thr3ads.net - Xen devel - [Xen-devel] About TSS initialization when Xen booting. [Aug 2007]

If this information is useful, please help other people find it:
Share via:

李亚琼

2007-Aug-06 12:45 UTC

[Xen-devel] About TSS initialization when Xen booting.

Hi,

         I have a doubt on the Tss initialization. In the function
cpu_init(),t->rsp0 = get_stack_bottom().

#define get_stack_bottom()                      \

           ((unsigned long)&get_cpu_info()->guest_cpu_user_regs.es)

         But I can’t find where the guest_cpu_user_regs has been
initialized. Anybody can help me?

 

-Techie.lee

 

 



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

Li, Xin B

2007-Aug-06 15:49 UTC

head link

RE: [Xen-devel] About TSS initialization when Xen booting.

in __context_switch, there is a memcpy to do this job.
-Xin 
 


________________________________

	From: xen-devel-bounces@lists.xensource.com
[mailto:xen-devel-bounces@lists.xensource.com] On Behalf Of 李亚琼
	Sent: Monday, August 06, 2007 8:46 PM
	To: xen-devel@lists.xensource.com
	Subject: [Xen-devel] About TSS initialization when Xen booting.
	
	

	Hi,

	         I have a doubt on the Tss initialization. In the function
cpu_init(),t->rsp0 = get_stack_bottom().

	#define get_stack_bottom()                      \

	           ((unsigned long)&get_cpu_info()->guest_cpu_user_regs.es)

	         But I can’t find where the guest_cpu_user_regs has been initialized.
Anybody can help me?

	 

	-Techie.lee

	 

	 



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

李亚琼

2007-Aug-07 01:27 UTC

head link

答复: [Xen-devel] About TSS initialization when Xen booting.

Thanks, Xin!

But __context_switch will be invoked only when system is scheduled. During
the initialization, Xen invokes percpu_traps_init and cpu_init to initialize
a physical cpu. In this two functions, tss will be assigned with a value
from guest_cpu_user_regs. But till this time, __context_switch has never
been invoked. In other words, idle_domain’s tss is assigned with a trivial
value. After all, when Xen is in the idle domain, a stack switch will occur
when a interrupt occurs. Is it correct?

Another question is about percpu_traps_init whick code is following:

void __init percpu_traps_init(void)

{

char *stack_bottom, *stack;

.

.

.

    stack_bottom = (char *)get_stack_bottom();

    stack        = (char *)((unsigned long)stack_bottom & ~(STACK_SIZE -
1));

 

    /* Double-fault handler has its own per-CPU 2kB stack. */

    init_tss[cpu].ist[0] = (unsigned long)&stack[2048];

 

    /* NMI handler has its own per-CPU 1kB stack. */

    init_tss[cpu].ist[1] = (unsigned long)&stack[3072];

 

    /*

     * Trampoline for SYSCALL entry from long mode.

     */

 

    /* Skip the NMI and DF stacks. */

    stack = &stack[3072];

    wrmsr(MSR_LSTAR, (unsigned long)stack, ((unsigned long)stack>>32));

 

    /* movq %rsp, saversp(%rip) */

    stack[0] = 0x48;

    stack[1] = 0x89;

    stack[2] = 0x25;

*(u32 *)&stack[3] = (stack_bottom - &stack[7]) - 16;

.

.

.

}

The question is about the variables,  “stack”  and “stack_bottom”. At
this moment, if guest_cpu_user_regs has not been initialized,
“stack_bottom” will be zero when the cpu is the first physical cpu ( its
stack is initialized with zero in file boot.S). And then “stack” will be
zero, too. Is it correct?

 

-Techie, lee

2007-8-7

  _____  

发件人: Li, Xin B [mailto:xin.b.li@intel.com] 
发送时间: 2007年8月6日 23:50
收件人: 李亚琼; xen-devel@lists.xensource.com
主题: RE: [Xen-devel] About TSS initialization when Xen booting.

 

In __context_switch, there is a memcpy to do this job.

-Xin 

 

 


  _____  


From: xen-devel-bounces@lists.xensource.com [mailto:xen-devel-bounces@lists.
xensource.com] On Behalf Of 李亚琼
Sent: Monday, August 06, 2007 8:46 PM
To: xen-devel@lists.xensource.com
Subject: [Xen-devel] About TSS initialization when Xen booting.

Hi,

         I have a doubt on the Tss initialization. In the function
cpu_init(),t->rsp0 = get_stack_bottom().

#define get_stack_bottom()                      \

           ((unsigned long)&get_cpu_info()->guest_cpu_user_regs.es)

         But I can’t find where the guest_cpu_user_regs has been
initialized. Anybody can help me?

 

-Techie.lee

 

 



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

Ian Campbell

2007-Aug-07 06:52 UTC

head link

Re: 答复: [Xen-devel] About TSS initialization when Xen booting.

On Tue, 2007-08-07 at 09:27 +0800, 李亚琼 wrote:> Thanks, Xin!
> 
> But __context_switch will be invoked only when system is scheduled.
> During the initialization, Xen invokes percpu_traps_init and cpu_init
> to initialize a physical cpu. In this two functions, tss will be
> assigned with a value from guest_cpu_user_regs. But till this time,
> __context_switch has never been invoked. In other words, idle_domain’s
> tss is assigned with a trivial value. After all, when Xen is in the
> idle domain, a stack switch will occur when a interrupt occurs. Is it
> correct?
get_cpu_info returns a pointer into the stack. guest_cpu_user_regs is a
member of the structure this pointer points to (but not a pointer
itself). It has been arranged so that guest_cpu_user_regs overlaps the
part of the stack where entry.S will save the guests registers (mainly
in the SAVE_ALL macro).

This should make sense because the TSS is being initialised with exactly
the value you wish the stack to have when transitioning to ring0.
> The question is about the variables,  “stack”  and “stack_bottom”. At
> this moment, if guest_cpu_user_regs has not been initialized,
> “stack_bottom” will be zero when the cpu is the first physical cpu
> ( its stack is initialized with zero in file boot.S). And then “stack”
> will be zero, too. Is it correct?
Remember that guest_cpu_user_regs isn''t a pointer, it''s just a
member of
a struct so it''s address will currently be the address of that struct
plus an offset.

Ian.



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

李亚琼

2007-Aug-07 07:20 UTC

head link

答复: 答复: [Xen-devel] About TSSinitialization when Xen booting.

Hi,Ian
	I know what you mean. But the problem is that guest_cpu_user_regs.es
(the macro of get_frame_bottom() returns this value) is used to initialize
the TSS when it has not been initialized. If TSS has not been initialized
correctly, Xen can''t switch to the kernel stack of the current cpu when
a
interrupt or trap occurs. Tss initialization should correctly be initialized
before a interrupt or trap occurs. But I can''t find where
guest_cpu_user_regs.es is initialized before used to initialize TSS in
function percpu_traps_init().
	Hope your reply!

-Techie.lee
2007-8-7

-----邮件原件-----
发件人: Ian Campbell [mailto:Ian.Campbell@XenSource.com] 
发送时间: 2007年8月7日 14:53
收件人: 李亚琼
抄送: ''Li, Xin B''; 
主题: Re: 答复: [Xen-devel] About TSSinitialization when Xen booting.

On Tue, 2007-08-07 at 09:27 +0800, 李亚琼 wrote:> Thanks, Xin!
> 
> But __context_switch will be invoked only when system is scheduled.
> During the initialization, Xen invokes percpu_traps_init and cpu_init
> to initialize a physical cpu. In this two functions, tss will be
> assigned with a value from guest_cpu_user_regs. But till this time,
> __context_switch has never been invoked. In other words, idle_domain’s
> tss is assigned with a trivial value. After all, when Xen is in the
> idle domain, a stack switch will occur when a interrupt occurs. Is it
> correct?
get_cpu_info returns a pointer into the stack. guest_cpu_user_regs is a
member of the structure this pointer points to (but not a pointer
itself). It has been arranged so that guest_cpu_user_regs overlaps the
part of the stack where entry.S will save the guests registers (mainly
in the SAVE_ALL macro).

This should make sense because the TSS is being initialised with exactly
the value you wish the stack to have when transitioning to ring0.
> The question is about the variables,  “stack”  and “stack_bottom”. At
> this moment, if guest_cpu_user_regs has not been initialized,
> “stack_bottom” will be zero when the cpu is the first physical cpu
> ( its stack is initialized with zero in file boot.S). And then “stack”
> will be zero, too. Is it correct?
Remember that guest_cpu_user_regs isn''t a pointer, it''s just a
member of
a struct so it''s address will currently be the address of that struct
plus an offset.

Ian.





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

Ian Campbell

2007-Aug-07 07:30 UTC

head link

Re: 答复: 答复: [Xen-devel] About TSSinitialization when Xen booting.

On Tue, 2007-08-07 at 15:20 +0800, 李亚琼 wrote:> Hi,Ian
> 	I know what you mean. But the problem is that guest_cpu_user_regs.es
> (the macro of get_frame_bottom() returns this value) is used to initialize
> the TSS when it has not been initialized.
It''s the _address_ of guest_cpu_user_regs.es, isn''t it?

((unsigned long)&get_cpu_info()->guest_cpu_user_regs.es)
                ^
>  If TSS has not been initialized
> correctly, Xen can''t switch to the kernel stack of the current cpu
when a
> interrupt or trap occurs. Tss initialization should correctly be
initialized
> before a interrupt or trap occurs. But I can''t find where
> guest_cpu_user_regs.es is initialized before used to initialize TSS in
> function percpu_traps_init().
The TSS setup in percpu_traps_init() is for the double fault TSS. The
regular TSS is setup in cpu_init using init_tss[cpu].

Ian.



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

Xen devel - Aug 2007 - About TSS initialization when Xen booting.

[Xen-devel] About TSS initialization when Xen booting.

RE: [Xen-devel] About TSS initialization when Xen booting.

答复: [Xen-devel] About TSS initialization when Xen booting.

Re: 答复: [Xen-devel] About TSS initialization when Xen booting.

答复: 答复: [Xen-devel] About TSSinitialization when Xen booting.

Re: 答复: 答复: [Xen-devel] About TSSinitialization when Xen booting.