xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Jan Beulich <JBeulich@suse.com>
Cc: xen-devel <xen-devel@lists.xenproject.org>, Keir Fraser <keir@xen.org>
Subject: Re: [PATCH] x86: fix cache flushing condition in map_pages_to_xen()
Date: Wed, 17 Jul 2013 16:40:12 +0100	[thread overview]
Message-ID: <51E6BADC.7000805@citrix.com> (raw)
In-Reply-To: <51E6CFC802000078000E5AB8@nat28.tlf.novell.com>


[-- Attachment #1.1: Type: text/plain, Size: 6568 bytes --]

On 17/07/13 16:09, Jan Beulich wrote:
> This fixes yet another shortcoming of the function (exposed by 8bfaa2c2
> ["x86: add locking to map_pages_to_xen()"]'s adjustment to
> msix_put_fixmap()): It must not flush caches when transitioning to a
> non-present mapping. Doing so causes the CLFLUSH to fault, if used in
> favor of WBINVD.
>
> To help code readability, factor out the whole flush flags updating
> in map_pages_to_xen() into a helper macro.
>
> Signed-off-by: Jan Beulich <jbeulich@suse.com>
>
> --- a/xen/arch/x86/mm.c
> +++ b/xen/arch/x86/mm.c
> @@ -5430,6 +5430,15 @@ l1_pgentry_t *virt_to_xen_l1e(unsigned l
>                           flush_area_local((const void *)v, f) : \
>                           flush_area_all((const void *)v, f))
>  
> +#define flush_flags(oldf) do {                                  \
> +    unsigned int o_ = (oldf);                                   \
> +    if ( (o_) & _PAGE_GLOBAL )                                  \
> +        flush_flags |= FLUSH_TLB_GLOBAL;                        \
> +    if ( (flags & _PAGE_PRESENT) &&                             \
> +         (((o_) ^ flags) & PAGE_CACHE_ATTRS) )                  \
> +        flush_flags |= FLUSH_CACHE;                             \
> +} while (0)
> +

I have to admit to being surprised that the compiler is even happy with
a macro aliasing a variable, but please can it be renamed to something
else (perhaps "set_flush_flags" ?) for the sanity of other humans trying
to read the code.

Furthermore, are we not introducing consistency errors?

Previously, we occasionally decided to flush specific cache lines, and
are now conditionally not flushing the cache lines depending on the
mappings.

Should the fix not be "If we need to flush parts of the cache and dont
have mappings to what we want to flush, use wbinvd()" ?

~Andrew

>  int map_pages_to_xen(
>      unsigned long virt,
>      unsigned long mfn,
> @@ -5465,11 +5474,7 @@ int map_pages_to_xen(
>  
>                  if ( l3e_get_flags(ol3e) & _PAGE_PSE )
>                  {
> -                    if ( l3e_get_flags(ol3e) & _PAGE_GLOBAL )
> -                        flush_flags |= FLUSH_TLB_GLOBAL;
> -                    if ( (lNf_to_l1f(l3e_get_flags(ol3e)) ^ flags) &
> -                         PAGE_CACHE_ATTRS )
> -                        flush_flags |= FLUSH_CACHE;
> +                    flush_flags(lNf_to_l1f(l3e_get_flags(ol3e)));
>                      flush_area(virt, flush_flags);
>                  }
>                  else
> @@ -5481,27 +5486,14 @@ int map_pages_to_xen(
>                          if ( !(l2e_get_flags(ol2e) & _PAGE_PRESENT) )
>                              continue;
>                          if ( l2e_get_flags(ol2e) & _PAGE_PSE )
> -                        {
> -                            if ( l2e_get_flags(ol2e) & _PAGE_GLOBAL )
> -                                flush_flags |= FLUSH_TLB_GLOBAL;
> -                            if ( (lNf_to_l1f(l2e_get_flags(ol2e)) ^ flags) &
> -                                 PAGE_CACHE_ATTRS )
> -                                flush_flags |= FLUSH_CACHE;
> -                        }
> +                            flush_flags(lNf_to_l1f(l2e_get_flags(ol2e)));
>                          else
>                          {
>                              unsigned int j;
>  
>                              pl1e = l2e_to_l1e(ol2e);
>                              for ( j = 0; j < L1_PAGETABLE_ENTRIES; j++ )
> -                            {
> -                                ol1e = pl1e[j];
> -                                if ( l1e_get_flags(ol1e) & _PAGE_GLOBAL )
> -                                    flush_flags |= FLUSH_TLB_GLOBAL;
> -                                if ( (l1e_get_flags(ol1e) ^ flags) &
> -                                     PAGE_CACHE_ATTRS )
> -                                    flush_flags |= FLUSH_CACHE;
> -                            }
> +                                flush_flags(l1e_get_flags(pl1e[j]));
>                          }
>                      }
>                      flush_area(virt, flush_flags);
> @@ -5595,24 +5587,14 @@ int map_pages_to_xen(
>  
>                  if ( l2e_get_flags(ol2e) & _PAGE_PSE )
>                  {
> -                    if ( l2e_get_flags(ol2e) & _PAGE_GLOBAL )
> -                        flush_flags |= FLUSH_TLB_GLOBAL;
> -                    if ( (lNf_to_l1f(l2e_get_flags(ol2e)) ^ flags) &
> -                         PAGE_CACHE_ATTRS )
> -                        flush_flags |= FLUSH_CACHE;
> +                    flush_flags(lNf_to_l1f(l2e_get_flags(ol2e)));
>                      flush_area(virt, flush_flags);
>                  }
>                  else
>                  {
>                      pl1e = l2e_to_l1e(ol2e);
>                      for ( i = 0; i < L1_PAGETABLE_ENTRIES; i++ )
> -                    {
> -                        if ( l1e_get_flags(pl1e[i]) & _PAGE_GLOBAL )
> -                            flush_flags |= FLUSH_TLB_GLOBAL;
> -                        if ( (l1e_get_flags(pl1e[i]) ^ flags) &
> -                             PAGE_CACHE_ATTRS )
> -                            flush_flags |= FLUSH_CACHE;
> -                    }
> +                        flush_flags(l1e_get_flags(pl1e[i]));
>                      flush_area(virt, flush_flags);
>                      free_xen_pagetable(pl1e);
>                  }
> @@ -5687,10 +5669,8 @@ int map_pages_to_xen(
>              if ( (l1e_get_flags(ol1e) & _PAGE_PRESENT) )
>              {
>                  unsigned int flush_flags = FLUSH_TLB | FLUSH_ORDER(0);
> -                if ( l1e_get_flags(ol1e) & _PAGE_GLOBAL )
> -                    flush_flags |= FLUSH_TLB_GLOBAL;
> -                if ( (l1e_get_flags(ol1e) ^ flags) & PAGE_CACHE_ATTRS )
> -                    flush_flags |= FLUSH_CACHE;
> +
> +                flush_flags(l1e_get_flags(ol1e));
>                  flush_area(virt, flush_flags);
>              }
>  
> @@ -5769,6 +5749,8 @@ int map_pages_to_xen(
>      return 0;
>  }
>  
> +#undef flush_flags
> +
>  void destroy_xen_mappings(unsigned long s, unsigned long e)
>  {
>      bool_t locking = system_state > SYS_STATE_boot;
> @@ -5908,6 +5890,8 @@ void destroy_xen_mappings(unsigned long 
>      flush_area(NULL, FLUSH_TLB_GLOBAL);
>  }
>  
> +#undef flush_area
> +
>  void __set_fixmap(
>      enum fixed_addresses idx, unsigned long mfn, unsigned long flags)
>  {
>
>
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel


[-- Attachment #1.2: Type: text/html, Size: 7336 bytes --]

[-- Attachment #2: Type: text/plain, Size: 126 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

  reply	other threads:[~2013-07-17 15:40 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-07-17 15:09 [PATCH] x86: fix cache flushing condition in map_pages_to_xen() Jan Beulich
2013-07-17 15:40 ` Andrew Cooper [this message]
2013-07-17 16:01   ` Jan Beulich
2013-07-17 16:31     ` Keir Fraser
2013-07-18  5:49       ` Jan Beulich
2013-07-18  7:37         ` Keir Fraser
2013-07-17 16:31 ` Keir Fraser
2013-07-17 20:41   ` Sander Eikelenboom

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=51E6BADC.7000805@citrix.com \
    --to=andrew.cooper3@citrix.com \
    --cc=JBeulich@suse.com \
    --cc=keir@xen.org \
    --cc=xen-devel@lists.xenproject.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).