Code for initializing/tearing down PMU for PV guests
Signed-off-by: Boris Ostrovsky <boris.ostrov...@oracle.com>
Acked-by: Kevin Tian <kevin.t...@intel.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.w...@oracle.com>
Reviewed-by: Dietmar Hahn <dietmar.h...@ts.fujitsu.com>
Tested-by: Dietmar Hahn <dietmar.h...@ts.fujitsu.com>
---
tools/flask/policy/policy/modules/xen/xen.te | 4 ++
xen/arch/x86/domain.c | 2 +
xen/arch/x86/hvm/hvm.c | 1 +
xen/arch/x86/hvm/svm/svm.c | 4 +-
xen/arch/x86/hvm/svm/vpmu.c | 44 +++++++++----
xen/arch/x86/hvm/vmx/vmx.c | 4 +-
xen/arch/x86/hvm/vmx/vpmu_core2.c | 81 ++++++++++++++++-------
xen/arch/x86/hvm/vpmu.c | 98 +++++++++++++++++++++++++++-
xen/common/event_channel.c | 1 +
xen/include/asm-x86/hvm/vpmu.h | 2 +
xen/include/public/pmu.h | 2 +
xen/include/public/xen.h | 1 +
xen/include/xsm/dummy.h | 3 +
xen/xsm/flask/hooks.c | 4 ++
xen/xsm/flask/policy/access_vectors | 2 +
15 files changed, 212 insertions(+), 41 deletions(-)
diff --git a/tools/flask/policy/policy/modules/xen/xen.te
b/tools/flask/policy/policy/modules/xen/xen.te
index ae7bf3c..9d84004 100644
--- a/tools/flask/policy/policy/modules/xen/xen.te
+++ b/tools/flask/policy/policy/modules/xen/xen.te
@@ -120,6 +120,10 @@ domain_comms(dom0_t, dom0_t)
# Allow all domains to use (unprivileged parts of) the tmem hypercall
allow domain_type xen_t:xen tmem_op;
+# Allow all domains to use PMU (but not to change its settings --- that's what
+# pmu_ctrl is for)
+allow domain_type xen_t:xen2 pmu_use;
+
###############################################################################
#
# Domain creation
diff --git a/xen/arch/x86/domain.c b/xen/arch/x86/domain.c
index da5bdf4..ce1d187 100644
--- a/xen/arch/x86/domain.c
+++ b/xen/arch/x86/domain.c
@@ -445,6 +445,8 @@ int vcpu_initialise(struct vcpu *v)
vmce_init_vcpu(v);
+ spin_lock_init(&v->arch.vpmu.vpmu_lock);
+
if ( has_hvm_container_domain(d) )
{
rc = hvm_vcpu_initialise(v);
diff --git a/xen/arch/x86/hvm/hvm.c b/xen/arch/x86/hvm/hvm.c
index 8f49b44..ec9c89a 100644
--- a/xen/arch/x86/hvm/hvm.c
+++ b/xen/arch/x86/hvm/hvm.c
@@ -4820,6 +4820,7 @@ static hvm_hypercall_t *const
pvh_hypercall64_table[NR_hypercalls] = {
HYPERCALL(hvm_op),
HYPERCALL(sysctl),
HYPERCALL(domctl),
+ HYPERCALL(xenpmu_op),
[ __HYPERVISOR_arch_1 ] = (hvm_hypercall_t *)paging_domctl_continuation
};
diff --git a/xen/arch/x86/hvm/svm/svm.c b/xen/arch/x86/hvm/svm/svm.c
index 8aca6e6..b1c4845 100644
--- a/xen/arch/x86/hvm/svm/svm.c
+++ b/xen/arch/x86/hvm/svm/svm.c
@@ -1157,7 +1157,9 @@ static int svm_vcpu_initialise(struct vcpu *v)
return rc;
}
- vpmu_initialise(v);
+ /* PVH's VPMU is initialized via hypercall */
+ if ( is_hvm_vcpu(v) )
+ vpmu_initialise(v);
svm_guest_osvw_init(v);
diff --git a/xen/arch/x86/hvm/svm/vpmu.c b/xen/arch/x86/hvm/svm/vpmu.c
index 8460d7b..fe852ed 100644
--- a/xen/arch/x86/hvm/svm/vpmu.c
+++ b/xen/arch/x86/hvm/svm/vpmu.c
@@ -363,17 +363,19 @@ static void amd_vpmu_destroy(struct vcpu *v)
if ( !vpmu_is_set(vpmu, VPMU_CONTEXT_ALLOCATED) )
return;
- if ( has_hvm_container_vcpu(v) && is_msr_bitmap_on(vpmu) )
- amd_vpmu_unset_msr_bitmap(v);
+ if ( has_hvm_container_vcpu(v) )
+ {
+ if ( is_msr_bitmap_on(vpmu) )
+ amd_vpmu_unset_msr_bitmap(v);
- xfree(vpmu->context);
- vpmu_reset(vpmu, VPMU_CONTEXT_ALLOCATED);
+ if ( is_hvm_vcpu(v) )
+ xfree(vpmu->context);
- if ( vpmu_is_set(vpmu, VPMU_RUNNING) )
- {
- vpmu_reset(vpmu, VPMU_RUNNING);
release_pmu_ownship(PMU_OWNER_HVM);
}
+
+ vpmu->context = NULL;
+ vpmu_clear(vpmu);
}
/* VPMU part of the 'q' keyhandler */
@@ -439,15 +441,19 @@ int svm_vpmu_initialise(struct vcpu *v)
if ( !counters )
return -EINVAL;
- ctxt = xzalloc_bytes(sizeof(*ctxt) +
- 2 * sizeof(uint64_t) * num_counters);
- if ( !ctxt )
+ if ( is_hvm_vcpu(v) )
{
- printk(XENLOG_G_WARNING "Insufficient memory for PMU, "
- " PMU feature is unavailable on domain %d vcpu %d.\n",
- v->vcpu_id, v->domain->domain_id);
- return -ENOMEM;
+ ctxt = xzalloc_bytes(sizeof(*ctxt) +
+ 2 * sizeof(uint64_t) * num_counters);
+ if ( !ctxt )
+ {
+ printk(XENLOG_G_WARNING "%pv: Insufficient memory for PMU, "
+ " PMU feature is unavailable\n", v);
+ return -ENOMEM;
+ }
}
+ else
+ ctxt = &v->arch.vpmu.xenpmu_data->pmu.c.amd;
ctxt->counters = sizeof(*ctxt);
ctxt->ctrls = ctxt->counters + sizeof(uint64_t) * num_counters;
@@ -489,6 +495,16 @@ int __init amd_vpmu_init(void)
return -EINVAL;
}
+ if ( sizeof(struct xen_pmu_data) +
+ 2 * sizeof(uint64_t) * num_counters > PAGE_SIZE )
+ {
+ printk(XENLOG_WARNING
+ "VPMU: Register bank does not fit into VPMU shared page\n");
+ counters = ctrls = NULL;
+ num_counters = 0;
+ return -ENOSPC;
+ }
+
return 0;
}
diff --git a/xen/arch/x86/hvm/vmx/vmx.c b/xen/arch/x86/hvm/vmx/vmx.c
index 0bf92b2..a7c3a7a 100644
--- a/xen/arch/x86/hvm/vmx/vmx.c
+++ b/xen/arch/x86/hvm/vmx/vmx.c
@@ -116,7 +116,9 @@ static int vmx_vcpu_initialise(struct vcpu *v)
return rc;
}
- vpmu_initialise(v);
+ /* PVH's VPMU is initialized via hypercall */
+ if ( is_hvm_vcpu(v) )
+ vpmu_initialise(v);
vmx_install_vlapic_mapping(v);
diff --git a/xen/arch/x86/hvm/vmx/vpmu_core2.c
b/xen/arch/x86/hvm/vmx/vpmu_core2.c
index e199367..03dc981 100644
--- a/xen/arch/x86/hvm/vmx/vpmu_core2.c
+++ b/xen/arch/x86/hvm/vmx/vpmu_core2.c
@@ -362,24 +362,34 @@ static int core2_vpmu_alloc_resource(struct vcpu *v)
struct xen_pmu_intel_ctxt *core2_vpmu_cxt = NULL;
uint64_t *p = NULL;
- if ( !acquire_pmu_ownership(PMU_OWNER_HVM) )
- return 0;
-
- wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0);
- if ( vmx_add_host_load_msr(MSR_CORE_PERF_GLOBAL_CTRL) )
+ p = xzalloc_bytes(sizeof(uint64_t));
+ if ( !p )
goto out_err;
- if ( vmx_add_guest_msr(MSR_CORE_PERF_GLOBAL_CTRL) )
- goto out_err;
- vmx_write_guest_msr(MSR_CORE_PERF_GLOBAL_CTRL, 0);
-
- core2_vpmu_cxt = xzalloc_bytes(sizeof(*core2_vpmu_cxt) +
- sizeof(uint64_t) * fixed_pmc_cnt +
- sizeof(struct xen_pmu_cntr_pair) *
- arch_pmc_cnt);
- p = xzalloc(uint64_t);
- if ( !core2_vpmu_cxt || !p )
- goto out_err;
+ if ( has_hvm_container_vcpu(v) )
+ {
+ if ( is_hvm_vcpu(v) && !acquire_pmu_ownership(PMU_OWNER_HVM) )
+ goto out_err;
+
+ wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0);
+ if ( vmx_add_host_load_msr(MSR_CORE_PERF_GLOBAL_CTRL) )
+ goto out_err_hvm;
+ if ( vmx_add_guest_msr(MSR_CORE_PERF_GLOBAL_CTRL) )
+ goto out_err_hvm;
+ vmx_write_guest_msr(MSR_CORE_PERF_GLOBAL_CTRL, 0);
+ }
+
+ if ( is_hvm_vcpu(v) )
+ {
+ core2_vpmu_cxt = xzalloc_bytes(sizeof(*core2_vpmu_cxt) +
+ sizeof(uint64_t) * fixed_pmc_cnt +
+ sizeof(struct xen_pmu_cntr_pair) *
+ arch_pmc_cnt);
+ if ( !core2_vpmu_cxt )
+ goto out_err_hvm;
+ }
+ else
+ core2_vpmu_cxt = &v->arch.vpmu.xenpmu_data->pmu.c.intel;
core2_vpmu_cxt->fixed_counters = sizeof(*core2_vpmu_cxt);
core2_vpmu_cxt->arch_counters = core2_vpmu_cxt->fixed_counters +
@@ -392,10 +402,12 @@ static int core2_vpmu_alloc_resource(struct vcpu *v)
return 1;
-out_err:
- release_pmu_ownship(PMU_OWNER_HVM);
-
+ out_err_hvm:
xfree(core2_vpmu_cxt);
+ if ( is_hvm_vcpu(v) )
+ release_pmu_ownship(PMU_OWNER_HVM);
+
+ out_err:
xfree(p);
printk("Failed to allocate VPMU resources for domain %u vcpu %u\n",
@@ -715,12 +727,20 @@ static void core2_vpmu_destroy(struct vcpu *v)
if ( !vpmu_is_set(vpmu, VPMU_CONTEXT_ALLOCATED) )
return;
- xfree(vpmu->context);
+ if ( has_hvm_container_vcpu(v) )
+ {
+ if ( cpu_has_vmx_msr_bitmap )
+ core2_vpmu_unset_msr_bitmap(v->arch.hvm_vmx.msr_bitmap);
+
+ if ( is_hvm_vcpu(v) )
+ xfree(vpmu->context);
+
+ release_pmu_ownship(PMU_OWNER_HVM);
+ }
+
xfree(vpmu->priv_context);
- if ( has_hvm_container_vcpu(v) && cpu_has_vmx_msr_bitmap )
- core2_vpmu_unset_msr_bitmap(v->arch.hvm_vmx.msr_bitmap);
- release_pmu_ownship(PMU_OWNER_HVM);
- vpmu_reset(vpmu, VPMU_CONTEXT_ALLOCATED);
+ vpmu->context = NULL;
+ vpmu_clear(vpmu);
}
struct arch_vpmu_ops core2_vpmu_ops = {
@@ -830,6 +850,10 @@ int vmx_vpmu_initialise(struct vcpu *v)
ds_warned = 1;
func_out:
+ /* PV domains can allocate resources immediately */
+ if ( is_pv_vcpu(v) && !core2_vpmu_alloc_resource(v) )
+ return -EIO;
+
vpmu->arch_vpmu_ops = &core2_vpmu_ops;
return 0;
@@ -895,5 +919,14 @@ int __init core2_vpmu_init(void)
check_pmc_quirk();
+ if ( sizeof(struct xen_pmu_data) + sizeof(uint64_t) * fixed_pmc_cnt +
+ sizeof(struct xen_pmu_cntr_pair) * arch_pmc_cnt > PAGE_SIZE )
+ {
+ printk(XENLOG_WARNING
+ "VPMU: Register bank does not fit into VPMU share page\n");
+ arch_pmc_cnt = fixed_pmc_cnt = 0;
+ return -ENOSPC;
+ }
+
return 0;
}
diff --git a/xen/arch/x86/hvm/vpmu.c b/xen/arch/x86/hvm/vpmu.c
index ed3b99a..2ad9832 100644
--- a/xen/arch/x86/hvm/vpmu.c
+++ b/xen/arch/x86/hvm/vpmu.c
@@ -26,6 +26,7 @@
#include <asm/regs.h>
#include <asm/types.h>
#include <asm/msr.h>
+#include <asm/p2m.h>
#include <asm/hvm/support.h>
#include <asm/hvm/vmx/vmx.h>
#include <asm/hvm/vmx/vmcs.h>
@@ -252,6 +253,7 @@ void vpmu_initialise(struct vcpu *v)
vpmu_destroy(v);
vpmu_clear(vpmu);
vpmu->context = NULL;
+ vpmu->hw_lapic_lvtpc = PMU_APIC_VECTOR | APIC_LVT_MASKED;
switch ( vendor )
{
@@ -278,7 +280,89 @@ void vpmu_destroy(struct vcpu *v)
struct vpmu_struct *vpmu = vcpu_vpmu(v);
if ( vpmu->arch_vpmu_ops && vpmu->arch_vpmu_ops->arch_vpmu_destroy )
+ {
+ /* Unload VPMU first. This will stop counters */
+ on_selected_cpus(cpumask_of(vcpu_vpmu(v)->last_pcpu),
+ vpmu_save_force, v, 1);
+
vpmu->arch_vpmu_ops->arch_vpmu_destroy(v);
+ }
+}
+
+static int pvpmu_init(struct domain *d, xen_pmu_params_t *params)
+{
+ struct vcpu *v;
+ struct vpmu_struct *vpmu;
+ struct page_info *page;
+ uint64_t gfn = params->val;
+
+ if ( vpmu_mode == XENPMU_MODE_OFF )
+ return -EINVAL;
+
+ if ( (params->vcpu >= d->max_vcpus) || (d->vcpu == NULL) ||
+ (d->vcpu[params->vcpu] == NULL) )
+ return -EINVAL;
+
+ page = get_page_from_gfn(d, gfn, NULL, P2M_ALLOC);
+ if ( !page )
+ return -EINVAL;
+
+ if ( !get_page_type(page, PGT_writable_page) )
+ {
+ put_page(page);
+ return -EINVAL;
+ }
+
+ v = d->vcpu[params->vcpu];
+ vpmu = vcpu_vpmu(v);
+ spin_lock(&vpmu->vpmu_lock);
+
+ v->arch.vpmu.xenpmu_data = __map_domain_page_global(page);
+ if ( !v->arch.vpmu.xenpmu_data )
+ {
+ put_page_and_type(page);
+ spin_unlock(&vpmu->vpmu_lock);
+ return -EINVAL;
+ }
+
+ vpmu_initialise(v);
+
+ spin_unlock(&vpmu->vpmu_lock);
+
+ return 0;
+}
+
+static void pvpmu_finish(struct domain *d, xen_pmu_params_t *params)
+{
+ struct vcpu *v;
+ struct vpmu_struct *vpmu;
+ uint64_t mfn;
+
+ if ( (params->vcpu >= d->max_vcpus) || (d->vcpu == NULL) ||
+ (d->vcpu[params->vcpu] == NULL) )
+ return;
+
+ v = d->vcpu[params->vcpu];
+ if ( v != current )
+ vcpu_pause(v);
+
+ vpmu = vcpu_vpmu(v);
+ spin_lock(&vpmu->vpmu_lock);
+
+ if ( v->arch.vpmu.xenpmu_data )
+ {
+ mfn = domain_page_map_to_mfn(v->arch.vpmu.xenpmu_data);
+ ASSERT(mfn != 0);
+ unmap_domain_page_global(v->arch.vpmu.xenpmu_data);
+ put_page_and_type(mfn_to_page(mfn));
+ v->arch.vpmu.xenpmu_data = NULL;
+ }
+ vpmu_destroy(v);
+
+ spin_unlock(&vpmu->vpmu_lock);
+
+ if ( v != current )
+ vcpu_unpause(v);
}
/* Dump some vpmu informations on console. Used in keyhandler dump_domains().
*/
@@ -402,7 +486,7 @@ long do_xenpmu_op(int op,
XEN_GUEST_HANDLE_PARAM(xen_pmu_params_t) arg)
if ( copy_to_guest(arg, &pmu_params, 1) )
return -EFAULT;
- break;
+ break;
case XENPMU_feature_set:
if ( copy_from_guest(&pmu_params, arg, 1) )
@@ -420,6 +504,18 @@ long do_xenpmu_op(int op,
XEN_GUEST_HANDLE_PARAM(xen_pmu_params_t) arg)
return -EFAULT;
break;
+ case XENPMU_init:
+ if ( copy_from_guest(&pmu_params, arg, 1) )
+ return -EFAULT;
+ ret = pvpmu_init(current->domain, &pmu_params);
+ break;
+
+ case XENPMU_finish:
+ if ( copy_from_guest(&pmu_params, arg, 1) )
+ return -EFAULT;
+ pvpmu_finish(current->domain, &pmu_params);
+ break;
+
default:
ret = -EINVAL;
}
diff --git a/xen/common/event_channel.c b/xen/common/event_channel.c
index 7d6de54..a991b2d 100644
--- a/xen/common/event_channel.c
+++ b/xen/common/event_channel.c
@@ -108,6 +108,7 @@ static int virq_is_global(uint32_t virq)
case VIRQ_TIMER:
case VIRQ_DEBUG:
case VIRQ_XENOPROF:
+ case VIRQ_XENPMU:
rc = 0;
break;
case VIRQ_ARCH_0 ... VIRQ_ARCH_7:
diff --git a/xen/include/asm-x86/hvm/vpmu.h b/xen/include/asm-x86/hvm/vpmu.h
index 97fe17c..ada2ab7 100644
--- a/xen/include/asm-x86/hvm/vpmu.h
+++ b/xen/include/asm-x86/hvm/vpmu.h
@@ -64,6 +64,8 @@ struct vpmu_struct {
void *context; /* May be shared with PV guest */
void *priv_context; /* hypervisor-only */
struct arch_vpmu_ops *arch_vpmu_ops;
+ struct xen_pmu_data *xenpmu_data;
+ spinlock_t vpmu_lock;
};
/* VPMU states */
diff --git a/xen/include/public/pmu.h b/xen/include/public/pmu.h
index 66cc494..afb4ca1 100644
--- a/xen/include/public/pmu.h
+++ b/xen/include/public/pmu.h
@@ -25,6 +25,8 @@
#define XENPMU_mode_set 1
#define XENPMU_feature_get 2
#define XENPMU_feature_set 3
+#define XENPMU_init 4
+#define XENPMU_finish 5
/* ` } */
/* Parameters structure for HYPERVISOR_xenpmu_op call */
diff --git a/xen/include/public/xen.h b/xen/include/public/xen.h
index 0766790..e4d0b79 100644
--- a/xen/include/public/xen.h
+++ b/xen/include/public/xen.h
@@ -161,6 +161,7 @@ DEFINE_XEN_GUEST_HANDLE(xen_ulong_t);
#define VIRQ_MEM_EVENT 10 /* G. (DOM0) A memory event has occured */
#define VIRQ_XC_RESERVED 11 /* G. Reserved for XenClient */
#define VIRQ_ENOMEM 12 /* G. (DOM0) Low on heap memory */
+#define VIRQ_XENPMU 13 /* V. PMC interrupt */
/* Architecture-specific VIRQ definitions. */
#define VIRQ_ARCH_0 16
diff --git a/xen/include/xsm/dummy.h b/xen/include/xsm/dummy.h
index c637454..ae47135 100644
--- a/xen/include/xsm/dummy.h
+++ b/xen/include/xsm/dummy.h
@@ -665,6 +665,9 @@ static XSM_INLINE int xsm_pmu_op (XSM_DEFAULT_ARG struct
domain *d, int op)
case XENPMU_feature_set:
case XENPMU_feature_get:
return xsm_default_action(XSM_PRIV, d, current->domain);
+ case XENPMU_init:
+ case XENPMU_finish:
+ return xsm_default_action(XSM_HOOK, d, current->domain);
default:
return -EPERM;
}
diff --git a/xen/xsm/flask/hooks.c b/xen/xsm/flask/hooks.c
index 8e1914e..76f2cf1 100644
--- a/xen/xsm/flask/hooks.c
+++ b/xen/xsm/flask/hooks.c
@@ -1515,6 +1515,10 @@ static int flask_pmu_op (struct domain *d, int op)
case XENPMU_feature_get:
return avc_has_perm(dsid, SECINITSID_XEN, SECCLASS_XEN2,
XEN2__PMU_CTRL, NULL);
+ case XENPMU_init:
+ case XENPMU_finish:
+ return avc_has_perm(dsid, SECINITSID_XEN, SECCLASS_XEN2,
+ XEN2__PMU_USE, NULL);
default:
return -EPERM;
}
diff --git a/xen/xsm/flask/policy/access_vectors
b/xen/xsm/flask/policy/access_vectors
index 3289d98..b84af31 100644
--- a/xen/xsm/flask/policy/access_vectors
+++ b/xen/xsm/flask/policy/access_vectors
@@ -87,6 +87,8 @@ class xen2
get_symbol
# PMU control
pmu_ctrl
+# PMU use (domains, including unprivileged ones, will be using this operation)
+ pmu_use
}
# Classes domain and domain2 consist of operations that a domain performs on
--
1.8.1.4
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
