count is 0, wait for use count to drain before finishing the close.
This is necessary in order for multiple processes to safely share file
descriptor tables.
rather than just "value / sizeof (char *)". While in these cases the latter
works as well, the former's a better example for other uses to copy (since
it handles the non-integral multiple case correctly).
1. the mask is passed by value, not by reference
2. compare the rest of the mask bits not the valid ones
3. return 0, not the current fd's flags.
Now appletviewer works on an inet display. It will not work
with :0 because /tmp/.X11-pipe/X0 is a named pipe, not a
socket like our /tmp/.X11-unix/X0. Maybe I'll kludge svr4_sys_open
to return an fd to a unix socket when someone attempts to call
open on a unix socket.
version:
1. The stack_t returned before was SS_DISABLED; now we enable the stack
and return a 16K stack which is 16K under the current stack pointer.
This seems to be what solaris does.
2. Preserve uc_link.
would appear that overloading the argument this way is used by SVR4 to `fall
off' the main context (uc_link being NULL) after a function set up using
makecontext() returns.
RLIM_INFINITY like we do. Java uses rlim_max to allocate a data structure
per file descriptor.
what we do is bogus anyway, because:
setrlimit(RLIM_NOFILE, {RLIM_INFINITY, RLIM_INFINITY}}
sets the limit to { maxfiles, maxfiles }
I undestand that this is so that if we change maxfiles via sysctl, processes
can still use the new maximum, but it is probably more appropriate to return
the current value of maxfiles (and maxproc) at every getrlimit call, even
if the current value for either is RLIM_INFINITY.
Anyway, java now works on the the sparc, on the i386 loses on setcontext().
sparc, and unknown for the rest. Also make sure that the string is terminated
properly and the length is returned as the specification calls. Maybe sometime
we can read the prom on the sparc and return the actual model.
pk