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

Pull more kvm updates from Paolo Bonzini:

 - Xen timer fixes

 - Documentation formatting fixes

 - Make rseq selftest compatible with glibc-2.35

 - Fix handling of illegal LEA reg, reg

 - Cleanup creation of debugfs entries

 - Fix steal time cache handling bug

 - Fixes for MMIO caching

 - Optimize computation of number of LBRs

 - Fix uninitialized field in guest_maxphyaddr < host_maxphyaddr path

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (26 commits)
  KVM: x86/MMU: properly format KVM_CAP_VM_DISABLE_NX_HUGE_PAGES capability table
  Documentation: KVM: extend KVM_CAP_VM_DISABLE_NX_HUGE_PAGES heading underline
  KVM: VMX: Adjust number of LBR records for PERF_CAPABILITIES at refresh
  KVM: VMX: Use proper type-safe functions for vCPU => LBRs helpers
  KVM: x86: Refresh PMU after writes to MSR_IA32_PERF_CAPABILITIES
  KVM: selftests: Test all possible "invalid" PERF_CAPABILITIES.LBR_FMT vals
  KVM: selftests: Use getcpu() instead of sched_getcpu() in rseq_test
  KVM: selftests: Make rseq compatible with glibc-2.35
  KVM: Actually create debugfs in kvm_create_vm()
  KVM: Pass the name of the VM fd to kvm_create_vm_debugfs()
  KVM: Get an fd before creating the VM
  KVM: Shove vcpu stats_id init into kvm_vcpu_init()
  KVM: Shove vm stats_id init into kvm_create_vm()
  KVM: x86/mmu: Add sanity check that MMIO SPTE mask doesn't overlap gen
  KVM: x86/mmu: rename trace function name for asynchronous page fault
  KVM: x86/xen: Stop Xen timer before changing IRQ
  KVM: x86/xen: Initialize Xen timer only once
  KVM: SVM: Disable SEV-ES support if MMIO caching is disable
  KVM: x86/mmu: Fully re-evaluate MMIO caching when SPTE masks change
  KVM: x86: Tag kvm_mmu_x86_module_init() with __init
  ...

3 files changed, 45 insertions, 37 deletions

diff --git a/tools/testing/selftests/kvm/Makefile b/tools/testing/selftests/kvm/Makefile
index c7f47429d6cd..4c122f1b1737 100644
--- a/tools/testing/selftests/kvm/Makefile
+++ b/tools/testing/selftests/kvm/Makefile
@@ -4,6 +4,8 @@ include ../../../build/Build.include
 all:
 
 top_srcdir = ../../../..
+include $(top_srcdir)/scripts/subarch.include
+ARCH            ?= $(SUBARCH)
 
 # For cross-builds to work, UNAME_M has to map to ARCH and arch specific
 # directories and targets in this Makefile. "uname -m" doesn't map to
@@ -197,7 +199,8 @@ endif
 CFLAGS += -Wall -Wstrict-prototypes -Wuninitialized -O2 -g -std=gnu99 \
 	-fno-stack-protector -fno-PIE -I$(LINUX_TOOL_INCLUDE) \
 	-I$(LINUX_TOOL_ARCH_INCLUDE) -I$(LINUX_HDR_PATH) -Iinclude \
-	-I$(<D) -Iinclude/$(UNAME_M) -I.. $(EXTRA_CFLAGS) $(KHDR_INCLUDES)
+	-I$(<D) -Iinclude/$(UNAME_M) -I ../rseq -I.. $(EXTRA_CFLAGS) \
+	$(KHDR_INCLUDES)
 
 no-pie-option := $(call try-run, echo 'int main() { return 0; }' | \
         $(CC) -Werror -no-pie -x c - -o "$$TMP", -no-pie)
@@ -206,7 +209,7 @@ no-pie-option := $(call try-run, echo 'int main() { return 0; }' | \
 pgste-option = $(call try-run, echo 'int main() { return 0; }' | \
 	$(CC) -Werror -Wl$(comma)--s390-pgste -x c - -o "$$TMP",-Wl$(comma)--s390-pgste)
 
-
+LDLIBS += -ldl
 LDFLAGS += -pthread $(no-pie-option) $(pgste-option)
 
 # After inclusion, $(OUTPUT) is defined and
diff --git a/tools/testing/selftests/kvm/rseq_test.c b/tools/testing/selftests/kvm/rseq_test.c
index a54d4d05a058..fac248a43666 100644
--- a/tools/testing/selftests/kvm/rseq_test.c
+++ b/tools/testing/selftests/kvm/rseq_test.c
@@ -20,15 +20,7 @@
 #include "processor.h"
 #include "test_util.h"
 
-static __thread volatile struct rseq __rseq = {
-	.cpu_id = RSEQ_CPU_ID_UNINITIALIZED,
-};
-
-/*
- * Use an arbitrary, bogus signature for configuring rseq, this test does not
- * actually enter an rseq critical section.
- */
-#define RSEQ_SIG 0xdeadbeef
+#include "../rseq/rseq.c"
 
 /*
  * Any bug related to task migration is likely to be timing-dependent; perform
@@ -49,12 +41,16 @@ static void guest_code(void)
 		GUEST_SYNC(0);
 }
 
-static void sys_rseq(int flags)
+/*
+ * We have to perform direct system call for getcpu() because it's
+ * not available until glic 2.29.
+ */
+static void sys_getcpu(unsigned *cpu)
 {
 	int r;
 
-	r = syscall(__NR_rseq, &__rseq, sizeof(__rseq), flags, RSEQ_SIG);
-	TEST_ASSERT(!r, "rseq failed, errno = %d (%s)", errno, strerror(errno));
+	r = syscall(__NR_getcpu, cpu, NULL, NULL);
+	TEST_ASSERT(!r, "getcpu failed, errno = %d (%s)", errno, strerror(errno));
 }
 
 static int next_cpu(int cpu)
@@ -101,7 +97,7 @@ static void *migration_worker(void *__rseq_tid)
 		atomic_inc(&seq_cnt);
 
 		/*
-		 * Ensure the odd count is visible while sched_getcpu() isn't
+		 * Ensure the odd count is visible while getcpu() isn't
 		 * stable, i.e. while changing affinity is in-progress.
 		 */
 		smp_wmb();
@@ -142,10 +138,10 @@ static void *migration_worker(void *__rseq_tid)
 		 *     check completes.
 		 *
 		 *  3. To ensure the read-side makes efficient forward progress,
-		 *     e.g. if sched_getcpu() involves a syscall.  Stalling the
-		 *     read-side means the test will spend more time waiting for
-		 *     sched_getcpu() to stabilize and less time trying to hit
-		 *     the timing-dependent bug.
+		 *     e.g. if getcpu() involves a syscall. Stalling the read-side
+		 *     means the test will spend more time waiting for getcpu()
+		 *     to stabilize and less time trying to hit the timing-dependent
+		 *     bug.
 		 *
 		 * Because any bug in this area is likely to be timing-dependent,
 		 * run with a range of delays at 1us intervals from 1us to 10us
@@ -218,7 +214,9 @@ int main(int argc, char *argv[])
 
 	calc_min_max_cpu();
 
-	sys_rseq(0);
+	r = rseq_register_current_thread();
+	TEST_ASSERT(!r, "rseq_register_current_thread failed, errno = %d (%s)",
+		    errno, strerror(errno));
 
 	/*
 	 * Create and run a dummy VM that immediately exits to userspace via
@@ -238,9 +236,9 @@ int main(int argc, char *argv[])
 
 		/*
 		 * Verify rseq's CPU matches sched's CPU.  Ensure migration
-		 * doesn't occur between sched_getcpu() and reading the rseq
-		 * cpu_id by rereading both if the sequence count changes, or
-		 * if the count is odd (migration in-progress).
+		 * doesn't occur between getcpu() and reading the rseq cpu_id
+		 * by rereading both if the sequence count changes, or if the
+		 * count is odd (migration in-progress).
 		 */
 		do {
 			/*
@@ -250,13 +248,13 @@ int main(int argc, char *argv[])
 			snapshot = atomic_read(&seq_cnt) & ~1;
 
 			/*
-			 * Ensure reading sched_getcpu() and rseq.cpu_id
-			 * complete in a single "no migration" window, i.e. are
-			 * not reordered across the seq_cnt reads.
+			 * Ensure calling getcpu() and reading rseq.cpu_id complete
+			 * in a single "no migration" window, i.e. are not reordered
+			 * across the seq_cnt reads.
 			 */
 			smp_rmb();
-			cpu = sched_getcpu();
-			rseq_cpu = READ_ONCE(__rseq.cpu_id);
+			sys_getcpu(&cpu);
+			rseq_cpu = rseq_current_cpu_raw();
 			smp_rmb();
 		} while (snapshot != atomic_read(&seq_cnt));
 
@@ -267,9 +265,9 @@ int main(int argc, char *argv[])
 	/*
 	 * Sanity check that the test was able to enter the guest a reasonable
 	 * number of times, e.g. didn't get stalled too often/long waiting for
-	 * sched_getcpu() to stabilize.  A 2:1 migration:KVM_RUN ratio is a
-	 * fairly conservative ratio on x86-64, which can do _more_ KVM_RUNs
-	 * than migrations given the 1us+ delay in the migration task.
+	 * getcpu() to stabilize.  A 2:1 migration:KVM_RUN ratio is a fairly
+	 * conservative ratio on x86-64, which can do _more_ KVM_RUNs than
+	 * migrations given the 1us+ delay in the migration task.
 	 */
 	TEST_ASSERT(i > (NR_TASK_MIGRATIONS / 2),
 		    "Only performed %d KVM_RUNs, task stalled too much?\n", i);
@@ -278,7 +276,7 @@ int main(int argc, char *argv[])
 
 	kvm_vm_free(vm);
 
-	sys_rseq(RSEQ_FLAG_UNREGISTER);
+	rseq_unregister_current_thread();
 
 	return 0;
 }
diff --git a/tools/testing/selftests/kvm/x86_64/vmx_pmu_caps_test.c b/tools/testing/selftests/kvm/x86_64/vmx_pmu_caps_test.c
index 6ec901dab61e..069589c52f41 100644
--- a/tools/testing/selftests/kvm/x86_64/vmx_pmu_caps_test.c
+++ b/tools/testing/selftests/kvm/x86_64/vmx_pmu_caps_test.c
@@ -59,6 +59,7 @@ int main(int argc, char *argv[])
 	int ret;
 	union cpuid10_eax eax;
 	union perf_capabilities host_cap;
+	uint64_t val;
 
 	host_cap.capabilities = kvm_get_feature_msr(MSR_IA32_PERF_CAPABILITIES);
 	host_cap.capabilities &= (PMU_CAP_FW_WRITES | PMU_CAP_LBR_FMT);
@@ -91,11 +92,17 @@ int main(int argc, char *argv[])
 	vcpu_set_msr(vcpu, MSR_IA32_PERF_CAPABILITIES, host_cap.lbr_format);
 	ASSERT_EQ(vcpu_get_msr(vcpu, MSR_IA32_PERF_CAPABILITIES), (u64)host_cap.lbr_format);
 
-	/* testcase 3, check invalid LBR format is rejected */
-	/* Note, on Arch LBR capable platforms, LBR_FMT in perf capability msr is 0x3f,
-	 * to avoid the failure, use a true invalid format 0x30 for the test. */
-	ret = _vcpu_set_msr(vcpu, MSR_IA32_PERF_CAPABILITIES, 0x30);
-	TEST_ASSERT(ret == 0, "Bad PERF_CAPABILITIES didn't fail.");
+	/*
+	 * Testcase 3, check that an "invalid" LBR format is rejected.  Only an
+	 * exact match of the host's format (and 0/disabled) is allowed.
+	 */
+	for (val = 1; val <= PMU_CAP_LBR_FMT; val++) {
+		if (val == (host_cap.capabilities & PMU_CAP_LBR_FMT))
+			continue;
+
+		ret = _vcpu_set_msr(vcpu, MSR_IA32_PERF_CAPABILITIES, val);
+		TEST_ASSERT(!ret, "Bad LBR FMT = 0x%lx didn't fail", val);
+	}
 
 	printf("Completed perf capability tests.\n");
 	kvm_vm_free(vm);