From: Waiman Long <longman@redhat.com>
To: Christopher Lameter <cl@linux.com>
Cc: Oleg Nesterov <oleg@redhat.com>,
Matthew Wilcox <willy@infradead.org>,
Andrew Morton <akpm@linux-foundation.org>,
Pekka Enberg <penberg@kernel.org>,
David Rientjes <rientjes@google.com>,
Joonsoo Kim <iamjoonsoo.kim@lge.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
selinux@vger.kernel.org, Paul Moore <paul@paul-moore.com>,
Stephen Smalley <sds@tycho.nsa.gov>,
Eric Paris <eparis@parisplace.org>,
"Peter Zijlstra (Intel)" <peterz@infradead.org>
Subject: Re: [PATCH 2/4] signal: Make flush_sigqueue() use free_q to release memory
Date: Fri, 22 Mar 2019 14:12:01 -0400 [thread overview]
Message-ID: <93523469-48b0-07c8-54fd-300678af3163@redhat.com> (raw)
In-Reply-To: <01000169a686689d-bc18fecd-95e1-4b3e-8cd5-dad1b1c570cc-000000@email.amazonses.com>
On 03/22/2019 01:50 PM, Christopher Lameter wrote:
> On Fri, 22 Mar 2019, Waiman Long wrote:
>
>> I am looking forward to it.
> There is also alrady rcu being used in these paths. kfree_rcu() would not
> be enough? It is an estalished mechanism that is mature and well
> understood.
>
In this case, the memory objects are from kmem caches, so they can't
freed using kfree_rcu().
There are certainly overhead using the kfree_rcu(), or a
kfree_rcu()-like mechanism. Also I think the actual freeing is done at
SoftIRQ context which can be a problem if there are too many memory
objects to free.
I think what Oleg is trying to do is probably the most efficient way.
Cheers,
Longman
next prev parent reply other threads:[~2019-03-22 18:12 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-21 21:45 [PATCH 0/4] Signal: Fix hard lockup problem in flush_sigqueue() Waiman Long
2019-03-21 21:45 ` [PATCH 1/4] mm: Implement kmem objects freeing queue Waiman Long
2019-03-22 17:47 ` Christopher Lameter
2019-03-21 21:45 ` [PATCH 2/4] signal: Make flush_sigqueue() use free_q to release memory Waiman Long
2019-03-22 1:52 ` Matthew Wilcox
2019-03-22 11:16 ` Oleg Nesterov
2019-03-22 16:10 ` Waiman Long
2019-03-22 17:50 ` Christopher Lameter
2019-03-22 18:12 ` Waiman Long [this message]
2019-03-22 19:39 ` Christopher Lameter
2019-03-22 19:59 ` Matthew Wilcox
2019-03-25 14:15 ` Christopher Lameter
2019-03-25 15:26 ` Matthew Wilcox
2019-03-25 16:16 ` Christopher Lameter
2019-03-26 13:36 ` Oleg Nesterov
2019-03-26 13:29 ` Oleg Nesterov
2019-03-21 21:45 ` [PATCH 3/4] signal: Add free_uid_to_q() Waiman Long
2019-03-21 21:45 ` [PATCH 4/4] mm: Do periodic rescheduling when freeing objects in kmem_free_up_q() Waiman Long
2019-03-21 22:00 ` Peter Zijlstra
2019-03-22 14:35 ` Waiman Long
2019-03-22 10:15 ` [PATCH 0/4] Signal: Fix hard lockup problem in flush_sigqueue() Matthew Wilcox
2019-03-22 11:49 ` Oleg Nesterov
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=93523469-48b0-07c8-54fd-300678af3163@redhat.com \
--to=longman@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=cl@linux.com \
--cc=eparis@parisplace.org \
--cc=iamjoonsoo.kim@lge.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=oleg@redhat.com \
--cc=paul@paul-moore.com \
--cc=penberg@kernel.org \
--cc=peterz@infradead.org \
--cc=rientjes@google.com \
--cc=sds@tycho.nsa.gov \
--cc=selinux@vger.kernel.org \
--cc=willy@infradead.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).