DPDK patches and discussions
 help / color / mirror / Atom feed
From: Vipin Varghese <vipin.varghese@intel.com>
To: dev@dpdk.org, harry.van.haaren@intel.com
Cc: deepak.k.jain@intel.com, Vipin Varghese <vipin.varghese@intel.com>
Subject: [dpdk-dev] [PATCH v1] service: fix memory leak by rte_service_init
Date: Sun, 31 Dec 2017 20:16:08 +0530	[thread overview]
Message-ID: <1514731568-3979-1-git-send-email-vipin.varghese@intel.com> (raw)

This patch fixes the memory leak created by rte_service_init, when
run from secondary application. Running secondary application which
shares the huge page memory from primary multiple times causes memory
to be initialized but not free when application exit.

The rte_service_deinit check if the service is initialized. If yes, it
frees up rte_services & lcore_states. The API has to be called at end of
application run.

Signed-off-by: Vipin Varghese <vipin.varghese@intel.com>
---
 lib/librte_eal/common/include/rte_service.h |  9 +++++++++
 lib/librte_eal/common/rte_service.c         | 14 ++++++++++++++
 lib/librte_eal/rte_eal_version.map          |  1 +
 3 files changed, 24 insertions(+)

diff --git a/lib/librte_eal/common/include/rte_service.h b/lib/librte_eal/common/include/rte_service.h
index 9272440..ad2649e 100644
--- a/lib/librte_eal/common/include/rte_service.h
+++ b/lib/librte_eal/common/include/rte_service.h
@@ -420,6 +420,15 @@ int32_t rte_service_run_iter_on_app_lcore(uint32_t id,
  */
 int32_t rte_service_dump(FILE *f, uint32_t id);
 
+/**
+ * @warning
+ * @b EXPERIMENTAL: this API may change without prior notice
+ *
+ * Fress all memory that is internally used.
+ */
+
+void rte_service_deinit(void);
+
 #ifdef __cplusplus
 }
 #endif
diff --git a/lib/librte_eal/common/rte_service.c b/lib/librte_eal/common/rte_service.c
index ae97e6b..017bc67 100644
--- a/lib/librte_eal/common/rte_service.c
+++ b/lib/librte_eal/common/rte_service.c
@@ -98,6 +98,20 @@ struct core_state {
 static struct core_state *lcore_states;
 static uint32_t rte_service_library_initialized;
 
+void rte_service_deinit(void)
+{
+	if (rte_service_library_initialized) {
+		if (rte_services)
+			rte_free(rte_services);
+		if (lcore_states)
+			rte_free(lcore_states);
+
+		rte_service_library_initialized = 0;
+	}
+	return;
+}
+
+
 int32_t rte_service_init(void)
 {
 	if (rte_service_library_initialized) {
diff --git a/lib/librte_eal/rte_eal_version.map b/lib/librte_eal/rte_eal_version.map
index f4f46c1..0f14409 100644
--- a/lib/librte_eal/rte_eal_version.map
+++ b/lib/librte_eal/rte_eal_version.map
@@ -234,5 +234,6 @@ EXPERIMENTAL {
 	rte_service_set_runstate_mapped_check;
 	rte_service_set_stats_enable;
 	rte_service_start_with_defaults;
+	rte_service_deinit;
 
 } DPDK_17.11;
-- 
1.9.1

             reply	other threads:[~2017-12-31  9:05 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-31 14:46 Vipin Varghese [this message]
2018-01-08 10:17 ` Van Haaren, Harry
2018-01-08 14:40   ` Bruce Richardson
2018-01-11 18:20 ` [dpdk-dev] [PATCH v2] " Vipin Varghese
2018-01-25 22:10   ` Thomas Monjalon
2018-01-26 10:10     ` Van Haaren, Harry
2018-01-26 13:18       ` Thomas Monjalon
2018-01-26 20:55   ` [dpdk-dev] [PATCH v3] " Vipin Varghese
2018-01-26 15:20     ` Van Haaren, Harry
2018-01-26 16:54       ` Thomas Monjalon

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=1514731568-3979-1-git-send-email-vipin.varghese@intel.com \
    --to=vipin.varghese@intel.com \
    --cc=deepak.k.jain@intel.com \
    --cc=dev@dpdk.org \
    --cc=harry.van.haaren@intel.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).