From: Gilles Chanteperdrix <gilles.chanteperdrix@xenomai.org>
To: Wolfgang Denk <wd@domain.hid>
Cc: xenomai@xenomai.org
Subject: Re: [Xenomai-help] Some problems with shared memory
Date: Mon, 04 May 2009 10:49:25 +0200 [thread overview]
Message-ID: <49FEAC15.5060502@domain.hid> (raw)
In-Reply-To: <20090504072304.A7E2483420E8@domain.hid>
Wolfgang Denk wrote:
> Dear Gilles,
>
> in message <49FDFCC7.6000009@domain.hid> you wrote:
>>> 1.) If I don´t page align the size of shared memory (multiple of 4096)
>> I get the following error message :
>>> "createshm mmap: No such device or address"
> ...
>> Are you sure this issue is specific to xenomai, do not you have the same
>> effect with plain Linux shared memories ?
>
> To avoid misunderstandings, we should first point out that we are
> here in fact *not* discussing classic shared memory in the sense of
> the SysV SHM code, but the creation of shared memory regions by using
> the mmap() system call.
Yes, obviously.
>
>
> For completeness: shmget(2) *does* allow to allocate sizes which are
> not a multiple of pages sizes; but it then rounds up internally:
>
> SHMGET(2) ... DESCRIPTION:
>
> ... A new shared memory segment, with size equal to the value
> of size rounded up to a multiple of PAGE_SIZE, is created ...
>
> As far as mmap() is concerned, there is no restriction on the
> "length" parameter (though it is clear that internally this is
> rounded up to a multiple of pages, too, but only "length" bytes are
> visible to the user).
>
> The restrictions for page allignments affect the addr and offset
> paraemters only.
>
>
> To me it seems clear that mmap() in plain Linux and Xenomai behave
> differently.
It will be clear to me when someone will have run the tests.
--
Gilles.
next prev parent reply other threads:[~2009-05-04 8:49 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-03-30 9:31 [Xenomai-help] Some problems with shared memory roderik.wildenburg
2009-03-30 12:14 ` Gilles Chanteperdrix
2009-03-31 11:52 ` roderik.wildenburg
2009-04-06 8:47 ` roderik.wildenburg
2009-05-03 20:21 ` Gilles Chanteperdrix
2009-05-04 7:23 ` Wolfgang Denk
2009-05-04 8:49 ` Gilles Chanteperdrix [this message]
2009-05-04 15:24 ` Thomas Lockhart
2009-05-04 18:01 ` Wolfgang Denk
2009-05-04 18:33 ` Gilles Chanteperdrix
2009-05-05 12:16 ` roderik.wildenburg
2009-05-29 18:42 ` Gilles Chanteperdrix
2009-06-08 8:25 ` roderik.wildenburg
2009-06-08 20:36 ` Gilles Chanteperdrix
2009-06-09 6:30 ` roderik.wildenburg
2009-06-09 13:28 ` Philippe Gerum
2009-06-09 13:38 ` Philippe Gerum
2009-06-10 11:47 ` roderik.wildenburg
2009-06-12 14:06 ` Philippe Gerum
2009-06-16 13:20 ` roderik.wildenburg
2009-06-16 13:45 ` Philippe Gerum
2009-06-17 11:40 ` roderik.wildenburg
2009-06-17 14:19 ` Philippe Gerum
2009-06-18 8:37 ` roderik.wildenburg
2009-06-18 8:51 ` Philippe Gerum
2009-05-05 16:36 ` Thomas Lockhart
2009-05-05 19:03 ` Wolfgang Denk
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=49FEAC15.5060502@domain.hid \
--to=gilles.chanteperdrix@xenomai.org \
--cc=wd@domain.hid \
--cc=xenomai@xenomai.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 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.