[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 3 Mar 2017 20:06:16 +0800
From: Baoquan He <bhe@...hat.com>
To: Borislav Petkov <bp@...e.de>
Cc: tglx@...utronix.de, hpa@...or.com, mingo@...hat.com,
linux-kernel@...r.kernel.org, x86@...nel.org,
keescook@...omium.org, yinghai@...nel.org, anderson@...hat.com,
luto@...nel.org, thgarnie@...gle.com, kuleshovmail@...il.com
Subject: Re: [PATCH v4 1/3] x86: Introduce a new constant KERNEL_MAPPING_SIZE
On 03/03/17 at 12:43pm, Borislav Petkov wrote:
> On Sun, Feb 26, 2017 at 12:09:08PM +0800, Baoquan He wrote:
> > Am I right on understanding it?
>
> That's exactly what I mean: KERNEL_IMAGE_SIZE is 512M by default but
> we're not hard-constrained to it - we're hard-constrained to a 1G limit
> as this is the 1G which is covered by level2_kernel_pgt.
>
> And in thinking about this more, I know I suggested making the
> KERNEL_IMAGE_SIZE by default 1G in order to simplify things.
>
> But you're adding another KERNEL_MAPPING_SIZE which confuses things
> more. And I fail to see why we absolutely need it.
OK, I am trying to make things clearer, seems I failed. I thought kernel
iamge size is only allowed to be 512M at most, but can be mapped into 1G
region.
>
> So we suggest kernel image size should be 512M but then we still will
> be using a whole 1G mapping for it anyway and a whole page of PMDs at
> level2_kernel_pgt.
>
> So why even bother?
>
> Just make it 1G and don't introduce anything new.
It's fine to me, thing can be solved anyway. Will repost with
KERNEL_IMAGE_SIZE by default 1G.
Thanks
Baoquan
Powered by blists - more mailing lists