Richard Braun, le Wed 04 Jul 2012 08:22:13 +0200, a écrit : > On Tue, Jul 03, 2012 at 09:13:36PM -0300, Samuel Thibault wrote: > > Richard Braun, le Tue 03 Jul 2012 23:13:26 +0200, a écrit : > > > * ext2fs/inode.c (diskfs_node_iterate): allocate the temporary node > > > table from the heap instead of the stack. > > > > That can be quite a lot for the stack indeed, thanks! > > It's also quite a lot of heap too, and every 5 seconds. We should > consider increasing the interval between global syncs to reduce the > impact of this potentially heavy operation.
It would increase the amount of work you loose on crash. Did you measure the CPU use that you get with a typical filesystem, by e.g. touching a file every 4 seconds to trigger the sync all the time? I guess the real fix would be to be better at knowing which inodes need to be recorded, and filter on that. Samuel
