linux/kernel/fork.c, branch v2.6.24

fix clone(CLONE_NEWPID)

2007-12-05T17:21:18Z

Currently we are complicating the code in copy_process, the clone ABI, and if we fix the bugs sys_setsid itself, with an unnecessary open coded version of sys_setsid. So just simplify everything and don't special case the session and pgrp of the initial process in a pid namespace. Having this special case actually presents to user space the classic linux startup conditions with session == pgrp == 0 for /sbin/init. We already handle sending signals to processes in a child pid namespace. We need to handle sending signals to processes in a parent pid namespace for cases like SIGCHILD and SIGIO. This makes nothing extra visible inside a pid namespace. So this extra special case appears to have no redeeming merits. Further removing this special case increases the flexibility of how we can use pid namespaces, by not requiring the initial process in a pid namespace to be a daemon. Signed-off-by: Eric W. Biederman Cc: Oleg Nesterov Cc: Pavel Emelyanov Cc: Sukadev Bhattiprolu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

sched: fix copy_namespace() <-> sched_fork() dependency in do_fork

2007-11-09T21:39:39Z

Sukadev Bhattiprolu reported a kernel crash with control groups. There are couple of problems discovered by Suka's test: - The test requires the cgroup filesystem to be mounted with atleast the cpu and ns options (i.e both namespace and cpu controllers are active in the same hierarchy). # mkdir /dev/cpuctl # mount -t cgroup -ocpu,ns none cpuctl (or simply) # mount -t cgroup none cpuctl -> Will activate all controllers in same hierarchy. - The test invokes clone() with CLONE_NEWNS set. This causes a a new child to be created, also a new group (do_fork->copy_namespaces->ns_cgroup_clone-> cgroup_clone) and the child is attached to the new group (cgroup_clone-> attach_task->sched_move_task). At this point in time, the child's scheduler related fields are uninitialized (including its on_rq field, which it has inherited from parent). As a result sched_move_task thinks its on runqueue, when it isn't. As a solution to this problem, I moved sched_fork() call, which initializes scheduler related fields on a new task, before copy_namespaces(). I am not sure though whether moving up will cause other side-effects. Do you see any issue? - The second problem exposed by this test is that task_new_fair() assumes that parent and child will be part of the same group (which needn't be as this test shows). As a result, cfs_rq->curr can be NULL for the child. The solution is to test for curr pointer being NULL in task_new_fair(). With the patch below, I could run ns_exec() fine w/o a crash. Reported-by: Sukadev Bhattiprolu Signed-off-by: Srivatsa Vaddagiri Signed-off-by: Ingo Molnar

sched: fix /proc//stat stime/utime monotonicity, part 2

2007-10-29T23:26:32Z

Extend Peter's patch to fix accounting issues, by keeping stime monotonic too. Signed-off-by: Balbir Singh Signed-off-by: Ingo Molnar Tested-by: Frans Pop

sched: keep utime/stime monotonic

2007-10-29T20:18:11Z

keep utime/stime monotonic. cpustats use utime/stime as a ratio against sum_exec_runtime, as a consequence it can happen - when the ratio changes faster than time accumulates - that either can be appear to go backwards. Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar

Uninline fork.c/exit.c

2007-10-19T18:53:56Z

Save ~650 bytes here. add/remove: 4/0 grow/shrink: 0/7 up/down: 430/-1088 (-658) function old new delta __copy_fs_struct - 202 +202 __put_fs_struct - 112 +112 __exit_fs - 58 +58 __exit_files - 58 +58 exit_files 58 2 -56 put_fs_struct 112 5 -107 exit_fs 161 2 -159 sys_unshare 774 590 -184 copy_process 4031 3840 -191 do_exit 1791 1597 -194 copy_fs_struct 202 5 -197 No difference in lmbench lat_proc tests on 2-way Opteron 246. Smaaaal degradation on UP P4 (within errors). [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Alexey Dobriyan Cc: Arjan van de Ven Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

kernel/fork.c: remove unneeded variable initialization in copy_process()

2007-10-19T18:53:56Z

This initialization of is not needed so just remove it. Signed-off-by: Mariusz Kozlowski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

Isolate the explicit usage of signal->pgrp

2007-10-19T18:53:43Z

The pgrp field is not used widely around the kernel so it is now marked as deprecated with appropriate comment. The initialization of INIT_SIGNALS is trimmed because a) they are set to 0 automatically; b) gcc cannot properly initialize two anonymous (the second one is the one with the session) unions. In this particular case to make it compile we'd have to add some field initialized right before the .pgrp. This is the same patch as the 1ec320afdc9552c92191d5f89fcd1ebe588334ca one (from Cedric), but for the pgrp field. Some progress report: We have to deprecate the pid, tgid, session and pgrp fields on struct task_struct and struct signal_struct. The session and pgrp are already deprecated. The tgid value is close to being such - the worst known usage in in fs/locks.c and audit code. The pid field deprecation is mainly blocked by numerous printk-s around the kernel that print the tsk->pid to log. Signed-off-by: Pavel Emelyanov Cc: Oleg Nesterov Cc: Sukadev Bhattiprolu Cc: Cedric Le Goater Cc: Serge Hallyn Cc: "Eric W. Biederman" Cc: Herbert Poetzl Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

Fix tsk->exit_state usage

2007-10-19T18:53:42Z

tsk->exit_state can only be 0, EXIT_ZOMBIE, or EXIT_DEAD. A non-zero test is the same as tsk->exit_state & (EXIT_ZOMBIE | EXIT_DEAD), so just testing tsk->exit_state is sufficient. Signed-off-by: Eugene Teo Cc: Roland McGrath Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

pid namespaces: changes to show virtual ids to user

2007-10-19T18:53:40Z

This is the largest patch in the set. Make all (I hope) the places where the pid is shown to or get from user operate on the virtual pids. The idea is: - all in-kernel data structures must store either struct pid itself or the pid's global nr, obtained with pid_nr() call; - when seeking the task from kernel code with the stored id one should use find_task_by_pid() call that works with global pids; - when showing pid's numerical value to the user the virtual one should be used, but however when one shows task's pid outside this task's namespace the global one is to be used; - when getting the pid from userspace one need to consider this as the virtual one and use appropriate task/pid-searching functions. [akpm@linux-foundation.org: build fix] [akpm@linux-foundation.org: nuther build fix] [akpm@linux-foundation.org: yet nuther build fix] [akpm@linux-foundation.org: remove unneeded casts] Signed-off-by: Pavel Emelyanov Signed-off-by: Alexey Dobriyan Cc: Sukadev Bhattiprolu Cc: Oleg Nesterov Cc: Paul Menage Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

pid namespaces: initialize the namespace's proc_mnt

2007-10-19T18:53:40Z

The namespace's proc_mnt must be kern_mount-ed to make this pointer always valid, independently of whether the user space mounted the proc or not. This solves raced in proc_flush_task, etc. with the proc_mnt switching from NULL to not-NULL. The initialization is done after the init's pid is created and hashed to make proc_get_sb() finr it and get for root inode. Sice the namespace holds the vfsmnt, vfsmnt holds the superblock and the superblock holds the namespace we must explicitly break this circle to destroy all the stuff. This is done after the init of the namespace dies. Running a few steps forward - when init exits it will kill all its children, so no proc_mnt will be needed after its death. Signed-off-by: Pavel Emelyanov Cc: Oleg Nesterov Cc: Sukadev Bhattiprolu Cc: Paul Menage Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds