linux-kernel - Re: [Intel-wired-lan] [PATCH] igc: Do not use link uninitialized in igc_check_for_copper

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, 16 Jul 2020 19:12:35 -0700
From:   Nathan Chancellor <natechancellor@...il.com>
To:     "Neftin, Sasha" <sasha.neftin@...el.com>
Cc:     Jeff Kirsher <jeffrey.t.kirsher@...el.com>, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org, clang-built-linux@...glegroups.com,
        intel-wired-lan@...ts.osuosl.org, Jakub Kicinski <kuba@...nel.org>,
        "David S. Miller" <davem@...emloft.net>,
        "Lifshits, Vitaly" <vitaly.lifshits@...el.com>,
        "Nguyen, Anthony L" <anthony.l.nguyen@...el.com>
Subject: Re: [Intel-wired-lan] [PATCH] igc: Do not use link uninitialized in
 igc_check_for_copper_link

On Thu, Jul 16, 2020 at 07:29:03PM +0300, Neftin, Sasha wrote:
> On 7/16/2020 07:49, Nathan Chancellor wrote:
> > Clang warns:
> > 
> Hello Nathan,
> Thanks for tracking our code.Please, look at https://patchwork.ozlabs.org/project/intel-wired-lan/patch/20200709073416.14126-1-sasha.neftin@intel.com/
> - I hope this patch already address this Clang warns - please, let me know.

I have not explicitly tested it but it seems obvious that it will. Let's
go with that.

Cheers,
Nathan

> > drivers/net/ethernet/intel/igc/igc_mac.c:374:6: warning: variable 'link'
> > is used uninitialized whenever 'if' condition is true
> > [-Wsometimes-uninitialized]
> >          if (!mac->get_link_status) {
> >              ^~~~~~~~~~~~~~~~~~~~~
> > drivers/net/ethernet/intel/igc/igc_mac.c:424:33: note: uninitialized use
> > occurs here
> >          ret_val = igc_set_ltr_i225(hw, link);
> >                                         ^~~~
> > drivers/net/ethernet/intel/igc/igc_mac.c:374:2: note: remove the 'if' if
> > its condition is always false
> >          if (!mac->get_link_status) {
> >          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > drivers/net/ethernet/intel/igc/igc_mac.c:367:11: note: initialize the
> > variable 'link' to silence this warning
> >          bool link;
> >                   ^
> >                    = 0
> > 1 warning generated.
> > 
> > It is not wrong, link is only uninitialized after this through
> > igc_phy_has_link. Presumably, if we skip the majority of this function
> > when get_link_status is false, we should skip calling igc_set_ltr_i225
> > as well. Just directly return 0 in this case, rather than bothering with
> > adding another label or initializing link in the if statement.
> > 
> > Fixes: 707abf069548 ("igc: Add initial LTR support")
> > Link: https://github.com/ClangBuiltLinux/linux/issues/1095
> > Signed-off-by: Nathan Chancellor <natechancellor@...il.com>
> > ---
> >   drivers/net/ethernet/intel/igc/igc_mac.c | 6 ++----
> >   1 file changed, 2 insertions(+), 4 deletions(-)
> > 
> > diff --git a/drivers/net/ethernet/intel/igc/igc_mac.c b/drivers/net/ethernet/intel/igc/igc_mac.c
> > index b47e7b0a6398..26e3c56a4a8b 100644
> > --- a/drivers/net/ethernet/intel/igc/igc_mac.c
> > +++ b/drivers/net/ethernet/intel/igc/igc_mac.c
> > @@ -371,10 +371,8 @@ s32 igc_check_for_copper_link(struct igc_hw *hw)
> >   	 * get_link_status flag is set upon receiving a Link Status
> >   	 * Change or Rx Sequence Error interrupt.
> >   	 */
> > -	if (!mac->get_link_status) {
> > -		ret_val = 0;
> > -		goto out;
> > -	}
> > +	if (!mac->get_link_status)
> > +		return 0;
> >   	/* First we want to see if the MII Status Register reports
> >   	 * link.  If so, then we want to get the current speed/duplex
> > 
> > base-commit: ca0e494af5edb59002665bf12871e94b4163a257
> > 
> Thanks,
> Sasha