linux-kernel - Re: [PATCH v2] x86-64: use 32-bit XOR to zero registers

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]

Message-ID: <20180709083359.GB7312@amd>
Date:   Mon, 9 Jul 2018 10:33:59 +0200
From:   Pavel Machek <pavel@....cz>
To:     Ingo Molnar <mingo@...nel.org>
Cc:     Jan Beulich <JBeulich@...e.com>, mingo@...e.hu, tglx@...utronix.de,
        hpa@...or.com, davem@...emloft.net, herbert@...dor.apana.org.au,
        rjw@...ysocki.net, Juergen Gross <jgross@...e.com>,
        linux-kernel@...r.kernel.org, Alok Kataria <akataria@...are.com>
Subject: Re: [PATCH v2] x86-64: use 32-bit XOR to zero registers

On Thu 2018-07-05 09:12:16, Ingo Molnar wrote:
> 
> * Pavel Machek <pavel@....cz> wrote:
> 
> > On Mon 2018-07-02 04:31:54, Jan Beulich wrote:
> > > Some Intel CPUs don't recognize 64-bit XORs as zeroing idioms. Zeroing
> > > idioms don't require execution bandwidth, as they're being taken care
> > > of in the frontend (through register renaming). Use 32-bit XORs instead.
> > > 
> > > Signed-off-by: Jan Beulich <jbeulich@...e.com>
> > 
> > > @@ -702,7 +702,7 @@ _no_extra_mask_1_\@:
> > >  
> > >  	# GHASH computation for the last <16 Byte block
> > >  	GHASH_MUL \AAD_HASH, %xmm13, %xmm0, %xmm10, %xmm11, %xmm5, %xmm6
> > > -	xor	%rax,%rax
> > > +	xor	%eax, %eax
> > >  
> > >  	mov	%rax, PBlockLen(%arg2)
> > >  	jmp	_dec_done_\@
> > 
> > This is rather subtle... and looks like a bug. To zero 64-bit
> > register, you zero its lower half, relying on implicit zeroing of the
> > upper half. Wow.
> > 
> > Perhaps we should get comments in the code? Because the explicit code
> > is more readable...
> 
> The automatic zero-extension of 32-bit ops to the full 64-bit register is a basic, 
> fundamental and well-known x86-64 idiom in use in literally hundreds of places in 
> x86-64 assembly code.
> 
> We sometimes document zero-extension on entry boundaries where we want to make it 
> really clear what information gets (and what doesn't get) into the kernel, but 
> generally it only needs documentation is the (very rare) cases where it's *not* 
> done.
> 
> Also, why would it be a bug?

Not a bug, just looks like one :-). That's why I'd add a comment.

Anyway, normally assembler is the one who chooses instruction
encoding.

xor %rax, %rax is equivalent to xor %eax, %eax; first variant is
slower on some CPUs, second variant will take one extra byte due to
operand size prefix IIRC... Should the assembler generate right
variant according to the CPU type?
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

Download attachment "signature.asc" of type "application/pgp-signature" (182 bytes)