android_kernel_xiaomi_sm8350/arch/x86_64
Ashok Raj 884d9e40b4 [PATCH] x86_64: Dont use broadcast shortcut to make it cpu hotplug safe.
Broadcast IPI's provide un-expected behaviour for cpu hotplug.  CPU's in
offline state also end up receiving the IPI.  Once the cpus become online they
receive these stale IPI's which are bad and introduce unexpected behaviour.

This is easily avoided by not sending a broadcast and addressing just the
CPU's in online map.  Doing prelim cycle counts it appears there is no big
overhead and numbers seem around 0x3000-0x3900 on an average on x86 and x86_64
systems with CPUS running 3G, both for broadcast and mask version of the
API's.

The shortcuts are useful only for flat mode (where the perf shows no
degradation), and in cluster mode, its unicast anyway.  Its simpler to just
not use broadcast anymore.

Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Acked-by: Andi Kleen <ak@muc.de>
Acked-by: Zwane Mwaikambo <zwane@arm.linux.org.uk>
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-06-25 16:24:31 -07:00
..
boot [PATCH] use ${CROSS_COMPILE}installkernel in arch/*/boot/install.sh 2005-06-23 09:45:07 -07:00
ia32 [PATCH] x86_64: never block forced SIGSEGV 2005-06-23 09:45:13 -07:00
kernel [PATCH] x86_64: Dont use broadcast shortcut to make it cpu hotplug safe. 2005-06-25 16:24:31 -07:00
lib [PATCH] Platform SMIs and their interferance with tsc based delay calibration 2005-06-23 09:45:08 -07:00
mm [PATCH] x86_64: Change init sections for CPU hotplug support 2005-06-25 16:24:30 -07:00
oprofile Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
pci [PATCH] x86/x86_64: pcibus_to_node 2005-06-23 09:45:08 -07:00
defconfig [PATCH] x86_64: Update defconfig 2005-05-17 07:59:17 -07:00
Kconfig [PATCH] x86_64: CPU hotplug support 2005-06-25 16:24:30 -07:00
Kconfig.debug Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00
Makefile Linux-2.6.12-rc2 2005-04-16 15:20:36 -07:00