linux/kernel/fork.c, branch v3.12

Merge git://git.kvack.org/~bcrl/aio-next

2013-09-13T17:55:58Z

Pull aio changes from Ben LaHaise: "First off, sorry for this pull request being late in the merge window. Al had raised a couple of concerns about 2 items in the series below. I addressed the first issue (the race introduced by Gu's use of mm_populate()), but he has not provided any further details on how he wants to rework the anon_inode.c changes (which were sent out months ago but have yet to be commented on). The bulk of the changes have been sitting in the -next tree for a few months, with all the issues raised being addressed" * git://git.kvack.org/~bcrl/aio-next: (22 commits) aio: rcu_read_lock protection for new rcu_dereference calls aio: fix race in ring buffer page lookup introduced by page migration support aio: fix rcu sparse warnings introduced by ioctx table lookup patch aio: remove unnecessary debugging from aio_free_ring() aio: table lookup: verify ctx pointer staging/lustre: kiocb->ki_left is removed aio: fix error handling and rcu usage in "convert the ioctx list to table lookup v3" aio: be defensive to ensure request batching is non-zero instead of BUG_ON() aio: convert the ioctx list to table lookup v3 aio: double aio_max_nr in calculations aio: Kill ki_dtor aio: Kill ki_users aio: Kill unneeded kiocb members aio: Kill aio_rw_vect_retry() aio: Don't use ctx->tail unnecessarily aio: io_cancel() no longer returns the io_event aio: percpu ioctx refcount aio: percpu reqs_available aio: reqs_active -> reqs_available aio: fix build when migration is disabled ...

mm: mempolicy: turn vma_set_policy() into vma_dup_policy()

2013-09-11T22:57:00Z

Simple cleanup. Every user of vma_set_policy() does the same work, this looks a bit annoying imho. And the new trivial helper which does mpol_dup() + vma_set_policy() to simplify the callers. Signed-off-by: Oleg Nesterov Cc: KOSAKI Motohiro Cc: Mel Gorman Cc: Rik van Riel Cc: Andi Kleen Cc: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

fork: unify and tighten up CLONE_NEWUSER/CLONE_NEWPID checks

2013-09-11T22:56:20Z

do_fork() denies CLONE_THREAD | CLONE_PARENT if NEWUSER | NEWPID. Then later copy_process() denies CLONE_SIGHAND if the new process will be in a different pid namespace (task_active_pid_ns() doesn't match current->nsproxy->pid_ns). This looks confusing and inconsistent. CLONE_NEWPID is very similar to the case when ->pid_ns was already unshared, we want the same restrictions so copy_process() should also nack CLONE_PARENT. And it would be better to deny CLONE_NEWUSER && CLONE_SIGHAND as well just for consistency. Kill the "CLONE_NEWUSER | CLONE_NEWPID" check in do_fork() and change copy_process() to do the same check along with ->pid_ns check we already have. Signed-off-by: Oleg Nesterov Acked-by: Andy Lutomirski Cc: "Eric W. Biederman" Cc: Colin Walters Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

pidns: kill the unnecessary CLONE_NEWPID in copy_process()

2013-09-11T22:56:19Z

Commit 8382fcac1b81 ("pidns: Outlaw thread creation after unshare(CLONE_NEWPID)") nacks CLONE_NEWPID if the forking process unshared pid_ns. This is correct but unnecessary, copy_pid_ns() does the same check. Remove the CLONE_NEWPID check to cleanup the code and prepare for the next change. Test-case: static int child(void *arg) { return 0; } static char stack[16 * 1024]; int main(void) { pid_t pid; assert(unshare(CLONE_NEWUSER | CLONE_NEWPID) == 0); pid = clone(child, stack + sizeof(stack) / 2, CLONE_NEWPID | SIGCHLD, NULL); assert(pid < 0 && errno == EINVAL); return 0; } clone(CLONE_NEWPID) correctly fails with or without this change. Signed-off-by: Oleg Nesterov Acked-by: Andy Lutomirski Cc: "Eric W. Biederman" Cc: Colin Walters Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

pidns: fix vfork() after unshare(CLONE_NEWPID)

2013-09-11T22:56:19Z

Commit 8382fcac1b81 ("pidns: Outlaw thread creation after unshare(CLONE_NEWPID)") nacks CLONE_VM if the forking process unshared pid_ns, this obviously breaks vfork: int main(void) { assert(unshare(CLONE_NEWUSER | CLONE_NEWPID) == 0); assert(vfork() >= 0); _exit(0); return 0; } fails without this patch. Change this check to use CLONE_SIGHAND instead. This also forbids CLONE_THREAD automatically, and this is what the comment implies. We could probably even drop CLONE_SIGHAND and use CLONE_THREAD, but it would be safer to not do this. The current check denies CLONE_SIGHAND implicitely and there is no reason to change this. Eric said "CLONE_SIGHAND is fine. CLONE_THREAD would be even better. Having shared signal handling between two different pid namespaces is the case that we are fundamentally guarding against." Signed-off-by: Oleg Nesterov Reported-by: Colin Walters Acked-by: Andy Lutomirski Reviewed-by: "Eric W. Biederman" Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace

2013-09-07T21:35:32Z

Pull namespace changes from Eric Biederman: "This is an assorted mishmash of small cleanups, enhancements and bug fixes. The major theme is user namespace mount restrictions. nsown_capable is killed as it encourages not thinking about details that need to be considered. A very hard to hit pid namespace exiting bug was finally tracked and fixed. A couple of cleanups to the basic namespace infrastructure. Finally there is an enhancement that makes per user namespace capabilities usable as capabilities, and an enhancement that allows the per userns root to nice other processes in the user namespace" * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace: userns: Kill nsown_capable it makes the wrong thing easy capabilities: allow nice if we are privileged pidns: Don't have unshare(CLONE_NEWPID) imply CLONE_THREAD userns: Allow PR_CAPBSET_DROP in a user namespace. namespaces: Simplify copy_namespaces so it is clear what is going on. pidns: Fix hang in zap_pid_ns_processes by sending a potentially extra wakeup sysfs: Restrict mounting sysfs userns: Better restrictions on when proc and sysfs can be mounted vfs: Don't copy mount bind mounts of /proc//ns/mnt between namespaces kernel/nsproxy.c: Improving a snippet of code. proc: Restrict mounting the proc filesystem vfs: Lock in place mounts from more privileged users

pidns: Don't have unshare(CLONE_NEWPID) imply CLONE_THREAD

2013-08-31T06:44:00Z

I goofed when I made unshare(CLONE_NEWPID) only work in a single-threaded process. There is no need for that requirement and in fact I analyzied things right for setns. The hard requirement is for tasks that share a VM to all be in the pid namespace and we properly prevent that in do_fork. Just to be certain I took a look through do_wait and forget_original_parent and there are no cases that make it any harder for children to be in the multiple pid namespaces than it is for children to be in the same pid namespace. I also performed a check to see if there were in uses of task->nsproxy_pid_ns I was not familiar with, but it is only used when allocating a new pid for a new task, and in checks to prevent craziness from happening. Acked-by: Serge Hallyn Signed-off-by: "Eric W. Biederman"

Rename nsproxy.pid_ns to nsproxy.pid_ns_for_children

2013-08-27T17:52:52Z

nsproxy.pid_ns is *not* the task's pid namespace. The name should clarify that. This makes it more obvious that setns on a pid namespace is weird -- it won't change the pid namespace shown in procfs. Signed-off-by: Andy Lutomirski Reviewed-by: "Eric W. Biederman" Signed-off-by: David S. Miller

microblaze: fix clone syscall

2013-08-14T00:57:48Z

Fix inadvertent breakage in the clone syscall ABI for Microblaze that was introduced in commit f3268edbe6fe ("microblaze: switch to generic fork/vfork/clone"). The Microblaze syscall ABI for clone takes the parent tid address in the 4th argument; the third argument slot is used for the stack size. The incorrectly-used CLONE_BACKWARDS type assigned parent tid to the 3rd slot. This commit restores the original ABI so that existing userspace libc code will work correctly. All kernel versions from v3.8-rc1 were affected. Signed-off-by: Michal Simek Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

aio: convert the ioctx list to table lookup v3

2013-07-30T16:56:36Z

On Wed, Jun 12, 2013 at 11:14:40AM -0700, Kent Overstreet wrote: > On Mon, Apr 15, 2013 at 02:40:55PM +0300, Octavian Purdila wrote: > > When using a large number of threads performing AIO operations the > > IOCTX list may get a significant number of entries which will cause > > significant overhead. For example, when running this fio script: > > > > rw=randrw; size=256k ;directory=/mnt/fio; ioengine=libaio; iodepth=1 > > blocksize=1024; numjobs=512; thread; loops=100 > > > > on an EXT2 filesystem mounted on top of a ramdisk we can observe up to > > 30% CPU time spent by lookup_ioctx: > > > > 32.51% [guest.kernel] [g] lookup_ioctx > > 9.19% [guest.kernel] [g] __lock_acquire.isra.28 > > 4.40% [guest.kernel] [g] lock_release > > 4.19% [guest.kernel] [g] sched_clock_local > > 3.86% [guest.kernel] [g] local_clock > > 3.68% [guest.kernel] [g] native_sched_clock > > 3.08% [guest.kernel] [g] sched_clock_cpu > > 2.64% [guest.kernel] [g] lock_release_holdtime.part.11 > > 2.60% [guest.kernel] [g] memcpy > > 2.33% [guest.kernel] [g] lock_acquired > > 2.25% [guest.kernel] [g] lock_acquire > > 1.84% [guest.kernel] [g] do_io_submit > > > > This patchs converts the ioctx list to a radix tree. For a performance > > comparison the above FIO script was run on a 2 sockets 8 core > > machine. This are the results (average and %rsd of 10 runs) for the > > original list based implementation and for the radix tree based > > implementation: > > > > cores 1 2 4 8 16 32 > > list 109376 ms 69119 ms 35682 ms 22671 ms 19724 ms 16408 ms > > %rsd 0.69% 1.15% 1.17% 1.21% 1.71% 1.43% > > radix 73651 ms 41748 ms 23028 ms 16766 ms 15232 ms 13787 ms > > %rsd 1.19% 0.98% 0.69% 1.13% 0.72% 0.75% > > % of radix > > relative 66.12% 65.59% 66.63% 72.31% 77.26% 83.66% > > to list > > > > To consider the impact of the patch on the typical case of having > > only one ctx per process the following FIO script was run: > > > > rw=randrw; size=100m ;directory=/mnt/fio; ioengine=libaio; iodepth=1 > > blocksize=1024; numjobs=1; thread; loops=100 > > > > on the same system and the results are the following: > > > > list 58892 ms > > %rsd 0.91% > > radix 59404 ms > > %rsd 0.81% > > % of radix > > relative 100.87% > > to list > > So, I was just doing some benchmarking/profiling to get ready to send > out the aio patches I've got for 3.11 - and it looks like your patch is > causing a ~1.5% throughput regression in my testing :/ ... I've got an alternate approach for fixing this wart in lookup_ioctx()... Instead of using an rbtree, just use the reserved id in the ring buffer header to index an array pointing the ioctx. It's not finished yet, and it needs to be tidied up, but is most of the way there. -ben -- "Thought is the essence of where you are now." -- kmo> And, a rework of Ben's code, but this was entirely his idea kmo> -Kent bcrl> And fix the code to use the right mm_struct in kill_ioctx(), actually free memory. Signed-off-by: Benjamin LaHaise