On an SSD used by a typical chromebook (i.e. the one on my desk right now), it takes about 300us to read a random 4k page, but it takes less than 10us to lzo-decompress a page from the zram device. Code compresses reasonably well (down to almost 50% for x86_64, although only 66% for ARM32), so I may be better off swapping file pages to zram, rather than reading them back from the SSD. Before I even get started trying to do this, can anybody tell me if this is a stupid idea? Or possibly a good idea, but totally impractical from an implementation perspective? Thanks! -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
