Linux-kselftest-mirror July 2025

linux-kselftest-mirror@lists.linaro.org

3 participants
3 discussions

[PATCH net-next] selftests: devmem: configure HDS threshold

by Taehee Yoo

The devmem TCP requires the hds-thresh value to be 0, but it doesn't change it automatically. Therefore, configure_hds_thresh() is added to handle this. The run_devmem_tests() now tests hds_thresh, but it skips test if the hds_thresh_max value is 0. Signed-off-by: Taehee Yoo <ap420073(a)gmail.com> --- .../selftests/drivers/net/hw/ncdevmem.c | 86 +++++++++++++++++++ 1 file changed, 86 insertions(+) diff --git a/tools/testing/selftests/drivers/net/hw/ncdevmem.c b/tools/testing/selftests/drivers/net/hw/ncdevmem.c index cc9b40d9c5d5..d78b5e5697d7 100644 --- a/tools/testing/selftests/drivers/net/hw/ncdevmem.c +++ b/tools/testing/selftests/drivers/net/hw/ncdevmem.c @@ -349,6 +349,72 @@ static int configure_headersplit(bool on) return ret; } +static int configure_hds_thresh(int len) +{ + struct ethtool_rings_get_req *get_req; + struct ethtool_rings_get_rsp *get_rsp; + struct ethtool_rings_set_req *req; + struct ynl_error yerr; + struct ynl_sock *ys; + int ret; + + ys = ynl_sock_create(&ynl_ethtool_family, &yerr); + if (!ys) { + fprintf(stderr, "YNL: %s\n", yerr.msg); + return -1; + } + + req = ethtool_rings_set_req_alloc(); + ethtool_rings_set_req_set_header_dev_index(req, ifindex); + ethtool_rings_set_req_set_hds_thresh(req, len); + ret = ethtool_rings_set(ys, req); + if (ret < 0) + fprintf(stderr, "YNL failed: %s\n", ys->err.msg); + ethtool_rings_set_req_free(req); + + if (ret == 0) { + get_req = ethtool_rings_get_req_alloc(); + ethtool_rings_get_req_set_header_dev_index(get_req, ifindex); + get_rsp = ethtool_rings_get(ys, get_req); + ethtool_rings_get_req_free(get_req); + if (get_rsp) + fprintf(stderr, "HDS threshold: %d\n", + get_rsp->hds_thresh); + ethtool_rings_get_rsp_free(get_rsp); + } + + ynl_sock_destroy(ys); + + return ret; +} + +static int get_hds_thresh_max(void) +{ + struct ethtool_rings_get_req *get_req; + struct ethtool_rings_get_rsp *get_rsp; + struct ynl_error yerr; + unsigned int ret = 0; + struct ynl_sock *ys; + + ys = ynl_sock_create(&ynl_ethtool_family, &yerr); + if (!ys) { + fprintf(stderr, "YNL: %s\n", yerr.msg); + return -1; + } + + get_req = ethtool_rings_get_req_alloc(); + ethtool_rings_get_req_set_header_dev_index(get_req, ifindex); + get_rsp = ethtool_rings_get(ys, get_req); + ethtool_rings_get_req_free(get_req); + if (get_rsp) + ret = get_rsp->hds_thresh_max; + ethtool_rings_get_rsp_free(get_rsp); + + ynl_sock_destroy(ys); + + return ret; +} + static int configure_rss(void) { return run_command("sudo ethtool -X %s equal %d >&2", ifname, start_queue); @@ -565,6 +631,9 @@ static int do_server(struct memory_buffer *mem) if (configure_headersplit(1)) error(1, 0, "Failed to enable TCP header split\n"); + if (configure_hds_thresh(0)) + error(1, 0, "Failed to set HDS threshold\n"); + /* Configure RSS to divert all traffic from our devmem queues */ if (configure_rss()) error(1, 0, "Failed to configure rss\n"); @@ -725,10 +794,14 @@ static int do_server(struct memory_buffer *mem) void run_devmem_tests(void) { struct memory_buffer *mem; + int hds_thresh_max = 0; struct ynl_sock *ys; mem = provider->alloc(getpagesize() * NUM_PAGES); + hds_thresh_max = get_hds_thresh_max(); + if (hds_thresh_max < 0) + error(1, 0, "Failed to get hds_thresh_max"); /* Configure RSS to divert all traffic from our devmem queues */ if (configure_rss()) error(1, 0, "rss error\n"); @@ -750,6 +823,19 @@ void run_devmem_tests(void) if (configure_headersplit(1)) error(1, 0, "Failed to configure header split\n"); + /* Do not test hds_thresh if hds_thresh_max value is 0 */ + if (hds_thresh_max) { + if (configure_hds_thresh(1)) + error(1, 0, "Failed to configure HDS threshold\n"); + + if (!bind_rx_queue(ifindex, mem->fd, create_queues(), + num_queues, &ys)) + error(1, 0, "Configure dmabuf with non-zero HDS threshold should have failed\n"); + + if (configure_hds_thresh(0)) + error(1, 0, "Failed to configure HDS threshold\n"); + } + if (bind_rx_queue(ifindex, mem->fd, create_queues(), num_queues, &ys)) error(1, 0, "Failed to bind\n"); -- 2.34.1

1 hour, 18 minutes

[PATCH v2] selftests/mm: Add process_madvise() tests

by wang lian

This patch adds tests for the process_madvise(), focusing on verifying behavior under various conditions including valid usage and error cases. Signed-off-by: wang lian<lianux.mm(a)gmail.com> Suggested-by: Lorenzo Stoakes <lorenzo.stoakes(a)oracle.com> Suggested-by: David Hildenbrand <david(a)redhat.com> --- Changelog v2: - Drop MADV_DONTNEED tests based on feedback - Focus solely on process_madvise() syscall - Improve error handling and structure - Add future-proof flag test - Style and comment cleanups tools/testing/selftests/mm/.gitignore | 1 + tools/testing/selftests/mm/Makefile | 1 + tools/testing/selftests/mm/process_madv.c | 414 ++++++++++++++++++++++ tools/testing/selftests/mm/run_vmtests.sh | 5 + 4 files changed, 421 insertions(+) create mode 100644 tools/testing/selftests/mm/process_madv.c diff --git a/tools/testing/selftests/mm/.gitignore b/tools/testing/selftests/mm/.gitignore index 911f39d634be..a8c3be02188c 100644 --- a/tools/testing/selftests/mm/.gitignore +++ b/tools/testing/selftests/mm/.gitignore @@ -42,6 +42,7 @@ memfd_secret hugetlb_dio pkey_sighandler_tests_32 pkey_sighandler_tests_64 +process_madv soft-dirty split_huge_page_test ksm_tests diff --git a/tools/testing/selftests/mm/Makefile b/tools/testing/selftests/mm/Makefile index 2352252f3914..725612e09582 100644 --- a/tools/testing/selftests/mm/Makefile +++ b/tools/testing/selftests/mm/Makefile @@ -86,6 +86,7 @@ TEST_GEN_FILES += mseal_test TEST_GEN_FILES += on-fault-limit TEST_GEN_FILES += pagemap_ioctl TEST_GEN_FILES += pfnmap +TEST_GEN_FILES += process_madv TEST_GEN_FILES += thuge-gen TEST_GEN_FILES += transhuge-stress TEST_GEN_FILES += uffd-stress diff --git a/tools/testing/selftests/mm/process_madv.c b/tools/testing/selftests/mm/process_madv.c new file mode 100644 index 000000000000..73999c8e3570 --- /dev/null +++ b/tools/testing/selftests/mm/process_madv.c @@ -0,0 +1,414 @@ +// SPDX-License-Identifier: GPL-2.0-or-later + +#define _GNU_SOURCE +#include "../kselftest_harness.h" +#include <errno.h> +#include <setjmp.h> +#include <signal.h> +#include <stdbool.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <sys/mman.h> +#include <sys/syscall.h> +#include <unistd.h> +#include <sched.h> +#include <sys/pidfd.h> +#include "vm_util.h" + +#include "../pidfd/pidfd.h" + +/* + * Ignore the checkpatch warning, as per the C99 standard, section 7.14.1.1: + * + * "If the signal occurs other than as the result of calling the abort or raise + * function, the behavior is undefined if the signal handler refers to any + * object with static storage duration other than by assigning a value to an + * object declared as volatile sig_atomic_t" + */ +static volatile sig_atomic_t signal_jump_set; +static sigjmp_buf signal_jmp_buf; + +/* + * Ignore the checkpatch warning, we must read from x but don't want to do + * anything with it in order to trigger a read page fault. We therefore must use + * volatile to stop the compiler from optimising this away. + */ +#define FORCE_READ(x) (*(volatile typeof(x) *)x) + +static void handle_fatal(int c) +{ + if (!signal_jump_set) + return; + + siglongjmp(signal_jmp_buf, c); +} + +FIXTURE(process_madvise) +{ + int pidfd; + int flag; +}; + +static void setup_sighandler(void) +{ + struct sigaction act = { + .sa_handler = &handle_fatal, + .sa_flags = SA_NODEFER, + }; + + sigemptyset(&act.sa_mask); + if (sigaction(SIGSEGV, &act, NULL)) + ksft_exit_fail_perror("sigaction"); +} + +static void teardown_sighandler(void) +{ + struct sigaction act = { + .sa_handler = SIG_DFL, + .sa_flags = SA_NODEFER, + }; + + sigemptyset(&act.sa_mask); + sigaction(SIGSEGV, &act, NULL); +} + +FIXTURE_SETUP(process_madvise) +{ + self->pidfd = PIDFD_SELF; + self->flag = 0; + setup_sighandler(); +}; + +FIXTURE_TEARDOWN_PARENT(process_madvise) +{ + teardown_sighandler(); +} + +static ssize_t sys_process_madvise(int pidfd, const struct iovec *iovec, + size_t vlen, int advice, unsigned int flags) +{ + return syscall(__NR_process_madvise, pidfd, iovec, vlen, advice, flags); +} + +/* + * Enable our signal catcher and try to read/write the specified buffer. The + * return value indicates whether the read/write succeeds without a fatal + * signal. + */ +static bool try_access_buf(char *ptr, bool write) +{ + bool failed; + + /* Tell signal handler to jump back here on fatal signal. */ + signal_jump_set = true; + /* If a fatal signal arose, we will jump back here and failed is set. */ + failed = sigsetjmp(signal_jmp_buf, 0) != 0; + + if (!failed) { + if (write) + *ptr = 'x'; + else + FORCE_READ(ptr); + } + + signal_jump_set = false; + return !failed; +} + +/* Try and read from a buffer, return true if no fatal signal. */ +static bool try_read_buf(char *ptr) +{ + return try_access_buf(ptr, false); +} + +TEST_F(process_madvise, basic) +{ + const unsigned long pagesize = (unsigned long)sysconf(_SC_PAGESIZE); + const int madvise_pages = 4; + char *map; + ssize_t ret; + struct iovec vec[madvise_pages]; + + /* + * Create a single large mapping. We will pick pages from this + * mapping to advise on. This ensures we test non-contiguous iovecs. + */ + map = mmap(NULL, pagesize * 10, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + ASSERT_NE(map, MAP_FAILED); + + /* Fill the entire region with a known pattern. */ + memset(map, 'A', pagesize * 10); + + /* + * Setup the iovec to point to 4 non-contiguous pages + * within the mapping. + */ + vec[0].iov_base = &map[0 * pagesize]; + vec[0].iov_len = pagesize; + vec[1].iov_base = &map[3 * pagesize]; + vec[1].iov_len = pagesize; + vec[2].iov_base = &map[5 * pagesize]; + vec[2].iov_len = pagesize; + vec[3].iov_base = &map[8 * pagesize]; + vec[3].iov_len = pagesize; + + ret = sys_process_madvise(PIDFD_SELF, vec, madvise_pages, MADV_DONTNEED, + 0); + if (ret == -1 && errno == EPERM) + ksft_exit_skip( + "process_madvise() unsupported or permission denied, try running as root.\n"); + else if (errno == EINVAL) + ksft_exit_skip( + "process_madvise() unsupported or parameter invalid, please check arguments.\n"); + + /* The call should succeed and report the total bytes processed. */ + ASSERT_EQ(ret, madvise_pages * pagesize); + + /* Check that advised pages are now zero. */ + for (int i = 0; i < madvise_pages; i++) { + char *advised_page = (char *)vec[i].iov_base; + + /* Access should be successful (kernel provides a new page). */ + ASSERT_TRUE(try_read_buf(advised_page)); + /* Content must be 0, not 'A'. */ + ASSERT_EQ(*advised_page, 0); + } + + /* Check that an un-advised page in between is still 'A'. */ + char *unadvised_page = &map[1 * pagesize]; + + ASSERT_TRUE(try_read_buf(unadvised_page)); + ASSERT_EQ(*unadvised_page, 'A'); + + /* Cleanup. */ + ASSERT_EQ(munmap(map, pagesize * 10), 0); +} + +static long get_smaps_anon_huge_pages(pid_t pid, void *addr) +{ + char smaps_path[64]; + char *line = NULL; + unsigned long start, end; + long anon_huge_kb; + size_t len; + FILE *f; + bool in_vma; + + in_vma = false; + sprintf(smaps_path, "/proc/%d/smaps", pid); + f = fopen(smaps_path, "r"); + if (!f) + return -1; + + while (getline(&line, &len, f) != -1) { + /* Check if the line describes a VMA range */ + if (sscanf(line, "%lx-%lx", &start, &end) == 2) { + if ((unsigned long)addr >= start && + (unsigned long)addr < end) + in_vma = true; + else + in_vma = false; + continue; + } + + /* If we are in the correct VMA, look for the AnonHugePages field */ + if (in_vma && + sscanf(line, "AnonHugePages: %ld kB", &anon_huge_kb) == 1) + break; + } + + free(line); + fclose(f); + + return (anon_huge_kb > 0) ? (anon_huge_kb * 1024) : 0; +} + +/** + * TEST_F(process_madvise, remote_collapse) + * + * This test deterministically validates process_madvise() with MADV_COLLAPSE + * on a remote process, other advices are difficult to verify reliably. + * + * The test verifies that a memory region in a child process, initially + * backed by small pages, can be collapsed into a Transparent Huge Page by a + * request from the parent. The result is verified by parsing the child's + * /proc/<pid>/smaps file. + */ +TEST_F(process_madvise, remote_collapse) +{ + const unsigned long pagesize = (unsigned long)sysconf(_SC_PAGESIZE); + pid_t child_pid; + int pidfd; + long huge_page_size; + int pipe_info[2]; + ssize_t ret; + struct iovec vec; + + struct child_info { + pid_t pid; + void *map_addr; + } info; + + huge_page_size = default_huge_page_size(); + if (huge_page_size <= 0) + ksft_exit_skip("Could not determine a valid huge page size.\n"); + + ASSERT_EQ(pipe(pipe_info), 0); + + child_pid = fork(); + ASSERT_NE(child_pid, -1); + + if (child_pid == 0) { + char *map; + size_t map_size = 2 * huge_page_size; + + close(pipe_info[0]); + + map = mmap(NULL, map_size, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + ASSERT_NE(map, MAP_FAILED); + + /* Fault in as small pages */ + for (size_t i = 0; i < map_size; i += pagesize) + map[i] = 'A'; + + /* Send info and pause */ + info.pid = getpid(); + info.map_addr = map; + ret = write(pipe_info[1], &info, sizeof(info)); + ASSERT_EQ(ret, sizeof(info)); + close(pipe_info[1]); + + pause(); + exit(0); + } + + close(pipe_info[1]); + + /* Receive child info */ + ret = read(pipe_info[0], &info, sizeof(info)); + if (ret <= 0) { + waitpid(child_pid, NULL, 0); + ksft_exit_skip("Failed to read child info from pipe.\n"); + } + ASSERT_EQ(ret, sizeof(info)); + close(pipe_info[0]); + child_pid = info.pid; + + pidfd = pidfd_open(child_pid, 0); + ASSERT_GE(pidfd, 0); + + /* Baseline Check from Parent's perspective */ + ASSERT_EQ(get_smaps_anon_huge_pages(child_pid, info.map_addr), 0); + + vec.iov_base = info.map_addr; + vec.iov_len = huge_page_size; + ret = sys_process_madvise(pidfd, &vec, 1, MADV_COLLAPSE, 0); + if (ret == -1) { + if (errno == EINVAL) + ksft_exit_skip( + "PROCESS_MADV_ADVISE is not supported.\n"); + else if (errno == EPERM) + ksft_exit_skip( + "No process_madvise() permissions, try running as root.\n"); + goto cleanup; + } + ASSERT_EQ(ret, huge_page_size); + + ASSERT_EQ(get_smaps_anon_huge_pages(child_pid, info.map_addr), + huge_page_size); + + ksft_test_result_pass( + "MADV_COLLAPSE successfully verified via smaps.\n"); + +cleanup: + /* Cleanup */ + kill(child_pid, SIGKILL); + waitpid(child_pid, NULL, 0); + if (pidfd >= 0) + close(pidfd); +} + +/* + * Test process_madvise() with various invalid pidfds to ensure correct error + * handling. This includes negative fds, non-pidfd fds, and pidfds for + * processes that no longer exist. + */ +TEST_F(process_madvise, invalid_pidfd) +{ + struct iovec vec; + pid_t child_pid; + ssize_t ret; + int pidfd; + + vec.iov_base = (void *)0x1234; + vec.iov_len = 4096; + + /* Using an invalid fd number (-1) should fail with EBADF. */ + ret = sys_process_madvise(-1, &vec, 1, MADV_DONTNEED, 0); + ASSERT_EQ(ret, -1); + ASSERT_EQ(errno, EBADF); + + /* + * Using a valid fd that is not a pidfd (e.g. stdin) should fail + * with EBADF. + */ + ret = sys_process_madvise(STDIN_FILENO, &vec, 1, MADV_DONTNEED, 0); + ASSERT_EQ(ret, -1); + ASSERT_EQ(errno, EBADF); + + /* + * Using a pidfd for a process that has already exited should fail + * with ESRCH. + */ + child_pid = fork(); + ASSERT_NE(child_pid, -1); + + if (child_pid == 0) + exit(0); + + pidfd = pidfd_open(child_pid, 0); + ASSERT_GE(pidfd, 0); + + /* Wait for the child to ensure it has terminated. */ + waitpid(child_pid, NULL, 0); + + ret = sys_process_madvise(pidfd, &vec, 1, MADV_DONTNEED, 0); + ASSERT_EQ(ret, -1); + ASSERT_EQ(errno, ESRCH); + close(pidfd); +} + +/* + * Test process_madvise() with an invalid flag value. Now we only support flag=0 + * future we will use it support sync so reserve this test. + */ +TEST_F(process_madvise, flag) +{ + const unsigned long pagesize = (unsigned long)sysconf(_SC_PAGESIZE); + unsigned int invalid_flag; + struct iovec vec; + char *map; + ssize_t ret; + + map = mmap(NULL, pagesize, PROT_READ, MAP_PRIVATE | MAP_ANONYMOUS, -1, + 0); + ASSERT_NE(map, MAP_FAILED); + + vec.iov_base = map; + vec.iov_len = pagesize; + + invalid_flag = 0x80000000; + + ret = sys_process_madvise(PIDFD_SELF, &vec, 1, MADV_DONTNEED, + invalid_flag); + ASSERT_EQ(ret, -1); + ASSERT_EQ(errno, EINVAL); + + /* Cleanup. */ + ASSERT_EQ(munmap(map, pagesize), 0); +} + +TEST_HARNESS_MAIN \ No newline at end of file diff --git a/tools/testing/selftests/mm/run_vmtests.sh b/tools/testing/selftests/mm/run_vmtests.sh index f96d43153fc0..5c28ebcf1ea9 100755 --- a/tools/testing/selftests/mm/run_vmtests.sh +++ b/tools/testing/selftests/mm/run_vmtests.sh @@ -61,6 +61,8 @@ separated by spaces: ksm tests that require >=2 NUMA nodes - pkey memory protection key tests +- process_madvise + test process_madvise - soft_dirty test soft dirty page bit semantics - pagemap @@ -424,6 +426,9 @@ CATEGORY="hmm" run_test bash ./test_hmm.sh smoke # MADV_GUARD_INSTALL and MADV_GUARD_REMOVE tests CATEGORY="madv_guard" run_test ./guard-regions +# PROCESS_MADVISE TEST +CATEGORY="process_madv" run_test ./process_madv + # MADV_DONTNEED and PROCESS_DONTNEED tests CATEGORY="madv_dontneed" run_test ./madv_dontneed -- 2.43.0

1 hour, 24 minutes

[PATCH net-next] selftests/tc-testing: Enable CONFIG_IP_SET

by Sebastian Andrzej Siewior

The config snippet specifies CONFIG_NET_EMATCH_IPSET. This option depends on CONFIG_IP_SET. Set CONFIG_IP_SET to be enabled at part for tc-testing. Signed-off-by: Sebastian Andrzej Siewior <bigeasy(a)linutronix.de> --- tools/testing/selftests/tc-testing/config | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/tc-testing/config b/tools/testing/selftests/tc-testing/config index db176fe7d0c3f..8e902f7f1a181 100644 --- a/tools/testing/selftests/tc-testing/config +++ b/tools/testing/selftests/tc-testing/config @@ -21,6 +21,7 @@ CONFIG_NF_NAT=m CONFIG_NETFILTER_XT_TARGET_LOG=m CONFIG_NET_SCHED=y +CONFIG_IP_SET=m # # Queueing/Scheduling -- 2.50.0

4 hours, 7 minutes

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-kselftest-mirror July 2025