Hi, getrlimit(2) says:
> Because this information is stored in the per-process information, this > system call must be executed directly by the shell if it is to affect all > future processes created by the shell; limit is thus a built-in command > to csh(1) and ulimit is the sh(1) equivalent. sh(1) hasn't mentioned ulimit since jmc@ rewrote it for revision 1.101. POSIX.1-2008 says ulimit is an XSI option here: http://pubs.opengroup.org/onlinepubs/9699919799/utilities/ulimit.html sh(1) talks about how it is mostly POSIX.1-2008 compliant in its STANDARDS section. You can use ulimit from /bin/sh, with or without POSIX mode. ksh(1) defines ulimit in all its splendor, with its considerable capabilities (relative to the POSIX.1-2008 spec), but does not mention anything about POSIX, XSI, etc. I think the simplest correction is to just change getrlimit(2) to say "ksh" instead of "sh" (trivial patch below), so that the documentation in the system itself is consistent... but elsewhere in the manpages XSI extensions are marked and such, so I don't know what else, if anything, should be changed. Something just seems off, hence bugs@. Thoughts? -- Scott Cheloha Index: getrlimit.2 =================================================================== RCS file: /cvs/src/lib/libc/sys/getrlimit.2,v retrieving revision 1.27 diff -u -p -r1.27 getrlimit.2 --- getrlimit.2 7 Oct 2016 15:48:55 -0000 1.27 +++ getrlimit.2 15 Mar 2017 23:00:25 -0000 @@ -148,7 +148,7 @@ is thus a built-in command to and .Ic ulimit is the -.Xr sh 1 +.Xr ksh 1 equivalent. .Pp The system refuses to extend the data or stack space when the limits @@ -206,7 +206,7 @@ and the caller is not the superuser. .El .Sh SEE ALSO .Xr csh 1 , -.Xr sh 1 , +.Xr ksh 1 , .Xr quotactl 2 , .Xr sigaction 2 , .Xr sigaltstack 2 ,
