summaryrefslogtreecommitdiff
path: root/sys/kern/kern_fork.c
AgeCommit message (Collapse)Author
2014-09-08Delete procfs; it's always had races and is now unused: no one noticed forPhilip Guenther
months that I broke it before the 5.5 release. confirmed as not being required by ports by sthen@, ajacoutot@, dcoppa@
2014-07-13KERNEL_ASSERT_LOCKED(9): Assertion for kernel lock (Rev. 3)Masao Uebayashi
This adds a new assertion macro, KERNEL_ASSERT_LOCKED(), to assert that kernel_lock is held. In the long process of removing kernel_lock, there will be a lot (hundreds or thousands) of use of this; virtually almost all functions in !MP-safe subsystems should have this assertion. Thus this assertion should have a short, good name. Not only that "KERNEL_ASSERT_LOCKED" is consistent with other KERNEL_* and SCHED_ASSERT_LOCKED() macros. Input from dlg@ guenther@ kettenis@. OK dlg@ guenther@
2014-07-12add a size argument to free. will be used soon, but for now default to 0.Ted Unangst
after discussions with beck deraadt kettenis.
2014-07-11Use membar_producer() to force visibility instead of misusingPhilip Guenther
atomic_setbits_int() ok kettenis@
2014-07-08decouple struct uvmexp into a new file, so that uvm_extern.h and sysctl.hTheo de Raadt
don't need to be married. ok guenther miod beck jsing kettenis
2014-07-07fork1: FORK_THREAD requires FORK_SHAREFILESMatthew Dempsky
Now that file descriptor tables have moved from a per-thread resource to per-process, it's impossible to fork a thread without sharing the file descriptor table. ok guenther
2014-07-04Track whether a process is a zombie or not yet fully built via flagsPhilip Guenther
PS_{ZOMBIE,EMBRYO} on the process instead of peeking into the process's thread data. This eliminates the need for the thread-level SDEAD state. Change kvm_getprocs() (both the sysctl() and kvm backends) to report the "most active" scheduler state for the process's threads. tweaks kettenis@ feedback and ok matthew@
2014-05-15Move from struct proc to process the reference-count-holding pointersPhilip Guenther
to the process's vmspace and filedescs. struct proc continues to keep copies of the pointers, copying them on fork, clearing them on exit, and (for vmspace) refreshing on exec. Also, make uvm_swapout_threads() thread aware, eliminating p_swtime in kernel. particular testing by ajacoutot@ and sebastia@
2014-05-06Include <sys/vmmeter.h> directly instead of relying on it beingMartin Pieuchot
pulled by <uvm/uvm_extern.h> and turn uvm_total() into a private function. The preferred way to get memory stats is through the VM_UVMEXP sysctl(3) since VM_METER is just a wrapper on top of it. In the kernel, use `uvmexp' directly instead of uvm_total(). This change does not remove <sys/vmmeter.h> from <uvm/uvm_extern.h> to give some more time to port maintainers to fix their ports. ok guenther@ as part of a larger diff.
2014-05-04Add PS_SYSTEM, the process-level mirror of the thread-level P_SYSTEM,Philip Guenther
and FORK_SYSTEM as a flag to set them. This eliminates needing to peek into other processes threads in various places. Inspired by NetBSD ok miod@ matthew@
2014-05-03Move the u-area allocation and pmap-magic logic to its own functionPhilip Guenther
uvm_uarea_alloc() function name from NetBSD; arm testing by miod@
2014-04-18Have each thread keeps its own (counted!) reference to the process's ucredsPhilip Guenther
to avoid possible use-after-free references when swapping ids in threaded processes. "Do I have the right creds?" checks are always made with the threads creds. Inspired by FreeBSD and NetBSD "right time" deraadt@
2014-03-30Eliminates struct pcred by moving the real and saved ugids intoPhilip Guenther
struct ucred; struct process then directly links to the ucred Based on a discussion at c2k10 or so before noting that FreeBSD and NetBSD did this too. ok matthew@
2014-03-28Reduce uvm include madness. Use <uvm/uvm_extern.h> instead ofMartin Pieuchot
<uvm/uvm.h> if possible and remove double inclusions. ok beck@, mlarkin@, deraadt@
2014-03-26Move p_emul and p_sigcode from proc to process.Philip Guenther
Tweak the handling of ktrace EMUL when changing ktracing: only generate one per process (not one per thread) and pass the correct proc pointer down to the VFS layer. Permit generating of NAMI and CSW records inside ktrace(2) itself. ok deraadt@ millert@
2014-03-22Move p_sigacts from struct proc to struct process.Philip Guenther
testing help mpi@
2014-02-12Eliminate the exit sig handling, which was only invokable via thePhilip Guenther
Linux-compat clone() syscall when *not* using CLONE_THREAD. pirofti@ confirms Opera runs in compat without this, so out it goes; one less hair to choke on in kern_exit.c ok tedu@ pirofti@
2014-02-10arc4random_uniform() returns a value strictly less than its argument; fixPhilip Guenther
arithmetic so that PID_MAX can be reached. ok otto@ zhuk@ miod@
2014-01-20Threads can't be zombies, only processes, so change zombproc to zombprocess,Philip Guenther
make it a list of processes, and change P_NOZOMBIE and P_STOPPED from thread flags to process flags. Add allprocess list for the code that just wants to see processes. ok tedu@
2014-01-20Move p_textvp from struct proc to struct process so that the exit codePhilip Guenther
can be further simplified. ok kettenis@
2013-10-08Fix delivery of SIGPROF and SIGVTALRM to threaded processes by havingPhilip Guenther
hardclock() set a flag on the running thread and force AST processing, and then have the thread signal itself from userret(). idea and flag names from FreeBSD ok jsing@
2013-08-14The last user of the old __tfork() was updated to the current one,Philip Guenther
so COMPAT_O51 can go. The complete ABI role means COMPAT_O53 can be removed as well. ok jsing@ tedu@
2013-06-11convert some easy bcopy to memcpy and clean up fdexpand a bit.Ted Unangst
ok kettenis
2013-06-06Prevent idle thread from being stolen on startup.Christiano F. Haesbaert
There is a race condition which might trigger a case where two cpus try to run the same idle thread. The problem arises when one cpu steals the idle proc of another cpu and this other cpu ends up running the idle thread via spc->spc_idleproc, resulting in two cpus trying to cpu_switchto(idleX). On startup, idle procs are scaterred around different runqueues, the decision for scheduling is: 1 look at my runqueue. 2 if empty, look at other dudes runqueue. 3 if empty, select idle proc via spc->spc_idleproc. The problem is that cpu0's idle0 might be running on cpu1 due to step 1 or 2 and cpu0 hits step 3. So cpu0 will select idle0, while cpu1 is in fact running it already. The solution is to never place idle on a runqueue, therefore being only selectable through spc->spc_idleproc. This race can be more easily triggered on a HT cpu on virtualized environments, where the guest more often than not doesn't have the cpu for itself, so timing gets shuffled. ok tedu@ guenther@ go ahead after t2k13 deraadt@
2013-06-05factor out pid allocation to functions. add a small cache of recentlyTed Unangst
exited pids that won't get recycled. ok deraadt
2013-06-03When creating a thread, don't add it to the process's thread listPhilip Guenther
until it's fully built, so that it can't get a signal from realitexpire(), as seen by sthen@ and espie@ in ports builds. Exact bits moved down worked out with tedu@, ok deraadt@
2013-06-03Convert some internal APIs to use timespecs instead of timevalsPhilip Guenther
ok matthew@ deraadt@
2013-06-01As found by kurt, there's a twisty race between exit1 and fork1Ted Unangst
with threaded processes. Fix this by checking for an attempt to go single threaded in fork1 and account for the new thread as well. ok espie guenther kurt
2013-04-06rthreads are always enabled. remove the sysctl.Ted Unangst
ok deraadt guenther kettenis matthew
2013-03-14the 5.1 era tfork syscall claws its way out of the grave. we failed toTed Unangst
fully deprecate it (notably the go port was still using it as of 5.3) so to give users a little more time to update, __tfork51 lives again. okish deraadt guenther
2013-03-02No longer need the 5.1 version of the __tfork syscallPhilip Guenther
ok deraadt@
2012-11-19If uvm_km_kmemalloc_pla() fails when just creating a thread (and not aPhilip Guenthe
process), then don't decrement the total and per-user counts of processes. ok deraadt@ miod@
2012-08-02Apply profiling to all threads instead of just the thread that calledPhilip Guenthe
profil() by moving P_PROFIL from proc->p_flag to process->ps_flags with matching adjustment in fork1() and exit1() ok matthew@
2012-06-21__tfork() needs to set the stack address of the new thread in the kernel,Philip Guenthe
so that it can't get a signal while still running on the parent thread's stack. Also, pass in sizeof(struct __tfork) to provide forward compat when more members are added. This is an ABI change, so switch syscall numbers and bump lib majors this time. ok deraadt@ matthew@
2012-05-10Only set a process's start time when starting the main thread. There'sPhilip Guenthe
also no need to protect that and the setting of the AFORK accounting flag with the scheduler lock. ok mikeb@
2012-04-13First stab at making ptrace(2) usable for debugging multi-threaded programs.Mark Kettenis
It implements a full-stop model where all threads are stopped before handing over control to the debugger. Events are reported as before through wait(2); you will have to call ptrace(PT_GET_PROCESS_STATE, ...) to find out which thread hit the event. Since this changes the size of struct ptrace_state, you will have to recompile gdb. ok guenther@
2012-04-12remove rfork(); ok guenther miodTheo de Raadt
2012-04-12move accounting flags to struct process; idea and ok guentherMike Belopuhov
2012-04-10Make the KERN_NPROCS and KERN_MAXPROC sysctl()s and the RLIMIT_NPROC rlimitPhilip Guenthe
count processes instead of threads. New sysctl()s KERN_NTHREADS and KERN_MAXTHREAD count and limit threads. The nprocs and maxproc kernel variables are replaced by nprocess, maxprocess, nthreads, and maxthread. ok tedu@ mikeb@
2012-03-23Make rusage totals, itimers, and profile settings per-process insteadPhilip Guenthe
of per-rthread. Handling of per-thread tick and runtime counters inspired by how FreeBSD does it. ok kettenis@
2012-02-20First steps for making ptrace work with rthreads:Philip Guenthe
- move the P_TRACED and P_INEXEC flags, and p_oppid, p_ptmask, and p_ptstat member from struct proc to struct process - sort the PT_* requests into those that take a PID vs those that can also take a TID - stub in PT_GET_THREAD_FIRST and PT_GET_THREAD_NEXT ok kettenis@
2011-12-14Handle rthreads consistently in ktrace by moving the flags and vnode intoPhilip Guenthe
struct process; KTRFAC_ACTIVE becomes P_INKTR. Also, save the credentials used to open the file in sys_ktrace() and use them for all writes to the vnode. much feedback and ok jsing@
2011-11-22Move struct proc's sigaltstack struct from the zeroed area into theJoshua Elsasser
copied area, and initialize it properly in the FORK_THREAD case. This restores the behavior of a forked process inheriting its parent's signal stack. ok guenther@
2011-11-09Change fork1() and kthread_create() to match the rest of the treePhilip Guenthe
and use curp vs p instead of p1 vs p2. Add curpr and pr variables for the respective struct processes. Make sigactsshare() return the shared sigacts intead of taking the struct proc to update. ok deraadt@
2011-11-05I had moved earlier the adding of processes to the pgrp and children listsPhilip Guenthe
during the big rework at c2k10, but it's too early as signals can be posted before the process is fully built. Move those list adds back down to the late stage they were before. Problem seen on sebastia@'s sparc. ok deraadt@ miod@
2011-10-15"TLS-lite": add kernel support for a per-thread userspace pointer,Philip Guenthe
for pointing to the thread-control-block. Support for mapping this to the correct hardware register can be added as it's finished; start with support for amd64, sparc, and sparc64. Includes syscalls for getting and setting it (for a portable __errno implementation) as well as creating a new thread with an initial value for it. discussed with miod@, kettenis@, deraadt@; committing to get the syscalls in with the impending libc bump and do further refinements in tree
2011-07-07Functions used in files other than where they are defined should bePhilip Guenthe
declared in .h files, not in each .c. Apply that rule to endtsleep(), scheduler_start(), updatepri(), and realitexpire() ok deraadt@ tedu@
2011-07-06Clean up after P_BIGLOCK removal.Artur Grabowski
KERNEL_PROC_LOCK -> KERNEL_LOCK KERNEL_PROC_UNLOCK -> KERNEL_UNLOCK oga@ ok
2011-06-06push kernel malloc(9) and kernel stacks into non-dma memory, since thatTheo de Raadt
appears to be safe now. If not, we'll know soon where the bugs lie, so that we can fix them. This diff has been in snapshots for many months. ok oga miod
2011-04-03Move PPWAIT flag from struct proc to process, so that rthreads inPhilip Guenthe
a vforked child behave correctly. Have the parent in a vfork() wait on a (different) flag in *its* process instead of the child to prevent a possible use-after-free. When ktracing the child return from a fork, call it rfork if an rthread was created. ok blambert@
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-05 07:28:09 +0000