Another possible improvement to ext
Posted Jun 15, 2006 6:48 UTC (Thu) by barryn
Parent article: Time for ext4?
This actually raises the possibility, combined with another patch, of taking ext2 and improving it so that, after a crash, only part of the FS needs to be fscked (i.e. decrease the time spent doing fsck without having to resort to journaling).
Just mentioning this so that it doesn't get completely buried by the discussion of extents and 48-bit block addresses...
to post comments)