- 29 Sep, 2006 40 commits
-
-
Andrew Morton authored
This appears to be a verbatim copy-n-paste of the GPL. Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Paul Jackson authored
The cpuset code handling hot unplug of CPUs or Memory Nodes was incorrect - it could remove a CPU or Node from the top cpuset, while leaving it still in some child cpusets. One basic rule of cpusets is that each cpusets cpus and mems are subsets of its parents. The cpuset hot unplug code violated this rule. So the cpuset hotunplug handler must walk down the tree, removing any removed CPU or Node from all cpusets. However, it is not allowed to make a cpusets cpus or mems become empty. They can only transition from empty to non-empty, not back. So if the last CPU or Node would be removed from a cpuset by the above walk, we scan back up the cpuset hierarchy, finding the nearest ancestor that still has something online, and copy its CPU or Memory placement. Signed-off-by:
Paul Jackson <pj@sgi.com> Cc: Nathan Lynch <ntl@pobox.com> Cc: Anton Blanchard <anton@samba.org> Signed-off-by:
-
Paul Jackson authored
Change the list of memory nodes allowed to tasks in the top (root) nodeset to dynamically track what cpus are online, using a call to a cpuset hook from the memory hotplug code. Make this top cpus file read-only. On systems that have cpusets configured in their kernel, but that aren't actively using cpusets (for some distros, this covers the majority of systems) all tasks end up in the top cpuset. If that system does support memory hotplug, then these tasks cannot make use of memory nodes that are added after system boot, because the memory nodes are not allowed in the top cpuset. This is a surprising regression over earlier kernels that didn't have cpusets enabled. One key motivation for this change is to remain consistent with the behaviour for the top_cpuset's 'cpus', which is also read-only, and which automatically tracks the cpu_online_map. This change also has the minor benefit that it fixes a long standing, little noticed, minor bug in cpusets. The cpuset performance tweak to short circuit the cpuset_zone_allowed() check on systems with just a single cpuset (see 'number_of_cpusets', in linux/cpuset.h) meant that simply changing the 'mems' of the top_cpuset had no affect, even though the change (the write system call) appeared to succeed. With the following change, that write to the 'mems' file fails -EACCES, and the 'mems' file stubbornly refuses to be changed via user space writes. Thus no one should be mislead into thinking they've changed the top_cpusets's 'mems' when in affect they haven't. In order to keep the behaviour of cpusets consistent between systems actively making use of them and systems not using them, this patch changes the behaviour of the 'mems' file in the top (root) cpuset, making it read only, and making it automatically track the value of node_online_map. Thus tasks in the top cpuset will have automatic use of hot plugged memory nodes allowed by their cpuset. [akpm@osdl.org: build fix] [bunk@stusta.de: build fix] Signed-off-by:
Adrian Bunk <bunk@stusta.de> Signed-off-by:
-
Pavel Machek authored
Fix simple typo in RTC_HCTOSYS option. Signed-off-by:
Pavel Machek <pavel@suse.cz> Signed-off-by:
-
Nick Piggin authored
A previous patch to allow an exiting task to OOM kill itself (and thereby avoid a little deadlock) introduced a problem. We don't want the PF_EXITING task, even if it is 'current', to access mem reserves if there is already a TIF_MEMDIE process in the system sucking up reserves. Also make the commenting a little bit clearer, and note that our current scheme of effectively single threading the OOM killer is not itself perfect. Signed-off-by:
Nick Piggin <npiggin@suse.de> Signed-off-by:
-
Oleg Nesterov authored
- It is not possible to have task->mm == &init_mm. - task_lock() buys nothing for 'if (!p->mm)' check. Signed-off-by:
Oleg Nesterov <oleg@tv-sign.ru> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by:
-
Oleg Nesterov authored
No logic changes, but imho easier to read. Signed-off-by:
-
Oleg Nesterov authored
The only one usage of TASK_DEAD outside of last schedule path, select_bad_process: for_each_task(p) { if (!p->mm) continue; ... if (p->state == TASK_DEAD) continue; ... TASK_DEAD state is set at the end of do_exit(), this means that p->mm was already set == NULL by exit_mm(), so this task was already rejected by 'if (!p->mm)' above. Note also that the caller holds tasklist_lock, this means that p can't pass exit_notify() and then set TASK_DEAD when p->mm != NULL. Also, remove open-coded is_init(). Signed-off-by: -
Oleg Nesterov authored
I am not sure about this patch, I am asking Ingo to take a decision. task_struct->state == EXIT_DEAD is a very special case, to avoid a confusion it makes sense to introduce a new state, TASK_DEAD, while EXIT_DEAD should live only in ->exit_state as documented in sched.h. Note that this state is not visible to user-space, get_task_state() masks off unsuitable states. Signed-off-by:
-
Oleg Nesterov authored
After the previous change (->flags & PF_DEAD) <=> (->state == EXIT_DEAD), we don't need PF_DEAD any longer. Signed-off-by:
-
Oleg Nesterov authored
schedule() checks PF_DEAD on every context switch and sets ->state = EXIT_DEAD to ensure that the exiting task will be deactivated. Note that this EXIT_DEAD is in fact a "random" value, we can use any bit except normal TASK_XXX values. It is better to set this state in do_exit() along with PF_DEAD flag and remove that check in schedule(). We are safe wrt concurrent try_to_wake_up() (for example ptrace, tkill), it can not change task's ->state: the 'state' argument of try_to_wake_up() can't have EXIT_DEAD bit. And in case when try_to_wake_up() sees a stale value of ->state == TASK_RUNNING it will do nothing. Signed-off-by:
-
Arjan van de Ven authored
Introduce the disable_irq_nosync_lockdep_irqsave() and enable_irq_lockdep_irqrestore() APIs. These are needed for NE2000; basically NE2000 calls disable_irq and enable_irq as locking against the IRQ handler, but both in cases where interrupts are on and off. This means that lockdep needs to track the old state of the virtual irq flags on disable_irq, and restore these at enable_irq time. Signed-off-by:
Arjan van de Ven <arjan@linux.intel.com> Signed-off-by:
Ingo Molnar <mingo@elte.hu> Cc: Jeff Garzik <jeff@garzik.org> Signed-off-by:
-
Alexey Dobriyan authored
Everyone passes valid pointer there. Signed-off-by:
Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by:
-
Alexey Dobriyan authored
If register_filesystem() fails mux workqueue must be killed. Signed-off-by:
-
Alexey Dobriyan authored
It always returns 0, so relying on it is useless. The only caller isn't checking return value. In general, un-, de-, -free functions should return void. Signed-off-by:
-
Alexey Dobriyan authored
If register_filesystem() fails, vxfs_inode cache must be destroyed. Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Alexey Dobriyan authored
Two lines -- two bugs. :-( Signed-off-by:
-
Josh Triplett authored
Currently, __acquire and __release take a lock expression, but __cond_lock takes only a condition, not the lock acquired if the expression evaluates to true. Change __cond_lock to accept a lock expression, and change all the callers to pass in a lock expression. Signed-off-by:
Josh Triplett <josh@freedesktop.org> Signed-off-by:
-
Randy Dunlap authored
Fix "quiet" parameter doc. No trailing '=' sign, no value after it. And it disables "most" kernel messages, not all of them. Signed-off-by:
Randy Dunlap <rdunlap@xenotime.net> Signed-off-by:
-
Frederik Deweerdt authored
At the beginning of the routine, "copied" is set to 0, but it is no good because in lines 805 and 812 it is set to other values. Finally, the routine returns as if it copied 12 (=ENOMEM) bytes less than it actually did. Signed-off-by:
Frederik Deweerdt <frederik.deweerdt@gmail.com> Acked-by:
Eric Biederman <ebiederm@xmission.com> Signed-off-by:
-
Jason Baron authored
In the case below we are locking the whole disk not a partition. This change simply brings the code in line with the piece above where when we are the 'first' opener, and we are a partition. Signed-off-by:
Jason Baron <jbaron@redhat.com> Acked-by:
Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by:
-
Josh Triplett authored
spin_trylock_irq and spin_trylock_irqsave use _spin_trylock, which does not use the __cond_lock wrapper annotation and thus does not affect the lock context; change them to use spin_trylock instead, which does use __cond_lock. Signed-off-by:
-
Josh Triplett authored
The lock annotations used on spinlocks and rwlocks currently use __{acquires,releases}(spinlock_t) and __{acquires,releases}(rwlock_t), respectively. This loses the information of which lock actually got acquired or released, and assumes a different type for the parameter of __acquires and __releases than the rest of the kernel. While the current implementations of __acquires and __releases throw away their argument, this will not always remain the case. Change this to use the lock parameter instead, to preserve this information and increase consistency in usage of __acquires and __releases. Signed-off-by: -
Alexey Dobriyan authored
Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Alan Cox authored
If your driver implements "break on" and "break off" this ensures you won't get multiple overlapping requests or requests in parallel. If your driver has its own break handling then its still your problem as the driver author. Break is also now serialized against writes from user space properly but no new guarantees are made driver level about writes from the line discipline itself (eg flow control or echo) Signed-off-by:
Alan Cox <alan@redhat.com> Signed-off-by:
-
Alan Cox authored
[akpm@osdl.org: build fix] [akpm@osdl.org: warning fix] Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
-
Henrik Kretzschmar authored
Adds a missing exit, if the file that should be parsed couldn't be opened. Without it crashes with a segfault, cause the filedescriptor is accessed even if the file could not be opened. Signed-off-by:
Henrik Kretzschmar <henne@nachtwindheim.de> Acked-by:
-
Oleg Nesterov authored
use rcu locks for find_task_by_pid(). Signed-off-by:
-
Oleg Nesterov authored
It is ok to do find_task_by_pid() + get_task_struct() under rcu_read_lock(), we cand drop tasklist_lock. Note that testing of ->exit_state is racy with or without tasklist anyway. Signed-off-by:
-
Ian Kent authored
During testing I've found that the mount pending flag can be left set at exit from autofs4_lookup after a failed mount request. This shouldn't be allowed to happen and causes incorrect error returns. Signed-off-by:
Ian Kent <raven@themaw.net> Signed-off-by:
-
Ian Kent authored
The check for an empty directory in the autofs4_follow_link method fails occassionally due to old dentrys. We had the same problem autofs4_revalidate ages ago. I thought we wouldn't need this in autofs4_follow_link, silly me. Signed-off-by:
-
Oleg Nesterov authored
copy_process: // holds tasklist_lock + ->siglock /* * inherit ioprio */ p->ioprio = current->ioprio; Why? ->ioprio was already copied in dup_task_struct(). I guess this is needed to ensure that the child can't escape sys_ioprio_set(IOPRIO_WHO_{PGRP,USER}), yes? In that case we don't need ->siglock held, and the comment should be updated. Signed-off-by: -
Oleg Nesterov authored
Remove open-coded has_rt_policy(), no changes in kernel/exit.o Signed-off-by:
-
Oleg Nesterov authored
I am not sure this patch is correct: I can't understand what the current code does, and I don't know what it was supposed to do. The comment says: * can't change policy, except between SCHED_NORMAL * and SCHED_BATCH: The code: if (((policy != SCHED_NORMAL && p->policy != SCHED_BATCH) && (policy != SCHED_BATCH && p->policy != SCHED_NORMAL)) && But this is equivalent to: if ( (is_rt_policy(policy) && has_rt_policy(p)) && which means something different. We can't _decrease_ the current ->rt_priority with such a check (if rlim[RLIMIT_RTPRIO] == 0). Probably, it was supposed to be: if ( !(policy == SCHED_NORMAL && p->policy == SCHED_BATCH) && !(policy == SCHED_BATCH && p->policy == SCHED_NORMAL) this matches the comment, but strange: it doesn't allow to _drop_ the realtime priority when rlim[RLIMIT_RTPRIO] == 0. I think the right check would be: /* can't set/change rt policy */ if (is_rt_policy(policy) && policy != p->policy && !rlim_rtprio) return -EPERM; Signed-off-by:
-
Oleg Nesterov authored
Imho, makes the code a bit easier to read. Signed-off-by:
-
Oleg Nesterov authored
Use rcu locks instead. sched_setscheduler() now takes ->siglock before reading ->signal->rlim[]. Signed-off-by:
-
Cal Peake authored
Get rid of an extraneous printk in kernel_restart(). Signed-off-by:
Cal Peake <cp@absolutedigital.net> Acked-by:
-
Björn Steinbrink authored
If ____call_usermodehelper fails, we're not interested in the child process' exit value, but the real error, so let's stop wait_for_helper from overwriting it in that case. Issue discovered by Benedikt Böhm while working on a Linux-VServer usermode helper. Signed-off-by:
Björn Steinbrink <B.Steinbrink@gmx.de> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by:
-
Rolf Eike Beer authored
Currently this module just returns 1 if anything on module init fails. Store the error code of the different function calls and return their error on problems. Signed-off-by:
Rolf Eike Beer <eike-kernel@sf-tec.de> Cc: Greg KH <greg@kroah.com> Signed-off-by:
-
Roland McGrath authored
It is sure confusing that linux/ptrace.h has: #define PTRACE_SINGLESTEP 9 #define PTRACE_ATTACH 0x10 #define PTRACE_DETACH 0x11 #define PTRACE_SYSCALL 24 All the low-numbered constants are in decimal, but the last two in hex. It sure makes it likely that someone will look at this and think that 9, 10, 11 are used, and that 16 and 17 are not used. How about we use the same notation for all the numbers [0,24] in the same short list? Signed-off-by:
Roland McGrath <roland@redhat.com> Signed-off-by:
-
