All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Rusty Russell <rusty@rustcorp.com.au>
Cc: Ingo Molnar <mingo@elte.hu>, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] x86: fix pte_flags() to only return flags, fix lguest.
Date: Mon, 21 Jul 2008 21:49:46 -0700	[thread overview]
Message-ID: <488566EA.5040805@goop.org> (raw)
In-Reply-To: <200807221431.58991.rusty@rustcorp.com.au>

Rusty Russell wrote:
> Change a15af1c9ea2750a9ff01e51615c45950bad8221b 'x86/paravirt: add
> pte_flags to just get pte flags' removed lguest's private pte_flags()
> in favor of a generic one.
>
> Unfortunately, the generic one doesn't filter out the non-flags bits:
> this results in lguest creating corrupt shadow page tables and blowing
> up host memory.
>
> Since noone is supposed to use the pfn part of pte_flags(), it seems
> safest to always do the filtering.
>   

Thinking about this, I wonder if it needs to be a pv_op at all.  
Generality says "yes", but there are no users which set it to anything 
other than native_pte_flags.  The point of it is to return the flags 
as-is, without needing more complex stuff (like Xen's mfn to pfn 
conversion).

In most cases, the surrounding code will be applying its own mask 
anyway, so making it a simple inline will allow all the masks to get 
folded together.

If a user which wants to do something other than return the bare flags, 
it would be easy enough to put the pv_op back.

> Cc: Jeremy Fitzhardinge <jeremy@goop.org>
> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
>
> diff -r ee1a6adad3d2 arch/x86/kernel/paravirt.c
> --- a/arch/x86/kernel/paravirt.c	Mon Jul 21 12:49:25 2008 +1000
> +++ b/arch/x86/kernel/paravirt.c	Tue Jul 22 14:09:33 2008 +1000
> @@ -428,7 +428,7 @@ struct pv_mmu_ops pv_mmu_ops = {
>  #endif /* PAGETABLE_LEVELS >= 3 */
>  
>  	.pte_val = native_pte_val,
> -	.pte_flags = native_pte_val,
> +	.pte_flags = native_pte_flags,
>  	.pgd_val = native_pgd_val,
>  
>  	.make_pte = native_make_pte,
> diff -r ee1a6adad3d2 include/asm-x86/page.h
> --- a/include/asm-x86/page.h	Mon Jul 21 12:49:25 2008 +1000
> +++ b/include/asm-x86/page.h	Tue Jul 22 14:09:33 2008 +1000
> @@ -144,6 +144,18 @@ static inline pteval_t native_pte_val(pt
>  	return pte.pte;
>  }
>  
> +/* This belongs in pgtable.h, but our includes are too much of a mess. */
> +#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
> +#define PTE_FLAGS	_AC(0x8000000000000FFF, ULL)
> +#else
> +#define PTE_FLAGS	0x00000FFF
> +#endif
>   
We already have this, it's called PTE_MASK.

    J

  parent reply	other threads:[~2008-07-22  4:50 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-07-22  4:31 [PATCH] x86: fix pte_flags() to only return flags, fix lguest Rusty Russell
2008-07-22  4:38 ` Stephen Rothwell
2008-07-22  4:51   ` Jeremy Fitzhardinge
2008-07-22  4:52   ` Rusty Russell
2008-07-22  4:49 ` Jeremy Fitzhardinge [this message]
2008-07-22  5:40   ` [PATCH] x86: fix pte_flags() to only return flags, fix lguest (updated) Rusty Russell
2008-07-22  5:59     ` [PATCH 1/2] x86: rename PTE_MASK to PTE_PFN_MASK Jeremy Fitzhardinge
2008-07-22  8:36       ` Ingo Molnar
2008-07-22 10:58         ` Rusty Russell
2008-07-22 11:55           ` Ingo Molnar
2008-07-22 13:03       ` Johannes Weiner
2008-07-22 14:52         ` Jeremy Fitzhardinge
2008-07-22 15:18           ` Johannes Weiner
2008-07-22 15:23             ` Johannes Weiner
2008-07-22 15:33               ` Ingo Molnar
2008-07-22 15:43                 ` Johannes Weiner
2008-07-22  5:59     ` [PATCH 2/2] x86: add PTE_FLAGS_MASK Jeremy Fitzhardinge
2008-07-22  9:04     ` [PATCH] x86: fix pte_flags() to only return flags, fix lguest (updated) Ingo Molnar
2008-07-23  0:59       ` Rusty Russell
2008-07-24 11:31         ` Ingo Molnar
2008-07-25  1:55           ` Rusty Russell
2008-07-28 15:11             ` Ingo Molnar

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=488566EA.5040805@goop.org \
    --to=jeremy@goop.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=rusty@rustcorp.com.au \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.