[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <c589992a-0c7e-ecf2-9544-3c82dfb96f9f@colorfullife.com>
Date: Tue, 20 Nov 2018 20:41:57 +0100
From: Manfred Spraul <manfred@...orfullife.com>
To: Matthew Wilcox <willy@...radead.org>,
Waiman Long <longman@...hat.com>
Cc: "Luis R. Rodriguez" <mcgrof@...nel.org>,
Kees Cook <keescook@...omium.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Jonathan Corbet <corbet@....net>, linux-kernel@...r.kernel.org,
linux-fsdevel@...r.kernel.org, linux-doc@...r.kernel.org,
Al Viro <viro@...iv.linux.org.uk>,
"Eric W. Biederman" <ebiederm@...ssion.com>,
Takashi Iwai <tiwai@...e.de>, Davidlohr Bueso <dbueso@...e.de>
Subject: Re: [PATCH v11 2/3] ipc: Conserve sequence numbers in ipcmni_extend
mode
Hi Matthew,
On 11/10/18 8:41 AM, Matthew Wilcox wrote:
> On Fri, Nov 09, 2018 at 03:11:31PM -0500, Waiman Long wrote:
>> The mixing in of a sequence number into the IPC IDs is probably to
>> avoid ID reuse in userspace as much as possible. With ipcmni_extend
>> mode, the number of usable sequence numbers is greatly reduced leading
>> to higher chance of ID reuse.
>>
>> To address this issue, we need to conserve the sequence number space
>> as much as possible. Right now, the sequence number is incremented
>> for every new ID created. In reality, we only need to increment the
>> sequence number when one or more IDs have been removed previously to
>> make sure that those IDs will not be reused when a new one is built.
>> This is being done irrespective of the ipcmni mode.
> That's not what I said. Increment the sequence ID when the cursor wraps,
> not when there's been a deletion.
Something like the attached patch?
Unfortunately, idr_alloc_cyclic cannot be used, this creates some
copy&paste from lib/idr.c to ipc/util.c
[as potential replacement for patch 2 and 3 from the series]
--
Manfred
View attachment "0002-ipc-util.c-use-idr_alloc_cyclic-for-ipc-allocations.patch" of type "text/x-patch" (3489 bytes)
Powered by blists - more mailing lists