why only domain0 using timer interrupt

why only domain0 using timer interrupt

[zyy9812]

[-- Attachment #1.1: Type: text/plain, Size: 1757 bytes --]

Hi,
  When I modprobe oprofile in domain0, the dmesg show below:
    [root@sd02138 ~]# dmesg | tail -n 3 
   sha256_ssse3: Using SSSE3 optimized SHA-256 implementation 
   SELinux: initialized (dev cifs, type cifs), uses genfs_contexts 
   oprofile: using timer interrupt.
  But in domainU it the oprofile using the NMI interrup rightly.
 [root@Domain1 ~]# modprobe oprofile 
 [root@Domain1 ~]# dmesg | tail -n 3 
 CIFS VFS: default security mechanism requested. The default security mechanism will be upgraded from ntlm to ntlmv2 in kernel release 3.3 
 SELinux: initialized (dev cifs, type cifs), uses genfs_contexts 
 oprofile: using NMI interrupt.


 I have patch the oprofile correctly,the opcontrol show:
[root@sd02138 ~]# opcontrol
……
--xen=file Xen image (for Xen only) 
--active-domains=id[,ids] list of domains in multiple domain profiling session (Xen) 
(detailed profiling of user level and kernel modules code) 
(requires running oprofile on these domains) 
--passive-domains=id[,ids] list of domains to be profiled (Xen). 
or --domains=id[,ids] (coarse profiling of user level and kernel modules code) 
(no need to run oprofile on these domains) 
--passive-images=file[,files] list of kernel images associated with each passive domain 
or 
--domain-images=file[,files]ll patch the oprofile correctly,the opcontrol show:

cpu flags showed the apic flag
[root@sd02138 ~]# cat /proc/cpuinfo | grep flags 
flags : fpu de tsc msr pae mce cx8 apic sep mca cmov pat clflush acpi mmx fxsr sse sse2 ss ht syscall nx lm constant_tsc rep_good nopl nonstop_tsc pni pclmulqdq monitor est ssse3 cx16 sse4_1 sse4_2 popcnt aes hypervisor lahf_lm arat epb dtherm

Does xenoprof does not supprot X86_64? 


zyy9812@126.com

[-- Attachment #1.2: Type: text/html, Size: 3699 bytes --]

[-- Attachment #2: Type: text/plain, Size: 126 bytes --]

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

Is: oprofile in Linux with Xen. Was:Re: why only domain0 using timer interrupt

[Konrad Rzeszutek Wilk]

On Thu, Jan 22, 2015 at 08:06:50PM +0800, zyy9812@126.com wrote:
> Hi,
>   When I modprobe oprofile in domain0, the dmesg show below:
>     [root@sd02138 ~]# dmesg | tail -n 3 
>    sha256_ssse3: Using SSSE3 optimized SHA-256 implementation 
>    SELinux: initialized (dev cifs, type cifs), uses genfs_contexts 
>    oprofile: using timer interrupt.
>   But in domainU it the oprofile using the NMI interrup rightly.
>  [root@Domain1 ~]# modprobe oprofile 
>  [root@Domain1 ~]# dmesg | tail -n 3 
>  CIFS VFS: default security mechanism requested. The default security mechanism will be upgraded from ntlm to ntlmv2 in kernel release 3.3 
>  SELinux: initialized (dev cifs, type cifs), uses genfs_contexts 
>  oprofile: using NMI interrupt.
> 
> 
>  I have patch the oprofile correctly,the opcontrol show:
> [root@sd02138 ~]# opcontrol
> ……
> --xen=file Xen image (for Xen only) 
> --active-domains=id[,ids] list of domains in multiple domain profiling session (Xen) 
> (detailed profiling of user level and kernel modules code) 
> (requires running oprofile on these domains) 
> --passive-domains=id[,ids] list of domains to be profiled (Xen). 
> or --domains=id[,ids] (coarse profiling of user level and kernel modules code) 
> (no need to run oprofile on these domains) 
> --passive-images=file[,files] list of kernel images associated with each passive domain 
> or 
> --domain-images=file[,files]ll patch the oprofile correctly,the opcontrol show:
> 
> cpu flags showed the apic flag
> [root@sd02138 ~]# cat /proc/cpuinfo | grep flags 
> flags : fpu de tsc msr pae mce cx8 apic sep mca cmov pat clflush acpi mmx fxsr sse sse2 ss ht syscall nx lm constant_tsc rep_good nopl nonstop_tsc pni pclmulqdq monitor est ssse3 cx16 sse4_1 sse4_2 popcnt aes hypervisor lahf_lm arat epb dtherm
> 
> Does xenoprof does not supprot X86_64? 

It does, but to to use oprofile in Linux you will need patches to enable it.
You might be better of just using 'perf'.


> 
> 
> zyy9812@126.com

> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel


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