Small x86 fixes.

-----BEGIN PGP SIGNATURE-----
 
 iQFIBAABCAAyFiEE8TM4V0tmI4mGbHaCv/vSX3jHroMFAmFXQUoUHHBib256aW5p
 QHJlZGhhdC5jb20ACgkQv/vSX3jHroMglgf/egh3zb9/+BUQWe0xWfhcINNzpsVk
 PJtiBmJc3nQLbZbTSLp63rouy1lNgR0s2DiMwP7G1u39OwW8W3LHMrBUSqF1F01+
 gntb4GGiRTiTPJI64K4z6ytORd3tuRarHq8TUIa2zvki9ZW5Obgkm1i1RsNMOo+s
 AOA7whhpS8e/a5fBbtbS9bTZb30PKTZmbW4oMjvO9Sw4Eb76IauqPSEtRPSuCAc7
 r7z62RTlm10Qk0JR3tW1iXMxTJHZk+tYPJ8pclUAWVX5bZqWa/9k8R0Z5i/miFiZ
 glW/y3R4+aUwIQV2v7V3Jx9MOKDhZxniMtnqZG/Hp9NVDtWIz37V/U37vw==
 =zQQ1
 -----END PGP SIGNATURE-----

Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull more kvm fixes from Paolo Bonzini:
 "Small x86 fixes"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: selftests: Ensure all migrations are performed when test is affined
  KVM: x86: Swap order of CPUID entry "index" vs. "significant flag" checks
  ptp: Fix ptp_kvm_getcrosststamp issue for x86 ptp_kvm
  x86/kvmclock: Move this_cpu_pvti into kvmclock.h
  selftests: KVM: Don't clobber XMM register when read
  KVM: VMX: Fix a TSX_CTRL_CPUID_CLEAR field mask issue

arch/x86/include/asm/kvmclock.h

@@ -2,6 +2,20 @@
 #ifndef _ASM_X86_KVM_CLOCK_H
 #define _ASM_X86_KVM_CLOCK_H
 
+#include <linux/percpu.h>
+
 extern struct clocksource kvm_clock;
 
+DECLARE_PER_CPU(struct pvclock_vsyscall_time_info *, hv_clock_per_cpu);
+
+static inline struct pvclock_vcpu_time_info *this_cpu_pvti(void)
+{
+	return &this_cpu_read(hv_clock_per_cpu)->pvti;
+}
+
+static inline struct pvclock_vsyscall_time_info *this_cpu_hvclock(void)
+{
+	return this_cpu_read(hv_clock_per_cpu);
+}
+
 #endif /* _ASM_X86_KVM_CLOCK_H */

arch/x86/kernel/kvmclock.c

@@ -49,18 +49,9 @@ early_param("no-kvmclock-vsyscall", parse_no_kvmclock_vsyscall);
 static struct pvclock_vsyscall_time_info
 			hv_clock_boot[HVC_BOOT_ARRAY_SIZE] __bss_decrypted __aligned(PAGE_SIZE);
 static struct pvclock_wall_clock wall_clock __bss_decrypted;
-static DEFINE_PER_CPU(struct pvclock_vsyscall_time_info *, hv_clock_per_cpu);
 static struct pvclock_vsyscall_time_info *hvclock_mem;
-
-static inline struct pvclock_vcpu_time_info *this_cpu_pvti(void)
-{
-	return &this_cpu_read(hv_clock_per_cpu)->pvti;
-}
-
-static inline struct pvclock_vsyscall_time_info *this_cpu_hvclock(void)
-{
-	return this_cpu_read(hv_clock_per_cpu);
-}
+DEFINE_PER_CPU(struct pvclock_vsyscall_time_info *, hv_clock_per_cpu);
+EXPORT_PER_CPU_SYMBOL_GPL(hv_clock_per_cpu);
 
 /*
  * The wallclock is the time of day when we booted. Since then, some time may

arch/x86/kvm/cpuid.c

@@ -65,8 +65,8 @@ static inline struct kvm_cpuid_entry2 *cpuid_entry2_find(
 	for (i = 0; i < nent; i++) {
 		e = &entries[i];
 
-		if (e->function == function && (e->index == index ||
-		    !(e->flags & KVM_CPUID_FLAG_SIGNIFCANT_INDEX)))
+		if (e->function == function &&
+		    (!(e->flags & KVM_CPUID_FLAG_SIGNIFCANT_INDEX) || e->index == index))
 			return e;
 	}
 

arch/x86/kvm/vmx/vmx.c

@@ -6848,7 +6848,7 @@ static int vmx_create_vcpu(struct kvm_vcpu *vcpu)
 		 */
 		tsx_ctrl = vmx_find_uret_msr(vmx, MSR_IA32_TSX_CTRL);
 		if (tsx_ctrl)
-			vmx->guest_uret_msrs[i].mask = ~(u64)TSX_CTRL_CPUID_CLEAR;
+			tsx_ctrl->mask = ~(u64)TSX_CTRL_CPUID_CLEAR;
 	}
 
 	err = alloc_loaded_vmcs(&vmx->vmcs01);

drivers/ptp/ptp_kvm_x86.c

@@ -15,8 +15,6 @@
 #include <linux/ptp_clock_kernel.h>
 #include <linux/ptp_kvm.h>
 
-struct pvclock_vsyscall_time_info *hv_clock;
-
 static phys_addr_t clock_pair_gpa;
 static struct kvm_clock_pairing clock_pair;
 
@@ -28,8 +26,7 @@ int kvm_arch_ptp_init(void)
 		return -ENODEV;
 
 	clock_pair_gpa = slow_virt_to_phys(&clock_pair);
-	hv_clock = pvclock_get_pvti_cpu0_va();
-	if (!hv_clock)
+	if (!pvclock_get_pvti_cpu0_va())
 		return -ENODEV;
 
 	ret = kvm_hypercall2(KVM_HC_CLOCK_PAIRING, clock_pair_gpa,
@@ -64,10 +61,8 @@ int kvm_arch_ptp_get_crosststamp(u64 *cycle, struct timespec64 *tspec,
 	struct pvclock_vcpu_time_info *src;
 	unsigned int version;
 	long ret;
-	int cpu;
 
-	cpu = smp_processor_id();
-	src = &hv_clock[cpu].pvti;
+	src = this_cpu_pvti();
 
 	do {
 		/*

tools/testing/selftests/kvm/include/x86_64/processor.h

@@ -315,7 +315,7 @@ static inline void set_xmm(int n, unsigned long val)
 #define GET_XMM(__xmm)							\
 ({									\
 	unsigned long __val;						\
-	asm volatile("movq %%"#__xmm", %0" : "=r"(__val) : : #__xmm);	\
+	asm volatile("movq %%"#__xmm", %0" : "=r"(__val));		\
 	__val;								\
 })
 

tools/testing/selftests/kvm/rseq_test.c

@@ -10,6 +10,7 @@
 #include <signal.h>
 #include <syscall.h>
 #include <sys/ioctl.h>
+#include <sys/sysinfo.h>
 #include <asm/barrier.h>
 #include <linux/atomic.h>
 #include <linux/rseq.h>
@@ -39,6 +40,7 @@ static __thread volatile struct rseq __rseq = {
 
 static pthread_t migration_thread;
 static cpu_set_t possible_mask;
+static int min_cpu, max_cpu;
 static bool done;
 
 static atomic_t seq_cnt;
@@ -57,20 +59,37 @@ static void sys_rseq(int flags)
 	TEST_ASSERT(!r, "rseq failed, errno = %d (%s)", errno, strerror(errno));
 }
 
+static int next_cpu(int cpu)
+{
+	/*
+	 * Advance to the next CPU, skipping those that weren't in the original
+	 * affinity set.  Sadly, there is no CPU_SET_FOR_EACH, and cpu_set_t's
+	 * data storage is considered as opaque.  Note, if this task is pinned
+	 * to a small set of discontigous CPUs, e.g. 2 and 1023, this loop will
+	 * burn a lot cycles and the test will take longer than normal to
+	 * complete.
+	 */
+	do {
+		cpu++;
+		if (cpu > max_cpu) {
+			cpu = min_cpu;
+			TEST_ASSERT(CPU_ISSET(cpu, &possible_mask),
+				    "Min CPU = %d must always be usable", cpu);
+			break;
+		}
+	} while (!CPU_ISSET(cpu, &possible_mask));
+
+	return cpu;
+}
+
 static void *migration_worker(void *ign)
 {
 	cpu_set_t allowed_mask;
-	int r, i, nr_cpus, cpu;
+	int r, i, cpu;
 
 	CPU_ZERO(&allowed_mask);
 
-	nr_cpus = CPU_COUNT(&possible_mask);
-
-	for (i = 0; i < NR_TASK_MIGRATIONS; i++) {
-		cpu = i % nr_cpus;
-		if (!CPU_ISSET(cpu, &possible_mask))
-			continue;
-
+	for (i = 0, cpu = min_cpu; i < NR_TASK_MIGRATIONS; i++, cpu = next_cpu(cpu)) {
 		CPU_SET(cpu, &allowed_mask);
 
 		/*
@@ -154,6 +173,36 @@ static void *migration_worker(void *ign)
 	return NULL;
 }
 
+static int calc_min_max_cpu(void)
+{
+	int i, cnt, nproc;
+
+	if (CPU_COUNT(&possible_mask) < 2)
+		return -EINVAL;
+
+	/*
+	 * CPU_SET doesn't provide a FOR_EACH helper, get the min/max CPU that
+	 * this task is affined to in order to reduce the time spent querying
+	 * unusable CPUs, e.g. if this task is pinned to a small percentage of
+	 * total CPUs.
+	 */
+	nproc = get_nprocs_conf();
+	min_cpu = -1;
+	max_cpu = -1;
+	cnt = 0;
+
+	for (i = 0; i < nproc; i++) {
+		if (!CPU_ISSET(i, &possible_mask))
+			continue;
+		if (min_cpu == -1)
+			min_cpu = i;
+		max_cpu = i;
+		cnt++;
+	}
+
+	return (cnt < 2) ? -EINVAL : 0;
+}
+
 int main(int argc, char *argv[])
 {
 	int r, i, snapshot;
@@ -167,8 +216,8 @@ int main(int argc, char *argv[])
 	TEST_ASSERT(!r, "sched_getaffinity failed, errno = %d (%s)", errno,
 		    strerror(errno));
 
-	if (CPU_COUNT(&possible_mask) < 2) {
-		print_skip("Only one CPU, task migration not possible\n");
+	if (calc_min_max_cpu()) {
+		print_skip("Only one usable CPU, task migration not possible");
 		exit(KSFT_SKIP);
 	}