- 09 May, 2007 40 commits
-
-
Jarek Poplawski authored
Signed-off-by:
Jarek Poplawski <jarkao2@o2.pl> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@elte.hu> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Eric Dumazet authored
Analysis of current linux futex code : -------------------------------------- A central hash table futex_queues[] holds all contexts (futex_q) of waiting threads. Each futex_wait()/futex_wait() has to obtain a spinlock on a hash slot to perform lookups or insert/deletion of a futex_q. When a futex_wait() is done, calling thread has to : 1) - Obtain a read lock on mmap_sem to be able to validate the user pointer (calling find_vma()). This validation tells us if the futex uses an inode based store (mapped file), or mm based store (anonymous mem) 2) - compute a hash key 3) - Atomic increment of reference counter on an inode or a mm_struct 4) - lock part of futex_queues[] hash table 5) - perform the test on value of futex. (rollback is value != expected_value, returns EWOULDBLOCK) (various loops if test triggers mm faults) 6) queue the context into hash table, release the lock got in 4) 7) - release the read_lock on mmap_sem <block> 8) Eventually unqueue the context (but rarely, as this part may be done by the futex_wake()) Futexes were designed to improve scalability but current implementation has various problems : - Central hashtable : This means scalability problems if many processes/threads want to use futexes at the same time. This means NUMA unbalance because this hashtable is located on one node. - Using mmap_sem on every futex() syscall : Even if mmap_sem is a rw_semaphore, up_read()/down_read() are doing atomic ops on mmap_sem, dirtying cache line : - lot of cache line ping pongs on SMP configurations. mmap_sem is also extensively used by mm code (page faults, mmap()/munmap()) Highly threaded processes might suffer from mmap_sem contention. mmap_sem is also used by oprofile code. Enabling oprofile hurts threaded programs because of contention on the mmap_sem cache line. - Using an atomic_inc()/atomic_dec() on inode ref counter or mm ref counter: It's also a cache line ping pong on SMP. It also increases mmap_sem hold time because of cache misses. Most of these scalability problems come from the fact that futexes are in one global namespace. As we use a central hash table, we must make sure they are all using the same reference (given by the mm subsystem). We chose to force all futexes be 'shared'. This has a cost. But fact is POSIX defined PRIVATE and SHARED, allowing clear separation, and optimal performance if carefuly implemented. Time has come for linux to have better threading performance. The goal is to permit new futex commands to avoid : - Taking the mmap_sem semaphore, conflicting with other subsystems. - Modifying a ref_count on mm or an inode, still conflicting with mm or fs. This is possible because, for one process using PTHREAD_PROCESS_PRIVATE futexes, we only need to distinguish futexes by their virtual address, no matter the underlying mm storage is. If glibc wants to exploit this new infrastructure, it should use new _PRIVATE futex subcommands for PTHREAD_PROCESS_PRIVATE futexes. And be prepared to fallback on old subcommands for old kernels. Using one global variable with the FUTEX_PRIVATE_FLAG or 0 value should be OK. PTHREAD_PROCESS_SHARED futexes should still use the old subcommands. Compatibility with old applications is preserved, they still hit the scalability problems, but new applications can fly :) Note : the same SHARED futex (mapped on a file) can be used by old binaries *and* new binaries, because both binaries will use the old subcommands. Note : Vast majority of futexes should be using PROCESS_PRIVATE semantic, as this is the default semantic. Almost all applications should benefit of this changes (new kernel and updated libc) Some bench results on a Pentium M 1.6 GHz (SMP kernel on a UP machine) /* calling futex_wait(addr, value) with value != *addr */ 433 cycles per futex(FUTEX_WAIT) call (mixing 2 futexes) 424 cycles per futex(FUTEX_WAIT) call (using one futex) 334 cycles per futex(FUTEX_WAIT_PRIVATE) call (mixing 2 futexes) 334 cycles per futex(FUTEX_WAIT_PRIVATE) call (using one futex) For reference : 187 cycles per getppid() call 188 cycles per umask() call 181 cycles per ni_syscall() call Signed-off-by:Eric Dumazet <dada1@cosmosbay.com> Pierre Peiffer <pierre.peiffer@bull.net> Cc: "Ulrich Drepper" <drepper@gmail.com> Cc: "Nick Piggin" <nickpiggin@yahoo.com.au> Cc: "Ingo Molnar" <mingo@elte.hu> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by:
-
Pierre Peiffer authored
This patch provides the futex_requeue_pi functionality, which allows some threads waiting on a normal futex to be requeued on the wait-queue of a PI-futex. This provides an optimization, already used for (normal) futexes, to be used with the PI-futexes. This optimization is currently used by the glibc in pthread_broadcast, when using "normal" mutexes. With futex_requeue_pi, it can be used with PRIO_INHERIT mutexes too. Signed-off-by:
Pierre Peiffer <pierre.peiffer@bull.net> Cc: Ingo Molnar <mingo@elte.hu> Cc: Ulrich Drepper <drepper@redhat.com> Signed-off-by:
-
Pierre Peiffer authored
This patch modifies futex_wait() to use an hrtimer + schedule() in place of schedule_timeout(). schedule_timeout() is tick based, therefore the timeout granularity is the tick (1 ms, 4 ms or 10 ms depending on HZ). By using a high resolution timer for timeout wakeup, we can attain a much finer timeout granularity (in the microsecond range). This parallels what is already done for futex_lock_pi(). The timeout passed to the syscall is no longer converted to jiffies and is therefore passed to do_futex() and futex_wait() as an absolute ktime_t therefore keeping nanosecond resolution. Also this removes the need to pass the nanoseconds timeout part to futex_lock_pi() in val2. In futex_wait(), if there is no timeout then a regular schedule() is performed. Otherwise, an hrtimer is fired before schedule() is called. [akpm@linux-foundation.org: fix `make headers_check'] Signed-off-by:
Sebastien Dugue <sebastien.dugue@bull.net> Signed-off-by:
-
Pierre Peiffer authored
Today, all threads waiting for a given futex are woken in FIFO order (first waiter woken first) instead of priority order. This patch makes use of plist (pirotity ordered lists) instead of simple list in futex_hash_bucket. All non-RT threads are stored with priority MAX_RT_PRIO, causing them to be woken last, in FIFO order (RT-threads are woken first, in priority order). Signed-off-by:
-
Andrew Morton authored
Some smarty went and inflicted ktime_t as a typedef upon us, so we cannot forward declare it. Create a new `union ktime', map ktime_t onto that. Now we need to kill off this ktime_t thing. Cc: Ingo Molnar <mingo@elte.hu> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: john stultz <johnstul@us.ibm.com> Signed-off-by:
-
Andrew Morton authored
Stick an unlikely() around is_aio(): I assert that most IO is synchronous. Cc: Suparna Bhattacharya <suparna@in.ibm.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Benjamin LaHaise <bcrl@kvack.org> Cc: Zach Brown <zach.brown@oracle.com> Cc: Ulrich Drepper <drepper@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by:
-
NeilBrown authored
When a lookup request arrives, nfsd uses information provided by userspace (mountd) to find the right filesystem. It then assumes that the same filehandle type as the incoming filehandle can be used to create an outgoing filehandle. However if mountd is buggy, or maybe just being creative, the filesystem may not support that filesystem type, and the kernel could oops, particularly if 'ex_uuid' is NULL but a FSID_UUID* filehandle type is used. So add some proper checking that the fsid version/type from the incoming filehandle is actually supportable, and ignore that information if it isn't supportable. Signed-off-by:
Neil Brown <neilb@suse.de> Signed-off-by:
-
NeilBrown authored
1/ decode_sattr and decode_sattr3 never return NULL, so remove several checks for that. ditto for xdr_decode_hyper. 2/ replace some open coded XDR_QUADLEN calls with calls to XDR_QUADLEN 3/ in decode_writeargs, simply an 'if' to use a single calculation. .page_len is the length of that part of the packet that did not fit in the first page (the head). So the length of the data part is the remainder of the head, plus page_len. 3/ other minor cleanups. Signed-off-by:
-
Christoph Hellwig authored
kbuild directly interprets <modulename>-y as objects to build into a module, no need to assign it to the old foo-objs variable. Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
NeilBrown authored
This while loop has an overly complex condition, which performs a couple of assignments. This hurts readability. We don't really need a loop at all. We can just return -EAGAIN and (providing we set SK_DATA), the function will be called again. So discard the loop, make the complex conditional become a few clear function calls, and hopefully improve readability. Signed-off-by:
-
Wei Yongjun authored
If I send a RPC_GSS_PROC_DESTROY message to NFSv4 server, it will reply with a bad rpc reply which lacks an authentication verifier. Maybe this patch is needed. Send/recv packets as following: send: RemoteProcedureCall xid rpcvers = 2 prog = 100003 vers = 4 proc = 0 cred = AUTH_GSS version = 1 gss_proc = 3 (RPCSEC_GSS_DESTROY) service = 1 (RPC_GSS_SVC_NONE) verf = AUTH_GSS checksum reply: RemoteProcedureReply xid msg_type reply_stat accepted_reply Signed-off-by:Wei Yongjun <yjwei@cn.fujitsu.com> Signed-off-by:
"J. Bruce Fields" <bfields@citi.umich.edu> Signed-off-by:
-
Frank Filz authored
I have been investigating a module reference count leak on the server for rpcsec_gss_krb5.ko. It turns out the problem is a reference count leak for the security context in net/sunrpc/auth_gss/svcauth_gss.c. The problem is that gss_write_init_verf() calls gss_svc_searchbyctx() which does a rsc_lookup() but never releases the reference to the context. There is another issue that rpc.svcgssd sets an "end of time" expiration for the context By adding a cache_put() call in gss_svc_searchbyctx(), and setting an expiration timeout in the downcall, cache_clean() does clean up the context and the module reference count now goes to zero after unmount. I also verified that if the context expires and then the client makes a new request, a new context is established. Here is the patch to fix the kernel, I will start a separate thread to discuss what expiration time should be set by rpc.svcgssd. Acked-by:
Frank Filz <ffilzlnx@us.ibm.com> Signed-off-by:
-
NeilBrown authored
It's not necessarily correct to assume that the xdr_buf used to hold the server's reply must have page data whenever it has tail data. And there's no need for us to deal with that case separately anyway. Acked-by:
-
NeilBrown authored
We need to zero various parts of 'exp' before any 'goto out', otherwise when we go to free the contents... we die. Signed-off-by:
-
Akinobu Mita authored
register_rpc_pipefs() needs to clean up rpc_inode_cache by kmem_cache_destroy() on register_filesystem() failure. init_sunrpc() needs to unregister rpc_pipe_fs by unregister_rpc_pipefs() when rpc_init_mempool() returns error. Signed-off-by:
Akinobu Mita <akinobu.mita@gmail.com> Cc: Neil Brown <neilb@suse.de> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by:
-
Jeff Layton authored
When the kernel calls svc_reserve to downsize the expected size of an RPC reply, it fails to account for the possibility of a checksum at the end of the packet. If a client mounts a NFSv2/3 with sec=krb5i/p, and does I/O then you'll generally see messages similar to this in the server's ring buffer: RPC request reserved 164 but used 208 While I was never able to verify it, I suspect that this problem is also the root cause of some oopses I've seen under these conditions: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=227726 This is probably also a problem for other sec= types and for NFSv4. The large reserved size for NFSv4 compound packets seems to generally paper over the problem, however. This patch adds a wrapper for svc_reserve that accounts for the possibility of a checksum. It also fixes up the appropriate callers of svc_reserve to call the wrapper. For now, it just uses a hardcoded value that I determined via testing. That value may need to be revised upward as things change, or we may want to eventually add a new auth_op that attempts to calculate this somehow. Unfortunately, there doesn't seem to be a good way to reliably determine the expected checksum length prior to actually calculating it, particularly with schemes like spkm3. Signed-off-by:
Jeff Layton <jlayton@redhat.com> Acked-by:
-
Eric W. Biederman authored
Acked-by:
Eric W. Biederman <ebiederm@xmission.com> Signed-off-by:
-
NeilBrown authored
Now that sk_defer_lock protects two different things, make the name more generic. Also don't bother with disabling _bh as the lock is only ever taken from process context. Signed-off-by:
-
Peter Staubach authored
The NFSv2 and NFSv3 servers do not handle WRITE requests for 0 bytes correctly. The specifications indicate that the server should accept the request, but it should mostly turn into a no-op. Currently, the server will return an XDR decode error, which it should not. Attached is a patch which addresses this issue. It also adds some boundary checking to ensure that the request contains as much data as was requested to be written. It also correctly handles an NFSv3 request which requests to write more data than the server has stated that it is prepared to handle. Previously, there was some support which looked like it should work, but wasn't quite right. Signed-off-by:
Peter Staubach <staubach@redhat.com> Acked-by:
-
Adrian Bunk authored
nfs4_acl_add_ace() can now be removed. Signed-off-by:
Adrian Bunk <bunk@stusta.de> Acked-by:
Neil Brown <neilb@cse.unsw.edu.au> Acked-by:
-
Oleg Nesterov authored
Thanks to Jarek Poplawski for the ideas and for spotting the bug in the initial draft patch. cancel_rearming_delayed_work() currently has many limitations, because it requires that dwork always re-arms itself via queue_delayed_work(). So it hangs forever if dwork doesn't do this, or cancel_rearming_delayed_work/ cancel_delayed_work was already called. It uses flush_workqueue() in a loop, so it can't be used if workqueue was freezed, and it is potentially live- lockable on busy system if delay is small. With this patch cancel_rearming_delayed_work() doesn't make any assumptions about dwork, it can re-arm itself via queue_delayed_work(), or queue_work(), or do nothing. As a "side effect", cancel_work_sync() was changed to handle re-arming works as well. Disadvantages: - this patch adds wmb() to insert_work(). - slowdowns the fast path (when del_timer() succeeds on entry) of cancel_rearming_delayed_work(), because wait_on_work() is called unconditionally. In that case, compared to the old version, we are doing "unneeded" lock/unlock for each online CPU. On the other hand, this means we don't need to use cancel_work_sync() after cancel_rearming_delayed_work(). - complicates the code (.text grows by 130 bytes). [akpm@linux-foundation.org: fix speling] Signed-off-by:
Oleg Nesterov <oleg@tv-sign.ru> Cc: David Chinner <dgc@sgi.com> Cc: David Howells <dhowells@redhat.com> Cc: Gautham Shenoy <ego@in.ibm.com> Acked-by:
-
Gautham R Shenoy authored
We are anyway kthread_stop()ping other per-cpu kernel threads after move_task_off_dead_cpu(), so we can do it with the stop_machine_run thread as well. I just checked with Vatsa if there was any subtle reason why they had put in the kthread_bind() in cpu.c. Vatsa cannot seem to recollect any and I can't see any. So let us just remove the kthread_bind. Signed-off-by:
Gautham R Shenoy <ego@in.ibm.com> Cc: Oleg Nesterov <oleg@tv-sign.ru> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Signed-off-by:
-
Oleg Nesterov authored
Currently kernel threads use sigprocmask(SIG_BLOCK) to protect against signals. This doesn't prevent the signal delivery, this only blocks signal_wake_up(). Every "killall -33 kthreadd" means a "struct siginfo" leak. Change kthreadd_setup() to set all handlers to SIG_IGN instead of blocking them (make a new helper ignore_signals() for that). If the kernel thread needs some signal, it should use allow_signal() anyway, and in that case it should not use CLONE_SIGHAND. Note that we can't change daemonize() (should die!) in the same way, because it can be used along with CLONE_SIGHAND. This means that allow_signal() still should unblock the signal to work correctly with daemonize()ed threads. However, disallow_signal() doesn't block the signal any longer but ignores it. NOTE: with or without this patch the kernel threads are not protected from handle_stop_signal(), this seems harmless, but not good. Signed-off-by:
-
Oleg Nesterov authored
worker_thread() inherits ignored SIGCHLD and numa_default_policy() from its parent, kthreadd. No need to setup this again. Signed-off-by:
-
Oleg Nesterov authored
allow_signal(SIGCHLD) does all necessary job, no need to call do_sigaction() prior to. Signed-off-by:
-
Oleg Nesterov authored
usbatm_do_heavy_init() calls allow_signal() which plays with parent process's ->sighand. Signed-off-by:
Duncan Sands <duncan.sands@free.fr> Cc: Greg KH <greg@kroah.com> Signed-off-by:
-
Eric W. Biederman authored
When a kernel thread calls daemonize, instead of reparenting the thread to init reparent the thread to kthreadd next to the threads created by kthread_create. This is really just a stop gap until daemonize goes away, but it does ensure no kernel threads are under init and they are all in one place that is easy to find. Signed-off-by:
-
Eric W. Biederman authored
Currently there is a circular reference between work queue initialization and kthread initialization. This prevents the kthread infrastructure from initializing until after work queues have been initialized. We want the properties of tasks created with kthread_create to be as close as possible to the init_task and to not be contaminated by user processes. The later we start our kthreadd that creates these tasks the harder it is to avoid contamination from user processes and the more of a mess we have to clean up because the defaults have changed on us. So this patch modifies the kthread support to not use work queues but to instead use a simple list of structures, and to have kthreadd start from init_task immediately after our kernel thread that execs /sbin/init. By being a true child of init_task we only have to change those process settings that we want to have different from init_task, such as our process name, the cpus that are allowed, blocking all signals and setting SIGCHLD to SIG_IGN so that all of our children are reaped automatically. By being a true child of init_task we also naturally get our ppid set to 0 and do not wind up as a child of PID == 1. Ensuring that tasks generated by kthread_create will not slow down the functioning of the wait family of functions. [akpm@linux-foundation.org: use interruptible sleeps] Signed-off-by:
-
Oleg Nesterov authored
____call_usermodehelper() has no reason for flush_signals(). It is a fresh forked process which is going to exec a user-space application or exit on failure. Signed-off-by:
-
Oleg Nesterov authored
flush_work(wq, work) doesn't need the first parameter, we can use cwq->wq (this was possible from the very beginnig, I missed this). So we can unify flush_work_keventd and flush_work. Also, rename flush_work() to cancel_work_sync() and fix all callers. Perhaps this is not the best name, but "flush_work" is really bad. (akpm: this is why the earlier patches bypassed maintainers) Signed-off-by:
-
Christoph Lameter authored
Shutdown the cache_reaper if the cpu is brought down and set the cache_reap.func to NULL. Otherwise hotplug shuts down the reaper for good. Signed-off-by:
Christoph Lameter <clameter@sgi.com> Cc: Oleg Nesterov <oleg@tv-sign.ru> Signed-off-by:
-
Oleg Nesterov authored
We already depend on fact that all sub-threads have ->exit_signal == -1, no need to set it in zap_other_threads(). Signed-off-by:
-
Oleg Nesterov authored
worker_thread() can miss freeze_process()->signal_wake_up() if it happens between try_to_freeze() and prepare_to_wait(). We should check freezing() before entering schedule(). This race was introduced by me in [PATCH 1/1] workqueue: don't migrate pending works from the dead CPU Looks like mm/vmscan.c:kswapd() has the same race. Signed-off-by:
-
Oleg Nesterov authored
worker_thread() doesn't need to "Block and flush all signals", this was already done by its caller, kthread(). Signed-off-by:
-
Oleg Nesterov authored
We don't have any users, and it is not so trivial to use NOAUTOREL works correctly. It is better to simplify API. Delete NOAUTOREL support and rename work_release to work_clear_pending to avoid a confusion. Signed-off-by:
David Howells <dhowells@redhat.com> Signed-off-by:
-
Oleg Nesterov authored
net/ipv4/ipvs/ip_vs_core.c module_exit ip_vs_cleanup ip_vs_control_cleanup cancel_rearming_delayed_work // done This is unsafe. The module may be unloaded and the memory may be freed while defense_work's handler is still running/preempted. Do flush_work(&defense_work.work) after cancel_rearming_delayed_work(). Alternatively, we could add flush_work() to cancel_rearming_delayed_work(), but note that we can't change cancel_delayed_work() in the same manner because it may be called from atomic context. Signed-off-by:
-
Oleg Nesterov authored
cancel_rearming_delayed_workqueue(wq, dwork) doesn't need the first parameter. We don't hang on un-queued dwork any longer, and work->data doesn't change its type. This means we can always figure out "wq" from dwork when it is needed. Remove this parameter, and rename the function to cancel_rearming_delayed_work(). Re-create an inline "obsolete" cancel_rearming_delayed_workqueue(wq) which just calls cancel_rearming_delayed_work(). Signed-off-by:
-
Oleg Nesterov authored
Cleanup. A number of per_cpu_ptr(wq->cpu_wq, cpu) users have to check that cpu is valid for this wq. Make a simple helper. Signed-off-by:
-
Oleg Nesterov authored
Change queue_delayed_work() to use queue_delayed_work_on() to avoid the code duplication (saves 133 bytes). Q: queue_delayed_work() enqueues &dwork->work directly when delay == 0, why? [jirislaby@gmail.com: oops fix] Signed-off-by:
Jiri Slaby <jirislaby@gmail.com> Signed-off-by:
-
