All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] EAL: Take reserved hugepages into account
@ 2014-04-16 11:11 Burakov, Anatoly
       [not found] ` <C6ECDF3AB251BE4894318F4E451236975928B89E-kPTMFJFq+rGvNW/NfzhIbrfspsVTdybXVpNB7YpNyf8@public.gmane.org>
  0 siblings, 1 reply; 3+ messages in thread
From: Burakov, Anatoly @ 2014-04-16 11:11 UTC (permalink / raw)
  To: dev-VfR2kkLFssw

Some applications reserve hugepages for later use, but DPDK doesn't take reserved pages into account when calculating number of available number of hugepages. This patch adds reading from "resv_hugepages" file in addition to "free_hugepages".
---
 lib/librte_eal/linuxapp/eal/eal_hugepage_info.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/lib/librte_eal/linuxapp/eal/eal_hugepage_info.c b/lib/librte_eal/linuxapp/eal/eal_hugepage_info.c
index ef4cd1d..4ffeda5 100644
--- a/lib/librte_eal/linuxapp/eal/eal_hugepage_info.c
+++ b/lib/librte_eal/linuxapp/eal/eal_hugepage_info.c
@@ -66,8 +66,16 @@ static int32_t
 get_num_hugepages(const char *subdir)
 {
 	char path[PATH_MAX];
-	long unsigned num_pages = 0;
+	long unsigned resv_pages, num_pages = 0;
 	const char *nr_hp_file;
+	const char *nr_rsvd_file = "resv_hugepages";
+
+	/* first, check how many reserved pages kernel reports */
+	rte_snprintf(path, sizeof(path), "%s/%s/%s",
+			sys_dir_path, subdir, nr_rsvd_file);
+
+	if (eal_parse_sysfs_value(path, &resv_pages) < 0)
+		return 0;
 
 	/* if secondary process, just look at the number of hugepages,
 	 * otherwise look at number of free hugepages */
@@ -76,6 +84,8 @@ get_num_hugepages(const char *subdir)
 	else
 		nr_hp_file = "free_hugepages";
 
+	memset(path, 0, sizeof(path));
+
 	rte_snprintf(path, sizeof(path), "%s/%s/%s",
 			sys_dir_path, subdir, nr_hp_file);
 
@@ -86,6 +96,10 @@ get_num_hugepages(const char *subdir)
 		RTE_LOG(WARNING, EAL, "No free hugepages reported in %s\n",
 				subdir);
 
+	/* adjust num_pages in case of primary process */
+	if (num_pages > 0 && internal_config.process_type == RTE_PROC_PRIMARY)
+		num_pages -= resv_pages;
+
 	return (int32_t)num_pages;
 }
 
-- 
1.8.1.4

^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH] EAL: Take reserved hugepages into account
       [not found] ` <C6ECDF3AB251BE4894318F4E451236975928B89E-kPTMFJFq+rGvNW/NfzhIbrfspsVTdybXVpNB7YpNyf8@public.gmane.org>
@ 2014-04-17  3:22   ` Stephen Hemminger
  2014-05-13 14:25   ` Thomas Monjalon
  1 sibling, 0 replies; 3+ messages in thread
From: Stephen Hemminger @ 2014-04-17  3:22 UTC (permalink / raw)
  To: Burakov, Anatoly; +Cc: dev-VfR2kkLFssw

On Wed, 16 Apr 2014 11:11:12 +0000
"Burakov, Anatoly" <anatoly.burakov-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org> wrote:

> Some applications reserve hugepages for later use, but DPDK doesn't take reserved pages into account when calculating number of available number of hugepages. This patch adds reading from "resv_hugepages" file in addition to "free_hugepages".
> ---
>  lib/librte_eal/linuxapp/eal/eal_hugepage_info.c | 16 +++++++++++++++-
>  1 file changed, 15 insertions(+), 1 deletion(-)

Would it be worth using libhuge instead of doing heavy lifting in DPDK?
>  

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH] EAL: Take reserved hugepages into account
       [not found] ` <C6ECDF3AB251BE4894318F4E451236975928B89E-kPTMFJFq+rGvNW/NfzhIbrfspsVTdybXVpNB7YpNyf8@public.gmane.org>
  2014-04-17  3:22   ` Stephen Hemminger
@ 2014-05-13 14:25   ` Thomas Monjalon
  1 sibling, 0 replies; 3+ messages in thread
From: Thomas Monjalon @ 2014-05-13 14:25 UTC (permalink / raw)
  To: Burakov, Anatoly; +Cc: dev-VfR2kkLFssw

2014-04-16 11:11, Burakov, Anatoly:
> Some applications reserve hugepages for later use, but DPDK doesn't take
> reserved pages into account when calculating number of available number of
> hugepages. This patch adds reading from "resv_hugepages" file in addition
> to "free_hugepages".

Acked-by: Thomas Monjalon <thomas.monjalon-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>

Applied for version 1.7.0

-- 
Thomas

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2014-05-13 14:25 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-04-16 11:11 [PATCH] EAL: Take reserved hugepages into account Burakov, Anatoly
     [not found] ` <C6ECDF3AB251BE4894318F4E451236975928B89E-kPTMFJFq+rGvNW/NfzhIbrfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2014-04-17  3:22   ` Stephen Hemminger
2014-05-13 14:25   ` Thomas Monjalon

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.