> Intuition should tell you that 2.6.24 would be slower than the rest. > But the opposite is the case. Benchmarking shows 2.6.24 to be > the fastest among all its predecessors. FWIW, 2.6.24 has been way slower (20-25%, reads and writes) than 2.6.22 (each Debian packaged kernels) for me, at least when testing atop md-raids that (without dm-crypt) are much faster than the CPUs' crypt performance. 2.6.24 is much better when accessing multiple dm-crypted devices at the same time. Where 2.6.22 would effectively only allow one device to be accessed at the same time, with all others blocking for seconds, 2.6.24 reduces this latency to the a-few-100ms order. Still noticeable, but better. C. --------------------------------------------------------------------- dm-crypt mailing list - http://www.saout.de/misc/dm-crypt/ To unsubscribe, e-mail: dm-crypt-unsubscribe@xxxxxxxx For additional commands, e-mail: dm-crypt-help@xxxxxxxx
