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

Merge 5.14-rc5 into driver-core-next

Browse Source 
We need the driver core fixes in here as well.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit is contained in:
Greg Kroah-Hartman
2021-08-09 09:03:47 +02:00
parent 96ba6c6e89 36a21d5172
commit bd935a7b21
518 changed files with 5466 additions and 2330 deletions
Download Patch File Download Diff File Expand all files Collapse all files

229
tools/testing/selftests/bpf/verifier/value_ptr_arith.c
View File

@@ -1,3 +1,232 @@
{
"map access: known scalar += value_ptr unknown vs const",
.insns = {
BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
offsetof(struct __sk_buff, len)),
BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 9),
BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 4),
BPF_MOV64_IMM(BPF_REG_1, 6),
BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
BPF_JMP_IMM(BPF_JA, 0, 0, 1),
BPF_MOV64_IMM(BPF_REG_1, 3),
BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
BPF_MOV64_IMM(BPF_REG_0, 1),
BPF_EXIT_INSN(),
},
.fixup_map_hash_16b = { 5 },
.fixup_map_array_48b = { 8 },
.result_unpriv = REJECT,
.errstr_unpriv = "R1 tried to add from different maps, paths or scalars",
.result = ACCEPT,
.retval = 1,
},
{
"map access: known scalar += value_ptr const vs unknown",
.insns = {
BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
offsetof(struct __sk_buff, len)),
BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 9),
BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 2),
BPF_MOV64_IMM(BPF_REG_1, 3),
BPF_JMP_IMM(BPF_JA, 0, 0, 3),
BPF_MOV64_IMM(BPF_REG_1, 6),
BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
BPF_MOV64_IMM(BPF_REG_0, 1),
BPF_EXIT_INSN(),
},
.fixup_map_hash_16b = { 5 },
.fixup_map_array_48b = { 8 },
.result_unpriv = REJECT,
.errstr_unpriv = "R1 tried to add from different maps, paths or scalars",
.result = ACCEPT,
.retval = 1,
},
{
"map access: known scalar += value_ptr const vs const (ne)",
.insns = {
BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
offsetof(struct __sk_buff, len)),
BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 2),
BPF_MOV64_IMM(BPF_REG_1, 3),
BPF_JMP_IMM(BPF_JA, 0, 0, 1),
BPF_MOV64_IMM(BPF_REG_1, 5),
BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
BPF_MOV64_IMM(BPF_REG_0, 1),
BPF_EXIT_INSN(),
},
.fixup_map_hash_16b = { 5 },
.fixup_map_array_48b = { 8 },
.result_unpriv = REJECT,
.errstr_unpriv = "R1 tried to add from different maps, paths or scalars",
.result = ACCEPT,
.retval = 1,
},
{
"map access: known scalar += value_ptr const vs const (eq)",
.insns = {
BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
offsetof(struct __sk_buff, len)),
BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 7),
BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 2),
BPF_MOV64_IMM(BPF_REG_1, 5),
BPF_JMP_IMM(BPF_JA, 0, 0, 1),
BPF_MOV64_IMM(BPF_REG_1, 5),
BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
BPF_MOV64_IMM(BPF_REG_0, 1),
BPF_EXIT_INSN(),
},
.fixup_map_hash_16b = { 5 },
.fixup_map_array_48b = { 8 },
.result = ACCEPT,
.retval = 1,
},
{
"map access: known scalar += value_ptr unknown vs unknown (eq)",
.insns = {
BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
offsetof(struct __sk_buff, len)),
BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 4),
BPF_MOV64_IMM(BPF_REG_1, 6),
BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
BPF_JMP_IMM(BPF_JA, 0, 0, 3),
BPF_MOV64_IMM(BPF_REG_1, 6),
BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
BPF_MOV64_IMM(BPF_REG_0, 1),
BPF_EXIT_INSN(),
},
.fixup_map_hash_16b = { 5 },
.fixup_map_array_48b = { 8 },
.result = ACCEPT,
.retval = 1,
},
{
"map access: known scalar += value_ptr unknown vs unknown (lt)",
.insns = {
BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
offsetof(struct __sk_buff, len)),
BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 4),
BPF_MOV64_IMM(BPF_REG_1, 6),
BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x3),
BPF_JMP_IMM(BPF_JA, 0, 0, 3),
BPF_MOV64_IMM(BPF_REG_1, 6),
BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
BPF_MOV64_IMM(BPF_REG_0, 1),
BPF_EXIT_INSN(),
},
.fixup_map_hash_16b = { 5 },
.fixup_map_array_48b = { 8 },
.result_unpriv = REJECT,
.errstr_unpriv = "R1 tried to add from different maps, paths or scalars",
.result = ACCEPT,
.retval = 1,
},
{
"map access: known scalar += value_ptr unknown vs unknown (gt)",
.insns = {
BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1,
offsetof(struct __sk_buff, len)),
BPF_ST_MEM(BPF_DW, BPF_REG_10, -8, 0),
BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 1, 3),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 1, 2),
BPF_LD_MAP_FD(BPF_REG_1, 0),
BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 11),
BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_0, 0),
BPF_JMP_IMM(BPF_JEQ, BPF_REG_4, 1, 4),
BPF_MOV64_IMM(BPF_REG_1, 6),
BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x7),
BPF_JMP_IMM(BPF_JA, 0, 0, 3),
BPF_MOV64_IMM(BPF_REG_1, 6),
BPF_ALU64_IMM(BPF_NEG, BPF_REG_1, 0),
BPF_ALU64_IMM(BPF_AND, BPF_REG_1, 0x3),
BPF_ALU64_REG(BPF_ADD, BPF_REG_1, BPF_REG_0),
BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
BPF_MOV64_IMM(BPF_REG_0, 1),
BPF_EXIT_INSN(),
},
.fixup_map_hash_16b = { 5 },
.fixup_map_array_48b = { 8 },
.result_unpriv = REJECT,
.errstr_unpriv = "R1 tried to add from different maps, paths or scalars",
.result = ACCEPT,
.retval = 1,
},
{
"map access: known scalar += value_ptr from different maps",
.insns = {

1
tools/testing/selftests/kvm/.gitignore vendored
View File

@@ -38,6 +38,7 @@
/x86_64/xen_vmcall_test
/x86_64/xss_msr_test
/x86_64/vmx_pmu_msrs_test
/access_tracking_perf_test
/demand_paging_test
/dirty_log_test
/dirty_log_perf_test

1
tools/testing/selftests/kvm/Makefile
View File

@@ -71,6 +71,7 @@ TEST_GEN_PROGS_x86_64 += x86_64/tsc_msrs_test
TEST_GEN_PROGS_x86_64 += x86_64/vmx_pmu_msrs_test
TEST_GEN_PROGS_x86_64 += x86_64/xen_shinfo_test
TEST_GEN_PROGS_x86_64 += x86_64/xen_vmcall_test
TEST_GEN_PROGS_x86_64 += access_tracking_perf_test
TEST_GEN_PROGS_x86_64 += demand_paging_test
TEST_GEN_PROGS_x86_64 += dirty_log_test
TEST_GEN_PROGS_x86_64 += dirty_log_perf_test

3
tools/testing/selftests/kvm/aarch64/get-reg-list.c
View File

@@ -1019,7 +1019,8 @@ static __u64 sve_rejects_set[] = {
#define VREGS_SUBLIST \
{ "vregs", .regs = vregs, .regs_n = ARRAY_SIZE(vregs), }
#define PMU_SUBLIST \
{ "pmu", .regs = pmu_regs, .regs_n = ARRAY_SIZE(pmu_regs), }
{ "pmu", .capability = KVM_CAP_ARM_PMU_V3, .feature = KVM_ARM_VCPU_PMU_V3, \
.regs = pmu_regs, .regs_n = ARRAY_SIZE(pmu_regs), }
#define SVE_SUBLIST \
{ "sve", .capability = KVM_CAP_ARM_SVE, .feature = KVM_ARM_VCPU_SVE, .finalize = true, \
.regs = sve_regs, .regs_n = ARRAY_SIZE(sve_regs), \

429
tools/testing/selftests/kvm/access_tracking_perf_test.c Normal file
View File

@@ -0,0 +1,429 @@
// SPDX-License-Identifier: GPL-2.0
/*
* access_tracking_perf_test
*
* Copyright (C) 2021, Google, Inc.
*
* This test measures the performance effects of KVM's access tracking.
* Access tracking is driven by the MMU notifiers test_young, clear_young, and
* clear_flush_young. These notifiers do not have a direct userspace API,
* however the clear_young notifier can be triggered by marking a pages as idle
* in /sys/kernel/mm/page_idle/bitmap. This test leverages that mechanism to
* enable access tracking on guest memory.
*
* To measure performance this test runs a VM with a configurable number of
* vCPUs that each touch every page in disjoint regions of memory. Performance
* is measured in the time it takes all vCPUs to finish touching their
* predefined region.
*
* Note that a deterministic correctness test of access tracking is not possible
* by using page_idle as it exists today. This is for a few reasons:
*
* 1. page_idle only issues clear_young notifiers, which lack a TLB flush. This
* means subsequent guest accesses are not guaranteed to see page table
* updates made by KVM until some time in the future.
*
* 2. page_idle only operates on LRU pages. Newly allocated pages are not
* immediately allocated to LRU lists. Instead they are held in a "pagevec",
* which is drained to LRU lists some time in the future. There is no
* userspace API to force this drain to occur.
*
* These limitations are worked around in this test by using a large enough
* region of memory for each vCPU such that the number of translations cached in
* the TLB and the number of pages held in pagevecs are a small fraction of the
* overall workload. And if either of those conditions are not true this test
* will fail rather than silently passing.
*/
#include <inttypes.h>
#include <limits.h>
#include <pthread.h>
#include <sys/mman.h>
#include <sys/types.h>
#include <sys/stat.h>
#include "kvm_util.h"
#include "test_util.h"
#include "perf_test_util.h"
#include "guest_modes.h"
/* Global variable used to synchronize all of the vCPU threads. */
static int iteration = -1;
/* Defines what vCPU threads should do during a given iteration. */
static enum {
/* Run the vCPU to access all its memory. */
ITERATION_ACCESS_MEMORY,
/* Mark the vCPU's memory idle in page_idle. */
ITERATION_MARK_IDLE,
} iteration_work;
/* Set to true when vCPU threads should exit. */
static bool done;
/* The iteration that was last completed by each vCPU. */
static int vcpu_last_completed_iteration[KVM_MAX_VCPUS];
/* Whether to overlap the regions of memory vCPUs access. */
static bool overlap_memory_access;
struct test_params {
/* The backing source for the region of memory. */
enum vm_mem_backing_src_type backing_src;
/* The amount of memory to allocate for each vCPU. */
uint64_t vcpu_memory_bytes;
/* The number of vCPUs to create in the VM. */
int vcpus;
};
static uint64_t pread_uint64(int fd, const char *filename, uint64_t index)
{
uint64_t value;
off_t offset = index * sizeof(value);
TEST_ASSERT(pread(fd, &value, sizeof(value), offset) == sizeof(value),
"pread from %s offset 0x%" PRIx64 " failed!",
filename, offset);
return value;
}
#define PAGEMAP_PRESENT (1ULL << 63)
#define PAGEMAP_PFN_MASK ((1ULL << 55) - 1)
static uint64_t lookup_pfn(int pagemap_fd, struct kvm_vm *vm, uint64_t gva)
{
uint64_t hva = (uint64_t) addr_gva2hva(vm, gva);
uint64_t entry;
uint64_t pfn;
entry = pread_uint64(pagemap_fd, "pagemap", hva / getpagesize());
if (!(entry & PAGEMAP_PRESENT))
return 0;
pfn = entry & PAGEMAP_PFN_MASK;
if (!pfn) {
print_skip("Looking up PFNs requires CAP_SYS_ADMIN");
exit(KSFT_SKIP);
}
return pfn;
}
static bool is_page_idle(int page_idle_fd, uint64_t pfn)
{
uint64_t bits = pread_uint64(page_idle_fd, "page_idle", pfn / 64);
return !!((bits >> (pfn % 64)) & 1);
}
static void mark_page_idle(int page_idle_fd, uint64_t pfn)
{
uint64_t bits = 1ULL << (pfn % 64);
TEST_ASSERT(pwrite(page_idle_fd, &bits, 8, 8 * (pfn / 64)) == 8,
"Set page_idle bits for PFN 0x%" PRIx64, pfn);
}
static void mark_vcpu_memory_idle(struct kvm_vm *vm, int vcpu_id)
{
uint64_t base_gva = perf_test_args.vcpu_args[vcpu_id].gva;
uint64_t pages = perf_test_args.vcpu_args[vcpu_id].pages;
uint64_t page;
uint64_t still_idle = 0;
uint64_t no_pfn = 0;
int page_idle_fd;
int pagemap_fd;
/* If vCPUs are using an overlapping region, let vCPU 0 mark it idle. */
if (overlap_memory_access && vcpu_id)
return;
page_idle_fd = open("/sys/kernel/mm/page_idle/bitmap", O_RDWR);
TEST_ASSERT(page_idle_fd > 0, "Failed to open page_idle.");
pagemap_fd = open("/proc/self/pagemap", O_RDONLY);
TEST_ASSERT(pagemap_fd > 0, "Failed to open pagemap.");
for (page = 0; page < pages; page++) {
uint64_t gva = base_gva + page * perf_test_args.guest_page_size;
uint64_t pfn = lookup_pfn(pagemap_fd, vm, gva);
if (!pfn) {
no_pfn++;
continue;
}
if (is_page_idle(page_idle_fd, pfn)) {
still_idle++;
continue;
}
mark_page_idle(page_idle_fd, pfn);
}
/*
* Assumption: Less than 1% of pages are going to be swapped out from
* under us during this test.
*/
TEST_ASSERT(no_pfn < pages / 100,
"vCPU %d: No PFN for %" PRIu64 " out of %" PRIu64 " pages.",
vcpu_id, no_pfn, pages);
/*
* Test that at least 90% of memory has been marked idle (the rest might
* not be marked idle because the pages have not yet made it to an LRU
* list or the translations are still cached in the TLB). 90% is
* arbitrary; high enough that we ensure most memory access went through
* access tracking but low enough as to not make the test too brittle
* over time and across architectures.
*/
TEST_ASSERT(still_idle < pages / 10,
"vCPU%d: Too many pages still idle (%"PRIu64 " out of %"
PRIu64 ").\n",
vcpu_id, still_idle, pages);
close(page_idle_fd);
close(pagemap_fd);
}
static void assert_ucall(struct kvm_vm *vm, uint32_t vcpu_id,
uint64_t expected_ucall)
{
struct ucall uc;
uint64_t actual_ucall = get_ucall(vm, vcpu_id, &uc);
TEST_ASSERT(expected_ucall == actual_ucall,
"Guest exited unexpectedly (expected ucall %" PRIu64
", got %" PRIu64 ")",
expected_ucall, actual_ucall);
}
static bool spin_wait_for_next_iteration(int *current_iteration)
{
int last_iteration = *current_iteration;
do {
if (READ_ONCE(done))
return false;
*current_iteration = READ_ONCE(iteration);
} while (last_iteration == *current_iteration);
return true;
}
static void *vcpu_thread_main(void *arg)
{
struct perf_test_vcpu_args *vcpu_args = arg;
struct kvm_vm *vm = perf_test_args.vm;
int vcpu_id = vcpu_args->vcpu_id;
int current_iteration = -1;
vcpu_args_set(vm, vcpu_id, 1, vcpu_id);
while (spin_wait_for_next_iteration(&current_iteration)) {
switch (READ_ONCE(iteration_work)) {
case ITERATION_ACCESS_MEMORY:
vcpu_run(vm, vcpu_id);
assert_ucall(vm, vcpu_id, UCALL_SYNC);
break;
case ITERATION_MARK_IDLE:
mark_vcpu_memory_idle(vm, vcpu_id);
break;
};
vcpu_last_completed_iteration[vcpu_id] = current_iteration;
}
return NULL;
}
static void spin_wait_for_vcpu(int vcpu_id, int target_iteration)
{
while (READ_ONCE(vcpu_last_completed_iteration[vcpu_id]) !=
target_iteration) {
continue;
}
}
/* The type of memory accesses to perform in the VM. */
enum access_type {
ACCESS_READ,
ACCESS_WRITE,
};
static void run_iteration(struct kvm_vm *vm, int vcpus, const char *description)
{
struct timespec ts_start;
struct timespec ts_elapsed;
int next_iteration;
int vcpu_id;
/* Kick off the vCPUs by incrementing iteration. */
next_iteration = ++iteration;
clock_gettime(CLOCK_MONOTONIC, &ts_start);
/* Wait for all vCPUs to finish the iteration. */
for (vcpu_id = 0; vcpu_id < vcpus; vcpu_id++)
spin_wait_for_vcpu(vcpu_id, next_iteration);
ts_elapsed = timespec_elapsed(ts_start);
pr_info("%-30s: %ld.%09lds\n",
description, ts_elapsed.tv_sec, ts_elapsed.tv_nsec);
}
static void access_memory(struct kvm_vm *vm, int vcpus, enum access_type access,
const char *description)
{
perf_test_args.wr_fract = (access == ACCESS_READ) ? INT_MAX : 1;
sync_global_to_guest(vm, perf_test_args);
iteration_work = ITERATION_ACCESS_MEMORY;
run_iteration(vm, vcpus, description);
}
static void mark_memory_idle(struct kvm_vm *vm, int vcpus)
{
/*
* Even though this parallelizes the work across vCPUs, this is still a
* very slow operation because page_idle forces the test to mark one pfn
* at a time and the clear_young notifier serializes on the KVM MMU
* lock.
*/
pr_debug("Marking VM memory idle (slow)...\n");
iteration_work = ITERATION_MARK_IDLE;
run_iteration(vm, vcpus, "Mark memory idle");
}
static pthread_t *create_vcpu_threads(int vcpus)
{
pthread_t *vcpu_threads;
int i;
vcpu_threads = malloc(vcpus * sizeof(vcpu_threads[0]));
TEST_ASSERT(vcpu_threads, "Failed to allocate vcpu_threads.");
for (i = 0; i < vcpus; i++) {
vcpu_last_completed_iteration[i] = iteration;
pthread_create(&vcpu_threads[i], NULL, vcpu_thread_main,
&perf_test_args.vcpu_args[i]);
}
return vcpu_threads;
}
static void terminate_vcpu_threads(pthread_t *vcpu_threads, int vcpus)
{
int i;
/* Set done to signal the vCPU threads to exit */
done = true;
for (i = 0; i < vcpus; i++)
pthread_join(vcpu_threads[i], NULL);
}
static void run_test(enum vm_guest_mode mode, void *arg)
{
struct test_params *params = arg;
struct kvm_vm *vm;
pthread_t *vcpu_threads;
int vcpus = params->vcpus;
vm = perf_test_create_vm(mode, vcpus, params->vcpu_memory_bytes,
params->backing_src);
perf_test_setup_vcpus(vm, vcpus, params->vcpu_memory_bytes,
!overlap_memory_access);
vcpu_threads = create_vcpu_threads(vcpus);
pr_info("\n");
access_memory(vm, vcpus, ACCESS_WRITE, "Populating memory");
/* As a control, read and write to the populated memory first. */
access_memory(vm, vcpus, ACCESS_WRITE, "Writing to populated memory");
access_memory(vm, vcpus, ACCESS_READ, "Reading from populated memory");
/* Repeat on memory that has been marked as idle. */
mark_memory_idle(vm, vcpus);
access_memory(vm, vcpus, ACCESS_WRITE, "Writing to idle memory");
mark_memory_idle(vm, vcpus);
access_memory(vm, vcpus, ACCESS_READ, "Reading from idle memory");
terminate_vcpu_threads(vcpu_threads, vcpus);
free(vcpu_threads);
perf_test_destroy_vm(vm);
}
static void help(char *name)
{
puts("");
printf("usage: %s [-h] [-m mode] [-b vcpu_bytes] [-v vcpus] [-o] [-s mem_type]\n",
name);
puts("");
printf(" -h: Display this help message.");
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");
printf(" -v: specify the number of vCPUs to run.\n");
printf(" -o: Overlap guest memory accesses instead of partitioning\n"
" them into a separate region of memory for each vCPU.\n");
printf(" -s: specify the type of memory that should be used to\n"
" back the guest data region.\n\n");
backing_src_help();
puts("");
exit(0);
}
int main(int argc, char *argv[])
{
struct test_params params = {
.backing_src = VM_MEM_SRC_ANONYMOUS,
.vcpu_memory_bytes = DEFAULT_PER_VCPU_MEM_SIZE,
.vcpus = 1,
};
int page_idle_fd;
int opt;
guest_modes_append_default();
while ((opt = getopt(argc, argv, "hm:b:v:os:")) != -1) {
switch (opt) {
case 'm':
guest_modes_cmdline(optarg);
break;
case 'b':
params.vcpu_memory_bytes = parse_size(optarg);
break;
case 'v':
params.vcpus = atoi(optarg);
break;
case 'o':
overlap_memory_access = true;
break;
case 's':
params.backing_src = parse_backing_src_type(optarg);
break;
case 'h':
default:
help(argv[0]);
break;
}
}
page_idle_fd = open("/sys/kernel/mm/page_idle/bitmap", O_RDWR);
if (page_idle_fd < 0) {
print_skip("CONFIG_IDLE_PAGE_TRACKING is not enabled");
exit(KSFT_SKIP);
}
close(page_idle_fd);
for_each_guest_mode(run_test, &params);
return 0;
}

1
tools/testing/selftests/kvm/dirty_log_perf_test.c
View File

@@ -312,6 +312,7 @@ int main(int argc, char *argv[])
break;
case 'o':
p.partition_vcpu_memory_access = false;
break;
case 's':
p.backing_src = parse_backing_src_type(optarg);
break;

5
tools/testing/selftests/kvm/include/x86_64/hyperv.h
View File

@@ -117,7 +117,7 @@
#define HV_X64_GUEST_DEBUGGING_AVAILABLE BIT(1)
#define HV_X64_PERF_MONITOR_AVAILABLE BIT(2)
#define HV_X64_CPU_DYNAMIC_PARTITIONING_AVAILABLE BIT(3)
#define HV_X64_HYPERCALL_PARAMS_XMM_AVAILABLE BIT(4)
#define HV_X64_HYPERCALL_XMM_INPUT_AVAILABLE BIT(4)
#define HV_X64_GUEST_IDLE_STATE_AVAILABLE BIT(5)
#define HV_FEATURE_FREQUENCY_MSRS_AVAILABLE BIT(8)
#define HV_FEATURE_GUEST_CRASH_MSR_AVAILABLE BIT(10)
@@ -182,4 +182,7 @@
#define HV_STATUS_INVALID_CONNECTION_ID 18
#define HV_STATUS_INSUFFICIENT_BUFFERS 19
/* hypercall options */
#define HV_HYPERCALL_FAST_BIT BIT(16)
#endif /* !SELFTEST_KVM_HYPERV_H */

2
tools/testing/selftests/kvm/steal_time.c
View File

@@ -320,7 +320,7 @@ int main(int ac, char **av)
run_delay = get_run_delay();
pthread_create(&thread, &attr, do_steal_time, NULL);
do
pthread_yield();
sched_yield();
while (get_run_delay() - run_delay < MIN_RUN_DELAY_NS);
pthread_join(thread, NULL);
run_delay = get_run_delay() - run_delay;

2
tools/testing/selftests/kvm/x86_64/hyperv_clock.c
View File

@@ -215,7 +215,7 @@ int main(void)
vcpu_set_hv_cpuid(vm, VCPU_ID);
tsc_page_gva = vm_vaddr_alloc_page(vm);
memset(addr_gpa2hva(vm, tsc_page_gva), 0x0, getpagesize());
memset(addr_gva2hva(vm, tsc_page_gva), 0x0, getpagesize());
TEST_ASSERT((addr_gva2gpa(vm, tsc_page_gva) & (getpagesize() - 1)) == 0,
"TSC page has to be page aligned\n");
vcpu_args_set(vm, VCPU_ID, 2, tsc_page_gva, addr_gva2gpa(vm, tsc_page_gva));

41
tools/testing/selftests/kvm/x86_64/hyperv_features.c
View File

@@ -47,6 +47,7 @@ static void do_wrmsr(u32 idx, u64 val)
}
static int nr_gp;
static int nr_ud;
static inline u64 hypercall(u64 control, vm_vaddr_t input_address,
vm_vaddr_t output_address)
@@ -80,6 +81,12 @@ static void guest_gp_handler(struct ex_regs *regs)
regs->rip = (uint64_t)&wrmsr_end;
}
static void guest_ud_handler(struct ex_regs *regs)
{
nr_ud++;
regs->rip += 3;
}
struct msr_data {
uint32_t idx;
bool available;
@@ -90,6 +97,7 @@ struct msr_data {
struct hcall_data {
uint64_t control;
uint64_t expect;
bool ud_expected;
};
static void guest_msr(struct msr_data *msr)
@@ -117,13 +125,26 @@ static void guest_msr(struct msr_data *msr)
static void guest_hcall(vm_vaddr_t pgs_gpa, struct hcall_data *hcall)
{
int i = 0;
u64 res, input, output;
wrmsr(HV_X64_MSR_GUEST_OS_ID, LINUX_OS_ID);
wrmsr(HV_X64_MSR_HYPERCALL, pgs_gpa);
while (hcall->control) {
GUEST_ASSERT(hypercall(hcall->control, pgs_gpa,
pgs_gpa + 4096) == hcall->expect);
nr_ud = 0;
if (!(hcall->control & HV_HYPERCALL_FAST_BIT)) {
input = pgs_gpa;
output = pgs_gpa + 4096;
} else {
input = output = 0;
}
res = hypercall(hcall->control, input, output);
if (hcall->ud_expected)
GUEST_ASSERT(nr_ud == 1);
else
GUEST_ASSERT(res == hcall->expect);
GUEST_SYNC(i++);
}
@@ -552,8 +573,18 @@ static void guest_test_hcalls_access(struct kvm_vm *vm, struct hcall_data *hcall
recomm.ebx = 0xfff;
hcall->expect = HV_STATUS_SUCCESS;
break;
case 17:
/* XMM fast hypercall */
hcall->control = HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE | HV_HYPERCALL_FAST_BIT;
hcall->ud_expected = true;
break;
case 18:
feat.edx |= HV_X64_HYPERCALL_XMM_INPUT_AVAILABLE;
hcall->ud_expected = false;
hcall->expect = HV_STATUS_SUCCESS;
break;
case 19:
/* END */
hcall->control = 0;
break;
@@ -625,6 +656,10 @@ int main(void)
/* Test hypercalls */
vm = vm_create_default(VCPU_ID, 0, guest_hcall);
vm_init_descriptor_tables(vm);
vcpu_init_descriptor_tables(vm, VCPU_ID);
vm_install_exception_handler(vm, UD_VECTOR, guest_ud_handler);
/* Hypercall input/output */
hcall_page = vm_vaddr_alloc_pages(vm, 2);
memset(addr_gva2hva(vm, hcall_page), 0x0, 2 * getpagesize());

165
tools/testing/selftests/net/ipsec.c
View File

@@ -484,13 +484,16 @@ enum desc_type {
MONITOR_ACQUIRE,
EXPIRE_STATE,
EXPIRE_POLICY,
SPDINFO_ATTRS,
};
const char *desc_name[] = {
"create tunnel",
"alloc spi",
"monitor acquire",
"expire state",
"expire policy"
"expire policy",
"spdinfo attributes",
""
};
struct xfrm_desc {
enum desc_type type;
@@ -1593,6 +1596,155 @@ out_close:
return ret;
}
static int xfrm_spdinfo_set_thresh(int xfrm_sock, uint32_t *seq,
unsigned thresh4_l, unsigned thresh4_r,
unsigned thresh6_l, unsigned thresh6_r,
bool add_bad_attr)
{
struct {
struct nlmsghdr nh;
union {
uint32_t unused;
int error;
};
char attrbuf[MAX_PAYLOAD];
} req;
struct xfrmu_spdhthresh thresh;
memset(&req, 0, sizeof(req));
req.nh.nlmsg_len = NLMSG_LENGTH(sizeof(req.unused));
req.nh.nlmsg_type = XFRM_MSG_NEWSPDINFO;
req.nh.nlmsg_flags = NLM_F_REQUEST | NLM_F_ACK;
req.nh.nlmsg_seq = (*seq)++;
thresh.lbits = thresh4_l;
thresh.rbits = thresh4_r;
if (rtattr_pack(&req.nh, sizeof(req), XFRMA_SPD_IPV4_HTHRESH, &thresh, sizeof(thresh)))
return -1;
thresh.lbits = thresh6_l;
thresh.rbits = thresh6_r;
if (rtattr_pack(&req.nh, sizeof(req), XFRMA_SPD_IPV6_HTHRESH, &thresh, sizeof(thresh)))
return -1;
if (add_bad_attr) {
BUILD_BUG_ON(XFRMA_IF_ID <= XFRMA_SPD_MAX + 1);
if (rtattr_pack(&req.nh, sizeof(req), XFRMA_IF_ID, NULL, 0)) {
pr_err("adding attribute failed: no space");
return -1;
}
}
if (send(xfrm_sock, &req, req.nh.nlmsg_len, 0) < 0) {
pr_err("send()");
return -1;
}
if (recv(xfrm_sock, &req, sizeof(req), 0) < 0) {
pr_err("recv()");
return -1;
} else if (req.nh.nlmsg_type != NLMSG_ERROR) {
printk("expected NLMSG_ERROR, got %d", (int)req.nh.nlmsg_type);
return -1;
}
if (req.error) {
printk("NLMSG_ERROR: %d: %s", req.error, strerror(-req.error));
return -1;
}
return 0;
}
static int xfrm_spdinfo_attrs(int xfrm_sock, uint32_t *seq)
{
struct {
struct nlmsghdr nh;
union {
uint32_t unused;
int error;
};
char attrbuf[MAX_PAYLOAD];
} req;
if (xfrm_spdinfo_set_thresh(xfrm_sock, seq, 32, 31, 120, 16, false)) {
pr_err("Can't set SPD HTHRESH");
return KSFT_FAIL;
}
memset(&req, 0, sizeof(req));
req.nh.nlmsg_len = NLMSG_LENGTH(sizeof(req.unused));
req.nh.nlmsg_type = XFRM_MSG_GETSPDINFO;
req.nh.nlmsg_flags = NLM_F_REQUEST;
req.nh.nlmsg_seq = (*seq)++;
if (send(xfrm_sock, &req, req.nh.nlmsg_len, 0) < 0) {
pr_err("send()");
return KSFT_FAIL;
}
if (recv(xfrm_sock, &req, sizeof(req), 0) < 0) {
pr_err("recv()");
return KSFT_FAIL;
} else if (req.nh.nlmsg_type == XFRM_MSG_NEWSPDINFO) {
size_t len = NLMSG_PAYLOAD(&req.nh, sizeof(req.unused));
struct rtattr *attr = (void *)req.attrbuf;
int got_thresh = 0;
for (; RTA_OK(attr, len); attr = RTA_NEXT(attr, len)) {
if (attr->rta_type == XFRMA_SPD_IPV4_HTHRESH) {
struct xfrmu_spdhthresh *t = RTA_DATA(attr);
got_thresh++;
if (t->lbits != 32 || t->rbits != 31) {
pr_err("thresh differ: %u, %u",
t->lbits, t->rbits);
return KSFT_FAIL;
}
}
if (attr->rta_type == XFRMA_SPD_IPV6_HTHRESH) {
struct xfrmu_spdhthresh *t = RTA_DATA(attr);
got_thresh++;
if (t->lbits != 120 || t->rbits != 16) {
pr_err("thresh differ: %u, %u",
t->lbits, t->rbits);
return KSFT_FAIL;
}
}
}
if (got_thresh != 2) {
pr_err("only %d thresh returned by XFRM_MSG_GETSPDINFO", got_thresh);
return KSFT_FAIL;
}
} else if (req.nh.nlmsg_type != NLMSG_ERROR) {
printk("expected NLMSG_ERROR, got %d", (int)req.nh.nlmsg_type);
return KSFT_FAIL;
} else {
printk("NLMSG_ERROR: %d: %s", req.error, strerror(-req.error));
return -1;
}
/* Restore the default */
if (xfrm_spdinfo_set_thresh(xfrm_sock, seq, 32, 32, 128, 128, false)) {
pr_err("Can't restore SPD HTHRESH");
return KSFT_FAIL;
}
/*
* At this moment xfrm uses nlmsg_parse_deprecated(), which
* implies NL_VALIDATE_LIBERAL - ignoring attributes with
* (type > maxtype). nla_parse_depricated_strict() would enforce
* it. Or even stricter nla_parse().
* Right now it's not expected to fail, but to be ignored.
*/
if (xfrm_spdinfo_set_thresh(xfrm_sock, seq, 32, 32, 128, 128, true))
return KSFT_PASS;
return KSFT_PASS;
}
static int child_serv(int xfrm_sock, uint32_t *seq,
unsigned int nr, int cmd_fd, void *buf, struct xfrm_desc *desc)
{
@@ -1717,6 +1869,9 @@ static int child_f(unsigned int nr, int test_desc_fd, int cmd_fd, void *buf)
case EXPIRE_POLICY:
ret = xfrm_expire_policy(xfrm_sock, &seq, nr, &desc);
break;
case SPDINFO_ATTRS:
ret = xfrm_spdinfo_attrs(xfrm_sock, &seq);
break;
default:
printk("Unknown desc type %d", desc.type);
exit(KSFT_FAIL);
@@ -1994,8 +2149,10 @@ static int write_proto_plan(int fd, int proto)
* sizeof(xfrm_user_polexpire) = 168 | sizeof(xfrm_user_polexpire) = 176
*
* Check the affected by the UABI difference structures.
* Also, check translation for xfrm_set_spdinfo: it has it's own attributes
* which needs to be correctly copied, but not translated.
*/
const unsigned int compat_plan = 4;
const unsigned int compat_plan = 5;
static int write_compat_struct_tests(int test_desc_fd)
{
struct xfrm_desc desc = {};
@@ -2019,6 +2176,10 @@ static int write_compat_struct_tests(int test_desc_fd)
if (__write_desc(test_desc_fd, &desc))
return -1;
desc.type = SPDINFO_ATTRS;
if (__write_desc(test_desc_fd, &desc))
return -1;
return 0;
}