A GPU kernel can read sensitive data from another GPU kernel (even from another user or app) through an optimized GPU memory region called _local memory_ on various architectures.
The product does not sufficiently track and release allocated memory after it has been used, making the memory unavailable for reallocation and reuse.
Link | Tags |
---|---|
https://registry.khronos.org/OpenCL/specs/3.0-unified/html/OpenCL_API.html#_fundamental_memory_regions | technical description |
https://registry.khronos.org/vulkan/specs/1.3-extensions/html/index.html | vendor advisory |
https://kb.cert.org/vuls/id/446598 | third party advisory us government resource |
https://blog.trailofbits.com | mitigation third party advisory exploit |
https://www.kb.cert.org/vuls/id/446598 | third party advisory us government resource |