I needed to reimplement fs/inode.c due to some reasons I will explain later.
Here are the main features of the new implementation:
- dramatically faster
- allows more concurrency
- intended to be MP-safe (not tested)
This is really nice work. Can I make one suggestion though? Could
you use the SLAB allocator to acquire the inodes when more are
necessary for the pool? The incentive to do it this way is two fold:
1) Cache friendly (== faster)
2) Can use constructors to init common inode struct state
---------------------------------------------////
Yow! 11.26 MB/s remote host TCP bandwidth & ////
199 usec remote TCP latency over 100Mb/s ////
ethernet. Beat that! ////
-----------------------------------------////__________ o
David S. Miller, davem@caip.rutgers.edu /_____________/ / // /_/ ><