Go to the first, previous, next, last section, table of contents.

Algorithms

• I understand that a new release of zlib has been made, which corrects a bug that in rare circumstances results in incorrect compression. The same bug probably exists in e2compr, so try to integrate this change into e2compr.
• Make the algorithms into kernel modules. This would reduce physical memory usage, and may have other advantages, e.g. adding a new algorithm without rebooting.
• Build e2decompress with Checker, ElectricFence or something, and try to get it (and hence the kernel, which shares the compression algorithm source code) to handle erroneous data better (i.e. without crashing). This isn't all that important because we store a checksum on the compressed data, which should usually protect us from trying to decompress corrupted data.

Go to the first, previous, next, last section, table of contents.