Hello Liyang, Thanks for the contribution! On Tue, 08/09 14:48, Dou Liyang wrote: > This document describes how to use cpu hotplug in QEMU. > > Change log v1: > From Igor's advice: > 1. Remove any mentioning of apic-id from the document. > 2. Remove the "device_del qom_path" from the CPU hot-unplug. > 3. Fix some comment.
In the future, please put the revision log after the "---" line below your signed-off-by line, so that it doesn't get committed to the git history (this information is not useful there) once get merged. I'm not a native English speaker either, but I'll try to give some suggestions on the documentation below. Please don't mind, and take with a grain of salt. :) > > Signed-off-by: Dou Liyang <douly.f...@cn.fujitsu.com> > --- > docs/cpu-hotplug.txt | 124 > +++++++++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 124 insertions(+) > create mode 100644 docs/cpu-hotplug.txt > > diff --git a/docs/cpu-hotplug.txt b/docs/cpu-hotplug.txt > new file mode 100644 > index 0000000..2c51e27 > --- /dev/null > +++ b/docs/cpu-hotplug.txt > @@ -0,0 +1,124 @@ > +QEMU CPU hotplug > +=================== I think a prettier way to write markdown style docs is to let the underlines be as long as the text, like: QEMU CPU hotplug ================ The same applies to following section titles as well. > + > +This document explains how to use the CPU hotplug feature in QEMU, > +which is present since v2.7. This isn't quite accurate, cpu-add has been available since 1.5 according to http://wiki.qemu.org/Features/CPUHotplug, I guess you are talking about something different. So you probably want to revise the sentence to clarify. > + > +Guest support is required for CPU hotplug to work. > + > +CPU hot-plug > +------------------- > + > +In order to be able to hotplug CPUs, QEMU has to be told what is the s/what is//, or put "is" after "grow". > +maximum amount of CPUs the guest can grow. This is done at startup > +time by means of the -smp command-line option, which has the following > +format: > + > + -smp [cpus=]n[,maxcpus=cpus][,cores=cores][,threads=threads] > + [,sockets=sockets] > + > +Where, > + > + - "cpus" set the number of CPUs to 'n' [default=1] Not very consistent in the syntax used in this list. There is a verb in "cpus" but none in the others. I suggest adding a verb to each item. > + - "maxcpus" maximum number of CPUs, including offline VCPUs for > + hotplug, etc > + - "cores" number of CPU cores on one socket > + - "threads= number of threads on one CPU core > + - "sockets= number of discrete sockets in the system(on sPAPR, s/system(on/system (on)/ > + sockets have no real meaning, it has no real effect on the > + guest.) > + > + > +For example, the following command-line: > + > + qemu [...] -smp 3,maxcpus=10,sockets=2,cores=2,threads=2 > + > +Creates a guest with 3 VCPUs and it support up to 10 VCPUs. The CPU s/support/supports/ > +topology is sockets (2) * cores (2) * threads (2) and can't greater s/greater/be greater/ > +than maxcpus. When the guest is just booted, the guest will see 3 s/just// > +VCPUs. > + > +There are seven VCPUs can be hotplugged by socket/core/thread-ids s/There are seven/Seven more/ > +which should be obtained from the properties advertised by QEMU via s/should be obtained/can be observed/ > +the QMP command query-hotpluggable-cpus or the corresponding HMP > +command "info hotpluggable-cpus". This paragraph seems to be repeating what is following. Maybe be more brief and say "more on this below"? > + > +Query possible available CPU objects > +-------------------------------------- > + > +Before add the VCPUs, we should know the topology properties of the s/add/adding/ > +possible available CPUs objects, so that we can find out the available > +place (socket,core,thread) for a new VCPU. > + > +A monitor commands are used to list the possible CPU objects: s/commands are used/command can be used/ > + > + (qemu) info hotpluggable-cpus > + > +Select the hotpluggable CPUs including "CPUInstance Properties" for > +hotpluging. Such as this: > + > + ... > + type: "qemu64-x86_64-cpu" > + vcpus_count: "1" > + CPUInstance Properties: > + socket-id: "0" > + core-id: "1" > + thread-id: "0" > + ... > + > +Hotplug CPUs > +---------------------- > + > +A monitor commands are used to hotplug CPUs: s/commands are used/command can be used/ > + > + - "device_add": creates a VCPU device and inserts it into the > + specific place as a device > + > +For example, the following commands add a VCPU which id is cpu1 in s/commands add/command adds/ s/which id is/which has id/ > +the last position of the guest's CPU sockets which was discussed > +earlier by using the socket/core/thread-ids: > + > + (qemu) device_add qemu64-x86_64-cpu,id=cpu9,socket-id=2,core-id=0, > + thread-id=1 > + > +Where, > + > + - "qemu64-x86_64-cpu" is the CPU modle. s/modle/model/ > + - "id" is the unique identifier in the device sets. there must have > + it. s/there must have it/It is required./ > + - "socket-id/core-id/thread-id" is represented the designated position > + which are gained form the above possible list of CPUs. s/is represented/represent/ s/are gained/is obtained/ > + > +It's also possible to start a guest with cpu cold-plugged into the > +specific place (socket,core,thread). > + > +In the following command-lines example, a guest which has 3 VCPUs is > +created: > + > + qemu [...] -smp 1,maxcpus=10,sockets=2,cores=2,threads=2 \ > + -device qemu64-x86_64-cpu,id=cpu1,socket-id=1, \ > + core-id=1,thread-id=0 \ > + -device qemu64-x86_64-cpu,id=cpu2,socket-id=1, \ > + core-id=1,thread-id=1 \ > + > +Two VCPUs are cold-plugged by "-device" parameter. They are in the > +same socket and core with different thread-id. After that, the guest s/thread-id/thread-ids/ > +has additional seven VCPUs to be hot-plugged when needed. > + > +CPU hot-unplug > +------------------------ > + > +In order to be able to hot unplug cpu device, QEMU has the way > +to remove cpu device. > + > + Using the ids which were assigned when you hot plugged cpus. Why the indentation for this line? > + > +A monitor commands are used to hot unplug cpus: s/commands are used/command can be used/ > + > + - "device_del": deletes a cpu device > + > +For example, assuming that the cpu device with id "cpu1" exists, > +the following commands tries to remove it. s/commands/command/ > + > + (qemu) device_del cpu1 > + Can you also document what should be done in guest to make this operation successful? Fam
