linux/arch
Heiko Carstens c10fde0d9e [S390] Vertical cpu management.
If vertical cpu polarization is active then the hypervisor will
dispatch certain cpus for a longer time than other cpus for maximum
performance. For example if a guest would have three virtual cpus,
each of them with a share of 33 percent, then in case of vertical
cpu polarization all of the processing time would be combined to a
single cpu which would run all the time, while the other two cpus
would get nearly no cpu time.

There are three different types of vertical cpus: high, medium and
low. Low cpus hardly get any real cpu time, while high cpus get a
full real cpu. Medium cpus get something in between.

In order to switch between the two possible modes (default is
horizontal) a 0 for horizontal polarization or a 1 for vertical
polarization must be written to the dispatching sysfs attribute:

/sys/devices/system/cpu/dispatching

The polarization of each single cpu can be figured out by the
polarization sysfs attribute of each cpu:

/sys/devices/system/cpu/cpuX/polarization

horizontal, vertical:high, vertical:medium, vertical:low or unknown.

When switching polarization the polarization attribute may contain
the value unknown until the configuration change is done and the
kernel has figured out the new polarization of each cpu.

Note that running a system with different types of vertical cpus may
result in significant performance regressions. If possible only one
type of vertical cpus should be used. All other cpus should be
offlined.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
2008-04-17 07:47:01 +02:00
..
alpha alpha: fix ALSA DMA mmap crash 2008-04-02 15:28:19 -07:00
arm [ARM] 4873/1: Fix ITE 8152 interrupt demux 2008-03-28 15:07:46 +00:00
avr32 avr32 mustn't select HAVE_IDE 2008-04-17 01:14:32 +02:00
blackfin [Blackfin] arch: fix atomic and32/xor32 comments and ENDPROC markings 2008-03-03 17:44:14 -07:00
cris cris: correct usage of __user for copy to and from user space in lib/usercopy and uaccess.h 2008-03-04 16:35:16 -08:00
frv FRV: Correctly determine the address of an illegal instruction 2008-04-14 07:55:15 -07:00
h8300 h8300: defconfig update 2008-02-23 17:12:16 -08:00
ia64 acpi: unneccessary to scan the PCI bus already scanned 2008-04-15 19:35:41 -07:00
m32r ide: introduce HAVE_IDE 2008-02-09 10:46:40 +01:00
m68k m68k: update defconfigs for 2.6.25 2008-04-04 14:42:30 -07:00
m68knommu m68k{,nommu}: Wire up new timerfd syscalls 2008-03-04 08:04:11 -08:00
mips Pb1200/DBAu1200: fix bad IDE resource size 2008-04-17 01:14:33 +02:00
mn10300 MN10300: define HZ as a config option 2008-02-23 17:12:13 -08:00
parisc [PARISC] fix signal trampoline cache flushing 2008-04-15 15:51:36 -07:00
powerpc [POWERPC] Fix MPC5200 (not B!) device tree so FEC ethernet works 2008-04-03 22:11:12 +11:00
ppc [PPC] 8xx: swap bug-fix 2008-03-07 16:56:54 -06:00
s390 [S390] Vertical cpu management. 2008-04-17 07:47:01 +02:00
sh sh: export empty_zero_page 2008-04-16 02:21:47 +09:00
sparc [SPARC]: Fix several regset and ptrace bugs. 2008-04-09 19:39:25 -07:00
sparc64 [SPARC]: Fix several regset and ptrace bugs. 2008-04-09 19:39:25 -07:00
um [UML]: uml-net: don't set IFF_ALLMULTI in set_multicast_list 2008-03-26 00:16:29 -07:00
v850 ide: introduce HAVE_IDE 2008-02-09 10:46:40 +01:00
x86 acpi: unneccessary to scan the PCI bus already scanned 2008-04-15 19:35:41 -07:00
xtensa [XTENSA] Allow debugger to modify the WINDOWBASE register. 2008-02-13 17:45:36 -08:00
.gitignore arch: Ignore arch/i386 and arch/x86_64 2008-01-19 21:29:39 -08:00
Kconfig Kprobes: indicate kretprobe support in Kconfig 2008-03-04 16:35:11 -08:00