[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] Branch Trace Storage for guests andVPMUinitialization

On 02/26/2015 12:57 PM, Kevin.Mayer@xxxxxxxx wrote:

-----UrsprÃngliche Nachricht-----
Von: Boris Ostrovsky [mailto:boris.ostrovsky@xxxxxxxxxx]
Gesendet: Donnerstag, 26. Februar 2015 17:35
An: Dietmar Hahn; xen-devel@xxxxxxxxxxxxx
Cc: Mayer, Kevin
Betreff: Re: [Xen-devel] Branch Trace Storage for guests and

On 02/26/2015 03:56 AM, Dietmar Hahn wrote:
Am Mittwoch 25 Februar 2015, 11:31:31 schrieb Boris Ostrovsky:
On 02/25/2015 10:12 AM, Kevin.Mayer@xxxxxxxx wrote:
-----UrsprÃngliche Nachricht-----
Von: Boris Ostrovsky [mailto:boris.ostrovsky@xxxxxxxxxx]
Gesendet: Dienstag, 24. Februar 2015 18:13
An: Mayer, Kevin; xen-devel@xxxxxxxxxxxxx
Betreff: Re: [Xen-devel] Branch Trace Storage for guests and VPMU

On 02/24/2015 10:27 AM, Kevin.Mayer@xxxxxxxx wrote:
Hi guys

I`m trying to set up the BTS so that I can log the branches taken
in the guest using Xen 4.4.1 with a WinXP SP3 guest on a Core i7
Sandy Bridge.

I added the vpmu=bts boot parameter to my grub2 configuration and
extended the libxl,libxc,domctl,â with an own command so that I
can trigger the activation of the BTS whenever I want.

I am not sure why you are doing all these changes to Xen code. BTS
is supposed to be managed from the guest. For example, a Fedora
guest will produce this:

[root@dhcp-burlington7-2nd-B-east-10-152-55-140 ~]# perf record -e
branches:u -c 1 -d sleep 1 [ perf record: Woken up 3838 times to
write data ] [ perf record: Captured and wrote 0.704 MB perf.data
(~30756 samples) ]
[root@dhcp-burlington7-2nd-B-east-10-152-55-140 ~]# perf script -f
ip,addr,sym,dso,symoff --show-kernel-path
     ffffffff8167c347 native_irq_return_iret+0x0 (/proc/kcore) =>
328c001590 [unknown] (/proc/kcore)
     ffffffff8167c347 native_irq_return_iret+0x0 (/proc/kcore) =>
328c001590 [unknown] ([unknown])
           328c001593 [unknown] ([unknown]) =>       328c004b70 [unknown]

I want to be able to log the taken branches (of the guest) without the
need to modify the guest at all.
This means I have to do all the logic in the hypervisor, or am I wrong?
In that case, yes. But then you have to make sure that at least
    * you don't load guest's VPMU (or, at least, BTS-related
registers) on context switch
But you need to modify PMU registers when switching to/from the guest
context to get PMU running.

I was thinking that all BTS stuff can be controlled from dom0 and so we can
use dom0's version of these registers. I didn't realize that DS_AREA would
have to be accessed in guest's address space (and that DEBUGCTL is loaded
from VMCS).

Which is what I think I said in response to this message (which didn't show up
on the list because Kevin accidentally dropped xen-devel).


Terribly sorry about that...

So the VPMU doesnât get loaded when there is a VMENTER?

Not exactly. For BTS, DEBUGCTL register, which lives in VMCS, does get loaded. But not DS_AREA --- it gets loaded by SW during context_switch()->vpmu_load().

(As for general VPMU registers such as counters --- they are also loaded during context_switch(). But I don't think you care about those. From what little I know about BTS, DEBUGCTL and DS_AREA are the only two registers you are interested in)

I thought I could set the domU->vcpu->vpmu to enable BTS while in dom0 (with 
modified versions of msr_write_intercept, vpmu_do_wrmsr and core2_vpmu_do_wrmsr of 
course since the build in ones use the current-vcpu which would be the dom0-vcpu)
and as soon as there is a context switch to domU the vpmu gets loaded and the 
guest starts logging.

And it should work, provided that DS_AREA is set up correctly.

If the described behavior is correct the only problem I can see is with 
allocating memory in dom0 in a way that the guest can access it.

This sounds right. All you have to do now is implementation details ;-)


But if I got it wrong please explain how the vpmu really works.



I didn't think of using the VPMU stuff with modifying the context from
outside the guest.

    * You don't send the interrupt to the guest (meaning that you will
need to somehow inform dom0 of the BTS interrupt)

and probably more.

Essentially, you want dom0 to profile the guest. I have been working
on patches that would allow that but they are still under review.

In this command I do the following:

I set up the memory region for the BTS Buffer and the DS Buffer
Management Area using xzalloc_bytes

I don't think you should be allocating BTS buffers in the
hypervisor, they are in guest's memory.
I agree. As I said I think this is where my main problem is at the moment.
Is there any way I can allocate memory in the hypervisor in a way the
guest can access it?
I am not sure this is what you want since you seem to *not* want the
guest to process the samples, right?

But yes, you can. E.g. something like what map_vcpu_info() does. (I
have no idea how you'd do this from Windows.)
The DS buffer has to be mapped within the guests address space so the
CPU running in guest context can access this area. Otherwise you get
this triple fault.
So I would think you need a mixture of writing some stuff in Windows
and patching the hypervisor.


Of course the guest must not be able to use this memory in its normal
operations but just for BTS.
Is this even possible? I am rather confused at the moment. :-D

Then I write the pointer to the BTS Buffer into the DS Buffer
Management Area at +0x0 and +0x8 (BTS Buffer Base and BTS Index)

When I use vmx_msr_write_intercept to store the value in
MSR_IA32_DS_AREA the host reboots (my idea is he tries to access a
vpmu-struct that isnÂt there in the current vcpu and panics).

Who is trying to write to MSR_IA32_DS_AREA? The guest or dom0? I
thought you said that you want dom0 to do sampling. Or are you trying
to setup DS area from your guest and control it from dom0? I am
somewhat confused.

Can you post hypervisor log? (hard to say how helpful it will be
without seeing your code changes though)

Right after enabling the BTS I get a triple fault.
hvm.c:1357:d2 Triple fault on VCPU0 - invoking HVM shutdown action 1.

That's not host reboot, this is your guest dying.

When I use a modified version of vmx_msr_write_intercept I donât
get any crashes as long as I donât enable BTS and TR in the
GUEST_IA32_DEBUGCTL (BTR works). When I enable the BTS (and TR)
the guest crashes. I suppose he gets killed by the hypervisor for
accessing forbidden memory.

Possibly because DS area point to hypervisor memory.

Having said all this, I am not sure how well BTS works. You did
notice this in the hypervisor log:

(XEN) ** WARNING: Emulation of BTS Feature is switched on **
(XEN) ** Using this processor feature in a virtualized **
(XEN) ** environment is not 100% safe. **
(XEN) ** Setting the DS buffer address with wrong values **
(XEN) ** may lead to hypervisor hangs or crashes. **
(XEN) ** It is NOT recommended for production use! **

Yes, I saw that. It doesnât state that BTS is not working at all, just that it 
not that safe to use.
As I understand it as long as I set the DS buffer address correctly I should
be fine, right?
Right. Except that I am not convinced you did set this buffer
correctly, which is possibly why your hypervisor crashed (I am not
sure I understood under what circumstances though).


Since I donât want to use for production that is fine with me. At least for


The modified version of vmx_msr_write_intercept takes a
vcpu-struct as a parameter and uses this instead of the current vcpu.

Instead of

staticint vmx_msr_write_intercept(unsigned int msr, uint64_t

       struct vcpu *v = current;

I just have

staticint own_vmx_msr_write_intercept(unsigned int msr, uint64_t
msr_content, struct vcpu *v)

I get this vcpu by d->vcpu[0] as I have limited my guest domain to
one vcpu atm.

Of course I also use similarly modified version of the called

IÂm pretty sure that my problem is with a wrong scope/usage of the
vcpus/memory, but I have no idea how to fix this.

I can see a potential problem with the memory allocation (in the
host) into which the cpu in guest-mode is supposed to write.

Or maybe I got the principle of a vcpu/vpmu all wrong.

Since I couldnât find any project that uses the BTS for the guest,
I am wondering if anyone has ever done this and if it is possible at all.

Any input is welcome as I am pretty much stuck atmâ



Virus checked by G Data MailSecurity
Version: AVA 25.435 dated 26.02.2015
Virus news: www.antiviruslab.com

Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.