From: Huichao cai <chcchc88@163.com>
To: jerinj@marvell.com, kirankumark@marvell.com,
ndabilpuram@marvell.com, yanzhirun_163@163.com
Cc: dev@dpdk.org
Subject: [PATCH] graph: optimize graph search when scheduling nodes
Date: Thu, 7 Nov 2024 16:04:42 +0800 [thread overview]
Message-ID: <1730966682-2632-1-git-send-email-chcchc88@163.com> (raw)
In the function __rte_graph_ccore_ispatch_stched_node_dequeue,
use a slower loop to search for the graph, modify the search logic
to record the result of the first search, and use this record for
subsequent searches to improve search speed.
Signed-off-by: Huichao cai <chcchc88@163.com>
---
lib/graph/rte_graph_model_mcore_dispatch.c | 11 +++++++----
lib/graph/rte_graph_worker_common.h | 1 +
2 files changed, 8 insertions(+), 4 deletions(-)
diff --git a/lib/graph/rte_graph_model_mcore_dispatch.c b/lib/graph/rte_graph_model_mcore_dispatch.c
index a590fc9..a81d338 100644
--- a/lib/graph/rte_graph_model_mcore_dispatch.c
+++ b/lib/graph/rte_graph_model_mcore_dispatch.c
@@ -118,11 +118,14 @@
struct rte_graph_rq_head *rq)
{
const unsigned int lcore_id = node->dispatch.lcore_id;
- struct rte_graph *graph;
+ struct rte_graph *graph = node->dispatch.graph;
- SLIST_FOREACH(graph, rq, next)
- if (graph->dispatch.lcore_id == lcore_id)
- break;
+ if (unlikely((!graph) || (graph->dispatch.lcore_id != lcore_id))) {
+ SLIST_FOREACH(graph, rq, next)
+ if (graph->dispatch.lcore_id == lcore_id)
+ break;
+ node->dispatch.graph = graph;
+ }
return graph != NULL ? __graph_sched_node_enqueue(node, graph) : false;
}
diff --git a/lib/graph/rte_graph_worker_common.h b/lib/graph/rte_graph_worker_common.h
index a518af2..4c2432b 100644
--- a/lib/graph/rte_graph_worker_common.h
+++ b/lib/graph/rte_graph_worker_common.h
@@ -110,6 +110,7 @@ struct __rte_cache_aligned rte_node {
unsigned int lcore_id; /**< Node running lcore. */
uint64_t total_sched_objs; /**< Number of objects scheduled. */
uint64_t total_sched_fail; /**< Number of scheduled failure. */
+ struct rte_graph *graph; /**< Graph corresponding to lcore_id. */
} dispatch;
};
rte_graph_off_t xstat_off; /**< Offset to xstat counters. */
--
1.8.3.1
next reply other threads:[~2024-11-07 8:04 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-11-07 8:04 Huichao cai [this message]
2024-11-07 9:37 ` [EXTERNAL] " Jerin Jacob
2024-11-08 1:39 ` Huichao Cai
2024-11-08 12:22 ` Jerin Jacob
2024-11-08 13:38 ` David Marchand
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1730966682-2632-1-git-send-email-chcchc88@163.com \
--to=chcchc88@163.com \
--cc=dev@dpdk.org \
--cc=jerinj@marvell.com \
--cc=kirankumark@marvell.com \
--cc=ndabilpuram@marvell.com \
--cc=yanzhirun_163@163.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).