Hi there! The ionice man page [1] says: "Programs inherit the CPU nice setting for io priorities.". However, on my 2.6.24 kernel (Ubuntu Hardy), the cpu nice level does not seem to affect the io priority at all: $ cat /sys/block/sda/queue/scheduler noop anticipatory deadline [cfq] $ ionice none: prio 0 $ nice -19 sleep 20s & [1] 30160 $ ionice -p 30160 none: prio 0 Note that when i put the shell into the best-effort class the effect is the same. The mailing list archive and the kernel documentation [2] has not given an explaination for this behavior. Is the manpage wrong or have i misunderstood it? The other thing that puzzled me about this and that is not explained in the manpage: There is scheduling class "none" and it seems to be the default. The manpage [1] says - in contradiction to this: "Best effort This is the default scheduling class for any process that hasn't asked for a specific io priority." What is the "none" class? Is it higher or lower priority than best effort? Could the answer be included in the manpage? Best regards, Jakob [1] http://git.kernel.org/?p=utils/util-linux-ng/util-linux-ng.git;a=blob;f=schedutils/ionice.1;h=2eca2b8b19a642382af478b261b95ef5f65468d6;hb=HEAD [2] http://www.mjmwired.net/kernel/Documentation/block/ioprio.txt -- To unsubscribe from this list: send the line "unsubscribe util-linux-ng" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
