lib: Demultiplex the fast path

The fast path will now use an incoming ring buffer per flow per
process. This necessitated the development of a new method for the
asynchronous io call, which is now based on an event queue system for
scalability (fqueue). The ipcpd's and tools have been updated to this
API.

1 files changed, 0 insertions, 661 deletions

diff --git a/src/lib/shm_ap_rbuff.c b/src/lib/shm_ap_rbuff.c
deleted file mode 100644
index 5cbf5bd0..00000000
--- a/src/lib/shm_ap_rbuff.c
+++ /dev/null
@@ -1,661 +0,0 @@
-/*
- * Ouroboros - Copyright (C) 2016
- *
- * Ring buffer for application processes
- *
- *    Dimitri Staessens <[email protected]>
- *
- * 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., 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-#include <ouroboros/config.h>
-#include <ouroboros/shm_ap_rbuff.h>
-#include <ouroboros/lockfile.h>
-#include <ouroboros/time_utils.h>
-#include <ouroboros/errno.h>
-
-#define OUROBOROS_PREFIX "shm_ap_rbuff"
-
-#include <ouroboros/logs.h>
-
-#include <pthread.h>
-#include <sys/mman.h>
-#include <fcntl.h>
-#include <stdlib.h>
-#include <string.h>
-#include <stdint.h>
-#include <unistd.h>
-#include <signal.h>
-#include <sys/stat.h>
-#include <assert.h>
-
-#define FN_MAX_CHARS 255
-
-#define SHM_RBUFF_FILE_SIZE (SHM_BUFFER_SIZE * sizeof(struct rb_entry)         \
-                             + IRMD_MAX_FLOWS * sizeof(int8_t)                 \
-                             + IRMD_MAX_FLOWS * sizeof (ssize_t)               \
-                             + 2 * sizeof(size_t) + sizeof(pthread_mutex_t)    \
-                             + 2 * sizeof (pthread_cond_t))
-
-#define shm_rbuff_used(rb)((*rb->head + SHM_BUFFER_SIZE - *rb->tail)   \
-                          & (SHM_BUFFER_SIZE - 1))
-#define shm_rbuff_free(rb)(shm_rbuff_used(rb) + 1 < SHM_BUFFER_SIZE)
-#define shm_rbuff_empty(rb) (*rb->head == *rb->tail)
-#define head_el_ptr(rb) (rb->shm_base + *rb->head)
-#define tail_el_ptr(rb) (rb->shm_base + *rb->tail)
-
-struct shm_ap_rbuff {
-        struct rb_entry * shm_base; /* start of entry                */
-        size_t *          head;     /* start of ringbuffer head      */
-        size_t *          tail;     /* start of ringbuffer tail      */
-        int8_t *          acl;      /* start of port_id access table */
-        ssize_t *         cntrs;    /* start of port_id counters     */
-        pthread_mutex_t * lock;     /* lock all free space in shm    */
-        pthread_cond_t *  add;      /* SDU arrived                   */
-        pthread_cond_t *  del;      /* SDU removed                   */
-        pid_t             api;      /* api to which this rb belongs  */
-        int               fd;
-};
-
-struct shm_ap_rbuff * shm_ap_rbuff_create()
-{
-        struct shm_ap_rbuff * rb;
-        int                   shm_fd;
-        struct rb_entry *     shm_base;
-        pthread_mutexattr_t   mattr;
-        pthread_condattr_t    cattr;
-        char                  fn[FN_MAX_CHARS];
-        mode_t                mask;
-        int                   i;
-
-        sprintf(fn, SHM_AP_RBUFF_PREFIX "%d", getpid());
-
-        rb = malloc(sizeof(*rb));
-        if (rb == NULL) {
-                LOG_DBG("Could not allocate struct.");
-                return NULL;
-        }
-
-        mask = umask(0);
-
-        shm_fd = shm_open(fn, O_CREAT | O_EXCL | O_RDWR, 0666);
-        if (shm_fd == -1) {
-                LOG_DBG("Failed creating ring buffer.");
-                free(rb);
-                return NULL;
-        }
-
-        umask(mask);
-
-        if (ftruncate(shm_fd, SHM_RBUFF_FILE_SIZE - 1) < 0) {
-                LOG_DBG("Failed to extend ringbuffer.");
-                free(rb);
-                return NULL;
-        }
-#ifndef __APPLE__
-        if (write(shm_fd, "", 1) != 1) {
-                LOG_DBG("Failed to finalise extension of ringbuffer.");
-                free(rb);
-                return NULL;
-        }
-#endif
-        shm_base = mmap(NULL,
-                        SHM_RBUFF_FILE_SIZE,
-                        PROT_READ | PROT_WRITE,
-                        MAP_SHARED,
-                        shm_fd,
-                        0);
-
-        if (shm_base == MAP_FAILED) {
-                LOG_DBG("Failed to map shared memory.");
-                if (close(shm_fd) == -1)
-                        LOG_DBG("Failed to close invalid shm.");
-
-                if (shm_unlink(fn) == -1)
-                        LOG_DBG("Failed to remove invalid shm.");
-
-                free(rb);
-                return NULL;
-        }
-
-        rb->shm_base = shm_base;
-        rb->head     = (size_t *) (rb->shm_base + SHM_BUFFER_SIZE);
-        rb->tail     = rb->head + 1;
-        rb->acl      = (int8_t *) (rb->tail + 1);
-        rb->cntrs    = (ssize_t *) (rb->acl + IRMD_MAX_FLOWS);
-        rb->lock     = (pthread_mutex_t *) (rb->cntrs + IRMD_MAX_FLOWS);
-        rb->add      = (pthread_cond_t *) (rb->lock + 1);
-        rb->del      = rb->add + 1;
-
-        pthread_mutexattr_init(&mattr);
-#ifndef __APPLE__
-        pthread_mutexattr_setrobust(&mattr, PTHREAD_MUTEX_ROBUST);
-#endif
-        pthread_mutexattr_setpshared(&mattr, PTHREAD_PROCESS_SHARED);
-        pthread_mutex_init(rb->lock, &mattr);
-
-        pthread_condattr_init(&cattr);
-        pthread_condattr_setpshared(&cattr, PTHREAD_PROCESS_SHARED);
-#ifndef __APPLE__
-        pthread_condattr_setclock(&cattr, PTHREAD_COND_CLOCK);
-#endif
-        for (i = 0; i < IRMD_MAX_FLOWS; ++i) {
-                rb->cntrs[i] = 0;
-                rb->acl[i] = -1;
-        }
-
-        pthread_cond_init(rb->add, &cattr);
-        pthread_cond_init(rb->del, &cattr);
-
-        *rb->head = 0;
-        *rb->tail = 0;
-
-        rb->fd  = shm_fd;
-        rb->api = getpid();
-
-        return rb;
-}
-
-struct shm_ap_rbuff * shm_ap_rbuff_open(pid_t api)
-{
-        struct shm_ap_rbuff * rb;
-        int                   shm_fd;
-        struct rb_entry *     shm_base;
-        char                  fn[FN_MAX_CHARS];
-
-        sprintf(fn, SHM_AP_RBUFF_PREFIX "%d", api);
-
-        rb = malloc(sizeof(*rb));
-        if (rb == NULL) {
-                LOG_DBG("Could not allocate struct.");
-                return NULL;
-        }
-
-        shm_fd = shm_open(fn, O_RDWR, 0666);
-        if (shm_fd == -1) {
-                LOG_DBG("%d failed opening shared memory %s.", getpid(), fn);
-                free(rb);
-                return NULL;
-        }
-
-        shm_base = mmap(NULL,
-                        SHM_RBUFF_FILE_SIZE,
-                        PROT_READ | PROT_WRITE,
-                        MAP_SHARED,
-                        shm_fd,
-                        0);
-
-        if (shm_base == MAP_FAILED) {
-                LOG_DBG("Failed to map shared memory.");
-                if (close(shm_fd) == -1)
-                        LOG_DBG("Failed to close invalid shm.");
-
-                if (shm_unlink(fn) == -1)
-                        LOG_DBG("Failed to remove invalid shm.");
-
-                free(rb);
-                return NULL;
-        }
-
-        rb->shm_base = shm_base;
-        rb->head     = (size_t *) (rb->shm_base + SHM_BUFFER_SIZE);
-        rb->tail     = rb->head + 1;
-        rb->acl      = (int8_t *) (rb->tail + 1);
-        rb->cntrs    = (ssize_t *) (rb->acl + IRMD_MAX_FLOWS);
-        rb->lock     = (pthread_mutex_t *) (rb->cntrs + IRMD_MAX_FLOWS);
-        rb->add      = (pthread_cond_t *) (rb->lock + 1);
-        rb->del      = rb->add + 1;
-
-        rb->fd = shm_fd;
-        rb->api = api;
-
-        return rb;
-}
-
-void shm_ap_rbuff_close(struct shm_ap_rbuff * rb)
-{
-        assert(rb);
-
-        if (close(rb->fd) < 0)
-                LOG_DBG("Couldn't close shared memory.");
-
-        if (munmap(rb->shm_base, SHM_RBUFF_FILE_SIZE) == -1)
-                LOG_DBG("Couldn't unmap shared memory.");
-
-        free(rb);
-}
-
-void shm_ap_rbuff_open_port(struct shm_ap_rbuff * rb, int port_id)
-{
-        assert(rb);
-
-#ifdef __APPLE__
-        pthread_mutex_lock(rb->lock);
-#else
-        if (pthread_mutex_lock(rb->lock) == EOWNERDEAD) {
-                LOG_DBG("Recovering dead mutex.");
-                pthread_mutex_consistent(rb->lock);
-        }
-#endif
-        rb->acl[port_id] = 0; /* open */
-
-        pthread_mutex_unlock(rb->lock);
-}
-
-int shm_ap_rbuff_close_port(struct shm_ap_rbuff * rb, int port_id)
-{
-        int ret = 0;
-
-        assert(rb);
-
-#ifdef __APPLE__
-        pthread_mutex_lock(rb->lock);
-#else
-        if (pthread_mutex_lock(rb->lock) == EOWNERDEAD) {
-                LOG_DBG("Recovering dead mutex.");
-                pthread_mutex_consistent(rb->lock);
-        }
-#endif
-        rb->acl[port_id] = -1;
-
-        if (rb->cntrs[port_id] > 0)
-                ret = -EBUSY;
-
-        pthread_mutex_unlock(rb->lock);
-
-        return ret;
-}
-
-void shm_ap_rbuff_destroy(struct shm_ap_rbuff * rb)
-{
-        char fn[25];
-        struct lockfile * lf = NULL;
-
-        assert(rb);
-
-        if (rb->api != getpid()) {
-                lf = lockfile_open();
-                if (lf == NULL)
-                        return;
-                if (lockfile_owner(lf) == getpid()) {
-                        LOG_DBG("Ringbuffer %d destroyed by IRMd %d.",
-                                 rb->api, getpid());
-                        lockfile_close(lf);
-                } else {
-                        LOG_ERR("AP-I %d tried to destroy rbuff owned by %d.",
-                                getpid(), rb->api);
-                        lockfile_close(lf);
-                        return;
-                }
-        }
-
-        if (close(rb->fd) < 0)
-                LOG_DBG("Couldn't close shared memory.");
-
-        sprintf(fn, SHM_AP_RBUFF_PREFIX "%d", rb->api);
-
-        if (munmap(rb->shm_base, SHM_RBUFF_FILE_SIZE) == -1)
-                LOG_DBG("Couldn't unmap shared memory.");
-
-        if (shm_unlink(fn) == -1)
-                LOG_DBG("Failed to unlink shm.");
-
-        free(rb);
-}
-
-int shm_ap_rbuff_write(struct shm_ap_rbuff * rb, struct rb_entry * e)
-{
-        assert(rb);
-        assert(e);
-
-#ifdef __APPLE__
-        pthread_mutex_lock(rb->lock);
-#else
-        if (pthread_mutex_lock(rb->lock) == EOWNERDEAD) {
-                LOG_DBG("Recovering dead mutex.");
-                pthread_mutex_consistent(rb->lock);
-        }
-#endif
-        if (rb->acl[e->port_id]) {
-                pthread_mutex_unlock(rb->lock);
-                return -ENOTALLOC;
-        }
-
-        if (!shm_rbuff_free(rb)) {
-                pthread_mutex_unlock(rb->lock);
-                return -1;
-        }
-
-        if (shm_rbuff_empty(rb))
-                pthread_cond_broadcast(rb->add);
-
-        *head_el_ptr(rb) = *e;
-        *rb->head = (*rb->head + 1) & (SHM_BUFFER_SIZE -1);
-
-        ++rb->cntrs[e->port_id];
-
-        pthread_mutex_unlock(rb->lock);
-
-        return 0;
-}
-
-int shm_ap_rbuff_pop_idx(struct shm_ap_rbuff * rb)
-{
-        int ret = 0;
-
-        assert(rb);
-
-#ifdef __APPLE__
-        pthread_mutex_lock(rb->lock);
-#else
-        if (pthread_mutex_lock(rb->lock) == EOWNERDEAD) {
-                LOG_DBG("Recovering dead mutex.");
-                pthread_mutex_consistent(rb->lock);
-        }
-#endif
-        if (shm_rbuff_empty(rb)) {
-                pthread_mutex_unlock(rb->lock);
-                return -1;
-        }
-
-        ret = tail_el_ptr(rb)->index;
-        --rb->cntrs[tail_el_ptr(rb)->port_id];
-        *rb->tail = (*rb->tail + 1) & (SHM_BUFFER_SIZE -1);
-
-        pthread_mutex_unlock(rb->lock);
-
-        return ret;
-}
-
-static int shm_ap_rbuff_peek_b_all(struct shm_ap_rbuff * rb,
-                                   const struct timespec * timeout)
-{
-        struct timespec abstime;
-        int ret = 0;
-
-        assert(rb);
-
-        if (timeout != NULL) {
-                clock_gettime(PTHREAD_COND_CLOCK, &abstime);
-                ts_add(&abstime, timeout, &abstime);
-        }
-
-        pthread_cleanup_push((void(*)(void *))pthread_mutex_unlock,
-                             (void *) rb->lock);
-#ifdef __APPLE__
-        pthread_mutex_lock(rb->lock);
-#else
-        if (pthread_mutex_lock(rb->lock) == EOWNERDEAD) {
-                LOG_DBG("Recovering dead mutex.");
-                pthread_mutex_consistent(rb->lock);
-        }
-#endif
-        while (shm_rbuff_empty(rb)) {
-                if (timeout != NULL)
-                        ret = pthread_cond_timedwait(rb->add,
-                                                     rb->lock,
-                                                     &abstime);
-                else
-                        ret = pthread_cond_wait(rb->add, rb->lock);
-#ifndef __APPLE__
-                if (ret == EOWNERDEAD) {
-                        LOG_DBG("Recovering dead mutex.");
-                        pthread_mutex_consistent(rb->lock);
-                }
-#endif
-                if (ret == ETIMEDOUT)
-                        break;
-        }
-
-        if (ret != ETIMEDOUT)
-                ret = tail_el_ptr(rb)->port_id;
-        else
-                ret = -ETIMEDOUT;
-
-        pthread_cleanup_pop(true);
-
-        return ret;
-}
-
-int shm_ap_rbuff_peek_b(struct shm_ap_rbuff *   rb,
-                        bool *                  set,
-                        const struct timespec * timeout)
-{
-        struct timespec abstime;
-        int ret;
-
-        assert(rb);
-
-        if (set == NULL)
-                return shm_ap_rbuff_peek_b_all(rb, timeout);
-
-#ifdef __APPLE__
-        pthread_mutex_lock(rb->lock);
-#else
-        if (pthread_mutex_lock(rb->lock) == EOWNERDEAD) {
-                LOG_DBG("Recovering dead mutex.");
-                pthread_mutex_consistent(rb->lock);
-        }
-#endif
-        if (timeout != NULL) {
-                clock_gettime(PTHREAD_COND_CLOCK, &abstime);
-                ts_add(&abstime, timeout, &abstime);
-        }
-
-        pthread_cleanup_push((void(*)(void *))pthread_mutex_unlock,
-                             (void *) rb->lock);
-
-        while ((shm_rbuff_empty(rb) || !set[tail_el_ptr(rb)->port_id])
-               && (ret != ETIMEDOUT)) {
-                while (shm_rbuff_empty(rb)) {
-                        if (timeout != NULL)
-                                ret = pthread_cond_timedwait(rb->add,
-                                                             rb->lock,
-                                                             &abstime);
-                        else
-                                ret = pthread_cond_wait(rb->add, rb->lock);
-
-#ifndef __APPLE__
-                        if (ret == EOWNERDEAD) {
-                                LOG_DBG("Recovering dead mutex.");
-                                pthread_mutex_consistent(rb->lock);
-                        }
-#endif
-                        if (ret == ETIMEDOUT)
-                                break;
-                }
-
-                while (!set[tail_el_ptr(rb)->port_id]) {
-                        if (timeout != NULL)
-                                ret = pthread_cond_timedwait(rb->del,
-                                                             rb->lock,
-                                                             &abstime);
-                        else
-                                ret = pthread_cond_wait(rb->del, rb->lock);
-
-#ifndef __APPLE__
-                        if (ret == EOWNERDEAD) {
-                                LOG_DBG("Recovering dead mutex.");
-                                pthread_mutex_consistent(rb->lock);
-                        }
-#endif
-                        if (ret == ETIMEDOUT)
-                                break;
-                }
-        }
-
-        if (ret != ETIMEDOUT)
-                ret = tail_el_ptr(rb)->port_id;
-        else
-                ret = -ETIMEDOUT;
-
-        pthread_cleanup_pop(true);
-
-        return ret;
-}
-
-
-struct rb_entry * shm_ap_rbuff_read(struct shm_ap_rbuff * rb)
-{
-        struct rb_entry * e = NULL;
-
-        assert(rb);
-
-        pthread_cleanup_push((void(*)(void *))pthread_mutex_unlock,
-                             (void *) rb->lock);
-#ifdef __APPLE__
-        pthread_mutex_lock(rb->lock);
-#else
-        if (pthread_mutex_lock(rb->lock) == EOWNERDEAD) {
-                LOG_DBG("Recovering dead mutex.");
-                pthread_mutex_consistent(rb->lock);
-        }
-#endif
-        while (shm_rbuff_empty(rb))
-#ifdef __APPLE__
-                pthread_cond_wait(rb->add, rb->lock);
-#else
-                if (pthread_cond_wait(rb->add, rb->lock) == EOWNERDEAD) {
-                        LOG_DBG("Recovering dead mutex.");
-                        pthread_mutex_consistent(rb->lock);
-                }
-#endif
-        e = malloc(sizeof(*e));
-        if (e != NULL) {
-                *e = *(rb->shm_base + *rb->tail);
-                --rb->cntrs[e->port_id];
-                *rb->tail = (*rb->tail + 1) & (SHM_BUFFER_SIZE -1);
-        }
-
-        pthread_cleanup_pop(true);
-
-        return e;
-}
-
-ssize_t shm_ap_rbuff_read_port(struct shm_ap_rbuff * rb, int port_id)
-{
-        ssize_t idx = -1;
-
-        assert(rb);
-
-#ifdef __APPLE__
-        pthread_mutex_lock(rb->lock);
-#else
-        if (pthread_mutex_lock(rb->lock) == EOWNERDEAD) {
-                LOG_DBG("Recovering dead mutex.");
-                pthread_mutex_consistent(rb->lock);
-        }
-#endif
-        if (shm_rbuff_empty(rb) || tail_el_ptr(rb)->port_id != port_id) {
-                pthread_mutex_unlock(rb->lock);
-                return -1;
-        }
-
-        idx = tail_el_ptr(rb)->index;
-        --rb->cntrs[port_id];
-        *rb->tail = (*rb->tail + 1) & (SHM_BUFFER_SIZE -1);
-
-        pthread_cond_broadcast(rb->del);
-        pthread_mutex_unlock(rb->lock);
-
-        return idx;
-}
-
-ssize_t shm_ap_rbuff_read_port_b(struct shm_ap_rbuff *   rb,
-                                 int                     port_id,
-                                 const struct timespec * timeout)
-{
-        struct timespec abstime;
-        int ret = 0;
-        ssize_t idx = -1;
-
-        assert(rb);
-
-#ifdef __APPLE__
-        pthread_mutex_lock(rb->lock);
-#else
-        if (pthread_mutex_lock(rb->lock) == EOWNERDEAD) {
-                LOG_DBG("Recovering dead mutex.");
-                pthread_mutex_consistent(rb->lock);
-        }
-#endif
-        if (timeout != NULL) {
-                idx = -ETIMEDOUT;
-                clock_gettime(PTHREAD_COND_CLOCK, &abstime);
-                ts_add(&abstime, timeout, &abstime);
-        }
-
-        pthread_cleanup_push((void(*)(void *))pthread_mutex_unlock,
-                             (void *) rb->lock);
-
-        while ((shm_rbuff_empty(rb) || tail_el_ptr(rb)->port_id != port_id)
-               && (ret != ETIMEDOUT)) {
-                while (shm_rbuff_empty(rb)) {
-                        if (timeout != NULL)
-                                ret = pthread_cond_timedwait(rb->add,
-                                                             rb->lock,
-                                                             &abstime);
-                        else
-                                ret = pthread_cond_wait(rb->add, rb->lock);
-#ifndef __APPLE__
-                        if (ret == EOWNERDEAD) {
-                                LOG_DBG("Recovering dead mutex.");
-                                pthread_mutex_consistent(rb->lock);
-                        }
-#endif
-                        if (ret == ETIMEDOUT)
-                                break;
-                }
-
-                while (tail_el_ptr(rb)->port_id != port_id) {
-                        if (timeout != NULL)
-                                ret = pthread_cond_timedwait(rb->del,
-                                                             rb->lock,
-                                                             &abstime);
-                        else
-                                ret = pthread_cond_wait(rb->del, rb->lock);
-#ifndef __APPLE__
-                        if (ret == EOWNERDEAD) {
-                                LOG_DBG("Recovering dead mutex.");
-                                pthread_mutex_consistent(rb->lock);
-                        }
-#endif
-                        if (ret == ETIMEDOUT)
-                                break;
-                }
-        }
-
-        if (ret != ETIMEDOUT) {
-                idx = tail_el_ptr(rb)->index;
-                --rb->cntrs[port_id];
-                *rb->tail = (*rb->tail + 1) & (SHM_BUFFER_SIZE -1);
-
-                pthread_cond_broadcast(rb->del);
-        }
-
-        pthread_cleanup_pop(true);
-
-        return idx;
-}
-
-void shm_ap_rbuff_reset(struct shm_ap_rbuff * rb)
-{
-        assert(rb);
-
-        pthread_mutex_lock(rb->lock);
-        *rb->tail = 0;
-        *rb->head = 0;
-        pthread_mutex_unlock(rb->lock);
-}