Missing Release of Memory after Effective Lifetime

Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Missing Release of Memory after Effective Lifetime via the decode-worker process. An attacker can cause memory exhaustion by submitting requests wit...

CVE-2026-93436 vLLM through 0.29.0 Memory Exhaustion via Rejected Requests

vLLM through 0.29.0 fails to properly clean up decode-side metadata for rejected inference requests in prefill/decode disaggregated deployments. Remote attackers can submit requests with maxtokens=0 to exhaust decode-worker memory without bound until the worker restarts...

CVE-2026-93436

vLLM through version 0.29.0 is vulnerable to a memory exhaustion issue in prefill/decode disaggregated deployments. The root cause is a failure to properly clean up decode-side metadata for rejected inference requests. A remote, unauthenticated attacker can submit requests with max_tokens=0 to ex...

freerdp: FreeRDP: Arbitrary code execution via heap buffer overflow in Opus audio decode

A flaw was found in FreeRDP, a free implementation of the Remote Desktop Protocol. A malicious Remote Desktop Protocol RDP server can exploit a heap buffer overflow vulnerability during Opus audio decoding. This occurs when the server negotiates Opus audio with a vulnerable client, causing the...

vLLM: Request-selected PyNvVideoCodec GPU decode bypasses static VRAM reservation

Summary Current vLLM main lets an inference request choose the PyNvVideoCodec GPU video decoder through mediaiokwargs.video.videobackend, but engine GPU memory reservation is computed only from static startup configuration and VLLMVIDEOLOADERBACKEND. If the server starts with the default...

CVE-2026-86040 libp2p: Unbounded RPC decode + synchronous subscription processing in @libp2p/floodsub allows unauthenticated DoS

libp2p is a JavaScript implementation of the libp2p networking stack. Prior to 11.0.26, @libp2p/floodsub accepts unauthenticated RPC frames on /floodsub/1.0.0 through PeerStreams.attachInboundStream in packages/floodsub/src/peer-streams.ts without protobuf element limits, then processRpc and...

CVE-2026-86040 libp2p: Unbounded RPC decode + synchronous subscription processing in @libp2p/floodsub allows unauthenticated DoS

libp2p is a JavaScript implementation of the libp2p networking stack. Prior to 11.0.26, @libp2p/floodsub accepts unauthenticated RPC frames on /floodsub/1.0.0 through PeerStreams.attachInboundStream in packages/floodsub/src/peer-streams.ts without protobuf element limits, then processRpc and...

CVE-2026-86040

CVE-2026-86040 affects @libp2p/floodsub in the libp2p JavaScript networking stack, prior to version 11.0.26 . The vulnerability stems from unauthenticated RPC frames on /floodsub/1.0.0 being decoded without protobuf element limits, and the subscription array being processed synchronously without ...

CVE-2026-24073

Memory corruption when processing decode statistics due to insufficient validation of offset against structure size...

CVE-2026-24073 Out-of-bounds Write in Video

Memory corruption when processing decode statistics due to insufficient validation of offset against structure size...

EUVD-2026-81320

Memory corruption when processing decode statistics due to insufficient validation of offset against structure size...

CVE-2026-24073 Out-of-bounds Write in Video

Memory corruption when processing decode statistics due to insufficient validation of offset against structure size...

CVE-2026-24073

CVE-2026-24073 is an out-of-bounds write in Qualcomm's Video component, specifically in the processing of decode statistics . The root cause is insufficient validation of an offset against structure size , leading to memory corruption. The vulnerability is rated CVSS 3.1 7.8 (HIGH) with a local a...

EUVD-2026-81317

Canva Desktop before v1.125.0 performed double decoding in the deeplink handler. A threat actor could cause the application to load arbitrary same-origin content under the user’s session...

CVE-2026-24073: Out-of-bounds Write

Memory corruption when processing decode statistics due to insufficient validation of offset against structure size...

CVE-2026-92972: Missing Authentication for Critical Function

SGLang through 0.5.19 in prefill/decode disaggregation mode contains an unauthenticated PUT /route endpoint on the prefill bootstrap service that allows attackers to poison the KV transfer routing table. Attackers can supply arbitrary rankip and rankport values to redirect decode workers to...

CVE-2026-93436: Missing Release of Memory after Effective Lifetime

vLLM through 0.29.0 fails to properly clean up decode-side metadata for rejected inference requests in prefill/decode disaggregated deployments. Remote attackers can submit requests with maxtokens=0 to exhaust decode-worker memory without bound until the worker restarts...

CVE-2026-54604: Reliance on Undefined, Unspecified, or Implementation-Defined Behavior

OpenSlide is a C library for reading whole slide image files. Prior to 4.0.1, a behavior change in libtiff 4.7.1 causes the indirect TIFF tile path in src/openslide-decode-tiff.c and openslidetiffreadtile to request a full-height destination for a partial bottom tile row, allowing uninitialized...

DEBIAN-CVE-2026-89827

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: avoid force-completing uninitialized UVD rings uvdv70swinit does not initialize the UVD decode ring for an SR-IOV VF. However, amdgpuuvdresume unconditionally force-completes the decode ring when restoring its fence...

CVE-2026-89827

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: avoid force-completing uninitialized UVD rings uvdv70swinit does not initialize the UVD decode ring for an SR-IOV VF. However, amdgpuuvdresume unconditionally force-completes the decode ring when restoring its fence...