All of lore.kernel.org
 help / color / mirror / Atom feed
* [ help ] vcpu handle MWAIT different
@ 2017-06-09  3:47 Longpeng (Mike)
  2017-06-09  3:52 ` Longpeng (Mike)
  0 siblings, 1 reply; 4+ messages in thread
From: Longpeng (Mike) @ 2017-06-09  3:47 UTC (permalink / raw)
  To: Radim Krčmář, Paolo Bonzini, Wanpeng Li
  Cc: kvm, Michael S. Tsirkin, Gonglei

Hi guys,

I wrote a testcase which repeat execute MWAIT insn in kvm-unit-test and I tested
it on different hardware, but got different results.


1) cpu: E5620 (hardware don't support MONITOR/MWAIT)

This works well because I saw the following info in trace:

       CPU 0/KVM-7301  [015] .... 1534719.099573: kvm_exit: reason
MWAIT_INSTRUCTION rip 0x400387 info 0 0
       CPU 0/KVM-7301  [015] d... 1534719.099573: kvm_entry: vcpu 0
       CPU 0/KVM-7301  [015] .... 1534719.099574: kvm_exit: reason
MWAIT_INSTRUCTION rip 0x400387 info 0 0
       CPU 0/KVM-7301  [015] d... 1534719.099574: kvm_entry: vcpu 0


2) cpu: E5-2620 v3 and E5-2690 v4 (both hardwares support MONITOR/MWAIT)

This...

       CPU 0/KVM-1683  [017] .... 4844125.856164: kvm_exit: reason EXCEPTION_NMI
rip 0x400382 info 0 80000306
       CPU 0/KVM-1683  [017] .... 4844125.856166: kvm_emulate_insn: 0:400382:0f
01 c9 (prot64)
       CPU 0/KVM-1683  [017] .... 4844125.856166: kvm_inj_exception: #UD (0x0)
       CPU 0/KVM-1683  [017] d... 4844125.856167: kvm_entry: vcpu 0

and the kvm-unit-test guest reported:

Unhandled exception 6 #UD at ip 0000000000400387
error_code=0000      rflags=00010002      cs=00000008
rax=0000000000000000 rcx=0000000000000000 rdx=0000000200000000 rbx=0000000000000000
rbp=000000000044db60 rsi=0000000000000001 rdi=0000000000004020
 r8=000000000000000a  r9=00000000000003f8 r10=000000000044d1e2 r11=000000000044d1e1
r12=0000000000000000 r13=0000000000000000 r14=0000000000000000 r15=0000000000000000
cr0=0000000080010011 cr2=0000000000000000 cr3=0000000007fff000 cr4=0000000000000020
cr8=0000000000000000
	STACK: @400387 40028f
0x0000000000400387: mwait at x86/perf_exit.c:17
        	/* "mwait %eax, %ecx;" */
      > 	asm volatile(".byte 0x0f, 0x01, 0xc9;"
        		:: "a" (eax), "c" (ecx));
 (inlined by) main at x86/perf_exit.c:93
        		mb();
      > 		mwait(0, 0);
        		mb();
0x000000000040028e: start64 at x86/cstart64.S:206
        	lea __argv(%rip), %rsi
      > 	call main
        	mov %eax, %edi

In result 2, the trace shows that vcpu is in 64bit protect-mode (the testcase
didn't jump to ring3 manually, so it's in ring0), as intel vol2 said the MWAIT
will #UD in this situation when:

#UD 	If the current privilege level is not 0.
	If CPUID.01H:ECX.MONITOR[bit 3] = 0.

I think CPUID.01H:ECX.MONITOR does not matter, so I'm so curious about the
result 2 .

Do you have any idea ? Thanks.

PS: Wanpeng tested this on his laptop (maybe the cpu is Core-i7) and it worked
well as result 1.

kvm version : 4.4.11
host kernel : 3.10


The testcase is:

#include "libcflat.h"
#include "smp.h"
#include "vm.h"
#include "processor.h"
#include "desc.h"
#include "apic.h"
#include "apic-defs.h"
#include "asm/barrier.h"

[...]

static inline void mwait(unsigned long eax, unsigned long ecx)
{
        /* "mwait %eax, %ecx;" */
        asm volatile(".byte 0x0f, 0x01, 0xc9;"
                :: "a" (eax), "c" (ecx));
}

[...]

int main(int argc, char **argv)
{
        setup_vm();
        smp_init();

	while (test_cnt--) {
                [...]
                mwait(0, 0);
                [...]
        }
	[...]
}


-- 
Regards,
Longpeng(Mike)

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [ help ] vcpu handle MWAIT different
  2017-06-09  3:47 [ help ] vcpu handle MWAIT different Longpeng (Mike)
@ 2017-06-09  3:52 ` Longpeng (Mike)
  2017-06-09  4:51   ` Nadav Amit
  0 siblings, 1 reply; 4+ messages in thread
From: Longpeng (Mike) @ 2017-06-09  3:52 UTC (permalink / raw)
  To: Radim Krčmář, Paolo Bonzini, Wanpeng Li
  Cc: kvm, Michael S. Tsirkin, Gonglei


On 2017/6/9 11:47, Longpeng (Mike) wrote:

> Hi guys,
> 
> I wrote a testcase which repeat execute MWAIT insn in kvm-unit-test and I tested
> it on different hardware, but got different results.
> 
> 
> 1) cpu: E5620 (hardware don't support MONITOR/MWAIT)
> 
> This works well because I saw the following info in trace:
> 
>        CPU 0/KVM-7301  [015] .... 1534719.099573: kvm_exit: reason
> MWAIT_INSTRUCTION rip 0x400387 info 0 0
>        CPU 0/KVM-7301  [015] d... 1534719.099573: kvm_entry: vcpu 0
>        CPU 0/KVM-7301  [015] .... 1534719.099574: kvm_exit: reason
> MWAIT_INSTRUCTION rip 0x400387 info 0 0
>        CPU 0/KVM-7301  [015] d... 1534719.099574: kvm_entry: vcpu 0
> 
> 
> 2) cpu: E5-2620 v3 and E5-2690 v4 (both hardwares support MONITOR/MWAIT)
> 
> This...
> 
>        CPU 0/KVM-1683  [017] .... 4844125.856164: kvm_exit: reason EXCEPTION_NMI
> rip 0x400382 info 0 80000306
>        CPU 0/KVM-1683  [017] .... 4844125.856166: kvm_emulate_insn: 0:400382:0f
> 01 c9 (prot64)
>        CPU 0/KVM-1683  [017] .... 4844125.856166: kvm_inj_exception: #UD (0x0)
>        CPU 0/KVM-1683  [017] d... 4844125.856167: kvm_entry: vcpu 0
> 


Sorry, I made a mistake, the trace of result 2 is:

       CPU 0/KVM-14333 [017] .... 4842658.749717: kvm_exit: reason EXCEPTION_NMI
rip 0x400387 info 0 80000306
       CPU 0/KVM-14333 [017] .... 4842658.749719: kvm_emulate_insn: 0:400387:0f
01 c9 (prot64)
       CPU 0/KVM-14333 [017] .... 4842658.749719: kvm_inj_exception: #UD (0x0)
       CPU 0/KVM-14333 [017] d... 4842658.749720: kvm_entry: vcpu 0


> and the kvm-unit-test guest reported:
> 
> Unhandled exception 6 #UD at ip 0000000000400387
> error_code=0000      rflags=00010002      cs=00000008
> rax=0000000000000000 rcx=0000000000000000 rdx=0000000200000000 rbx=0000000000000000
> rbp=000000000044db60 rsi=0000000000000001 rdi=0000000000004020
>  r8=000000000000000a  r9=00000000000003f8 r10=000000000044d1e2 r11=000000000044d1e1
> r12=0000000000000000 r13=0000000000000000 r14=0000000000000000 r15=0000000000000000
> cr0=0000000080010011 cr2=0000000000000000 cr3=0000000007fff000 cr4=0000000000000020
> cr8=0000000000000000
> 	STACK: @400387 40028f
> 0x0000000000400387: mwait at x86/perf_exit.c:17
>         	/* "mwait %eax, %ecx;" */
>       > 	asm volatile(".byte 0x0f, 0x01, 0xc9;"
>         		:: "a" (eax), "c" (ecx));
>  (inlined by) main at x86/perf_exit.c:93
>         		mb();
>       > 		mwait(0, 0);
>         		mb();
> 0x000000000040028e: start64 at x86/cstart64.S:206
>         	lea __argv(%rip), %rsi
>       > 	call main
>         	mov %eax, %edi
> 
> In result 2, the trace shows that vcpu is in 64bit protect-mode (the testcase
> didn't jump to ring3 manually, so it's in ring0), as intel vol2 said the MWAIT
> will #UD in this situation when:
> 
> #UD 	If the current privilege level is not 0.
> 	If CPUID.01H:ECX.MONITOR[bit 3] = 0.
> 
> I think CPUID.01H:ECX.MONITOR does not matter, so I'm so curious about the
> result 2 .
> 
> Do you have any idea ? Thanks.
> 
> PS: Wanpeng tested this on his laptop (maybe the cpu is Core-i7) and it worked
> well as result 1.
> 
> kvm version : 4.4.11
> host kernel : 3.10
> 
> 
> The testcase is:
> 
> #include "libcflat.h"
> #include "smp.h"
> #include "vm.h"
> #include "processor.h"
> #include "desc.h"
> #include "apic.h"
> #include "apic-defs.h"
> #include "asm/barrier.h"
> 
> [...]
> 
> static inline void mwait(unsigned long eax, unsigned long ecx)
> {
>         /* "mwait %eax, %ecx;" */
>         asm volatile(".byte 0x0f, 0x01, 0xc9;"
>                 :: "a" (eax), "c" (ecx));
> }
> 
> [...]
> 
> int main(int argc, char **argv)
> {
>         setup_vm();
>         smp_init();
> 
> 	while (test_cnt--) {
>                 [...]
>                 mwait(0, 0);
>                 [...]
>         }
> 	[...]
> }
> 
> 


-- 
Regards,
Longpeng(Mike)

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [ help ] vcpu handle MWAIT different
  2017-06-09  3:52 ` Longpeng (Mike)
@ 2017-06-09  4:51   ` Nadav Amit
  2017-06-09  5:57     ` Longpeng (Mike)
  0 siblings, 1 reply; 4+ messages in thread
From: Nadav Amit @ 2017-06-09  4:51 UTC (permalink / raw)
  To: Longpeng (Mike)
  Cc: Radim Krčmář, Paolo Bonzini, Wanpeng Li, kvm,
	Michael S. Tsirkin, Gonglei

Maybe you disabled MWAIT it in the BIOS. Try to read IA32_MISC_ENABLE[18]
(MSR 0x1a0). If it is clear, MWAIT/MONITOR is disabled.




> On Jun 8, 2017, at 8:52 PM, Longpeng (Mike) <longpeng2@huawei.com> wrote:
> 
> 
> On 2017/6/9 11:47, Longpeng (Mike) wrote:
> 
>> Hi guys,
>> 
>> I wrote a testcase which repeat execute MWAIT insn in kvm-unit-test and I tested
>> it on different hardware, but got different results.
>> 
>> 
>> 1) cpu: E5620 (hardware don't support MONITOR/MWAIT)
>> 
>> This works well because I saw the following info in trace:
>> 
>>       CPU 0/KVM-7301  [015] .... 1534719.099573: kvm_exit: reason
>> MWAIT_INSTRUCTION rip 0x400387 info 0 0
>>       CPU 0/KVM-7301  [015] d... 1534719.099573: kvm_entry: vcpu 0
>>       CPU 0/KVM-7301  [015] .... 1534719.099574: kvm_exit: reason
>> MWAIT_INSTRUCTION rip 0x400387 info 0 0
>>       CPU 0/KVM-7301  [015] d... 1534719.099574: kvm_entry: vcpu 0
>> 
>> 
>> 2) cpu: E5-2620 v3 and E5-2690 v4 (both hardwares support MONITOR/MWAIT)
>> 
>> This...
>> 
>>       CPU 0/KVM-1683  [017] .... 4844125.856164: kvm_exit: reason EXCEPTION_NMI
>> rip 0x400382 info 0 80000306
>>       CPU 0/KVM-1683  [017] .... 4844125.856166: kvm_emulate_insn: 0:400382:0f
>> 01 c9 (prot64)
>>       CPU 0/KVM-1683  [017] .... 4844125.856166: kvm_inj_exception: #UD (0x0)
>>       CPU 0/KVM-1683  [017] d... 4844125.856167: kvm_entry: vcpu 0
>> 
> 
> 
> Sorry, I made a mistake, the trace of result 2 is:
> 
>       CPU 0/KVM-14333 [017] .... 4842658.749717: kvm_exit: reason EXCEPTION_NMI
> rip 0x400387 info 0 80000306
>       CPU 0/KVM-14333 [017] .... 4842658.749719: kvm_emulate_insn: 0:400387:0f
> 01 c9 (prot64)
>       CPU 0/KVM-14333 [017] .... 4842658.749719: kvm_inj_exception: #UD (0x0)
>       CPU 0/KVM-14333 [017] d... 4842658.749720: kvm_entry: vcpu 0
> 
> 
>> and the kvm-unit-test guest reported:
>> 
>> Unhandled exception 6 #UD at ip 0000000000400387
>> error_code=0000      rflags=00010002      cs=00000008
>> rax=0000000000000000 rcx=0000000000000000 rdx=0000000200000000 rbx=0000000000000000
>> rbp=000000000044db60 rsi=0000000000000001 rdi=0000000000004020
>> r8=000000000000000a  r9=00000000000003f8 r10=000000000044d1e2 r11=000000000044d1e1
>> r12=0000000000000000 r13=0000000000000000 r14=0000000000000000 r15=0000000000000000
>> cr0=0000000080010011 cr2=0000000000000000 cr3=0000000007fff000 cr4=0000000000000020
>> cr8=0000000000000000
>> 	STACK: @400387 40028f
>> 0x0000000000400387: mwait at x86/perf_exit.c:17
>>        	/* "mwait %eax, %ecx;" */
>>> 	asm volatile(".byte 0x0f, 0x01, 0xc9;"
>>        		:: "a" (eax), "c" (ecx));
>> (inlined by) main at x86/perf_exit.c:93
>>        		mb();
>>> 		mwait(0, 0);
>>        		mb();
>> 0x000000000040028e: start64 at x86/cstart64.S:206
>>        	lea __argv(%rip), %rsi
>>> 	call main
>>        	mov %eax, %edi
>> 
>> In result 2, the trace shows that vcpu is in 64bit protect-mode (the testcase
>> didn't jump to ring3 manually, so it's in ring0), as intel vol2 said the MWAIT
>> will #UD in this situation when:
>> 
>> #UD 	If the current privilege level is not 0.
>> 	If CPUID.01H:ECX.MONITOR[bit 3] = 0.
>> 
>> I think CPUID.01H:ECX.MONITOR does not matter, so I'm so curious about the
>> result 2 .
>> 
>> Do you have any idea ? Thanks.
>> 
>> PS: Wanpeng tested this on his laptop (maybe the cpu is Core-i7) and it worked
>> well as result 1.
>> 
>> kvm version : 4.4.11
>> host kernel : 3.10
>> 
>> 
>> The testcase is:
>> 
>> #include "libcflat.h"
>> #include "smp.h"
>> #include "vm.h"
>> #include "processor.h"
>> #include "desc.h"
>> #include "apic.h"
>> #include "apic-defs.h"
>> #include "asm/barrier.h"
>> 
>> [...]
>> 
>> static inline void mwait(unsigned long eax, unsigned long ecx)
>> {
>>        /* "mwait %eax, %ecx;" */
>>        asm volatile(".byte 0x0f, 0x01, 0xc9;"
>>                :: "a" (eax), "c" (ecx));
>> }
>> 
>> [...]
>> 
>> int main(int argc, char **argv)
>> {
>>        setup_vm();
>>        smp_init();
>> 
>> 	while (test_cnt--) {
>>                [...]
>>                mwait(0, 0);
>>                [...]
>>        }
>> 	[...]
>> }
>> 
>> 
> 
> 
> -- 
> Regards,
> Longpeng(Mike)

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [ help ] vcpu handle MWAIT different
  2017-06-09  4:51   ` Nadav Amit
@ 2017-06-09  5:57     ` Longpeng (Mike)
  0 siblings, 0 replies; 4+ messages in thread
From: Longpeng (Mike) @ 2017-06-09  5:57 UTC (permalink / raw)
  To: Nadav Amit
  Cc: Radim Krčmář, Paolo Bonzini, Wanpeng Li, kvm,
	Michael S. Tsirkin, Gonglei

Hi Nadav,

Thanks for your suggestion, you're right.

I checked my hardwares, the E5620/E5-2620 v3/E5-2690 v4 all support
MONITOR/MWAIT, but my E5-2620 server and E5-2690 server disable MWAIT in the
BIOS, and E5620 server enable. So I made a mistake, sorry guys.

-- 
Regards,
Longpeng(Mike)

On 2017/6/9 12:51, Nadav Amit wrote:

> Maybe you disabled MWAIT it in the BIOS. Try to read IA32_MISC_ENABLE[18]
> (MSR 0x1a0). If it is clear, MWAIT/MONITOR is disabled.
> 
> 
> 
> 
>> On Jun 8, 2017, at 8:52 PM, Longpeng (Mike) <longpeng2@huawei.com> wrote:
>>
>>
>> On 2017/6/9 11:47, Longpeng (Mike) wrote:
>>
>>> Hi guys,
>>>
>>> I wrote a testcase which repeat execute MWAIT insn in kvm-unit-test and I tested
>>> it on different hardware, but got different results.
>>>
>>>
>>> 1) cpu: E5620 (hardware don't support MONITOR/MWAIT)
>>>
>>> This works well because I saw the following info in trace:
>>>
>>>       CPU 0/KVM-7301  [015] .... 1534719.099573: kvm_exit: reason
>>> MWAIT_INSTRUCTION rip 0x400387 info 0 0
>>>       CPU 0/KVM-7301  [015] d... 1534719.099573: kvm_entry: vcpu 0
>>>       CPU 0/KVM-7301  [015] .... 1534719.099574: kvm_exit: reason
>>> MWAIT_INSTRUCTION rip 0x400387 info 0 0
>>>       CPU 0/KVM-7301  [015] d... 1534719.099574: kvm_entry: vcpu 0
>>>
>>>
>>> 2) cpu: E5-2620 v3 and E5-2690 v4 (both hardwares support MONITOR/MWAIT)
>>>
>>> This...
>>>
>>>       CPU 0/KVM-1683  [017] .... 4844125.856164: kvm_exit: reason EXCEPTION_NMI
>>> rip 0x400382 info 0 80000306
>>>       CPU 0/KVM-1683  [017] .... 4844125.856166: kvm_emulate_insn: 0:400382:0f
>>> 01 c9 (prot64)
>>>       CPU 0/KVM-1683  [017] .... 4844125.856166: kvm_inj_exception: #UD (0x0)
>>>       CPU 0/KVM-1683  [017] d... 4844125.856167: kvm_entry: vcpu 0
>>>
>>
>>
>> Sorry, I made a mistake, the trace of result 2 is:
>>
>>       CPU 0/KVM-14333 [017] .... 4842658.749717: kvm_exit: reason EXCEPTION_NMI
>> rip 0x400387 info 0 80000306
>>       CPU 0/KVM-14333 [017] .... 4842658.749719: kvm_emulate_insn: 0:400387:0f
>> 01 c9 (prot64)
>>       CPU 0/KVM-14333 [017] .... 4842658.749719: kvm_inj_exception: #UD (0x0)
>>       CPU 0/KVM-14333 [017] d... 4842658.749720: kvm_entry: vcpu 0
>>
>>
>>> and the kvm-unit-test guest reported:
>>>
>>> Unhandled exception 6 #UD at ip 0000000000400387
>>> error_code=0000      rflags=00010002      cs=00000008
>>> rax=0000000000000000 rcx=0000000000000000 rdx=0000000200000000 rbx=0000000000000000
>>> rbp=000000000044db60 rsi=0000000000000001 rdi=0000000000004020
>>> r8=000000000000000a  r9=00000000000003f8 r10=000000000044d1e2 r11=000000000044d1e1
>>> r12=0000000000000000 r13=0000000000000000 r14=0000000000000000 r15=0000000000000000
>>> cr0=0000000080010011 cr2=0000000000000000 cr3=0000000007fff000 cr4=0000000000000020
>>> cr8=0000000000000000
>>> 	STACK: @400387 40028f
>>> 0x0000000000400387: mwait at x86/perf_exit.c:17
>>>        	/* "mwait %eax, %ecx;" */
>>>> 	asm volatile(".byte 0x0f, 0x01, 0xc9;"
>>>        		:: "a" (eax), "c" (ecx));
>>> (inlined by) main at x86/perf_exit.c:93
>>>        		mb();
>>>> 		mwait(0, 0);
>>>        		mb();
>>> 0x000000000040028e: start64 at x86/cstart64.S:206
>>>        	lea __argv(%rip), %rsi
>>>> 	call main
>>>        	mov %eax, %edi
>>>
>>> In result 2, the trace shows that vcpu is in 64bit protect-mode (the testcase
>>> didn't jump to ring3 manually, so it's in ring0), as intel vol2 said the MWAIT
>>> will #UD in this situation when:
>>>
>>> #UD 	If the current privilege level is not 0.
>>> 	If CPUID.01H:ECX.MONITOR[bit 3] = 0.
>>>
>>> I think CPUID.01H:ECX.MONITOR does not matter, so I'm so curious about the
>>> result 2 .
>>>
>>> Do you have any idea ? Thanks.
>>>
>>> PS: Wanpeng tested this on his laptop (maybe the cpu is Core-i7) and it worked
>>> well as result 1.
>>>
>>> kvm version : 4.4.11
>>> host kernel : 3.10
>>>
>>>
>>> The testcase is:
>>>
>>> #include "libcflat.h"
>>> #include "smp.h"
>>> #include "vm.h"
>>> #include "processor.h"
>>> #include "desc.h"
>>> #include "apic.h"
>>> #include "apic-defs.h"
>>> #include "asm/barrier.h"
>>>
>>> [...]
>>>
>>> static inline void mwait(unsigned long eax, unsigned long ecx)
>>> {
>>>        /* "mwait %eax, %ecx;" */
>>>        asm volatile(".byte 0x0f, 0x01, 0xc9;"
>>>                :: "a" (eax), "c" (ecx));
>>> }
>>>
>>> [...]
>>>
>>> int main(int argc, char **argv)
>>> {
>>>        setup_vm();
>>>        smp_init();
>>>
>>> 	while (test_cnt--) {
>>>                [...]
>>>                mwait(0, 0);
>>>                [...]
>>>        }
>>> 	[...]
>>> }
>>>
>>>
>>
>>
>> -- 
>> Regards,
>> Longpeng(Mike)
> 
> 
> .
> 


-- 
Regards,
Longpeng(Mike)

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2017-06-09  6:00 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-06-09  3:47 [ help ] vcpu handle MWAIT different Longpeng (Mike)
2017-06-09  3:52 ` Longpeng (Mike)
2017-06-09  4:51   ` Nadav Amit
2017-06-09  5:57     ` Longpeng (Mike)

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.