Mardi 6 octobre 2026

SixLabors ImageSharp

ImageSharp is a 2D graphics library. From 1.0.0-beta0001 until 4.1.2, ICC CLUT parsing calculates allocation sizes from attacker-declared channel and grid dimensions before confirming that the profile contains the declared values. IccDataReader.ReadClutF32 can request a large float array, and earlier public IccProfile.Entries parsing paths can allocate a large jagged representation, from a short truncated profile. In version 4, automatic image conversion reaches the parser when DecoderOptions.ColorProfileHandling is Convert; the default Preserve mode avoids that conversion path. The demonstrated impact is memory pressure and input-validation failure, not unhandled process termination. This issue is fixed in version 4.1.2.

SixLabors ImageSharp

ImageSharp is a 2D graphics library. From 2.0.0 until 4.1.2, ExifReader.ReadValues64 trusts the 64-bit BigTIFF IFD entry count and iterates once per declared entry. When fewer than 20 bytes remain, ExifReader.ReadValue64 returns without advancing the stream or terminating the outer loop, so a small malformed BigTIFF can keep one decoder thread executing for an attacker-controlled duration. This report does not claim worker-pool exhaustion. This issue is fixed in version 4.1.2.

SixLabors ImageSharp

ImageSharp is a 2D graphics library. From 4.0.0 until 4.1.2, ExrBaseDecompressor.UndoZipCompression accepts a nonempty ZIP or ZIPS inflate result that is shorter than the EXR block's required size. ZipExrCompression.Decompress reconstructs the returned prefix while ExrDecoderCore processes the full expected block from a buffer obtained through Configuration.Default, allowing bytes retained from a completed prior ImageSharp operation to appear in decoded pixels. Applications that expose pixels or output from the later attacker-controlled EXR decode can disclose process-local image data. This issue is fixed in version 4.1.2.

vllm-project vllm · Vllm vllm

vLLM before 0.29.0 contains a resource-limit bypass vulnerability in PyNvVideoCodec decoder allocation where sampler subclass shadowing allows independent counter increments. Unauthenticated attackers can select different sampler subclasses in video requests to exceed configured decoder limits and exhaust unaccounted GPU memory.

vllm-project vllm · Vllm vllm

vLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate. When the request contains a 'features' (multimodal) payload, vllm/entrypoints/serve/disagg/serving.py builds a multimodal EngineInput directly from the caller-supplied token_ids, and GenerateRequest.token_ids (vllm/entrypoints/serve/disagg/protocol.py) is not checked against model_config.max_model_len. For multimodal processors that report skip_prompt_length_check=True (for example Nemotron Parse, Whisper, and FireRedLID), InputProcessor._validate_prompt_len() returns immediately for both encoder and decoder prompts, so an overlong prompt becomes an EngineCoreRequest and reaches the worker input-batch copy into a fixed max_model_len-wide NumPy row. A client able to reach the endpoint on an affected model configuration can therefore submit an overlong token_ids list to trigger a worker failure and denial of service. Fixed in 0.29.0.

vgmstream

A vulnerability was detected in vgmstream up to r2117. Affected by this vulnerability is the function ps_find_padding of the file src/coding/psx_decoder.c of the component VAG File Handler. Performing a manipulation results in out-of-bounds read. The attack is possible to be carried out remotely. The patch is named 4b8316652a30d40f99ad43310bed273fd1f8a7a3. It is suggested to install a patch to address this issue.

vgmstream

A security vulnerability has been detected in vgmstream up to r2117. Affected is the function decode_ms_ima of the file src/coding/ima_decoder.c of the component Microsoft IMA Decoder. Such manipulation leads to divide by zero. The attack can be executed remotely.