linux-kernel - Re: [PATCH][RESEND] checkpatch: Add warning for p0-patches

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20081114093853.GA3152@pengutronix.de>
Date:	Fri, 14 Nov 2008 10:38:53 +0100
From:	Wolfram Sang <w.sang@...gutronix.de>
To:	petkovbb@...il.com, Andy Whitcroft <apw@...dowen.org>,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH][RESEND] checkpatch: Add warning for p0-patches

Hi Boris,

> --- /dev/null   2008-11-09 02:46:02.525014459 +0100
> +++ arch/x86/kernel/tsc_resync.c        2008-11-14 07:22:34.000000000 +0100
> @@ -0,0 +1 @@
> +This is a new file
> 
> and, as you can see, it is a -p0 patch. Now, in the code you do:
> 
> if ($tree && -e "$root/$p1_prefix") {
> 	WARN("Patch prefix '$p1_prefix' exists. Is it maybe a p0-patch?\n");
> 
> and the "$root/$p1_prefix" won't exist - as a matter of fact - would
> lose its "arch" part due to the regex before and the if-condition won't
> trigger.

Careful. My approach is a bit different (inverse so to say) from yours
which I missed back then. $p1_prefix is the part which _was_ cut off and
it is wrong if it _does_ exist. See:

-                       $realfile =~ s@^[^/]*/@@;
+                       $realfile =~ s@^([^/]*)/@@;
+
+                       $p1_prefix = $1;

(So, a way to fool this algorithm is to give your kernel root dir the
same name as a directory inside the root dir, like:

--- drivers.orig/drivers/...
+++ drivers/drivers/...

This will generate a false positive. Oh, well...)

I decided to go this way intentionally to handle the new file problem.
So, in your case (I tried) it will cut off "arch", find the "arch"
directory and will complain. (Did you actually apply this patch? ;))

I thought the variable name 'p1_prefix' would speak for itself, but as
you misinterpreted it, maybe it should be renamed?

All the best,

   Wolfram

-- 
  Dipl.-Ing. Wolfram Sang | http://www.pengutronix.de
 Pengutronix - Linux Solutions for Science and Industry

Download attachment "signature.asc" of type "application/pgp-signature" (198 bytes)