tags 596492 fixed-upstream thanks On Sun, Sep 12, 2010 at 3:06 AM, Tanaka Akira <[email protected]> wrote: > Package: manpages-dev > Version: 3.05-1 > > I found that the manual page of getrlimit doesn't describe the unit of > the amount of CPU time for RLIMIT_RTTIME. > > | RLIMIT_RTTIME (Since Linux 2.6.25) > | Specifies a limit on the amount of CPU time that a process > | scheduled under a real-time scheduling policy may consume with- > | out making a blocking system call. For the purpose of this > | limit, each time a process makes a blocking system call, the > | count of its consumed CPU time is reset to zero. The CPU time > | count is not reset if the process continues trying to use the > | CPU but is preempted, its time slice expires, or it calls > | sched_yield(2). > | > | Upon reaching the soft limit, the process is sent a SIGXCPU sig- > | nal. If the process catches or ignores this signal and contin- > | ues consuming CPU time, then SIGXCPU will be generated once each > | second until the hard limit is reached, at which point the pro- > | cess is sent a SIGKILL signal. > | > | The intended use of this limit is to stop a runaway real-time > | process from locking up the system. > > I guess the unit is micro seconds. > http://lkml.org/lkml/2007/12/18/218 > http://linux.derkeiler.com/Mailing-Lists/Kernel/2008-04/msg03922.html
Hello Tanaka Akira, Thanks for this report. You are correct that the units should be microseconds. I have updated the upstream man-pages to mention this. The change will be in the next release, 3.27. Thanks, Michael -- Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ Author of "The Linux Programming Interface"; http://man7.org/tlpi/ -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
