1.7 KiB
CVE-2023-52497
Description
In the Linux kernel, the following vulnerability has been resolved:erofs: fix lz4 inplace decompressionCurrently EROFS can map another compressed buffer for inplacedecompression, that was used to handle the cases that some pages ofcompressed data are actually not in-place I/O.However, like most simple LZ77 algorithms, LZ4 expects the compresseddata is arranged at the end of the decompressed buffer and itexplicitly uses memmove() to handle overlapping: __________________________________________________________ |_ direction of decompression --> ____ |_ compressed data _|Although EROFS arranges compressed data like this, it typically maps twoindividual virtual buffers so the relative order is uncertain.Previously, it was hardly observed since LZ4 only uses memmove() forshort overlapped literals and x86/arm64 memmove implementations seem tocompletely cover it up and they don't have this issue. Juhyung reportedthat EROFS data corruption can be found on a new Intel x86 processor.After some analysis, it seems that recent x86 processors with the newFSRM feature expose this issue with "rep movsb".Let's strictly use the decompressed buffer for lz4 inplacedecompression for now. Later, as an useful improvement, we could tryto tie up these two buffers together in the correct order.
POC
Reference
No PoCs from references.