All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Jan Beulich <JBeulich@suse.com>,
	xen-devel <xen-devel@lists.xenproject.org>
Cc: Keir Fraser <keir@xen.org>
Subject: Re: [PATCH 1/3] x86: tolerate running on EFI runtime services page tables in map_domain_page()
Date: Thu, 23 Oct 2014 14:57:19 +0100	[thread overview]
Message-ID: <5449093F.6080500@citrix.com> (raw)
In-Reply-To: <5449226E020000780004176D@mail.emea.novell.com>


[-- Attachment #1.1: Type: text/plain, Size: 2503 bytes --]

On 23/10/14 14:44, Jan Beulich wrote:
> In the event of a #PF while inan EFI runtime service function we

"in an"

> otherwise can't dump the page tables, making the analysis of the
> problem more cumbersome.
>
> Signed-off-by: Jan Beulich <jbeulich@suse.com>
>
> --- a/xen/arch/x86/domain_page.c
> +++ b/xen/arch/x86/domain_page.c
> @@ -7,6 +7,7 @@
>   */
>  
>  #include <xen/domain_page.h>
> +#include <xen/efi.h>
>  #include <xen/mm.h>
>  #include <xen/perfc.h>
>  #include <xen/pfn.h>
> @@ -37,11 +38,14 @@ static inline struct vcpu *mapcache_curr
>       */
>      if ( unlikely(pagetable_is_null(v->arch.guest_table)) && is_pv_vcpu(v) )
>      {
> +        unsigned long cr3;
> +

This will need an __maybe_unused to compile in a non-debug build.

>          /* If we really are idling, perform lazy context switch now. */
>          if ( (v = idle_vcpu[smp_processor_id()]) == current )
>              sync_local_execstate();
>          /* We must now be running on the idle page table. */
> -        ASSERT(read_cr3() == __pa(idle_pg_table));
> +        ASSERT((cr3 = read_cr3()) == __pa(idle_pg_table) ||
> +               (efi_enabled && cr3 == efi_rs_page_table()));
>      }
>  
>      return v;
> --- a/xen/arch/x86/efi/stub.c
> +++ b/xen/arch/x86/efi/stub.c
> @@ -9,6 +9,12 @@ const bool_t efi_enabled = 0;
>  
>  void __init efi_init_memory(void) { }
>  
> +paddr_t efi_rs_page_table(void)
> +{
> +    BUG();
> +    return 0;

Is the return strictly needed?  The __builtin_unreachable() in BUG()
should prevent the compiler from complaining.

~Andrew

> +}
> +
>  unsigned long efi_get_time(void)
>  {
>      BUG();
> --- a/xen/common/efi/runtime.c
> +++ b/xen/common/efi/runtime.c
> @@ -98,6 +98,11 @@ void efi_rs_leave(unsigned long cr3)
>      stts();
>  }
>  
> +paddr_t efi_rs_page_table(void)
> +{
> +    return virt_to_maddr(efi_l4_pgtable);
> +}
> +
>  unsigned long efi_get_time(void)
>  {
>      EFI_TIME time;
> --- a/xen/include/xen/efi.h
> +++ b/xen/include/xen/efi.h
> @@ -28,6 +28,7 @@ struct xenpf_efi_runtime_call;
>  struct compat_pf_efi_runtime_call;
>  
>  void efi_init_memory(void);
> +paddr_t efi_rs_page_table(void);
>  unsigned long efi_get_time(void);
>  void efi_halt_system(void);
>  void efi_reset_system(bool_t warm);
>
>
>
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel


[-- Attachment #1.2: Type: text/html, Size: 3502 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:[~2014-10-23 13:58 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-10-23 13:38 [PATCH 0/3] EFI: runtime services related improvements Jan Beulich
2014-10-23 13:44 ` [PATCH 1/3] x86: tolerate running on EFI runtime services page tables in map_domain_page() Jan Beulich
2014-10-23 13:57   ` Andrew Cooper [this message]
2014-10-23 15:18     ` Jan Beulich
2014-10-24  9:52       ` Andrew Cooper
2014-10-23 13:45 ` [PATCH 2/3] EFI: allow to suppress the use of runtime services Jan Beulich
2014-10-23 14:47   ` Andrew Cooper
2014-10-23 13:45 ` [PATCH 3/3] EFI: constify a few table pointers Jan Beulich
2014-10-23 14:48   ` Andrew Cooper

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=5449093F.6080500@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 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.