linux-kernel - Re: [patch 14/48] KVM: Reduce kvm stack usage in kvm_arch_vm

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: <1252244589.14385.22402.camel@nimitz>
Date:	Sun, 06 Sep 2009 06:43:09 -0700
From:	Dave Hansen <dave@...ux.vnet.ibm.com>
To:	Pavel Machek <pavel@....cz>
Cc:	Greg KH <gregkh@...e.de>, linux-kernel@...r.kernel.org,
	stable@...nel.org, stable-review@...nel.org,
	torvalds@...ux-foundation.org, akpm@...ux-foundation.org,
	alan@...rguk.ukuu.org.uk, Avi Kivity <avi@...hat.com>,
	Avi Kivity <avi@...ranet.com>
Subject: Re: [patch 14/48] KVM: Reduce kvm stack usage in
 kvm_arch_vm_ioctl()

On Sun, 2009-09-06 at 07:47 +0200, Pavel Machek wrote:
> On Fri 2009-09-04 13:07:26, Greg KH wrote:
> > On my machine with gcc 3.4, kvm uses ~2k of stack in a few
> > select functions.  This is mostly because gcc fails to
> > notice that the different case: statements could have their
> > stack usage combined.  It overflows very nicely if interrupts
> > happen during one of these large uses.
> > 
> > This patch uses two methods for reducing stack usage.
> > 1. dynamically allocate large objects instead of putting
> >    on the stack.
> > 2. Use a union{} member for all of the case variables. This
> >    tricks gcc into combining them all into a single stack
> >    allocation. (There's also a comment on this)
> 
> Are the 'reduce stack usage' patches suitable for stable? The rules
> said that fix must be for 'serious problem', not 'theoretical
> issue'...

I guess some context got dropped at some point.  I was getting really
consistent oopses and goofy memory corruption when running KVM:

	http://lkml.org/lkml/2008/3/25/340

Not theoretical at all.  I think it cost me a few new gray hairs.
 
> > Signed-off-by: Dave Hansen <dave@...ux.vnet.ibm.com>
> > Signed-off-by: Avi Kivity <avi@...ranet.com>
> > Signed-off-by: Greg Kroah-Hartman <gregkh@...e.de>
> > ---
> >  arch/x86/kvm/x86.c |   72 +++++++++++++++++++++++++++++++++--------------------
> >  1 file changed, 45 insertions(+), 27 deletions(-)
> > 
> > --- a/arch/x86/kvm/x86.c
> > +++ b/arch/x86/kvm/x86.c
> > @@ -1630,6 +1630,15 @@ long kvm_arch_vm_ioctl(struct file *filp
> >  	struct kvm *kvm = filp->private_data;
> >  	void __user *argp = (void __user *)arg;
> >  	int r = -EINVAL;
> > +	/*
> > +	 * This union makes it completely explicit to gcc-3.x
> > +	 * that these two variables' stack usage should be
> > +	 * combined, not added together.
> > +	 */
> > +	union {
> > +		struct kvm_pit_state ps;
> > +		struct kvm_memory_alias alias;
> > +	} u;
> >  
> >  	switch (ioctl) {
> >  	case KVM_SET_TSS_ADDR:
> 
> ...plus this is really ugly hack. Just declare the variable inside the
> case block that needs it?

Do we need to give that a better comment?  It's explained a bit better
here:

http://lkml.org/lkml/2008/7/17/12
http://lkml.org/lkml/2008/7/17/16

Would this comment help?

/*
 * gcc-3.x will sum the stack usage of two stack variables
 * if they are declared in two different case blocks.  This
 * union makes it explicit that their stack space can be
 * shared which greatly reduces stack usage.
 */

-- Dave

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