|
|
|
|
|
|
|
|
|
|
xen-devel
RE: [Xen-devel] Question about machine time delay in guest domains(Full
Hi Christi,
I can reproduce the time delay by just one x86_64 rhel5u1
guest under specjbb2005, although the delay is not so great like your data. I
found it would be slow 100 secs, after running specjbb2005 2000 secs. I also
tried timer_mode=1 or 2. They all met time slow issue.
BTW, I only catch this issue on x86_64 RHEL5u1 guest. With more
guests running, the guest timer would be slower. But x86_32 RHEL5u1
guest doesn't meet this issue. And without workload, guest timer is
acceptable.
Best Regards,
Yongkang You
Hi, Yongkang,
Thanks so much!!
> Did you try if only
creating 1 domains with 2 cpus?
I did the following 2 patterns.
1) 2 domains with 2 vcpus (each domain shares 2 physical cpus) 2) 4
domains with 2 vcpus (each domain shares 2 physical cpus)
>What's
your platform and guest OS version?
Could you see the following
info?
platform: x86_64 OS: RHEL5u1 2.6.18 (details are
below)
-------------------- #uname -a Linux DL360-dom0-RHEL5u1
2.6.18-53.el5xen #1 SMP Wed Oct 10 16:48:44 EDT 2007 x86_64 x86_64 x86_64
GNU/Linux
# xm info host : DL360-dom0-RHEL5u1 release :
2.6.18-53.el5xen version : #1 SMP Wed Oct 10 16:48:44 EDT 2007 machine :
x86_64 nr_cpus : 2 nr_nodes : 1 sockets_per_node :
1 cores_per_socket : 2 threads_per_core : 1 cpu_mhz : 1866 hw_caps :
bfebfbff:20000800:00000000:00000140:0004e3bd:00000000:00000001 total_memory :
16381 free_memory : 1 xen_major : 3 xen_minor : 1 xen_extra :
.0-53.el5 xen_caps : xen-3.0-x86_64 xen-3.0-x86_32p hvm-3.0-x86_32
hvm-3.0-x86_32p hvm-3.0-x86_64 xen_pagesize : 4096 platform_params :
virt_start=0xffff800000000000 xen_changeset : unavailable cc_compiler :
gcc version 4.1.2 20070626 (Red Hat 4.1.2-14) cc_compile_by :
brewbuilder cc_compile_domain : build.redhat.com cc_compile_date : Wed Oct
10 16:30:57 EDT 2007 xend_config_format : 2
# cat
/proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family :
6 model : 15 model name : Intel(R) Xeon(R) CPU E5320 @ 1.86GHz stepping
: 7 cpu MHz : 1866.730 cache size : 4096 KB physical id :
0 siblings : 1 core id : 0 cpu cores : 1 fpu : yes fpu_exception
: yes cpuid level : 10 wp : yes flags : fpu tsc msr pae mce cx8 apic
mtrr mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall
lm constant_tsc pni monitor ds_cpl vmx est tm2 cx16 xtpr lahf_lm bogomips
: 4669.03 clflush size : 64 cache_alignment : 64 address sizes : 36
bits physical, 48 bits virtual power management:
processor :
1 vendor_id : GenuineIntel cpu family : 6 model : 15 model name :
Intel(R) Xeon(R) CPU E5320 @ 1.86GHz stepping : 7 cpu MHz :
1866.730 cache size : 4096 KB physical id : 1 siblings : 1 core id :
0 cpu cores : 1 fpu : yes fpu_exception : yes cpuid level : 10 wp
: yes flags : fpu tsc msr pae mce cx8 apic mtrr mca cmov pat pse36 clflush
dts acpi mmx fxsr sse sse2 ss ht tm syscall lm constant_tsc pni monitor
ds_cpl vmx est tm2 cx16 xtpr lahf_lm bogomips : 4669.03 clflush size :
64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits
virtual power management:
-------------------------
Thanks
again, I really appreciate your kind help! ;)
-Christi
2008/1/9, You, Yongkang <yongkang.you@xxxxxxxxx>:
On
Tuesday, January 08, 2008 10:03 PM christie kumiko goto wrote:
>
------------------------------------------------------ >
<<Environment>> > -Host machine has 2 cores (core number
0-1) > -Guest OSs: Full virtualization > > *test 1 > 4
domains share 2 core (each domain has 2 vcpu) > ->result: actual test
time : test time measured in guest domains = 4:1 > > *test
2 > 2 domains share 1 core and the other 2 domains share the other 1
core. > ->result: actual test time : test time measured in guest
domains = 2:1 > ------------------------------------------------------
Hi Christi,
Did you try if only creating 1 domains with 2
cpus? What's your platform and guest OS version? There are some new
configurations for guest timer, such as timer_mode=1/2/3 (default timer mode
is 0). You can set it in the guest config file. For some new OS, we can use
timer_mode=1/2 to make the timer more acccurate. But I am not sure if they can
work in your testing.
Best Regards, Yongkang
You
_______________________________________________ Xen-devel
mailing list Xen-devel@xxxxxxxxxxxxxxxxxxx http://lists.xensource.com/xen-devel
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
|
|
|
|