| Message ID | 157010411139.246126.16419749660388287086.stgit@bahia.lan (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | spapr: Use less XIVE HW resources in KVM | expand |
On 03/10/2019 14:01, Greg Kurz wrote: > The XIVE KVM devices now has an attribute to configure the number of > interrupt servers. This allows to greatly optimize the usage of the VP > space in the XIVE HW, and thus to start a lot more VMs. > > Only set this attribute if available in order to support older POWER9 > KVM. > > The XIVE KVM device now reports the exhaustion of VPs upon the > connection of the first VCPU. Check that in order to have a chance > to provide an hint to the user. > > Signed-off-by: Greg Kurz <groug@kaod.org> Reviewed-by: Cédric Le Goater <clg@kaod.org> > --- > hw/intc/spapr_xive_kvm.c | 22 ++++++++++++++++++++-- > 1 file changed, 20 insertions(+), 2 deletions(-) > > diff --git a/hw/intc/spapr_xive_kvm.c b/hw/intc/spapr_xive_kvm.c > index 51b334b676a1..2a3a9ef22b6f 100644 > --- a/hw/intc/spapr_xive_kvm.c > +++ b/hw/intc/spapr_xive_kvm.c > @@ -152,7 +152,8 @@ void kvmppc_xive_cpu_synchronize_state(XiveTCTX *tctx, Error **errp) > > void kvmppc_xive_cpu_connect(XiveTCTX *tctx, Error **errp) > { > - SpaprXive *xive = SPAPR_MACHINE(qdev_get_machine())->xive; > + MachineState *ms = MACHINE(qdev_get_machine()); > + SpaprXive *xive = SPAPR_MACHINE(ms)->xive; > unsigned long vcpu_id; > int ret; > > @@ -171,8 +172,15 @@ void kvmppc_xive_cpu_connect(XiveTCTX *tctx, Error **errp) > ret = kvm_vcpu_enable_cap(tctx->cs, KVM_CAP_PPC_IRQ_XIVE, 0, xive->fd, > vcpu_id, 0); > if (ret < 0) { > - error_setg(errp, "XIVE: unable to connect CPU%ld to KVM device: %s", > + Error *err = NULL; > + > + error_setg(&err, "XIVE: unable to connect CPU%ld to KVM device: %s", > vcpu_id, strerror(errno)); > + if (errno == ENOSPC) { > + error_append_hint(&local_err, "Try -smp maxcpus=N with N < %u\n", > + ms->smp.max_cpus); > + } > + error_propagate(errp, err); > return; > } > > @@ -768,6 +776,16 @@ void kvmppc_xive_connect(SpaprXive *xive, Error **errp) > return; > } > > + /* Tell KVM about the # of VCPUs we may have */ > + if (kvm_device_check_attr(xive->fd, KVM_DEV_XIVE_GRP_CTRL, > + KVM_DEV_XIVE_NR_SERVERS)) { > + if (kvm_device_access(xive->fd, KVM_DEV_XIVE_GRP_CTRL, > + KVM_DEV_XIVE_NR_SERVERS, &xive->nr_servers, true, > + &local_err)) { > + goto fail; > + } > + } > + > /* > * 1. Source ESB pages - KVM mapping > */ >
diff --git a/hw/intc/spapr_xive_kvm.c b/hw/intc/spapr_xive_kvm.c index 51b334b676a1..2a3a9ef22b6f 100644 --- a/hw/intc/spapr_xive_kvm.c +++ b/hw/intc/spapr_xive_kvm.c @@ -152,7 +152,8 @@ void kvmppc_xive_cpu_synchronize_state(XiveTCTX *tctx, Error **errp) void kvmppc_xive_cpu_connect(XiveTCTX *tctx, Error **errp) { - SpaprXive *xive = SPAPR_MACHINE(qdev_get_machine())->xive; + MachineState *ms = MACHINE(qdev_get_machine()); + SpaprXive *xive = SPAPR_MACHINE(ms)->xive; unsigned long vcpu_id; int ret; @@ -171,8 +172,15 @@ void kvmppc_xive_cpu_connect(XiveTCTX *tctx, Error **errp) ret = kvm_vcpu_enable_cap(tctx->cs, KVM_CAP_PPC_IRQ_XIVE, 0, xive->fd, vcpu_id, 0); if (ret < 0) { - error_setg(errp, "XIVE: unable to connect CPU%ld to KVM device: %s", + Error *err = NULL; + + error_setg(&err, "XIVE: unable to connect CPU%ld to KVM device: %s", vcpu_id, strerror(errno)); + if (errno == ENOSPC) { + error_append_hint(&local_err, "Try -smp maxcpus=N with N < %u\n", + ms->smp.max_cpus); + } + error_propagate(errp, err); return; } @@ -768,6 +776,16 @@ void kvmppc_xive_connect(SpaprXive *xive, Error **errp) return; } + /* Tell KVM about the # of VCPUs we may have */ + if (kvm_device_check_attr(xive->fd, KVM_DEV_XIVE_GRP_CTRL, + KVM_DEV_XIVE_NR_SERVERS)) { + if (kvm_device_access(xive->fd, KVM_DEV_XIVE_GRP_CTRL, + KVM_DEV_XIVE_NR_SERVERS, &xive->nr_servers, true, + &local_err)) { + goto fail; + } + } + /* * 1. Source ESB pages - KVM mapping */
The XIVE KVM devices now has an attribute to configure the number of interrupt servers. This allows to greatly optimize the usage of the VP space in the XIVE HW, and thus to start a lot more VMs. Only set this attribute if available in order to support older POWER9 KVM. The XIVE KVM device now reports the exhaustion of VPs upon the connection of the first VCPU. Check that in order to have a chance to provide an hint to the user. Signed-off-by: Greg Kurz <groug@kaod.org> --- hw/intc/spapr_xive_kvm.c | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-)