This does not buy us new functionnality for now, because we still have to
discover how mach_init (which acts as a name server, enabling processes to
discover each other's ports) is able to receive messages from other processes
(this is a bootstrap problem, and the bootstrap port might be the place to
search).
While we are there:
- removed a lot of debug which is now available using ktrace.
- reworked message handling to avoid mutliple copyin/copyout of the
same data. ktrace of Mach message now uses the in-kernel copy of the
message instead of copying it from userland.
- packed mach trap handlers arguments into a structure to avoid modifying
everything next time we have to add an argument.
and make the sleep length depend on value of variable forkfsleep;
it's set to zero by default (no sleep)
this is a preparation for making the sleep length settable via sysctl
These are of use to userland code which previously depended on the
hard-coded values of LABELSECTOR and LABELOFFSET to figure out the
location of the disklabel for a particular platform.
With the introduction of umbrella ports such as evbarm, evbmips, etc,
the location of the disklabel may vary between kernels for the same
MACHINE. This sysctl will allow userland programs to remain independent
of the particular flavour of MACHINE in such cases.
and seems like generally sensible (more sensible than not doing so), so done
in generic code rather than compat glue only
Change proposed in PR kern/18767 by Emmanuel Dreyfus.
- leave 5 processes for root-only use, the previous value of 1
was unsufficient to execute additional commands once logged, and
perhaps also not enough to actually login remotely with recent (open)sshd
- protect the log of "proc: table full" with ratecheck(), so that
the message is only logged once per 10 seconds; though syslogd normally
doesn't pass the repeated messages through, this avoids flooding
syslogd and potentially also screen/logs
- If the process hits either system limit of number of processes in system,
or user's limit of same, force the process to sleep for 0.5 seconds
before returning failure. This turns 2000 rampaging fork monsters into
2000 harmlessly snoozing fork monsters.
The sleep is intentionally uninterruptible by signals.
These are not intended as ultimate protection agains fork-bombs.
Determined attacker can eat CPU differently than via repeating
fork() calls. But this is good enough to help protect against
programming mistakes or simple-minded tests.
Based on FreeBSD kern_fork.c change in revision 1.132 by
Mike Silbersack <silby at FreeBSD org>
Change also discussed on tech-kern@NetBSD.org, thread
'Fork bomb protection patch'.
until after wakeup event, so we can't clear the SI_COLL flag
in selrecord(). Thus, effectively back rev. 1.57 off.
Problem reported in PR kern/17517 by David Laight, program triggering
the problem is in regress/sys/kern/poll/poll3w.c.
need to reparent the process to initproc, so that child wouldn't
have its p_pptr pointer still pointing on the exited parent
pointed out by Dave Sainty in private mail (the patch in kern/14443
didn't have this bug)
in this case, and even if not, the process would be already woken up by the
wakeup() call
change sent as part of kern/17517 by David Laight
XXX perhaps should KASSERT() sel_pid is zero in the SI_COLL case
the same file multiple times because of recursive loading (ie: libx require
liby and libz and liby require libz, so libz would be loaded twice)
This is probably suboptimal, but it enable /bin/sh to load on the PowerPC,
so it's a good interim solution until we figure precisely how things should
work.
I'm not sure whether this makes the excessive recursive check useless or not.
reparented back to original parent before it's killed.
This makes the original parent aware that the child has exited if
the debugger failed to wait() on the debugged zombie before exiting.
Since we clear tracing flags before killing the child, the reparenting
logic in wait4() wouldn't be triggered, so it's necessary to do it here.
Problem reported and fix provided in kern/14443 by David Sainty.
macho_hdr, argc, *argv, NULL, *envp, NULL, progname, NULL,
*progname, **argv, **envp
Where progname is a pointer to the program name as given in the first
argument to execve(), and macho_hdr a pointer to the Mach-O header at
the beginning of the executable file.
backed by physical pages (ie. because it reused a previously-freed one),
so that we can skip a bunch of useless work in that case.
this fixes the underlying problem behind PR 18543, and also speeds up fork()
quite a bit (eg. 7% on my pc, 1% on my ultra2) when we get a cache hit.
gmcgarry