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 11:17:24 +1000
From:	Dave Airlie <airlied@...il.com>
To:	Steven Rostedt <rostedt@...dmis.org>
Cc:	Matt Fleming <matt@...sole-pimps.org>, Ingo Molnar <mingo@...e.hu>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Benjamin Herrenschmidt <benh@...nel.crashing.org>,
	Paul Mackerras <paulus@...ba.org>,
	linux-kernel@...r.kernel.org, Dave Jones <davej@...emonkey.org.uk>
Subject: Re: Please revert fc4c73554c9d93b3e495f2f7acae1323b0d5db84. Re: 
	[PATCH 1/2] ftrace: Fix the conditional that updates $ref_func

On Thu, Aug 6, 2009 at 7:21 AM, Steven Rostedt<rostedt@...dmis.org> wrote:
>
> On Wed, 5 Aug 2009, Steven Rostedt wrote:
>
>>
>> On Wed, 5 Aug 2009, Dave Airlie wrote:
>>
>> > On Fri, Jul 24, 2009 at 2:16 AM, Matt Fleming<matt@...sole-pimps.org> wrote:
>> > > Fix the conditional that checks if we already have a $ref_func and that
>> > > the new function is weak. The code as previously checking whether either
>> > > condition was false, and we really need to only update $ref_func is both
>> > > cconditions are false.
>> > >
>> >
>> > This breaks the powerpc build on Fedora.
>> >
>> > When building on ppc64 this commit causes the links of drivers/hwmon/lm93.o
>> > to fail.
>> >
>> > It introduces an undefined symbol
>> >                  U .LM93_IN_FROM_REG
>> >
>> > that isn't produced when this patch is reverted
>> >
>> > This was found when the Fedora kernel failed to build when I pulled in
>> > -rc5-git3.
>> >
>> > can be seen at the end of:
>> > http://koji.fedoraproject.org/koji/getfile?taskID=1582002&name=build.log
>> >
>> > So can we revert this (at this stage in -rc5 or maybe fix it).
>> >
>>
>> I was not able to reproduce it, but could you see if this patch fixes the
>> issue?
>>
>
> Forget that one. I was finally able to reproduce it. This patch fixed it
> for me:
>

Doesn't seem to work here, I put the patch in on top of the Fedora
kernel instead of the previous one and got a link failure again in lm93.o

I'll reconfirm it now.

Dave.


> -- Steve
>
> diff --git a/scripts/recordmcount.pl b/scripts/recordmcount.pl
> index d29baa2..98e4afb 100755
> --- a/scripts/recordmcount.pl
> +++ b/scripts/recordmcount.pl
> @@ -135,6 +135,7 @@ $mv = "mv" if ((length $mv) == 0);
>
>  my %locals;            # List of local (static) functions
>  my %weak;              # List of weak functions
> +my %symbols;           # List of symbols that are not local or weak
>  my %convert;           # List of local functions used that needs conversion
>
>  my $type;
> @@ -318,6 +319,8 @@ while (<IN>) {
>        $locals{$1} = 1;
>     } elsif (/^[0-9a-fA-F]+\s+([wW])\s+(\S+)/) {
>        $weak{$2} = $1;
> +    } elsif (/^[0-9a-fA-F]+\s+(\S+)\s+(\S+)/) {
> +       $symbols{$2} = $1;
>     }
>  }
>  close(IN);
> @@ -408,13 +411,15 @@ while (<IN>) {
>        # if this is either a local function or a weak function
>        # keep looking for functions that are global that
>        # we can use safely.
> -       if (!defined($locals{$text}) && !defined($weak{$text})) {
> +       if (!defined($locals{$text}) && !defined($weak{$text}) &&
> +           defined($symbols{$text})) {
>            $ref_func = $text;
>            $read_function = 0;
>            $offset = hex $1;
>        } else {
>            # if we already have a function, and this is weak, skip it
> -           if (!defined($ref_func) && !defined($weak{$text})) {
> +           if (!defined($ref_func) && defined($locals{$text}) &&
> +               !defined($weak{$text})) {
>                $ref_func = $text;
>                $offset = hex $1;
>            }
>
--
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/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ