xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Mats Petersson <mats.petersson@citrix.com>
To: xen-devel@lists.xen.org
Subject: Re: [PATCH 1/2, v2] x86: use MOV instead of PUSH/POP when saving/restoring register state
Date: Tue, 30 Oct 2012 15:19:46 +0000	[thread overview]
Message-ID: <508FF012.30901@citrix.com> (raw)
In-Reply-To: <508FF1E202000078000A5786@nat28.tlf.novell.com>

On 30/10/12 14:27, Jan Beulich wrote:
> Signed-off-by: Jan Beulich <jbeulich@suse.com>
> Acked-by: Keir Fraser <keir@xen.org>
> ---
> This patch did not change from its v1 submission.
>
> --- a/xen/arch/x86/x86_64/compat/entry.S
> +++ b/xen/arch/x86/x86_64/compat/entry.S
> @@ -21,8 +21,7 @@ ENTRY(compat_hypercall)
>   UNLIKELY_START(ne, msi_check)
>           movl  $HYPERCALL_VECTOR,%edi
>           call  check_for_unexpected_msi
> -        RESTORE_ALL
> -        SAVE_ALL
> +        LOAD_C_CLOBBERED
>   UNLIKELY_END(msi_check)
>   
>           GET_CURRENT(%rbx)
> @@ -173,8 +172,7 @@ compat_bad_hypercall:
>   /* %rbx: struct vcpu, interrupts disabled */
>   compat_restore_all_guest:
>           ASSERT_INTERRUPTS_DISABLED
> -        RESTORE_ALL
> -        addq  $8,%rsp
> +        RESTORE_ALL adj=8
>   .Lft0:  iretq
>   
>   .section .fixup,"ax"
> --- a/xen/arch/x86/x86_64/entry.S
> +++ b/xen/arch/x86/x86_64/entry.S
> @@ -47,12 +47,10 @@ restore_all_guest:
>           cmpl  $1,%ecx
>           ja    .Lforce_iret
>   
> -        addq  $8,%rsp
> -        popq  %rcx                    # RIP
> -        popq  %r11                    # CS
> -        cmpw  $FLAT_USER_CS32,%r11
> -        popq  %r11                    # RFLAGS
> -        popq  %rsp                    # RSP
> +        cmpw  $FLAT_USER_CS32,16(%rsp)# CS
> +        movq  8(%rsp),%rcx            # RIP
> +        movq  24(%rsp),%r11           # RFLAGS
> +        movq  32(%rsp),%rsp           # RSP
>           je    1f
>           sysretq
>   1:      sysretl
> @@ -101,8 +99,7 @@ failsafe_callback:
>           ALIGN
>   /* No special register assumptions. */
>   restore_all_xen:
> -        RESTORE_ALL
> -        addq  $8,%rsp
> +        RESTORE_ALL adj=8
>           iretq
>   
>   /*
> @@ -311,8 +308,7 @@ ENTRY(int80_direct_trap)
>   UNLIKELY_START(ne, msi_check)
>           movl  $0x80,%edi
>           call  check_for_unexpected_msi
> -        RESTORE_ALL
> -        SAVE_ALL
> +        LOAD_C_CLOBBERED
>   UNLIKELY_END(msi_check)
>   
>           GET_CURRENT(%rbx)
> --- a/xen/include/asm-x86/x86_64/asm_defns.h
> +++ b/xen/include/asm-x86/x86_64/asm_defns.h
> @@ -5,11 +5,11 @@
>   
>   #ifdef CONFIG_FRAME_POINTER
>   /* Indicate special exception stack frame by inverting the frame pointer. */
> -#define SETUP_EXCEPTION_FRAME_POINTER           \
> -        movq  %rsp,%rbp;                        \
> +#define SETUP_EXCEPTION_FRAME_POINTER(offs)     \
> +        leaq  offs(%rsp),%rbp;                  \
>           notq  %rbp
>   #else
> -#define SETUP_EXCEPTION_FRAME_POINTER
> +#define SETUP_EXCEPTION_FRAME_POINTER(off)
Is it too pedantic to say that "off" here should be "offs"?
Just for consistency...

--
Mats

  reply	other threads:[~2012-10-30 15:19 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-02 15:20 [PATCH 0/3] x86: adjust entry frame generation Jan Beulich
2012-10-02 15:26 ` [PATCH 1/3] x86: use MOV instead of PUSH/POP when saving/restoring register state Jan Beulich
2012-10-02 17:01   ` Keir Fraser
2012-10-02 17:57   ` Konrad Rzeszutek Wilk
2012-10-03 13:01     ` Jan Beulich
2012-10-02 15:26 ` [PATCH 2/3] x86: consolidate frame state manipulation functions Jan Beulich
2012-10-02 17:01   ` Keir Fraser
2012-10-02 15:27 ` [PATCH 3/3] x86: save/restore only partial register state where possible Jan Beulich
2012-10-02 17:02   ` Keir Fraser
2012-10-03 13:05     ` Jan Beulich
2012-10-03 14:35       ` Keir Fraser
2012-10-30 14:26         ` Jan Beulich
2012-10-30 14:20 ` [PATCH 0/2, v2] x86: adjust entry frame generation Jan Beulich
2012-10-30 14:27   ` [PATCH 1/2, v2] x86: use MOV instead of PUSH/POP when saving/restoring register state Jan Beulich
2012-10-30 15:19     ` Mats Petersson [this message]
2012-10-30 14:36       ` Keir Fraser
2012-10-30 15:33       ` Jan Beulich
2012-10-30 14:29   ` [PATCH 2/2, v2] x86: save/restore only partial register state where possible Jan Beulich
2012-10-30 14:35   ` [PATCH 0/2, v2] x86: adjust entry frame generation Keir Fraser

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=508FF012.30901@citrix.com \
    --to=mats.petersson@citrix.com \
    --cc=xen-devel@lists.xen.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).