From: Jan Kiszka <jan.kiszka@domain.hid>
To: Gilles Chanteperdrix <gilles.chanteperdrix@xenomai.org>
Cc: Xenomai core <Xenomai-core@domain.hid>
Subject: Re: [Xenomai-core] [Xenomai-git] Jan Kiszka : nucleus: Fix race between gatekeeper and thread deletion
Date: Sat, 16 Jul 2011 10:13:47 +0200 [thread overview]
Message-ID: <4E21483B.7050503@domain.hid> (raw)
In-Reply-To: <4E203C55.3080605@domain.hid>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 2011-07-15 15:10, Jan Kiszka wrote:
> But... right now it looks like we found our primary regression:
> "nucleus/shadow: shorten the uninterruptible path to secondary mode".
> It opens a short windows during relax where the migrated task may be
> active under both schedulers. We are currently evaluating a revert
> (looks good so far), and I need to work out my theory in more
> details.
Looks like this commit just made a long-standing flaw in Xenomai's
interrupt handling more visible: We reschedule over the interrupt stack
in the Xenomai interrupt handler tails, at least on x86-64. Not sure if
other archs have interrupt stacks, the point is Xenomai's design wrongly
assumes there are no such things. We were lucky so far that the values
saved on this shared stack were apparently "compatible", means we were
overwriting them with identical or harmless values. But that's no longer
true when interrupts are hitting us in the xnpod_suspend_thread path of
a relaxing shadow.
Likely the only possible fix is establishing a reschedule hook for
Xenomai in the interrupt exit path after the original stack is restored
- - just like Linux works. Requires changes to both ipipe and Xenomai
unfortunately.
Jan
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.16 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
iEYEARECAAYFAk4hSDsACgkQitSsb3rl5xSmOACfbZfcNKyO9YDvPE+R5H75d0ky
DX0An32BrZW+lpEnxnLLCHSQ5r8itnE9
=n6u8
-----END PGP SIGNATURE-----
next prev parent reply other threads:[~2011-07-16 8:13 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <E1QfDvt-0003TN-G7@domain.hid>
2011-07-11 18:53 ` [Xenomai-core] [Xenomai-git] Jan Kiszka : nucleus: Fix race between gatekeeper and thread deletion Gilles Chanteperdrix
2011-07-11 19:10 ` Jan Kiszka
2011-07-11 19:16 ` Jan Kiszka
2011-07-11 19:51 ` Gilles Chanteperdrix
2011-07-11 19:59 ` Jan Kiszka
2011-07-11 20:02 ` Gilles Chanteperdrix
2011-07-11 20:06 ` Jan Kiszka
2011-07-11 20:09 ` Gilles Chanteperdrix
2011-07-11 20:12 ` Jan Kiszka
2011-07-12 6:41 ` Gilles Chanteperdrix
2011-07-12 7:22 ` Jan Kiszka
2011-07-12 7:49 ` Gilles Chanteperdrix
2011-07-12 10:59 ` Gilles Chanteperdrix
2011-07-12 11:00 ` Jan Kiszka
2011-07-12 11:04 ` Gilles Chanteperdrix
2011-07-12 11:06 ` Jan Kiszka
2011-07-12 11:08 ` Gilles Chanteperdrix
2011-07-12 11:10 ` Jan Kiszka
2011-07-12 11:26 ` Gilles Chanteperdrix
2011-07-12 11:29 ` Jan Kiszka
2011-07-12 11:41 ` Gilles Chanteperdrix
2011-07-12 11:56 ` Jan Kiszka
2011-07-12 11:58 ` Jan Kiszka
2011-07-12 12:06 ` Gilles Chanteperdrix
2011-07-12 12:13 ` Jan Kiszka
2011-07-12 12:57 ` Jan Kiszka
2011-07-12 15:48 ` Philippe Gerum
2011-07-12 16:18 ` Jan Kiszka
2011-07-12 17:31 ` Gilles Chanteperdrix
2011-07-12 17:34 ` Jan Kiszka
2011-07-12 17:38 ` Gilles Chanteperdrix
2011-07-12 17:43 ` Jan Kiszka
2011-07-13 18:39 ` Gilles Chanteperdrix
2011-07-13 19:04 ` Jan Kiszka
2011-07-13 19:12 ` Gilles Chanteperdrix
2011-07-14 20:57 ` Jan Kiszka
2011-07-15 12:30 ` Gilles Chanteperdrix
2011-07-15 13:10 ` Jan Kiszka
2011-07-16 8:13 ` Jan Kiszka [this message]
2011-07-16 8:52 ` Philippe Gerum
2011-07-16 9:15 ` Jan Kiszka
2011-07-16 9:56 ` Philippe Gerum
2011-07-16 17:16 ` Jan Kiszka
2011-07-13 19:35 ` Philippe Gerum
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=4E21483B.7050503@domain.hid \
--to=jan.kiszka@domain.hid \
--cc=Xenomai-core@domain.hid \
--cc=gilles.chanteperdrix@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.