kevent: Core files.
Subsystems (11)
A message that is not in the archive (sent off-list or not yet received): [email protected]
cover kevent: Generic event handling mechanism.
Generic event handling mechanism.
Kevent is a generic subsytem which allows to handle event notifications. It supports both level and edge triggered events. It is similar to poll/epoll in some cases, but it is more scalable, it is faster and allows to work with essentially eny kind of events.
Events are provided into kernel through control syscall and can be read back through ring buffer or using usual syscalls.
Kevent update (i.e. readiness switching) happens directly from internals of the appropriate state machine of the underlying subsytem (like network, filesystem, timer or any other).
Homepage:
http://tservice.net.ru/~s0mbre/old/?section=projects&item=kevent
Documentation page:
http://linux-net.osdl.org/index.php/Kevent
I installed slightly used, but still functional (bought on ebay) remote mind reader, and set it up to read Ulrich's alpha brain waves (I hope he agrees that it is a good decision), which took me the whole week. So I think the last ring buffer implementation is what we all wanted. Details in documentation part.
It seems that setup was correct and we finially found what we wanted from interface part.
No changes since previous version, since it is a resend of the previous release made more than a week ago, which did not received any feedback.
Consider for inclusion.
Changes from 'take25' patchset:
* use timespec as timeout parameter.
* added high-resolution timer to handle absolute timeouts.
* added flags to waiting and initialization syscalls.
* kevent_commit() has new_uidx parameter.
* kevent_wait() has old_uidx parameter, which, if not equal to u->uidx, results in immediate wakeup (usefull for the case when entries
are added asynchronously from kernel (not supported for now)).
* added interface to mark any event as ready.
* event POSIX timers support.
* return -ENOSYS if there is no registered event type.
* provided file descriptor must be checked for fifo type (spotted by Eric Dumazet).
* signal notifications.
* documentation update.
* lighttpd patch updated (the latest benchmarks with lighttpd patch can be found in blog).
Changes from 'take24' patchset:
* new (old (new)) ring buffer implementation with kernel and user indexes.
* added initialization syscall instead of opening /dev/kevent
* kevent_commit() syscall to commit ring buffer entries
* changed KEVENT_REQ_WAKEUP_ONE flag to KEVENT_REQ_WAKEUP_ALL, kevent wakes only first thread always if that flag is not set
* KEVENT_REQ_ALWAYS_QUEUE flag. If set, kevent will be queued into ready queue instead of copying back to userspace when kevent is ready immediately when it is added.
* lighttpd patch (Hail! Although nothing really outstanding compared to epoll)
Changes from 'take23' patchset:
* kevent PIPE notifications
* KEVENT_REQ_LAST_CHECK flag, which allows to perform last check at dequeueing time
* fixed poll/select notifications (were broken due to tree manipulations)
* made Documentation/kevent.txt look nice in 80-col terminal
* fix for copy_to_user() failure report for the first kevent (Andrew Morton)
* minor function renames
Changes from 'take22' patchset:
* new ring buffer implementation in process' memory
* wakeup-one-thread flag
* edge-triggered behaviour
Changes from 'take21' patchset:
* minor cleanups (different return values, removed unneded variables, whitespaces and so on)
* fixed bug in kevent removal in case when kevent being removed
is the same as overflow_kevent (spotted by Eric Dumazet)
Changes from 'take20' patchset:
* new ring buffer implementation
* removed artificial limit on possible number of kevents
Changes from 'take19' patchset:
* use __init instead of __devinit
* removed 'default N' from config for user statistic
* removed kevent_user_fini() since kevent can not be unloaded
* use KERN_INFO for statistic output
Changes from 'take18' patchset:
* use __init instead of __devinit
* removed 'default N' from config for user statistic
* removed kevent_user_fini() since kevent can not be unloaded
* use KERN_INFO for statistic output
Changes from 'take17' patchset:
* Use RB tree instead of hash table.
At least for a web sever, frequency of addition/deletion of new kevent is comparable with number of search access, i.e. most of the time events are added, accesed only couple of times and then removed, so it justifies RB tree usage over AVL tree, since the latter does have much slower deletion time (max O(log(N)) compared to 3 ops),
although faster search time (1.44*O(log(N)) vs. 2*O(log(N))).
So for kevents I use RB tree for now and later, when my AVL tree implementation is ready, it will be possible to compare them.
* Changed readiness check for socket notifications.
With both above changes it is possible to achieve more than 3380 req/second compared to 2200, sometimes 2500 req/second for epoll() for trivial web-server and httperf client on the same hardware.
It is possible that above kevent limit is due to maximum allowed kevents in a time limit, which is 4096 events.
Changes from 'take16' patchset:
* misc cleanups (__read_mostly, const ...)
* created special macro which is used for mmap size (number of pages) calculation
* export kevent_socket_notify(), since it is used in network protocols which can be built as modules (IPv6 for example)
Changes from 'take15' patchset:
* converted kevent_timer to high-resolution timers, this forces timer API update at
http://linux-net.osdl.org/index.php/Kevent
* use struct ukevent* instead of void * in syscalls (documentation has been updated)
* added warning in kevent_add_ukevent() if ring has broken index (for testing)
Changes from 'take14' patchset:
* added kevent_wait()
This syscall waits until either timeout expires or at least one event becomes ready. It also commits that @num events from @start are processed by userspace and thus can be be removed or rearmed (depending on it's flags). It can be used for commit events read by userspace through mmap interface. Example userspace code (evtest.c) can be found on project's homepage.
* added socket notifications (send/recv/accept)
Changes from 'take13' patchset:
* do not get lock aroung user data check in __kevent_search()
* fail early if there were no registered callbacks for given type of kevent
* trailing whitespace cleanup
Changes from 'take12' patchset:
* remove non-chardev interface for initialization
* use pointer to kevent_mring instead of unsigned longs
* use aligned 64bit type in raw user data (can be used by high-res timer if needed)
* simplified enqueue/dequeue callbacks and kevent initialization
* use nanoseconds for timeout
* put number of milliseconds into timer's return data
* move some definitions into user-visible header
* removed filenames from comments
Changes from 'take11' patchset:
* include missing headers into patchset
* some trivial code cleanups (use goto instead of if/else games and so on)
* some whitespace cleanups
* check for ready_callback() callback before main loop which should save us some ticks
Changes from 'take10' patchset:
* removed non-existent prototypes
* added helper function for kevent_registered_callbacks
* fixed 80 lines comments issues
* added shared between userspace and kernelspace header instead of embedd them in one
* core restructuring to remove forward declarations
* s o m e w h i t e s p a c e c o d y n g s t y l e c l e a n u p
* use vm_insert_page() instead of remap_pfn_range()
Changes from 'take9' patchset:
* fixed ->nopage method
Changes from 'take8' patchset:
* fixed mmap release bug
* use module_init() instead of late_initcall()
* use better structures for timer notifications
Changes from 'take7' patchset:
* new mmap interface (not tested, waiting for other changes to be acked)
- use nopage() method to dynamically substitue pages
- allocate new page for events only when new added kevent requres it
- do not use ugly index dereferencing, use structure instead
- reduced amount of data in the ring (id and flags), maximum 12 pages on x86 per kevent fd
Changes from 'take6' patchset:
* a lot of comments!
* do not use list poisoning for detection of the fact, that entry is in the list
* return number of ready kevents even if copy*user() fails
* strict check for number of kevents in syscall
* use ARRAY_SIZE for array size calculation
* changed superblock magic number
* use SLAB_PANIC instead of direct panic() call
* changed -E* return values
* a lot of small cleanups and indent fixes
Changes from 'take5' patchset:
* removed compilation warnings about unused wariables when lockdep is not turned on
* do not use internal socket structures, use appropriate (exported) wrappers instead
* removed default 1 second timeout
* removed AIO stuff from patchset
Changes from 'take4' patchset:
* use miscdevice instead of chardevice
* comments fixes
Changes from 'take3' patchset:
* removed serializing mutex from kevent_user_wait()
* moved storage list processing to RCU
* removed lockdep screaming - all storage locks are initialized in the same function, so it was
learned
to differentiate between various cases
* remove kevent from storage if is marked as broken after callback
* fixed a typo in mmaped buffer implementation which would end up in wrong index calcualtion
Changes from 'take2' patchset:
* split kevent_finish_user() to locked and unlocked variants
* do not use KEVENT_STAT ifdefs, use inline functions instead
* use array of callbacks of each type instead of each kevent callback initialization
* changed name of ukevent guarding lock
* use only one kevent lock in kevent_user for all hash buckets instead of per-bucket locks
* do not use kevent_user_ctl structure instead provide needed arguments as syscall parameters
* various indent cleanups
* added optimisation, which is aimed to help when a lot of kevents are being copied from
userspace
* mapped buffer (initial) implementation (no userspace yet)
Changes from 'take1' patchset:
- rebased against 2.6.18-git tree
- removed ioctl controlling
- added new syscall kevent_get_events(int fd, unsigned int min_nr, unsigned int max_nr, unsigned int timeout, void __user *buf, unsigned flags)
- use old syscall kevent_ctl for creation/removing, modification and initial kevent initialization
- use mutuxes instead of semaphores
- added file descriptor check and return error if provided descriptor does not match kevent file operations
- various indent fixes
- removed aio_sendfile() declarations.
Thank you.
PATCH 1/8 kevent: Description.
Description.
Documentation/kevent.txt new file +240 -0
@@ -0,0 +1,240 @@ +Description. + +int kevent_init(struct kevent_ring *ring, unsigned int ring_size, + unsigned int flags); + +num - size of the ring buffer in events +ring - pointer to allocated ring buffer +flags - various flags, see KEVENT_FLAGS_* definitions. + +Return value: kevent control file descriptor or negative error value. + + struct kevent_ring + { + unsigned int ring_kidx, ring_over; + struct ukevent event[0]; + } + +ring_kidx - index in the ring buffer where kernel will put new events + when kevent_wait() or kevent_get_events() is called +ring_over - number of overflows of ring_uidx happend from the start. + Overflow counter is used to prevent situation when two threads + are going to free the same events, but one of them was scheduled + away for too long, so ring indexes were wrapped, so when that + thread will be awakened, it will free not those events, which + it suppose to free. + +Example userspace code (ring_buffer.c) can be found on project's homepage. + +Each kevent syscall can be so called cancellation point in glibc, i.e. when +thread has been cancelled in kevent syscall, thread can be safely removed +and no events will be lost, since each syscall (kevent_wait() or +kevent_get_events()) will copy event into special ring buffer, accessible +from other threads or even processes (if shared memory is used). + +When kevent is removed (not dequeued when it is ready, but just removed), +even if it was ready, it is not copied into ring buffer, since if it is +removed, no one cares about it (otherwise user would wait until it becomes +ready and got it through usual way using kevent_get_events() or kevent_wait()) +and thus no need to copy it to the ring buffer. + +------------------------------------------------------------------------------- + + +int kevent_ctl(int fd, unsigned int cmd, unsigned int num, struct ukevent *arg); + +fd - is the file descriptor referring to the kevent queue to manipulate. +It is created by opening "/dev/kevent" char device, which is created with +dynamic minor number and major number assigned for misc devices. + +cmd - is the requested operation. It can be one of the following: + KEVENT_CTL_ADD - add event notification + KEVENT_CTL_REMOVE - remove event notification + KEVENT_CTL_MODIFY - modify existing notification + KEVENT_CTL_READY - mark existing events as ready, if number of events is zero, + it just wakes up parked in syscall thread + +num - number of struct ukevent in the array pointed to by arg +arg - array of struct ukevent + +Return value: + number of events processed or negative error value. + +When called, kevent_ctl will carry out the operation specified in the +cmd parameter. +------------------------------------------------------------------------------- + + int kevent_get_events(int ctl_fd, unsigned int min_nr, unsigned int max_nr, + struct timespec timeout, struct ukevent *buf, unsigned flags); + +ctl_fd - file descriptor referring to the kevent queue +min_nr - minimum number of completed events that kevent_get_events will block + waiting for +max_nr - number of struct ukevent in buf +timeout - time to wait before returning less than min_nr + events. If this is -1, then wait forever. +buf - pointer to an array of struct ukevent. +flags - various flags, see KEVENT_FLAGS_* definitions. + +Return value: + number of events copied or negative error value. + +kevent_get_events will wait timeout milliseconds for at least min_nr completed +events, copying completed struct ukevents to buf and deleting any +KEVENT_REQ_ONESHOT event requests. In nonblocking mode it returns as many +events as possible, but not more than max_nr. In blocking mode it waits until +timeout or if at least min_nr events are ready. + +This function copies event into ring buffer if it was initialized, if ring buffer +is full, KEVENT_RET_COPY_FAILED flag is set in ret_flags field. +------------------------------------------------------------------------------- + + int kevent_wait(int ctl_fd, unsigned int num, unsigned int old_uidx, + struct timespec timeout, unsigned int flags); + +ctl_fd - file descriptor referring to the kevent queue +num - number of processed kevents +old_uidx - the last index user is aware of +timeout - time to wait until there is free space in kevent queue +flags - various flags, see KEVENT_FLAGS_* definitions. + +Return value: + number of events copied into ring buffer or negative error value. + +This syscall waits until either timeout expires or at least one event becomes +ready. It also copies events into special ring buffer. If ring buffer is full, +it waits until there are ready events and then return. +If kevent is one-shot kevent it is removed in this syscall. +If kevent is edge-triggered (KEVENT_REQ_ET flag is set in 'req_flags') it is +requeued in this syscall for performance reasons. +------------------------------------------------------------------------------- + + int kevent_commit(int ctl_fd, unsigned int new_idx, unsigned int over); + +ctl_fd - file descriptor referring to the kevent queue +new_uidx - the last committed kevent +over - overflow count for given $new_idx value + +Return value: + number of committed kevents or negative error value. + +This function commits, i.e. marks as empty, slots in the ring buffer, so +they can be reused when userspace completes that entries processing. + +Overflow counter is used to prevent situation when two threads are going +to free the same events, but one of them was scheduled away for too long, +so ring indexes were wrapped, so when that thread will be awakened, it +will free not those events, which it suppose to free. + +It is possible that returned number of committed events will be smaller than +requested number - it is possible when several threads try to commit the +same events. +------------------------------------------------------------------------------- + +The bulk of the interface is entirely done through the ukevent struct. +It is used to add event requests, modify existing event requests, +specify which event requests to remove, and return completed events. + +struct ukevent contains the following members: + +struct kevent_id id + Id of this request, e.g. socket number, file descriptor and so on +__u32 type + Event type, e.g. KEVENT_SOCK, KEVENT_INODE, KEVENT_TIMER and so on +__u32 event + Event itself, e.g. SOCK_ACCEPT, INODE_CREATED, TIMER_FIRED +__u32 req_flags + Per-event request flags, + + KEVENT_REQ_ONESHOT + event will be removed when it is ready + + KEVENT_REQ_WAKEUP_ALL + Kevent wakes up only first thread interested in given event, + or all threads if this flag is set. + + KEVENT_REQ_ET + Edge Triggered behaviour. It is an optimisation which allows to move + ready and dequeued (i.e. copied to userspace) event to move into set + of interest for given storage (socket, inode and so on) again. It is + very usefull for cases when the same event should be used many times + (like reading from pipe). It is similar to epoll()'s EPOLLET flag. + + KEVENT_REQ_LAST_CHECK + if set allows to perform the last check on kevent (call appropriate + callback) when kevent is marked as ready and has been removed from + ready queue. If it will be confirmed that kevent is ready + (k->callbacks.callback(k) returns true) then kevent will be copied + to userspace, otherwise it will be requeued back to storage. + Second (checking) call is performed with this bit cleared, so callback + can detect when it was called from kevent_storage_ready() - bit is set, + or kevent_dequeue_ready() - bit is cleared. If kevent will be requeued, + bit will be set again. + + KEVENT_REQ_ALWAYS_QUEUE + If this flag is set kevent will be queued into ready queue if it is + ready at enqueue time, otherwise it will be copied back to userspace + and will not be queued into the storage. + +__u32 ret_flags + Per-event return flags + + KEVENT_RET_BROKEN + Kevent is broken + + KEVENT_RET_DONE + Kevent processing was finished successfully + + KEVENT_RET_COPY_FAILED + Kevent was not copied into ring buffer due to some error conditions. + +__u32 ret_data + Event return data. Event originator fills it with anything it likes + (for example timer notifications put number of milliseconds when timer + has fired +union { __u32 user[2]; void *ptr; } + User's data. It is not used, just copied to/from user. The whole structure + is aligned to 8 bytes already, so the last union is aligned properly. + +------------------------------------------------------------------------------- + +Kevent waiting syscall flags. + +KEVENT_FLAGS_ABSTIME - provided timespec parameter contains absolute time, + for example Aug 27, 2194, or time(NULL) + 10. + +------------------------------------------------------------------------------- + +Usage + +For KEVENT_CTL_ADD, all fields relevant to the event type must be filled +(id, type, event, req_flags). +After kevent_ctl(..., KEVENT_CTL_ADD, ...) returns each struct's ret_flags +should be checked to see if the event is already broken or done. + +For KEVENT_CTL_MODIFY, the id, req_flags, and user and event fields must be +set and an existing kevent request must have matching id and user fields. If +match is found, req_flags and event are replaced with the newly supplied +values and requeueing is started, so modified kevent can be checked and +probably marked as ready immediately. If a match can't be found, the +passed in ukevent's ret_flags has KEVENT_RET_BROKEN set. KEVENT_RET_DONE is +always set. + +For KEVENT_CTL_REMOVE, the id and user fields must be set and an existing +kevent request must have matching id and user fields. If a match is found, +the kevent request is removed. If a match can't be found, the passed in +ukevent's ret_flags has KEVENT_RET_BROKEN set. KEVENT_RET_DONE is always set. + +For kevent_get_events, the entire structure is returned. + +------------------------------------------------------------------------------- + +Usage cases + +kevent_timer +struct ukevent should contain following fields: + type - KEVENT_TIMER + event - KEVENT_TIMER_FIRED + req_flags - KEVENT_REQ_ONESHOT if you want to fire that timer only once + id.raw[0] - number of seconds after commit when this timer shout expire + id.raw[0] - additional to number of seconds number of nanoseconds
PATCH 2/8 kevent: Core files.
Core files.
This patch includes core kevent files:
* userspace controlling
* kernelspace interfaces
* initialization
* notification state machines
Some bits of documentation can be found on project's homepage (and links from there):
http://tservice.net.ru/~s0mbre/old/?section=projects&item=kevent
arch/i386/kernel/syscall_table.S +5 -0
@@ -318,3 +318,8 @@ ENTRY(sys_call_table) .long sys_vmsplice .long sys_move_pages .long sys_getcpu + .long sys_kevent_get_events + .long sys_kevent_ctl /* 320 */ + .long sys_kevent_wait + .long sys_kevent_commit + .long sys_kevent_init
arch/x86_64/ia32/ia32entry.S +6 -1
@@ -714,8 +714,13 @@ ia32_sys_call_table: .quad compat_sys_get_robust_list .quad sys_splice .quad sys_sync_file_range - .quad sys_tee + .quad sys_tee /* 315 */ .quad compat_sys_vmsplice .quad compat_sys_move_pages .quad sys_getcpu + .quad sys_kevent_get_events + .quad sys_kevent_ctl /* 320 */ + .quad sys_kevent_wait + .quad sys_kevent_commit + .quad sys_kevent_init ia32_syscall_end:
include/asm-i386/unistd.h +6 -1
@@ -324,10 +324,15 @@#define __NR_vmsplice 316#define __NR_move_pages 317#define __NR_getcpu 318 +#define __NR_kevent_get_events 319 +#define __NR_kevent_ctl 320 +#define __NR_kevent_wait 321 +#define __NR_kevent_commit 322 +#define __NR_kevent_init 323#ifdef __KERNEL__ -#define NR_syscalls 319 +#define NR_syscalls 324#include <linux/err.h> /*
include/asm-x86_64/unistd.h +11 -1
@@ -619,10 +619,20 @@ __SYSCALL(__NR_sync_file_range, sys_sync_file_range) __SYSCALL(__NR_vmsplice, sys_vmsplice)#define __NR_move_pages 279 __SYSCALL(__NR_move_pages, sys_move_pages) +#define __NR_kevent_get_events 280 +__SYSCALL(__NR_kevent_get_events, sys_kevent_get_events) +#define __NR_kevent_ctl 281 +__SYSCALL(__NR_kevent_ctl, sys_kevent_ctl) +#define __NR_kevent_wait 282 +__SYSCALL(__NR_kevent_wait, sys_kevent_wait) +#define __NR_kevent_commit 283 +__SYSCALL(__NR_kevent_commit, sys_kevent_commit) +#define __NR_kevent_init 284 +__SYSCALL(__NR_kevent_init, sys_kevent_init)#ifdef __KERNEL__ -#define __NR_syscall_max __NR_move_pages +#define __NR_syscall_max __NR_kevent_init#include <linux/err.h> #ifndef __NO_STUBS
include/linux/kevent.h new file +238 -0
238 lines not on this page · raw message
include/linux/kevent_storage.h new file +11 -0
@@ -0,0 +1,11 @@ +#ifndef __KEVENT_STORAGE_H +#define __KEVENT_STORAGE_H + +struct kevent_storage +{ +void *origin;/* Originator's pointer, e.g. struct sock or struct file. Can be NULL. */ +struct list_head list;/* List of queued kevents. */ +spinlock_t lock;/* Protects users queue. */ +}; + +#endif /* __KEVENT_STORAGE_H */
include/linux/syscalls.h +9 -0
@@ -54,6 +54,8 @@ struct compat_stat;struct compat_timeval ;struct robust_list_head ;struct getcpu_cache ; +struct ukevent ; +struct kevent_ring ;#include <linux/types.h> #include <linux/aio_abi.h>
@@ -599,4 +601,11 @@ asmlinkage long sys_set_robust_list(struct robust_list_head __user *head, size_t len);asmlinkage long sys_getcpu (unsigned __user *cpu,unsigned __user *node,struct getcpu_cache __user *cache); +asmlinkage long sys_kevent_get_events (int ctl_fd,unsigned int min,unsigned int max, +struct timespec timeout,struct ukevent __user *buf,unsigned flags); +asmlinkage long sys_kevent_ctl (int ctl_fd,unsigned int cmd,unsigned int num,struct ukevent __user *buf); +asmlinkage long sys_kevent_wait (int ctl_fd,unsigned int num,unsigned int old_uidx, +struct timespec timeout,unsigned int flags); +asmlinkage long sys_kevent_commit (int ctl_fd,unsigned int new_uidx,unsigned int over); +asmlinkage long sys_kevent_init (int ctl_fd,struct kevent_ring __user *ring,unsigned int num,unsigned int flags);#endif
include/linux/ukevent.h new file +183 -0
183 lines not on this page · raw message
init/Kconfig +2 -0
@@ -201,6 +201,8 @@ config AUDITSYSCALL such as SELinux. To use audit's filesystem watch feature, please ensure that INOTIFY is configured. +source "kernel/kevent/Kconfig" + config IKCONFIG bool "Kernel .config support" ---help---
kernel/Makefile +1 -0
@@ -47,6 +47,7 @@ obj-$(CONFIG_DETECT_SOFTLOCKUP) += softlockup.o obj-$(CONFIG_GENERIC_HARDIRQS) += irq/ obj-$(CONFIG_SECCOMP) += seccomp.o obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o +obj-$(CONFIG_KEVENT) += kevent/ obj-$(CONFIG_RELAY) += relay.o obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o obj-$(CONFIG_TASKSTATS) += taskstats.o
kernel/kevent/Kconfig new file +60 -0
60 lines not on this page · raw message
kernel/kevent/Makefile new file +6 -0
@@ -0,0 +1,6 @@ +obj-y := kevent.o kevent_user.o +obj-$(CONFIG_KEVENT_TIMER) += kevent_timer.o +obj-$(CONFIG_KEVENT_POLL) += kevent_poll.o +obj-$(CONFIG_KEVENT_SOCKET) += kevent_socket.o +obj-$(CONFIG_KEVENT_PIPE) += kevent_pipe.o +obj-$(CONFIG_KEVENT_SIGNAL) += kevent_signal.o
kernel/kevent/kevent.c new file +247 -0
247 lines not on this page · raw message
kernel/kevent/kevent_user.c new file +1344 -0
1,344 lines not on this page · raw message
kernel/sys_ni.c +6 -0
@@ -122,6 +122,12 @@ cond_syscall(ppc_rtas);cond_syscall (sys_spu_run);cond_syscall (sys_spu_create); +cond_syscall (sys_kevent_get_events); +cond_syscall (sys_kevent_ctl); +cond_syscall (sys_kevent_wait); +cond_syscall (sys_kevent_commit); +cond_syscall (sys_kevent_init); +/* mmu depending weak syscall entries */ cond_syscall (sys_mprotect);cond_syscall (sys_msync);
PATCH 3/8 kevent: poll/select() notifications.
poll/select() notifications.
This patch includes generic poll/select notifications. kevent_poll works simialr to epoll and has the same issues (callback is invoked not from internal state machine of the caller, but through process awake, a lot of allocations and so on).
fs/file_table.c +3 -0
@@ -20,6 +20,7 @@#include <linux/cdev.h> #include <linux/fsnotify.h> #include <linux/sysctl.h> +#include <linux/kevent.h> #include <linux/percpu_counter.h> #include <asm/atomic.h>
@@ -119,6 +120,7 @@ struct file *get_empty_filp(void) f->f_uid = tsk->fsuid; f->f_gid = tsk->fsgid;eventpoll_init_file (f); +kevent_init_file (f);/* f->f_version: 0 */ return f;
@@ -164,6 +166,7 @@ void fastcall __fput(struct file *file)* in the file cleanup chain. */ eventpoll_release (file); +kevent_cleanup_file (file);locks_remove_flock (file);if (file->f_op && file->f_op->release)
include/linux/fs.h +8 -0
@@ -276,6 +276,7 @@ extern int dir_notify_enable;#include <linux/init.h> #include <linux/sched.h> #include <linux/mutex.h> +#include <linux/kevent_storage.h> #include <asm/atomic.h> #include <asm/semaphore.h>
@@ -586,6 +587,10 @@ struct inode {struct mutex inotify_mutex;/* protects the watches list */ #endif +#if definedCONFIG_KEVENT_SOCKET || definedCONFIG_KEVENT_PIPE +struct kevent_storage st; +#endif +unsigned long i_state;unsigned long dirtied_when;/* jiffies of first dirtying */
@@ -739,6 +744,9 @@ struct file {struct list_head f_ep_links;spinlock_t f_ep_lock;#endif /* #ifdef CONFIG_EPOLL */ +#ifdef CONFIG_KEVENT_POLL +struct kevent_storage st; +#endif struct address_space *f_mapping; };extern spinlock_t files_lock;
kernel/kevent/kevent_poll.c new file +232 -0
232 lines not on this page · raw message
PATCH 4/8 kevent: Socket notifications.
Socket notifications.
This patch includes socket send/recv/accept notifications. Using trivial web server based on kevent and this features instead of epoll it's performance increased more than noticebly. More details about various benchmarks and server itself (evserver_kevent.c) can be found on project's homepage.
fs/inode.c +7 -0
@@ -21,6 +21,7 @@#include <linux/cdev.h> #include <linux/bootmem.h> #include <linux/inotify.h> +#include <linux/kevent.h> #include <linux/mount.h> /*
@@ -164,12 +165,18 @@ static struct inode *alloc_inode(struct super_block *sb) } inode->i_private =0 ; inode->i_mapping = mapping; +#if definedCONFIG_KEVENT_SOCKET || definedCONFIG_KEVENT_PIPE +kevent_storage_init (inode, &inode->st); +#endif }return inode; }void destroy_inode (struct inode *inode) { +#if definedCONFIG_KEVENT_SOCKET || definedCONFIG_KEVENT_PIPE +kevent_storage_fini (&inode->st); +#endif BUG_ON (inode_has_buffers (inode));security_inode_free (inode);if (inode->i_sb->s_op->destroy_inode)
include/net/sock.h +17 -15
@@ -48,6 +48,7 @@#include <linux/netdevice.h> #include <linux/skbuff.h> /* struct sk_buff */ #include <linux/security.h> +#include <linux/kevent.h> #include <linux/filter.h>
@@ -450,6 +451,21 @@ static inline int sk_stream_memory_free(struct sock *sk)extern void sk_stream_rfree (struct sk_buff *skb); +struct socket_alloc { +struct socket socket; +struct inode vfs_inode; +}; + +static inline struct socket *SOCKET_I (struct inode *inode) +{ + return &container_of(inode, struct socket_alloc, vfs_inode)->socket; +} + +static inline struct inode *SOCK_INODE (struct socket *socket) +{ + return &container_of(socket, struct socket_alloc, socket)->vfs_inode; +} +static inline void sk_stream_set_owner_r (struct sk_buff *skb,struct sock *sk) { skb->sk = sk;
@@ -477,6 +493,7 @@ static inline void sk_add_backlog(struct sock *sk, struct sk_buff *skb) sk->sk_backlog.tail = skb; } skb->next =NULL ; +kevent_socket_notify (sk,KEVENT_SOCKET_RECV ); }#define sk_wait_event (__sk, __timeo, __condition) \
@@ -679,21 +696,6 @@ static inline struct kiocb *siocb_to_kiocb(struct sock_iocb *si)return si->kiocb; } -struct socket_alloc { -struct socket socket; -struct inode vfs_inode; -}; - -static inline struct socket *SOCKET_I (struct inode *inode) -{ - return &container_of(inode, struct socket_alloc, vfs_inode)->socket; -} - -static inline struct inode *SOCK_INODE (struct socket *socket) -{ - return &container_of(socket, struct socket_alloc, socket)->vfs_inode; -} -extern void __sk_stream_mem_reclaim (struct sock *sk);extern int sk_stream_mem_schedule (struct sock *sk,int size,int kind);
include/net/tcp.h +1 -0
@@ -857,6 +857,7 @@ static inline int tcp_prequeue(struct sock *sk, struct sk_buff *skb) tp->ucopy.memory =0 ; }else if (skb_queue_len (&tp->ucopy.prequeue) ==1 ) {wake_up_interruptible (sk->sk_sleep); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND );if (!inet_csk_ack_scheduled (sk)) inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, (3 * TCP_RTO_MIN) / 4,
kernel/kevent/kevent_socket.c new file +142 -0
142 lines not on this page · raw message
net/core/sock.c +9 -1
@@ -1402,6 +1402,7 @@ static void sock_def_wakeup(struct sock *sk)if (sk->sk_sleep &&waitqueue_active (sk->sk_sleep))wake_up_interruptible_all (sk->sk_sleep);read_unlock (&sk->sk_callback_lock); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND ); }static void sock_def_error_report (struct sock *sk)
@@ -1411,6 +1412,7 @@ static void sock_def_error_report(struct sock *sk)wake_up_interruptible (sk->sk_sleep);sk_wake_async (sk,0 ,POLL_ERR );read_unlock (&sk->sk_callback_lock); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND ); }static void sock_def_readable (struct sock *sk,int len)
@@ -1420,6 +1422,7 @@ static void sock_def_readable(struct sock *sk, int len)wake_up_interruptible (sk->sk_sleep);sk_wake_async (sk,1 ,POLL_IN );read_unlock (&sk->sk_callback_lock); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND ); }static void sock_def_write_space (struct sock *sk)
@@ -1439,6 +1442,7 @@ static void sock_def_write_space(struct sock *sk) }read_unlock (&sk->sk_callback_lock); +kevent_socket_notify (sk,KEVENT_SOCKET_SEND |KEVENT_SOCKET_RECV ); }static void sock_def_destruct (struct sock *sk)
@@ -1489,6 +1493,8 @@ void sock_init_data(struct socket *sock, struct sock *sk) sk->sk_state =TCP_CLOSE ; sk->sk_socket = sock; +kevent_sk_reinit (sk); +sock_set_flag (sk,SOCK_ZAPPED );if (sock)
@@ -1555,8 +1561,10 @@ void fastcall release_sock(struct sock *sk)if (sk->sk_backlog.tail)__release_sock (sk); sk->sk_lock.owner =NULL ; -if (waitqueue_active (&sk->sk_lock.wq)) +if (waitqueue_active (&sk->sk_lock.wq)) {wake_up (&sk->sk_lock.wq); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND ); + }spin_unlock_bh (&sk->sk_lock.slock); }EXPORT_SYMBOL (release_sock);
net/core/stream.c +1 -0
@@ -36,6 +36,7 @@ void sk_stream_write_space(struct sock *sk)wake_up_interruptible (sk->sk_sleep);if (sock->fasync_list && !(sk->sk_shutdown &SEND_SHUTDOWN ))sock_wake_async (sock,2 ,POLL_OUT ); +kevent_socket_notify (sk,KEVENT_SOCKET_SEND |KEVENT_SOCKET_RECV ); } }
net/ipv4/tcp_input.c +1 -0
@@ -3119,6 +3119,7 @@ static void tcp_ofo_queue(struct sock *sk)__skb_unlink (skb, &tp->out_of_order_queue);__skb_queue_tail (&sk->sk_receive_queue, skb); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV ); tp->rcv_nxt =TCP_SKB_CB (skb)->end_seq;if (skb->h.th->fin)tcp_fin (skb, sk, skb->h.th);
net/ipv4/tcp_ipv4.c +2 -0
@@ -61,6 +61,7 @@#include <linux/jhash.h> #include <linux/init.h> #include <linux/times.h> +#include <linux/kevent.h> #include <net/icmp.h> #include <net/inet_hashtables.h>
@@ -870,6 +871,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)reqsk_free (req); }else {inet_csk_reqsk_queue_hash_add (sk, req,TCP_TIMEOUT_INIT ); +kevent_socket_notify (sk,KEVENT_SOCKET_ACCEPT ); }return 0 ;
net/socket.c +3 -0
@@ -85,6 +85,7 @@#include <linux/kmod.h> #include <linux/audit.h> #include <linux/wireless.h> +#include <linux/kevent.h> #include <asm/uaccess.h> #include <asm/unistd.h>
@@ -490,6 +491,8 @@ static struct socket *sock_alloc(void) inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; +kevent_socket_reinit (sock); +get_cpu_var (sockets_in_use)++;put_cpu_var (sockets_in_use);return sock;
PATCH 5/8 kevent: Timer notifications.
Timer notifications.
Timer notifications can be used for fine grained per-process time management, since interval timers are very inconvenient to use, and they are limited.
This subsystem uses high-resolution timers.
id.raw[0] is used as number of seconds
id.raw[1] is used as number of nanoseconds
kernel/kevent/kevent_timer.c new file +112 -0
@@ -0,0 +1,112 @@ +/* +* 2006 Copyright (c) Evgeniy Polyakov <[email protected]> +* All rights reserved. +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation; either version 2 of the License, or +* (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, write to the Free Software +* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA +*/ + +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/list.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/hrtimer.h> +#include <linux/jiffies.h> +#include <linux/kevent.h> + +struct kevent_timer +{ +struct hrtimer ktimer; +struct kevent_storage ktimer_storage; +struct kevent *ktimer_event; +}; + +static int kevent_timer_func (struct hrtimer *timer) +{ + struct kevent_timer *t = container_of(timer, struct kevent_timer, ktimer); +struct kevent *k = t->ktimer_event; + +kevent_storage_ready (&t->ktimer_storage,NULL ,KEVENT_MASK_ALL ); +hrtimer_forward (timer, timer->base->softirq_time, +ktime_set (k->event.id.raw[0 ], k->event.id.raw[1 ])); +return HRTIMER_RESTART ; +} + +static struct lock_class_key kevent_timer_key; + +static int kevent_timer_enqueue (struct kevent *k) +{ +int err; +struct kevent_timer *t; + + t =kmalloc (sizeof (struct kevent_timer ),GFP_KERNEL ); +if (!t) +return -ENOMEM ; + +hrtimer_init (&t->ktimer,CLOCK_MONOTONIC ,HRTIMER_REL ); + t->ktimer.expires =ktime_set (k->event.id.raw[0 ], k->event.id.raw[1 ]); + t->ktimer.function = kevent_timer_func; + t->ktimer_event = k; + + err =kevent_storage_init (&t->ktimer, &t->ktimer_storage); +if (err) + goto err_out_free; +lockdep_set_class (&t->ktimer_storage.lock, &kevent_timer_key); + + err =kevent_storage_enqueue (&t->ktimer_storage, k); +if (err) + goto err_out_st_fini; + +hrtimer_start (&t->ktimer, t->ktimer.expires,HRTIMER_REL ); + +return 0 ; + +err_out_st_fini: +kevent_storage_fini (&t->ktimer_storage); +err_out_free: +kfree (t); + +return err; +} + +static int kevent_timer_dequeue (struct kevent *k) +{ +struct kevent_storage *st = k->st; + struct kevent_timer *t = container_of(st, struct kevent_timer, ktimer_storage); + +hrtimer_cancel (&t->ktimer); +kevent_storage_dequeue (st, k); +kfree (t); + +return 0 ; +} + +static int kevent_timer_callback (struct kevent *k) +{ + k->event.ret_data[0 ] =jiffies_to_msecs (jiffies); +return 1 ; +} + +static int __init kevent_init_timer (void ) +{ +struct kevent_callbacks tc = { + .callback = &kevent_timer_callback, + .enqueue = &kevent_timer_enqueue, + .dequeue = &kevent_timer_dequeue}; + +return kevent_add_callbacks (&tc,KEVENT_TIMER ); +} +module_init (kevent_init_timer); +
PATCH 6/8 kevent: Pipe notifications.
Pipe notifications.
fs/pipe.c +6 -0
@@ -16,6 +16,7 @@#include <linux/uio.h> #include <linux/highmem.h> #include <linux/pagemap.h> +#include <linux/kevent.h> #include <asm/uaccess.h> #include <asm/ioctls.h>
@@ -312,6 +313,7 @@ redo:break ; }if (do_wakeup) { +kevent_pipe_notify (inode,KEVENT_SOCKET_SEND );wake_up_interruptible_sync (&pipe->wait);kill_fasync (&pipe->fasync_writers,SIGIO ,POLL_OUT ); }
@@ -321,6 +323,7 @@ redo:/* Signal writers asynchronously that there is more room. */ if (do_wakeup) { +kevent_pipe_notify (inode,KEVENT_SOCKET_SEND );wake_up_interruptible (&pipe->wait);kill_fasync (&pipe->fasync_writers,SIGIO ,POLL_OUT ); }
@@ -490,6 +493,7 @@ redo2:break ; }if (do_wakeup) { +kevent_pipe_notify (inode,KEVENT_SOCKET_RECV );wake_up_interruptible_sync (&pipe->wait);kill_fasync (&pipe->fasync_readers,SIGIO ,POLL_IN ); do_wakeup =0 ;
@@ -501,6 +505,7 @@ redo2: out:mutex_unlock (&inode->i_mutex);if (do_wakeup) { +kevent_pipe_notify (inode,KEVENT_SOCKET_RECV );wake_up_interruptible (&pipe->wait);kill_fasync (&pipe->fasync_readers,SIGIO ,POLL_IN ); }
@@ -605,6 +610,7 @@ pipe_release(struct inode *inode, int decr, int decw)free_pipe_info (inode); }else {wake_up_interruptible (&pipe->wait); +kevent_pipe_notify (inode,KEVENT_SOCKET_SEND |KEVENT_SOCKET_RECV );kill_fasync (&pipe->fasync_readers,SIGIO ,POLL_IN );kill_fasync (&pipe->fasync_writers,SIGIO ,POLL_OUT ); }
kernel/kevent/kevent_pipe.c new file +121 -0
121 lines not on this page · raw message
PATCH 7/8 kevent: Signal notifications.
Signal notifications.
This type of notifications allows to deliver signals through kevent queue. One can find example application signal.c on project homepage.
If KEVENT_SIGNAL_NOMASK bit is set in raw_u64 id then signal will be delivered only through queue, otherwise both delivery types are used - old through update of mask of pending signals and through queue.
If signal is delivered only through kevent queue mask of pending signals is not updated at all, which is equal to putting signal into blocked mask, but with delivery of that signal through kevent queue.
include/linux/sched.h +5 -0
@@ -80,6 +80,7 @@ struct sched_param {#include <linux/resource.h> #include <linux/timer.h> #include <linux/hrtimer.h> +#include <linux/kevent_storage.h> #include <asm/processor.h>
@@ -1013,6 +1014,10 @@ struct task_struct {#ifdef CONFIG_TASK_DELAY_ACCT struct task_delay_info *delays;#endif +#ifdef CONFIG_KEVENT_SIGNAL +struct kevent_storage st; +u32 kevent_signals; +#endif };static inline pid_t process_group (struct task_struct *tsk)
kernel/fork.c +8 -0
@@ -46,6 +46,7 @@#include <linux/delayacct.h> #include <linux/taskstats_kern.h> #include <linux/random.h> +#include <linux/kevent.h> #include <asm/pgtable.h> #include <asm/pgalloc.h>
@@ -115,6 +116,9 @@ void __put_task_struct(struct task_struct *tsk)WARN_ON (atomic_read (&tsk->usage));WARN_ON (tsk == current); +#ifdef CONFIG_KEVENT_SIGNAL +kevent_storage_fini (&tsk->st); +#endif security_task_free (tsk);free_uid (tsk->user);put_group_info (tsk->group_info);
@@ -1121,6 +1125,10 @@ static struct task_struct *copy_process(unsigned long clone_flags,if (retval) goto bad_fork_cleanup_namespace; +#ifdef CONFIG_KEVENT_SIGNAL +kevent_storage_init (p, &p->st); +#endif + p->set_child_tid = (clone_flags &CLONE_CHILD_SETTID ) ? child_tidptr :NULL ;/* * Clear TID on mm_release()?
kernel/kevent/kevent_signal.c new file +92 -0
92 lines not on this page · raw message
kernel/signal.c +26 -0
@@ -23,6 +23,7 @@#include <linux/ptrace.h> #include <linux/signal.h> #include <linux/capability.h> +#include <linux/kevent.h> #include <asm/param.h> #include <asm/uaccess.h> #include <asm/unistd.h>
@@ -703,6 +704,9 @@ static int send_signal(int sig, struct siginfo *info, struct task_struct *t, {struct sigqueue * q =NULL ;int ret =0 ; + +if (kevent_signal_notify (t, sig)) +return 1 ;/* * fast-pathed signals for kernel-internal things like SIGSTOP
@@ -782,6 +786,17 @@ specific_send_sig_info(int sig, struct siginfo *info, struct task_struct *t) ret =send_signal (sig, info, t, &t->pending);if (!ret && !sigismember (&t->blocked, sig))signal_wake_up (t, sig ==SIGKILL ); +#ifdef CONFIG_KEVENT_SIGNAL +/* +* Kevent allows to deliver signals through kevent queue, +* it is possible to setup kevent to not deliver +* signal through the usual way, in that case send_signal() +* returns 1 and signal is delivered only through kevent queue. +* We simulate successfull delivery notification through this hack: +*/ +if (ret ==1 ) + ret =0 ; +#endif out:return ret; }
@@ -971,6 +986,17 @@ __group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p)* to avoid several races. */ ret =send_signal (sig, info, p, &p->signal->shared_pending); +#ifdef CONFIG_KEVENT_SIGNAL +/* +* Kevent allows to deliver signals through kevent queue, +* it is possible to setup kevent to not deliver +* signal through the usual way, in that case send_signal() +* returns 1 and signal is delivered only through kevent queue. +* We simulate successfull delivery notification through this hack: +*/ +if (ret ==1 ) + ret =0 ; +#endif if (unlikely (ret))return ret;
PATCH 8/8 kevent: Kevent posix timer notifications.
Kevent posix timer notifications.
Simple extensions to POSIX timers which allows
to deliver notification of the timer expiration
through kevent queue.
Example application posix_timer.c can be found
in archive on project homepage.
include/asm-generic/siginfo.h +3 -0
@@ -235,6 +235,7 @@ typedef struct siginfo {#define SIGEV_NONE 1/* other notification: meaningless */ #define SIGEV_THREAD 2/* deliver via thread creation */ #define SIGEV_THREAD_ID 4/* deliver to thread */ +#define SIGEV_KEVENT 8/* deliver through kevent queue */ /* * This works because the alignment is ok on all current architectures
@@ -260,6 +261,8 @@ typedef struct sigevent {void (*_function)(sigval_t );void *_attribute;/* really pthread_attr_t */ } _sigev_thread; + +int kevent_fd; } _sigev_un; } sigevent_t;
include/linux/posix-timers.h +4 -0
@@ -4,6 +4,7 @@#include <linux/spinlock.h> #include <linux/list.h> #include <linux/sched.h> +#include <linux/kevent_storage.h> union cpu_time_count {cputime_t cpu;
@@ -49,6 +50,9 @@ struct k_itimer {sigval_t it_sigev_value;/* value word of sigevent struct */ struct task_struct *it_process;/* process to send signal to */ struct sigqueue *sigq;/* signal queue entry. */ +#ifdef CONFIG_KEVENT_TIMER +struct kevent_storage st; +#endif union { struct { struct hrtimer timer;
kernel/posix-timers.c +168 -48
261 lines not on this page · raw message
Re: [take26-resend1 7/8] kevent: Signal notifications.
In reply to Evgeniy PolyakovHi Evgeniy,
I have used kobject_uevent() to notify userspace about some events. For instance, when memory comsumption reaches a predefined watermark, a signal is sent to userspace to allow applications to free resources. But I am not sure if kobject_uevent() is the more appropriate way for that since if I have many different levels of notifications (using kobject_uevent()) from kernel space to user space, so how the application could know or differentiate from which level of kernel notification the signal was sent from?
The application should perform a specific task according to different type of received notification. So I do not know if the current kernel provides something like that. Do you know any current kernel (2.6.19) implementation for that?
After reading about your Kevent implementation, I guess that your patches are able to do what I need, right? Will it be included in the mainline kernel? Do you have examples about how can I use your socket and/or signal notifications to establish kernel and user space communication?
BR,
Mauricio Lin.
Quoted 236 lines · Evgeniy Polyakov
Signal notifications.
This type of notifications allows to deliver signals through kevent queue. One can find example application
signal.con project homepage.If
KEVENT_SIGNAL_NOMASKbit is set inraw_u64id then signal will be delivered only through queue, otherwise both delivery types are used - old through update of mask of pending signals and through queue.If signal is delivered only through kevent queue mask of pending signals is not updated at all, which is equal to putting signal into blocked mask, but with delivery of that signal through kevent queue.
Signed-off-by Evgeniy Polyakov <[email protected]>include/linux/sched.h +5
-0@@ -80,6 +80,7 @@ struct sched_param {#include <linux/resource.h> #include <linux/timer.h> #include <linux/hrtimer.h> +#include <linux/kevent_storage.h> #include <asm/processor.h> @@ -1013,6 +1014,10 @@ struct task_struct {#ifdef CONFIG_TASK_DELAY_ACCT struct task_delay_info *delays;#endif +#ifdef CONFIG_KEVENT_SIGNAL +struct kevent_storage st; +u32 kevent_signals; +#endif };static inline pid_t process_group (struct task_struct *tsk)kernel/fork.c +8
-0@@ -46,6 +46,7 @@#include <linux/delayacct.h> #include <linux/taskstats_kern.h> #include <linux/random.h> +#include <linux/kevent.h> #include <asm/pgtable.h> #include <asm/pgalloc.h> @@ -115,6 +116,9 @@ void __put_task_struct(struct task_struct *tsk)WARN_ON (atomic_read (&tsk->usage));WARN_ON (tsk == current); +#ifdef CONFIG_KEVENT_SIGNAL +kevent_storage_fini (&tsk->st); +#endif security_task_free (tsk);free_uid (tsk->user);put_group_info (tsk->group_info);@@ -1121,6 +1125,10 @@ static struct task_struct *copy_process(unsigned long clone_flags,if (retval) goto bad_fork_cleanup_namespace; +#ifdef CONFIG_KEVENT_SIGNAL +kevent_storage_init (p, &p->st); +#endif + p->set_child_tid = (clone_flags &CLONE_CHILD_SETTID ) ? child_tidptr :NULL ;/* * Clear TID on mm_release()? kernel/kevent/kevent_signal.c new file +92
-0@@ -0,0 +1,92 @@ +/* +* kevent_signal.c +* +* 2006 Copyright (c) Evgeniy Polyakov <[email protected]> +* All rights reserved. +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation; either version 2 of the License, or +* (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, write to the Free Software +* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA +*/ + +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/file.h> +#include <linux/fs.h> +#include <linux/kevent.h> + +static int kevent_signal_callback (struct kevent *k) +{ +struct task_struct *tsk = k->st->origin; +int sig = k->event.id.raw[0 ]; +int ret =0 ; + +if (sig == tsk->kevent_signals) + ret =1 ; + +if (ret && (k->event.id.raw_u64 &KEVENT_SIGNAL_NOMASK )) + tsk->kevent_signals |=0x80000000 ; + +return ret; +} + +int kevent_signal_enqueue (struct kevent *k) +{ +int err; + + err =kevent_storage_enqueue (¤t->st, k); +if (err) + goto err_out_exit; + +if (k->event.req_flags &KEVENT_REQ_ALWAYS_QUEUE ) { +kevent_requeue (k); + err =0 ; + }else { + err = k->callbacks.callback (k); +if (err) + goto err_out_dequeue; + } + +return err; + +err_out_dequeue: +kevent_storage_dequeue (k->st, k); +err_out_exit: +return err; +} + +int kevent_signal_dequeue (struct kevent *k) +{ +kevent_storage_dequeue (k->st, k); +return 0 ; +} + +int kevent_signal_notify (struct task_struct *tsk,int sig) +{ + tsk->kevent_signals = sig; +kevent_storage_ready (&tsk->st,NULL ,KEVENT_SIGNAL_DELIVERY ); +return (tsk->kevent_signals &0x80000000 ); +} + +static int __init kevent_init_signal (void ) +{ +struct kevent_callbacks sc = { + .callback = &kevent_signal_callback, + .enqueue = &kevent_signal_enqueue, + .dequeue = &kevent_signal_dequeue}; + +return kevent_add_callbacks (&sc,KEVENT_SIGNAL ); +} +module_init (kevent_init_signal);kernel/signal.c +26
-1@@ -23,6 +23,7 @@#include <linux/ptrace.h> #include <linux/signal.h> #include <linux/capability.h> +#include <linux/kevent.h> #include <asm/param.h> #include <asm/uaccess.h> #include <asm/unistd.h> @@ -703,6 +704,9 @@ static int send_signal(int sig, struct siginfo *info, struct task_struct *t, {struct sigqueue * q =NULL ;int ret =0 ; + +if (kevent_signal_notify (t, sig)) +return 1 ;/* * fast-pathed signals for kernel-internal things like SIGSTOP @@ -782,6 +786,17 @@ specific_send_sig_info(int sig, struct siginfo *info, struct task_struct *t) ret =send_signal (sig, info, t, &t->pending);if (!ret && !sigismember (&t->blocked, sig))signal_wake_up (t, sig ==SIGKILL ); +#ifdef CONFIG_KEVENT_SIGNAL +/* +* Kevent allows to deliver signals through kevent queue, +* it is possible to setup kevent to not deliver +* signal through the usual way, in that case send_signal() +* returns 1 and signal is delivered only through kevent queue. +* We simulate successfull delivery notification through this hack: +*/ +if (ret ==1 ) + ret =0 ; +#endif out:return ret; }@@ -971,6 +986,17 @@ __group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p)* to avoid several races. */ ret =send_signal (sig, info, p, &p->signal->shared_pending); +#ifdef CONFIG_KEVENT_SIGNAL +/* +* Kevent allows to deliver signals through kevent queue, +* it is possible to setup kevent to not deliver +* signal through the usual way, in that case send_signal() +* returns 1 and signal is delivered only through kevent queue. +* We simulate successfull delivery notification through this hack: +*/ +if (ret ==1 ) + ret =0 ; +#endif if (unlikely (ret))return ret; -To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: [take26-resend1 7/8] kevent: Signal notifications.
Quoted 1 line · Mauricio Lin
Hi Evgeniy,
Hi Mauricio.
Quoted 14 lines
I have used
kobject_uevent()to notify userspace about some events. For instance, when memory comsumption reaches a predefined watermark, a signal is sent to userspace to allow applications to free resources. But I am not sure ifkobject_uevent()is the more appropriate way for that since if I have many different levels of notifications (usingkobject_uevent()) from kernel space to user space, so how the application could know or differentiate from which level of kernel notification the signal was sent from?The application should perform a specific task according to different type of received notification. So I do not know if the current kernel provides something like that. Do you know any current kernel (2.6.19) implementation for that?
Quoted 5 lines
After reading about your Kevent implementation, I guess that your patches are able to do what I need, right? Will it be included in the mainline kernel? Do you have examples about how can I use your socket and/or signal notifications to establish kernel and user space communication?
I do not know if it will be included or not, but would like to hear an opinion of people added to Cc: list on that point.
I have a lot of examples from trivial applications to real-world web server patched with kevent support. Although some applications might not compile with the latest kevent sources due to interface parameters changes, it is easily fixable looking into other examples.
According to your task - yes, it can be done through kevent - you need to create own kevent subsystem if you plan to use something special, register it with kevent and start commiting events.
But it is easier to use different notification mechanisms for that task:
I suggest using netlink based connector, gennetlink or kobject_uevent, although the latter is not the best choice definitely, and create own protocol embedded into that transports.
Quoted 3 lines
BR,
Mauricio Lin.
Signature · 1 line
Evgeniy Polyakov
Re: [take26-resend1 0/8] kevent: Generic event handling mechanism.
In reply to Evgeniy PolyakovComments:
* [oh, everybody will hate me for saying this, but...] to me, "kevent" implies an internal kernel subsystem. I would rather call it "uevent" or anything else lacking a 'k' prefix.
* I like the absolute timespec (and use of timespec itself)
* more on naming: I think kevent_open would be more natural than kevent_init, since it opens a file descriptor.
* why is KEVENT_MAX not equal to KEVENT_POSIX_TIMER? (perhaps answer this question in a comment, if it is not a mistake)
* Kill all the CONFIG_KEVENT_xxx sub-options, or hide them under CONFIG_EMBEDDED. Application developers should NOT be left wondering whether or support for KEVENT_INODE was compiled into the kernel.
Re: [take26-resend1 0/8] kevent: Generic event handling mechanism.
Quoted 5 lines · Jeff Garzik
Comments:
* [oh, everybody will hate me for saying this, but...] to me, "kevent" implies an internal kernel subsystem. I would rather call it "uevent" or anything else lacking a 'k' prefix.
It is kernel subsystem indeed, which exports some of its part to userspace.
I previously thought that prefix 'k' can only be confused with KDE.
Quoted 1 line
* I like the absolute timespec (and use of timespec itself)
And I do not, but I made them to make at least some progress.
Quoted 2 lines
* more on naming: I think
kevent_openwould be more natural thankevent_init, since it opens a file descriptor.
It is also initializes ring buffer.
Quoted 2 lines
* why is
KEVENT_MAXnot equal toKEVENT_POSIX_TIMER? (perhaps answer this question in a comment, if it is not a mistake)
I check for error number using '>=' and use it as array size, so it is always bigger than the last entry id.
I will add a comment.
Quoted 3 lines
* Kill all the
CONFIG_KEVENT_xxxsub-options, or hide them underCONFIG_EMBEDDED. Application developers should NOT be left wondering whether or support forKEVENT_INODEwas compiled into the kernel.
Ok, I will put them under !CONFIG_EMBEDDED.
Signature · 1 line
Evgeniy Polyakov
Re: [take26-resend1 0/8] kevent: Generic event handling mechanism.
Quoted 8 lines · Evgeniy Polyakov
Quoted 3 lines · Jeff Garzik
Comments:
* [oh, everybody will hate me for saying this, but...] to me, "kevent" implies an internal kernel subsystem. I would rather call it "uevent" or anything else lacking a 'k' prefix.
It is kernel subsystem indeed, which exports some of its part to userspace.
I previously thought that prefix 'k' can only be confused with KDE.
It is a true statement to say "without the kevent subsystem, userspace lacks uevent handling".
And let's be honest, the main consumers of this will be userspace apps, and a few in-kernel users pretending to be userspace apps (kernel threads).
Jeff
cover kevent: Generic event handling mechanism.
In reply to a message not in the archiveGeneric event handling mechanism.
Kevent is a generic subsytem which allows to handle event notifications. It supports both level and edge triggered events. It is similar to poll/epoll in some cases, but it is more scalable, it is faster and allows to work with essentially eny kind of events.
Events are provided into kernel through control syscall and can be read back through ring buffer or using usual syscalls.
Kevent update (i.e. readiness switching) happens directly from internals of the appropriate state machine of the underlying subsytem (like network, filesystem, timer or any other).
Homepage:
http://tservice.net.ru/~s0mbre/old/?section=projects&item=kevent
Documentation page:
http://linux-net.osdl.org/index.php/Kevent
Consider for inclusion.
Changes from 'take26' patchset:
* made kevent visible in config only in case of embedded setup.
* added comment about KEVENT_MAX number.
* spell fix.
Changes from 'take25' patchset:
* use timespec as timeout parameter.
* added high-resolution timer to handle absolute timeouts.
* added flags to waiting and initialization syscalls.
* kevent_commit() has new_uidx parameter.
* kevent_wait() has old_uidx parameter, which, if not equal to u->uidx, results in immediate wakeup (usefull for the case when entries
are added asynchronously from kernel (not supported for now)).
* added interface to mark any event as ready.
* event POSIX timers support.
* return -ENOSYS if there is no registered event type.
* provided file descriptor must be checked for fifo type (spotted by Eric Dumazet).
* signal notifications.
* documentation update.
* lighttpd patch updated (the latest benchmarks with lighttpd patch can be found in blog).
Changes from 'take24' patchset:
* new (old (new)) ring buffer implementation with kernel and user indexes.
* added initialization syscall instead of opening /dev/kevent
* kevent_commit() syscall to commit ring buffer entries
* changed KEVENT_REQ_WAKEUP_ONE flag to KEVENT_REQ_WAKEUP_ALL, kevent wakes only first thread always if that flag is not set
* KEVENT_REQ_ALWAYS_QUEUE flag. If set, kevent will be queued into ready queue instead of copying back to userspace when kevent is ready immediately when it is added.
* lighttpd patch (Hail! Although nothing really outstanding compared to epoll)
Changes from 'take23' patchset:
* kevent PIPE notifications
* KEVENT_REQ_LAST_CHECK flag, which allows to perform last check at dequeueing time
* fixed poll/select notifications (were broken due to tree manipulations)
* made Documentation/kevent.txt look nice in 80-col terminal
* fix for copy_to_user() failure report for the first kevent (Andrew Morton)
* minor function renames
Changes from 'take22' patchset:
* new ring buffer implementation in process' memory
* wakeup-one-thread flag
* edge-triggered behaviour
Changes from 'take21' patchset:
* minor cleanups (different return values, removed unneded variables, whitespaces and so on)
* fixed bug in kevent removal in case when kevent being removed
is the same as overflow_kevent (spotted by Eric Dumazet)
Changes from 'take20' patchset:
* new ring buffer implementation
* removed artificial limit on possible number of kevents
Changes from 'take19' patchset:
* use __init instead of __devinit
* removed 'default N' from config for user statistic
* removed kevent_user_fini() since kevent can not be unloaded
* use KERN_INFO for statistic output
Changes from 'take18' patchset:
* use __init instead of __devinit
* removed 'default N' from config for user statistic
* removed kevent_user_fini() since kevent can not be unloaded
* use KERN_INFO for statistic output
Changes from 'take17' patchset:
* Use RB tree instead of hash table.
At least for a web sever, frequency of addition/deletion of new kevent is comparable with number of search access, i.e. most of the time events are added, accesed only couple of times and then removed, so it justifies RB tree usage over AVL tree, since the latter does have much slower deletion time (max O(log(N)) compared to 3 ops),
although faster search time (1.44*O(log(N)) vs. 2*O(log(N))).
So for kevents I use RB tree for now and later, when my AVL tree implementation is ready, it will be possible to compare them.
* Changed readiness check for socket notifications.
With both above changes it is possible to achieve more than 3380 req/second compared to 2200, sometimes 2500 req/second for epoll() for trivial web-server and httperf client on the same hardware.
It is possible that above kevent limit is due to maximum allowed kevents in a time limit, which is 4096 events.
Changes from 'take16' patchset:
* misc cleanups (__read_mostly, const ...)
* created special macro which is used for mmap size (number of pages) calculation
* export kevent_socket_notify(), since it is used in network protocols which can be built as modules (IPv6 for example)
Changes from 'take15' patchset:
* converted kevent_timer to high-resolution timers, this forces timer API update at
http://linux-net.osdl.org/index.php/Kevent
* use struct ukevent* instead of void * in syscalls (documentation has been updated)
* added warning in kevent_add_ukevent() if ring has broken index (for testing)
Changes from 'take14' patchset:
* added kevent_wait()
This syscall waits until either timeout expires or at least one event becomes ready. It also commits that @num events from @start are processed by userspace and thus can be be removed or rearmed (depending on it's flags). It can be used for commit events read by userspace through mmap interface. Example userspace code (evtest.c) can be found on project's homepage.
* added socket notifications (send/recv/accept)
Changes from 'take13' patchset:
* do not get lock aroung user data check in __kevent_search()
* fail early if there were no registered callbacks for given type of kevent
* trailing whitespace cleanup
Changes from 'take12' patchset:
* remove non-chardev interface for initialization
* use pointer to kevent_mring instead of unsigned longs
* use aligned 64bit type in raw user data (can be used by high-res timer if needed)
* simplified enqueue/dequeue callbacks and kevent initialization
* use nanoseconds for timeout
* put number of milliseconds into timer's return data
* move some definitions into user-visible header
* removed filenames from comments
Changes from 'take11' patchset:
* include missing headers into patchset
* some trivial code cleanups (use goto instead of if/else games and so on)
* some whitespace cleanups
* check for ready_callback() callback before main loop which should save us some ticks
Changes from 'take10' patchset:
* removed non-existent prototypes
* added helper function for kevent_registered_callbacks
* fixed 80 lines comments issues
* added shared between userspace and kernelspace header instead of embedd them in one
* core restructuring to remove forward declarations
* s o m e w h i t e s p a c e c o d y n g s t y l e c l e a n u p
* use vm_insert_page() instead of remap_pfn_range()
Changes from 'take9' patchset:
* fixed ->nopage method
Changes from 'take8' patchset:
* fixed mmap release bug
* use module_init() instead of late_initcall()
* use better structures for timer notifications
Changes from 'take7' patchset:
* new mmap interface (not tested, waiting for other changes to be acked)
- use nopage() method to dynamically substitue pages
- allocate new page for events only when new added kevent requres it
- do not use ugly index dereferencing, use structure instead
- reduced amount of data in the ring (id and flags), maximum 12 pages on x86 per kevent fd
Changes from 'take6' patchset:
* a lot of comments!
* do not use list poisoning for detection of the fact, that entry is in the list
* return number of ready kevents even if copy*user() fails
* strict check for number of kevents in syscall
* use ARRAY_SIZE for array size calculation
* changed superblock magic number
* use SLAB_PANIC instead of direct panic() call
* changed -E* return values
* a lot of small cleanups and indent fixes
Changes from 'take5' patchset:
* removed compilation warnings about unused wariables when lockdep is not turned on
* do not use internal socket structures, use appropriate (exported) wrappers instead
* removed default 1 second timeout
* removed AIO stuff from patchset
Changes from 'take4' patchset:
* use miscdevice instead of chardevice
* comments fixes
Changes from 'take3' patchset:
* removed serializing mutex from kevent_user_wait()
* moved storage list processing to RCU
* removed lockdep screaming - all storage locks are initialized in the same function, so it was
learned
to differentiate between various cases
* remove kevent from storage if is marked as broken after callback
* fixed a typo in mmaped buffer implementation which would end up in wrong index calcualtion
Changes from 'take2' patchset:
* split kevent_finish_user() to locked and unlocked variants
* do not use KEVENT_STAT ifdefs, use inline functions instead
* use array of callbacks of each type instead of each kevent callback initialization
* changed name of ukevent guarding lock
* use only one kevent lock in kevent_user for all hash buckets instead of per-bucket locks
* do not use kevent_user_ctl structure instead provide needed arguments as syscall parameters
* various indent cleanups
* added optimisation, which is aimed to help when a lot of kevents are being copied from
userspace
* mapped buffer (initial) implementation (no userspace yet)
Changes from 'take1' patchset:
- rebased against 2.6.18-git tree
- removed ioctl controlling
- added new syscall kevent_get_events(int fd, unsigned int min_nr, unsigned int max_nr, unsigned int timeout, void __user *buf, unsigned flags)
- use old syscall kevent_ctl for creation/removing, modification and initial kevent initialization
- use mutuxes instead of semaphores
- added file descriptor check and return error if provided descriptor does not match kevent file operations
- various indent fixes
- removed aio_sendfile() declarations.
Thank you.
PATCH 1/8 kevent: Description.
Description.
Documentation/kevent.txt new file +240 -0
@@ -0,0 +1,240 @@ +Description. + +int kevent_init(struct kevent_ring *ring, unsigned int ring_size, + unsigned int flags); + +num - size of the ring buffer in events +ring - pointer to allocated ring buffer +flags - various flags, see KEVENT_FLAGS_* definitions. + +Return value: kevent control file descriptor or negative error value. + + struct kevent_ring + { + unsigned int ring_kidx, ring_over; + struct ukevent event[0]; + } + +ring_kidx - index in the ring buffer where kernel will put new events + when kevent_wait() or kevent_get_events() is called +ring_over - number of overflows of ring_uidx happend from the start. + Overflow counter is used to prevent situation when two threads + are going to free the same events, but one of them was scheduled + away for too long, so ring indexes were wrapped, so when that + thread will be awakened, it will free not those events, which + it suppose to free. + +Example userspace code (ring_buffer.c) can be found on project's homepage. + +Each kevent syscall can be so called cancellation point in glibc, i.e. when +thread has been cancelled in kevent syscall, thread can be safely removed +and no events will be lost, since each syscall (kevent_wait() or +kevent_get_events()) will copy event into special ring buffer, accessible +from other threads or even processes (if shared memory is used). + +When kevent is removed (not dequeued when it is ready, but just removed), +even if it was ready, it is not copied into ring buffer, since if it is +removed, no one cares about it (otherwise user would wait until it becomes +ready and got it through usual way using kevent_get_events() or kevent_wait()) +and thus no need to copy it to the ring buffer. + +------------------------------------------------------------------------------- + + +int kevent_ctl(int fd, unsigned int cmd, unsigned int num, struct ukevent *arg); + +fd - is the file descriptor referring to the kevent queue to manipulate. +It is created by opening "/dev/kevent" char device, which is created with +dynamic minor number and major number assigned for misc devices. + +cmd - is the requested operation. It can be one of the following: + KEVENT_CTL_ADD - add event notification + KEVENT_CTL_REMOVE - remove event notification + KEVENT_CTL_MODIFY - modify existing notification + KEVENT_CTL_READY - mark existing events as ready, if number of events is zero, + it just wakes up parked in syscall thread + +num - number of struct ukevent in the array pointed to by arg +arg - array of struct ukevent + +Return value: + number of events processed or negative error value. + +When called, kevent_ctl will carry out the operation specified in the +cmd parameter. +------------------------------------------------------------------------------- + + int kevent_get_events(int ctl_fd, unsigned int min_nr, unsigned int max_nr, + struct timespec timeout, struct ukevent *buf, unsigned flags); + +ctl_fd - file descriptor referring to the kevent queue +min_nr - minimum number of completed events that kevent_get_events will block + waiting for +max_nr - number of struct ukevent in buf +timeout - time to wait before returning less than min_nr + events. If this is -1, then wait forever. +buf - pointer to an array of struct ukevent. +flags - various flags, see KEVENT_FLAGS_* definitions. + +Return value: + number of events copied or negative error value. + +kevent_get_events will wait timeout milliseconds for at least min_nr completed +events, copying completed struct ukevents to buf and deleting any +KEVENT_REQ_ONESHOT event requests. In nonblocking mode it returns as many +events as possible, but not more than max_nr. In blocking mode it waits until +timeout or if at least min_nr events are ready. + +This function copies event into ring buffer if it was initialized, if ring buffer +is full, KEVENT_RET_COPY_FAILED flag is set in ret_flags field. +------------------------------------------------------------------------------- + + int kevent_wait(int ctl_fd, unsigned int num, unsigned int old_uidx, + struct timespec timeout, unsigned int flags); + +ctl_fd - file descriptor referring to the kevent queue +num - number of processed kevents +old_uidx - the last index user is aware of +timeout - time to wait until there is free space in kevent queue +flags - various flags, see KEVENT_FLAGS_* definitions. + +Return value: + number of events copied into ring buffer or negative error value. + +This syscall waits until either timeout expires or at least one event becomes +ready. It also copies events into special ring buffer. If ring buffer is full, +it waits until there are ready events and then return. +If kevent is one-shot kevent it is removed in this syscall. +If kevent is edge-triggered (KEVENT_REQ_ET flag is set in 'req_flags') it is +requeued in this syscall for performance reasons. +------------------------------------------------------------------------------- + + int kevent_commit(int ctl_fd, unsigned int new_idx, unsigned int over); + +ctl_fd - file descriptor referring to the kevent queue +new_uidx - the last committed kevent +over - overflow count for given $new_idx value + +Return value: + number of committed kevents or negative error value. + +This function commits, i.e. marks as empty, slots in the ring buffer, so +they can be reused when userspace completes that entries processing. + +Overflow counter is used to prevent situation when two threads are going +to free the same events, but one of them was scheduled away for too long, +so ring indexes were wrapped, so when that thread will be awakened, it +will free not those events, which it suppose to free. + +It is possible that returned number of committed events will be smaller than +requested number - it is possible when several threads try to commit the +same events. +------------------------------------------------------------------------------- + +The bulk of the interface is entirely done through the ukevent struct. +It is used to add event requests, modify existing event requests, +specify which event requests to remove, and return completed events. + +struct ukevent contains the following members: + +struct kevent_id id + Id of this request, e.g. socket number, file descriptor and so on +__u32 type + Event type, e.g. KEVENT_SOCK, KEVENT_INODE, KEVENT_TIMER and so on +__u32 event + Event itself, e.g. SOCK_ACCEPT, INODE_CREATED, TIMER_FIRED +__u32 req_flags + Per-event request flags, + + KEVENT_REQ_ONESHOT + event will be removed when it is ready + + KEVENT_REQ_WAKEUP_ALL + Kevent wakes up only first thread interested in given event, + or all threads if this flag is set. + + KEVENT_REQ_ET + Edge Triggered behaviour. It is an optimisation which allows to move + ready and dequeued (i.e. copied to userspace) event to move into set + of interest for given storage (socket, inode and so on) again. It is + very usefull for cases when the same event should be used many times + (like reading from pipe). It is similar to epoll()'s EPOLLET flag. + + KEVENT_REQ_LAST_CHECK + if set allows to perform the last check on kevent (call appropriate + callback) when kevent is marked as ready and has been removed from + ready queue. If it will be confirmed that kevent is ready + (k->callbacks.callback(k) returns true) then kevent will be copied + to userspace, otherwise it will be requeued back to storage. + Second (checking) call is performed with this bit cleared, so callback + can detect when it was called from kevent_storage_ready() - bit is set, + or kevent_dequeue_ready() - bit is cleared. If kevent will be requeued, + bit will be set again. + + KEVENT_REQ_ALWAYS_QUEUE + If this flag is set kevent will be queued into ready queue if it is + ready at enqueue time, otherwise it will be copied back to userspace + and will not be queued into the storage. + +__u32 ret_flags + Per-event return flags + + KEVENT_RET_BROKEN + Kevent is broken + + KEVENT_RET_DONE + Kevent processing was finished successfully + + KEVENT_RET_COPY_FAILED + Kevent was not copied into ring buffer due to some error conditions. + +__u32 ret_data + Event return data. Event originator fills it with anything it likes + (for example timer notifications put number of milliseconds when timer + has fired +union { __u32 user[2]; void *ptr; } + User's data. It is not used, just copied to/from user. The whole structure + is aligned to 8 bytes already, so the last union is aligned properly. + +------------------------------------------------------------------------------- + +Kevent waiting syscall flags. + +KEVENT_FLAGS_ABSTIME - provided timespec parameter contains absolute time, + for example Aug 27, 2194, or time(NULL) + 10. + +------------------------------------------------------------------------------- + +Usage + +For KEVENT_CTL_ADD, all fields relevant to the event type must be filled +(id, type, event, req_flags). +After kevent_ctl(..., KEVENT_CTL_ADD, ...) returns each struct's ret_flags +should be checked to see if the event is already broken or done. + +For KEVENT_CTL_MODIFY, the id, req_flags, and user and event fields must be +set and an existing kevent request must have matching id and user fields. If +match is found, req_flags and event are replaced with the newly supplied +values and requeueing is started, so modified kevent can be checked and +probably marked as ready immediately. If a match can't be found, the +passed in ukevent's ret_flags has KEVENT_RET_BROKEN set. KEVENT_RET_DONE is +always set. + +For KEVENT_CTL_REMOVE, the id and user fields must be set and an existing +kevent request must have matching id and user fields. If a match is found, +the kevent request is removed. If a match can't be found, the passed in +ukevent's ret_flags has KEVENT_RET_BROKEN set. KEVENT_RET_DONE is always set. + +For kevent_get_events, the entire structure is returned. + +------------------------------------------------------------------------------- + +Usage cases + +kevent_timer +struct ukevent should contain following fields: + type - KEVENT_TIMER + event - KEVENT_TIMER_FIRED + req_flags - KEVENT_REQ_ONESHOT if you want to fire that timer only once + id.raw[0] - number of seconds after commit when this timer shout expire + id.raw[0] - additional to number of seconds number of nanoseconds
PATCH 2/8 kevent: Core files.
Core files.
This patch includes core kevent files:
* userspace controlling
* kernelspace interfaces
* initialization
* notification state machines
Some bits of documentation can be found on project's homepage (and links from there):
http://tservice.net.ru/~s0mbre/old/?section=projects&item=kevent
arch/i386/kernel/syscall_table.S +5 -0
@@ -318,3 +318,8 @@ ENTRY(sys_call_table) .long sys_vmsplice .long sys_move_pages .long sys_getcpu + .long sys_kevent_get_events + .long sys_kevent_ctl /* 320 */ + .long sys_kevent_wait + .long sys_kevent_commit + .long sys_kevent_init
arch/x86_64/ia32/ia32entry.S +6 -1
@@ -714,8 +714,13 @@ ia32_sys_call_table: .quad compat_sys_get_robust_list .quad sys_splice .quad sys_sync_file_range - .quad sys_tee + .quad sys_tee /* 315 */ .quad compat_sys_vmsplice .quad compat_sys_move_pages .quad sys_getcpu + .quad sys_kevent_get_events + .quad sys_kevent_ctl /* 320 */ + .quad sys_kevent_wait + .quad sys_kevent_commit + .quad sys_kevent_init ia32_syscall_end:
include/asm-i386/unistd.h +6 -1
@@ -324,10 +324,15 @@#define __NR_vmsplice 316#define __NR_move_pages 317#define __NR_getcpu 318 +#define __NR_kevent_get_events 319 +#define __NR_kevent_ctl 320 +#define __NR_kevent_wait 321 +#define __NR_kevent_commit 322 +#define __NR_kevent_init 323#ifdef __KERNEL__ -#define NR_syscalls 319 +#define NR_syscalls 324#include <linux/err.h> /*
include/asm-x86_64/unistd.h +11 -1
@@ -619,10 +619,20 @@ __SYSCALL(__NR_sync_file_range, sys_sync_file_range) __SYSCALL(__NR_vmsplice, sys_vmsplice)#define __NR_move_pages 279 __SYSCALL(__NR_move_pages, sys_move_pages) +#define __NR_kevent_get_events 280 +__SYSCALL(__NR_kevent_get_events, sys_kevent_get_events) +#define __NR_kevent_ctl 281 +__SYSCALL(__NR_kevent_ctl, sys_kevent_ctl) +#define __NR_kevent_wait 282 +__SYSCALL(__NR_kevent_wait, sys_kevent_wait) +#define __NR_kevent_commit 283 +__SYSCALL(__NR_kevent_commit, sys_kevent_commit) +#define __NR_kevent_init 284 +__SYSCALL(__NR_kevent_init, sys_kevent_init)#ifdef __KERNEL__ -#define __NR_syscall_max __NR_move_pages +#define __NR_syscall_max __NR_kevent_init#include <linux/err.h> #ifndef __NO_STUBS
include/linux/kevent.h new file +238 -0
238 lines not on this page · raw message
include/linux/kevent_storage.h new file +11 -0
@@ -0,0 +1,11 @@ +#ifndef __KEVENT_STORAGE_H +#define __KEVENT_STORAGE_H + +struct kevent_storage +{ +void *origin;/* Originator's pointer, e.g. struct sock or struct file. Can be NULL. */ +struct list_head list;/* List of queued kevents. */ +spinlock_t lock;/* Protects users queue. */ +}; + +#endif /* __KEVENT_STORAGE_H */
include/linux/syscalls.h +9 -0
@@ -54,6 +54,8 @@ struct compat_stat;struct compat_timeval ;struct robust_list_head ;struct getcpu_cache ; +struct ukevent ; +struct kevent_ring ;#include <linux/types.h> #include <linux/aio_abi.h>
@@ -599,4 +601,11 @@ asmlinkage long sys_set_robust_list(struct robust_list_head __user *head, size_t len);asmlinkage long sys_getcpu (unsigned __user *cpu,unsigned __user *node,struct getcpu_cache __user *cache); +asmlinkage long sys_kevent_get_events (int ctl_fd,unsigned int min,unsigned int max, +struct timespec timeout,struct ukevent __user *buf,unsigned flags); +asmlinkage long sys_kevent_ctl (int ctl_fd,unsigned int cmd,unsigned int num,struct ukevent __user *buf); +asmlinkage long sys_kevent_wait (int ctl_fd,unsigned int num,unsigned int old_uidx, +struct timespec timeout,unsigned int flags); +asmlinkage long sys_kevent_commit (int ctl_fd,unsigned int new_uidx,unsigned int over); +asmlinkage long sys_kevent_init (int ctl_fd,struct kevent_ring __user *ring,unsigned int num,unsigned int flags);#endif
include/linux/ukevent.h new file +185 -0
185 lines not on this page · raw message
init/Kconfig +2 -0
@@ -201,6 +201,8 @@ config AUDITSYSCALL such as SELinux. To use audit's filesystem watch feature, please ensure that INOTIFY is configured. +source "kernel/kevent/Kconfig" + config IKCONFIG bool "Kernel .config support" ---help---
kernel/Makefile +1 -0
@@ -47,6 +47,7 @@ obj-$(CONFIG_DETECT_SOFTLOCKUP) += softlockup.o obj-$(CONFIG_GENERIC_HARDIRQS) += irq/ obj-$(CONFIG_SECCOMP) += seccomp.o obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o +obj-$(CONFIG_KEVENT) += kevent/ obj-$(CONFIG_RELAY) += relay.o obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o obj-$(CONFIG_TASKSTATS) += taskstats.o
kernel/kevent/Kconfig new file +65 -0
65 lines not on this page · raw message
kernel/kevent/Makefile new file +6 -0
@@ -0,0 +1,6 @@ +obj-y := kevent.o kevent_user.o +obj-$(CONFIG_KEVENT_TIMER) += kevent_timer.o +obj-$(CONFIG_KEVENT_POLL) += kevent_poll.o +obj-$(CONFIG_KEVENT_SOCKET) += kevent_socket.o +obj-$(CONFIG_KEVENT_PIPE) += kevent_pipe.o +obj-$(CONFIG_KEVENT_SIGNAL) += kevent_signal.o
kernel/kevent/kevent.c new file +247 -0
247 lines not on this page · raw message
kernel/kevent/kevent_user.c new file +1344 -0
1,344 lines not on this page · raw message
kernel/sys_ni.c +6 -0
@@ -122,6 +122,12 @@ cond_syscall(ppc_rtas);cond_syscall (sys_spu_run);cond_syscall (sys_spu_create); +cond_syscall (sys_kevent_get_events); +cond_syscall (sys_kevent_ctl); +cond_syscall (sys_kevent_wait); +cond_syscall (sys_kevent_commit); +cond_syscall (sys_kevent_init); +/* mmu depending weak syscall entries */ cond_syscall (sys_mprotect);cond_syscall (sys_msync);
PATCH 3/8 kevent: poll/select() notifications.
poll/select() notifications.
This patch includes generic poll/select notifications. kevent_poll works simialr to epoll and has the same issues (callback is invoked not from internal state machine of the caller, but through process awake, a lot of allocations and so on).
fs/file_table.c +3 -0
@@ -20,6 +20,7 @@#include <linux/cdev.h> #include <linux/fsnotify.h> #include <linux/sysctl.h> +#include <linux/kevent.h> #include <linux/percpu_counter.h> #include <asm/atomic.h>
@@ -119,6 +120,7 @@ struct file *get_empty_filp(void) f->f_uid = tsk->fsuid; f->f_gid = tsk->fsgid;eventpoll_init_file (f); +kevent_init_file (f);/* f->f_version: 0 */ return f;
@@ -164,6 +166,7 @@ void fastcall __fput(struct file *file)* in the file cleanup chain. */ eventpoll_release (file); +kevent_cleanup_file (file);locks_remove_flock (file);if (file->f_op && file->f_op->release)
include/linux/fs.h +8 -0
@@ -276,6 +276,7 @@ extern int dir_notify_enable;#include <linux/init.h> #include <linux/sched.h> #include <linux/mutex.h> +#include <linux/kevent_storage.h> #include <asm/atomic.h> #include <asm/semaphore.h>
@@ -586,6 +587,10 @@ struct inode {struct mutex inotify_mutex;/* protects the watches list */ #endif +#if definedCONFIG_KEVENT_SOCKET || definedCONFIG_KEVENT_PIPE +struct kevent_storage st; +#endif +unsigned long i_state;unsigned long dirtied_when;/* jiffies of first dirtying */
@@ -739,6 +744,9 @@ struct file {struct list_head f_ep_links;spinlock_t f_ep_lock;#endif /* #ifdef CONFIG_EPOLL */ +#ifdef CONFIG_KEVENT_POLL +struct kevent_storage st; +#endif struct address_space *f_mapping; };extern spinlock_t files_lock;
kernel/kevent/kevent_poll.c new file +232 -0
232 lines not on this page · raw message
PATCH 4/8 kevent: Socket notifications.
Socket notifications.
This patch includes socket send/recv/accept notifications. Using trivial web server based on kevent and this features instead of epoll it's performance increased more than noticebly. More details about various benchmarks and server itself (evserver_kevent.c) can be found on project's homepage.
fs/inode.c +7 -0
@@ -21,6 +21,7 @@#include <linux/cdev.h> #include <linux/bootmem.h> #include <linux/inotify.h> +#include <linux/kevent.h> #include <linux/mount.h> /*
@@ -164,12 +165,18 @@ static struct inode *alloc_inode(struct super_block *sb) } inode->i_private =0 ; inode->i_mapping = mapping; +#if definedCONFIG_KEVENT_SOCKET || definedCONFIG_KEVENT_PIPE +kevent_storage_init (inode, &inode->st); +#endif }return inode; }void destroy_inode (struct inode *inode) { +#if definedCONFIG_KEVENT_SOCKET || definedCONFIG_KEVENT_PIPE +kevent_storage_fini (&inode->st); +#endif BUG_ON (inode_has_buffers (inode));security_inode_free (inode);if (inode->i_sb->s_op->destroy_inode)
include/net/sock.h +17 -15
@@ -48,6 +48,7 @@#include <linux/netdevice.h> #include <linux/skbuff.h> /* struct sk_buff */ #include <linux/security.h> +#include <linux/kevent.h> #include <linux/filter.h>
@@ -450,6 +451,21 @@ static inline int sk_stream_memory_free(struct sock *sk)extern void sk_stream_rfree (struct sk_buff *skb); +struct socket_alloc { +struct socket socket; +struct inode vfs_inode; +}; + +static inline struct socket *SOCKET_I (struct inode *inode) +{ + return &container_of(inode, struct socket_alloc, vfs_inode)->socket; +} + +static inline struct inode *SOCK_INODE (struct socket *socket) +{ + return &container_of(socket, struct socket_alloc, socket)->vfs_inode; +} +static inline void sk_stream_set_owner_r (struct sk_buff *skb,struct sock *sk) { skb->sk = sk;
@@ -477,6 +493,7 @@ static inline void sk_add_backlog(struct sock *sk, struct sk_buff *skb) sk->sk_backlog.tail = skb; } skb->next =NULL ; +kevent_socket_notify (sk,KEVENT_SOCKET_RECV ); }#define sk_wait_event (__sk, __timeo, __condition) \
@@ -679,21 +696,6 @@ static inline struct kiocb *siocb_to_kiocb(struct sock_iocb *si)return si->kiocb; } -struct socket_alloc { -struct socket socket; -struct inode vfs_inode; -}; - -static inline struct socket *SOCKET_I (struct inode *inode) -{ - return &container_of(inode, struct socket_alloc, vfs_inode)->socket; -} - -static inline struct inode *SOCK_INODE (struct socket *socket) -{ - return &container_of(socket, struct socket_alloc, socket)->vfs_inode; -} -extern void __sk_stream_mem_reclaim (struct sock *sk);extern int sk_stream_mem_schedule (struct sock *sk,int size,int kind);
include/net/tcp.h +1 -0
@@ -857,6 +857,7 @@ static inline int tcp_prequeue(struct sock *sk, struct sk_buff *skb) tp->ucopy.memory =0 ; }else if (skb_queue_len (&tp->ucopy.prequeue) ==1 ) {wake_up_interruptible (sk->sk_sleep); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND );if (!inet_csk_ack_scheduled (sk)) inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, (3 * TCP_RTO_MIN) / 4,
kernel/kevent/kevent_socket.c new file +142 -0
142 lines not on this page · raw message
net/core/sock.c +9 -1
@@ -1402,6 +1402,7 @@ static void sock_def_wakeup(struct sock *sk)if (sk->sk_sleep &&waitqueue_active (sk->sk_sleep))wake_up_interruptible_all (sk->sk_sleep);read_unlock (&sk->sk_callback_lock); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND ); }static void sock_def_error_report (struct sock *sk)
@@ -1411,6 +1412,7 @@ static void sock_def_error_report(struct sock *sk)wake_up_interruptible (sk->sk_sleep);sk_wake_async (sk,0 ,POLL_ERR );read_unlock (&sk->sk_callback_lock); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND ); }static void sock_def_readable (struct sock *sk,int len)
@@ -1420,6 +1422,7 @@ static void sock_def_readable(struct sock *sk, int len)wake_up_interruptible (sk->sk_sleep);sk_wake_async (sk,1 ,POLL_IN );read_unlock (&sk->sk_callback_lock); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND ); }static void sock_def_write_space (struct sock *sk)
@@ -1439,6 +1442,7 @@ static void sock_def_write_space(struct sock *sk) }read_unlock (&sk->sk_callback_lock); +kevent_socket_notify (sk,KEVENT_SOCKET_SEND |KEVENT_SOCKET_RECV ); }static void sock_def_destruct (struct sock *sk)
@@ -1489,6 +1493,8 @@ void sock_init_data(struct socket *sock, struct sock *sk) sk->sk_state =TCP_CLOSE ; sk->sk_socket = sock; +kevent_sk_reinit (sk); +sock_set_flag (sk,SOCK_ZAPPED );if (sock)
@@ -1555,8 +1561,10 @@ void fastcall release_sock(struct sock *sk)if (sk->sk_backlog.tail)__release_sock (sk); sk->sk_lock.owner =NULL ; -if (waitqueue_active (&sk->sk_lock.wq)) +if (waitqueue_active (&sk->sk_lock.wq)) {wake_up (&sk->sk_lock.wq); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND ); + }spin_unlock_bh (&sk->sk_lock.slock); }EXPORT_SYMBOL (release_sock);
net/core/stream.c +1 -0
@@ -36,6 +36,7 @@ void sk_stream_write_space(struct sock *sk)wake_up_interruptible (sk->sk_sleep);if (sock->fasync_list && !(sk->sk_shutdown &SEND_SHUTDOWN ))sock_wake_async (sock,2 ,POLL_OUT ); +kevent_socket_notify (sk,KEVENT_SOCKET_SEND |KEVENT_SOCKET_RECV ); } }
net/ipv4/tcp_input.c +1 -0
@@ -3119,6 +3119,7 @@ static void tcp_ofo_queue(struct sock *sk)__skb_unlink (skb, &tp->out_of_order_queue);__skb_queue_tail (&sk->sk_receive_queue, skb); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV ); tp->rcv_nxt =TCP_SKB_CB (skb)->end_seq;if (skb->h.th->fin)tcp_fin (skb, sk, skb->h.th);
net/ipv4/tcp_ipv4.c +2 -0
@@ -61,6 +61,7 @@#include <linux/jhash.h> #include <linux/init.h> #include <linux/times.h> +#include <linux/kevent.h> #include <net/icmp.h> #include <net/inet_hashtables.h>
@@ -870,6 +871,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)reqsk_free (req); }else {inet_csk_reqsk_queue_hash_add (sk, req,TCP_TIMEOUT_INIT ); +kevent_socket_notify (sk,KEVENT_SOCKET_ACCEPT ); }return 0 ;
net/socket.c +3 -0
@@ -85,6 +85,7 @@#include <linux/kmod.h> #include <linux/audit.h> #include <linux/wireless.h> +#include <linux/kevent.h> #include <asm/uaccess.h> #include <asm/unistd.h>
@@ -490,6 +491,8 @@ static struct socket *sock_alloc(void) inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; +kevent_socket_reinit (sock); +get_cpu_var (sockets_in_use)++;put_cpu_var (sockets_in_use);return sock;
PATCH 5/8 kevent: Timer notifications.
Timer notifications.
Timer notifications can be used for fine grained per-process time management, since interval timers are very inconvenient to use, and they are limited.
This subsystem uses high-resolution timers.
id.raw[0] is used as number of seconds
id.raw[1] is used as number of nanoseconds
kernel/kevent/kevent_timer.c new file +112 -0
@@ -0,0 +1,112 @@ +/* +* 2006 Copyright (c) Evgeniy Polyakov <[email protected]> +* All rights reserved. +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation; either version 2 of the License, or +* (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, write to the Free Software +* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA +*/ + +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/list.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/hrtimer.h> +#include <linux/jiffies.h> +#include <linux/kevent.h> + +struct kevent_timer +{ +struct hrtimer ktimer; +struct kevent_storage ktimer_storage; +struct kevent *ktimer_event; +}; + +static int kevent_timer_func (struct hrtimer *timer) +{ + struct kevent_timer *t = container_of(timer, struct kevent_timer, ktimer); +struct kevent *k = t->ktimer_event; + +kevent_storage_ready (&t->ktimer_storage,NULL ,KEVENT_MASK_ALL ); +hrtimer_forward (timer, timer->base->softirq_time, +ktime_set (k->event.id.raw[0 ], k->event.id.raw[1 ])); +return HRTIMER_RESTART ; +} + +static struct lock_class_key kevent_timer_key; + +static int kevent_timer_enqueue (struct kevent *k) +{ +int err; +struct kevent_timer *t; + + t =kmalloc (sizeof (struct kevent_timer ),GFP_KERNEL ); +if (!t) +return -ENOMEM ; + +hrtimer_init (&t->ktimer,CLOCK_MONOTONIC ,HRTIMER_REL ); + t->ktimer.expires =ktime_set (k->event.id.raw[0 ], k->event.id.raw[1 ]); + t->ktimer.function = kevent_timer_func; + t->ktimer_event = k; + + err =kevent_storage_init (&t->ktimer, &t->ktimer_storage); +if (err) + goto err_out_free; +lockdep_set_class (&t->ktimer_storage.lock, &kevent_timer_key); + + err =kevent_storage_enqueue (&t->ktimer_storage, k); +if (err) + goto err_out_st_fini; + +hrtimer_start (&t->ktimer, t->ktimer.expires,HRTIMER_REL ); + +return 0 ; + +err_out_st_fini: +kevent_storage_fini (&t->ktimer_storage); +err_out_free: +kfree (t); + +return err; +} + +static int kevent_timer_dequeue (struct kevent *k) +{ +struct kevent_storage *st = k->st; + struct kevent_timer *t = container_of(st, struct kevent_timer, ktimer_storage); + +hrtimer_cancel (&t->ktimer); +kevent_storage_dequeue (st, k); +kfree (t); + +return 0 ; +} + +static int kevent_timer_callback (struct kevent *k) +{ + k->event.ret_data[0 ] =jiffies_to_msecs (jiffies); +return 1 ; +} + +static int __init kevent_init_timer (void ) +{ +struct kevent_callbacks tc = { + .callback = &kevent_timer_callback, + .enqueue = &kevent_timer_enqueue, + .dequeue = &kevent_timer_dequeue}; + +return kevent_add_callbacks (&tc,KEVENT_TIMER ); +} +module_init (kevent_init_timer); +
PATCH 6/8 kevent: Pipe notifications.
Pipe notifications.
fs/pipe.c +6 -0
@@ -16,6 +16,7 @@#include <linux/uio.h> #include <linux/highmem.h> #include <linux/pagemap.h> +#include <linux/kevent.h> #include <asm/uaccess.h> #include <asm/ioctls.h>
@@ -312,6 +313,7 @@ redo:break ; }if (do_wakeup) { +kevent_pipe_notify (inode,KEVENT_SOCKET_SEND );wake_up_interruptible_sync (&pipe->wait);kill_fasync (&pipe->fasync_writers,SIGIO ,POLL_OUT ); }
@@ -321,6 +323,7 @@ redo:/* Signal writers asynchronously that there is more room. */ if (do_wakeup) { +kevent_pipe_notify (inode,KEVENT_SOCKET_SEND );wake_up_interruptible (&pipe->wait);kill_fasync (&pipe->fasync_writers,SIGIO ,POLL_OUT ); }
@@ -490,6 +493,7 @@ redo2:break ; }if (do_wakeup) { +kevent_pipe_notify (inode,KEVENT_SOCKET_RECV );wake_up_interruptible_sync (&pipe->wait);kill_fasync (&pipe->fasync_readers,SIGIO ,POLL_IN ); do_wakeup =0 ;
@@ -501,6 +505,7 @@ redo2: out:mutex_unlock (&inode->i_mutex);if (do_wakeup) { +kevent_pipe_notify (inode,KEVENT_SOCKET_RECV );wake_up_interruptible (&pipe->wait);kill_fasync (&pipe->fasync_readers,SIGIO ,POLL_IN ); }
@@ -605,6 +610,7 @@ pipe_release(struct inode *inode, int decr, int decw)free_pipe_info (inode); }else {wake_up_interruptible (&pipe->wait); +kevent_pipe_notify (inode,KEVENT_SOCKET_SEND |KEVENT_SOCKET_RECV );kill_fasync (&pipe->fasync_readers,SIGIO ,POLL_IN );kill_fasync (&pipe->fasync_writers,SIGIO ,POLL_OUT ); }
kernel/kevent/kevent_pipe.c new file +121 -0
121 lines not on this page · raw message
PATCH 7/8 kevent: Signal notifications.
Signal notifications.
This type of notifications allows to deliver signals through kevent queue. One can find example application signal.c on project homepage.
If KEVENT_SIGNAL_NOMASK bit is set in raw_u64 id then signal will be delivered only through queue, otherwise both delivery types are used - old through update of mask of pending signals and through queue.
If signal is delivered only through kevent queue mask of pending signals is not updated at all, which is equal to putting signal into blocked mask, but with delivery of that signal through kevent queue.
include/linux/sched.h +5 -0
@@ -80,6 +80,7 @@ struct sched_param {#include <linux/resource.h> #include <linux/timer.h> #include <linux/hrtimer.h> +#include <linux/kevent_storage.h> #include <asm/processor.h>
@@ -1013,6 +1014,10 @@ struct task_struct {#ifdef CONFIG_TASK_DELAY_ACCT struct task_delay_info *delays;#endif +#ifdef CONFIG_KEVENT_SIGNAL +struct kevent_storage st; +u32 kevent_signals; +#endif };static inline pid_t process_group (struct task_struct *tsk)
kernel/fork.c +8 -0
@@ -46,6 +46,7 @@#include <linux/delayacct.h> #include <linux/taskstats_kern.h> #include <linux/random.h> +#include <linux/kevent.h> #include <asm/pgtable.h> #include <asm/pgalloc.h>
@@ -115,6 +116,9 @@ void __put_task_struct(struct task_struct *tsk)WARN_ON (atomic_read (&tsk->usage));WARN_ON (tsk == current); +#ifdef CONFIG_KEVENT_SIGNAL +kevent_storage_fini (&tsk->st); +#endif security_task_free (tsk);free_uid (tsk->user);put_group_info (tsk->group_info);
@@ -1121,6 +1125,10 @@ static struct task_struct *copy_process(unsigned long clone_flags,if (retval) goto bad_fork_cleanup_namespace; +#ifdef CONFIG_KEVENT_SIGNAL +kevent_storage_init (p, &p->st); +#endif + p->set_child_tid = (clone_flags &CLONE_CHILD_SETTID ) ? child_tidptr :NULL ;/* * Clear TID on mm_release()?
kernel/kevent/kevent_signal.c new file +92 -0
92 lines not on this page · raw message
kernel/signal.c +26 -0
@@ -23,6 +23,7 @@#include <linux/ptrace.h> #include <linux/signal.h> #include <linux/capability.h> +#include <linux/kevent.h> #include <asm/param.h> #include <asm/uaccess.h> #include <asm/unistd.h>
@@ -703,6 +704,9 @@ static int send_signal(int sig, struct siginfo *info, struct task_struct *t, {struct sigqueue * q =NULL ;int ret =0 ; + +if (kevent_signal_notify (t, sig)) +return 1 ;/* * fast-pathed signals for kernel-internal things like SIGSTOP
@@ -782,6 +786,17 @@ specific_send_sig_info(int sig, struct siginfo *info, struct task_struct *t) ret =send_signal (sig, info, t, &t->pending);if (!ret && !sigismember (&t->blocked, sig))signal_wake_up (t, sig ==SIGKILL ); +#ifdef CONFIG_KEVENT_SIGNAL +/* +* Kevent allows to deliver signals through kevent queue, +* it is possible to setup kevent to not deliver +* signal through the usual way, in that case send_signal() +* returns 1 and signal is delivered only through kevent queue. +* We simulate successfull delivery notification through this hack: +*/ +if (ret ==1 ) + ret =0 ; +#endif out:return ret; }
@@ -971,6 +986,17 @@ __group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p)* to avoid several races. */ ret =send_signal (sig, info, p, &p->signal->shared_pending); +#ifdef CONFIG_KEVENT_SIGNAL +/* +* Kevent allows to deliver signals through kevent queue, +* it is possible to setup kevent to not deliver +* signal through the usual way, in that case send_signal() +* returns 1 and signal is delivered only through kevent queue. +* We simulate successfull delivery notification through this hack: +*/ +if (ret ==1 ) + ret =0 ; +#endif if (unlikely (ret))return ret;
PATCH 8/8 kevent: Kevent posix timer notifications.
Kevent posix timer notifications.
Simple extensions to POSIX timers which allows
to deliver notification of the timer expiration
through kevent queue.
Example application posix_timer.c can be found
in archive on project homepage.
include/asm-generic/siginfo.h +3 -0
@@ -235,6 +235,7 @@ typedef struct siginfo {#define SIGEV_NONE 1/* other notification: meaningless */ #define SIGEV_THREAD 2/* deliver via thread creation */ #define SIGEV_THREAD_ID 4/* deliver to thread */ +#define SIGEV_KEVENT 8/* deliver through kevent queue */ /* * This works because the alignment is ok on all current architectures
@@ -260,6 +261,8 @@ typedef struct sigevent {void (*_function)(sigval_t );void *_attribute;/* really pthread_attr_t */ } _sigev_thread; + +int kevent_fd; } _sigev_un; } sigevent_t;
include/linux/posix-timers.h +4 -0
@@ -4,6 +4,7 @@#include <linux/spinlock.h> #include <linux/list.h> #include <linux/sched.h> +#include <linux/kevent_storage.h> union cpu_time_count {cputime_t cpu;
@@ -49,6 +50,9 @@ struct k_itimer {sigval_t it_sigev_value;/* value word of sigevent struct */ struct task_struct *it_process;/* process to send signal to */ struct sigqueue *sigq;/* signal queue entry. */ +#ifdef CONFIG_KEVENT_TIMER +struct kevent_storage st; +#endif union { struct { struct hrtimer timer;
kernel/posix-timers.c +168 -48
261 lines not on this page · raw message
cover kevent: Generic event handling mechanism.
In reply to a message not in the archiveGeneric event handling mechanism.
Kevent is a generic subsytem which allows to handle event notifications. It supports both level and edge triggered events. It is similar to poll/epoll in some cases, but it is more scalable, it is faster and allows to work with essentially eny kind of events.
Events are provided into kernel through control syscall and can be read back through ring buffer or using usual syscalls.
Kevent update (i.e. readiness switching) happens directly from internals of the appropriate state machine of the underlying subsytem (like network, filesystem, timer or any other).
Homepage:
http://tservice.net.ru/~s0mbre/old/?section=projects&item=kevent
Documentation page:
http://linux-net.osdl.org/index.php/Kevent
Consider for inclusion.
New benchmark, which can be a hoax though, can be found at
http://tservice.net.ru/~s0mbre/blog/2006/11/30#2006_11_30
where kevent on amd64 with 1gb of ram can handle more than 7200 events per second with 8000 requests concurrency with 'ab' benchmark and lighttpd. Although I tought it should not be published due to possible errors, I decided to send it for review.
With this release I start 3 days resending timeout - i.e. each third day I will send either new version (if something new was requested and agreed to be implemented) or resending with back counter started from three. When back counter hits zero after three resending I consider there is no interest in subsystem and I will stop further sending.
Thanks for understanding and your time.
Changes from 'take27' patchset:
* made kevent default yes in non embedded case.
* added falgs to callback structures - currently used to check if kevent can be requested from kernelspace only (posix timers) or userspace (all others)
Changes from 'take26' patchset:
* made kevent visible in config only in case of embedded setup.
* added comment about KEVENT_MAX number.
* spell fix.
Changes from 'take25' patchset:
* use timespec as timeout parameter.
* added high-resolution timer to handle absolute timeouts.
* added flags to waiting and initialization syscalls.
* kevent_commit() has new_uidx parameter.
* kevent_wait() has old_uidx parameter, which, if not equal to u->uidx, results in immediate wakeup (usefull for the case when entries
are added asynchronously from kernel (not supported for now)).
* added interface to mark any event as ready.
* event POSIX timers support.
* return -ENOSYS if there is no registered event type.
* provided file descriptor must be checked for fifo type (spotted by Eric Dumazet).
* signal notifications.
* documentation update.
* lighttpd patch updated (the latest benchmarks with lighttpd patch can be found in blog).
Changes from 'take24' patchset:
* new (old (new)) ring buffer implementation with kernel and user indexes.
* added initialization syscall instead of opening /dev/kevent
* kevent_commit() syscall to commit ring buffer entries
* changed KEVENT_REQ_WAKEUP_ONE flag to KEVENT_REQ_WAKEUP_ALL, kevent wakes only first thread always if that flag is not set
* KEVENT_REQ_ALWAYS_QUEUE flag. If set, kevent will be queued into ready queue instead of copying back to userspace when kevent is ready immediately when it is added.
* lighttpd patch (Hail! Although nothing really outstanding compared to epoll)
Changes from 'take23' patchset:
* kevent PIPE notifications
* KEVENT_REQ_LAST_CHECK flag, which allows to perform last check at dequeueing time
* fixed poll/select notifications (were broken due to tree manipulations)
* made Documentation/kevent.txt look nice in 80-col terminal
* fix for copy_to_user() failure report for the first kevent (Andrew Morton)
* minor function renames
Changes from 'take22' patchset:
* new ring buffer implementation in process' memory
* wakeup-one-thread flag
* edge-triggered behaviour
Changes from 'take21' patchset:
* minor cleanups (different return values, removed unneded variables, whitespaces and so on)
* fixed bug in kevent removal in case when kevent being removed
is the same as overflow_kevent (spotted by Eric Dumazet)
Changes from 'take20' patchset:
* new ring buffer implementation
* removed artificial limit on possible number of kevents
Changes from 'take19' patchset:
* use __init instead of __devinit
* removed 'default N' from config for user statistic
* removed kevent_user_fini() since kevent can not be unloaded
* use KERN_INFO for statistic output
Changes from 'take18' patchset:
* use __init instead of __devinit
* removed 'default N' from config for user statistic
* removed kevent_user_fini() since kevent can not be unloaded
* use KERN_INFO for statistic output
Changes from 'take17' patchset:
* Use RB tree instead of hash table.
At least for a web sever, frequency of addition/deletion of new kevent is comparable with number of search access, i.e. most of the time events are added, accesed only couple of times and then removed, so it justifies RB tree usage over AVL tree, since the latter does have much slower deletion time (max O(log(N)) compared to 3 ops),
although faster search time (1.44*O(log(N)) vs. 2*O(log(N))).
So for kevents I use RB tree for now and later, when my AVL tree implementation is ready, it will be possible to compare them.
* Changed readiness check for socket notifications.
With both above changes it is possible to achieve more than 3380 req/second compared to 2200, sometimes 2500 req/second for epoll() for trivial web-server and httperf client on the same hardware.
It is possible that above kevent limit is due to maximum allowed kevents in a time limit, which is 4096 events.
Changes from 'take16' patchset:
* misc cleanups (__read_mostly, const ...)
* created special macro which is used for mmap size (number of pages) calculation
* export kevent_socket_notify(), since it is used in network protocols which can be built as modules (IPv6 for example)
Changes from 'take15' patchset:
* converted kevent_timer to high-resolution timers, this forces timer API update at
http://linux-net.osdl.org/index.php/Kevent
* use struct ukevent* instead of void * in syscalls (documentation has been updated)
* added warning in kevent_add_ukevent() if ring has broken index (for testing)
Changes from 'take14' patchset:
* added kevent_wait()
This syscall waits until either timeout expires or at least one event becomes ready. It also commits that @num events from @start are processed by userspace and thus can be be removed or rearmed (depending on it's flags). It can be used for commit events read by userspace through mmap interface. Example userspace code (evtest.c) can be found on project's homepage.
* added socket notifications (send/recv/accept)
Changes from 'take13' patchset:
* do not get lock aroung user data check in __kevent_search()
* fail early if there were no registered callbacks for given type of kevent
* trailing whitespace cleanup
Changes from 'take12' patchset:
* remove non-chardev interface for initialization
* use pointer to kevent_mring instead of unsigned longs
* use aligned 64bit type in raw user data (can be used by high-res timer if needed)
* simplified enqueue/dequeue callbacks and kevent initialization
* use nanoseconds for timeout
* put number of milliseconds into timer's return data
* move some definitions into user-visible header
* removed filenames from comments
Changes from 'take11' patchset:
* include missing headers into patchset
* some trivial code cleanups (use goto instead of if/else games and so on)
* some whitespace cleanups
* check for ready_callback() callback before main loop which should save us some ticks
Changes from 'take10' patchset:
* removed non-existent prototypes
* added helper function for kevent_registered_callbacks
* fixed 80 lines comments issues
* added shared between userspace and kernelspace header instead of embedd them in one
* core restructuring to remove forward declarations
* s o m e w h i t e s p a c e c o d y n g s t y l e c l e a n u p
* use vm_insert_page() instead of remap_pfn_range()
Changes from 'take9' patchset:
* fixed ->nopage method
Changes from 'take8' patchset:
* fixed mmap release bug
* use module_init() instead of late_initcall()
* use better structures for timer notifications
Changes from 'take7' patchset:
* new mmap interface (not tested, waiting for other changes to be acked)
- use nopage() method to dynamically substitue pages
- allocate new page for events only when new added kevent requres it
- do not use ugly index dereferencing, use structure instead
- reduced amount of data in the ring (id and flags), maximum 12 pages on x86 per kevent fd
Changes from 'take6' patchset:
* a lot of comments!
* do not use list poisoning for detection of the fact, that entry is in the list
* return number of ready kevents even if copy*user() fails
* strict check for number of kevents in syscall
* use ARRAY_SIZE for array size calculation
* changed superblock magic number
* use SLAB_PANIC instead of direct panic() call
* changed -E* return values
* a lot of small cleanups and indent fixes
Changes from 'take5' patchset:
* removed compilation warnings about unused wariables when lockdep is not turned on
* do not use internal socket structures, use appropriate (exported) wrappers instead
* removed default 1 second timeout
* removed AIO stuff from patchset
Changes from 'take4' patchset:
* use miscdevice instead of chardevice
* comments fixes
Changes from 'take3' patchset:
* removed serializing mutex from kevent_user_wait()
* moved storage list processing to RCU
* removed lockdep screaming - all storage locks are initialized in the same function, so it was
learned
to differentiate between various cases
* remove kevent from storage if is marked as broken after callback
* fixed a typo in mmaped buffer implementation which would end up in wrong index calcualtion
Changes from 'take2' patchset:
* split kevent_finish_user() to locked and unlocked variants
* do not use KEVENT_STAT ifdefs, use inline functions instead
* use array of callbacks of each type instead of each kevent callback initialization
* changed name of ukevent guarding lock
* use only one kevent lock in kevent_user for all hash buckets instead of per-bucket locks
* do not use kevent_user_ctl structure instead provide needed arguments as syscall parameters
* various indent cleanups
* added optimisation, which is aimed to help when a lot of kevents are being copied from
userspace
* mapped buffer (initial) implementation (no userspace yet)
Changes from 'take1' patchset:
- rebased against 2.6.18-git tree
- removed ioctl controlling
- added new syscall kevent_get_events(int fd, unsigned int min_nr, unsigned int max_nr, unsigned int timeout, void __user *buf, unsigned flags)
- use old syscall kevent_ctl for creation/removing, modification and initial kevent initialization
- use mutuxes instead of semaphores
- added file descriptor check and return error if provided descriptor does not match kevent file operations
- various indent fixes
- removed aio_sendfile() declarations.
Thank you.
PATCH 1/8 kevent: Description.
Description.
Documentation/kevent.txt new file +240 -0
@@ -0,0 +1,240 @@ +Description. + +int kevent_init(struct kevent_ring *ring, unsigned int ring_size, + unsigned int flags); + +num - size of the ring buffer in events +ring - pointer to allocated ring buffer +flags - various flags, see KEVENT_FLAGS_* definitions. + +Return value: kevent control file descriptor or negative error value. + + struct kevent_ring + { + unsigned int ring_kidx, ring_over; + struct ukevent event[0]; + } + +ring_kidx - index in the ring buffer where kernel will put new events + when kevent_wait() or kevent_get_events() is called +ring_over - number of overflows of ring_uidx happend from the start. + Overflow counter is used to prevent situation when two threads + are going to free the same events, but one of them was scheduled + away for too long, so ring indexes were wrapped, so when that + thread will be awakened, it will free not those events, which + it suppose to free. + +Example userspace code (ring_buffer.c) can be found on project's homepage. + +Each kevent syscall can be so called cancellation point in glibc, i.e. when +thread has been cancelled in kevent syscall, thread can be safely removed +and no events will be lost, since each syscall (kevent_wait() or +kevent_get_events()) will copy event into special ring buffer, accessible +from other threads or even processes (if shared memory is used). + +When kevent is removed (not dequeued when it is ready, but just removed), +even if it was ready, it is not copied into ring buffer, since if it is +removed, no one cares about it (otherwise user would wait until it becomes +ready and got it through usual way using kevent_get_events() or kevent_wait()) +and thus no need to copy it to the ring buffer. + +------------------------------------------------------------------------------- + + +int kevent_ctl(int fd, unsigned int cmd, unsigned int num, struct ukevent *arg); + +fd - is the file descriptor referring to the kevent queue to manipulate. +It is created by opening "/dev/kevent" char device, which is created with +dynamic minor number and major number assigned for misc devices. + +cmd - is the requested operation. It can be one of the following: + KEVENT_CTL_ADD - add event notification + KEVENT_CTL_REMOVE - remove event notification + KEVENT_CTL_MODIFY - modify existing notification + KEVENT_CTL_READY - mark existing events as ready, if number of events is zero, + it just wakes up parked in syscall thread + +num - number of struct ukevent in the array pointed to by arg +arg - array of struct ukevent + +Return value: + number of events processed or negative error value. + +When called, kevent_ctl will carry out the operation specified in the +cmd parameter. +------------------------------------------------------------------------------- + + int kevent_get_events(int ctl_fd, unsigned int min_nr, unsigned int max_nr, + struct timespec timeout, struct ukevent *buf, unsigned flags); + +ctl_fd - file descriptor referring to the kevent queue +min_nr - minimum number of completed events that kevent_get_events will block + waiting for +max_nr - number of struct ukevent in buf +timeout - time to wait before returning less than min_nr + events. If this is -1, then wait forever. +buf - pointer to an array of struct ukevent. +flags - various flags, see KEVENT_FLAGS_* definitions. + +Return value: + number of events copied or negative error value. + +kevent_get_events will wait timeout milliseconds for at least min_nr completed +events, copying completed struct ukevents to buf and deleting any +KEVENT_REQ_ONESHOT event requests. In nonblocking mode it returns as many +events as possible, but not more than max_nr. In blocking mode it waits until +timeout or if at least min_nr events are ready. + +This function copies event into ring buffer if it was initialized, if ring buffer +is full, KEVENT_RET_COPY_FAILED flag is set in ret_flags field. +------------------------------------------------------------------------------- + + int kevent_wait(int ctl_fd, unsigned int num, unsigned int old_uidx, + struct timespec timeout, unsigned int flags); + +ctl_fd - file descriptor referring to the kevent queue +num - number of processed kevents +old_uidx - the last index user is aware of +timeout - time to wait until there is free space in kevent queue +flags - various flags, see KEVENT_FLAGS_* definitions. + +Return value: + number of events copied into ring buffer or negative error value. + +This syscall waits until either timeout expires or at least one event becomes +ready. It also copies events into special ring buffer. If ring buffer is full, +it waits until there are ready events and then return. +If kevent is one-shot kevent it is removed in this syscall. +If kevent is edge-triggered (KEVENT_REQ_ET flag is set in 'req_flags') it is +requeued in this syscall for performance reasons. +------------------------------------------------------------------------------- + + int kevent_commit(int ctl_fd, unsigned int new_idx, unsigned int over); + +ctl_fd - file descriptor referring to the kevent queue +new_uidx - the last committed kevent +over - overflow count for given $new_idx value + +Return value: + number of committed kevents or negative error value. + +This function commits, i.e. marks as empty, slots in the ring buffer, so +they can be reused when userspace completes that entries processing. + +Overflow counter is used to prevent situation when two threads are going +to free the same events, but one of them was scheduled away for too long, +so ring indexes were wrapped, so when that thread will be awakened, it +will free not those events, which it suppose to free. + +It is possible that returned number of committed events will be smaller than +requested number - it is possible when several threads try to commit the +same events. +------------------------------------------------------------------------------- + +The bulk of the interface is entirely done through the ukevent struct. +It is used to add event requests, modify existing event requests, +specify which event requests to remove, and return completed events. + +struct ukevent contains the following members: + +struct kevent_id id + Id of this request, e.g. socket number, file descriptor and so on +__u32 type + Event type, e.g. KEVENT_SOCK, KEVENT_INODE, KEVENT_TIMER and so on +__u32 event + Event itself, e.g. SOCK_ACCEPT, INODE_CREATED, TIMER_FIRED +__u32 req_flags + Per-event request flags, + + KEVENT_REQ_ONESHOT + event will be removed when it is ready + + KEVENT_REQ_WAKEUP_ALL + Kevent wakes up only first thread interested in given event, + or all threads if this flag is set. + + KEVENT_REQ_ET + Edge Triggered behaviour. It is an optimisation which allows to move + ready and dequeued (i.e. copied to userspace) event to move into set + of interest for given storage (socket, inode and so on) again. It is + very usefull for cases when the same event should be used many times + (like reading from pipe). It is similar to epoll()'s EPOLLET flag. + + KEVENT_REQ_LAST_CHECK + if set allows to perform the last check on kevent (call appropriate + callback) when kevent is marked as ready and has been removed from + ready queue. If it will be confirmed that kevent is ready + (k->callbacks.callback(k) returns true) then kevent will be copied + to userspace, otherwise it will be requeued back to storage. + Second (checking) call is performed with this bit cleared, so callback + can detect when it was called from kevent_storage_ready() - bit is set, + or kevent_dequeue_ready() - bit is cleared. If kevent will be requeued, + bit will be set again. + + KEVENT_REQ_ALWAYS_QUEUE + If this flag is set kevent will be queued into ready queue if it is + ready at enqueue time, otherwise it will be copied back to userspace + and will not be queued into the storage. + +__u32 ret_flags + Per-event return flags + + KEVENT_RET_BROKEN + Kevent is broken + + KEVENT_RET_DONE + Kevent processing was finished successfully + + KEVENT_RET_COPY_FAILED + Kevent was not copied into ring buffer due to some error conditions. + +__u32 ret_data + Event return data. Event originator fills it with anything it likes + (for example timer notifications put number of milliseconds when timer + has fired +union { __u32 user[2]; void *ptr; } + User's data. It is not used, just copied to/from user. The whole structure + is aligned to 8 bytes already, so the last union is aligned properly. + +------------------------------------------------------------------------------- + +Kevent waiting syscall flags. + +KEVENT_FLAGS_ABSTIME - provided timespec parameter contains absolute time, + for example Aug 27, 2194, or time(NULL) + 10. + +------------------------------------------------------------------------------- + +Usage + +For KEVENT_CTL_ADD, all fields relevant to the event type must be filled +(id, type, event, req_flags). +After kevent_ctl(..., KEVENT_CTL_ADD, ...) returns each struct's ret_flags +should be checked to see if the event is already broken or done. + +For KEVENT_CTL_MODIFY, the id, req_flags, and user and event fields must be +set and an existing kevent request must have matching id and user fields. If +match is found, req_flags and event are replaced with the newly supplied +values and requeueing is started, so modified kevent can be checked and +probably marked as ready immediately. If a match can't be found, the +passed in ukevent's ret_flags has KEVENT_RET_BROKEN set. KEVENT_RET_DONE is +always set. + +For KEVENT_CTL_REMOVE, the id and user fields must be set and an existing +kevent request must have matching id and user fields. If a match is found, +the kevent request is removed. If a match can't be found, the passed in +ukevent's ret_flags has KEVENT_RET_BROKEN set. KEVENT_RET_DONE is always set. + +For kevent_get_events, the entire structure is returned. + +------------------------------------------------------------------------------- + +Usage cases + +kevent_timer +struct ukevent should contain following fields: + type - KEVENT_TIMER + event - KEVENT_TIMER_FIRED + req_flags - KEVENT_REQ_ONESHOT if you want to fire that timer only once + id.raw[0] - number of seconds after commit when this timer shout expire + id.raw[0] - additional to number of seconds number of nanoseconds
PATCH 2/8 kevent: Core files.
Core files.
This patch includes core kevent files:
* userspace controlling
* kernelspace interfaces
* initialization
* notification state machines
Some bits of documentation can be found on project's homepage (and links from there):
http://tservice.net.ru/~s0mbre/old/?section=projects&item=kevent
arch/i386/kernel/syscall_table.S +5 -0
@@ -318,3 +318,8 @@ ENTRY(sys_call_table) .long sys_vmsplice .long sys_move_pages .long sys_getcpu + .long sys_kevent_get_events + .long sys_kevent_ctl /* 320 */ + .long sys_kevent_wait + .long sys_kevent_commit + .long sys_kevent_init
arch/x86_64/ia32/ia32entry.S +6 -1
@@ -714,8 +714,13 @@ ia32_sys_call_table: .quad compat_sys_get_robust_list .quad sys_splice .quad sys_sync_file_range - .quad sys_tee + .quad sys_tee /* 315 */ .quad compat_sys_vmsplice .quad compat_sys_move_pages .quad sys_getcpu + .quad sys_kevent_get_events + .quad sys_kevent_ctl /* 320 */ + .quad sys_kevent_wait + .quad sys_kevent_commit + .quad sys_kevent_init ia32_syscall_end:
include/asm-i386/unistd.h +6 -1
@@ -324,10 +324,15 @@#define __NR_vmsplice 316#define __NR_move_pages 317#define __NR_getcpu 318 +#define __NR_kevent_get_events 319 +#define __NR_kevent_ctl 320 +#define __NR_kevent_wait 321 +#define __NR_kevent_commit 322 +#define __NR_kevent_init 323#ifdef __KERNEL__ -#define NR_syscalls 319 +#define NR_syscalls 324#include <linux/err.h> /*
include/asm-x86_64/unistd.h +11 -1
@@ -619,10 +619,20 @@ __SYSCALL(__NR_sync_file_range, sys_sync_file_range) __SYSCALL(__NR_vmsplice, sys_vmsplice)#define __NR_move_pages 279 __SYSCALL(__NR_move_pages, sys_move_pages) +#define __NR_kevent_get_events 280 +__SYSCALL(__NR_kevent_get_events, sys_kevent_get_events) +#define __NR_kevent_ctl 281 +__SYSCALL(__NR_kevent_ctl, sys_kevent_ctl) +#define __NR_kevent_wait 282 +__SYSCALL(__NR_kevent_wait, sys_kevent_wait) +#define __NR_kevent_commit 283 +__SYSCALL(__NR_kevent_commit, sys_kevent_commit) +#define __NR_kevent_init 284 +__SYSCALL(__NR_kevent_init, sys_kevent_init)#ifdef __KERNEL__ -#define __NR_syscall_max __NR_move_pages +#define __NR_syscall_max __NR_kevent_init#include <linux/err.h> #ifndef __NO_STUBS
include/linux/kevent.h new file +244 -0
244 lines not on this page · raw message
include/linux/kevent_storage.h new file +11 -0
@@ -0,0 +1,11 @@ +#ifndef __KEVENT_STORAGE_H +#define __KEVENT_STORAGE_H + +struct kevent_storage +{ +void *origin;/* Originator's pointer, e.g. struct sock or struct file. Can be NULL. */ +struct list_head list;/* List of queued kevents. */ +spinlock_t lock;/* Protects users queue. */ +}; + +#endif /* __KEVENT_STORAGE_H */
include/linux/syscalls.h +9 -0
@@ -54,6 +54,8 @@ struct compat_stat;struct compat_timeval ;struct robust_list_head ;struct getcpu_cache ; +struct ukevent ; +struct kevent_ring ;#include <linux/types.h> #include <linux/aio_abi.h>
@@ -599,4 +601,11 @@ asmlinkage long sys_set_robust_list(struct robust_list_head __user *head, size_t len);asmlinkage long sys_getcpu (unsigned __user *cpu,unsigned __user *node,struct getcpu_cache __user *cache); +asmlinkage long sys_kevent_get_events (int ctl_fd,unsigned int min,unsigned int max, +struct timespec timeout,struct ukevent __user *buf,unsigned flags); +asmlinkage long sys_kevent_ctl (int ctl_fd,unsigned int cmd,unsigned int num,struct ukevent __user *buf); +asmlinkage long sys_kevent_wait (int ctl_fd,unsigned int num,unsigned int old_uidx, +struct timespec timeout,unsigned int flags); +asmlinkage long sys_kevent_commit (int ctl_fd,unsigned int new_uidx,unsigned int over); +asmlinkage long sys_kevent_init (int ctl_fd,struct kevent_ring __user *ring,unsigned int num,unsigned int flags);#endif
include/linux/ukevent.h new file +185 -0
185 lines not on this page · raw message
init/Kconfig +2 -0
@@ -201,6 +201,8 @@ config AUDITSYSCALL such as SELinux. To use audit's filesystem watch feature, please ensure that INOTIFY is configured. +source "kernel/kevent/Kconfig" + config IKCONFIG bool "Kernel .config support" ---help---
kernel/Makefile +1 -0
@@ -47,6 +47,7 @@ obj-$(CONFIG_DETECT_SOFTLOCKUP) += softlockup.o obj-$(CONFIG_GENERIC_HARDIRQS) += irq/ obj-$(CONFIG_SECCOMP) += seccomp.o obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o +obj-$(CONFIG_KEVENT) += kevent/ obj-$(CONFIG_RELAY) += relay.o obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o obj-$(CONFIG_TASKSTATS) += taskstats.o
kernel/kevent/Kconfig new file +66 -0
66 lines not on this page · raw message
kernel/kevent/Makefile new file +6 -0
@@ -0,0 +1,6 @@ +obj-y := kevent.o kevent_user.o +obj-$(CONFIG_KEVENT_TIMER) += kevent_timer.o +obj-$(CONFIG_KEVENT_POLL) += kevent_poll.o +obj-$(CONFIG_KEVENT_SOCKET) += kevent_socket.o +obj-$(CONFIG_KEVENT_PIPE) += kevent_pipe.o +obj-$(CONFIG_KEVENT_SIGNAL) += kevent_signal.o
kernel/kevent/kevent.c new file +265 -0
265 lines not on this page · raw message
kernel/kevent/kevent_user.c new file +1360 -0
1,360 lines not on this page · raw message
kernel/sys_ni.c +6 -0
@@ -122,6 +122,12 @@ cond_syscall(ppc_rtas);cond_syscall (sys_spu_run);cond_syscall (sys_spu_create); +cond_syscall (sys_kevent_get_events); +cond_syscall (sys_kevent_ctl); +cond_syscall (sys_kevent_wait); +cond_syscall (sys_kevent_commit); +cond_syscall (sys_kevent_init); +/* mmu depending weak syscall entries */ cond_syscall (sys_mprotect);cond_syscall (sys_msync);
PATCH 3/8 kevent: poll/select() notifications.
poll/select() notifications.
This patch includes generic poll/select notifications. kevent_poll works simialr to epoll and has the same issues (callback is invoked not from internal state machine of the caller, but through process awake, a lot of allocations and so on).
fs/file_table.c +3 -0
@@ -20,6 +20,7 @@#include <linux/cdev.h> #include <linux/fsnotify.h> #include <linux/sysctl.h> +#include <linux/kevent.h> #include <linux/percpu_counter.h> #include <asm/atomic.h>
@@ -119,6 +120,7 @@ struct file *get_empty_filp(void) f->f_uid = tsk->fsuid; f->f_gid = tsk->fsgid;eventpoll_init_file (f); +kevent_init_file (f);/* f->f_version: 0 */ return f;
@@ -164,6 +166,7 @@ void fastcall __fput(struct file *file)* in the file cleanup chain. */ eventpoll_release (file); +kevent_cleanup_file (file);locks_remove_flock (file);if (file->f_op && file->f_op->release)
include/linux/fs.h +8 -0
@@ -276,6 +276,7 @@ extern int dir_notify_enable;#include <linux/init.h> #include <linux/sched.h> #include <linux/mutex.h> +#include <linux/kevent_storage.h> #include <asm/atomic.h> #include <asm/semaphore.h>
@@ -586,6 +587,10 @@ struct inode {struct mutex inotify_mutex;/* protects the watches list */ #endif +#if definedCONFIG_KEVENT_SOCKET || definedCONFIG_KEVENT_PIPE +struct kevent_storage st; +#endif +unsigned long i_state;unsigned long dirtied_when;/* jiffies of first dirtying */
@@ -739,6 +744,9 @@ struct file {struct list_head f_ep_links;spinlock_t f_ep_lock;#endif /* #ifdef CONFIG_EPOLL */ +#ifdef CONFIG_KEVENT_POLL +struct kevent_storage st; +#endif struct address_space *f_mapping; };extern spinlock_t files_lock;
kernel/kevent/kevent_poll.c new file +234 -0
234 lines not on this page · raw message
PATCH 4/8 kevent: Socket notifications.
Socket notifications.
This patch includes socket send/recv/accept notifications. Using trivial web server based on kevent and this features instead of epoll it's performance increased more than noticebly. More details about various benchmarks and server itself (evserver_kevent.c) can be found on project's homepage.
fs/inode.c +7 -0
@@ -21,6 +21,7 @@#include <linux/cdev.h> #include <linux/bootmem.h> #include <linux/inotify.h> +#include <linux/kevent.h> #include <linux/mount.h> /*
@@ -164,12 +165,18 @@ static struct inode *alloc_inode(struct super_block *sb) } inode->i_private =0 ; inode->i_mapping = mapping; +#if definedCONFIG_KEVENT_SOCKET || definedCONFIG_KEVENT_PIPE +kevent_storage_init (inode, &inode->st); +#endif }return inode; }void destroy_inode (struct inode *inode) { +#if definedCONFIG_KEVENT_SOCKET || definedCONFIG_KEVENT_PIPE +kevent_storage_fini (&inode->st); +#endif BUG_ON (inode_has_buffers (inode));security_inode_free (inode);if (inode->i_sb->s_op->destroy_inode)
include/net/sock.h +17 -15
@@ -48,6 +48,7 @@#include <linux/netdevice.h> #include <linux/skbuff.h> /* struct sk_buff */ #include <linux/security.h> +#include <linux/kevent.h> #include <linux/filter.h>
@@ -450,6 +451,21 @@ static inline int sk_stream_memory_free(struct sock *sk)extern void sk_stream_rfree (struct sk_buff *skb); +struct socket_alloc { +struct socket socket; +struct inode vfs_inode; +}; + +static inline struct socket *SOCKET_I (struct inode *inode) +{ + return &container_of(inode, struct socket_alloc, vfs_inode)->socket; +} + +static inline struct inode *SOCK_INODE (struct socket *socket) +{ + return &container_of(socket, struct socket_alloc, socket)->vfs_inode; +} +static inline void sk_stream_set_owner_r (struct sk_buff *skb,struct sock *sk) { skb->sk = sk;
@@ -477,6 +493,7 @@ static inline void sk_add_backlog(struct sock *sk, struct sk_buff *skb) sk->sk_backlog.tail = skb; } skb->next =NULL ; +kevent_socket_notify (sk,KEVENT_SOCKET_RECV ); }#define sk_wait_event (__sk, __timeo, __condition) \
@@ -679,21 +696,6 @@ static inline struct kiocb *siocb_to_kiocb(struct sock_iocb *si)return si->kiocb; } -struct socket_alloc { -struct socket socket; -struct inode vfs_inode; -}; - -static inline struct socket *SOCKET_I (struct inode *inode) -{ - return &container_of(inode, struct socket_alloc, vfs_inode)->socket; -} - -static inline struct inode *SOCK_INODE (struct socket *socket) -{ - return &container_of(socket, struct socket_alloc, socket)->vfs_inode; -} -extern void __sk_stream_mem_reclaim (struct sock *sk);extern int sk_stream_mem_schedule (struct sock *sk,int size,int kind);
include/net/tcp.h +1 -0
@@ -857,6 +857,7 @@ static inline int tcp_prequeue(struct sock *sk, struct sk_buff *skb) tp->ucopy.memory =0 ; }else if (skb_queue_len (&tp->ucopy.prequeue) ==1 ) {wake_up_interruptible (sk->sk_sleep); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND );if (!inet_csk_ack_scheduled (sk)) inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, (3 * TCP_RTO_MIN) / 4,
kernel/kevent/kevent_socket.c new file +144 -0
144 lines not on this page · raw message
net/core/sock.c +9 -1
@@ -1402,6 +1402,7 @@ static void sock_def_wakeup(struct sock *sk)if (sk->sk_sleep &&waitqueue_active (sk->sk_sleep))wake_up_interruptible_all (sk->sk_sleep);read_unlock (&sk->sk_callback_lock); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND ); }static void sock_def_error_report (struct sock *sk)
@@ -1411,6 +1412,7 @@ static void sock_def_error_report(struct sock *sk)wake_up_interruptible (sk->sk_sleep);sk_wake_async (sk,0 ,POLL_ERR );read_unlock (&sk->sk_callback_lock); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND ); }static void sock_def_readable (struct sock *sk,int len)
@@ -1420,6 +1422,7 @@ static void sock_def_readable(struct sock *sk, int len)wake_up_interruptible (sk->sk_sleep);sk_wake_async (sk,1 ,POLL_IN );read_unlock (&sk->sk_callback_lock); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND ); }static void sock_def_write_space (struct sock *sk)
@@ -1439,6 +1442,7 @@ static void sock_def_write_space(struct sock *sk) }read_unlock (&sk->sk_callback_lock); +kevent_socket_notify (sk,KEVENT_SOCKET_SEND |KEVENT_SOCKET_RECV ); }static void sock_def_destruct (struct sock *sk)
@@ -1489,6 +1493,8 @@ void sock_init_data(struct socket *sock, struct sock *sk) sk->sk_state =TCP_CLOSE ; sk->sk_socket = sock; +kevent_sk_reinit (sk); +sock_set_flag (sk,SOCK_ZAPPED );if (sock)
@@ -1555,8 +1561,10 @@ void fastcall release_sock(struct sock *sk)if (sk->sk_backlog.tail)__release_sock (sk); sk->sk_lock.owner =NULL ; -if (waitqueue_active (&sk->sk_lock.wq)) +if (waitqueue_active (&sk->sk_lock.wq)) {wake_up (&sk->sk_lock.wq); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV |KEVENT_SOCKET_SEND ); + }spin_unlock_bh (&sk->sk_lock.slock); }EXPORT_SYMBOL (release_sock);
net/core/stream.c +1 -0
@@ -36,6 +36,7 @@ void sk_stream_write_space(struct sock *sk)wake_up_interruptible (sk->sk_sleep);if (sock->fasync_list && !(sk->sk_shutdown &SEND_SHUTDOWN ))sock_wake_async (sock,2 ,POLL_OUT ); +kevent_socket_notify (sk,KEVENT_SOCKET_SEND |KEVENT_SOCKET_RECV ); } }
net/ipv4/tcp_input.c +1 -0
@@ -3119,6 +3119,7 @@ static void tcp_ofo_queue(struct sock *sk)__skb_unlink (skb, &tp->out_of_order_queue);__skb_queue_tail (&sk->sk_receive_queue, skb); +kevent_socket_notify (sk,KEVENT_SOCKET_RECV ); tp->rcv_nxt =TCP_SKB_CB (skb)->end_seq;if (skb->h.th->fin)tcp_fin (skb, sk, skb->h.th);
net/ipv4/tcp_ipv4.c +2 -0
@@ -61,6 +61,7 @@#include <linux/jhash.h> #include <linux/init.h> #include <linux/times.h> +#include <linux/kevent.h> #include <net/icmp.h> #include <net/inet_hashtables.h>
@@ -870,6 +871,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)reqsk_free (req); }else {inet_csk_reqsk_queue_hash_add (sk, req,TCP_TIMEOUT_INIT ); +kevent_socket_notify (sk,KEVENT_SOCKET_ACCEPT ); }return 0 ;
net/socket.c +3 -0
@@ -85,6 +85,7 @@#include <linux/kmod.h> #include <linux/audit.h> #include <linux/wireless.h> +#include <linux/kevent.h> #include <asm/uaccess.h> #include <asm/unistd.h>
@@ -490,6 +491,8 @@ static struct socket *sock_alloc(void) inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; +kevent_socket_reinit (sock); +get_cpu_var (sockets_in_use)++;put_cpu_var (sockets_in_use);return sock;
PATCH 5/8 kevent: Timer notifications.
Timer notifications.
Timer notifications can be used for fine grained per-process time management, since interval timers are very inconvenient to use, and they are limited.
This subsystem uses high-resolution timers.
id.raw[0] is used as number of seconds
id.raw[1] is used as number of nanoseconds
kernel/kevent/kevent_timer.c new file +114 -0
@@ -0,0 +1,114 @@ +/* +* 2006 Copyright (c) Evgeniy Polyakov <[email protected]> +* All rights reserved. +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation; either version 2 of the License, or +* (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, write to the Free Software +* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA +*/ + +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/list.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/hrtimer.h> +#include <linux/jiffies.h> +#include <linux/kevent.h> + +struct kevent_timer +{ +struct hrtimer ktimer; +struct kevent_storage ktimer_storage; +struct kevent *ktimer_event; +}; + +static int kevent_timer_func (struct hrtimer *timer) +{ + struct kevent_timer *t = container_of(timer, struct kevent_timer, ktimer); +struct kevent *k = t->ktimer_event; + +kevent_storage_ready (&t->ktimer_storage,NULL ,KEVENT_MASK_ALL ); +hrtimer_forward (timer, timer->base->softirq_time, +ktime_set (k->event.id.raw[0 ], k->event.id.raw[1 ])); +return HRTIMER_RESTART ; +} + +static struct lock_class_key kevent_timer_key; + +static int kevent_timer_enqueue (struct kevent *k) +{ +int err; +struct kevent_timer *t; + + t =kmalloc (sizeof (struct kevent_timer ),GFP_KERNEL ); +if (!t) +return -ENOMEM ; + +hrtimer_init (&t->ktimer,CLOCK_MONOTONIC ,HRTIMER_REL ); + t->ktimer.expires =ktime_set (k->event.id.raw[0 ], k->event.id.raw[1 ]); + t->ktimer.function = kevent_timer_func; + t->ktimer_event = k; + + err =kevent_storage_init (&t->ktimer, &t->ktimer_storage); +if (err) + goto err_out_free; +lockdep_set_class (&t->ktimer_storage.lock, &kevent_timer_key); + + err =kevent_storage_enqueue (&t->ktimer_storage, k); +if (err) + goto err_out_st_fini; + +hrtimer_start (&t->ktimer, t->ktimer.expires,HRTIMER_REL ); + +return 0 ; + +err_out_st_fini: +kevent_storage_fini (&t->ktimer_storage); +err_out_free: +kfree (t); + +return err; +} + +static int kevent_timer_dequeue (struct kevent *k) +{ +struct kevent_storage *st = k->st; + struct kevent_timer *t = container_of(st, struct kevent_timer, ktimer_storage); + +hrtimer_cancel (&t->ktimer); +kevent_storage_dequeue (st, k); +kfree (t); + +return 0 ; +} + +static int kevent_timer_callback (struct kevent *k) +{ + k->event.ret_data[0 ] =jiffies_to_msecs (jiffies); +return 1 ; +} + +static int __init kevent_init_timer (void ) +{ +struct kevent_callbacks tc = { + .callback = &kevent_timer_callback, + .enqueue = &kevent_timer_enqueue, + .dequeue = &kevent_timer_dequeue, + .flags =0 , + }; + +return kevent_add_callbacks (&tc,KEVENT_TIMER ); +} +module_init (kevent_init_timer); +
PATCH 6/8 kevent: Pipe notifications.
Pipe notifications.
fs/pipe.c +6 -0
@@ -16,6 +16,7 @@#include <linux/uio.h> #include <linux/highmem.h> #include <linux/pagemap.h> +#include <linux/kevent.h> #include <asm/uaccess.h> #include <asm/ioctls.h>
@@ -312,6 +313,7 @@ redo:break ; }if (do_wakeup) { +kevent_pipe_notify (inode,KEVENT_SOCKET_SEND );wake_up_interruptible_sync (&pipe->wait);kill_fasync (&pipe->fasync_writers,SIGIO ,POLL_OUT ); }
@@ -321,6 +323,7 @@ redo:/* Signal writers asynchronously that there is more room. */ if (do_wakeup) { +kevent_pipe_notify (inode,KEVENT_SOCKET_SEND );wake_up_interruptible (&pipe->wait);kill_fasync (&pipe->fasync_writers,SIGIO ,POLL_OUT ); }
@@ -490,6 +493,7 @@ redo2:break ; }if (do_wakeup) { +kevent_pipe_notify (inode,KEVENT_SOCKET_RECV );wake_up_interruptible_sync (&pipe->wait);kill_fasync (&pipe->fasync_readers,SIGIO ,POLL_IN ); do_wakeup =0 ;
@@ -501,6 +505,7 @@ redo2: out:mutex_unlock (&inode->i_mutex);if (do_wakeup) { +kevent_pipe_notify (inode,KEVENT_SOCKET_RECV );wake_up_interruptible (&pipe->wait);kill_fasync (&pipe->fasync_readers,SIGIO ,POLL_IN ); }
@@ -605,6 +610,7 @@ pipe_release(struct inode *inode, int decr, int decw)free_pipe_info (inode); }else {wake_up_interruptible (&pipe->wait); +kevent_pipe_notify (inode,KEVENT_SOCKET_SEND |KEVENT_SOCKET_RECV );kill_fasync (&pipe->fasync_readers,SIGIO ,POLL_IN );kill_fasync (&pipe->fasync_writers,SIGIO ,POLL_OUT ); }
kernel/kevent/kevent_pipe.c new file +123 -0
123 lines not on this page · raw message
PATCH 7/8 kevent: Signal notifications.
Signal notifications.
This type of notifications allows to deliver signals through kevent queue. One can find example application signal.c on project homepage.
If KEVENT_SIGNAL_NOMASK bit is set in raw_u64 id then signal will be delivered only through queue, otherwise both delivery types are used - old through update of mask of pending signals and through queue.
If signal is delivered only through kevent queue mask of pending signals is not updated at all, which is equal to putting signal into blocked mask, but with delivery of that signal through kevent queue.
include/linux/sched.h +5 -0
@@ -80,6 +80,7 @@ struct sched_param {#include <linux/resource.h> #include <linux/timer.h> #include <linux/hrtimer.h> +#include <linux/kevent_storage.h> #include <asm/processor.h>
@@ -1013,6 +1014,10 @@ struct task_struct {#ifdef CONFIG_TASK_DELAY_ACCT struct task_delay_info *delays;#endif +#ifdef CONFIG_KEVENT_SIGNAL +struct kevent_storage st; +u32 kevent_signals; +#endif };static inline pid_t process_group (struct task_struct *tsk)
kernel/fork.c +8 -0
@@ -46,6 +46,7 @@#include <linux/delayacct.h> #include <linux/taskstats_kern.h> #include <linux/random.h> +#include <linux/kevent.h> #include <asm/pgtable.h> #include <asm/pgalloc.h>
@@ -115,6 +116,9 @@ void __put_task_struct(struct task_struct *tsk)WARN_ON (atomic_read (&tsk->usage));WARN_ON (tsk == current); +#ifdef CONFIG_KEVENT_SIGNAL +kevent_storage_fini (&tsk->st); +#endif security_task_free (tsk);free_uid (tsk->user);put_group_info (tsk->group_info);
@@ -1121,6 +1125,10 @@ static struct task_struct *copy_process(unsigned long clone_flags,if (retval) goto bad_fork_cleanup_namespace; +#ifdef CONFIG_KEVENT_SIGNAL +kevent_storage_init (p, &p->st); +#endif + p->set_child_tid = (clone_flags &CLONE_CHILD_SETTID ) ? child_tidptr :NULL ;/* * Clear TID on mm_release()?
kernel/kevent/kevent_signal.c new file +94 -0
94 lines not on this page · raw message
kernel/signal.c +26 -0
@@ -23,6 +23,7 @@#include <linux/ptrace.h> #include <linux/signal.h> #include <linux/capability.h> +#include <linux/kevent.h> #include <asm/param.h> #include <asm/uaccess.h> #include <asm/unistd.h>
@@ -703,6 +704,9 @@ static int send_signal(int sig, struct siginfo *info, struct task_struct *t, {struct sigqueue * q =NULL ;int ret =0 ; + +if (kevent_signal_notify (t, sig)) +return 1 ;/* * fast-pathed signals for kernel-internal things like SIGSTOP
@@ -782,6 +786,17 @@ specific_send_sig_info(int sig, struct siginfo *info, struct task_struct *t) ret =send_signal (sig, info, t, &t->pending);if (!ret && !sigismember (&t->blocked, sig))signal_wake_up (t, sig ==SIGKILL ); +#ifdef CONFIG_KEVENT_SIGNAL +/* +* Kevent allows to deliver signals through kevent queue, +* it is possible to setup kevent to not deliver +* signal through the usual way, in that case send_signal() +* returns 1 and signal is delivered only through kevent queue. +* We simulate successfull delivery notification through this hack: +*/ +if (ret ==1 ) + ret =0 ; +#endif out:return ret; }
@@ -971,6 +986,17 @@ __group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p)* to avoid several races. */ ret =send_signal (sig, info, p, &p->signal->shared_pending); +#ifdef CONFIG_KEVENT_SIGNAL +/* +* Kevent allows to deliver signals through kevent queue, +* it is possible to setup kevent to not deliver +* signal through the usual way, in that case send_signal() +* returns 1 and signal is delivered only through kevent queue. +* We simulate successfull delivery notification through this hack: +*/ +if (ret ==1 ) + ret =0 ; +#endif if (unlikely (ret))return ret;
PATCH 8/8 kevent: Kevent posix timer notifications.
Kevent posix timer notifications.
Simple extensions to POSIX timers which allows
to deliver notification of the timer expiration
through kevent queue.
Example application posix_timer.c can be found
in archive on project homepage.
include/asm-generic/siginfo.h +3 -0
@@ -235,6 +235,7 @@ typedef struct siginfo {#define SIGEV_NONE 1/* other notification: meaningless */ #define SIGEV_THREAD 2/* deliver via thread creation */ #define SIGEV_THREAD_ID 4/* deliver to thread */ +#define SIGEV_KEVENT 8/* deliver through kevent queue */ /* * This works because the alignment is ok on all current architectures
@@ -260,6 +261,8 @@ typedef struct sigevent {void (*_function)(sigval_t );void *_attribute;/* really pthread_attr_t */ } _sigev_thread; + +int kevent_fd; } _sigev_un; } sigevent_t;
include/linux/posix-timers.h +4 -0
@@ -4,6 +4,7 @@#include <linux/spinlock.h> #include <linux/list.h> #include <linux/sched.h> +#include <linux/kevent_storage.h> union cpu_time_count {cputime_t cpu;
@@ -49,6 +50,9 @@ struct k_itimer {sigval_t it_sigev_value;/* value word of sigevent struct */ struct task_struct *it_process;/* process to send signal to */ struct sigqueue *sigq;/* signal queue entry. */ +#ifdef CONFIG_KEVENT_TIMER +struct kevent_storage st; +#endif union { struct { struct hrtimer timer;
kernel/posix-timers.c +171 -48
264 lines not on this page · raw message