linux-kernel - Re: [kbuild] arch/x86/kvm/x86.c:4988 kvm_arch_tsc_set_attr() warn: check for integer overflow 'offset'

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] [day] [month] [year] [list]

Message-ID: <Y0Zf6g7ahY2b7MBK@kadam>
Date:   Wed, 12 Oct 2022 09:34:18 +0300
From:   Dan Carpenter <dan.carpenter@...cle.com>
To:     Paolo Bonzini <pbonzini@...hat.com>
Cc:     Sean Christopherson <seanjc@...gle.com>, kbuild@...ts.01.org,
        Oliver Upton <oupton@...gle.com>, lkp@...el.com,
        kbuild-all@...ts.01.org, linux-kernel@...r.kernel.org
Subject: Re: [kbuild] arch/x86/kvm/x86.c:4988 kvm_arch_tsc_set_attr() warn:
 check for integer overflow 'offset'

On Tue, Oct 11, 2022 at 03:02:28PM +0200, Paolo Bonzini wrote:
> On 10/10/22 20:39, Sean Christopherson wrote:
> > > 828ca89628bfcb Oliver Upton 2021-09-16 @4988  		tsc = kvm_scale_tsc(vcpu, rdtsc(), vcpu->arch.l1_tsc_scaling_ratio) + offset;
> > > 
> > > Smatch hates obvious user triggerable integer overflows...  No checking
> > > on offset.
> > 
> > This is ok, and even necessary, e.g. if the host TSC > guest TSC.
> 
> (which in fact is the common case).  Also this is unsigned which is fine
> according to the C standard, though I understand that static analyzers want
> to be stricter.
> 
> > Is there anything
> > we can do in KVM to help Smatch avoid false positives?  Or do you/Smatch already
> > maintain a list of known false positives?
> 
> Seconded.

Thanks for your feedback.

I could probably make a rule to ignore clock related stuff.  That's
honestly been a known source of false positives for a while.  I kind of
have the infrastructure so it's not super hard to do actually... I'll do
that.

regards,
dan carpenter