diff options
author | Ankur Arora <ankur.a.arora@oracle.com> | 2017-06-02 17:05:59 -0700 |
---|---|---|
committer | Juergen Gross <jgross@suse.com> | 2017-06-13 16:05:17 +0200 |
commit | 0b64ffb8db4e310f77a01079ca752d946a8526b5 (patch) | |
tree | 0975f7c4353148b0526752e84462945af44e7802 /arch/x86/xen/enlighten.c | |
parent | ad73fd595c2ab168fdd01a266cbe6e4df95f8db0 (diff) | |
download | linux-0b64ffb8db4e310f77a01079ca752d946a8526b5.tar.bz2 |
xen/pvh*: Support > 32 VCPUs at domain restore
When Xen restores a PVHVM or PVH guest, its shared_info only holds
up to 32 CPUs. The hypercall VCPUOP_register_vcpu_info allows
us to setup per-page areas for VCPUs. This means we can boot
PVH* guests with more than 32 VCPUs. During restore the per-cpu
structure is allocated freshly by the hypervisor (vcpu_info_mfn is
set to INVALID_MFN) so that the newly restored guest can make a
VCPUOP_register_vcpu_info hypercall.
However, we end up triggering this condition in Xen:
/* Run this command on yourself or on other offline VCPUS. */
if ( (v != current) && !test_bit(_VPF_down, &v->pause_flags) )
which means we are unable to setup the per-cpu VCPU structures
for running VCPUS. The Linux PV code paths makes this work by
iterating over cpu_possible in xen_vcpu_restore() with:
1) is target CPU up (VCPUOP_is_up hypercall?)
2) if yes, then VCPUOP_down to pause it
3) VCPUOP_register_vcpu_info
4) if it was down, then VCPUOP_up to bring it back up
With Xen commit 192df6f9122d ("xen/x86: allow HVM guests to use
hypercalls to bring up vCPUs") this is available for non-PV guests.
As such first check if VCPUOP_is_up is actually possible before
trying this dance.
As most of this dance code is done already in xen_vcpu_restore()
let's make it callable on PV, PVH and PVHVM.
Based-on-patch-by: Konrad Wilk <konrad.wilk@oracle.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Diffstat (limited to 'arch/x86/xen/enlighten.c')
-rw-r--r-- | arch/x86/xen/enlighten.c | 45 |
1 files changed, 32 insertions, 13 deletions
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 96b745e3f56c..276cc21619ec 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -106,6 +106,21 @@ int xen_cpuhp_setup(int (*cpu_up_prepare_cb)(unsigned int), return rc >= 0 ? 0 : rc; } +static void xen_vcpu_setup_restore(int cpu) +{ + /* Any per_cpu(xen_vcpu) is stale, so reset it */ + xen_vcpu_info_reset(cpu); + + /* + * For PVH and PVHVM, setup online VCPUs only. The rest will + * be handled by hotplug. + */ + if (xen_pv_domain() || + (xen_hvm_domain() && cpu_online(cpu))) { + xen_vcpu_setup(cpu); + } +} + /* * On restore, set the vcpu placement up again. * If it fails, then we're in a bad state, since @@ -117,17 +132,23 @@ void xen_vcpu_restore(void) for_each_possible_cpu(cpu) { bool other_cpu = (cpu != smp_processor_id()); - bool is_up = HYPERVISOR_vcpu_op(VCPUOP_is_up, xen_vcpu_nr(cpu), - NULL); + bool is_up; + + if (xen_vcpu_nr(cpu) == XEN_VCPU_ID_INVALID) + continue; + + /* Only Xen 4.5 and higher support this. */ + is_up = HYPERVISOR_vcpu_op(VCPUOP_is_up, + xen_vcpu_nr(cpu), NULL) > 0; if (other_cpu && is_up && HYPERVISOR_vcpu_op(VCPUOP_down, xen_vcpu_nr(cpu), NULL)) BUG(); - xen_setup_runstate_info(cpu); + if (xen_pv_domain() || xen_feature(XENFEAT_hvm_safe_pvclock)) + xen_setup_runstate_info(cpu); - if (xen_have_vcpu_info_placement) - xen_vcpu_setup(cpu); + xen_vcpu_setup_restore(cpu); if (other_cpu && is_up && HYPERVISOR_vcpu_op(VCPUOP_up, xen_vcpu_nr(cpu), NULL)) @@ -163,11 +184,11 @@ void xen_vcpu_setup(int cpu) BUG_ON(HYPERVISOR_shared_info == &xen_dummy_shared_info); /* - * This path is called twice on PVHVM - first during bootup via - * smp_init -> xen_hvm_cpu_notify, and then if the VCPU is being - * hotplugged: cpu_up -> xen_hvm_cpu_notify. - * As we can only do the VCPUOP_register_vcpu_info once lets - * not over-write its result. + * This path is called on PVHVM at bootup (xen_hvm_smp_prepare_boot_cpu) + * and at restore (xen_vcpu_restore). Also called for hotplugged + * VCPUs (cpu_init -> xen_hvm_cpu_prepare_hvm). + * However, the hypercall can only be done once (see below) so if a VCPU + * is offlined and comes back online then let's not redo the hypercall. * * For PV it is called during restore (xen_vcpu_restore) and bootup * (xen_setup_vcpu_info_placement). The hotplug mechanism does not @@ -178,8 +199,6 @@ void xen_vcpu_setup(int cpu) return; } - xen_vcpu_info_reset(cpu); - if (xen_have_vcpu_info_placement) { vcpup = &per_cpu(xen_vcpu_info, cpu); info.mfn = arbitrary_virt_to_mfn(vcpup); @@ -214,7 +233,7 @@ void xen_vcpu_setup(int cpu) if (!xen_have_vcpu_info_placement) { if (cpu >= MAX_VIRT_CPUS) clamp_max_cpus(); - return; + xen_vcpu_info_reset(cpu); } } |