linux-kernel - Re: drivers/staging/meilhaus build broken with ftrace

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]

Date:	Thu, 6 Aug 2009 18:47:02 -0400 (EDT)
From:	Steven Rostedt <rostedt@...dmis.org>
To:	Greg KH <gregkh@...e.de>
cc:	Roland Dreier <rdreier@...co.com>,
	Matt Fleming <matt@...sole-pimps.org>,
	David Kiliani <mail@...idkiliani.de>,
	LKML <linux-kernel@...r.kernel.org>, Ingo Molnar <mingo@...e.hu>
Subject: Re: drivers/staging/meilhaus build broken with ftrace


On Thu, 6 Aug 2009, Greg KH wrote:

> On Thu, Aug 06, 2009 at 05:46:57PM -0400, Steven Rostedt wrote:
> > 
> > Hi Roland,
> > 
> > On Thu, 6 Aug 2009, Roland Dreier wrote:
> > 
> > > Commit fc4c7355 ("ftrace: Fix the conditional that updates $ref_func")
> > > breaks the build of drivers/staging/meilhaus with ftrace enabled:
> > > 
> > >       CC [M]  drivers/staging/meilhaus/me0600_device.o
> > >     No function to reference at scripts/recordmcount.pl line 339, <IN> line 258.
> > >     make[4]: *** [drivers/staging/meilhaus/me0600_device.o] Error 9
> > > 
> > > Yes, that's crazy, but the situation is the following:
> > > 
> > >  - for some reason I can't fathom, the meilhaus driver declares
> > >    me0600_pci_constructor() as weak in drivers/staging/meilhaus/me0600_device.h
> > > 
> > >  - me0600_pci_constructor() is the only function in .text in
> > >    drivers/staging/meilhaus/me0600_device.c
> > > 
> > > so recordmcount.pl ends up finding only one weak function in .text, and
> > > so it never sets $ref_func before calling update_funcs().
> > 
> > Ug, I figured this would happen. I think I have a patch somewhere that can
> > handle this.
> 
> Don't make ftrace do a lot of work if it's a bug in the staging driver.
> Odds are the code in there should be fixed instead, I'll gladly take a
> patch to do that.

Yes the module did something funky, but what broke recordmcount.pl was not 
something that is not allowed. The code broke because a weak function was 
the only function in a section. This patch keeps it from crashing.

Hmm, maybe I should make it warn, because a weak function by itself in a 
section will not be traced.

-- Steve

diff --git a/scripts/recordmcount.pl b/scripts/recordmcount.pl
index d29baa2..979da4b 100755
--- a/scripts/recordmcount.pl
+++ b/scripts/recordmcount.pl
@@ -393,7 +393,7 @@ while (<IN>) {
 	    $read_function = 0;
 	}
 	# print out any recorded offsets
-	update_funcs() if ($text_found);
+	update_funcs() if (defined($ref_func));
 
 	# reset all markers and arrays
 	$text_found = 0;
@@ -441,7 +441,7 @@ while (<IN>) {
 }
 
 # dump out anymore offsets that may have been found
-update_funcs() if ($text_found);
+update_funcs() if (defined($ref_func));
 
 # If we did not find any mcount callers, we are done (do nothing).
 if (!$opened) {
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/