This patch allows a worker thread to enable enqueue and move streams of
objects to the next nodes over different cores.
1. add graph_sched_wq_node to hold graph scheduling workqueue node
stream
2. add workqueue help functions to create/destroy/enqueue/dequeue
Signed-off-by: Haiyue Wang <haiyue.w...@intel.com>
Signed-off-by: Cunming Liang <cunming.li...@intel.com>
Signed-off-by: Zhirun Yan <zhirun....@intel.com>
---
lib/graph/graph_populate.c | 1 +
lib/graph/graph_private.h | 38 +++++++
lib/graph/graph_sched.c | 194 +++++++++++++++++++++++++++++++++++
lib/graph/meson.build | 2 +
lib/graph/rte_graph_worker.h | 45 ++++++++
lib/graph/version.map | 3 +
6 files changed, 283 insertions(+)
create mode 100644 lib/graph/graph_sched.c
diff --git a/lib/graph/graph_populate.c b/lib/graph/graph_populate.c
index 102fd6c29b..26f9670406 100644
--- a/lib/graph/graph_populate.c
+++ b/lib/graph/graph_populate.c
@@ -84,6 +84,7 @@ graph_nodes_populate(struct graph *_graph)
}
node->id = graph_node->node->id;
node->parent_id = pid;
+ node->lcore_id = graph_node->node->lcore_id;
nb_edges = graph_node->node->nb_edges;
node->nb_edges = nb_edges;
off += sizeof(struct rte_node);
diff --git a/lib/graph/graph_private.h b/lib/graph/graph_private.h
index d53ef289d4..e50a207764 100644
--- a/lib/graph/graph_private.h
+++ b/lib/graph/graph_private.h
@@ -59,6 +59,17 @@ struct node {
char next_nodes[][RTE_NODE_NAMESIZE]; /**< Names of next nodes. */
};
+/**
+ * @internal
+ *
+ * Structure that holds the graph scheduling workqueue node stream.
+ */
+struct graph_sched_wq_node {
+ rte_graph_off_t node_off;
+ uint16_t nb_objs;
+ void *objs[RTE_GRAPH_BURST_SIZE];
+} __rte_cache_aligned;
+
/**
* @internal
*
@@ -349,4 +360,31 @@ void graph_dump(FILE *f, struct graph *g);
*/
void node_dump(FILE *f, struct node *n);
+/**
+ * @internal
+ *
+ * Create the graph schedule work queue. And all cloned graphs attached to the
+ * parent graph MUST be destroied together for fast schedule design limitation.
+ *
+ * @param _graph
+ * The graph object
+ * @param _parent_graph
+ * The parent graph object which holds the run-queue head.
+ *
+ * @return
+ * - 0: Success.
+ * - <0: Graph schedule work queue related error.
+ */
+int graph_sched_wq_create(struct graph *_graph, struct graph *_parent_graph);
+
+/**
+ * @internal
+ *
+ * Destroy the graph schedule work queue.
+ *
+ * @param _graph
+ * The graph object
+ */
+void graph_sched_wq_destroy(struct graph *_graph);
+
#endif /* _RTE_GRAPH_PRIVATE_H_ */
diff --git a/lib/graph/graph_sched.c b/lib/graph/graph_sched.c
new file mode 100644
index 0000000000..465148796b
--- /dev/null
+++ b/lib/graph/graph_sched.c
@@ -0,0 +1,194 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(C) 2022 Intel Corporation
+ */
+
+#include <rte_common.h>
+#include <rte_errno.h>
+#include <rte_malloc.h>
+#include <rte_mempool.h>
+#include <rte_memzone.h>
+#include <rte_ring.h>
+
+#include "graph_private.h"
+
+#define GRAPH_SCHED_WQ_SIZE_MULTIPLIER 8
+#define GRAPH_SCHED_WQ_SIZE(nb_nodes) \
+ ((typeof(nb_nodes))((nb_nodes) * GRAPH_SCHED_WQ_SIZE_MULTIPLIER))
+
+static __rte_always_inline bool
+graph_src_node_avail(struct graph *graph)
+{
+ struct graph_node *graph_node;
+
+ STAILQ_FOREACH(graph_node, &graph->node_list, next)
+ if ((graph_node->node->flags & RTE_NODE_SOURCE_F) &&
+ (graph_node->node->lcore_id == RTE_MAX_LCORE ||
+ graph->lcore_id == graph_node->node->lcore_id))
+ return true;
+
+ return false;
+}
+
+int
+graph_sched_wq_create(struct graph *_graph, struct graph *_parent_graph)
+{
+ struct rte_graph *parent_graph = _parent_graph->graph;
+ struct rte_graph *graph = _graph->graph;
+ unsigned int wq_size;
+
+ /* cloned graph doesn't access input node if not affinity rx core*/
+ if (!graph_src_node_avail(_graph))
+ graph->head = 0;
+
+ wq_size = GRAPH_SCHED_WQ_SIZE(graph->nb_nodes);
+ wq_size = rte_align32pow2(wq_size + 1);
+
+ graph->wq = rte_ring_create(graph->name, wq_size, graph->socket,
+ RING_F_SC_DEQ);
+ if (graph->wq == NULL)
+ SET_ERR_JMP(EIO, fail, "Failed to allocate graph WQ");
+
+ graph->mp = rte_mempool_create(graph->name, wq_size,
+ sizeof(struct graph_sched_wq_node),
+ 0, 0, NULL, NULL, NULL, NULL,
+ graph->socket, MEMPOOL_F_SP_PUT);
+ if (graph->mp == NULL)
+ SET_ERR_JMP(EIO, fail_mp,
+ "Failed to allocate graph WQ schedule entry");
+
+ graph->lcore_id = _graph->lcore_id;
+
+ if (parent_graph->rq == NULL) {
+ parent_graph->rq = &parent_graph->rq_head;
+ SLIST_INIT(parent_graph->rq);
+ }
+
+ graph->rq = parent_graph->rq;
+ SLIST_INSERT_HEAD(graph->rq, graph, rq_next);
+
+ return 0;
+
+fail_mp:
+ rte_ring_free(graph->wq);
+ graph->wq = NULL;
+fail:
+ return -rte_errno;
+}
+
+void
+graph_sched_wq_destroy(struct graph *_graph)
+{
+ struct rte_graph *graph = _graph->graph;
+
+ if (graph == NULL)
+ return;
+
+ rte_ring_free(graph->wq);
+ graph->wq = NULL;
+
+ rte_mempool_free(graph->mp);
+ graph->mp = NULL;
+}
+
+static __rte_always_inline bool
+__graph_sched_node_enqueue(struct rte_node *node, struct rte_graph *graph)
+{
+ struct graph_sched_wq_node *wq_node;
+ uint16_t off = 0;
+ uint16_t size;
+
+submit_again:
+ if (rte_mempool_get(graph->mp, (void **)&wq_node) < 0)
+ goto fallback;
+
+ size = RTE_MIN(node->idx, RTE_DIM(wq_node->objs));
+ wq_node->node_off = node->off;
+ wq_node->nb_objs = size;
+ rte_memcpy(wq_node->objs, &node->objs[off], size * sizeof(void *));
+
+ while (rte_ring_mp_enqueue_bulk_elem(graph->wq, (void *)&wq_node,
+ sizeof(wq_node), 1, NULL) == 0)
+ rte_pause();
+
+ off += size;
+ node->total_sched_objs += size;
+ node->idx -= size;
+ if (node->idx > 0)
+ goto submit_again;
+
+ return true;
+
+fallback:
+ if (off != 0)
+ memmove(&node->objs[0], &node->objs[off],
+ node->idx * sizeof(void *));
+
+ node->total_sched_fail += node->idx;
+
+ return false;
+}
+
+bool __rte_noinline
+__rte_graph_sched_node_enqueue(struct rte_node *node,
+ struct rte_graph_rq_head *rq)
+{
+ const unsigned int lcore_id = node->lcore_id;
+ struct rte_graph *graph;
+
+ SLIST_FOREACH(graph, rq, rq_next)
+ if (graph->lcore_id == lcore_id)
+ break;
+
+ return graph != NULL ? __graph_sched_node_enqueue(node, graph) : false;
+}
+
+void __rte_noinline
+__rte_graph_sched_wq_process(struct rte_graph *graph)
+{
+ struct graph_sched_wq_node *wq_node;
+ struct rte_mempool *mp = graph->mp;
+ struct rte_ring *wq = graph->wq;
+ uint16_t idx, free_space;
+ struct rte_node *node;
+ unsigned int i, n;
+ uint64_t start;
+ uint16_t rc;
+ void **objs;
+ struct graph_sched_wq_node *wq_nodes[32];
+
+ n = rte_ring_sc_dequeue_burst_elem(wq, wq_nodes, sizeof(wq_nodes[0]),
+ RTE_DIM(wq_nodes), NULL);
+ if (n == 0)
+ return;
+
+ for (i = 0; i < n; i++) {
+ wq_node = wq_nodes[i];
+ node = RTE_PTR_ADD(graph, wq_node->node_off);
+ RTE_ASSERT(node->fence == RTE_GRAPH_FENCE);
+ idx = node->idx;
+ free_space = node->size - idx;
+
+ if (unlikely(free_space < wq_node->nb_objs))
+ __rte_node_stream_alloc_size(graph, node, node->size +
wq_node->nb_objs);
+
+ memmove(&node->objs[idx], wq_node->objs, wq_node->nb_objs *
sizeof(void *));
+ memset(wq_node->objs, 0, wq_node->nb_objs * sizeof(void *));
+ node->idx = idx + wq_node->nb_objs;
+ objs = node->objs;
+ rte_prefetch0(objs);
+
+ if (rte_graph_has_stats_feature()) {
+ start = rte_rdtsc();
+ rc = node->process(graph, node, objs, wq_node->nb_objs);
+ node->total_cycles += rte_rdtsc() - start;
+ node->total_calls++;
+ node->total_objs += rc;
+ } else {
+ node->process(graph, node, objs, wq_node->nb_objs);
+ }
+ wq_node->nb_objs = 0;
+ node->idx = 0;
+ }
+
+ rte_mempool_put_bulk(mp, (void **)wq_nodes, n);
+}
diff --git a/lib/graph/meson.build b/lib/graph/meson.build
index c7327549e8..440063b24e 100644
--- a/lib/graph/meson.build
+++ b/lib/graph/meson.build
@@ -13,8 +13,10 @@ sources = files(
'graph_ops.c',
'graph_debug.c',
'graph_stats.c',
+ 'graph_sched.c',
'graph_populate.c',
)
headers = files('rte_graph.h', 'rte_graph_worker.h')
deps += ['eal']
+deps += ['mempool', 'ring']
diff --git a/lib/graph/rte_graph_worker.h b/lib/graph/rte_graph_worker.h
index ca0185ec36..faf3f31ddc 100644
--- a/lib/graph/rte_graph_worker.h
+++ b/lib/graph/rte_graph_worker.h
@@ -28,6 +28,13 @@
extern "C" {
#endif
+/**
+ * @internal
+ *
+ * Singly-linked list head for graph schedule run-queue.
+ */
+SLIST_HEAD(rte_graph_rq_head, rte_graph);
+
/**
* @internal
*
@@ -39,6 +46,14 @@ struct rte_graph {
uint32_t cir_mask; /**< Circular buffer wrap around mask. */
rte_node_t nb_nodes; /**< Number of nodes in the graph. */
rte_graph_off_t *cir_start; /**< Pointer to circular buffer. */
+ /* Graph schedule area --BEGIN-- */
+ struct rte_graph_rq_head *rq __rte_cache_aligned; /* The run-queue */
+ struct rte_graph_rq_head rq_head; /* The head for run-queue list */
+ SLIST_ENTRY(rte_graph) rq_next; /* The next for run-queue list */
+ unsigned int lcore_id; /**< The graph running Lcore. */
+ struct rte_ring *wq; /**< The work-queue for pending streams. */
+ struct rte_mempool *mp; /**< The mempool for scheduling streams. */
+ /* Graph schedule area --END-- */
rte_graph_off_t nodes_start; /**< Offset at which node memory starts. */
rte_graph_t id; /**< Graph identifier. */
int socket; /**< Socket ID where memory is allocated. */
@@ -63,6 +78,8 @@ struct rte_node {
char parent[RTE_NODE_NAMESIZE]; /**< Parent node name. */
char name[RTE_NODE_NAMESIZE]; /**< Name of the node. */
+ /* Fast schedule area */
+ unsigned int lcore_id __rte_cache_aligned; /**< Node running Lcore. */
/* Fast path area */
#define RTE_NODE_CTX_SZ 16
uint8_t ctx[RTE_NODE_CTX_SZ] __rte_cache_aligned; /**< Node Context. */
@@ -118,6 +135,34 @@ __rte_experimental
void __rte_node_stream_alloc_size(struct rte_graph *graph,
struct rte_node *node, uint16_t req_size);
+/**
+ * @internal
+ *
+ * Schedule the node to the right graph's work queue.
+ *
+ * @param node
+ * Pointer to the scheduled node object.
+ * @param rq
+ * Pointer to the scheduled run-queue for all graphs.
+ *
+ * @return
+ * True on success, false otherwise.
+ */
+__rte_experimental
+bool __rte_graph_sched_node_enqueue(struct rte_node *node,
+ struct rte_graph_rq_head *rq);
+
+/**
+ * @internal
+ *
+ * Process all nodes (streams) in the graph's work queue.
+ *
+ * @param graph
+ * Pointer to the graph object.
+ */
+__rte_experimental
+void __rte_graph_sched_wq_process(struct rte_graph *graph);
+
/**
* Perform graph walk on the circular buffer and invoke the process function
* of the nodes and collect the stats.
diff --git a/lib/graph/version.map b/lib/graph/version.map
index 6fc43e4411..99f82eb21a 100644
--- a/lib/graph/version.map
+++ b/lib/graph/version.map
@@ -1,6 +1,9 @@
EXPERIMENTAL {
global:
+ __rte_graph_sched_node_enqueue;
+ __rte_graph_sched_wq_process;
+
__rte_node_register;
__rte_node_stream_alloc;
__rte_node_stream_alloc_size;
--
2.25.1
