Re: [Xen-devel] [PATCH] x86/spec_ctrl: Fix several bugs in SPEC_CTRL_ENTRY_FROM_INTR_IST

2018-02-14 Thread Andrew Cooper 
On 14/02/18 11:47, Roger Pau Monné wrote:
>
>> diff --git a/xen/include/asm-x86/spec_ctrl_asm.h 
>> b/xen/include/asm-x86/spec_ctrl_asm.h
>> index 814f53d..1f78599 100644
>> --- a/xen/include/asm-x86/spec_ctrl_asm.h
>> +++ b/xen/include/asm-x86/spec_ctrl_asm.h
>> @@ -79,10 +79,10 @@
>>   *  - SPEC_CTRL_EXIT_TO_GUEST
>>   */
>>  
>> -.macro DO_OVERWRITE_RSB
>> +.macro DO_OVERWRITE_RSB tmp=%rax
>>  /*
>>   * Requires nothing
>> - * Clobbers %rax, %rcx
>> + * Clobbers \tmp (%rax by default), %rcx
> Is it worth changing %rcx to %ecx here...

No, because a mov to %ecx clobbers %rcx.  I'll adjust the comment.

~Andrew

___
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH] x86/spec_ctrl: Fix several bugs in SPEC_CTRL_ENTRY_FROM_INTR_IST

2018-02-14 Thread Wei Liu 
On Wed, Feb 14, 2018 at 11:10:55AM +, Andrew Cooper wrote:
> DO_OVERWRITE_RSB clobbers %rax, meaning in practice that that the bti_ist_info
> field gets zeroed.  Older versions of this code had the DO_OVERWRITE_RSB
> register selectable, so reintroduce this ability and use it to cause the
> INTR_IST path to use %rdx instead.
> 
> The use of %dl for the %cs.rpl check means that when an IST interrupt hits
> Xen, we try to load 1 into the high 32 bits of MSR_SPEC_CTRL, suffering a #GP
> fault instead.
> 
> Also, drop an unused label which was a copy/paste mistake.
> 
> Reported-by: Boris Ostrovsky 
> Reported-by: Zhenzhong Duan 
> Signed-off-by: Andrew Cooper 

Reviewed-by: Wei Liu 

___
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH] x86/spec_ctrl: Fix several bugs in SPEC_CTRL_ENTRY_FROM_INTR_IST

2018-02-14 Thread Roger Pau Monné 
On Wed, Feb 14, 2018 at 11:10:55AM +, Andrew Cooper wrote:
> DO_OVERWRITE_RSB clobbers %rax, meaning in practice that that the bti_ist_info
   ^dup
> field gets zeroed.  Older versions of this code had the DO_OVERWRITE_RSB
> register selectable, so reintroduce this ability and use it to cause the
> INTR_IST path to use %rdx instead.
> 
> The use of %dl for the %cs.rpl check means that when an IST interrupt hits
> Xen, we try to load 1 into the high 32 bits of MSR_SPEC_CTRL, suffering a #GP
> fault instead.
> 
> Also, drop an unused label which was a copy/paste mistake.
> 
> Reported-by: Boris Ostrovsky 
> Reported-by: Zhenzhong Duan 
> Signed-off-by: Andrew Cooper 

Reviewed-by: Roger Pau Monné 

> ---
> CC: Jan Beulich 
> CC: Zhenzhong Duan 
> CC: Boris Ostrovsky 
> CC: Wei Liu 
> CC: Roger Pau Monné 
> ---
>  xen/include/asm-x86/spec_ctrl_asm.h | 12 ++--
>  1 file changed, 6 insertions(+), 6 deletions(-)
> 
> diff --git a/xen/include/asm-x86/spec_ctrl_asm.h 
> b/xen/include/asm-x86/spec_ctrl_asm.h
> index 814f53d..1f78599 100644
> --- a/xen/include/asm-x86/spec_ctrl_asm.h
> +++ b/xen/include/asm-x86/spec_ctrl_asm.h
> @@ -79,10 +79,10 @@
>   *  - SPEC_CTRL_EXIT_TO_GUEST
>   */
>  
> -.macro DO_OVERWRITE_RSB
> +.macro DO_OVERWRITE_RSB tmp=%rax
>  /*
>   * Requires nothing
> - * Clobbers %rax, %rcx
> + * Clobbers \tmp (%rax by default), %rcx

Is it worth changing %rcx to %ecx here...

>   *
>   * Requires 256 bytes of stack space, but %rsp has no net change. Based on
>   * Google's performance numbers, the loop is unrolled to 16 iterations and 
> two
> @@ -97,7 +97,7 @@
>   * optimised with mov-elimination in modern cores.
>   */
>  mov $16, %ecx   /* 16 iterations, two calls per loop */
> -mov %rsp, %rax  /* Store the current %rsp */
> +mov %rsp, \tmp  /* Store the current %rsp */
>  
>  .L\@_fill_rsb_loop:
>  
> @@ -114,7 +114,7 @@
>  
>  sub $1, %ecx
>  jnz .L\@_fill_rsb_loop
> -mov %rax, %rsp  /* Restore old %rsp */
> +mov \tmp, %rsp  /* Restore old %rsp */
>  .endm
>  
>  .macro DO_SPEC_CTRL_ENTRY_FROM_VMEXIT ibrs_val:req
> @@ -274,7 +274,7 @@
>  testb $BTI_IST_RSB, %al
>  jz .L\@_skip_rsb
>  
> -DO_OVERWRITE_RSB
> +DO_OVERWRITE_RSB tmp=%rdx /* Clobbers %ecx/%rdx */

... to match the comment here?

Thanks, Roger.

___
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: [Xen-devel] [PATCH] x86/spec_ctrl: Fix several bugs in SPEC_CTRL_ENTRY_FROM_INTR_IST

2018-02-14 Thread Jan Beulich 
>>> On 14.02.18 at 12:10,  wrote:
> DO_OVERWRITE_RSB clobbers %rax, meaning in practice that that the bti_ist_info
> field gets zeroed.  Older versions of this code had the DO_OVERWRITE_RSB
> register selectable, so reintroduce this ability and use it to cause the
> INTR_IST path to use %rdx instead.
> 
> The use of %dl for the %cs.rpl check means that when an IST interrupt hits
> Xen, we try to load 1 into the high 32 bits of MSR_SPEC_CTRL, suffering a #GP
> fault instead.
> 
> Also, drop an unused label which was a copy/paste mistake.
> 
> Reported-by: Boris Ostrovsky 
> Reported-by: Zhenzhong Duan 
> Signed-off-by: Andrew Cooper 

Reviewed-by: Jan Beulich 
with one remark/suggestion:

> --- a/xen/include/asm-x86/spec_ctrl_asm.h
> +++ b/xen/include/asm-x86/spec_ctrl_asm.h
> @@ -79,10 +79,10 @@
>   *  - SPEC_CTRL_EXIT_TO_GUEST
>   */
>  
> -.macro DO_OVERWRITE_RSB
> +.macro DO_OVERWRITE_RSB tmp=%rax

If only registers are supposed to be passed as macro arguments,
I generally consider it better to leave specifying at least the % to
the macro body - that way nothing else (like a memory operand,
perhaps referencing %rsp in a dangerous way) can be passed, as
the assembler will choke on invalid names following %. In order to
have the flexibility to also use the 32-bit register name inside the
macro, I would also generally omit the r (but this is less of a concern
here, as it's quite unlikely for the 32-bit name to be needed).

Jan

Jan


___
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel