection; this could be relevant where the rocket accelerator driver is deployed in automotive systems. Confirm applicability through the product SBOM or dependency inventory.
A flaw in the Linux kernel's ACPI APEI GHES error-handling code miscalculates the remaining length of an ARM error record, which can cause the parser to read past the end of the error section when the record count is large. This could lead to out-of-bounds memory access in affected systems. This could be relevant where the Linux kernel is deployed in automotive systems; confirm applicability through the product SBOM or dependency inventory.
A flaw in the Linux kernel's auxdisplay charlcd driver, when built with the CHARLCD_BL_FLASH option, can leave a scheduled backlight work item pointing at memory that has already been freed if device registration fails. An attacker able to trigger that registration failure could cause the kernel to access freed memory, which may lead to a crash or other unpredictable behavior. The advisory does not state a severity rating. This could be relevant where the Linux kernel is deployed in automotive systems; confirm applicability through the product SBOM or dependency inventory.
A flaw in the Linux kernel's ATA SCSI translation code causes TRIM/discard operations to fail completely on storage devices with logical sector sizes larger than 2048 bytes, such as 4Kn drives, because the TRIM descriptor is incorrectly sized against the logical sector size instead of the fixed 512-byte page format. An attacker or normal user attempting a TRIM on such a device would trigger repeated "Parameter list length error" rejections and kernel WARN_ON() splats, effectively breaking discard functionality. The advisory does not state an explicit severity rating. This could be relevant where Linux-based storage components are deployed in automotive systems; confirm applicability through the product SBOM or dependency inventory.
The Linux kernel block layer did not validate user-space I/O vectors against a device's DMA alignment when building bios, so misaligned requests could reach drivers that do not check alignment, potentially causing improper handling. An attacker able to submit crafted I/O could trigger this misalignment, though the advisory does not state a specific severity rating. This could be relevant where the Linux kernel block layer is deployed in automotive systems; confirm applicability through the product SBOM or dependency inventory.
The advisory describes a double free in the bnx2x_init_firmware() error path in the bnx2x network driver, which could allow an attacker to corrupt memory or crash the affected system. This could be relevant where the bnx2x driver is deployed in automotive systems, though the advisory does not confirm any automotive use. Confirm applicability through the product SBOM or dependency inventory.
The Linux kernel's dm-io and dm-raid1 device-mapper components were updated so that non-retryable errors (BLK_STS_NOTSUPP and BLK_STS_INVAL) are tracked separately and no longer cause a RAID1 leg to fail, fixing misbehavior when unaligned bio vectors are used. An attacker or faulty workload could previously trigger unnecessary leg failures through such errors, though the advisory does not state a severity rating. This could be relevant where the Linux kernel is deployed in automotive systems; confirm applicability through the product SBOM or dependency inventory.
A flaw in the Linux kernel's dm-io component, used for device-mapper I/O, could cause I/O operations to hang indefinitely when handling misaligned direct I/O to a dm-mirror device, because the code's sector-based accounting could lose a sub-sector fragment and loop forever submitting empty bios. The fix clones the source bio instead of rebuilding its biovec, removing the faulty copy path for bio-backed I/O. This could be relevant where the Linux kernel's device-mapper subsystem is deployed in automotive systems; confirm applicability through the product SBOM or dependency inventory.
The Linux kernel's dm-era device-mapper target has a bug where a failed snapshot operation permanently leaks a metadata block, and repeated failures can gradually exhaust metadata space. This could be relevant where dm-era is deployed in automotive systems, since it is a generic kernel component rather than an automotive-specific one. Confirm applicability through the product SBOM or dependency inventory.
The advisory describes a fix in the Linux kernel's dm raid1 code that reserves space for a NUL-terminator in build_constructor_string(), a change that prevents a string from being improperly terminated. The advisory does not state a specific attacker capability or severity, so no technical impact or severity rating can be confirmed from the supplied text. This could be relevant where the Linux kernel is deployed in automotive systems, so confirm applicability through the product SBOM or dependency inventory.
A flaw in the Linux kernel's device-mapper array code allows a crafted metadata block to be accepted even when its value size does not match what the caller expects, which can cause out-of-bounds memory access. An attacker able to supply crafted metadata could trigger this, though the advisory does not state a severity rating. This could be relevant where the Linux kernel device-mapper subsystem is deployed in automotive systems; confirm applicability through the product SBOM or dependency inventory.