From: Greg Kurz <groug@kaod.org>
To: David Gibson <david@gibson.dropbear.id.au>
Cc: qemu-ppc@nongnu.org, "Cédric Le Goater" <clg@kaod.org>,
qemu-devel@nongnu.org
Subject: Re: [PATCH 7/7] spapr: Set VSMT to smp_threads by default
Date: Mon, 14 Oct 2019 13:31:49 +0200 [thread overview]
Message-ID: <20191014133149.686b681e@bahia.lan> (raw)
In-Reply-To: <20191014061247.GZ4080@umbus.fritz.box>
[-- Attachment #1: Type: text/plain, Size: 3798 bytes --]
On Mon, 14 Oct 2019 17:12:47 +1100
David Gibson <david@gibson.dropbear.id.au> wrote:
> On Thu, Oct 03, 2019 at 02:02:00PM +0200, Greg Kurz wrote:
> > Support for setting VSMT is available in KVM since linux-4.13. Most distros
> > that support KVM on POWER already have it. It thus seem reasonable enough
> > to have the default machine to set VSMT to smp_threads.
> >
> > This brings contiguous VCPU ids and thus brings their upper bound down to
> > the machine's max_cpus. This is especially useful for XIVE KVM devices,
> > which may thus allocate only one VP descriptor per VCPU.
> >
Just to clarify that without the other patches in this series, XIVE
KVM devices still allocate a fixed block of 2048 VPs, no matter what.
ie, the last sentence in the changelog may be slightly misleading.
> > Signed-off-by: Greg Kurz <groug@kaod.org>
>
> Applied to ppc-for-4.2, thanks.
>
> > ---
> > hw/ppc/spapr.c | 7 ++++++-
> > include/hw/ppc/spapr.h | 1 +
> > 2 files changed, 7 insertions(+), 1 deletion(-)
> >
> > diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
> > index 8f59f08c102e..473ce1d04775 100644
> > --- a/hw/ppc/spapr.c
> > +++ b/hw/ppc/spapr.c
> > @@ -2503,6 +2503,7 @@ static CPUArchId *spapr_find_cpu_slot(MachineState *ms, uint32_t id, int *idx)
> > static void spapr_set_vsmt_mode(SpaprMachineState *spapr, Error **errp)
> > {
> > MachineState *ms = MACHINE(spapr);
> > + SpaprMachineClass *smc = SPAPR_MACHINE_GET_CLASS(spapr);
> > Error *local_err = NULL;
> > bool vsmt_user = !!spapr->vsmt;
> > int kvm_smt = kvmppc_smt_threads();
> > @@ -2529,7 +2530,7 @@ static void spapr_set_vsmt_mode(SpaprMachineState *spapr, Error **errp)
> > goto out;
> > }
> > /* In this case, spapr->vsmt has been set by the command line */
> > - } else {
> > + } else if (!smc->smp_threads_vsmt) {
> > /*
> > * Default VSMT value is tricky, because we need it to be as
> > * consistent as possible (for migration), but this requires
> > @@ -2538,6 +2539,8 @@ static void spapr_set_vsmt_mode(SpaprMachineState *spapr, Error **errp)
> > * overwhelmingly common case in production systems.
> > */
> > spapr->vsmt = MAX(8, smp_threads);
> > + } else {
> > + spapr->vsmt = smp_threads;
> > }
> >
> > /* KVM: If necessary, set the SMT mode: */
> > @@ -4452,6 +4455,7 @@ static void spapr_machine_class_init(ObjectClass *oc, void *data)
> > smc->irq = &spapr_irq_dual;
> > smc->dr_phb_enabled = true;
> > smc->linux_pci_probe = true;
> > + smc->smp_threads_vsmt = true;
> > }
> >
> > static const TypeInfo spapr_machine_info = {
> > @@ -4519,6 +4523,7 @@ static void spapr_machine_4_1_class_options(MachineClass *mc)
> >
> > spapr_machine_4_2_class_options(mc);
> > smc->linux_pci_probe = false;
> > + smc->smp_threads_vsmt = false;
> > compat_props_add(mc->compat_props, hw_compat_4_1, hw_compat_4_1_len);
> > compat_props_add(mc->compat_props, compat, G_N_ELEMENTS(compat));
> > }
> > diff --git a/include/hw/ppc/spapr.h b/include/hw/ppc/spapr.h
> > index cbd1a4c9f390..2009eb64f9cb 100644
> > --- a/include/hw/ppc/spapr.h
> > +++ b/include/hw/ppc/spapr.h
> > @@ -122,6 +122,7 @@ struct SpaprMachineClass {
> > bool broken_host_serial_model; /* present real host info to the guest */
> > bool pre_4_1_migration; /* don't migrate hpt-max-page-size */
> > bool linux_pci_probe;
> > + bool smp_threads_vsmt; /* set VSMT to smp_threads by default */
> >
> > void (*phb_placement)(SpaprMachineState *spapr, uint32_t index,
> > uint64_t *buid, hwaddr *pio,
> >
>
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
prev parent reply other threads:[~2019-10-14 11:43 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-03 12:00 [PATCH 0/7] spapr: Use less XIVE HW resources in KVM Greg Kurz
2019-10-03 12:00 ` [PATCH 1/7] spapr, xics: Get number of servers with a XICSFabricClass method Greg Kurz
2019-10-03 12:24 ` Cédric Le Goater
2019-10-03 12:49 ` Greg Kurz
2019-10-03 12:58 ` Cédric Le Goater
2019-10-03 13:02 ` Greg Kurz
2019-10-03 13:19 ` Cédric Le Goater
2019-10-03 13:41 ` Greg Kurz
2019-10-03 13:59 ` Cédric Le Goater
2019-10-03 14:58 ` Greg Kurz
2019-10-03 12:01 ` [PATCH 2/7] spapr, xive: Turn "nr-ends" property into "nr-servers" property Greg Kurz
2019-10-03 12:21 ` Cédric Le Goater
2019-10-03 12:44 ` Greg Kurz
2019-10-04 4:07 ` David Gibson
2019-10-04 5:53 ` Cédric Le Goater
2019-10-04 6:52 ` Greg Kurz
2019-10-04 7:27 ` Cédric Le Goater
2019-10-04 6:51 ` Greg Kurz
2019-10-05 10:23 ` David Gibson
2019-10-03 12:01 ` [PATCH 3/7] spapr, xics, xive: Drop nr_servers argument in DT-related functions Greg Kurz
2019-10-03 12:25 ` Cédric Le Goater
2019-10-03 12:52 ` Greg Kurz
2019-10-03 12:01 ` [PATCH RFC 4/7] linux-headers: Update against 5.3-rc2 Greg Kurz
2019-10-03 12:01 ` [PATCH 5/7] spapr/xics: Configure number of servers in KVM Greg Kurz
2019-10-03 12:29 ` Cédric Le Goater
2019-10-03 12:55 ` Greg Kurz
2019-10-03 12:01 ` [PATCH 6/7] spapr/xive: " Greg Kurz
2019-10-03 12:30 ` Cédric Le Goater
2019-10-03 12:02 ` [PATCH 7/7] spapr: Set VSMT to smp_threads by default Greg Kurz
2019-10-14 6:12 ` David Gibson
2019-10-14 11:31 ` Greg Kurz [this message]
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=20191014133149.686b681e@bahia.lan \
--to=groug@kaod.org \
--cc=clg@kaod.org \
--cc=david@gibson.dropbear.id.au \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.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).