- Linux-kselftest-mirror - lists.linaro.org

[PATCH v3] selftests: tmpfs: Add kselftest support to tmpfs

by Shivam Chaudhary

Add kselftest support for openat, linkat, unshare, mount tests - Replace direct error handling with 'ksft_test_result_*' , 'ksft_print_msg' macros for better reporting. - Add `ksft_print_header()` and `ksft_set_plan()` to structure test outputs more effectively. - Improve the test flow by adding more detailed pass/fail reporting for unshare, mounting, file opening, and linking operations. - Skip the test if it's not run as root, providing an appropriate Warning. Test logs: Before change: - Without root error: unshare, errno 1 - With root No, output After change: - Without root TAP version 13 1..1 ok 1 # SKIP This test needs root to run - With root TAP version 13 1..1 unshare(): Creat new mount namespace: Success. mount(): Root filesystem private mount: Success mount(): Mounting tmpfs on /tmp: Success openat(): Open first temporary file: Success linkat(): Linking the temporary file: Success openat(): Opening the second temporary file: Success ok 1 Test : Success Totals: pass:1 fail:0 xfail:0 xpass:0 skip:0 error:0 Signed-off-by: Shivam Chaudhary <cvam0000(a)gmail.com> --- Notes: Changes in v3: - Remove extra ksft_set_plan() - Remove function for unshare() - Fix the comment style link to v2: https://lore.kernel.org/all/20241026191621.2860376-1-cvam0000@gmail.com/ Changes in v2: - Make the commit message more clear. link to v1: https://lore.kernel.org/all/20241024200228.1075840-1-cvam0000@gmail.com/T/#u .../selftests/tmpfs/bug-link-o-tmpfile.c | 69 +++++++++++++++---- 1 file changed, 57 insertions(+), 12 deletions(-) diff --git a/tools/testing/selftests/tmpfs/bug-link-o-tmpfile.c b/tools/testing/selftests/tmpfs/bug-link-o-tmpfile.c index b5c3ddb90942..9ca1245784d9 100644 --- a/tools/testing/selftests/tmpfs/bug-link-o-tmpfile.c +++ b/tools/testing/selftests/tmpfs/bug-link-o-tmpfile.c @@ -23,45 +23,90 @@ #include <sys/mount.h> #include <unistd.h> +#include "../kselftest.h" + + int main(void) { int fd; + /* Setting up kselftest framework */ + ksft_print_header(); + ksft_set_plan(1); + + /* Check if test is run as root */ + if (geteuid()) { + ksft_test_result_skip("This test needs root to run!\n"); + return 1; + } + + if (unshare(CLONE_NEWNS) == -1) { if (errno == ENOSYS || errno == EPERM) { - fprintf(stderr, "error: unshare, errno %d\n", errno); - return 4; + ksft_test_result_fail("unshare() error: unshare, errno %d\n", errno); + return 1; } - fprintf(stderr, "error: unshare, errno %d\n", errno); - return 1; + else{ + fprintf(stderr, "unshare() error: unshare, errno %d\n", errno); + ksft_test_result_fail("unshare() error: unshare, errno %d\n", errno); + return 1; + + } + } + + else { + ksft_print_msg("unshare(): Creat new mount namespace: Success.\n"); + } - if (mount(NULL, "/", NULL, MS_PRIVATE|MS_REC, NULL) == -1) { - fprintf(stderr, "error: mount '/', errno %d\n", errno); + + + + if (mount(NULL, "/", NULL, MS_PRIVATE | MS_REC, NULL) == -1) { + ksft_test_result_fail("mount() error: Root filesystem private mount: Fail %d\n", errno); return 1; + } else { + ksft_print_msg("mount(): Root filesystem private mount: Success\n"); } + /* Our heroes: 1 root inode, 1 O_TMPFILE inode, 1 permanent inode. */ if (mount(NULL, "/tmp", "tmpfs", 0, "nr_inodes=3") == -1) { - fprintf(stderr, "error: mount tmpfs, errno %d\n", errno); + ksft_test_result_fail("mount() error: Mounting tmpfs on /tmp: Fail %d\n", errno); return 1; + } else { + ksft_print_msg("mount(): Mounting tmpfs on /tmp: Success\n"); } - fd = openat(AT_FDCWD, "/tmp", O_WRONLY|O_TMPFILE, 0600); + + fd = openat(AT_FDCWD, "/tmp", O_WRONLY | O_TMPFILE, 0600); if (fd == -1) { - fprintf(stderr, "error: open 1, errno %d\n", errno); + ksft_test_result_fail("openat() error: Open first temporary file: Fail %d\n", errno); return 1; + } else { + ksft_print_msg("openat(): Open first temporary file: Success\n"); } + + if (linkat(fd, "", AT_FDCWD, "/tmp/1", AT_EMPTY_PATH) == -1) { - fprintf(stderr, "error: linkat, errno %d\n", errno); + ksft_test_result_fail("linkat() error: Linking the temporary file: Fail %d\n", errno); + /* Ensure fd is closed on failure */ + close(fd); return 1; + } else { + ksft_print_msg("linkat(): Linking the temporary file: Success\n"); } close(fd); - fd = openat(AT_FDCWD, "/tmp", O_WRONLY|O_TMPFILE, 0600); + + fd = openat(AT_FDCWD, "/tmp", O_WRONLY | O_TMPFILE, 0600); if (fd == -1) { - fprintf(stderr, "error: open 2, errno %d\n", errno); + ksft_test_result_fail("openat() error: Opening the second temporary file: Fail %d\n", errno); return 1; + } else { + ksft_print_msg("openat(): Opening the second temporary file: Success\n"); } + ksft_test_result_pass("Test : Success\n"); + ksft_exit_pass(); return 0; } -- 2.45.2

10 months, 2 weeks

2
1
0 0

[PATCH v2] selftests/mount_setattr: fix idmap_mount_tree_invalid failed to run

by zhouyuhang

From: zhouyuhang <zhouyuhang(a)kylinos.cn> Test case idmap_mount_tree_invalid failed to run on the newer kernel with the following output: # RUN mount_setattr_idmapped.idmap_mount_tree_invalid ... # mount_setattr_test.c:1428:idmap_mount_tree_invalid:Expected sys_mount_setattr(open_tree_fd, "", AT_EMPTY_PATH, &attr, sizeof(attr)) (0) ! = 0 (0) # idmap_mount_tree_invalid: Test terminated by assertion This is because tmpfs is mounted at "/mnt/A", and tmpfs already contains the flag FS_ALLOW_IDMAP after the commit 7a80e5b8c6fa ("shmem: support idmapped mounts for tmpfs"). So calling sys_mount_setattr here returns 0 instead of -EINVAL as expected. Ramfs does not support idmap mounts, so we can use it here to test invalid mounts, which allows the test case to pass with the following output: # Starting 1 tests from 1 test cases. # RUN mount_setattr_idmapped.idmap_mount_tree_invalid ... # OK mount_setattr_idmapped.idmap_mount_tree_invalid ok 1 mount_setattr_idmapped.idmap_mount_tree_invalid # PASSED: 1 / 1 tests passed. Signed-off-by: zhouyuhang <zhouyuhang(a)kylinos.cn> --- .../testing/selftests/mount_setattr/mount_setattr_test.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/tools/testing/selftests/mount_setattr/mount_setattr_test.c b/tools/testing/selftests/mount_setattr/mount_setattr_test.c index c6a8c732b802..68801e1a9ec2 100644 --- a/tools/testing/selftests/mount_setattr/mount_setattr_test.c +++ b/tools/testing/selftests/mount_setattr/mount_setattr_test.c @@ -1414,6 +1414,13 @@ TEST_F(mount_setattr_idmapped, idmap_mount_tree_invalid) ASSERT_EQ(expected_uid_gid(-EBADF, "/tmp/B/b", 0, 0, 0), 0); ASSERT_EQ(expected_uid_gid(-EBADF, "/tmp/B/BB/b", 0, 0, 0), 0); + ASSERT_EQ(mount("testing", "/mnt/A", "ramfs", MS_NOATIME | MS_NODEV, + "size=100000,mode=700"), 0); + + ASSERT_EQ(mkdir("/mnt/A/AA", 0777), 0); + + ASSERT_EQ(mount("/tmp", "/mnt/A/AA", NULL, MS_BIND | MS_REC, NULL), 0); + open_tree_fd = sys_open_tree(-EBADF, "/mnt/A", AT_RECURSIVE | AT_EMPTY_PATH | @@ -1433,6 +1440,8 @@ TEST_F(mount_setattr_idmapped, idmap_mount_tree_invalid) ASSERT_EQ(expected_uid_gid(-EBADF, "/tmp/B/BB/b", 0, 0, 0), 0); ASSERT_EQ(expected_uid_gid(open_tree_fd, "B/b", 0, 0, 0), 0); ASSERT_EQ(expected_uid_gid(open_tree_fd, "B/BB/b", 0, 0, 0), 0); + + (void)umount2("/mnt/A", MNT_DETACH); } TEST_F(mount_setattr, mount_attr_nosymfollow) -- 2.27.0

10 months, 2 weeks

3
2
0 0

Re: Dear Respectable, I wait for your response to this my message

by Evan Bohdan

Good day, My name is Evan from Ukraine.I have tried to reach you but find it difficult due to internet scarcity here in this town. I am contacting you because I want to come over to your country,I have some huge funds I plan to invest in your country because I want to relocate my Late Father business due to ongoing war in my country Ukraine and visit your country to set up new investment business you may advice profitable in your area. I also have some millions of dollars belonging to my late father that i want to invest and 995kg of 24+carat 99.9% purity gold I want to ship to your country and establish gold jewelry manufacturing company. I will offer you good monetary rewards for your help,due to how russian drone bombards this town frequently,internet network is disrupted because of attacks on network installations,please for fast discussion, you can add me on whatsapp and let's talk faster as i want to leave this war zone as soon as possible,this is my whatsapp number below. +380-672575220 Email:evanbohdan@gmail.com I wait for your earliest response. Regards, Evan Whatsapp/Tel:+380-672575220 Email:evanbohdan@gmail.com

10 months, 2 weeks

1
0
0 0

[PATCH bpf-next] bpf: handle implicit declaration of function gettid in bpf_iter.c

by Jason Xing

From: Jason Xing <kernelxing(a)tencent.com> As we can see from the title, when I compiled the selftests/bpf, I saw the error: implicit declaration of function ‘gettid’ ; did you mean ‘getgid’? [-Werror=implicit-function-declaration] skel->bss->tid = gettid(); ^~~~~~ getgid Adding a define to fix it (referring to tools/perf/tests/shell/coresight/thread_loop/thread_loop.c file. Signed-off-by: Jason Xing <kernelxing(a)tencent.com> --- tools/testing/selftests/bpf/prog_tests/bpf_iter.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index f0a3a9c18e9e..a105759f3dcf 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -34,6 +34,8 @@ #include "bpf_iter_ksym.skel.h" #include "bpf_iter_sockmap.skel.h" +#define gettid() syscall(SYS_gettid) + static void test_btf_id_or_null(void) { struct bpf_iter_test_kern3 *skel; -- 2.37.3

10 months, 2 weeks

2
2
0 0

[PATCH 0/6] damon/{self,kunit}tests: minor fixups for DAMON debugfs interface tests

by SeongJae Park

Fixup small broken window panes in DAMON selftests and kunit tests. First four patches clean up DAMON debugfs interface selftests output, by fixing segmentation fault of a test program (patch 1), removing unnecessary debugging messages (patch 2), and hiding error messages from expected failures (patches 3 and 4). Following two patches fix copy-paste mistakes in DAMON Kconfig help message that copied from debugfs kunit test (patch 5) and a comment on the debugfs kunit test code (patch 6). Signed-off-by: SeongJae Park <sj(a)kernel.org> Andrew Paniakin (1): selftests/damon/huge_count_read_write: provide sufficiently large buffer for DEPRECATED file read SeongJae Park (5): selftests/damon/huge_count_read_write: remove unnecessary debugging message selftests/damon/_debugfs_common: hide expected error message from test_write_result() selftests/damon/debugfs_duplicate_context_creation: hide errors from expected file write failures mm/damon/Kconfig: update DBGFS_KUNIT prompt copy for SYSFS_KUNIT mm/damon/tests/dbgfs-kunit: fix the header double inclusion guarding ifdef comment mm/damon/Kconfig | 2 +- mm/damon/tests/dbgfs-kunit.h | 2 +- tools/testing/selftests/damon/_debugfs_common.sh | 7 ++++++- .../selftests/damon/debugfs_duplicate_context_creation.sh | 2 +- tools/testing/selftests/damon/huge_count_read_write.c | 4 +--- 5 files changed, 10 insertions(+), 7 deletions(-) base-commit: 13583c750117b4e10cdaf5578dcc7723b305ce4e -- 2.39.5

10 months, 2 weeks

1
5
0 0

[PATCH net-next v3] selftest/tcp-ao: Add filter tests

by Leo Stone

Add tests that check if getsockopt(TCP_AO_GET_KEYS) returns the right keys when using different filters. Sample output: > # ok 114 filter keys: by sndid, rcvid, address > # ok 115 filter keys: by is_current > # ok 116 filter keys: by is_rnext > # ok 117 filter keys: by sndid, rcvid > # ok 118 filter keys: correct nkeys when in.nkeys < matches Acked-by: Dmitry Safonov <0x7f454c46(a)gmail.com> Signed-off-by: Leo Stone <leocstone(a)gmail.com> --- v3: - Ordered locals in reverse xmas tree order - Separated socket fd declaration from assignment - Broke lines longer than 80 columns v2: https://lore.kernel.org/netdev/20241016055823.21299-1-leocstone@gmail.com/ - Changed 2 unnecessary test_error calls to test_fail - Added another test to make sure getsockopt returns the right nkeys value when the input nkeys is smaller than the number of matching keys - Removed the TODO that this patch addresses v1: https://lore.kernel.org/netdev/20241014213313.15100-1-leocstone@gmail.com/ Thanks to the reviewers for their time and feedback! --- .../selftests/net/tcp_ao/setsockopt-closed.c | 186 +++++++++++++++++- 1 file changed, 181 insertions(+), 5 deletions(-) diff --git a/tools/testing/selftests/net/tcp_ao/setsockopt-closed.c b/tools/testing/selftests/net/tcp_ao/setsockopt-closed.c index 084db4ecdff6..0abb9807d742 100644 --- a/tools/testing/selftests/net/tcp_ao/setsockopt-closed.c +++ b/tools/testing/selftests/net/tcp_ao/setsockopt-closed.c @@ -6,6 +6,8 @@ static union tcp_addr tcp_md5_client; +#define FILTER_TEST_NKEYS 16 + static int test_port = 7788; static void make_listen(int sk) { @@ -813,23 +815,197 @@ static void duplicate_tests(void) setsockopt_checked(sk, TCP_AO_ADD_KEY, &ao, EEXIST, "duplicate: SendID differs"); } +static void fetch_all_keys(int sk, struct tcp_ao_getsockopt *keys) +{ + socklen_t optlen = sizeof(struct tcp_ao_getsockopt); + + memset(keys, 0, sizeof(struct tcp_ao_getsockopt) * FILTER_TEST_NKEYS); + keys[0].get_all = 1; + keys[0].nkeys = FILTER_TEST_NKEYS; + if (getsockopt(sk, IPPROTO_TCP, TCP_AO_GET_KEYS, &keys[0], &optlen)) + test_error("getsockopt"); +} + +static int prepare_test_keys(struct tcp_ao_getsockopt *keys) +{ + const char *test_password = "Test password number "; + struct tcp_ao_add test_ao[FILTER_TEST_NKEYS]; + char test_password_scratch[64] = {}; + u8 rcvid = 100, sndid = 100; + int sk; + + sk = socket(test_family, SOCK_STREAM, IPPROTO_TCP); + if (sk < 0) + test_error("socket()"); + + for (int i = 0; i < FILTER_TEST_NKEYS; i++) { + snprintf(test_password_scratch, 64, "%s %d", test_password, i); + test_prepare_key(&test_ao[i], DEFAULT_TEST_ALGO, this_ip_dest, + false, false, DEFAULT_TEST_PREFIX, 0, sndid++, + rcvid++, 0, 0, strlen(test_password_scratch), + test_password_scratch); + } + test_ao[0].set_current = 1; + test_ao[1].set_rnext = 1; + /* One key with a different addr and overlapping sndid, rcvid */ + tcp_addr_to_sockaddr_in(&test_ao[2].addr, &this_ip_addr, 0); + test_ao[2].sndid = 100; + test_ao[2].rcvid = 100; + + /* Add keys in a random order */ + for (int i = 0; i < FILTER_TEST_NKEYS; i++) { + int randidx = rand() % (FILTER_TEST_NKEYS - i); + + if (setsockopt(sk, IPPROTO_TCP, TCP_AO_ADD_KEY, + &test_ao[randidx], sizeof(struct tcp_ao_add))) + test_error("setsockopt()"); + memcpy(&test_ao[randidx], &test_ao[FILTER_TEST_NKEYS - 1 - i], + sizeof(struct tcp_ao_add)); + } + + fetch_all_keys(sk, keys); + + return sk; +} + +/* Assumes passwords are unique */ +static int compare_mkts(struct tcp_ao_getsockopt *expected, int nexpected, + struct tcp_ao_getsockopt *actual, int nactual) +{ + int matches = 0; + + for (int i = 0; i < nexpected; i++) { + for (int j = 0; j < nactual; j++) { + if (memcmp(expected[i].key, actual[j].key, + TCP_AO_MAXKEYLEN) == 0) + matches++; + } + } + return nexpected - matches; +} + +static void filter_keys_checked(int sk, struct tcp_ao_getsockopt *filter, + struct tcp_ao_getsockopt *expected, + unsigned int nexpected, const char *tst) +{ + struct tcp_ao_getsockopt filtered_keys[FILTER_TEST_NKEYS] = {}; + struct tcp_ao_getsockopt all_keys[FILTER_TEST_NKEYS] = {}; + socklen_t len = sizeof(struct tcp_ao_getsockopt); + + fetch_all_keys(sk, all_keys); + memcpy(&filtered_keys[0], filter, sizeof(struct tcp_ao_getsockopt)); + filtered_keys[0].nkeys = FILTER_TEST_NKEYS; + if (getsockopt(sk, IPPROTO_TCP, TCP_AO_GET_KEYS, filtered_keys, &len)) + test_error("getsockopt"); + if (filtered_keys[0].nkeys != nexpected) { + test_fail("wrong nr of keys, expected %u got %u", nexpected, + filtered_keys[0].nkeys); + goto out_close; + } + if (compare_mkts(expected, nexpected, filtered_keys, + filtered_keys[0].nkeys)) { + test_fail("got wrong keys back"); + goto out_close; + } + test_ok("filter keys: %s", tst); + +out_close: + close(sk); + memset(filter, 0, sizeof(struct tcp_ao_getsockopt)); +} + +static void filter_tests(void) +{ + struct tcp_ao_getsockopt original_keys[FILTER_TEST_NKEYS]; + struct tcp_ao_getsockopt expected_keys[FILTER_TEST_NKEYS]; + struct tcp_ao_getsockopt filter = {}; + int sk, f, nmatches; + socklen_t len; + + f = 2; + sk = prepare_test_keys(original_keys); + filter.rcvid = original_keys[f].rcvid; + filter.sndid = original_keys[f].sndid; + memcpy(&filter.addr, &original_keys[f].addr, + sizeof(original_keys[f].addr)); + filter.prefix = original_keys[f].prefix; + filter_keys_checked(sk, &filter, &original_keys[f], 1, + "by sndid, rcvid, address"); + + f = -1; + sk = prepare_test_keys(original_keys); + for (int i = 0; i < original_keys[0].nkeys; i++) { + if (original_keys[i].is_current) { + f = i; + break; + } + } + if (f < 0) + test_error("No current key after adding one"); + filter.is_current = 1; + filter_keys_checked(sk, &filter, &original_keys[f], 1, "by is_current"); + + f = -1; + sk = prepare_test_keys(original_keys); + for (int i = 0; i < original_keys[0].nkeys; i++) { + if (original_keys[i].is_rnext) { + f = i; + break; + } + } + if (f < 0) + test_error("No rnext key after adding one"); + filter.is_rnext = 1; + filter_keys_checked(sk, &filter, &original_keys[f], 1, "by is_rnext"); + + f = -1; + nmatches = 0; + sk = prepare_test_keys(original_keys); + for (int i = 0; i < original_keys[0].nkeys; i++) { + if (original_keys[i].sndid == 100) { + f = i; + memcpy(&expected_keys[nmatches], &original_keys[i], + sizeof(struct tcp_ao_getsockopt)); + nmatches++; + } + } + if (f < 0) + test_error("No key for sndid 100"); + if (nmatches != 2) + test_error("Should have 2 keys with sndid 100"); + filter.rcvid = original_keys[f].rcvid; + filter.sndid = original_keys[f].sndid; + filter.addr.ss_family = test_family; + filter_keys_checked(sk, &filter, expected_keys, nmatches, + "by sndid, rcvid"); + + sk = prepare_test_keys(original_keys); + filter.get_all = 1; + filter.nkeys = FILTER_TEST_NKEYS / 2; + len = sizeof(struct tcp_ao_getsockopt); + if (getsockopt(sk, IPPROTO_TCP, TCP_AO_GET_KEYS, &filter, &len)) + test_error("getsockopt"); + if (filter.nkeys == FILTER_TEST_NKEYS) + test_ok("filter keys: correct nkeys when in.nkeys < matches"); + else + test_fail("filter keys: wrong nkeys, expected %u got %u", + FILTER_TEST_NKEYS, filter.nkeys); +} + static void *client_fn(void *arg) { if (inet_pton(TEST_FAMILY, __TEST_CLIENT_IP(2), &tcp_md5_client) != 1) test_error("Can't convert ip address"); extend_tests(); einval_tests(); + filter_tests(); duplicate_tests(); - /* - * TODO: check getsockopt(TCP_AO_GET_KEYS) with different filters - * returning proper nr & keys; - */ return NULL; } int main(int argc, char *argv[]) { - test_init(121, client_fn, NULL); + test_init(126, client_fn, NULL); return 0; } -- 2.43.0

10 months, 2 weeks

3
2
0 0

[PATCH net 0/3] mptcp: sched: fix some lock issues

by Matthieu Baerts (NGI0)

Two small fixes related to the MPTCP packets scheduler: - Patch 1: add missing rcu_read_(un)lock(). A fix for >= 6.6. - Patch 2: remove unneeded lock when listing packets schedulers. A fix for >= 6.10. And some modifications in the MPTCP selftests: - Patch 3: a small addition to the MPTCP selftests to cover more code. Signed-off-by: Matthieu Baerts (NGI0) <matttbe(a)kernel.org> --- Matthieu Baerts (NGI0) (3): mptcp: init: protect sched with rcu_read_lock mptcp: remove unneeded lock when listing scheds selftests: mptcp: list sysctl data net/mptcp/protocol.c | 2 ++ net/mptcp/sched.c | 2 -- tools/testing/selftests/net/mptcp/mptcp_connect.sh | 9 +++++++++ 3 files changed, 11 insertions(+), 2 deletions(-) --- base-commit: 3b05b9c36ddd01338e1352588f2ec1ea23f97d43 change-id: 20241021-net-mptcp-sched-lock-10dfc75d1e00 Best regards, -- Matthieu Baerts (NGI0) <matttbe(a)kernel.org>

10 months, 2 weeks

5
10
0 0

[PATCH v3 0/5] implement lightweight guard pages

by Lorenzo Stoakes

Userland library functions such as allocators and threading implementations often require regions of memory to act as 'guard pages' - mappings which, when accessed, result in a fatal signal being sent to the accessing process. The current means by which these are implemented is via a PROT_NONE mmap() mapping, which provides the required semantics however incur an overhead of a VMA for each such region. With a great many processes and threads, this can rapidly add up and incur a significant memory penalty. It also has the added problem of preventing merges that might otherwise be permitted. This series takes a different approach - an idea suggested by Vlasimil Babka (and before him David Hildenbrand and Jann Horn - perhaps more - the provenance becomes a little tricky to ascertain after this - please forgive any omissions!) - rather than locating the guard pages at the VMA layer, instead placing them in page tables mapping the required ranges. Early testing of the prototype version of this code suggests a 5 times speed up in memory mapping invocations (in conjunction with use of process_madvise()) and a 13% reduction in VMAs on an entirely idle android system and unoptimised code. We expect with optimisation and a loaded system with a larger number of guard pages this could significantly increase, but in any case these numbers are encouraging. This way, rather than having separate VMAs specifying which parts of a range are guard pages, instead we have a VMA spanning the entire range of memory a user is permitted to access and including ranges which are to be 'guarded'. After mapping this, a user can specify which parts of the range should result in a fatal signal when accessed. By restricting the ability to specify guard pages to memory mapped by existing VMAs, we can rely on the mappings being torn down when the mappings are ultimately unmapped and everything works simply as if the memory were not faulted in, from the point of view of the containing VMAs. This mechanism in effect poisons memory ranges similar to hardware memory poisoning, only it is an entirely software-controlled form of poisoning. The mechanism is implemented via madvise() behaviour - MADV_GUARD_INSTALL which installs page table-level guard page markers - and MADV_GUARD_REMOVE - which clears them. Guard markers can be installed across multiple VMAs and any existing mappings will be cleared, that is zapped, before installing the guard page markers in the page tables. There is no concept of 'nested' guard markers, multiple attempts to install guard markers in a range will, after the first attempt, have no effect. Importantly, removing guard markers over a range that contains both guard markers and ordinary backed memory has no effect on anything but the guard markers (including leaving huge pages un-split), so a user can safely remove guard markers over a range of memory leaving the rest intact. The actual mechanism by which the page table entries are specified makes use of existing logic - PTE markers, which are used for the userfaultfd UFFDIO_POISON mechanism. Unfortunately PTE_MARKER_POISONED is not suited for the guard page mechanism as it results in VM_FAULT_HWPOISON semantics in the fault handler, so we add our own specific PTE_MARKER_GUARD and adapt existing logic to handle it. We also extend the generic page walk mechanism to allow for installation of PTEs (carefully restricted to memory management logic only to prevent unwanted abuse). We ensure that zapping performed by MADV_DONTNEED and MADV_FREE do not remove guard markers, nor does forking (except when VM_WIPEONFORK is specified for a VMA which implies a total removal of memory characteristics). It's important to note that the guard page implementation is emphatically NOT a security feature, so a user can remove the markers if they wish. We simply implement it in such a way as to provide the least surprising behaviour. An extensive set of self-tests are provided which ensure behaviour is as expected and additionally self-documents expected behaviour of guard ranges. Suggested-by: Vlastimil Babka <vbabka(a)suse.cz> Suggested-by: Jann Horn <jannh(a)google.com> Suggested-by: David Hildenbrand <david(a)redhat.com> v3 * Cleaned up mm/pagewalk.c logic a bit to make things clearer, as suggested by Vlastiml. * Explicitly avoid splitting THP on PTE installation, as suggested by Vlastimil. Note this has no impact on the guard pages logic, which has page table entry handlers at PUD, PMD and PTE level. * Added WARN_ON_ONCE() to mm/hugetlb.c path where we don't expect a guard marker, as suggested by Vlastimil. * Reverted change to is_poisoned_swp_entry() to exclude guard pages which has the effect of MADV_FREE _not_ clearing guard pages. After discussion with Vlastimil, it became apparent that the ability to 'cancel' the freeing operation by writing to the mapping after having issued an MADV_FREE would mean that we would risk unexpected behaviour should the guard pages be removed, so we now do not remove markers here at all. * Added comment to PTE_MARKER_GUARD to highlight that memory tagged with the marker behaves as if it were a region mapped PROT_NONE, as highlighted by David. * Rename poison -> install, unpoison -> remove (i.e. MADV_GUARD_INSTALL / MADV_GUARD_REMOVE over MADV_GUARD_POISON / MADV_GUARD_REMOVE) at the request of David and John who both find the poison analogy confusing/overloaded. * After a lot of discussion, replace the looping behaviour should page faults race with guard page installation with a modest reattempt followed by returning -ERESTARTNOINTR to have the operation abort and re-enter, relieving lock contention and avoiding the possibility of allowing a malicious sandboxed process to impact the mmap lock or stall the overall process more than necessary, as suggested by Jann and Vlastimil having raised the issue. * Adjusted the page table walker so a populated huge PUD or PMD is correctly treated as being populated, necessitating a zap. In v2 we incorrectly skipped over these, which would cause the logic to wrongly proceed as if nothing were populated and the install succeeded. Instead, explicitly check to see if a huge page - if so, do not split but rather abort the operation and let zap take care of things. * Updated the guard remove logic to not unnecessarily split huge pages either. * Added a debug check to assert that the number of installed PTEs matches expectation, accounting for any existing guard pages. * Adapted vector_madvise() used by the process_madvise() system call to handle -ERESTARTNOINTR correctly. v2 * The macros in kselftest_harness.h seem to be broken - __EXPECT() is terminated by '} while (0); OPTIONAL_HANDLER(_assert)' meaning it is not safe in single line if / else or for /which blocks, however working around this results in checkpatch producing invalid warnings, as reported by Shuah. * Fixing these macros is out of scope for this series, so compromise and instead rewrite test blocks so as to use multiple lines by separating out a decl in most cases. This has the side effect of, for the most part, making things more readable. * Heavily document the use of the volatile keyword - we can't avoid checkpatch complaining about this, so we explain it, as reported by Shuah. * Updated commit message to highlight that we skip tests we lack permissions for, as reported by Shuah. * Replaced a perror() with ksft_exit_fail_perror(), as reported by Shuah. * Added user friendly messages to cases where tests are skipped due to lack of permissions, as reported by Shuah. * Update the tool header to include the new MADV_GUARD_POISON/UNPOISON defines and directly include asm-generic/mman.h to get the platform-neutral versions to ensure we import them. * Finally fixed Vlastimil's email address in Suggested-by tags from suze to suse, as reported by Vlastimil. * Added linux-api to cc list, as reported by Vlastimil. https://lore.kernel.org/all/cover.1729440856.git.lorenzo.stoakes@oracle.com/ v1 * Un-RFC'd as appears no major objections to approach but rather debate on implementation. * Fixed issue with arches which need mmu_context.h and tlbfush.h. header imports in pagewalker logic to be able to use update_mmu_cache() as reported by the kernel test bot. * Added comments in page walker logic to clarify who can use ops->install_pte and why as well as adding a check_ops_valid() helper function, as suggested by Christoph. * Pass false in full parameter in pte_clear_not_present_full() as suggested by Jann. * Stopped erroneously requiring a write lock for the poison operation as suggested by Jann and Suren. * Moved anon_vma_prepare() to the start of madvise_guard_poison() to be consistent with how this is used elsewhere in the kernel as suggested by Jann. * Avoid returning -EAGAIN if we are raced on page faults, just keep looping and duck out if a fatal signal is pending or a conditional reschedule is needed, as suggested by Jann. * Avoid needlessly splitting huge PUDs and PMDs by specifying ACTION_CONTINUE, as suggested by Jann. https://lore.kernel.org/all/cover.1729196871.git.lorenzo.stoakes@oracle.com/ RFC https://lore.kernel.org/all/cover.1727440966.git.lorenzo.stoakes@oracle.com/ Lorenzo Stoakes (5): mm: pagewalk: add the ability to install PTEs mm: add PTE_MARKER_GUARD PTE marker mm: madvise: implement lightweight guard page mechanism tools: testing: update tools UAPI header for mman-common.h selftests/mm: add self tests for guard page feature arch/alpha/include/uapi/asm/mman.h | 3 + arch/mips/include/uapi/asm/mman.h | 3 + arch/parisc/include/uapi/asm/mman.h | 3 + arch/xtensa/include/uapi/asm/mman.h | 3 + include/linux/mm_inline.h | 2 +- include/linux/pagewalk.h | 18 +- include/linux/swapops.h | 24 +- include/uapi/asm-generic/mman-common.h | 3 + mm/hugetlb.c | 4 + mm/internal.h | 12 + mm/madvise.c | 225 ++++ mm/memory.c | 18 +- mm/mprotect.c | 6 +- mm/mseal.c | 1 + mm/pagewalk.c | 227 +++- tools/include/uapi/asm-generic/mman-common.h | 3 + tools/testing/selftests/mm/.gitignore | 1 + tools/testing/selftests/mm/Makefile | 1 + tools/testing/selftests/mm/guard-pages.c | 1239 ++++++++++++++++++ 19 files changed, 1720 insertions(+), 76 deletions(-) create mode 100644 tools/testing/selftests/mm/guard-pages.c -- 2.47.0

10 months, 2 weeks

5
28
0 0

kselftest:arm64/FVP: arm64_check_buffer_fill - failed on Linux next-20241025

by Naresh Kamboju

The following kselftest arm64 and FVP failed with Linux next-20241025 on - Qemu-arm64 - FVP running Linux next-20241025 kernel. First seen on next-20241025 Good: next-20241024 BAD: next-20241025 kselftest-arm64, FVP * arm64_check_buffer_fill * arm64_check_mmap_options * arm64_check_child_memory Anyone have seen these failures ? Reported-by: Linux Kernel Functional Testing <lkft(a)linaro.org> Test log: ---------- # selftests: arm64: check_buffer_fill # 1..20 # ok 1 Check buffer correctness by byte with sync err mode and mmap memory # ok 2 Check buffer correctness by byte with async err mode and mmap memory # ok 3 Check buffer correctness by byte with sync err mode and mmap/mprotect memory # ok 4 Check buffer correctness by byte with async err mode and mmap/mprotect memory # ok 5 Check buffer write underflow by byte with sync mode and mmap memory # ok 6 Check buffer write underflow by byte with async mode and mmap memory # ok 7 Check buffer write underflow by byte with tag check fault ignore and mmap memory # ok 8 Check buffer write underflow by byte with sync mode and mmap memory # ok 9 Check buffer write underflow by byte with async mode and mmap memory # ok 10 Check buffer write underflow by byte with tag check fault ignore and mmap memory # ok 11 Check buffer write overflow by byte with sync mode and mmap memory # ok 12 Check buffer write overflow by byte with async mode and mmap memory # ok 13 Check buffer write overflow by byte with tag fault ignore mode and mmap memory # ok 14 Check buffer write correctness by block with sync mode and mmap memory # ok 15 Check buffer write correctness by block with async mode and mmap memory # ok 16 Check buffer write correctness by block with tag fault ignore and mmap memory # # FAIL: mmap allocation # # FAIL: memory allocation # not ok 17 Check initial tags with private mapping, sync error mode and mmap memory # ok 18 Check initial tags with private mapping, sync error mode and mmap/mprotect memory # # FAIL: mmap allocation # # FAIL: memory allocation # not ok 19 Check initial tags with shared mapping, sync error mode and mmap memory # ok 20 Check initial tags with shared mapping, sync error mode and mmap/mprotect memory # # Totals: pass:18 fail:2 xfail:0 xpass:0 skip:0 error:0 not ok 21 selftests: arm64: check_buffer_fill # exit=1 # timeout set to 45 # selftests: arm64: check_child_memory # 1..12 # ok 1 Check child anonymous memory with private mapping, precise mode and mmap memory # ok 2 Check child anonymous memory with shared mapping, precise mode and mmap memory # ok 3 Check child anonymous memory with private mapping, imprecise mode and mmap memory # ok 4 Check child anonymous memory with shared mapping, imprecise mode and mmap memory # ok 5 Check child anonymous memory with private mapping, precise mode and mmap/mprotect memory # ok 6 Check child anonymous memory with shared mapping, precise mode and mmap/mprotect memory # # FAIL: mmap allocation # # FAIL: memory allocation # not ok 7 Check child file memory with private mapping, precise mode and mmap memory # # FAIL: mmap allocation # # FAIL: memory allocation # not ok 8 Check child file memory with shared mapping, precise mode and mmap memory # ok 9 Check child file memory with private mapping, imprecise mode and mmap memory # ok 10 Check child file memory with shared mapping, imprecise mode and mmap memory # ok 11 Check child file memory with private mapping, precise mode and mmap/mprotect memory # ok 12 Check child file memory with shared mapping, precise mode and mmap/mprotect memory # # Totals: pass:10 fail:2 xfail:0 xpass:0 skip:0 error:0 not ok 22 selftests: arm64: check_child_memory # exit=1 boot Log links, -------- - https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20241028/te… - https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20241028/te… Test results history: ---------- - https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20241028/te… - https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20241028/te… metadata: ---- git describe: next-20241028 git repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next git sha: dec9255a128e19c5fcc3bdb18175d78094cc624d kernel config: https://storage.tuxsuite.com/public/linaro/lkft/builds/2o3tMqzOtHXYQjlvfR5t… build url: https://storage.tuxsuite.com/public/linaro/lkft/builds/2o3tMqzOtHXYQjlvfR5t… toolchain: gcc-13 Steps to reproduce: --------- - https://tuxapi.tuxsuite.com/v1/groups/linaro/projects/lkft/tests/2o3tON5kNi… - https://tuxapi.tuxsuite.com/v1/groups/linaro/projects/lkft/tests/2o3tON5kNi… -- Linaro LKFT https://lkft.linaro.org

10 months, 2 weeks

2
1
0 0

[PATCH v1 00/10] iommufd: Add VIOMMU infrastructure (Part-2 VIRQ)

by Nicolin Chen

As the part-2 of the VIOMMU infrastructure, this series introduces a VIRQ object after repurposing the existing FAULT object, which provides a nice notification pathway to the user space already. So, the first thing to do is reworking the FAULT object. Mimicing the HWPT structures, add a common EVENT structure to support its derivatives: EVENT_IOPF (the prior FAULT object) and EVENT_VIRQ (new one). IOMMUFD_CMD_VIRQ_ALLOC is introduced to allocate EVENT_VIRQ for a VIOMMU. One VIOMMU can have multiple VIRQs in different types but can not support multiple VIRQs with the same types. Drivers might need the VIOMMU's vdev_id list or the exact vdev_id link of the passthrough device's to forward IRQs/events via the VIOMMU framework. Thus, extend the set/unset_vdev_id ioctls down to the driver using VIOMMU ops. This allows drivers to take the control of a vdev_id's lifecycle. The forwarding part is fairly simple but might need to replace a physical device ID with a virtual device ID. So, there comes with some helpers for drivers to use. As usual, this series comes with the selftest coverage for this new VIRQ, and with a real world use case in the ARM SMMUv3 driver. This must be based on the VIOMMU Part-1 series. It's on Github: https://github.com/nicolinc/iommufd/commits/iommufd_virq-v1 Paring QEMU branch for testing: https://github.com/nicolinc/qemu/commits/wip/for_iommufd_virq-v1 Thanks! Nicolin Nicolin Chen (10): iommufd: Rename IOMMUFD_OBJ_FAULT to IOMMUFD_OBJ_EVENT_IOPF iommufd: Rename fault.c to event.c iommufd: Add IOMMUFD_OBJ_EVENT_VIRQ and IOMMUFD_CMD_VIRQ_ALLOC iommufd/viommu: Allow drivers to control vdev_id lifecycle iommufd/viommu: Add iommufd_vdev_id_to_dev helper iommufd/viommu: Add iommufd_viommu_report_irq helper iommufd/selftest: Implement mock_viommu_set/unset_vdev_id iommufd/selftest: Add IOMMU_TEST_OP_TRIGGER_VIRQ for VIRQ coverage iommufd/selftest: Add EVENT_VIRQ test coverage iommu/arm-smmu-v3: Report virtual IRQ for device in user space drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 109 +++- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 2 + drivers/iommu/iommufd/Makefile | 2 +- drivers/iommu/iommufd/device.c | 2 + drivers/iommu/iommufd/event.c | 613 ++++++++++++++++++ drivers/iommu/iommufd/fault.c | 443 ------------- drivers/iommu/iommufd/hw_pagetable.c | 12 +- drivers/iommu/iommufd/iommufd_private.h | 147 ++++- drivers/iommu/iommufd/iommufd_test.h | 10 + drivers/iommu/iommufd/main.c | 13 +- drivers/iommu/iommufd/selftest.c | 66 ++ drivers/iommu/iommufd/viommu.c | 25 +- drivers/iommu/iommufd/viommu_api.c | 54 ++ include/linux/iommufd.h | 28 + include/uapi/linux/iommufd.h | 46 ++ tools/testing/selftests/iommu/iommufd.c | 11 + tools/testing/selftests/iommu/iommufd_utils.h | 64 ++ 17 files changed, 1130 insertions(+), 517 deletions(-) create mode 100644 drivers/iommu/iommufd/event.c delete mode 100644 drivers/iommu/iommufd/fault.c -- 2.43.0

10 months, 2 weeks

2
16
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-kselftest-mirror