lkml.org 
[lkml]   [2009]   [May]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH] Fix Warnining in arch/x86/kernel/signal.c
From
Date
On Wed, 2009-05-13 at 11:20 +0900, Hiroshi Shimamoto wrote:
> Subrata Modak wrote:
> > Hi,
> >
> >> On Tue, 2009-05-12 at 17:16 +0200, Andi Kleen wrote:
> >> On Tue, May 12, 2009 at 05:16:14PM +0530, Subrata Modak wrote:
> >>
> >> Hi Subrata,
> >>
> >>> With gcc (GCC) 4.4.1 20090429 (prerelease), i get the following build warning:
> >> Patch looks good (you can add a
> >> Reviewed-by: Andi Kleen <ak@linux.intel.com>)
> >> But I don't maintain this code anymore. Please resend to x86@kernel.org
> >> cc linux-kernel@vger.kernel.org for merge.
> >>
> >> Thanks,
> >>
> >> -Andi
> >
> >
> > With gcc (GCC) 4.4.1 20090429 (prerelease), i get the following build warning:
> >
> > CC arch/x86/kernel/signal.o
> > arch/x86/kernel/signal.c: In function ‘sys_sigreturn’:
> > arch/x86/kernel/signal.c:573: warning: ‘set.sig[1]’ may be used uninitialized in this function
> >
> > On investigation i found that this is because of the evaluation
> > precedence of the expression below:
> >
> > 569 unsigned long sys_sigreturn(struct pt_regs *regs)
> > 570 {
> > 571 struct sigframe __user *frame;
> > 572 unsigned long ax;
> > 573 sigset_t set;
> > 574
> > 575 frame = (struct sigframe __user *)(regs->sp - 8);
> > 576
> > 577 if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
> > 578 goto badframe;
> > 579 if (__get_user(set.sig[0], &frame->sc.oldmask) || (_NSIG_WORDS > 1
> > 580 && __copy_from_user(&set.sig[1], &frame->extramask,
> > 581 sizeof(frame->extramask))))
> >
> > The initialization for set.sig[1] may not occur if
> > __get_user(set.sig[0], &frame->sc.oldmask)
> > evalutes to true. So, the compiler is complaining.
> >
> > I have devised a small patch for this which wanes away this warning
> > without changing the conditional evaluation criteria. Let me know if
> > you like this patch.
> >
> > 582 goto badframe;
> > 583
> > 584 sigdelsetmask(&set, ~_BLOCKABLE);
> > 585 spin_lock_irq(&current->sighand->siglock);
> > 586 current->blocked = set;
> > 587 recalc_sigpending();
> > 588 spin_unlock_irq(&current->sighand->siglock);
> > 589
> > 590 if (restore_sigcontext(regs, &frame->sc, &ax))
> > 591 goto badframe;
> > 592 return ax;
> > 593
> > 594 badframe:
> > 595 signal_fault(regs, frame, "sigreturn");
> > 596
> > 597 return 0;
> > 598 }
> >
> > Signed-Off-By: Subrata Modak <subrata@linux.vnet.ibm.com>
> > Reviewed-by: Andi Kleen <ak@linux.intel.com>
> > To: <x86@kernel.org>
> > Cc: Linux Kernel <linux-kernel@vger.kernel.org>
> > Cc: Andi Kleen <andi@firstfloor.org>
> > Cc: Andi Kleen <ak@linux.intel.com>
> > Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
> > Cc: Sachin P Sant <sachinp@linux.vnet.ibm.com>
> > Subject: [PATCH] Fix Warnining in arch/x86/kernel/signal.c
> > ---
> >
> > --- a/arch/x86/kernel/signal.c 2009-05-12 10:59:24.000000000 +0530
> > +++ b/arch/x86/kernel/signal.c 2009-05-12 16:57:32.000000000 +0530
> > @@ -576,9 +576,10 @@ unsigned long sys_sigreturn(struct pt_re
> >
> > if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
> > goto badframe;
> > - if (__get_user(set.sig[0], &frame->sc.oldmask) || (_NSIG_WORDS > 1
> > - && __copy_from_user(&set.sig[1], &frame->extramask,
> > - sizeof(frame->extramask))))
> > +
> > + if ( (__copy_from_user(&set.sig[1], &frame->extramask,
> > + sizeof(frame->extramask)) && _NSIG_WORDS > 1) ||
> > + __get_user(set.sig[0], &frame->sc.oldmask))
> > goto badframe;
>
> I'm not sure why this eliminates that warning.
> set.sig[0] may not be initialized too, if __copy_from_user() failed.

True, but only when either or both of __copy_from_user() and
(_NSIG_WORDS > 1) fails. But in all instances set.sig[1] gets
initialized.

> I don't have enough time to look at this right now, sorry.
>
> Another question, __copy_from_user() will be called even if
> _NSIG_WORDS is less than 2, perhaps it never occurs.
> I think, to check _NSIG_WORDS > 1 before calling __copy_from_user()
> is better.

Fine. Let Ingo/Thomas/Peter decide whether they would like this fix or
drop it.

Regards--
Subrata

>
> Thanks,
> Hiroshi
>

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2009-05-13 11:09    [W:0.046 / U:1.228 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site