From: Anatoly Burakov <anatoly.burakov@intel.com>
To: dev@dpdk.org
Cc: andras.kovacs@ericsson.com, laszlo.vadkeri@ericsson.com,
keith.wiles@intel.com, benjamin.walker@intel.com,
bruce.richardson@intel.com, thomas@monjalon.net
Subject: [dpdk-dev] [RFC 13/23] eal/memory: make use of dynamic memory allocation for init
Date: Tue, 19 Dec 2017 11:04:38 +0000 [thread overview]
Message-ID: <b7133f61831f5742efc7cc60ee071028025ed0f3.1513594170.git.anatoly.burakov@intel.com> (raw)
In-Reply-To: <cover.1513680516.git.anatoly.burakov@intel.com>
In-Reply-To: <cover.1513594170.git.anatoly.burakov@intel.com>
Add a new (non-legacy) memory init path for EAL. It uses the
new dynamic allocation facilities, although it's only being run
at startup.
If no -m or --socket-mem switches were specified, the new init
will not allocate anything, whereas if those switches were passed,
appropriate amounts of pages would be requested, just like for
legacy init.
Since rte_malloc support for dynamic allocation comes in later
patches, running DPDK without --socket-mem or -m switches will
fail in this patch.
Also, allocated pages will be physically discontiguous (or rather,
they're not guaranteed to be physically contiguous - they may still
be, by accident) unless IOVA_AS_VA mode is used.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
---
lib/librte_eal/linuxapp/eal/eal_memory.c | 60 ++++++++++++++++++++++++++++++++
1 file changed, 60 insertions(+)
diff --git a/lib/librte_eal/linuxapp/eal/eal_memory.c b/lib/librte_eal/linuxapp/eal/eal_memory.c
index 59f6889..7cc4a55 100644
--- a/lib/librte_eal/linuxapp/eal/eal_memory.c
+++ b/lib/librte_eal/linuxapp/eal/eal_memory.c
@@ -68,6 +68,7 @@
#include <rte_string_fns.h>
#include "eal_private.h"
+#include "eal_memalloc.h"
#include "eal_internal_cfg.h"
#include "eal_filesystem.h"
#include "eal_hugepages.h"
@@ -1322,6 +1323,61 @@ eal_legacy_hugepage_init(void)
return -1;
}
+static int
+eal_hugepage_init(void) {
+ struct hugepage_info used_hp[MAX_HUGEPAGE_SIZES];
+ uint64_t memory[RTE_MAX_NUMA_NODES];
+ int hp_sz_idx, socket_id;
+
+ test_phys_addrs_available();
+
+ memset(used_hp, 0, sizeof(used_hp));
+
+ for (hp_sz_idx = 0;
+ hp_sz_idx < (int) internal_config.num_hugepage_sizes;
+ hp_sz_idx++) {
+ /* meanwhile, also initialize used_hp hugepage sizes in used_hp */
+ struct hugepage_info *hpi;
+ hpi = &internal_config.hugepage_info[hp_sz_idx];
+ used_hp[hp_sz_idx].hugepage_sz = hpi->hugepage_sz;
+ }
+
+ /* make a copy of socket_mem, needed for balanced allocation. */
+ for (hp_sz_idx = 0; hp_sz_idx < RTE_MAX_NUMA_NODES; hp_sz_idx++)
+ memory[hp_sz_idx] = internal_config.socket_mem[hp_sz_idx];
+
+ /* calculate final number of pages */
+ if (calc_num_pages_per_socket(memory,
+ internal_config.hugepage_info, used_hp,
+ internal_config.num_hugepage_sizes) < 0)
+ return -1;
+
+ for (int hp_sz_idx = 0;
+ hp_sz_idx < (int) internal_config.num_hugepage_sizes;
+ hp_sz_idx++) {
+ for (socket_id = 0; socket_id < RTE_MAX_NUMA_NODES;
+ socket_id++) {
+ struct hugepage_info *hpi = &used_hp[hp_sz_idx];
+ unsigned num_pages = hpi->num_pages[socket_id];
+ int num_pages_alloc;
+
+ if (num_pages == 0)
+ continue;
+
+ RTE_LOG(DEBUG, EAL, "Allocating %u pages of size %luM on socket %i\n",
+ num_pages, hpi->hugepage_sz >> 20, socket_id);
+
+ num_pages_alloc = eal_memalloc_alloc_page_bulk(NULL,
+ num_pages,
+ hpi->hugepage_sz, socket_id,
+ true);
+ if (num_pages_alloc < 0)
+ return -1;
+ }
+ }
+ return 0;
+}
+
/*
* uses fstat to report the size of a file on disk
*/
@@ -1533,6 +1589,8 @@ int
rte_eal_hugepage_init(void) {
if (internal_config.legacy_mem)
return eal_legacy_hugepage_init();
+ else
+ return eal_hugepage_init();
return -1;
}
@@ -1540,6 +1598,8 @@ int
rte_eal_hugepage_attach(void) {
if (internal_config.legacy_mem)
return eal_legacy_hugepage_attach();
+ else
+ RTE_LOG(ERR, EAL, "Secondary processes aren't supported yet\n");
return -1;
}
--
2.7.4
next prev parent reply other threads:[~2017-12-19 11:05 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-12-19 11:04 [dpdk-dev] [PATCH 00/23] Dynamic memory allocation for DPDK Anatoly Burakov
[not found] ` <cover.1513594170.git.anatoly.burakov@intel.com>
2017-12-19 11:04 ` [dpdk-dev] [RFC 01/23] eal/memory: move get_virtual_area out of linuxapp eal_memory.c Anatoly Burakov
2017-12-19 11:15 ` Burakov, Anatoly
2017-12-19 11:04 ` [dpdk-dev] [RFC 02/23] eal/lcore: add function to report number of detected sockets Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 04/23] eal/malloc: move all locking to heap Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 05/23] eal/malloc: protect malloc heap stats with a lock Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 06/23] eal/malloc: make malloc a doubly-linked list Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 07/23] eal/malloc: make malloc_elem_join_adjacent_free public Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 10/23] eal: populate hugepage counts from socket-specific sysfs path Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 12/23] eal/memalloc: add support for dynamic memory allocation Anatoly Burakov
2017-12-19 11:04 ` Anatoly Burakov [this message]
2017-12-19 11:04 ` [dpdk-dev] [RFC 14/23] eal/memory: add support for dynamic unmapping of pages Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 15/23] eal/memalloc: add function to check if memory is physically contiguous Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 16/23] eal/malloc: enable dynamic memory allocation/free on malloc/free Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 17/23] eal/malloc: add backend support for contiguous memory allocation Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 18/23] eal/malloc: add rte_malloc support for allocating contiguous memory Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 19/23] eal/memzone: add support for reserving physically contiguous memzones Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 20/23] eal/memzone: make memzones use rte_fbarray Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 21/23] lib/mempool: add support for the new memory allocation methods Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [RFC 23/23] eal/memalloc: register/unregister memory with VFIO when alloc/free pages Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 01/23] eal: move get_virtual_area out of linuxapp eal_memory.c Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 02/23] eal: add function to report number of detected sockets Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 03/23] eal: add rte_fbarray Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 04/23] eal: move all locking to heap Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 05/23] eal: protect malloc heap stats with a lock Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 06/23] eal: make malloc a doubly-linked list Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 07/23] eal: make malloc_elem_join_adjacent_free public Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 08/23] eal: add "single file segments" command-line option Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 09/23] eal: add "legacy memory" option Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 10/23] eal: read hugepage counts from node-specific sysfs path Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 11/23] eal: replace memseg with memseg lists Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 12/23] eal: add support for dynamic memory allocation Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 13/23] eal: make use of dynamic memory allocation for init Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 14/23] eal: add support for dynamic unmapping of pages Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 15/23] eal: add API to check if memory is physically contiguous Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 16/23] eal: enable dynamic memory allocation/free on malloc/free Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 17/23] eal: add backend support for contiguous memory allocation Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 18/23] eal: add rte_malloc support for allocating contiguous memory Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 19/23] eal: enable reserving physically contiguous memzones Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 20/23] eal: make memzones use rte_fbarray Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 21/23] mempool: add support for the new memory allocation methods Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 22/23] vfio: allow to map other memory regions Anatoly Burakov
2017-12-19 11:04 ` [dpdk-dev] [PATCH 23/23] eal: map/unmap memory with VFIO when alloc/free pages Anatoly Burakov
2017-12-19 11:15 ` [dpdk-dev] [PATCH 00/23] Dynamic memory allocation for DPDK Burakov, Anatoly
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=b7133f61831f5742efc7cc60ee071028025ed0f3.1513594170.git.anatoly.burakov@intel.com \
--to=anatoly.burakov@intel.com \
--cc=andras.kovacs@ericsson.com \
--cc=benjamin.walker@intel.com \
--cc=bruce.richardson@intel.com \
--cc=dev@dpdk.org \
--cc=keith.wiles@intel.com \
--cc=laszlo.vadkeri@ericsson.com \
--cc=thomas@monjalon.net \
/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).