Re: Rethinking sigcontext's xfeatures slightly for PKRU's benefit?

From: Dave Hansen <dave.hansen@linux.intel.com>
To: Andy Lutomirski <luto@amacapital.net>
Cc: Thomas Gleixner <tglx@linutronix.de>,
	Borislav Petkov <bp@alien8.de>, Oleg Nesterov <oleg@redhat.com>,
	Ingo Molnar <mingo@kernel.org>, Christoph Hellwig <hch@lst.de>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Brian Gerst <brgerst@gmail.com>, "H. Peter Anvin" <hpa@zytor.com>,
	Linus Torvalds <torvalds@linux-foundation.org>
Subject: Re: Rethinking sigcontext's xfeatures slightly for PKRU's benefit?
Date: Mon, 21 Dec 2015 15:00:08 -0800	[thread overview]
Message-ID: <56788478.9050805@linux.intel.com> (raw)
In-Reply-To: <CALCETrW26qunk5CatBXXsTL4oBPoQ05q9Rch=tkbnKJU8GjJug@mail.gmail.com>

On 12/21/2015 02:52 PM, Andy Lutomirski wrote:
> Perhaps this is silly, but what if the default were changed to deny
> reads and writes for unallocated keys?  Is there a use case that
> breaks?

It's probably a reasonable debugging feature.

But, anything that takes an XSAVE feature out of its "init state" has
the potential to do a bit of harm because it increases the potential
size of writes during XSAVE.  XSAVEOPT will _help_ here, but we probably
don't want to go out of our way to take things out of the init state
when we're unsure of the benefits.