x86/VPMU: implement ipc and arch filter flags

This introduces a way to have a restricted VPMU, by specifying one of two
predefined groups of PMCs to make available. For secure environments, this
allows the VPMU to be used without needing to enable all PMCs.

Signed-off-by: Brendan Gregg <bgregg@netflix.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Acked-by: Kevin Tian <kevin.tian@intel.com>

diff --git a/docs/misc/xen-command-line.markdown b/docs/misc/xen-command-line.markdown
index 467dc8f760cbcfe344702b8fe00aa8128ba860e0..5ed0730783107510a2d8191c4ab3ca51c4caa8f7 100644 (file)
--- a/docs/misc/xen-command-line.markdown
+++ b/docs/misc/xen-command-line.markdown
@@ -1476,7 +1476,7 @@ Use Virtual Processor ID support if available.  This prevents the need for TLB
 flushes on VM entry and exit, increasing performance.
 
 ### vpmu
-> `= ( bts )`
+> `= ( <boolean> | { bts | ipc | arch [, ...] } )`
 
 > Default: `off`
 
@@ -1492,6 +1492,19 @@ wrong behaviour (see handle\_pmc\_quirk()).
 If 'vpmu=bts' is specified the virtualisation of the Branch Trace Store (BTS)
 feature is switched on on Intel processors supporting this feature.
 
+vpmu=ipc enables performance monitoring, but restricts the counters to the
+most minimum set possible: instructions, cycles, and reference cycles. These
+can be used to calculate instructions per cycle (IPC).
+
+vpmu=arch enables performance monitoring, but restricts the counters to the
+pre-defined architectural events only. These are exposed by cpuid, and listed
+in the Pre-Defined Architectural Performance Events table from the Intel 64
+and IA-32 Architectures Software Developer's Manual, Volume 3B, System
+Programming Guide, Part 2.
+
+If a boolean is not used, combinations of flags are allowed, comma separated.
+For example, vpmu=arch,bts.
+
 Note that if **watchdog** option is also specified vpmu will be turned off.
 
 *Warning:*

diff --git a/xen/arch/x86/cpu/vpmu.c b/xen/arch/x86/cpu/vpmu.c
index 4856e98d841c1181c957673a54a1216e48d3ce81..237b5ffd873621598a3935a73eff8b004596ac3e 100644 (file)
--- a/xen/arch/x86/cpu/vpmu.c
+++ b/xen/arch/x86/cpu/vpmu.c
@@ -43,34 +43,59 @@ CHECK_pmu_data;
 CHECK_pmu_params;
 
 /*
- * "vpmu" :     vpmu generally enabled
- * "vpmu=off" : vpmu generally disabled
- * "vpmu=bts" : vpmu enabled and Intel BTS feature switched on.
+ * "vpmu" :     vpmu generally enabled (all counters)
+ * "vpmu=off"  : vpmu generally disabled
+ * "vpmu=bts"  : vpmu enabled and Intel BTS feature switched on.
+ * "vpmu=ipc"  : vpmu enabled for IPC counters only (most restrictive)
+ * "vpmu=arch" : vpmu enabled for predef arch counters only (restrictive)
+ * flag combinations are allowed, eg, "vpmu=ipc,bts".
  */
 static unsigned int __read_mostly opt_vpmu_enabled;
 unsigned int __read_mostly vpmu_mode = XENPMU_MODE_OFF;
 unsigned int __read_mostly vpmu_features = 0;
-static void parse_vpmu_param(char *s);
-custom_param("vpmu", parse_vpmu_param);
+static void parse_vpmu_params(char *s);
+custom_param("vpmu", parse_vpmu_params);
 
 static DEFINE_SPINLOCK(vpmu_lock);
 static unsigned vpmu_count;
 
 static DEFINE_PER_CPU(struct vcpu *, last_vcpu);
 
-static void __init parse_vpmu_param(char *s)
+static int parse_vpmu_param(char *s, unsigned int len)
 {
+    if ( !*s || !len )
+        return 0;
+    if ( !strncmp(s, "bts", len) )
+        vpmu_features |= XENPMU_FEATURE_INTEL_BTS;
+    else if ( !strncmp(s, "ipc", len) )
+        vpmu_features |= XENPMU_FEATURE_IPC_ONLY;
+    else if ( !strncmp(s, "arch", len) )
+        vpmu_features |= XENPMU_FEATURE_ARCH_ONLY;
+    else
+        return 1;
+    return 0;
+}
+
+static void __init parse_vpmu_params(char *s)
+{
+    char *sep, *p = s;
+
     switch ( parse_bool(s) )
     {
     case 0:
         break;
     default:
-        if ( !strcmp(s, "bts") )
-            vpmu_features |= XENPMU_FEATURE_INTEL_BTS;
-        else if ( *s )
+        for ( ; ; )
         {
-            printk("VPMU: unknown flag: %s - vpmu disabled!\n", s);
-            break;
+            sep = strchr(p, ',');
+            if ( sep == NULL )
+                sep = strchr(p, 0);
+            if ( parse_vpmu_param(p, sep - p) )
+                goto error;
+            if ( !*sep )
+                /* reached end of flags */
+                break;
+            p = sep + 1;
         }
         /* fall through */
     case 1:
@@ -79,6 +104,10 @@ static void __init parse_vpmu_param(char *s)
         opt_vpmu_enabled = 1;
         break;
     }
+    return;
+
+ error:
+    printk("VPMU: unknown flags: %s - vpmu disabled!\n", s);
 }
 
 void vpmu_lvtpc_update(uint32_t val)
@@ -708,7 +737,9 @@ long do_xenpmu_op(unsigned int op, XEN_GUEST_HANDLE_PARAM(xen_pmu_params_t) arg)
         break;
 
     case XENPMU_feature_set:
-        if ( pmu_params.val & ~XENPMU_FEATURE_INTEL_BTS )
+        if ( pmu_params.val & ~(XENPMU_FEATURE_INTEL_BTS |
+                                XENPMU_FEATURE_IPC_ONLY |
+                                XENPMU_FEATURE_ARCH_ONLY))
             return -EINVAL;
 
         spin_lock(&vpmu_lock);

diff --git a/xen/arch/x86/cpu/vpmu_intel.c b/xen/arch/x86/cpu/vpmu_intel.c
index a179717161041c0eb5a4f7c98f2e0665bf3157a1..50b583115006a7044a1852227547a769f0ce1163 100644 (file)
--- a/xen/arch/x86/cpu/vpmu_intel.c
+++ b/xen/arch/x86/cpu/vpmu_intel.c
@@ -605,11 +605,16 @@ static int core2_vpmu_do_wrmsr(unsigned int msr, uint64_t msr_content,
                  "MSR_PERF_GLOBAL_STATUS(0x38E)!\n");
         return -EINVAL;
     case MSR_IA32_PEBS_ENABLE:
+        if ( vpmu_features & (XENPMU_FEATURE_IPC_ONLY |
+                              XENPMU_FEATURE_ARCH_ONLY) )
+            return -EINVAL;
         if ( msr_content )
             /* PEBS is reported as unavailable in MSR_IA32_MISC_ENABLE */
             return -EINVAL;
         return 0;
     case MSR_IA32_DS_AREA:
+        if ( !(vpmu_features & XENPMU_FEATURE_INTEL_BTS) )
+            return -EINVAL;
         if ( vpmu_is_set(vpmu, VPMU_CPU_HAS_DS) )
         {
             if ( !(has_hvm_container_vcpu(v)
@@ -658,12 +663,52 @@ static int core2_vpmu_do_wrmsr(unsigned int msr, uint64_t msr_content,
         tmp = msr - MSR_P6_EVNTSEL(0);
         if ( tmp >= 0 && tmp < arch_pmc_cnt )
         {
+            bool_t blocked = 0;
+            uint64_t umaskevent = msr_content & MSR_IA32_CMT_EVTSEL_UE_MASK;
             struct xen_pmu_cntr_pair *xen_pmu_cntr_pair =
                 vpmu_reg_pointer(core2_vpmu_cxt, arch_counters);
 
             if ( msr_content & ARCH_CTRL_MASK )
                 return -EINVAL;
 
+            /* PMC filters */
+            if ( vpmu_features & (XENPMU_FEATURE_IPC_ONLY |
+                                  XENPMU_FEATURE_ARCH_ONLY) )
+            {
+                blocked = 1;
+                switch ( umaskevent )
+                {
+                /*
+                 * See the Pre-Defined Architectural Performance Events table
+                 * from the Intel 64 and IA-32 Architectures Software
+                 * Developer's Manual, Volume 3B, System Programming Guide,
+                 * Part 2.
+                 */
+                case 0x003c:   /* UnHalted Core Cycles */
+                case 0x013c:   /* UnHalted Reference Cycles */
+                case 0x00c0:   /* Instructions Retired */
+                    blocked = 0;
+                    break;
+                }
+            }
+
+            if ( vpmu_features & XENPMU_FEATURE_ARCH_ONLY )
+            {
+                /* Additional counters beyond IPC only; blocked already set. */
+                switch ( umaskevent )
+                {
+                case 0x4f2e:   /* Last Level Cache References */
+                case 0x412e:   /* Last Level Cache Misses */
+                case 0x00c4:   /* Branch Instructions Retired */
+                case 0x00c5:   /* All Branch Mispredict Retired */
+                    blocked = 0;
+                    break;
+               }
+            }
+
+            if ( blocked )
+                return -EINVAL;
+
             if ( has_hvm_container_vcpu(v) )
                 vmx_read_guest_msr(MSR_CORE_PERF_GLOBAL_CTRL,
                                    &core2_vpmu_cxt->global_ctrl);

diff --git a/xen/include/asm-x86/msr-index.h b/xen/include/asm-x86/msr-index.h
index 2b9786169d0cd5fbf2b19f940c7df920b1ddf35e..84da631ad082bab8d9a6ec9c33a5889d3a6f85cb 100644 (file)
--- a/xen/include/asm-x86/msr-index.h
+++ b/xen/include/asm-x86/msr-index.h
@@ -329,6 +329,7 @@
 
 /* Platform Shared Resource MSRs */
 #define MSR_IA32_CMT_EVTSEL            0x00000c8d
+#define MSR_IA32_CMT_EVTSEL_UE_MASK    0x0000ffff
 #define MSR_IA32_CMT_CTR               0x00000c8e
 #define MSR_IA32_PSR_ASSOC             0x00000c8f
 #define MSR_IA32_PSR_L3_QOS_CFG        0x00000c81

diff --git a/xen/include/public/pmu.h b/xen/include/public/pmu.h
index 7753df012b66525f870d5b49873cfef8831ff0ac..0e1312cfe50d6b924ea07dad8c8aaaf9ec9bde78 100644 (file)
--- a/xen/include/public/pmu.h
+++ b/xen/include/public/pmu.h
@@ -84,9 +84,19 @@ DEFINE_XEN_GUEST_HANDLE(xen_pmu_params_t);
 
 /*
  * PMU features:
- * - XENPMU_FEATURE_INTEL_BTS: Intel BTS support (ignored on AMD)
+ * - XENPMU_FEATURE_INTEL_BTS:  Intel BTS support (ignored on AMD)
+ * - XENPMU_FEATURE_IPC_ONLY:   Restrict PMCs to the most minimum set possible.
+ *                              Instructions, cycles, and ref cycles. Can be
+ *                              used to calculate instructions-per-cycle (IPC)
+ *                              (ignored on AMD).
+ * - XENPMU_FEATURE_ARCH_ONLY:  Restrict PMCs to the Intel Pre-Defined
+ *                              Architectural Performance Events exposed by
+ *                              cpuid and listed in the Intel developer's manual
+ *                              (ignored on AMD).
  */
-#define XENPMU_FEATURE_INTEL_BTS  1
+#define XENPMU_FEATURE_INTEL_BTS  (1<<0)
+#define XENPMU_FEATURE_IPC_ONLY   (1<<1)
+#define XENPMU_FEATURE_ARCH_ONLY  (1<<2)
 
 /*
  * Shared PMU data between hypervisor and PV(H) domains.