linux-kernel - Re: [PATCH v3 3/3] x86: Support compiling out userspace I/O (iopl and ioperm)

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: <20141029171754.GA18888@cloud>
Date:	Wed, 29 Oct 2014 10:17:54 -0700
From:	josh@...htriplett.org
To:	Kees Cook <keescook@...omium.org>
Cc:	"H. Peter Anvin" <hpa@...or.com>, Ingo Molnar <mingo@...hat.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	LKML <linux-kernel@...r.kernel.org>,
	"virtualization@...ts.linux-foundation.org" 
	<virtualization@...ts.linux-foundation.org>,
	"x86@...nel.org" <x86@...nel.org>, xen-devel@...ts.xenproject.org
Subject: Re: [PATCH v3 3/3] x86: Support compiling out userspace I/O (iopl
 and ioperm)

On Wed, Oct 29, 2014 at 09:59:25AM -0700, Kees Cook wrote:
> On Wed, Oct 29, 2014 at 9:10 AM, Josh Triplett <josh@...htriplett.org> wrote:
> > --- a/arch/x86/kernel/process-io.h
> > +++ b/arch/x86/kernel/process-io.h
> > @@ -1,9 +1,17 @@
> >  #ifndef _X86_KERNEL_PROCESS_IO_H
> >  #define _X86_KERNEL_PROCESS_IO_H
> >
> > +static inline void clear_thread_io_bitmap(struct task_struct *p)
> > +{
> > +#ifdef CONFIG_X86_IOPORT
> > +       p->thread.io_bitmap_ptr = NULL;
> > +#endif /* CONFIG_X86_IOPORT */
> > +}
> 
> Personally, I prefer seeing these kinds of optional functions declared
> in a single block rather than having the #ifdefs inside the functions:
> 
> #ifdef CONFIG_X86_IOPORT
> static inline void clear_thread_io_bitmap(struct task_struct *p)
> {
>     ...
> }
> 
> static inline int copy_io_bitmap(struct task_struct *me,
>                                   struct task_struct *p)
> {
>     ...
> }
> 
> ...remaining_functions...
> 
> #else
> static inline void clear_thread_io_bitmap(struct task_struct *p) { }
> static inline int copy_io_bitmap(struct task_struct *me,
>                                   struct task_struct *p)
> {
>     return 0;
> }
> ...remaining functions...
> #endif /* CONFIG_X86_IOPORT */
> 
> But this is entirely a style decision, so I leave it up to the x86
> maintainers ...

I can certainly do that if the x86 maintainers prefer, but that tends to
produce a net increase in lines of code, as well as duplicating all the
function prototypes, which to me seems more error-prone.  If the
stub versions contained any code, rather than just becoming no-ops, I'd
definitely do that.

> Another nit may be that we should call this CONFIG_SYSCALL_IOPL or
> CONFIG_SYSCALL_IOPERM in keeping with the other CONFIG_SYSCALL_*
> naming thread? Again, I don't really care strongly beyond really
> wanting to use this new feature! :)

I don't feel strongly about the naming.  Ingo?

> Thanks for working on this!

No problem.  I look forward to seeing it used, in Chrome OS and
elsewhere. :)

- Josh Triplett
--
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/