iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

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

Browse Source 
Pull kvm fixes from Paolo Bonzini:
 "x86:
   - Fixes for the new scalable MMU
   - Fixes for migration of nested hypervisors on AMD
   - Fix for clang integrated assembler
   - Fix for left shift by 64 (UBSAN)
   - Small cleanups
   - Straggler SEV-ES patch

  ARM:
   - VM init cleanups
   - PSCI relay cleanups
   - Kill CONFIG_KVM_ARM_PMU
   - Fixup __init annotations
   - Fixup reg_to_encoding()
   - Fix spurious PMCR_EL0 access

  Misc:
   - selftests cleanups"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (38 commits)
  KVM: x86: __kvm_vcpu_halt can be static
  KVM: SVM: Add support for booting APs in an SEV-ES guest
  KVM: nSVM: cancel KVM_REQ_GET_NESTED_STATE_PAGES on nested vmexit
  KVM: nSVM: mark vmcb as dirty when forcingly leaving the guest mode
  KVM: nSVM: correctly restore nested_run_pending on migration
  KVM: x86/mmu: Clarify TDP MMU page list invariants
  KVM: x86/mmu: Ensure TDP MMU roots are freed after yield
  kvm: check tlbs_dirty directly
  KVM: x86: change in pv_eoi_get_pending() to make code more readable
  MAINTAINERS: Really update email address for Sean Christopherson
  KVM: x86: fix shift out of bounds reported by UBSAN
  KVM: selftests: Implement perf_test_util more conventionally
  KVM: selftests: Use vm_create_with_vcpus in create_vm
  KVM: selftests: Factor out guest mode code
  KVM/SVM: Remove leftover __svm_vcpu_run prototype from svm.c
  KVM: SVM: Add register operand to vmsave call in sev_es_vcpu_load
  KVM: x86/mmu: Optimize not-present/MMIO SPTE check in get_mmio_spte()
  KVM: x86/mmu: Use raw level to index into MMIO walks' sptes array
  KVM: x86/mmu: Get root level from walkers when retrieving MMIO SPTE
  KVM: x86/mmu: Use -1 to flag an undefined spte in get_mmio_spte()
  ...
This commit is contained in:
Linus Torvalds
2021-01-08 15:06:02 -08:00
parent 3e2a590acb 872f36eb0b
commit 2a190b22aa
45 changed files with 666 additions and 654 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tools/testing/selftests/kvm/Makefile
View File

@ -33,7 +33,7 @@ ifeq ($(ARCH),s390)
UNAME_M := s390x
endif
LIBKVM = lib/assert.c lib/elf.c lib/io.c lib/kvm_util.c lib/sparsebit.c lib/test_util.c
LIBKVM = lib/assert.c lib/elf.c lib/io.c lib/kvm_util.c lib/sparsebit.c lib/test_util.c lib/guest_modes.c lib/perf_test_util.c
LIBKVM_x86_64 = lib/x86_64/processor.c lib/x86_64/vmx.c lib/x86_64/svm.c lib/x86_64/ucall.c lib/x86_64/handlers.S
LIBKVM_aarch64 = lib/aarch64/processor.c lib/aarch64/ucall.c
LIBKVM_s390x = lib/s390x/processor.c lib/s390x/ucall.c lib/s390x/diag318_test_handler.c

116
tools/testing/selftests/kvm/demand_paging_test.c
View File

@ -7,23 +7,20 @@
* Copyright (C) 2019, Google, Inc.
*/
#define _GNU_SOURCE /* for program_invocation_name */
#define _GNU_SOURCE /* for pipe2 */
#include <stdio.h>
#include <stdlib.h>
#include <sys/syscall.h>
#include <unistd.h>
#include <asm/unistd.h>
#include <time.h>
#include <poll.h>
#include <pthread.h>
#include <linux/bitmap.h>
#include <linux/bitops.h>
#include <linux/userfaultfd.h>
#include <sys/syscall.h>
#include "perf_test_util.h"
#include "processor.h"
#include "kvm_util.h"
#include "test_util.h"
#include "perf_test_util.h"
#include "guest_modes.h"
#ifdef __NR_userfaultfd
@ -39,12 +36,14 @@
#define PER_VCPU_DEBUG(...) _no_printf(__VA_ARGS__)
#endif
static int nr_vcpus = 1;
static uint64_t guest_percpu_mem_size = DEFAULT_PER_VCPU_MEM_SIZE;
static char *guest_data_prototype;
static void *vcpu_worker(void *data)
{
int ret;
struct vcpu_args *vcpu_args = (struct vcpu_args *)data;
struct perf_test_vcpu_args *vcpu_args = (struct perf_test_vcpu_args *)data;
int vcpu_id = vcpu_args->vcpu_id;
struct kvm_vm *vm = perf_test_args.vm;
struct kvm_run *run;
@ -248,9 +247,14 @@ static int setup_demand_paging(struct kvm_vm *vm,
return 0;
}
static void run_test(enum vm_guest_mode mode, bool use_uffd,
useconds_t uffd_delay)
struct test_params {
bool use_uffd;
useconds_t uffd_delay;
};
static void run_test(enum vm_guest_mode mode, void *arg)
{
struct test_params *p = arg;
pthread_t *vcpu_threads;
pthread_t *uffd_handler_threads = NULL;
struct uffd_handler_args *uffd_args = NULL;
@ -261,7 +265,7 @@ static void run_test(enum vm_guest_mode mode, bool use_uffd,
int vcpu_id;
int r;
vm = create_vm(mode, nr_vcpus, guest_percpu_mem_size);
vm = perf_test_create_vm(mode, nr_vcpus, guest_percpu_mem_size);
perf_test_args.wr_fract = 1;
@ -273,9 +277,9 @@ static void run_test(enum vm_guest_mode mode, bool use_uffd,
vcpu_threads = malloc(nr_vcpus * sizeof(*vcpu_threads));
TEST_ASSERT(vcpu_threads, "Memory allocation failed");
add_vcpus(vm, nr_vcpus, guest_percpu_mem_size);
perf_test_setup_vcpus(vm, nr_vcpus, guest_percpu_mem_size);
if (use_uffd) {
if (p->use_uffd) {
uffd_handler_threads =
malloc(nr_vcpus * sizeof(*uffd_handler_threads));
TEST_ASSERT(uffd_handler_threads, "Memory allocation failed");
@ -308,7 +312,7 @@ static void run_test(enum vm_guest_mode mode, bool use_uffd,
r = setup_demand_paging(vm,
&uffd_handler_threads[vcpu_id],
pipefds[vcpu_id * 2],
uffd_delay, &uffd_args[vcpu_id],
p->uffd_delay, &uffd_args[vcpu_id],
vcpu_hva, guest_percpu_mem_size);
if (r < 0)
exit(-r);
@ -339,7 +343,7 @@ static void run_test(enum vm_guest_mode mode, bool use_uffd,
pr_info("All vCPU threads joined\n");
if (use_uffd) {
if (p->use_uffd) {
char c;
/* Tell the user fault fd handler threads to quit */
@ -357,43 +361,23 @@ static void run_test(enum vm_guest_mode mode, bool use_uffd,
perf_test_args.vcpu_args[0].pages * nr_vcpus /
((double)ts_diff.tv_sec + (double)ts_diff.tv_nsec / 100000000.0));
ucall_uninit(vm);
kvm_vm_free(vm);
perf_test_destroy_vm(vm);
free(guest_data_prototype);
free(vcpu_threads);
if (use_uffd) {
if (p->use_uffd) {
free(uffd_handler_threads);
free(uffd_args);
free(pipefds);
}
}
struct guest_mode {
bool supported;
bool enabled;
};
static struct guest_mode guest_modes[NUM_VM_MODES];
#define guest_mode_init(mode, supported, enabled) ({ \
guest_modes[mode] = (struct guest_mode){ supported, enabled }; \
})
static void help(char *name)
{
int i;
puts("");
printf("usage: %s [-h] [-m mode] [-u] [-d uffd_delay_usec]\n"
" [-b memory] [-v vcpus]\n", name);
printf(" -m: specify the guest mode ID to test\n"
" (default: test all supported modes)\n"
" This option may be used multiple times.\n"
" Guest mode IDs:\n");
for (i = 0; i < NUM_VM_MODES; ++i) {
printf(" %d: %s%s\n", i, vm_guest_mode_string(i),
guest_modes[i].supported ? " (supported)" : "");
}
guest_modes_help();
printf(" -u: use User Fault FD to handle vCPU page\n"
" faults.\n");
printf(" -d: add a delay in usec to the User Fault\n"
@ -410,53 +394,22 @@ static void help(char *name)
int main(int argc, char *argv[])
{
int max_vcpus = kvm_check_cap(KVM_CAP_MAX_VCPUS);
bool mode_selected = false;
unsigned int mode;
int opt, i;
bool use_uffd = false;
useconds_t uffd_delay = 0;
struct test_params p = {};
int opt;
#ifdef __x86_64__
guest_mode_init(VM_MODE_PXXV48_4K, true, true);
#endif
#ifdef __aarch64__
guest_mode_init(VM_MODE_P40V48_4K, true, true);
guest_mode_init(VM_MODE_P40V48_64K, true, true);
{
unsigned int limit = kvm_check_cap(KVM_CAP_ARM_VM_IPA_SIZE);
if (limit >= 52)
guest_mode_init(VM_MODE_P52V48_64K, true, true);
if (limit >= 48) {
guest_mode_init(VM_MODE_P48V48_4K, true, true);
guest_mode_init(VM_MODE_P48V48_64K, true, true);
}
}
#endif
#ifdef __s390x__
guest_mode_init(VM_MODE_P40V48_4K, true, true);
#endif
guest_modes_append_default();
while ((opt = getopt(argc, argv, "hm:ud:b:v:")) != -1) {
switch (opt) {
case 'm':
if (!mode_selected) {
for (i = 0; i < NUM_VM_MODES; ++i)
guest_modes[i].enabled = false;
mode_selected = true;
}
mode = strtoul(optarg, NULL, 10);
TEST_ASSERT(mode < NUM_VM_MODES,
"Guest mode ID %d too big", mode);
guest_modes[mode].enabled = true;
guest_modes_cmdline(optarg);
break;
case 'u':
use_uffd = true;
p.use_uffd = true;
break;
case 'd':
uffd_delay = strtoul(optarg, NULL, 0);
TEST_ASSERT(uffd_delay >= 0,
"A negative UFFD delay is not supported.");
p.uffd_delay = strtoul(optarg, NULL, 0);
TEST_ASSERT(p.uffd_delay >= 0, "A negative UFFD delay is not supported.");
break;
case 'b':
guest_percpu_mem_size = parse_size(optarg);
@ -473,14 +426,7 @@ int main(int argc, char *argv[])
}
}
for (i = 0; i < NUM_VM_MODES; ++i) {
if (!guest_modes[i].enabled)
continue;
TEST_ASSERT(guest_modes[i].supported,
"Guest mode ID %d (%s) not supported.",
i, vm_guest_mode_string(i));
run_test(i, use_uffd, uffd_delay);
}
for_each_guest_mode(run_test, &p);
return 0;
}

145
tools/testing/selftests/kvm/dirty_log_perf_test.c
View File

@ -8,29 +8,28 @@
* Copyright (C) 2020, Google, Inc.
*/
#define _GNU_SOURCE /* for program_invocation_name */
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <time.h>
#include <pthread.h>
#include <linux/bitmap.h>
#include <linux/bitops.h>
#include "kvm_util.h"
#include "perf_test_util.h"
#include "processor.h"
#include "test_util.h"
#include "perf_test_util.h"
#include "guest_modes.h"
/* How many host loops to run by default (one KVM_GET_DIRTY_LOG for each loop)*/
#define TEST_HOST_LOOP_N 2UL
static int nr_vcpus = 1;
static uint64_t guest_percpu_mem_size = DEFAULT_PER_VCPU_MEM_SIZE;
/* Host variables */
static u64 dirty_log_manual_caps;
static bool host_quit;
static uint64_t iteration;
static uint64_t vcpu_last_completed_iteration[MAX_VCPUS];
static uint64_t vcpu_last_completed_iteration[KVM_MAX_VCPUS];
static void *vcpu_worker(void *data)
{
@ -42,7 +41,7 @@ static void *vcpu_worker(void *data)
struct timespec ts_diff;
struct timespec total = (struct timespec){0};
struct timespec avg;
struct vcpu_args *vcpu_args = (struct vcpu_args *)data;
struct perf_test_vcpu_args *vcpu_args = (struct perf_test_vcpu_args *)data;
int vcpu_id = vcpu_args->vcpu_id;
vcpu_args_set(vm, vcpu_id, 1, vcpu_id);
@ -89,9 +88,15 @@ static void *vcpu_worker(void *data)
return NULL;
}
static void run_test(enum vm_guest_mode mode, unsigned long iterations,
uint64_t phys_offset, int wr_fract)
struct test_params {
unsigned long iterations;
uint64_t phys_offset;
int wr_fract;
};
static void run_test(enum vm_guest_mode mode, void *arg)
{
struct test_params *p = arg;
pthread_t *vcpu_threads;
struct kvm_vm *vm;
unsigned long *bmap;
@ -106,9 +111,9 @@ static void run_test(enum vm_guest_mode mode, unsigned long iterations,
struct kvm_enable_cap cap = {};
struct timespec clear_dirty_log_total = (struct timespec){0};
vm = create_vm(mode, nr_vcpus, guest_percpu_mem_size);
vm = perf_test_create_vm(mode, nr_vcpus, guest_percpu_mem_size);
perf_test_args.wr_fract = wr_fract;
perf_test_args.wr_fract = p->wr_fract;
guest_num_pages = (nr_vcpus * guest_percpu_mem_size) >> vm_get_page_shift(vm);
guest_num_pages = vm_adjust_num_guest_pages(mode, guest_num_pages);
@ -124,7 +129,7 @@ static void run_test(enum vm_guest_mode mode, unsigned long iterations,
vcpu_threads = malloc(nr_vcpus * sizeof(*vcpu_threads));
TEST_ASSERT(vcpu_threads, "Memory allocation failed");
add_vcpus(vm, nr_vcpus, guest_percpu_mem_size);
perf_test_setup_vcpus(vm, nr_vcpus, guest_percpu_mem_size);
sync_global_to_guest(vm, perf_test_args);
@ -150,13 +155,13 @@ static void run_test(enum vm_guest_mode mode, unsigned long iterations,
/* Enable dirty logging */
clock_gettime(CLOCK_MONOTONIC, &start);
vm_mem_region_set_flags(vm, TEST_MEM_SLOT_INDEX,
vm_mem_region_set_flags(vm, PERF_TEST_MEM_SLOT_INDEX,
KVM_MEM_LOG_DIRTY_PAGES);
ts_diff = timespec_diff_now(start);
pr_info("Enabling dirty logging time: %ld.%.9lds\n\n",
ts_diff.tv_sec, ts_diff.tv_nsec);
while (iteration < iterations) {
while (iteration < p->iterations) {
/*
* Incrementing the iteration number will start the vCPUs
* dirtying memory again.
@ -177,7 +182,7 @@ static void run_test(enum vm_guest_mode mode, unsigned long iterations,
iteration, ts_diff.tv_sec, ts_diff.tv_nsec);
clock_gettime(CLOCK_MONOTONIC, &start);
kvm_vm_get_dirty_log(vm, TEST_MEM_SLOT_INDEX, bmap);
kvm_vm_get_dirty_log(vm, PERF_TEST_MEM_SLOT_INDEX, bmap);
ts_diff = timespec_diff_now(start);
get_dirty_log_total = timespec_add(get_dirty_log_total,
@ -187,7 +192,7 @@ static void run_test(enum vm_guest_mode mode, unsigned long iterations,
if (dirty_log_manual_caps) {
clock_gettime(CLOCK_MONOTONIC, &start);
kvm_vm_clear_dirty_log(vm, TEST_MEM_SLOT_INDEX, bmap, 0,
kvm_vm_clear_dirty_log(vm, PERF_TEST_MEM_SLOT_INDEX, bmap, 0,
host_num_pages);
ts_diff = timespec_diff_now(start);
@ -205,43 +210,30 @@ static void run_test(enum vm_guest_mode mode, unsigned long iterations,
/* Disable dirty logging */
clock_gettime(CLOCK_MONOTONIC, &start);
vm_mem_region_set_flags(vm, TEST_MEM_SLOT_INDEX, 0);
vm_mem_region_set_flags(vm, PERF_TEST_MEM_SLOT_INDEX, 0);
ts_diff = timespec_diff_now(start);
pr_info("Disabling dirty logging time: %ld.%.9lds\n",
ts_diff.tv_sec, ts_diff.tv_nsec);
avg = timespec_div(get_dirty_log_total, iterations);
avg = timespec_div(get_dirty_log_total, p->iterations);
pr_info("Get dirty log over %lu iterations took %ld.%.9lds. (Avg %ld.%.9lds/iteration)\n",
iterations, get_dirty_log_total.tv_sec,
p->iterations, get_dirty_log_total.tv_sec,
get_dirty_log_total.tv_nsec, avg.tv_sec, avg.tv_nsec);
if (dirty_log_manual_caps) {
avg = timespec_div(clear_dirty_log_total, iterations);
avg = timespec_div(clear_dirty_log_total, p->iterations);
pr_info("Clear dirty log over %lu iterations took %ld.%.9lds. (Avg %ld.%.9lds/iteration)\n",
iterations, clear_dirty_log_total.tv_sec,
p->iterations, clear_dirty_log_total.tv_sec,
clear_dirty_log_total.tv_nsec, avg.tv_sec, avg.tv_nsec);
}
free(bmap);
free(vcpu_threads);
ucall_uninit(vm);
kvm_vm_free(vm);
perf_test_destroy_vm(vm);
}
struct guest_mode {
bool supported;
bool enabled;
};
static struct guest_mode guest_modes[NUM_VM_MODES];
#define guest_mode_init(mode, supported, enabled) ({ \
guest_modes[mode] = (struct guest_mode){ supported, enabled }; \
})
static void help(char *name)
{
int i;
puts("");
printf("usage: %s [-h] [-i iterations] [-p offset] "
"[-m mode] [-b vcpu bytes] [-v vcpus]\n", name);
@ -250,14 +242,7 @@ static void help(char *name)
TEST_HOST_LOOP_N);
printf(" -p: specify guest physical test memory offset\n"
" Warning: a low offset can conflict with the loaded test code.\n");
printf(" -m: specify the guest mode ID to test "
"(default: test all supported modes)\n"
" This option may be used multiple times.\n"
" Guest mode IDs:\n");
for (i = 0; i < NUM_VM_MODES; ++i) {
printf(" %d: %s%s\n", i, vm_guest_mode_string(i),
guest_modes[i].supported ? " (supported)" : "");
}
guest_modes_help();
printf(" -b: specify the size of the memory region which should be\n"
" dirtied by each vCPU. e.g. 10M or 3G.\n"
" (default: 1G)\n");
@ -272,74 +257,43 @@ static void help(char *name)
int main(int argc, char *argv[])
{
unsigned long iterations = TEST_HOST_LOOP_N;
bool mode_selected = false;
uint64_t phys_offset = 0;
unsigned int mode;
int opt, i;
int wr_fract = 1;
int max_vcpus = kvm_check_cap(KVM_CAP_MAX_VCPUS);
struct test_params p = {
.iterations = TEST_HOST_LOOP_N,
.wr_fract = 1,
};
int opt;
dirty_log_manual_caps =
kvm_check_cap(KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2);
dirty_log_manual_caps &= (KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE |
KVM_DIRTY_LOG_INITIALLY_SET);
#ifdef __x86_64__
guest_mode_init(VM_MODE_PXXV48_4K, true, true);
#endif
#ifdef __aarch64__
guest_mode_init(VM_MODE_P40V48_4K, true, true);
guest_mode_init(VM_MODE_P40V48_64K, true, true);
{
unsigned int limit = kvm_check_cap(KVM_CAP_ARM_VM_IPA_SIZE);
if (limit >= 52)
guest_mode_init(VM_MODE_P52V48_64K, true, true);
if (limit >= 48) {
guest_mode_init(VM_MODE_P48V48_4K, true, true);
guest_mode_init(VM_MODE_P48V48_64K, true, true);
}
}
#endif
#ifdef __s390x__
guest_mode_init(VM_MODE_P40V48_4K, true, true);
#endif
guest_modes_append_default();
while ((opt = getopt(argc, argv, "hi:p:m:b:f:v:")) != -1) {
switch (opt) {
case 'i':
iterations = strtol(optarg, NULL, 10);
p.iterations = strtol(optarg, NULL, 10);
break;
case 'p':
phys_offset = strtoull(optarg, NULL, 0);
p.phys_offset = strtoull(optarg, NULL, 0);
break;
case 'm':
if (!mode_selected) {
for (i = 0; i < NUM_VM_MODES; ++i)
guest_modes[i].enabled = false;
mode_selected = true;
}
mode = strtoul(optarg, NULL, 10);
TEST_ASSERT(mode < NUM_VM_MODES,
"Guest mode ID %d too big", mode);
guest_modes[mode].enabled = true;
guest_modes_cmdline(optarg);
break;
case 'b':
guest_percpu_mem_size = parse_size(optarg);
break;
case 'f':
wr_fract = atoi(optarg);
TEST_ASSERT(wr_fract >= 1,
p.wr_fract = atoi(optarg);
TEST_ASSERT(p.wr_fract >= 1,
"Write fraction cannot be less than one");
break;
case 'v':
nr_vcpus = atoi(optarg);
TEST_ASSERT(nr_vcpus > 0,
"Must have a positive number of vCPUs");
TEST_ASSERT(nr_vcpus <= MAX_VCPUS,
"This test does not currently support\n"
"more than %d vCPUs.", MAX_VCPUS);
TEST_ASSERT(nr_vcpus > 0 && nr_vcpus <= max_vcpus,
"Invalid number of vcpus, must be between 1 and %d", max_vcpus);
break;
case 'h':
default:
@ -348,18 +302,11 @@ int main(int argc, char *argv[])
}
}
TEST_ASSERT(iterations >= 2, "The test should have at least two iterations");
TEST_ASSERT(p.iterations >= 2, "The test should have at least two iterations");
pr_info("Test iterations: %"PRIu64"\n", iterations);
pr_info("Test iterations: %"PRIu64"\n", p.iterations);
for (i = 0; i < NUM_VM_MODES; ++i) {
if (!guest_modes[i].enabled)
continue;
TEST_ASSERT(guest_modes[i].supported,
"Guest mode ID %d (%s) not supported.",
i, vm_guest_mode_string(i));
run_test(i, iterations, phys_offset, wr_fract);
}
for_each_guest_mode(run_test, &p);
return 0;
}

125
tools/testing/selftests/kvm/dirty_log_test.c
View File

@ -9,8 +9,6 @@
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <time.h>
#include <pthread.h>
#include <semaphore.h>
#include <sys/types.h>
@ -20,8 +18,9 @@
#include <linux/bitops.h>
#include <asm/barrier.h>
#include "test_util.h"
#include "kvm_util.h"
#include "test_util.h"
#include "guest_modes.h"
#include "processor.h"
#define VCPU_ID 1
@ -673,9 +672,15 @@ static struct kvm_vm *create_vm(enum vm_guest_mode mode, uint32_t vcpuid,
#define DIRTY_MEM_BITS 30 /* 1G */
#define PAGE_SHIFT_4K 12
static void run_test(enum vm_guest_mode mode, unsigned long iterations,
unsigned long interval, uint64_t phys_offset)
struct test_params {
unsigned long iterations;
unsigned long interval;
uint64_t phys_offset;
};
static void run_test(enum vm_guest_mode mode, void *arg)
{
struct test_params *p = arg;
struct kvm_vm *vm;
unsigned long *bmap;
@ -709,12 +714,12 @@ static void run_test(enum vm_guest_mode mode, unsigned long iterations,
host_page_size = getpagesize();
host_num_pages = vm_num_host_pages(mode, guest_num_pages);
if (!phys_offset) {
if (!p->phys_offset) {
guest_test_phys_mem = (vm_get_max_gfn(vm) -
guest_num_pages) * guest_page_size;
guest_test_phys_mem &= ~(host_page_size - 1);
} else {
guest_test_phys_mem = phys_offset;
guest_test_phys_mem = p->phys_offset;
}
#ifdef __s390x__
@ -758,9 +763,9 @@ static void run_test(enum vm_guest_mode mode, unsigned long iterations,
pthread_create(&vcpu_thread, NULL, vcpu_worker, vm);
while (iteration < iterations) {
while (iteration < p->iterations) {
/* Give the vcpu thread some time to dirty some pages */
usleep(interval * 1000);
usleep(p->interval * 1000);
log_mode_collect_dirty_pages(vm, TEST_MEM_SLOT_INDEX,
bmap, host_num_pages);
vm_dirty_log_verify(mode, bmap);
@ -783,20 +788,8 @@ static void run_test(enum vm_guest_mode mode, unsigned long iterations,
kvm_vm_free(vm);
}
struct guest_mode {
bool supported;
bool enabled;
};
static struct guest_mode guest_modes[NUM_VM_MODES];
#define guest_mode_init(mode, supported, enabled) ({ \
guest_modes[mode] = (struct guest_mode){ supported, enabled }; \
})
static void help(char *name)
{
int i;
puts("");
printf("usage: %s [-h] [-i iterations] [-I interval] "
"[-p offset] [-m mode]\n", name);
@ -813,51 +806,23 @@ static void help(char *name)
printf(" -M: specify the host logging mode "
"(default: run all log modes). Supported modes: \n\t");
log_modes_dump();
printf(" -m: specify the guest mode ID to test "
"(default: test all supported modes)\n"
" This option may be used multiple times.\n"
" Guest mode IDs:\n");
for (i = 0; i < NUM_VM_MODES; ++i) {
printf(" %d: %s%s\n", i, vm_guest_mode_string(i),
guest_modes[i].supported ? " (supported)" : "");
}
guest_modes_help();
puts("");
exit(0);
}
int main(int argc, char *argv[])
{
unsigned long iterations = TEST_HOST_LOOP_N;
unsigned long interval = TEST_HOST_LOOP_INTERVAL;
bool mode_selected = false;
uint64_t phys_offset = 0;
unsigned int mode;
int opt, i, j;
struct test_params p = {
.iterations = TEST_HOST_LOOP_N,
.interval = TEST_HOST_LOOP_INTERVAL,
};
int opt, i;
sem_init(&dirty_ring_vcpu_stop, 0, 0);
sem_init(&dirty_ring_vcpu_cont, 0, 0);
#ifdef __x86_64__
guest_mode_init(VM_MODE_PXXV48_4K, true, true);
#endif
#ifdef __aarch64__
guest_mode_init(VM_MODE_P40V48_4K, true, true);
guest_mode_init(VM_MODE_P40V48_64K, true, true);
{
unsigned int limit = kvm_check_cap(KVM_CAP_ARM_VM_IPA_SIZE);
if (limit >= 52)
guest_mode_init(VM_MODE_P52V48_64K, true, true);
if (limit >= 48) {
guest_mode_init(VM_MODE_P48V48_4K, true, true);
guest_mode_init(VM_MODE_P48V48_64K, true, true);
}
}
#endif
#ifdef __s390x__
guest_mode_init(VM_MODE_P40V48_4K, true, true);
#endif
guest_modes_append_default();
while ((opt = getopt(argc, argv, "c:hi:I:p:m:M:")) != -1) {
switch (opt) {
@ -865,24 +830,16 @@ int main(int argc, char *argv[])
test_dirty_ring_count = strtol(optarg, NULL, 10);
break;
case 'i':
iterations = strtol(optarg, NULL, 10);
p.iterations = strtol(optarg, NULL, 10);
break;
case 'I':
interval = strtol(optarg, NULL, 10);
p.interval = strtol(optarg, NULL, 10);
break;
case 'p':
phys_offset = strtoull(optarg, NULL, 0);
p.phys_offset = strtoull(optarg, NULL, 0);
break;
case 'm':
if (!mode_selected) {
for (i = 0; i < NUM_VM_MODES; ++i)
guest_modes[i].enabled = false;
mode_selected = true;
}
mode = strtoul(optarg, NULL, 10);
TEST_ASSERT(mode < NUM_VM_MODES,
"Guest mode ID %d too big", mode);
guest_modes[mode].enabled = true;
guest_modes_cmdline(optarg);
break;
case 'M':
if (!strcmp(optarg, "all")) {
@ -911,32 +868,24 @@ int main(int argc, char *argv[])
}
}
TEST_ASSERT(iterations > 2, "Iterations must be greater than two");
TEST_ASSERT(interval > 0, "Interval must be greater than zero");
TEST_ASSERT(p.iterations > 2, "Iterations must be greater than two");
TEST_ASSERT(p.interval > 0, "Interval must be greater than zero");
pr_info("Test iterations: %"PRIu64", interval: %"PRIu64" (ms)\n",
iterations, interval);
p.iterations, p.interval);
srandom(time(0));
for (i = 0; i < NUM_VM_MODES; ++i) {
if (!guest_modes[i].enabled)
continue;
TEST_ASSERT(guest_modes[i].supported,
"Guest mode ID %d (%s) not supported.",
i, vm_guest_mode_string(i));
if (host_log_mode_option == LOG_MODE_ALL) {
/* Run each log mode */
for (j = 0; j < LOG_MODE_NUM; j++) {
pr_info("Testing Log Mode '%s'\n",
log_modes[j].name);
host_log_mode = j;
run_test(i, iterations, interval, phys_offset);
}
} else {
host_log_mode = host_log_mode_option;
run_test(i, iterations, interval, phys_offset);
if (host_log_mode_option == LOG_MODE_ALL) {
/* Run each log mode */
for (i = 0; i < LOG_MODE_NUM; i++) {
pr_info("Testing Log Mode '%s'\n", log_modes[i].name);
host_log_mode = i;
for_each_guest_mode(run_test, &p);
}
} else {
host_log_mode = host_log_mode_option;
for_each_guest_mode(run_test, &p);
}
return 0;

21
tools/testing/selftests/kvm/include/guest_modes.h Normal file
View File

@ -0,0 +1,21 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Copyright (C) 2020, Red Hat, Inc.
*/
#include "kvm_util.h"
struct guest_mode {
bool supported;
bool enabled;
};
extern struct guest_mode guest_modes[NUM_VM_MODES];
#define guest_mode_append(mode, supported, enabled) ({ \
guest_modes[mode] = (struct guest_mode){ supported, enabled }; \
})
void guest_modes_append_default(void);
void for_each_guest_mode(void (*func)(enum vm_guest_mode, void *), void *arg);
void guest_modes_help(void);
void guest_modes_cmdline(const char *arg);

9
tools/testing/selftests/kvm/include/kvm_util.h
View File

@ -16,6 +16,7 @@
#include "sparsebit.h"
#define KVM_MAX_VCPUS 512
/*
* Callers of kvm_util only have an incomplete/opaque description of the
@ -70,6 +71,14 @@ enum vm_guest_mode {
#define vm_guest_mode_string(m) vm_guest_mode_string[m]
extern const char * const vm_guest_mode_string[];
struct vm_guest_mode_params {
unsigned int pa_bits;
unsigned int va_bits;
unsigned int page_size;
unsigned int page_shift;
};
extern const struct vm_guest_mode_params vm_guest_mode_params[];
enum vm_mem_backing_src_type {
VM_MEM_SRC_ANONYMOUS,
VM_MEM_SRC_ANONYMOUS_THP,

167
tools/testing/selftests/kvm/include/perf_test_util.h
View File

@ -9,38 +9,15 @@
#define SELFTEST_KVM_PERF_TEST_UTIL_H
#include "kvm_util.h"
#include "processor.h"
#define MAX_VCPUS 512
#define PAGE_SHIFT_4K 12
#define PTES_PER_4K_PT 512
#define TEST_MEM_SLOT_INDEX 1
/* Default guest test virtual memory offset */
#define DEFAULT_GUEST_TEST_MEM 0xc0000000
#define DEFAULT_PER_VCPU_MEM_SIZE (1 << 30) /* 1G */
/*
* Guest physical memory offset of the testing memory slot.
* This will be set to the topmost valid physical address minus
* the test memory size.
*/
static uint64_t guest_test_phys_mem;
#define PERF_TEST_MEM_SLOT_INDEX 1
/*
* Guest virtual memory offset of the testing memory slot.
* Must not conflict with identity mapped test code.
*/
static uint64_t guest_test_virt_mem = DEFAULT_GUEST_TEST_MEM;
static uint64_t guest_percpu_mem_size = DEFAULT_PER_VCPU_MEM_SIZE;
/* Number of VCPUs for the test */
static int nr_vcpus = 1;
struct vcpu_args {
struct perf_test_vcpu_args {
uint64_t gva;
uint64_t pages;
@ -54,141 +31,21 @@ struct perf_test_args {
uint64_t guest_page_size;
int wr_fract;
struct vcpu_args vcpu_args[MAX_VCPUS];
struct perf_test_vcpu_args vcpu_args[KVM_MAX_VCPUS];
};
static struct perf_test_args perf_test_args;
extern struct perf_test_args perf_test_args;
/*
* Continuously write to the first 8 bytes of each page in the
* specified region.
* Guest physical memory offset of the testing memory slot.
* This will be set to the topmost valid physical address minus
* the test memory size.
*/
static void guest_code(uint32_t vcpu_id)
{
struct vcpu_args *vcpu_args = &perf_test_args.vcpu_args[vcpu_id];
uint64_t gva;
uint64_t pages;
int i;
extern uint64_t guest_test_phys_mem;
/* Make sure vCPU args data structure is not corrupt. */
GUEST_ASSERT(vcpu_args->vcpu_id == vcpu_id);
gva = vcpu_args->gva;
pages = vcpu_args->pages;
while (true) {
for (i = 0; i < pages; i++) {
uint64_t addr = gva + (i * perf_test_args.guest_page_size);
if (i % perf_test_args.wr_fract == 0)
*(uint64_t *)addr = 0x0123456789ABCDEF;
else
READ_ONCE(*(uint64_t *)addr);
}
GUEST_SYNC(1);
}
}
static struct kvm_vm *create_vm(enum vm_guest_mode mode, int vcpus,
uint64_t vcpu_memory_bytes)
{
struct kvm_vm *vm;
uint64_t pages = DEFAULT_GUEST_PHY_PAGES;
uint64_t guest_num_pages;
/* Account for a few pages per-vCPU for stacks */
pages += DEFAULT_STACK_PGS * vcpus;
/*
* Reserve twice the ammount of memory needed to map the test region and
* the page table / stacks region, at 4k, for page tables. Do the
* calculation with 4K page size: the smallest of all archs. (e.g., 64K
* page size guest will need even less memory for page tables).
*/
pages += (2 * pages) / PTES_PER_4K_PT;
pages += ((2 * vcpus * vcpu_memory_bytes) >> PAGE_SHIFT_4K) /
PTES_PER_4K_PT;
pages = vm_adjust_num_guest_pages(mode, pages);
pr_info("Testing guest mode: %s\n", vm_guest_mode_string(mode));
vm = vm_create(mode, pages, O_RDWR);
kvm_vm_elf_load(vm, program_invocation_name, 0, 0);
#ifdef __x86_64__
vm_create_irqchip(vm);
#endif
perf_test_args.vm = vm;
perf_test_args.guest_page_size = vm_get_page_size(vm);
perf_test_args.host_page_size = getpagesize();
TEST_ASSERT(vcpu_memory_bytes % perf_test_args.guest_page_size == 0,
"Guest memory size is not guest page size aligned.");
guest_num_pages = (vcpus * vcpu_memory_bytes) /
perf_test_args.guest_page_size;
guest_num_pages = vm_adjust_num_guest_pages(mode, guest_num_pages);
/*
* If there should be more memory in the guest test region than there
* can be pages in the guest, it will definitely cause problems.
*/
TEST_ASSERT(guest_num_pages < vm_get_max_gfn(vm),
"Requested more guest memory than address space allows.\n"
" guest pages: %lx max gfn: %x vcpus: %d wss: %lx]\n",
guest_num_pages, vm_get_max_gfn(vm), vcpus,
vcpu_memory_bytes);
TEST_ASSERT(vcpu_memory_bytes % perf_test_args.host_page_size == 0,
"Guest memory size is not host page size aligned.");
guest_test_phys_mem = (vm_get_max_gfn(vm) - guest_num_pages) *
perf_test_args.guest_page_size;
guest_test_phys_mem &= ~(perf_test_args.host_page_size - 1);
#ifdef __s390x__
/* Align to 1M (segment size) */
guest_test_phys_mem &= ~((1 << 20) - 1);
#endif
pr_info("guest physical test memory offset: 0x%lx\n", guest_test_phys_mem);
/* Add an extra memory slot for testing */
vm_userspace_mem_region_add(vm, VM_MEM_SRC_ANONYMOUS,
guest_test_phys_mem,
TEST_MEM_SLOT_INDEX,
guest_num_pages, 0);
/* Do mapping for the demand paging memory slot */
virt_map(vm, guest_test_virt_mem, guest_test_phys_mem, guest_num_pages, 0);
ucall_init(vm, NULL);
return vm;
}
static void add_vcpus(struct kvm_vm *vm, int vcpus, uint64_t vcpu_memory_bytes)
{
vm_paddr_t vcpu_gpa;
struct vcpu_args *vcpu_args;
int vcpu_id;
for (vcpu_id = 0; vcpu_id < vcpus; vcpu_id++) {
vcpu_args = &perf_test_args.vcpu_args[vcpu_id];
vm_vcpu_add_default(vm, vcpu_id, guest_code);
vcpu_args->vcpu_id = vcpu_id;
vcpu_args->gva = guest_test_virt_mem +
(vcpu_id * vcpu_memory_bytes);
vcpu_args->pages = vcpu_memory_bytes /
perf_test_args.guest_page_size;
vcpu_gpa = guest_test_phys_mem + (vcpu_id * vcpu_memory_bytes);
pr_debug("Added VCPU %d with test mem gpa [%lx, %lx)\n",
vcpu_id, vcpu_gpa, vcpu_gpa + vcpu_memory_bytes);
}
}
struct kvm_vm *perf_test_create_vm(enum vm_guest_mode mode, int vcpus,
uint64_t vcpu_memory_bytes);
void perf_test_destroy_vm(struct kvm_vm *vm);
void perf_test_setup_vcpus(struct kvm_vm *vm, int vcpus, uint64_t vcpu_memory_bytes);
#endif /* SELFTEST_KVM_PERF_TEST_UTIL_H */

70
tools/testing/selftests/kvm/lib/guest_modes.c Normal file
View File

@ -0,0 +1,70 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright (C) 2020, Red Hat, Inc.
*/
#include "guest_modes.h"
struct guest_mode guest_modes[NUM_VM_MODES];
void guest_modes_append_default(void)
{
guest_mode_append(VM_MODE_DEFAULT, true, true);
#ifdef __aarch64__
guest_mode_append(VM_MODE_P40V48_64K, true, true);
{
unsigned int limit = kvm_check_cap(KVM_CAP_ARM_VM_IPA_SIZE);
if (limit >= 52)
guest_mode_append(VM_MODE_P52V48_64K, true, true);
if (limit >= 48) {
guest_mode_append(VM_MODE_P48V48_4K, true, true);
guest_mode_append(VM_MODE_P48V48_64K, true, true);
}
}
#endif
}
void for_each_guest_mode(void (*func)(enum vm_guest_mode, void *), void *arg)
{
int i;
for (i = 0; i < NUM_VM_MODES; ++i) {
if (!guest_modes[i].enabled)
continue;
TEST_ASSERT(guest_modes[i].supported,
"Guest mode ID %d (%s) not supported.",
i, vm_guest_mode_string(i));
func(i, arg);
}
}
void guest_modes_help(void)
{
int i;
printf(" -m: specify the guest mode ID to test\n"
" (default: test all supported modes)\n"
" This option may be used multiple times.\n"
" Guest mode IDs:\n");
for (i = 0; i < NUM_VM_MODES; ++i) {
printf(" %d: %s%s\n", i, vm_guest_mode_string(i),
guest_modes[i].supported ? " (supported)" : "");
}
}
void guest_modes_cmdline(const char *arg)
{
static bool mode_selected;
unsigned int mode;
int i;
if (!mode_selected) {
for (i = 0; i < NUM_VM_MODES; ++i)
guest_modes[i].enabled = false;
mode_selected = true;
}
mode = strtoul(optarg, NULL, 10);
TEST_ASSERT(mode < NUM_VM_MODES, "Guest mode ID %d too big", mode);
guest_modes[mode].enabled = true;
}

9
tools/testing/selftests/kvm/lib/kvm_util.c
View File

@ -153,14 +153,7 @@ const char * const vm_guest_mode_string[] = {
_Static_assert(sizeof(vm_guest_mode_string)/sizeof(char *) == NUM_VM_MODES,
"Missing new mode strings?");
struct vm_guest_mode_params {
unsigned int pa_bits;
unsigned int va_bits;
unsigned int page_size;
unsigned int page_shift;
};
static const struct vm_guest_mode_params vm_guest_mode_params[] = {
const struct vm_guest_mode_params vm_guest_mode_params[] = {
{ 52, 48, 0x1000, 12 },
{ 52, 48, 0x10000, 16 },
{ 48, 48, 0x1000, 12 },

134
tools/testing/selftests/kvm/lib/perf_test_util.c Normal file
View File

@ -0,0 +1,134 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright (C) 2020, Google LLC.
*/
#include "kvm_util.h"
#include "perf_test_util.h"
#include "processor.h"
struct perf_test_args perf_test_args;
uint64_t guest_test_phys_mem;
/*
* Guest virtual memory offset of the testing memory slot.
* Must not conflict with identity mapped test code.
*/
static uint64_t guest_test_virt_mem = DEFAULT_GUEST_TEST_MEM;
/*
* Continuously write to the first 8 bytes of each page in the
* specified region.
*/
static void guest_code(uint32_t vcpu_id)
{
struct perf_test_vcpu_args *vcpu_args = &perf_test_args.vcpu_args[vcpu_id];
uint64_t gva;
uint64_t pages;
int i;
/* Make sure vCPU args data structure is not corrupt. */
GUEST_ASSERT(vcpu_args->vcpu_id == vcpu_id);
gva = vcpu_args->gva;
pages = vcpu_args->pages;
while (true) {
for (i = 0; i < pages; i++) {
uint64_t addr = gva + (i * perf_test_args.guest_page_size);
if (i % perf_test_args.wr_fract == 0)
*(uint64_t *)addr = 0x0123456789ABCDEF;
else
READ_ONCE(*(uint64_t *)addr);
}
GUEST_SYNC(1);
}
}
struct kvm_vm *perf_test_create_vm(enum vm_guest_mode mode, int vcpus,
uint64_t vcpu_memory_bytes)
{
struct kvm_vm *vm;
uint64_t guest_num_pages;
pr_info("Testing guest mode: %s\n", vm_guest_mode_string(mode));
perf_test_args.host_page_size = getpagesize();
perf_test_args.guest_page_size = vm_guest_mode_params[mode].page_size;
guest_num_pages = vm_adjust_num_guest_pages(mode,
(vcpus * vcpu_memory_bytes) / perf_test_args.guest_page_size);
TEST_ASSERT(vcpu_memory_bytes % perf_test_args.host_page_size == 0,
"Guest memory size is not host page size aligned.");
TEST_ASSERT(vcpu_memory_bytes % perf_test_args.guest_page_size == 0,
"Guest memory size is not guest page size aligned.");
vm = vm_create_with_vcpus(mode, vcpus,
(vcpus * vcpu_memory_bytes) / perf_test_args.guest_page_size,
0, guest_code, NULL);
perf_test_args.vm = vm;
/*
* If there should be more memory in the guest test region than there
* can be pages in the guest, it will definitely cause problems.
*/
TEST_ASSERT(guest_num_pages < vm_get_max_gfn(vm),
"Requested more guest memory than address space allows.\n"
" guest pages: %lx max gfn: %x vcpus: %d wss: %lx]\n",
guest_num_pages, vm_get_max_gfn(vm), vcpus,
vcpu_memory_bytes);
guest_test_phys_mem = (vm_get_max_gfn(vm) - guest_num_pages) *
perf_test_args.guest_page_size;
guest_test_phys_mem &= ~(perf_test_args.host_page_size - 1);
#ifdef __s390x__
/* Align to 1M (segment size) */
guest_test_phys_mem &= ~((1 << 20) - 1);
#endif
pr_info("guest physical test memory offset: 0x%lx\n", guest_test_phys_mem);
/* Add an extra memory slot for testing */
vm_userspace_mem_region_add(vm, VM_MEM_SRC_ANONYMOUS,
guest_test_phys_mem,
PERF_TEST_MEM_SLOT_INDEX,
guest_num_pages, 0);
/* Do mapping for the demand paging memory slot */
virt_map(vm, guest_test_virt_mem, guest_test_phys_mem, guest_num_pages, 0);
ucall_init(vm, NULL);
return vm;
}
void perf_test_destroy_vm(struct kvm_vm *vm)
{
ucall_uninit(vm);
kvm_vm_free(vm);
}
void perf_test_setup_vcpus(struct kvm_vm *vm, int vcpus, uint64_t vcpu_memory_bytes)
{
vm_paddr_t vcpu_gpa;
struct perf_test_vcpu_args *vcpu_args;
int vcpu_id;
for (vcpu_id = 0; vcpu_id < vcpus; vcpu_id++) {
vcpu_args = &perf_test_args.vcpu_args[vcpu_id];
vcpu_args->vcpu_id = vcpu_id;
vcpu_args->gva = guest_test_virt_mem +
(vcpu_id * vcpu_memory_bytes);
vcpu_args->pages = vcpu_memory_bytes /
perf_test_args.guest_page_size;
vcpu_gpa = guest_test_phys_mem + (vcpu_id * vcpu_memory_bytes);
pr_debug("Added VCPU %d with test mem gpa [%lx, %lx)\n",
vcpu_id, vcpu_gpa, vcpu_gpa + vcpu_memory_bytes);
}
}