An shm ringbuffer that is used for low overhead server-client communication.
Signalling is done through eventfd semaphores - it's based on pa_fdsem to avoid
syscalls if nothing is waiting on the other side.
Signed-off-by: David Henningsson <david.henningsson at canonical.com>
---
src/Makefile.am | 3 +-
src/pulsecore/srchannel.c | 297 ++++++++++++++++++++++++++++++++++++++++++++++
src/pulsecore/srchannel.h | 62 ++++++++++
3 files changed, 361 insertions(+), 1 deletion(-)
create mode 100644 src/pulsecore/srchannel.c
create mode 100644 src/pulsecore/srchannel.h
diff --git a/src/Makefile.am b/src/Makefile.am
index 5c2d5bc..9e72982 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -618,6 +618,7 @@ libpulsecommon_ at PA_MAJORMINOR@_la_SOURCES = \
pulsecore/creds.h \
pulsecore/dynarray.c pulsecore/dynarray.h \
pulsecore/endianmacros.h \
+ pulsecore/fdsem.c pulsecore/fdsem.h \
pulsecore/flist.c pulsecore/flist.h \
pulsecore/g711.c pulsecore/g711.h \
pulsecore/hashmap.c pulsecore/hashmap.h \
@@ -651,6 +652,7 @@ libpulsecommon_ at PA_MAJORMINOR@_la_SOURCES = \
pulsecore/queue.c pulsecore/queue.h \
pulsecore/random.c pulsecore/random.h \
pulsecore/refcnt.h \
+ pulsecore/srchannel.c pulsecore/srchannel.h \
pulsecore/sample-util.c pulsecore/sample-util.h \
pulsecore/shm.c pulsecore/shm.h \
pulsecore/bitset.c pulsecore/bitset.h \
@@ -880,7 +882,6 @@ libpulsecore_ at PA_MAJORMINOR@_la_SOURCES = \
pulsecore/core-scache.c pulsecore/core-scache.h \
pulsecore/core-subscribe.c pulsecore/core-subscribe.h \
pulsecore/core.c pulsecore/core.h \
- pulsecore/fdsem.c pulsecore/fdsem.h \
pulsecore/hook-list.c pulsecore/hook-list.h \
pulsecore/ltdl-helper.c pulsecore/ltdl-helper.h \
pulsecore/modargs.c pulsecore/modargs.h \
diff --git a/src/pulsecore/srchannel.c b/src/pulsecore/srchannel.c
new file mode 100644
index 0000000..15485b2
--- /dev/null
+++ b/src/pulsecore/srchannel.c
@@ -0,0 +1,297 @@
+/***
+ This file is part of PulseAudio.
+
+ Copyright 2014 David Henningsson, Canonical Ltd.
+
+ PulseAudio is free software; you can redistribute it and/or modify
+ it under the terms of the GNU Lesser General Public License as
+ published by the Free Software Foundation; either version 2.1 of the
+ License, or (at your option) any later version.
+
+ PulseAudio 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
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public
+ License along with PulseAudio; if not, write to the Free Software
+ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
+ USA.
+***/
+
+#ifdef HAVE_CONFIG_H
+#include <config.h>
+#endif
+
+#include "srchannel.h"
+
+#include <pulsecore/atomic.h>
+#include <pulse/xmalloc.h>
+
+/* #define DEBUG_SRCHANNEL */
+
+/* This ringbuffer might be useful in other contexts too, but
+ right now it's only used inside the srchannel, so let's keep it here
+ for the time being. */
+typedef struct pa_ringbuffer pa_ringbuffer;
+struct pa_ringbuffer {
+ pa_atomic_t *count;
+ int capacity;
+ uint8_t *memory;
+ int readindex, writeindex;
+};
+
+static void *pa_ringbuffer_peek(pa_ringbuffer *r, int *count) {
+ int c = pa_atomic_load(r->count);
+ if (r->readindex + c > r->capacity)
+ *count = r->capacity - r->readindex;
+ else
+ *count = c;
+ return r->memory + r->readindex;
+}
+
+/* Returns true only if the buffer was completely full before the drop. */
+static bool pa_ringbuffer_drop(pa_ringbuffer *r, int count) {
+ bool b = pa_atomic_sub(r->count, count) >= r->capacity;
+ r->readindex += count;
+ r->readindex %= r->capacity;
+ return b;
+}
+
+static void *pa_ringbuffer_begin_write(pa_ringbuffer *r, int *count) {
+ int c = pa_atomic_load(r->count);
+ *count = PA_MIN(r->capacity - r->writeindex, r->capacity - c);
+ return r->memory + r->writeindex;
+}
+
+static void pa_ringbuffer_end_write(pa_ringbuffer *r, int count) {
+ pa_atomic_add(r->count, count);
+ r->writeindex += count;
+ r->writeindex %= r->capacity;
+}
+
+struct pa_srchannel {
+ pa_ringbuffer rb_read, rb_write;
+ pa_fdsem *sem_read, *sem_write;
+ pa_memblock *memblock;
+ void *cb_userdata;
+ pa_srchannel_cb_t callback;
+ pa_io_event *read_event;
+ pa_mainloop_api *mainloop;
+};
+
+ssize_t pa_srchannel_write(pa_srchannel *sr, const void *data, size_t l) {
+ ssize_t written = 0;
+ while (l > 0) {
+ int towrite;
+ void *ptr = pa_ringbuffer_begin_write(&sr->rb_write, &towrite);
+ if ((size_t) towrite > l)
+ towrite = l;
+ if (towrite == 0) {
+#ifdef DEBUG_SRCHANNEL
+ pa_log("srchannel output buffer full");
+#endif
+ break;
+ }
+ memcpy(ptr, data, towrite);
+ pa_ringbuffer_end_write(&sr->rb_write, towrite);
+ written += towrite;
+ data = (uint8_t*) data + towrite;
+ l -= towrite;
+ }
+#ifdef DEBUG_SRCHANNEL
+ pa_log("Wrote %d bytes to srchannel, signalling fdsem", (int) written);
+#endif
+ pa_fdsem_post(sr->sem_write);
+ return written;
+}
+
+ssize_t pa_srchannel_read(pa_srchannel *sr, void *data, size_t l) {
+ ssize_t isread = 0;
+ while (l > 0) {
+ int toread;
+ void *ptr = pa_ringbuffer_peek(&sr->rb_read, &toread);
+ if ((size_t) toread > l)
+ toread = l;
+ if (toread == 0)
+ break;
+ memcpy(data, ptr, toread);
+ if (pa_ringbuffer_drop(&sr->rb_read, toread)) {
+#ifdef DEBUG_SRCHANNEL
+ pa_log("read from full output buffer, signalling fdsem");
+#endif
+ pa_fdsem_post(sr->sem_write);
+ }
+
+ isread += toread;
+ data = (uint8_t*) data + toread;
+ l -= toread;
+ }
+#ifdef DEBUG_SRCHANNEL
+ pa_log("Read %d bytes from srchannel", (int) isread);
+#endif
+ return isread;
+}
+
+/* This is the memory layout of the ringbuffer shm block. It is followed by
+ read and write ringbuffer memory. */
+struct srheader {
+ pa_atomic_t read_count;
+ pa_atomic_t write_count;
+ pa_fdsem_data read_semdata;
+ pa_fdsem_data write_semdata;
+ int capacity;
+ int readbuf_offset;
+ int writebuf_offset;
+ /* TODO: Maybe a marker here to make sure we talk to a server with equally sized struct */
+};
+
+static void read_loop(pa_srchannel* sr) {
+ do {
+ int q;
+ pa_ringbuffer_peek(&sr->rb_read, &q);
+#ifdef DEBUG_SRCHANNEL
+ pa_log("In read loop from srchannel, before callback, count = %d", q);
+#endif
+
+ if (q > 0 && sr->callback)
+ if (!sr->callback(sr, sr->cb_userdata)) {
+#ifdef DEBUG_SRCHANNEL
+ pa_log("Aborting read loop from srchannel");
+#endif
+ return;
+ }
+
+#ifdef DEBUG_SRCHANNEL
+ pa_ringbuffer_peek(&sr->rb_read, &q);
+ pa_log("In read loop from srchannel, after callback, count = %d", q);
+#endif
+
+ } while (pa_fdsem_before_poll(sr->sem_read) < 0);
+}
+
+static void read_cb(pa_mainloop_api *m, pa_io_event *e, int fd, pa_io_event_flags_t events, void *userdata) {
+ pa_srchannel* sr = userdata;
+
+ pa_fdsem_after_poll(sr->sem_read);
+ read_loop(sr);
+}
+
+pa_srchannel* pa_srchannel_new(pa_mainloop_api *m, pa_mempool *p) {
+ int capacity;
+ int readfd;
+ struct srheader *srh;
+
+ pa_srchannel* sr = pa_xmalloc0(sizeof(pa_srchannel));
+ sr->mainloop = m;
+ sr->memblock = pa_memblock_new_pool(p, -1);
+ srh = pa_memblock_acquire(sr->memblock);
+ pa_zero(*srh);
+
+ sr->rb_read.memory = (uint8_t*) srh + PA_ALIGN(sizeof(*srh));
+ srh->readbuf_offset = sr->rb_read.memory - (uint8_t*) srh;
+ capacity = (pa_memblock_get_length(sr->memblock) - srh->readbuf_offset) / 2;
+ sr->rb_write.memory = PA_ALIGN_PTR(sr->rb_read.memory + capacity);
+ srh->writebuf_offset = sr->rb_write.memory - (uint8_t*) srh;
+ capacity = PA_MIN(capacity, srh->writebuf_offset - srh->readbuf_offset);
+ pa_log_debug("SHM block is %d bytes, ringbuffer capacity is 2 * %d bytes",
+ (int) pa_memblock_get_length(sr->memblock), capacity);
+
+ srh->capacity = sr->rb_read.capacity = sr->rb_write.capacity = capacity;
+ sr->rb_read.count = &srh->read_count;
+ sr->rb_write.count = &srh->write_count;
+
+ sr->sem_read = pa_fdsem_new_shm(&srh->read_semdata);
+ sr->sem_write = pa_fdsem_new_shm(&srh->write_semdata);
+
+ readfd = pa_fdsem_get(sr->sem_read);
+#ifdef DEBUG_SRCHANNEL
+ pa_log("Enabling io event on fd %d", readfd);
+#endif
+ sr->read_event = m->io_new(m, readfd, PA_IO_EVENT_INPUT, read_cb, sr);
+ m->io_enable(sr->read_event, PA_IO_EVENT_INPUT);
+
+ return sr;
+}
+
+static void pa_srchannel_swap(pa_srchannel *sr) {
+ pa_srchannel temp = *sr;
+ sr->sem_read = temp.sem_write;
+ sr->sem_write = temp.sem_read;
+ sr->rb_read = temp.rb_write;
+ sr->rb_write = temp.rb_read;
+}
+
+pa_srchannel* pa_srchannel_new_from_template(pa_mainloop_api *m, pa_srchannel_template *t)
+{
+ int temp;
+ struct srheader *srh;
+ pa_srchannel* sr = pa_xmalloc0(sizeof(pa_srchannel));
+
+ sr->mainloop = m;
+ sr->memblock = t->memblock;
+ pa_memblock_ref(sr->memblock);
+ srh = pa_memblock_acquire(sr->memblock);
+
+ sr->rb_read.capacity = sr->rb_write.capacity = srh->capacity;
+ sr->rb_read.count = &srh->read_count;
+ sr->rb_write.count = &srh->write_count;
+ sr->rb_read.memory = (uint8_t*) srh + srh->readbuf_offset;
+ sr->rb_write.memory = (uint8_t*) srh + srh->writebuf_offset;
+
+ sr->sem_read = pa_fdsem_open_shm(&srh->read_semdata, t->readfd);
+ sr->sem_write = pa_fdsem_open_shm(&srh->write_semdata, t->writefd);
+
+ pa_srchannel_swap(sr);
+ temp = t->readfd; t->readfd = t->writefd; t->writefd = temp;
+
+#ifdef DEBUG_SRCHANNEL
+ pa_log("Enabling io event on fd %d", t->readfd);
+#endif
+ sr->read_event = m->io_new(m, t->readfd, PA_IO_EVENT_INPUT, read_cb, sr);
+ m->io_enable(sr->read_event, PA_IO_EVENT_INPUT);
+
+ return sr;
+}
+
+void pa_srchannel_export(pa_srchannel *sr, pa_srchannel_template *t) {
+ t->memblock = sr->memblock;
+ t->readfd = pa_fdsem_get(sr->sem_read);
+ t->writefd = pa_fdsem_get(sr->sem_write);
+}
+
+void pa_srchannel_set_callback(pa_srchannel *sr, pa_srchannel_cb_t callback, void *userdata) {
+ if (sr->callback)
+ pa_fdsem_after_poll(sr->sem_read);
+
+ sr->callback = callback;
+ sr->cb_userdata = userdata;
+
+ if (sr->callback)
+ /* Maybe deferred event? */
+ read_loop(sr);
+}
+
+void pa_srchannel_free(pa_srchannel *sr)
+{
+#ifdef DEBUG_SRCHANNEL
+ pa_log("Freeing srchannel");
+#endif
+ if (!sr)
+ return;
+
+ if (sr->read_event)
+ sr->mainloop->io_free(sr->read_event);
+
+ if (sr->sem_read)
+ pa_fdsem_free(sr->sem_read);
+ if (sr->sem_write)
+ pa_fdsem_free(sr->sem_write);
+
+ if (sr->memblock) {
+ pa_memblock_release(sr->memblock);
+ pa_memblock_unref(sr->memblock);
+ }
+
+ pa_xfree(sr);
+}
diff --git a/src/pulsecore/srchannel.h b/src/pulsecore/srchannel.h
new file mode 100644
index 0000000..8eac77f
--- /dev/null
+++ b/src/pulsecore/srchannel.h
@@ -0,0 +1,62 @@
+#ifndef foopulsesrchannelhfoo
+#define foopulsesrchannelhfoo
+
+/***
+ This file is part of PulseAudio.
+
+ Copyright 2014 David Henningsson, Canonical Ltd.
+
+ PulseAudio is free software; you can redistribute it and/or modify
+ it under the terms of the GNU Lesser General Public License as
+ published by the Free Software Foundation; either version 2.1 of the
+ License, or (at your option) any later version.
+
+ PulseAudio 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
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public
+ License along with PulseAudio; if not, write to the Free Software
+ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
+ USA.
+***/
+
+#include <pulse/mainloop-api.h>
+#include <pulsecore/fdsem.h>
+#include <pulsecore/memblock.h>
+
+/* An shm ringbuffer that is used for low overhead server-client communication.
+ Signaling is done through eventfd semaphores (pa_fdsem). */
+
+typedef struct pa_srchannel pa_srchannel;
+
+typedef struct pa_srchannel_template {
+ int readfd, writefd;
+ pa_memblock *memblock;
+} pa_srchannel_template;
+
+pa_srchannel* pa_srchannel_new(pa_mainloop_api *m, pa_mempool *p);
+/* Note: this creates a srchannel with swapped read and write. */
+pa_srchannel* pa_srchannel_new_from_template(pa_mainloop_api *m, pa_srchannel_template *t);
+
+void pa_srchannel_free(pa_srchannel *sr);
+
+void pa_srchannel_export(pa_srchannel *sr, pa_srchannel_template *t);
+
+ssize_t pa_srchannel_write(pa_srchannel *sr, const void *data, size_t l);
+ssize_t pa_srchannel_read(pa_srchannel *sr, void *data, size_t l);
+
+/* Set the callback function that is called whenever data becomes available for reading.
+ It can also be called if the output buffer was full and can now be written to.
+
+ Return false to abort all processing (e g if the srchannel has been freed during the callback).
+ Otherwise return true.
+
+ Note that if the ringbuffer is not empty when the callback is set up, the callback
+ will be called immediately.
+*/
+typedef bool (*pa_srchannel_cb_t)(pa_srchannel *sr, void *userdata);
+void pa_srchannel_set_callback(pa_srchannel *sr, pa_srchannel_cb_t callback, void *userdata);
+
+#endif
--
1.9.1
