ia64/linux-2.6.18-xen.hg

view Documentation/cpu-freq/cpufreq-stats.txt @ 897:329ea0ccb344

balloon: try harder to balloon up under memory pressure.

Currently if the balloon driver is unable to increase the guest's
reservation it assumes the failure was due to reaching its full
allocation, gives up on the ballooning operation and records the limit
it reached as the "hard limit". The driver will not try again until
the target is set again (even to the same value).

However it is possible that ballooning has in fact failed due to
memory pressure in the host and therefore it is desirable to keep
attempting to reach the target in case memory becomes available. The
most likely scenario is that some guests are ballooning down while
others are ballooning up and therefore there is temporary memory
pressure while things stabilise. You would not expect a well behaved
toolstack to ask a domain to balloon to more than its allocation nor
would you expect it to deliberately over-commit memory by setting
balloon targets which exceed the total host memory.

This patch drops the concept of a hard limit and causes the balloon
driver to retry increasing the reservation on a timer in the same
manner as when decreasing the reservation.

Also if we partially succeed in increasing the reservation
(i.e. receive less pages than we asked for) then we may as well keep
those pages rather than returning them to Xen.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
author Keir Fraser <keir.fraser@citrix.com>
date Fri Jun 05 14:01:20 2009 +0100 (2009-06-05)
parents 831230e53067
children
line source
2 CPU frequency and voltage scaling statictics in the Linux(TM) kernel
5 L i n u x c p u f r e q - s t a t s d r i v e r
7 - information for users -
10 Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
12 Contents
13 1. Introduction
14 2. Statistics Provided (with example)
15 3. Configuring cpufreq-stats
18 1. Introduction
20 cpufreq-stats is a driver that provices CPU frequency statistics for each CPU.
21 This statistics is provided in /sysfs as a bunch of read_only interfaces. This
22 interface (when configured) will appear in a seperate directory under cpufreq
23 in /sysfs (<sysfs root>/devices/system/cpu/cpuX/cpufreq/stats/) for each CPU.
24 Various statistics will form read_only files under this directory.
26 This driver is designed to be independent of any particular cpufreq_driver
27 that may be running on your CPU. So, it will work with any cpufreq_driver.
30 2. Statistics Provided (with example)
32 cpufreq stats provides following statistics (explained in detail below).
33 - time_in_state
34 - total_trans
35 - trans_table
37 All the statistics will be from the time the stats driver has been inserted
38 to the time when a read of a particular statistic is done. Obviously, stats
39 driver will not have any information about the frequency transitions before
40 the stats driver insertion.
42 --------------------------------------------------------------------------------
43 <mysystem>:/sys/devices/system/cpu/cpu0/cpufreq/stats # ls -l
44 total 0
45 drwxr-xr-x 2 root root 0 May 14 16:06 .
46 drwxr-xr-x 3 root root 0 May 14 15:58 ..
47 -r--r--r-- 1 root root 4096 May 14 16:06 time_in_state
48 -r--r--r-- 1 root root 4096 May 14 16:06 total_trans
49 -r--r--r-- 1 root root 4096 May 14 16:06 trans_table
50 --------------------------------------------------------------------------------
52 - time_in_state
53 This gives the amount of time spent in each of the frequencies supported by
54 this CPU. The cat output will have "<frequency> <time>" pair in each line, which
55 will mean this CPU spent <time> usertime units of time at <frequency>. Output
56 will have one line for each of the supported freuencies. usertime units here
57 is 10mS (similar to other time exported in /proc).
59 --------------------------------------------------------------------------------
60 <mysystem>:/sys/devices/system/cpu/cpu0/cpufreq/stats # cat time_in_state
61 3600000 2089
62 3400000 136
63 3200000 34
64 3000000 67
65 2800000 172488
66 --------------------------------------------------------------------------------
69 - total_trans
70 This gives the total number of frequency transitions on this CPU. The cat
71 output will have a single count which is the total number of frequency
72 transitions.
74 --------------------------------------------------------------------------------
75 <mysystem>:/sys/devices/system/cpu/cpu0/cpufreq/stats # cat total_trans
76 20
77 --------------------------------------------------------------------------------
79 - trans_table
80 This will give a fine grained information about all the CPU frequency
81 transitions. The cat output here is a two dimensional matrix, where an entry
82 <i,j> (row i, column j) represents the count of number of transitions from
83 Freq_i to Freq_j. Freq_i is in descending order with increasing rows and
84 Freq_j is in descending order with increasing columns. The output here also
85 contains the actual freq values for each row and column for better readability.
87 --------------------------------------------------------------------------------
88 <mysystem>:/sys/devices/system/cpu/cpu0/cpufreq/stats # cat trans_table
89 From : To
90 : 3600000 3400000 3200000 3000000 2800000
91 3600000: 0 5 0 0 0
92 3400000: 4 0 2 0 0
93 3200000: 0 1 0 2 0
94 3000000: 0 0 1 0 3
95 2800000: 0 0 0 2 0
96 --------------------------------------------------------------------------------
99 3. Configuring cpufreq-stats
101 To configure cpufreq-stats in your kernel
102 Config Main Menu
103 Power management options (ACPI, APM) --->
104 CPU Frequency scaling --->
105 [*] CPU Frequency scaling
106 <*> CPU frequency translation statistics
107 [*] CPU frequency translation statistics details
110 "CPU Frequency scaling" (CONFIG_CPU_FREQ) should be enabled to configure
111 cpufreq-stats.
113 "CPU frequency translation statistics" (CONFIG_CPU_FREQ_STAT) provides the
114 basic statistics which includes time_in_state and total_trans.
116 "CPU frequency translation statistics details" (CONFIG_CPU_FREQ_STAT_DETAILS)
117 provides fine grained cpufreq stats by trans_table. The reason for having a
118 seperate config option for trans_table is:
119 - trans_table goes against the traditional /sysfs rule of one value per
120 interface. It provides a whole bunch of value in a 2 dimensional matrix
121 form.
123 Once these two options are enabled and your CPU supports cpufrequency, you
124 will be able to see the CPU frequency statistics in /sysfs.