From: Andi Kleen <ak@muc.de>
To: Alexander Viro <viro@math.psu.edu>
Cc: Andi Kleen <ak@muc.de>,
engler@csl.Stanford.EDU, linux-kernel@vger.kernel.org
Subject: Re: [CHECKER] 37 stack variables >= 1K in 2.4.17
Date: Thu, 13 Jun 2002 21:01:30 +0200 [thread overview]
Message-ID: <20020613210130.A27417@averell> (raw)
In-Reply-To: <m3d6uvxdts.fsf@averell.firstfloor.org> <Pine.GSO.4.21.0206131420010.20315-100000@weyl.math.psu.edu>
On Thu, Jun 13, 2002 at 08:26:54PM +0200, Alexander Viro wrote:
> Ugh... OK, let me try again:
>
> One bit of apriory information needed to get anything interesting from
> this analysis: there is a set of mutually recursive functions (see above)
> and there is a limit (5) on the depth of recursion in that loop.
>
> It has to be known to checker. Explicitly, since
> a) automatically deducing it is not realistic
> b) cutting off the stuff behind that loop will cut off a _lot_ of
> things - both in filesystems and in VFS (and in block layer, while we are
> at it).
>
> I'm not saying that checker can't be used for that analysis - it can, but
> naive approach (find recursive stuff and cut it off) will not be too
if you see all possible paths through the program as a tree which branches
for every decision then you only need to cut off the branches that are
actually pointing upward the tree again. This would still allow to follow
down into the callees of the recursive function because there should be
at least one path that is non recursive (if not Checker should definitely
complain ;)
e.g.
----<-----------------+
v |
IF TRUE RECURSION
-------+------ some path ----+
|
ELSE non recursive path
+-------------------------- other functions ---------->
Other functions can be still checked, you only need to prune the cycle.
I have no idea if checker's algorithms actually work like this, but I could
imagine that it would be one possible implementation.
-Andi
next prev parent reply other threads:[~2002-06-13 19:01 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-06-10 3:56 [CHECKER] 37 stack variables >= 1K in 2.4.17 Dawson Engler
2002-06-12 8:43 ` Pavel Machek
2002-06-12 19:11 ` Nikita Danilov
2002-06-12 21:51 ` Benjamin LaHaise
2002-06-12 22:26 ` Alexander Viro
2002-06-12 22:38 ` Benjamin LaHaise
2002-06-12 22:44 ` Robert Love
2002-06-12 22:55 ` procfs documentation Tom Bradley
2002-06-13 11:17 ` John Levon
2002-06-13 0:20 ` [CHECKER] 37 stack variables >= 1K in 2.4.17 Alexander Viro
2002-06-13 8:30 ` Helge Hafting
2002-06-13 13:24 ` Roger Larsson
2002-06-14 10:06 ` Helge Hafting
2002-06-13 6:38 ` Dawson Engler
2002-06-13 6:59 ` Alexander Viro
2002-06-13 17:41 ` Daniel Phillips
2002-06-13 17:53 ` Alexander Viro
2002-06-13 18:45 ` Daniel Phillips
2002-06-13 17:56 ` Andi Kleen
2002-06-13 18:26 ` Alexander Viro
2002-06-13 19:01 ` Andi Kleen [this message]
2002-06-14 0:05 ` William Lee Irwin III
2002-06-13 21:50 ` Dawson Engler
2002-06-13 22:43 ` Oliver Xymoron
2002-06-14 0:25 ` Alexander Viro
2002-06-13 6:36 ` Dawson Engler
-- strict thread matches above, loose matches on Subject: below --
2002-06-16 0:48 Andries.Brouwer
2002-06-16 1:08 ` Alexander Viro
2002-06-16 7:47 Andries.Brouwer
2002-06-16 8:36 ` Alexander Viro
2002-06-16 9:59 Andries.Brouwer
2002-06-16 10:33 ` Alexander Viro
2002-06-16 10:56 Andries.Brouwer
2002-06-16 11:38 ` Alexander Viro
2002-06-16 13:13 Andries.Brouwer
2002-06-16 18:51 ` Alexander Viro
2002-06-16 20:41 Andries.Brouwer
2002-06-16 21:33 ` Andreas Dilger
2002-06-16 21:34 ` Alexander Viro
2002-06-17 10:09 ` David Woodhouse
2002-06-16 22:05 Andries.Brouwer
2002-06-16 23:57 ` Alexander Viro
2002-06-17 11:07 Andries.Brouwer
2002-06-17 12:00 ` David Woodhouse
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=20020613210130.A27417@averell \
--to=ak@muc.de \
--cc=engler@csl.Stanford.EDU \
--cc=linux-kernel@vger.kernel.org \
--cc=viro@math.psu.edu \
/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