From: Juergen Gross <jgross@suse.com>
To: Andrew Cooper <andrew.cooper3@citrix.com>,
xen-devel@lists.xenproject.org
Cc: "Stefano Stabellini" <sstabellini@kernel.org>,
"Wei Liu" <wei.liu2@citrix.com>,
"Konrad Rzeszutek Wilk" <konrad.wilk@oracle.com>,
"George Dunlap" <George.Dunlap@eu.citrix.com>,
"Ian Jackson" <ian.jackson@eu.citrix.com>,
"Tim Deegan" <tim@xen.org>, "Julien Grall" <julien.grall@arm.com>,
"Jan Beulich" <jbeulich@suse.com>,
"Roger Pau Monné" <roger.pau@citrix.com>
Subject: Re: [PATCH RFC 44/49] xen: round up max vcpus to scheduling granularity
Date: Tue, 2 Apr 2019 09:49:14 +0200 [thread overview]
Message-ID: <fa9da763-c3db-c298-d1e8-6220fe15e5e0@suse.com> (raw)
In-Reply-To: <6c259488-c0ab-ec96-ee09-b76a9f402901@suse.com>
On 01/04/2019 11:47, Juergen Gross wrote:
> On 01/04/2019 10:50, Andrew Cooper wrote:
>> On 29/03/2019 15:09, Juergen Gross wrote:
>>> Make sure the number of vcpus is always a multiple of the scheduling
>>> granularity. Note that we don't support a scheduling granularity above
>>> one on ARM.
>>
>> I'm afraid that I don't think this is a clever move. In turn, this
>> brings into question the approach to idle handling.
>>
>> Firstly, with a proposed socket granularity, this would be 128 on some
>> systems which exist today. Furthermore, consider the case where
>> cpupool0 has a granularity of 1, and a second pool has a granularity of
>> 2. A domain can be created with an odd number of vcpus and operate in
>> pool0 fine, but can't now be moved to pool1.
>
> For now granularity is the same for all pools, but I plan to enhance
> that in future.
>
> The answer to that problem might be either to allow for later addition
> of dummy vcpus (e.g. by sizing only the vcpu pointer array to the needed
> number), or to really disallow moving such a domain between pools.
>
>> If at all possible, I think it would be better to try and reuse the idle
>> cpus for holes like this. Seeing as you've been playing with this code
>> a lot, what is your assessment?
>
> This would be rather complicated. I'd either need to switch vcpus
> dynamically in schedule items, or I'd need to special case the idle
> vcpus in _lots_ of places.
I have thought more about this and maybe I have found a way to make that
less intrusive as I thought in the beginning.
I'll give it a try...
Juergen
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
next prev parent reply other threads:[~2019-04-02 7:49 UTC|newest]
Thread overview: 111+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-29 15:08 [PATCH RFC 00/49] xen: add core scheduling support Juergen Gross
2019-03-29 15:08 ` [PATCH RFC 01/49] xen/sched: call cpu_disable_scheduler() via cpu notifier Juergen Gross
2019-04-01 9:21 ` Julien Grall
2019-04-01 9:40 ` Juergen Gross
2019-04-01 10:29 ` Julien Grall
2019-04-01 10:37 ` Juergen Gross
2019-04-01 13:21 ` Julien Grall
2019-04-01 13:33 ` Juergen Gross
2019-04-01 14:01 ` Julien Grall
2019-04-01 14:23 ` Juergen Gross
2019-04-01 15:15 ` Julien Grall
2019-04-01 16:00 ` Juergen Gross
2019-04-01 17:17 ` Julien Grall
2019-04-16 19:34 ` Stefano Stabellini
2019-04-16 19:34 ` [Xen-devel] " Stefano Stabellini
2019-03-29 15:08 ` [PATCH RFC 02/49] xen: add helper for calling notifier_call_chain() to common/cpu.c Juergen Gross
2019-03-29 15:08 ` [PATCH RFC 03/49] xen: add new cpu notifier action CPU_RESUME_FAILED Juergen Gross
2019-03-29 15:08 ` [PATCH RFC 04/49] xen: don't free percpu areas during suspend Juergen Gross
2019-03-29 15:08 ` [PATCH RFC 05/49] xen/cpupool: simplify suspend/resume handling Juergen Gross
2019-03-29 15:08 ` [PATCH RFC 06/49] xen/sched: don't disable scheduler on cpus during suspend Juergen Gross
2019-03-29 15:08 ` [PATCH RFC 07/49] xen/sched: fix credit2 smt idle handling Juergen Gross
2019-03-29 18:22 ` Dario Faggioli
2019-03-29 15:08 ` [PATCH RFC 08/49] xen/sched: use new sched_item instead of vcpu in scheduler interfaces Juergen Gross
2019-03-29 18:42 ` Andrew Cooper
2019-03-30 10:24 ` Juergen Gross
2019-04-01 6:06 ` Juergen Gross
2019-04-01 7:05 ` Dario Faggioli
2019-04-01 8:19 ` Andrew Cooper
2019-04-01 8:49 ` Juergen Gross
2019-04-01 15:15 ` Dario Faggioli
2019-03-29 15:08 ` [PATCH RFC 09/49] xen/sched: alloc struct sched_item for each vcpu Juergen Gross
2019-03-29 15:08 ` [PATCH RFC 10/49] xen/sched: move per-vcpu scheduler private data pointer to sched_item Juergen Gross
2019-03-29 15:08 ` [PATCH RFC 11/49] xen/sched: build a linked list of struct sched_item Juergen Gross
2019-03-29 15:08 ` [PATCH RFC 12/49] xen/sched: introduce struct sched_resource Juergen Gross
2019-03-29 15:08 ` [PATCH RFC 13/49] xen/sched: let pick_cpu return a scheduler resource Juergen Gross
2019-03-29 15:08 ` [PATCH RFC 14/49] xen/sched: switch schedule_data.curr to point at sched_item Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 15/49] xen/sched: move per cpu scheduler private data into struct sched_resource Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 16/49] xen/sched: switch vcpu_schedule_lock to item_schedule_lock Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 17/49] xen/sched: move some per-vcpu items to struct sched_item Juergen Gross
2019-03-29 21:33 ` Andrew Cooper
2019-03-30 9:59 ` Juergen Gross
2019-04-01 5:59 ` Juergen Gross
2019-04-01 8:05 ` Jan Beulich
2019-04-01 8:26 ` Andrew Cooper
2019-04-01 8:41 ` Jan Beulich
2019-04-01 8:45 ` Juergen Gross
2019-04-01 8:01 ` Jan Beulich
2019-04-01 8:33 ` Andrew Cooper
2019-04-01 8:44 ` Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 18/49] xen/sched: add scheduler helpers hiding vcpu Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 19/49] xen/sched: add domain pointer to struct sched_item Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 20/49] xen/sched: add id " Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 21/49] xen/sched: rename scheduler related perf counters Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 22/49] xen/sched: switch struct task_slice from vcpu to sched_item Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 23/49] xen/sched: move is_running indicator to struct sched_item Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 24/49] xen/sched: make null scheduler vcpu agnostic Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 25/49] xen/sched: make rt " Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 26/49] xen/sched: make credit " Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 27/49] xen/sched: make credit2 " Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 28/49] xen/sched: make arinc653 " Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 29/49] xen: add sched_item_pause_nosync() and sched_item_unpause() Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 30/49] xen: let vcpu_create() select processor Juergen Gross
2019-03-29 19:17 ` Andrew Cooper
2019-03-30 10:23 ` Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 31/49] xen/sched: use sched_resource cpu instead smp_processor_id in schedulers Juergen Gross
2019-03-29 19:36 ` Andrew Cooper
2019-03-30 10:22 ` Juergen Gross
2019-04-01 8:10 ` Jan Beulich
2019-03-29 15:09 ` [PATCH RFC 32/49] xen/sched: switch schedule() from vcpus to sched_items Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 33/49] xen/sched: switch sched_move_irqs() to take sched_item as parameter Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 34/49] xen: switch from for_each_vcpu() to for_each_sched_item() Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 35/49] xen/sched: add runstate counters to struct sched_item Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 36/49] xen/sched: rework and rename vcpu_force_reschedule() Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 37/49] xen/sched: Change vcpu_migrate_*() to operate on schedule item Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 38/49] xen/sched: move struct task_slice into struct sched_item Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 39/49] xen/sched: add code to sync scheduling of all vcpus of a sched item Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 40/49] xen/sched: add support for multiple vcpus per sched item where missing Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 41/49] x86: make loading of GDT at context switch more modular Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 42/49] xen/sched: add support for guest vcpu idle Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 43/49] xen/sched: modify cpupool_domain_cpumask() to be an item mask Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 44/49] xen: round up max vcpus to scheduling granularity Juergen Gross
2019-04-01 8:50 ` Andrew Cooper
2019-04-01 9:47 ` Juergen Gross
2019-04-02 7:49 ` Juergen Gross [this message]
2019-03-29 15:09 ` [PATCH RFC 45/49] xen/sched: support allocating multiple vcpus into one sched item Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 46/49] xen/sched: add a scheduler_percpu_init() function Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 47/49] xen/sched: support core scheduling in continue_running() Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 48/49] xen/sched: make vcpu_wake() core scheduling aware Juergen Gross
2019-03-29 15:09 ` [PATCH RFC 49/49] xen/sched: add scheduling granularity enum Juergen Gross
2019-03-29 15:37 ` [PATCH RFC 00/49] xen: add core scheduling support Juergen Gross
2019-03-29 15:39 ` Jan Beulich
[not found] ` <5C9E3C3D0200007800222FB0@suse.com>
2019-03-29 15:46 ` Juergen Gross
2019-03-29 16:56 ` Dario Faggioli
2019-03-29 17:00 ` Juergen Gross
2019-03-29 17:29 ` Dario Faggioli
2019-03-29 17:39 ` Rian Quinn
2019-03-29 17:48 ` Andrew Cooper
2019-03-29 18:35 ` Rian Quinn
2019-03-29 18:16 ` Dario Faggioli
2019-03-30 9:55 ` Juergen Gross
2019-04-11 0:34 ` Dario Faggioli
2019-04-11 0:34 ` [Xen-devel] " Dario Faggioli
2019-04-11 7:16 ` Juergen Gross
2019-04-11 7:16 ` [Xen-devel] " Juergen Gross
2019-04-11 13:28 ` Dario Faggioli
2019-04-11 13:28 ` [Xen-devel] " Dario Faggioli
2019-04-01 6:41 ` Jan Beulich
[not found] ` <5CA1B285020000780022361D@suse.com>
2019-04-01 6:49 ` Juergen Gross
2019-04-01 7:10 ` Dario Faggioli
2019-04-01 7:15 ` Juergen Gross
2019-04-01 7:13 ` Jan Beulich
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=fa9da763-c3db-c298-d1e8-6220fe15e5e0@suse.com \
--to=jgross@suse.com \
--cc=George.Dunlap@eu.citrix.com \
--cc=andrew.cooper3@citrix.com \
--cc=ian.jackson@eu.citrix.com \
--cc=jbeulich@suse.com \
--cc=julien.grall@arm.com \
--cc=konrad.wilk@oracle.com \
--cc=roger.pau@citrix.com \
--cc=sstabellini@kernel.org \
--cc=tim@xen.org \
--cc=wei.liu2@citrix.com \
--cc=xen-devel@lists.xenproject.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).