263 matches found
CVE-2026-76042
Use of uninitialized resource in GPU in Google Chrome prior to 151.0.7922.169 allowed a remote attacker who had compromised the renderer process to read memory outside the sandbox via a crafted HTML page. Chromium security severity: High...
UBUNTU-CVE-2026-74447
In the Linux kernel, the following vulnerability has been resolved: drm/amdkfd: fix uint32t overflow in EOP ring buffer size alignment eopringbuffersize in struct queueproperties is a u32. In kfdqueueacquirebuffers the expected EOP buffer size is computed as ALIGNeopringbuffersize, PAGESIZE; ALIG...
CVE-2026-68260
In the Linux kernel, the following vulnerability has been resolved: drm/imagination: acquire vmctx-lock before mapping memory to GPU VM The drm gpuvm code doesn't protect find operation against map operation, and the driver needs to ensure a map operation shouldn't happen when a find operation is...
CVE-2026-68260
CVE-2026-68260 affects the Linux kernel, specifically the drm/imagination driver. The issue arises because the GPU VM code does not properly synchronize find operations with map/unmap operations, allowing a race that can lead to a NULL pointer dereference in certain memory operations. The resolve...
CVE-2026-68260 drm/imagination: acquire vm_ctx->lock before mapping memory to GPU VM
In the Linux kernel, the following vulnerability has been resolved: drm/imagination: acquire vmctx-lock before mapping memory to GPU VM The drm gpuvm code doesn't protect find operation against map operation, and the driver needs to ensure a map operation shouldn't happen when a find operation is...
SLAC: Access-Driven CPU-To-GPU Side-Channel Attacks Via System-Level Cache on Apple Silicon
Modern heterogeneous System-on-Chip designs integrate CPU cores and a GPU that share a last-level cache LLC or system-level cache SLC. This sharing exposes a new cross-domain attack surface, and existing attacks on integrated platforms either exploit coarse-grained cache-occupancy contention or...
EUVD-2026-48746
An integer overflow when calculating physical offsets for sparse PMRs may result in 32-bit truncation of address computations for PMRs larger than 4 GB. This can lead to incorrect GPU MMU mappings and may allow a non-privileged user to trigger access to unintended physical memory, resulting in...
CVE-2026-16280 GPU DDK - Integer overflow in _PMRLogicalOffsetToPhysicalOffset
An integer overflow when calculating physical offsets for sparse PMRs may result in 32-bit truncation of address computations for PMRs larger than 4 GB. This can lead to incorrect GPU MMU mappings and may allow a non-privileged user to trigger access to unintended physical memory, resulting in...
CVE-2026-49744 GPU DDK - Unchecked ui32TracePointer in rgxfw_log_ex()
Kernel software installed and running inside a Guest VM may post improper commands to the GPU Firmware to trigger a write of data outside the Guest's virtualised GPU memory. Out of bounds accesses triggered by malware introduced to a Guest KMD could allow privilege escalation which escapes...
PT-2026-64421
Name of the Vulnerable Software and Affected Versions GPU DDK affected versions not specified Description An integer overflow occurs during the calculation of physical offsets for sparse PMRs Partial Memory Regions within the PMRLogicalOffsetToPhysicalOffset function. For PMRs larger than 4 GB,...
vLLM: GGUF dequantize kernel int truncation exposes uninitialized GPU memory in multi-tenant serving
SummaryInteger truncation of tensor dimensions in vLLM's GGUF dequantize kernels csrc/quantization/gguf/ggufkernel.cu causes partial tensor processing. The output tensor is allocated at full size via torch::empty uninitialized memory, but the dequantize CUDA kernel processes only a truncated numb...
EUVD-2026-40609
Uninitialized Use in GPU in Google Chrome on Android prior to 150.0.7871.47 allowed a remote attacker to obtain potentially sensitive information from process memory via a crafted HTML page. Chromium security severity: Medium...
CVE-2026-13030
Uninitialized Use in GPU in Google Chrome on Android prior to 149.0.7827.197 allowed a remote attacker to obtain potentially sensitive information from process memory via a crafted HTML page. Chromium security severity: High...
CVE-2026-53923
vLLM is an inference and serving engine for large language models LLMs. From 0.5.5 until 0.23.1rc0, integer truncation of tensor dimensions in vLLM's GGUF dequantize kernels csrc/quantization/gguf/ggufkernel.cu causes partial tensor processing. The output tensor is allocated at full size via...
CVE-2026-53923 vLLM GGUF Kernels: int64_t to int truncation of tensor dimensions causes GPU buffer overflow
vLLM is an inference and serving engine for large language models LLMs. From 0.5.5 until 0.23.1rc0, integer truncation of tensor dimensions in vLLM's GGUF dequantize kernels csrc/quantization/gguf/ggufkernel.cu causes partial tensor processing. The output tensor is allocated at full size via...
CVE-2026-53923 vLLM GGUF Kernels: int64_t to int truncation of tensor dimensions causes GPU buffer overflow
vLLM is an inference and serving engine for large language models LLMs. From 0.5.5 until 0.23.1rc0, integer truncation of tensor dimensions in vLLM's GGUF dequantize kernels csrc/quantization/gguf/ggufkernel.cu causes partial tensor processing. The output tensor is allocated at full size via...
CVE-2026-53923: Incorrect Conversion between Numeric Types
vLLM is an inference and serving engine for large language models LLMs. From 0.5.5 until 0.23.1rc0, integer truncation of tensor dimensions in vLLM's GGUF dequantize kernels csrc/quantization/gguf/ggufkernel.cu causes partial tensor processing. The output tensor is allocated at full size via...
vLLM: GGUF dequantize kernel int truncation exposes uninitialized GPU memory in multi-tenant serving
Summary Integer truncation of tensor dimensions in vLLM's GGUF dequantize kernels csrc/quantization/gguf/ggufkernel.cu causes partial tensor processing. The output tensor is allocated at full size via torch::empty uninitialized memory, but the dequantize CUDA kernel processes only a truncated...
CVE-2026-10008
Uninitialized Use in GPU in Google Chrome on Android prior to 148.0.7778.216 allowed a remote attacker to obtain potentially sensitive information from process memory via a crafted HTML page. Chromium security severity: High...
CVE-2026-46229
The CVE-2026-46229 issue affects the Linux kernel’s DRM/AMDKFD path: KFD VRAM allocations could leave stale data because AMDGPU_GEM_CREATE_VRAM_CLEARED was not applied in the KFD code path, unlike the GEM/user paths which already set VRAM_CLEARED. This allowed stale page-table remnants to leak in...