linux-kernel - Re: [PATCH 01/12] x86_64: memset

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:   Fri, 21 Dec 2018 12:29:46 -0800
From:   Matthew Wilcox <willy@...radead.org>
To:     Cyrill Gorcunov <gorcunov@...il.com>
Cc:     Igor Stoppa <igor.stoppa@...il.com>,
        Andy Lutomirski <luto@...capital.net>,
        Peter Zijlstra <peterz@...radead.org>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Mimi Zohar <zohar@...ux.vnet.ibm.com>,
        Thiago Jung Bauermann <bauerman@...ux.ibm.com>,
        igor.stoppa@...wei.com, Nadav Amit <nadav.amit@...il.com>,
        Kees Cook <keescook@...omium.org>,
        Ahmed Soliman <ahmedsoliman@...a.vt.edu>,
        linux-integrity@...r.kernel.org,
        kernel-hardening@...ts.openwall.com, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH 01/12] x86_64: memset_user()

On Fri, Dec 21, 2018 at 11:05:46PM +0300, Cyrill Gorcunov wrote:
> On Fri, Dec 21, 2018 at 10:25:16AM -0800, Matthew Wilcox wrote:
> > On Fri, Dec 21, 2018 at 08:14:12PM +0200, Igor Stoppa wrote:
> > > +unsigned long __memset_user(void __user *addr, int c, unsigned long size)
> > > +{
> > > +	long __d0;
> > > +	unsigned long  pattern = 0;
> > > +	int i;
> > > +
> > > +	for (i = 0; i < 8; i++)
> > > +		pattern = (pattern << 8) | (0xFF & c);
> > 
> > That's inefficient.
> > 
> > 	pattern = (unsigned char)c;
> > 	pattern |= pattern << 8;
> > 	pattern |= pattern << 16;
> > 	pattern |= pattern << 32;
> 
> Won't
> 
> 	pattern = 0x0101010101010101 * c;
> 
> do the same but faster?

Depends on your CPU.  Some yes, some no.

(Also you need to cast 'c' to unsigned char to avoid someone passing in
0x1234 and getting 0x4646464646464634 instead of 0x3434343434343434)