dev.dpdk.org archive mirror
 help / color / mirror / Atom feed
From: =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?= <mannywang@tencent.com>
To: Stephen Hemminger <stephen@networkplumber.org>,
	Konstantin Ananyev <konstantin.ananyev@huawei.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [Internet]Re: [PATCH v6] acl: support custom memory allocators
Date: Tue, 9 Dec 2025 10:30:32 +0800	[thread overview]
Message-ID: <ACCDEF5FF0444B4A+74e02010-42f4-4b59-bb14-66aeccfe9bfd@tencent.com> (raw)
In-Reply-To: <20251208112916.2743f596@phoenix.local>

Just to add some background on how ACL currently allocates memory:

ACL has two types of allocations:
1. The memory used for the ACL match trie is NUMA-aware.
    This part relies on specific NUMA placement, so replacing rte_malloc 
with
    regular malloc/free is not feasible.

2. The temporary memory used during the build phase already uses 
malloc/free,
    since NUMA locality is not required there.

On 12/9/2025 3:29 AM, Stephen Hemminger wrote:
> A lot of the problems would go away if ACL just used regular malloc/free more,
> and rte_malloc/rte_free less. The existing rte_malloc is slow and fragments badly



  reply	other threads:[~2025-12-09  2:30 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-14  2:51 [RFC] rte_acl_build memory fragmentation concern and proposal for external memory support mannywang(王永峰)
2025-11-17 12:51 ` Konstantin Ananyev
2025-11-25  9:40   ` [PATCH] acl: support custom memory allocator =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-11-25 12:06   ` [PATCH v2] " =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-11-25 12:14   ` [PATCH v3] " =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-11-25 14:59     ` Stephen Hemminger
2025-11-26  2:37       ` [Internet]Re: " =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-11-25 18:01     ` Dmitry Kozlyuk
2025-11-26  2:44       ` [Internet]Re: " =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-11-26  7:57         ` Dmitry Kozlyuk
2025-11-26  8:09           ` =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-11-26 21:28             ` Stephen Hemminger
2025-11-27  2:05               ` [Internet]Re: " =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-11-28 13:26     ` Konstantin Ananyev
2025-11-28 15:07       ` =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-12-01 12:05       ` [PATCH v4] acl: support custom memory allocators in rte_acl_build =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-12-01 15:59         ` Patrick Robb
2025-12-01 16:42         ` Stephen Hemminger
2025-12-02  9:33           ` [Internet]Re: " =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-12-01 12:45       ` [PATCH v5] " =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-12-02  2:47         ` fengchengwen
2025-12-02  9:25           ` [PATCH v6] acl: support custom memory allocators =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-12-08  9:43             ` Konstantin Ananyev
2025-12-08 12:48               ` =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-12-08 19:29               ` Stephen Hemminger
2025-12-09  2:30                 ` =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?= [this message]
2025-12-09 11:06                 ` Konstantin Ananyev
2025-12-10  4:09                   ` fengchengwen
2025-12-08 12:57             ` [PATCH v7] " =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-12-09 10:59               ` Konstantin Ananyev
2025-12-09 12:56                 ` [Internet]RE: " =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-12-09 12:52               ` [PATCH v8] " =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-12-02  9:31           ` [Internet]Re: [PATCH v5] acl: support custom memory allocators in rte_acl_build =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-12-11  1:46     ` [PATCH v3] acl: support custom memory allocator Stephen Hemminger
2025-12-11  2:22       ` [Internet]Re: " =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-12-11  2:29       ` =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=
2025-12-11 13:04       ` =?gb18030?B?bWFubnl3YW5nKM3108C35Sk=?=

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=ACCDEF5FF0444B4A+74e02010-42f4-4b59-bb14-66aeccfe9bfd@tencent.com \
    --to=mannywang@tencent.com \
    --cc=dev@dpdk.org \
    --cc=konstantin.ananyev@huawei.com \
    --cc=stephen@networkplumber.org \
    /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).