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

[PATCH v7] selftests/mm: add process_madvise() tests

by wang lian

Add tests for 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> Suggested-by: Mark Brown <broonie(a)kernel.org> Acked-by: SeongJae Park <sj(a)kernel.org> Reviewed-by: Zi Yan <ziy(a)nvidia.com> Tested-by: Zi Yan <ziy(a)nvidia.com> --- Changelog v7: - In the remote_collapse test, replace default_huge_page_size() with read_pmd_pagesize() - Add a new test, invalid_vlen, to verify that process_madvise() correctly fails with EINVAL when the vlen argument exceeds UIO_MAXIOV. Changelog v6: https://lore.kernel.org/lkml/20250721114614.40996-1-lianux.mm@gmail.com/ - Refactor child process and pidfd management to use the kselftest fixture's setup and teardown mechanism. This ensures that child processes are reliably terminated and file descriptors are closed, even when a test is aborted by an ASSERT or SKIP macro. This resolves the issue where a failed assertion could lead to a leaked child process. Changelog v5: https://lore.kernel.org/lkml/20250714122533.3135-1-lianux.mm@gmail.com/ - Refactor the remote_collapse test to concentrate on its primary goal confirming the successful remote invocation of process_madvise() on a child process. - Split the validation logic for invalid pidfds out of the remote test and into two new (`exited_process_pidfd` and `bad_pidfd`). - Based mm-new branch, can ensure clean application Changelog v4: https://lore.kernel.org/lkml/20250710112249.58722-1-lianux.mm@gmail.com/ - Refine resource cleanup logic in test teardown to be more robust. - Improve remote_collapse test to correctly handle different THP (Transparent Huge Page) policies ('always', 'madvise', 'never'), including handling race conditions with khugepaged. - Resolve build errors Changelog v3: https://lore.kernel.org/lkml/20250703044326.65061-1-lianux.mm@gmail.com/ - Rebased onto the latest mm-stable branch to ensure clean application. - Refactor common signal handling logic into vm_util to reduce code duplication. - Improve test robustness and diagnostics based on community feedback. - Address minor code style and script corrections. Changelog v2: https://lore.kernel.org/lkml/20250630140957.4000-1-lianux.mm@gmail.com/ - 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. -V1: https://lore.kernel.org/lkml/20250621133003.4733-1-lianux.mm@gmail.com/ tools/testing/selftests/mm/.gitignore | 1 + tools/testing/selftests/mm/Makefile | 1 + tools/testing/selftests/mm/process_madv.c | 344 ++++++++++++++++++++++ tools/testing/selftests/mm/run_vmtests.sh | 5 + 4 files changed, 351 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 f2dafa0b700b..e7b23a8a05fe 100644 --- a/tools/testing/selftests/mm/.gitignore +++ b/tools/testing/selftests/mm/.gitignore @@ -21,6 +21,7 @@ on-fault-limit transhuge-stress pagemap_ioctl pfnmap +process_madv *.tmp* protection_keys protection_keys_32 diff --git a/tools/testing/selftests/mm/Makefile b/tools/testing/selftests/mm/Makefile index ae6f994d3add..d13b3cef2a2b 100644 --- a/tools/testing/selftests/mm/Makefile +++ b/tools/testing/selftests/mm/Makefile @@ -85,6 +85,7 @@ 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..471cae8427f1 --- /dev/null +++ b/tools/testing/selftests/mm/process_madv.c @@ -0,0 +1,344 @@ +// 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 <linux/mman.h> +#include <sys/syscall.h> +#include <unistd.h> +#include <sched.h> +#include "vm_util.h" + +#include "../pidfd/pidfd.h" + +FIXTURE(process_madvise) +{ + unsigned long page_size; + pid_t child_pid; + int remote_pidfd; + int pidfd; +}; + +FIXTURE_SETUP(process_madvise) +{ + self->page_size = (unsigned long)sysconf(_SC_PAGESIZE); + self->pidfd = PIDFD_SELF; + self->remote_pidfd = -1; + self->child_pid = -1; +}; + +FIXTURE_TEARDOWN_PARENT(process_madvise) +{ + /* This teardown is guaranteed to run, even if tests SKIP or ASSERT */ + if (self->child_pid > 0) { + kill(self->child_pid, SIGKILL); + waitpid(self->child_pid, NULL, 0); + } + + if (self->remote_pidfd >= 0) + close(self->remote_pidfd); +} + +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); +} + +/* + * This test uses PIDFD_SELF to target the current process. The main + * goal is to verify the basic behavior of process_madvise() with + * a vector of non-contiguous memory ranges, not its cross-process + * capabilities. + */ +TEST_F(process_madvise, basic) +{ + const unsigned long pagesize = self->page_size; + const int madvise_pages = 4; + struct iovec vec[madvise_pages]; + int pidfd = self->pidfd; + ssize_t ret; + char *map; + + /* + * 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); + if (map == MAP_FAILED) + SKIP(return, "mmap failed, not enough memory.\n"); + + /* 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, vec, madvise_pages, MADV_DONTNEED, 0); + if (ret == -1 && errno == EPERM) + SKIP(return, + "process_madvise() unsupported or permission denied, try running as root.\n"); + else if (errno == EINVAL) + SKIP(return, + "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; + + /* 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]; + + for (int i = 0; i < pagesize; i++) + ASSERT_EQ(unadvised_page[i], 'A'); + + /* Cleanup. */ + ASSERT_EQ(munmap(map, pagesize * 10), 0); +} + +/* + * 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, + * focus on process_madv remote result, only check addresses and lengths. + * The correctness of the MADV_COLLAPSE can be found in the relevant test examples in khugepaged. + */ +TEST_F(process_madvise, remote_collapse) +{ + const unsigned long pagesize = self->page_size; + 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 = read_pmd_pagesize(); + if (huge_page_size <= 0) + SKIP(return, "Could not determine a valid huge page size.\n"); + + ASSERT_EQ(pipe(pipe_info), 0); + + self->child_pid = fork(); + ASSERT_NE(self->child_pid, -1); + + if (self->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(self->child_pid, NULL, 0); + SKIP(return, "Failed to read child info from pipe.\n"); + } + ASSERT_EQ(ret, sizeof(info)); + close(pipe_info[0]); + self->child_pid = info.pid; + + self->remote_pidfd = syscall(__NR_pidfd_open, self->child_pid, 0); + ASSERT_GE(self->remote_pidfd, 0); + + vec.iov_base = info.map_addr; + vec.iov_len = huge_page_size; + + ret = sys_process_madvise(self->remote_pidfd, &vec, 1, MADV_COLLAPSE, + 0); + if (ret == -1) { + if (errno == EINVAL) + SKIP(return, "PROCESS_MADV_ADVISE is not supported.\n"); + else if (errno == EPERM) + SKIP(return, + "No process_madvise() permissions, try running as root.\n"); + return; + } + + ASSERT_EQ(ret, huge_page_size); +} + +/* + * Test process_madvise() with a pidfd for a process that has already + * exited to ensure correct error handling. + */ +TEST_F(process_madvise, exited_process_pidfd) +{ + const unsigned long pagesize = self->page_size; + struct iovec vec; + char *map; + ssize_t ret; + + map = mmap(NULL, pagesize, PROT_READ, MAP_PRIVATE | MAP_ANONYMOUS, -1, + 0); + if (map == MAP_FAILED) + SKIP(return, "mmap failed, not enough memory.\n"); + + vec.iov_base = map; + vec.iov_len = pagesize; + + /* + * Using a pidfd for a process that has already exited should fail + * with ESRCH. + */ + self->child_pid = fork(); + ASSERT_NE(self->child_pid, -1); + + if (self->child_pid == 0) + exit(0); + + self->remote_pidfd = syscall(__NR_pidfd_open, self->child_pid, 0); + ASSERT_GE(self->remote_pidfd, 0); + + /* Wait for the child to ensure it has terminated. */ + waitpid(self->child_pid, NULL, 0); + + ret = sys_process_madvise(self->remote_pidfd, &vec, 1, MADV_DONTNEED, + 0); + ASSERT_EQ(ret, -1); + ASSERT_EQ(errno, ESRCH); +} + +/* + * Test process_madvise() with bad pidfds to ensure correct error + * handling. + */ +TEST_F(process_madvise, bad_pidfd) +{ + const unsigned long pagesize = self->page_size; + struct iovec vec; + char *map; + ssize_t ret; + + map = mmap(NULL, pagesize, PROT_READ, MAP_PRIVATE | MAP_ANONYMOUS, -1, + 0); + if (map == MAP_FAILED) + SKIP(return, "mmap failed, not enough memory.\n"); + + vec.iov_base = map; + vec.iov_len = pagesize; + + /* 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); +} + +/* + * Test that process_madvise() rejects vlen > UIO_MAXIOV. + * The kernel should return -EINVAL when the number of iovecs exceeds 1024. + */ +TEST_F(process_madvise, invalid_vlen) +{ + const unsigned long pagesize = self->page_size; + int pidfd = self->pidfd; + struct iovec vec; + char *map; + ssize_t ret; + + map = mmap(NULL, pagesize, PROT_READ, MAP_PRIVATE | MAP_ANONYMOUS, -1, + 0); + if (map == MAP_FAILED) + SKIP(return, "mmap failed, not enough memory.\n"); + + vec.iov_base = map; + vec.iov_len = pagesize; + + ret = sys_process_madvise(pidfd, &vec, 1025, MADV_DONTNEED, 0); + ASSERT_EQ(ret, -1); + ASSERT_EQ(errno, EINVAL); + + /* Cleanup. */ + ASSERT_EQ(munmap(map, pagesize), 0); +} + +/* + * Test process_madvise() with an invalid flag value. Currently, only a flag + * value of 0 is supported. This test is reserved for the future, e.g., if + * synchronous flags are added. + */ +TEST_F(process_madvise, flag) +{ + const unsigned long pagesize = self->page_size; + unsigned int invalid_flag; + int pidfd = self->pidfd; + struct iovec vec; + char *map; + ssize_t ret; + + map = mmap(NULL, pagesize, PROT_READ, MAP_PRIVATE | MAP_ANONYMOUS, -1, + 0); + if (map == MAP_FAILED) + SKIP(return, "mmap failed, not enough memory.\n"); + + vec.iov_base = map; + vec.iov_len = pagesize; + + invalid_flag = 0x80000000; + + ret = sys_process_madvise(pidfd, &vec, 1, MADV_DONTNEED, invalid_flag); + ASSERT_EQ(ret, -1); + ASSERT_EQ(errno, EINVAL); + + /* Cleanup. */ + ASSERT_EQ(munmap(map, pagesize), 0); +} + +TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/mm/run_vmtests.sh b/tools/testing/selftests/mm/run_vmtests.sh index a38c984103ce..471e539d82b8 100755 --- a/tools/testing/selftests/mm/run_vmtests.sh +++ b/tools/testing/selftests/mm/run_vmtests.sh @@ -65,6 +65,8 @@ separated by spaces: test pagemap_scan IOCTL - pfnmap tests for VM_PFNMAP handling +- process_madv + test for process_madv - cow test copy-on-write semantics - thp @@ -425,6 +427,9 @@ CATEGORY="madv_guard" run_test ./guard-regions # MADV_POPULATE_READ and MADV_POPULATE_WRITE tests CATEGORY="madv_populate" run_test ./madv_populate +# PROCESS_MADV test +CATEGORY="process_madv" run_test ./process_madv + CATEGORY="vma_merge" run_test ./merge if [ -x ./memfd_secret ] -- 2.43.0

3 weeks, 4 days

1
0
0 0

[PATCH bpf-next v4 0/4] bpf: Show precise rejected function when attaching to __noreturn and deny list functions

by KaFai Wan

Show precise rejected function when attaching fexit/fmod_ret to __noreturn functions. Add log for attaching tracing programs to functions in deny list. Add selftest for attaching tracing programs to functions in deny list. Migrate fexit_noreturns case into tracing_failure test suite. changes: v4: - change tracing_deny case attaching function (Yonghong Song) - add Acked-by: Yafang Shao and Yonghong Song v3: - add tracing_deny case into existing files (Alexei) - migrate fexit_noreturns into tracing_failure - change SOB https://lore.kernel.org/bpf/20250722153434.20571-1-kafai.wan@linux.dev/ v2: - change verifier log message (Alexei) - add missing Suggested-by https://lore.kernel.org/bpf/20250714120408.1627128-1-mannkafai@gmail.com/ v1: https://lore.kernel.org/all/20250710162717.3808020-1-mannkafai@gmail.com/ --- KaFai Wan (4): bpf: Show precise rejected function when attaching fexit/fmod_ret to __noreturn functions bpf: Add log for attaching tracing programs to functions in deny list selftests/bpf: Add selftest for attaching tracing programs to functions in deny list selftests/bpf: Migrate fexit_noreturns case into tracing_failure test suite kernel/bpf/verifier.c | 5 +- .../bpf/prog_tests/fexit_noreturns.c | 9 ---- .../bpf/prog_tests/tracing_failure.c | 52 +++++++++++++++++++ .../selftests/bpf/progs/fexit_noreturns.c | 15 ------ .../selftests/bpf/progs/tracing_failure.c | 12 +++++ 5 files changed, 68 insertions(+), 25 deletions(-) delete mode 100644 tools/testing/selftests/bpf/prog_tests/fexit_noreturns.c delete mode 100644 tools/testing/selftests/bpf/progs/fexit_noreturns.c -- 2.43.0

3 weeks, 4 days

2
5
0 0

[PATCH] selftests: timers: improve adjtick output readability

by Vishal Parmar

Reformat the output of the `adjtick` test in tools/testing/selftests/timers/ to display results in a clean tabular format. Previously, the output was printed in a free-form manner like this: Each iteration takes about 15 seconds Estimating tick (act: 9000 usec, -100000 ppm): 9000 usec, -100000 ppm [OK] This format made it hard to visually compare values across iterations or parse results in scripts. The new output is aligned in a table with clearly labeled columns: Each iteration takes about 15 seconds --------------------------------------------------------------- | Requested (usec) | Expected (ppm) | Measured (ppm) | Result | |------------------|----------------|----------------|---------| | 9000 | -100000 | -100001 | [ OK ] | | 9250 | -75000 | -75000 | [ OK ] | ... --------------------------------------------------------------- This improves readability, consistency, and log usability for automated tooling. Signed-off-by: Vishal Parmar <vishistriker(a)gmail.com> --- tools/testing/selftests/timers/adjtick.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) diff --git a/tools/testing/selftests/timers/adjtick.c b/tools/testing/selftests/timers/adjtick.c index 777d9494b683..b6b3de04d6ae 100644 --- a/tools/testing/selftests/timers/adjtick.c +++ b/tools/testing/selftests/timers/adjtick.c @@ -128,18 +128,18 @@ int check_tick_adj(long tickval) sleep(1); ppm = ((long long)tickval * MILLION)/systick - MILLION; - printf("Estimating tick (act: %ld usec, %lld ppm): ", tickval, ppm); + printf(" | %-16ld | %-14lld |", tickval, ppm); eppm = get_ppm_drift(); - printf("%lld usec, %lld ppm", systick + (systick * eppm / MILLION), eppm); + printf(" %-14lld |", eppm); fflush(stdout); tx1.modes = 0; adjtimex(&tx1); if (tx1.offset || tx1.freq || tx1.tick != tickval) { - printf(" [ERROR]\n"); - printf("\tUnexpected adjtimex return values, make sure ntpd is not running.\n"); + printf(" [ERROR] |\n"); + printf(" Unexpected adjtimex return values, make sure ntpd is not running.\n"); return -1; } @@ -153,10 +153,10 @@ int check_tick_adj(long tickval) * room for interruptions during the measurement. */ if (llabs(eppm - ppm) > 100) { - printf(" [FAILED]\n"); + printf(" [FAILED]\n"); return -1; } - printf(" [OK]\n"); + printf(" [ OK ] |\n"); return 0; } @@ -175,7 +175,10 @@ int main(int argc, char **argv) return -1; } - printf("Each iteration takes about 15 seconds\n"); + printf("\n Each iteration takes about 15 seconds\n"); + printf(" ---------------------------------------------------------------\n"); + printf(" | Requested (usec) | Expected (ppm) | Measured (ppm) | Result |\n"); + printf(" |------------------|----------------|----------------|---------|\n"); systick = sysconf(_SC_CLK_TCK); systick = USEC_PER_SEC/sysconf(_SC_CLK_TCK); @@ -188,6 +191,7 @@ int main(int argc, char **argv) break; } } + printf(" ---------------------------------------------------------------\n"); /* Reset things to zero */ tx1.modes = ADJ_TICK; -- 2.39.5

3 weeks, 4 days

2
1
0 0

[PATCH v3 00/15] Consolidate iommu page table implementations (AMD)

by Jason Gunthorpe

[All the precursor patches are merged now and AMD/RISCV/VTD conversions are written] Currently each of the iommu page table formats duplicates all of the logic to maintain the page table and perform map/unmap/etc operations. There are several different versions of the algorithms between all the different formats. The io-pgtable system provides an interface to help isolate the page table code from the iommu driver, but doesn't provide tools to implement the common algorithms. This makes it very hard to improve the state of the pagetable code under the iommu domains as any proposed improvement needs to alter a large number of different driver code paths. Combined with a lack of software based testing this makes improvement in this area very hard. iommufd wants several new page table operations: - More efficient map/unmap operations, using iommufd's batching logic - unmap that returns the physical addresses into a batch as it progresses - cut that allows splitting areas so large pages can have holes poked in them dynamically (ie guestmemfd hitless shared/private transitions) - More agressive freeing of table memory to avoid waste - Fragmenting large pages so that dirty tracking can be more granular - Reassembling large pages so that VMs can run at full IO performance in migration/dirty tracking error flows - KHO integration for kernel live upgrade Together these are algorithmically complex enough to be a very significant task to go and implement in all the page table formats we support. Just the "server" focused drivers use almost all the formats (ARMv8 S1&S2 / x86 PAE / AMDv1 / VT-D SS / RISCV) Instead of doing the duplicated work, this series takes the first step to consolidate the algorithms into one places. In spirit it is similar to the work Christoph did a few years back to pull the redundant get_user_pages() implementations out of the arch code into core MM. This unlocked a great deal of improvement in that space in the following years. I would like to see the same benefit in iommu as well. My first RFC showed a bigger picture with all most all formats and more algorithms. This series reorganizes that to be narrowly focused on just enough to convert the AMD driver to use the new mechanism. kunit tests are provided that allow good testing of the algorithms and all formats on x86, nothing is arch specific. AMD is one of the simpler options as the HW is quite uniform with few different options/bugs while still requiring the complicated contiguous pages support. The HW also has a very simple range based invalidation approach that is easy to implement. The AMD v1 and AMD v2 page table formats are implemented bit for bit identical to the current code, tested using a compare kunit test that checks against the io-pgtable version (on github, see below). Updating the AMD driver to replace the io-pgtable layer with the new stuff is fairly straightforward now. The layering is fixed up in the new version so that all the invalidation goes through function pointers. Several small fixing patches have come out of this as I've been fixing the problems that the test suite uncovers in the current code, and implementing the fixed version in iommupt. On performance, there is a quite wide variety of implementation designs across all the drivers. Looking at some key performance across the main formats: iommu_map(): pgsz ,avg new,old ns, min new,old ns , min % (+ve is better) 2^12, 53,66 , 51,63 , 19.19 (AMDV1) 256*2^12, 386,1909 , 367,1795 , 79.79 256*2^21, 362,1633 , 355,1556 , 77.77 2^12, 56,62 , 52,59 , 11.11 (AMDv2) 256*2^12, 405,1355 , 357,1292 , 72.72 256*2^21, 393,1160 , 358,1114 , 67.67 2^12, 55,65 , 53,62 , 14.14 (VTD second stage) 256*2^12, 391,518 , 332,512 , 35.35 256*2^21, 383,635 , 336,624 , 46.46 2^12, 57,65 , 55,63 , 12.12 (ARM 64 bit) 256*2^12, 380,389 , 361,369 , 2.02 256*2^21, 358,419 , 345,400 , 13.13 iommu_unmap(): pgsz ,avg new,old ns, min new,old ns , min % (+ve is better) 2^12, 69,88 , 65,85 , 23.23 (AMDv1) 256*2^12, 353,6498 , 331,6029 , 94.94 256*2^21, 373,6014 , 360,5706 , 93.93 2^12, 71,72 , 66,69 , 4.04 (AMDv2) 256*2^12, 228,891 , 206,871 , 76.76 256*2^21, 254,721 , 245,711 , 65.65 2^12, 69,87 , 65,82 , 20.20 (VTD second stage) 256*2^12, 210,321 , 200,315 , 36.36 256*2^21, 255,349 , 238,342 , 30.30 2^12, 72,77 , 68,74 , 8.08 (ARM 64 bit) 256*2^12, 521,357 , 447,346 , -29.29 256*2^21, 489,358 , 433,345 , -25.25 * Above numbers include additional patches to remove the iommu_pgsize() overheads. gcc 13.3.0, i7-12700 This version provides fairly consistent performance across formats. ARM unmap performance is quite different because this version supports contiguous pages and uses a very different algorithm for unmapping. Though why it is so worse compared to AMDv1 I haven't figured out yet. The per-format commits include a more detailed chart. There is a second branch: https://github.com/jgunthorpe/linux/commits/iommu_pt_all Containing supporting work and future steps: - ARM short descriptor (32 bit), ARM long descriptor (64 bit) formats - RISCV format and RISCV conversion https://github.com/jgunthorpe/linux/commits/iommu_pt_riscv - Support for a DMA incoherent HW page table walker - VT-D second stage format and VT-D conversion https://github.com/jgunthorpe/linux/commits/iommu_pt_vtd - DART v1 & v2 format - Draft of a iommufd 'cut' operation to break down huge pages - A compare test that checks the iommupt formats against the iopgtable interface, including updating AMD to have a working iopgtable and patches to make VT-D have an iopgtable for testing. - A performance test to micro-benchmark map and unmap against iogptable My strategy is to go one by one for the drivers: - AMD driver conversion - RISCV page table and driver - Intel VT-D driver and VTDSS page table - Flushing improvements for RISCV - ARM SMMUv3 And concurrently work on the algorithm side: - debugfs content dump, like VT-D has - Cut support - Increase/Decrease page size support - map/unmap batching - KHO As we make more algorithm improvements the value to convert the drivers increases. This is on github: https://github.com/jgunthorpe/linux/commits/iommu_pt v2: - Rebase on v6.16-rc2 - s/PT_ENTRY_WORD_SIZE/PT_ITEM_WORD_SIZE/s to follow the language better - Comment and documentation updates - Add PT_TOP_PHYS_MASK to help manage alignment restrictions on the top pointer - Add missed force_aperture = true - Make pt_iommu_deinit() take care of the not-yet-inited error case internally as AMD/RISCV/VTD all shared this logic - Change gather_range() into gather_range_pages() so it also deals with the page list. This makes the following cache flushing series simpler - Fix missed update of unmap->unmapped in some error cases - Change clear_contig() to order the gather more logically - Remove goto from the error handling in __map_range_leaf() - s/log2_/oalog2_/ in places where the argument is an oaddr_t - Pass the pts to pt_table_install64/32() - Do not use SIGN_EXTEND for the AMDv2 page table because of Vasant's information on how PASID 0 works. v1: https://patch.msgid.link/r/0-v2-5c26bde5c22d+58b-iommu_pt_jgg@nvidia.com - AMD driver only, many code changes RFC: https://lore.kernel.org/all/0-v1-01fa10580981+1d-iommu_pt_jgg@nvidia.com/ Alejandro Jimenez (1): iommu/amd: Use the generic iommu page table Jason Gunthorpe (14): genpt: Generic Page Table base API genpt: Add Documentation/ files iommupt: Add the basic structure of the iommu implementation iommupt: Add the AMD IOMMU v1 page table format iommupt: Add iova_to_phys op iommupt: Add unmap_pages op iommupt: Add map_pages op iommupt: Add read_and_clear_dirty op iommupt: Add a kunit test for Generic Page Table iommupt: Add a mock pagetable format for iommufd selftest to use iommufd: Change the selftest to use iommupt instead of xarray iommupt: Add the x86 64 bit page table format iommu/amd: Remove AMD io_pgtable support iommupt: Add a kunit test for the IOMMU implementation .clang-format | 1 + Documentation/driver-api/generic_pt.rst | 140 ++ Documentation/driver-api/index.rst | 1 + drivers/iommu/Kconfig | 2 + drivers/iommu/Makefile | 1 + drivers/iommu/amd/Kconfig | 5 +- drivers/iommu/amd/Makefile | 2 +- drivers/iommu/amd/amd_iommu.h | 1 - drivers/iommu/amd/amd_iommu_types.h | 109 +- drivers/iommu/amd/io_pgtable.c | 560 -------- drivers/iommu/amd/io_pgtable_v2.c | 370 ------ drivers/iommu/amd/iommu.c | 516 ++++---- drivers/iommu/generic_pt/.kunitconfig | 13 + drivers/iommu/generic_pt/Kconfig | 72 ++ drivers/iommu/generic_pt/fmt/Makefile | 26 + drivers/iommu/generic_pt/fmt/amdv1.h | 409 ++++++ drivers/iommu/generic_pt/fmt/defs_amdv1.h | 21 + drivers/iommu/generic_pt/fmt/defs_x86_64.h | 21 + drivers/iommu/generic_pt/fmt/iommu_amdv1.c | 15 + drivers/iommu/generic_pt/fmt/iommu_mock.c | 10 + drivers/iommu/generic_pt/fmt/iommu_template.h | 48 + drivers/iommu/generic_pt/fmt/iommu_x86_64.c | 11 + drivers/iommu/generic_pt/fmt/x86_64.h | 248 ++++ drivers/iommu/generic_pt/iommu_pt.h | 1150 +++++++++++++++++ drivers/iommu/generic_pt/kunit_generic_pt.h | 717 ++++++++++ drivers/iommu/generic_pt/kunit_iommu.h | 183 +++ drivers/iommu/generic_pt/kunit_iommu_pt.h | 451 +++++++ drivers/iommu/generic_pt/pt_common.h | 354 +++++ drivers/iommu/generic_pt/pt_defs.h | 323 +++++ drivers/iommu/generic_pt/pt_fmt_defaults.h | 193 +++ drivers/iommu/generic_pt/pt_iter.h | 640 +++++++++ drivers/iommu/generic_pt/pt_log2.h | 130 ++ drivers/iommu/io-pgtable.c | 4 - drivers/iommu/iommufd/Kconfig | 1 + drivers/iommu/iommufd/iommufd_test.h | 11 +- drivers/iommu/iommufd/selftest.c | 439 +++---- include/linux/generic_pt/common.h | 166 +++ include/linux/generic_pt/iommu.h | 270 ++++ include/linux/io-pgtable.h | 2 - tools/testing/selftests/iommu/iommufd.c | 60 +- tools/testing/selftests/iommu/iommufd_utils.h | 12 + 41 files changed, 6119 insertions(+), 1589 deletions(-) create mode 100644 Documentation/driver-api/generic_pt.rst delete mode 100644 drivers/iommu/amd/io_pgtable.c delete mode 100644 drivers/iommu/amd/io_pgtable_v2.c create mode 100644 drivers/iommu/generic_pt/.kunitconfig create mode 100644 drivers/iommu/generic_pt/Kconfig create mode 100644 drivers/iommu/generic_pt/fmt/Makefile create mode 100644 drivers/iommu/generic_pt/fmt/amdv1.h create mode 100644 drivers/iommu/generic_pt/fmt/defs_amdv1.h create mode 100644 drivers/iommu/generic_pt/fmt/defs_x86_64.h create mode 100644 drivers/iommu/generic_pt/fmt/iommu_amdv1.c create mode 100644 drivers/iommu/generic_pt/fmt/iommu_mock.c create mode 100644 drivers/iommu/generic_pt/fmt/iommu_template.h create mode 100644 drivers/iommu/generic_pt/fmt/iommu_x86_64.c create mode 100644 drivers/iommu/generic_pt/fmt/x86_64.h create mode 100644 drivers/iommu/generic_pt/iommu_pt.h create mode 100644 drivers/iommu/generic_pt/kunit_generic_pt.h create mode 100644 drivers/iommu/generic_pt/kunit_iommu.h create mode 100644 drivers/iommu/generic_pt/kunit_iommu_pt.h create mode 100644 drivers/iommu/generic_pt/pt_common.h create mode 100644 drivers/iommu/generic_pt/pt_defs.h create mode 100644 drivers/iommu/generic_pt/pt_fmt_defaults.h create mode 100644 drivers/iommu/generic_pt/pt_iter.h create mode 100644 drivers/iommu/generic_pt/pt_log2.h create mode 100644 include/linux/generic_pt/common.h create mode 100644 include/linux/generic_pt/iommu.h base-commit: cd76b0248a38645a3e3f8ca4a48bffc591e9da19 -- 2.43.0

3 weeks, 5 days

4
24
0 0

[PATCH v14 net-next 00/14] AccECN protocol patch series

by chia-yu.chang＠nokia-bell-labs.com

From: Chia-Yu Chang <chia-yu.chang(a)nokia-bell-labs.com> Hello, Please find the v14 AccECN protocol patch series, which covers the core functionality of Accurate ECN, AccECN negotiation, AccECN TCP options, and AccECN failure handling. The Accurate ECN draft can be found in https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-28 This patch series is part of the full AccECN patch series, which is available at https://github.com/L4STeam/linux-net-next/commits/upstream_l4steam/ Best Regards, Chia-Yu --- v14 (22-Jul-2025) - Add missing const for struct tcp_sock of tcp_accecn_option_beacon_check() of #11 (Simon Horman <horms(a)kernel.org>) v13 (18-Jul-2025) - Implement tcp_accecn_extract_syn_ect() and tcp_accecn_reflector_flags() with static array lookup of patch #6 (Paolo Abeni <pabeni(a)redhat.com>) - Fix typos in comments of #6 and remove patch #7 of v12 about simulatenous connect (Paolo Abeni <pabeni(a)redhat.com>) - Move TCP_ACCECN_E1B_INIT_OFFSET, TCP_ACCECN_E0B_INIT_OFFSET, and TCP_ACCECN_CEB_INIT_OFFSET from patch #7 to #11 (Paolo Abeni <pabeni(a)redhat.com>) - Use static array lookup in tcp_accecn_optfield_to_ecnfield() of patch #11 (Paolo Abeni <pabeni(a)redhat.com>) - Return false when WARN_ON_ONCE() is true in tcp_accecn_process_option() of patch #11 (Paolo Abeni <pabeni(a)redhat.com>) - Make synack_ecn_bytes as static const array and use const u32 pointer in tcp_options_write() of #11 (Paolo Abeni <pabeni(a)redhat.com>) - Use ALIGN() and ALIGN_DOWN() in tcp_options_fit_accecn() to pad TCP AccECN option to dword of #11 (Paolo Abeni <pabeni(a)redhat.com>) - Return TCP_ACCECN_OPT_FAIL_SEEN if WARN_ON_ONCE() is true in tcp_accecn_option_init() of #12 (Paolo Abeni <pabeni(a)redhat.com>) v12 (04-Jul-2025) - Fix compilation issues with some intermediate patches in v11 - Add more comments for AccECN helpers of tcp_ecn.h v11 (03-Jul-2025) - Fix compilation issues with some intermediate patches in v10 v10 (02-Jul-2025) - Add new patch of separated header file include/net/tcp_ecn.h to include ECN and AccECN functions (Eric Dumazet <edumazet(a)google.com>) - Add comments on the AccECN helper functions in tcp_ecn.h (Eric Dumazet <edumazet(a)google.com>) - Add documentation of tcp_ecn, tcp_ecn_option, tcp_ecn_beacon in ip-sysctl.rst to the corresponding patch (Eric Dumazet <edumazet(a)google.com>) - Split wait third ACK functionality into a separated patch from AccECN negotiation patch (Eric Dumazet <edumazet(a)google.com>) - Add READ_ONCE() over every reads of sysctl for all patches in the series (Eric Dumazet <edumazet(a)google.com>) - Merge heuristics of AccECN option ceb/cep and ACE field multi-wrap into a single patch - Add a table of SACK block reduction and required AccECN field in patch #15 commit message (Eric Dumazet <edumazet(a)google.com>) v9 (21-Jun-2025) - Use tcp_data_ecn_check() to set TCP_ECN_SEE flag only for RFC3168 ECN (Paolo Abeni <pabeni(a)redhat.com>) - Add comments about setting TCP_ECN_SEEN flag for RFC3168 and Accruate ECN (Paolo Abeni <pabeni(a)redhat.com>) - Restruct the code in the for loop of tcp_accecn_process_option() (Paolo Abeni <pabeni(a)redhat.com>) - Remove ecn_bytes and add use_synack_ecn_bytes flag to identify whether syn_ack_bytes or received_ecn_bytes is used (Paolo Abeni <pabeni(a)redhat.com>) - Replace leftover_bytes and leftover_size with leftover_highbyte and leftover_lowbyte and add comments in tcp_options_write() (Paolo Abeni <pabeni(a)redhat.com>) - Add comments and commit message about the 1st retx SYN still attempt AccECN negotiation (Paolo Abeni <pabeni(a)redhat.com>) v8 (10-Jun-2025) - Add new helper function tcp_ecn_received_counters_payload() in #6 (Paolo Abeni <pabeni(a)redhat.com>) - Set opts->num_sack_blocks=0 to avoid potential undefined value in #8 (Paolo Abeni <pabeni(a)redhat.com>) - Reset leftover_size to 2 once leftover_bytes is used in #9 (Paolo Abeni <pabeni(a)redhat.com>) - Add new helper function tcp_accecn_opt_demand_min() in #10 (Paolo Abeni <pabeni(a)redhat.com>) - Add new helper function tcp_accecn_saw_opt_fail_recv() in #11 (Paolo Abeni <pabeni(a)redhat.com>) - Update tcp_options_fit_accecn() to avoid using recursion in #14 (Paolo Abeni <pabeni(a)redhat.com>) v7 (14-May-2025) - Modify group sizes of tcp_sock_write_txrx and tcp_sock_write_rx in #3 based on pahole results (Paolo Abeni <pabeni(a)redhat.com>) - Fix the issue in #4 and #5 where the RFC3168 ECN behavior in tcp_ecn_send() is changed (Paolo Abeni <pabeni(a)redhat.com>) - Modify group size of tcp_sock_write_txrx in #4 and #6 based on pahole results (Paolo Abeni <pabeni(a)redhat.com>) - Update commit message for #9 to explain the increase in tcp_sock_write_rx group size - Modify group size of tcp_sock_write_tx in #10 based on pahole results v6 (09-May-2025) - Add #3 to utilize exisintg holes of tcp_sock_write_txrx group for later patches (#4, #9, #10) with new u8 members (Paolo Abeni <pabeni(a)redhat.com>) - Add pahole outcomes before and after commit in #4, #5, #6, #9, #10, #15 (Paolo Abeni <pabeni(a)redhat.com>) - Define new helper function tcp_send_ack_reflect_ect() for sending ACK with reflected ECT in #5 (Paolo Abeni <pabeni(a)redhat.com>) - Add comments for function tcp_ecn_rcv_synack() in #5 (Paolo Abeni <pabeni(a)redhat.com>) - Add enum/define to be used by sysctl_tcp_ecn in #5, sysctl_tcp_ecn_option in #9, and sysctl_tcp_ecn_option_beacon in #10 (Paolo Abeni <pabeni(a)redhat.com>) - Move accecn_fail_mode and saw_accecn_opt in #5 and #11 to use exisintg holes of tcp_sock (Paolo Abeni <pabeni(a)redhat.com>) - Change data type of new members of tcp_request_sock and move them to the end of struct in #5 and #11 (Paolo Abeni <pabeni(a)redhat.com>) - Move new members of tcp_info to the end of struct in #6 (Paolo Abeni <pabeni(a)redhat.com>) - Merge previous #7 into #9 (Paolo Abeni <pabeni(a)redhat.com>) - Mask ecnfield with INET_ECN_MASK to remove WARN_ONCE in #9 (Paolo Abeni <pabeni(a)redhat.com>) - Reduce the indentation levels for reabability in #9 and #10 (Paolo Abeni <pabeni(a)redhat.com>) - Move delivered_ecn_bytes to the RX group in #9, accecn_opt_tstamp to the TX group in #10, pkts_acked_ewma to the RX group in #15 (Paolo Abeni <pabeni(a)redhat.com>) - Add changes in Documentation/networking/net_cachelines/tcp_sock.rst for new tcp_sock members in #3, #5, #6, #9, #10, #15 v5 (22-Apr-2025) - Further fix for 32-bit ARM alignment in tcp.c (Simon Horman <horms(a)kernel.org>) v4 (18-Apr-2025) - Fix 32-bit ARM assertion for alignment requirement (Simon Horman <horms(a)kernel.org>) v3 (14-Apr-2025) - Fix patch apply issue in v2 (Jakub Kicinski <kuba(a)kernel.org>) v2 (18-Mar-2025) - Add one missing patch from the previous AccECN protocol preparation patch series to this patch series. --- Chia-Yu Chang (5): tcp: reorganize tcp_sock_write_txrx group for variables later tcp: ecn functions in separated include file tcp: accecn: AccECN option send control tcp: accecn: AccECN option failure handling tcp: accecn: try to fit AccECN option with SACK Ilpo Järvinen (9): tcp: reorganize SYN ECN code tcp: fast path functions later tcp: AccECN core tcp: accecn: AccECN negotiation tcp: accecn: add AccECN rx byte counters tcp: accecn: AccECN needs to know delivered bytes tcp: sack option handling improvements tcp: accecn: AccECN option tcp: accecn: AccECN option ceb/cep and ACE field multi-wrap heuristics Documentation/networking/ip-sysctl.rst | 55 +- .../networking/net_cachelines/tcp_sock.rst | 12 + include/linux/tcp.h | 32 +- include/net/netns/ipv4.h | 2 + include/net/tcp.h | 87 ++- include/net/tcp_ecn.h | 649 ++++++++++++++++++ include/uapi/linux/tcp.h | 7 + net/ipv4/syncookies.c | 4 + net/ipv4/sysctl_net_ipv4.c | 19 + net/ipv4/tcp.c | 28 +- net/ipv4/tcp_input.c | 353 ++++++++-- net/ipv4/tcp_ipv4.c | 8 +- net/ipv4/tcp_minisocks.c | 40 +- net/ipv4/tcp_output.c | 294 ++++++-- net/ipv6/syncookies.c | 2 + net/ipv6/tcp_ipv6.c | 1 + 16 files changed, 1409 insertions(+), 184 deletions(-) create mode 100644 include/net/tcp_ecn.h -- 2.34.1

3 weeks, 5 days

3
16
0 0

[PATCH v20 4/5] binder: add transaction_report feature entry

by Carlos Llamas

From: Li Li <dualli(a)google.com> Add "transaction_report" to the binderfs feature list, to help userspace determine if the "BINDER_CMD_REPORT" generic netlink api is supported by the binder driver. Signed-off-by: Li Li <dualli(a)google.com> Signed-off-by: Carlos Llamas <cmllamas(a)google.com> --- drivers/android/binderfs.c | 8 ++++++++ .../selftests/filesystems/binderfs/binderfs_test.c | 1 + 2 files changed, 9 insertions(+) diff --git a/drivers/android/binderfs.c b/drivers/android/binderfs.c index 4f827152d18e..f74a7e380261 100644 --- a/drivers/android/binderfs.c +++ b/drivers/android/binderfs.c @@ -59,6 +59,7 @@ struct binder_features { bool oneway_spam_detection; bool extended_error; bool freeze_notification; + bool transaction_report; }; static const struct constant_table binderfs_param_stats[] = { @@ -76,6 +77,7 @@ static struct binder_features binder_features = { .oneway_spam_detection = true, .extended_error = true, .freeze_notification = true, + .transaction_report = true, }; static inline struct binderfs_info *BINDERFS_SB(const struct super_block *sb) @@ -616,6 +618,12 @@ static int init_binder_features(struct super_block *sb) if (IS_ERR(dentry)) return PTR_ERR(dentry); + dentry = binderfs_create_file(dir, "transaction_report", + &binder_features_fops, + &binder_features.transaction_report); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + return 0; } diff --git a/tools/testing/selftests/filesystems/binderfs/binderfs_test.c b/tools/testing/selftests/filesystems/binderfs/binderfs_test.c index 81db85a5cc16..39a68078a79b 100644 --- a/tools/testing/selftests/filesystems/binderfs/binderfs_test.c +++ b/tools/testing/selftests/filesystems/binderfs/binderfs_test.c @@ -65,6 +65,7 @@ static int __do_binderfs_test(struct __test_metadata *_metadata) "oneway_spam_detection", "extended_error", "freeze_notification", + "transaction_report", }; change_mountns(_metadata); -- 2.50.1.470.g6ba607880d-goog

3 weeks, 6 days

1
0
0 0

[PATCH net-next] selftests: bpf: fix legacy netfilter options

by Jakub Kicinski

Recent commit to add NETFILTER_XTABLES_LEGACY missed setting a couple of configs to y. They are still enabled but as modules which appears to have upset BPF CI, e.g.: test_bpf_nf_ct:FAIL:iptables-legacy -t raw -A PREROUTING -j CONNMARK --set-mark 42/0 unexpected error: 768 (errno 0) Fixes: 3c3ab65f00eb ("selftests: net: Enable legacy netfilter legacy options.") Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> --- Targeting net-next 'cause that's where the bad commit is. CC: ast(a)kernel.org CC: daniel(a)iogearbox.net CC: andrii(a)kernel.org CC: martin.lau(a)linux.dev CC: eddyz87(a)gmail.com CC: song(a)kernel.org CC: yonghong.song(a)linux.dev CC: john.fastabend(a)gmail.com CC: kpsingh(a)kernel.org CC: sdf(a)fomichev.me CC: haoluo(a)google.com CC: jolsa(a)kernel.org CC: mykolal(a)fb.com CC: shuah(a)kernel.org CC: pablo(a)netfilter.org CC: bigeasy(a)linutronix.de CC: fw(a)strlen.de CC: bpf(a)vger.kernel.org CC: linux-kselftest(a)vger.kernel.org --- tools/testing/selftests/bpf/config | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tools/testing/selftests/bpf/config b/tools/testing/selftests/bpf/config index 521836776733..e8c6c77b96cb 100644 --- a/tools/testing/selftests/bpf/config +++ b/tools/testing/selftests/bpf/config @@ -97,6 +97,8 @@ CONFIG_NF_TABLES_NETDEV=y CONFIG_NF_TABLES_IPV4=y CONFIG_NF_TABLES_IPV6=y CONFIG_NETFILTER_INGRESS=y +CONFIG_IP_NF_IPTABLES_LEGACY=y +CONFIG_IP6_NF_IPTABLES_LEGACY=y CONFIG_NETFILTER_XTABLES_LEGACY=y CONFIG_NF_FLOW_TABLE=y CONFIG_NF_FLOW_TABLE_INET=y -- 2.50.1

4 weeks

3
3
0 0

[PATCH net v2] selftests: rtnetlink.sh: remove esp4_offload after test

by Xiumei Mu

The esp4_offload module, loaded during IPsec offload tests, should be reset to its default settings after testing. Otherwise, leaving it enabled could unintentionally affect subsequence test cases by keeping offload active. Without this fix: $ lsmod | grep offload; ./rtnetlink.sh -t kci_test_ipsec_offload ; lsmod | grep offload; PASS: ipsec_offload esp4_offload 12288 0 esp4 32768 1 esp4_offload With this fix: $ lsmod | grep offload; ./rtnetlink.sh -t kci_test_ipsec_offload ; lsmod | grep offload; PASS: ipsec_offload Fixes: 2766a11161cc ("selftests: rtnetlink: add ipsec offload API test") Signed-off-by: Xiumei Mu <xmu(a)redhat.com> Reviewed-by: Shannon Nelson <sln(a)onemain.com> --- Changes in v2: - add test results in description - Enhanced logic for rmmod esp4_offload - fix shellcheck warning: SC2086 (The quoting issue) --- --- tools/testing/selftests/net/rtnetlink.sh | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/tools/testing/selftests/net/rtnetlink.sh b/tools/testing/selftests/net/rtnetlink.sh index 2e8243a65b50..d2298da320a6 100755 --- a/tools/testing/selftests/net/rtnetlink.sh +++ b/tools/testing/selftests/net/rtnetlink.sh @@ -673,6 +673,11 @@ kci_test_ipsec_offload() sysfsf=$sysfsd/ipsec sysfsnet=/sys/bus/netdevsim/devices/netdevsim0/net/ probed=false + esp4_offload_probed_default=false + + if lsmod | grep -q esp4_offload; then + esp4_offload_probed_default=true + fi if ! mount | grep -q debugfs; then mount -t debugfs none /sys/kernel/debug/ &> /dev/null @@ -766,6 +771,7 @@ EOF fi # clean up any leftovers + ! "$esp4_offload_probed_default" && lsmod | grep -q esp4_offload && rmmod esp4_offload echo 0 > /sys/bus/netdevsim/del_device $probed && rmmod netdevsim -- 2.50.1

4 weeks

3
2
0 0

[PATCH -next] selftests/ftrace: Prevent potential failure in subsystem-enable test case

by Tengda Wu

The first 100 lines of trace output don't always contain 3 or more distinct events. In busy systems, they may be dominated by repetitive events like sched_stat_runtime, causing the `$count -lt 3` check to fail. Example trace: $ head -n 100 trace | grep -v ^# systemd-timesyn-266 [006] d.h2. 738.778482: sched_stat_runtime: comm=systemd-timesyn pid=266 runtime=976854 [ns] ftracetest-8751 [001] d.h2. 738.778512: sched_stat_runtime: comm=ftracetest pid=8751 runtime=938335 [ns] systemd-timesyn-266 [006] d.h1. 738.779531: sched_stat_runtime: comm=systemd-timesyn pid=266 runtime=1044284 [ns] ftracetest-8751 [001] d.h2. 738.779541: sched_stat_runtime: comm=ftracetest pid=8751 runtime=1028575 [ns] systemd-1 [007] d.h5. 738.779657: sched_stat_runtime: comm=systemd pid=1 runtime=642624 [ns] [...] With trace cleared, simply check `$count -eq 0` to confirm subsystem enablement, just like toplevel-enable.tc does. Fixes: 1a4ea83a6e67 ("selftests/ftrace: Limit length in subsystem-enable tests") Signed-off-by: Tengda Wu <wutengda(a)huaweicloud.com> --- .../selftests/ftrace/test.d/event/subsystem-enable.tc | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tools/testing/selftests/ftrace/test.d/event/subsystem-enable.tc b/tools/testing/selftests/ftrace/test.d/event/subsystem-enable.tc index b7c8f29c09a9..3a28adc7b727 100644 --- a/tools/testing/selftests/ftrace/test.d/event/subsystem-enable.tc +++ b/tools/testing/selftests/ftrace/test.d/event/subsystem-enable.tc @@ -19,8 +19,8 @@ echo 'sched:*' > set_event yield count=`head -n 100 trace | grep -v ^# | awk '{ print $5 }' | sort -u | wc -l` -if [ $count -lt 3 ]; then - fail "at least fork, exec and exit events should be recorded" +if [ $count -eq 0 ]; then + fail "none of scheduler events are recorded" fi do_reset @@ -30,8 +30,8 @@ echo 1 > events/sched/enable yield count=`head -n 100 trace | grep -v ^# | awk '{ print $5 }' | sort -u | wc -l` -if [ $count -lt 3 ]; then - fail "at least fork, exec and exit events should be recorded" +if [ $count -eq 0 ]; then + fail "none of scheduler events are recorded" fi do_reset -- 2.34.1

4 weeks

2
7
0 0

[PATCH net v2] selftests: netfilter: ipvs.sh: Explicity disable rp_filter on interface tunl0

by Yi Chen

Although setup_ns() set net.ipv4.conf.default.rp_filter=0, loading certain module such as ipip will automatically create a tunl0 interface in all netns including new created ones. In the script, this is before than default.rp_filter=0 applied, as a result tunl0.rp_filter remains set to 1 which causes the test report FAIL when ipip module is preloaded. Before fix: Testing DR mode... Testing NAT mode... Testing Tunnel mode... ipvs.sh: FAIL After fix: Testing DR mode... Testing NAT mode... Testing Tunnel mode... ipvs.sh: PASS Fixes: 7c8b89ec506e ("selftests: netfilter: remove rp_filter configuration") v2: Fixed the format of Fixes tag. Signed-off-by: Yi Chen <yiche(a)redhat.com> --- tools/testing/selftests/net/netfilter/ipvs.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/netfilter/ipvs.sh b/tools/testing/selftests/net/netfilter/ipvs.sh index 6af2ea3ad6b8..9c9d5b38ab71 100755 --- a/tools/testing/selftests/net/netfilter/ipvs.sh +++ b/tools/testing/selftests/net/netfilter/ipvs.sh @@ -151,7 +151,7 @@ test_nat() { test_tun() { ip netns exec "${ns0}" ip route add "${vip_v4}" via "${gip_v4}" dev br0 - ip netns exec "${ns1}" modprobe -q ipip + modprobe -q ipip ip netns exec "${ns1}" ip link set tunl0 up ip netns exec "${ns1}" sysctl -qw net.ipv4.ip_forward=0 ip netns exec "${ns1}" sysctl -qw net.ipv4.conf.all.send_redirects=0 @@ -160,10 +160,10 @@ test_tun() { ip netns exec "${ns1}" ipvsadm -a -i -t "${vip_v4}:${port}" -r ${rip_v4}:${port} ip netns exec "${ns1}" ip addr add ${vip_v4}/32 dev lo:1 - ip netns exec "${ns2}" modprobe -q ipip ip netns exec "${ns2}" ip link set tunl0 up ip netns exec "${ns2}" sysctl -qw net.ipv4.conf.all.arp_ignore=1 ip netns exec "${ns2}" sysctl -qw net.ipv4.conf.all.arp_announce=2 + ip netns exec "${ns2}" sysctl -qw net.ipv4.conf.tunl0.rp_filter=0 ip netns exec "${ns2}" ip addr add "${vip_v4}/32" dev lo:1 test_service -- 2.50.1

4 weeks

2
1
0 0

[PATCH net-next] selftests: drv-net: Wait for bkg socat to start

by Mohsin Bashir

Currently, UDP exchange is prone to failure when cmd attempt to send data while socat in bkg is not ready. Since, the behavior is probabilistic, this can result in flakiness for XDP tests. While testing test_xdp_native_tx_mb() on netdevsim, a failure rate of around 1% in 500 500 iterations was observed. Use wait_port_listen() to ensure that the bkg socat is started and ready to receive before cmd start sending. With proposed changes, a re-run of the same test passed 100% of time. Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Mohsin Bashir <mohsin.bashr(a)gmail.com> --- tools/testing/selftests/drivers/net/xdp.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/drivers/net/xdp.py b/tools/testing/selftests/drivers/net/xdp.py index 887d662ad128..1dd8bf3bf6c9 100755 --- a/tools/testing/selftests/drivers/net/xdp.py +++ b/tools/testing/selftests/drivers/net/xdp.py @@ -13,7 +13,7 @@ from enum import Enum from lib.py import ksft_run, ksft_exit, ksft_eq, ksft_ne, ksft_pr from lib.py import KsftFailEx, NetDrvEpEnv, EthtoolFamily, NlError -from lib.py import bkg, cmd, rand_port +from lib.py import bkg, cmd, rand_port, wait_port_listen from lib.py import ip, bpftool, defer @@ -70,6 +70,7 @@ def _exchg_udp(cfg, port, test_string): tx_udp_cmd = f"echo -n {test_string} | socat -t 2 -u STDIN UDP:{cfg.baddr}:{port}" with bkg(rx_udp_cmd, exit_wait=True) as nc: + wait_port_listen(port, proto="udp") cmd(tx_udp_cmd, host=cfg.remote, shell=True) return nc.stdout.strip() @@ -310,6 +311,7 @@ def test_xdp_native_tx_mb(cfg): tx_udp = f"echo {test_string} | socat -t 2 -u STDIN UDP:{cfg.baddr}:{port}" with bkg(rx_udp, host=cfg.remote, exit_wait=True) as rnc: + wait_port_listen(port, proto="udp", host=cfg.remote) cmd(tx_udp, host=cfg.remote, shell=True) stats = _get_stats(prog_info['maps']['map_xdp_stats']) -- 2.47.3

4 weeks

2
1
0 0

[PATCH net-next v7] ipv6: add `force_forwarding` sysctl to enable per-interface forwarding

by Gabriel Goller

It is currently impossible to enable ipv6 forwarding on a per-interface basis like in ipv4. To enable forwarding on an ipv6 interface we need to enable it on all interfaces and disable it on the other interfaces using a netfilter rule. This is especially cumbersome if you have lots of interfaces and only want to enable forwarding on a few. According to the sysctl docs [0] the `net.ipv6.conf.all.forwarding` enables forwarding for all interfaces, while the interface-specific `net.ipv6.conf.<interface>.forwarding` configures the interface Host/Router configuration. Introduce a new sysctl flag `force_forwarding`, which can be set on every interface. The ip6_forwarding function will then check if the global forwarding flag OR the force_forwarding flag is active and forward the packet. To preserve backwards-compatibility reset the flag (on all interfaces) to 0 if the net.ipv6.conf.all.forwarding flag is set to 0. Add a short selftest that checks if a packet gets forwarded with and without `force_forwarding`. [0]: https://www.kernel.org/doc/Documentation/networking/ip-sysctl.txt Acked-by: Nicolas Dichtel <nicolas.dichtel(a)6wind.com> Signed-off-by: Gabriel Goller <g.goller(a)proxmox.com> --- v7: * rebase * fix typos in commit message v6: https://lore.kernel.org/netdev/20250711124243.526735-1-g.goller@proxmox.com/ * rebase * remove brackets around single line * add 'nodad' to addresses in selftest to avoid sporadic failures v5: https://lore.kernel.org/netdev/20250707094307.223975-1-g.goller@proxmox.com/ * update conf/all/forwarding docs * simplified backwards-compat comment * remove ASSERT_RTNL as it's guaranteed by __in6_dev_get_rtnl_net() already * cange ip6_forward logic so that it doesn't depend on the idev existing * move WRITE_ONCE inside device lock v4: https://lore.kernel.org/netdev/20250703160154.560239-1-g.goller@proxmox.com/ * actually write the sysctl value to the table * use ASSERT_RTNL() when forwarding the sysctl change * remove useless comments in function body * simplify forwarding and force_forwarding check in ip6_output.c * fix code backticks in Documentation (double instead of single) * add selftests v3: https://lore.kernel.org/netdev/20250702074619.139031-1-g.goller@proxmox.com/ * remove forwarding=0 setting force_forwarding=0 globally. * add min and max (0 and 1) value to sysctl. v2: https://lore.kernel.org/netdev/20250701140423.487411-1-g.goller@proxmox.com/ * rename from `do_forwarding` to `force_forwarding`. * add global `force_forwarding` flag which will enable `force_forwarding` on every interface like the `ipv4.all.forwarding` flag. * `forwarding`=0 will disable global and per-interface `force_forwarding`. * export option as NETCONFA_FORCE_FORWARDING. v1: https://lore.kernel.org/netdev/20250702074619.139031-1-g.goller@proxmox.com/ Documentation/networking/ip-sysctl.rst | 8 +- include/linux/ipv6.h | 1 + include/uapi/linux/ipv6.h | 1 + include/uapi/linux/netconf.h | 1 + include/uapi/linux/sysctl.h | 1 + net/ipv6/addrconf.c | 82 ++++++++++++++ net/ipv6/ip6_output.c | 3 +- tools/testing/selftests/net/Makefile | 1 + .../selftests/net/ipv6_force_forwarding.sh | 105 ++++++++++++++++++ 9 files changed, 200 insertions(+), 3 deletions(-) create mode 100755 tools/testing/selftests/net/ipv6_force_forwarding.sh diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst index 14700ea77e75..bb620f554598 100644 --- a/Documentation/networking/ip-sysctl.rst +++ b/Documentation/networking/ip-sysctl.rst @@ -2543,8 +2543,8 @@ conf/all/disable_ipv6 - BOOLEAN conf/all/forwarding - BOOLEAN Enable global IPv6 forwarding between all interfaces. - IPv4 and IPv6 work differently here; e.g. netfilter must be used - to control which interfaces may forward packets and which not. + IPv4 and IPv6 work differently here; the ``force_forwarding`` flag must + be used to control which interfaces may forward packets. This also sets all interfaces' Host/Router setting 'forwarding' to the specified value. See below for details. @@ -2561,6 +2561,10 @@ proxy_ndp - BOOLEAN Default: 0 (disabled) +force_forwarding - BOOLEAN + Enable forwarding on this interface only -- regardless of the setting on + ``conf/all/forwarding``. When setting ``conf.all.forwarding`` to 0, + the ``force_forwarding`` flag will be reset on all interfaces. fwmark_reflect - BOOLEAN Controls the fwmark of kernel-generated IPv6 reply packets that are not diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index db0eb0d86b64..bc6ec2959173 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -17,6 +17,7 @@ struct ipv6_devconf { __s32 hop_limit; __s32 mtu6; __s32 forwarding; + __s32 force_forwarding; __s32 disable_policy; __s32 proxy_ndp; __cacheline_group_end(ipv6_devconf_read_txrx); diff --git a/include/uapi/linux/ipv6.h b/include/uapi/linux/ipv6.h index cf592d7b630f..d4d3ae774b26 100644 --- a/include/uapi/linux/ipv6.h +++ b/include/uapi/linux/ipv6.h @@ -199,6 +199,7 @@ enum { DEVCONF_NDISC_EVICT_NOCARRIER, DEVCONF_ACCEPT_UNTRACKED_NA, DEVCONF_ACCEPT_RA_MIN_LFT, + DEVCONF_FORCE_FORWARDING, DEVCONF_MAX }; diff --git a/include/uapi/linux/netconf.h b/include/uapi/linux/netconf.h index fac4edd55379..1c8c84d65ae3 100644 --- a/include/uapi/linux/netconf.h +++ b/include/uapi/linux/netconf.h @@ -19,6 +19,7 @@ enum { NETCONFA_IGNORE_ROUTES_WITH_LINKDOWN, NETCONFA_INPUT, NETCONFA_BC_FORWARDING, + NETCONFA_FORCE_FORWARDING, __NETCONFA_MAX }; #define NETCONFA_MAX (__NETCONFA_MAX - 1) diff --git a/include/uapi/linux/sysctl.h b/include/uapi/linux/sysctl.h index 8981f00204db..63d1464cb71c 100644 --- a/include/uapi/linux/sysctl.h +++ b/include/uapi/linux/sysctl.h @@ -573,6 +573,7 @@ enum { NET_IPV6_ACCEPT_RA_FROM_LOCAL=26, NET_IPV6_ACCEPT_RA_RT_INFO_MIN_PLEN=27, NET_IPV6_RA_DEFRTR_METRIC=28, + NET_IPV6_FORCE_FORWARDING=29, __NET_IPV6_MAX }; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 4f1d7d110302..81a067a2e526 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -239,6 +239,7 @@ static struct ipv6_devconf ipv6_devconf __read_mostly = { .ndisc_evict_nocarrier = 1, .ra_honor_pio_life = 0, .ra_honor_pio_pflag = 0, + .force_forwarding = 0, }; static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = { @@ -303,6 +304,7 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = { .ndisc_evict_nocarrier = 1, .ra_honor_pio_life = 0, .ra_honor_pio_pflag = 0, + .force_forwarding = 0, }; /* Check if link is ready: is it up and is a valid qdisc available */ @@ -857,6 +859,9 @@ static void addrconf_forward_change(struct net *net, __s32 newf) idev = __in6_dev_get_rtnl_net(dev); if (idev) { int changed = (!idev->cnf.forwarding) ^ (!newf); + /* Disabling all.forwarding sets 0 to force_forwarding for all interfaces */ + if (newf == 0) + WRITE_ONCE(idev->cnf.force_forwarding, 0); WRITE_ONCE(idev->cnf.forwarding, newf); if (changed) @@ -5710,6 +5715,7 @@ static void ipv6_store_devconf(const struct ipv6_devconf *cnf, array[DEVCONF_ACCEPT_UNTRACKED_NA] = READ_ONCE(cnf->accept_untracked_na); array[DEVCONF_ACCEPT_RA_MIN_LFT] = READ_ONCE(cnf->accept_ra_min_lft); + array[DEVCONF_FORCE_FORWARDING] = READ_ONCE(cnf->force_forwarding); } static inline size_t inet6_ifla6_size(void) @@ -6738,6 +6744,75 @@ static int addrconf_sysctl_disable_policy(const struct ctl_table *ctl, int write return ret; } +static void addrconf_force_forward_change(struct net *net, __s32 newf) +{ + struct net_device *dev; + struct inet6_dev *idev; + + for_each_netdev(net, dev) { + idev = __in6_dev_get_rtnl_net(dev); + if (idev) { + int changed = (!idev->cnf.force_forwarding) ^ (!newf); + + WRITE_ONCE(idev->cnf.force_forwarding, newf); + if (changed) + inet6_netconf_notify_devconf(dev_net(dev), RTM_NEWNETCONF, + NETCONFA_FORCE_FORWARDING, + dev->ifindex, &idev->cnf); + } + } +} + +static int addrconf_sysctl_force_forwarding(const struct ctl_table *ctl, int write, + void *buffer, size_t *lenp, loff_t *ppos) +{ + struct inet6_dev *idev = ctl->extra1; + struct ctl_table tmp_ctl = *ctl; + struct net *net = ctl->extra2; + int *valp = ctl->data; + int new_val = *valp; + int old_val = *valp; + loff_t pos = *ppos; + int ret; + + tmp_ctl.extra1 = SYSCTL_ZERO; + tmp_ctl.extra2 = SYSCTL_ONE; + tmp_ctl.data = &new_val; + + ret = proc_douintvec_minmax(&tmp_ctl, write, buffer, lenp, ppos); + + if (write && old_val != new_val) { + if (!rtnl_net_trylock(net)) + return restart_syscall(); + + WRITE_ONCE(*valp, new_val); + + if (valp == &net->ipv6.devconf_dflt->force_forwarding) { + inet6_netconf_notify_devconf(net, RTM_NEWNETCONF, + NETCONFA_FORCE_FORWARDING, + NETCONFA_IFINDEX_DEFAULT, + net->ipv6.devconf_dflt); + } else if (valp == &net->ipv6.devconf_all->force_forwarding) { + inet6_netconf_notify_devconf(net, RTM_NEWNETCONF, + NETCONFA_FORCE_FORWARDING, + NETCONFA_IFINDEX_ALL, + net->ipv6.devconf_all); + + addrconf_force_forward_change(net, new_val); + } else { + inet6_netconf_notify_devconf(net, RTM_NEWNETCONF, + NETCONFA_FORCE_FORWARDING, + idev->dev->ifindex, + &idev->cnf); + } + rtnl_net_unlock(net); + } + + if (ret) + *ppos = pos; + return ret; +} + static int minus_one = -1; static const int two_five_five = 255; static u32 ioam6_if_id_max = U16_MAX; @@ -7208,6 +7283,13 @@ static const struct ctl_table addrconf_sysctl[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_TWO, }, + { + .procname = "force_forwarding", + .data = &ipv6_devconf.force_forwarding, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = addrconf_sysctl_force_forwarding, + }, }; static int __addrconf_sysctl_register(struct net *net, char *dev_name, diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 0412f8544695..1e1410237b6e 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -511,7 +511,8 @@ int ip6_forward(struct sk_buff *skb) u32 mtu; idev = __in6_dev_get_safely(dev_get_by_index_rcu(net, IP6CB(skb)->iif)); - if (READ_ONCE(net->ipv6.devconf_all->forwarding) == 0) + if (!READ_ONCE(net->ipv6.devconf_all->forwarding) && + (!idev || !READ_ONCE(idev->cnf.force_forwarding))) goto error; if (skb->pkt_type != PACKET_HOST) diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile index 13e2678d418b..b31a71f2b372 100644 --- a/tools/testing/selftests/net/Makefile +++ b/tools/testing/selftests/net/Makefile @@ -116,6 +116,7 @@ TEST_GEN_FILES += skf_net_off TEST_GEN_FILES += tfo TEST_PROGS += tfo_passive.sh TEST_PROGS += broadcast_pmtu.sh +TEST_PROGS += ipv6_force_forwarding.sh # YNL files, must be before "include ..lib.mk" YNL_GEN_FILES := busy_poller netlink-dumps diff --git a/tools/testing/selftests/net/ipv6_force_forwarding.sh b/tools/testing/selftests/net/ipv6_force_forwarding.sh new file mode 100755 index 000000000000..bf0243366caa --- /dev/null +++ b/tools/testing/selftests/net/ipv6_force_forwarding.sh @@ -0,0 +1,105 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# +# Test IPv6 force_forwarding interface property +# +# This test verifies that the force_forwarding property works correctly: +# - When global forwarding is disabled, packets are not forwarded normally +# - When force_forwarding is enabled on an interface, packets are forwarded +# regardless of the global forwarding setting + +source lib.sh + +cleanup() { + cleanup_ns $ns1 $ns2 $ns3 +} + +trap cleanup EXIT + +setup_test() { + # Create three namespaces: sender, router, receiver + setup_ns ns1 ns2 ns3 + + # Create veth pairs: ns1 <-> ns2 <-> ns3 + ip link add name veth12 type veth peer name veth21 + ip link add name veth23 type veth peer name veth32 + + # Move interfaces to namespaces + ip link set veth12 netns $ns1 + ip link set veth21 netns $ns2 + ip link set veth23 netns $ns2 + ip link set veth32 netns $ns3 + + # Configure interfaces + ip -n $ns1 addr add 2001:db8:1::1/64 dev veth12 nodad + ip -n $ns2 addr add 2001:db8:1::2/64 dev veth21 nodad + ip -n $ns2 addr add 2001:db8:2::1/64 dev veth23 nodad + ip -n $ns3 addr add 2001:db8:2::2/64 dev veth32 nodad + + # Bring up interfaces + ip -n $ns1 link set veth12 up + ip -n $ns2 link set veth21 up + ip -n $ns2 link set veth23 up + ip -n $ns3 link set veth32 up + + # Add routes + ip -n $ns1 route add 2001:db8:2::/64 via 2001:db8:1::2 + ip -n $ns3 route add 2001:db8:1::/64 via 2001:db8:2::1 + + # Disable global forwarding + ip netns exec $ns2 sysctl -qw net.ipv6.conf.all.forwarding=0 +} + +test_force_forwarding() { + local ret=0 + + echo "TEST: force_forwarding functionality" + + # Check if force_forwarding sysctl exists + if ! ip netns exec $ns2 test -f /proc/sys/net/ipv6/conf/veth21/force_forwarding; then + echo "SKIP: force_forwarding not available" + return $ksft_skip + fi + + # Test 1: Without force_forwarding, ping should fail + ip netns exec $ns2 sysctl -qw net.ipv6.conf.veth21.force_forwarding=0 + ip netns exec $ns2 sysctl -qw net.ipv6.conf.veth23.force_forwarding=0 + + if ip netns exec $ns1 ping -6 -c 1 -W 2 2001:db8:2::2 &>/dev/null; then + echo "FAIL: ping succeeded when forwarding disabled" + ret=1 + else + echo "PASS: forwarding disabled correctly" + fi + + # Test 2: With force_forwarding enabled, ping should succeed + ip netns exec $ns2 sysctl -qw net.ipv6.conf.veth21.force_forwarding=1 + ip netns exec $ns2 sysctl -qw net.ipv6.conf.veth23.force_forwarding=1 + + if ip netns exec $ns1 ping -6 -c 1 -W 2 2001:db8:2::2 &>/dev/null; then + echo "PASS: force_forwarding enabled forwarding" + else + echo "FAIL: ping failed with force_forwarding enabled" + ret=1 + fi + + return $ret +} + +echo "IPv6 force_forwarding test" +echo "==========================" + +setup_test +test_force_forwarding +ret=$? + +if [ $ret -eq 0 ]; then + echo "OK" + exit 0 +elif [ $ret -eq $ksft_skip ]; then + echo "SKIP" + exit $ksft_skip +else + echo "FAIL" + exit 1 +fi -- 2.39.5

4 weeks

2
1
0 0

[PATCH net-next] selftests: net: Skip test if IPv6 is not configured

by Breno Leitao

Extend the `check_for_dependencies()` function in `lib_netcons.sh` to check whether IPv6 is enabled by verifying the existence of `/proc/net/if_inet6`. Having IPv6 is a now a dependency of netconsole tests. If the file does not exist, the script will skip the test with an appropriate message suggesting to verify if `CONFIG_IPV6` is enabled. This prevents the test to misbehave if IPv6 is not configured. Signed-off-by: Breno Leitao <leitao(a)debian.org> --- tools/testing/selftests/drivers/net/lib/sh/lib_netcons.sh | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/tools/testing/selftests/drivers/net/lib/sh/lib_netcons.sh b/tools/testing/selftests/drivers/net/lib/sh/lib_netcons.sh index 258af805497b4..b6071e80ebbb6 100644 --- a/tools/testing/selftests/drivers/net/lib/sh/lib_netcons.sh +++ b/tools/testing/selftests/drivers/net/lib/sh/lib_netcons.sh @@ -281,6 +281,11 @@ function check_for_dependencies() { exit "${ksft_skip}" fi + if [ ! -f /proc/net/if_inet6 ]; then + echo "SKIP: IPv6 not configured. Check if CONFIG_IPV6 is enabled" >&2 + exit "${ksft_skip}" + fi + if [ ! -f "${NSIM_DEV_SYS_NEW}" ]; then echo "SKIP: file ${NSIM_DEV_SYS_NEW} does not exist. Check if CONFIG_NETDEVSIM is enabled" >&2 exit "${ksft_skip}" --- base-commit: dd500e4aecf25e48e874ca7628697969df679493 change-id: 20250723-netcons_test_ipv6-15b1b76bb231 Best regards, -- Breno Leitao <leitao(a)debian.org>

4 weeks

5
7
0 0

[PATCH v3 00/10] mm/mremap: permit mremap() move of multiple VMAs

by Lorenzo Stoakes

Historically we've made it a uAPI requirement that mremap() may only operate on a single VMA at a time. For instances where VMAs need to be resized, this makes sense, as it becomes very difficult to determine what a user actually wants should they indicate a desire to expand or shrink the size of multiple VMAs (truncate? Adjust sizes individually? Some other strategy?). However, in instances where a user is moving VMAs, it is restrictive to disallow this. This is especially the case when anonymous mapping remap may or may not be mergeable depending on whether VMAs have or have not been faulted due to anon_vma assignment and folio index alignment with vma->vm_pgoff. Often this can result in surprising impact where a moved region is faulted, then moved back and a user fails to observe a merge from otherwise compatible, adjacent VMAs. This change allows such cases to work without the user having to be cognizant of whether a prior mremap() move or other VMA operations has resulted in VMA fragmentation. In order to do this, this series performs a large amount of refactoring, most pertinently - grouping sanity checks together, separately those that check input parameters and those relating to VMAs. we also simplify the post-mmap lock drop processing for uffd and mlock()'d VMAs. With this done, we can then fairly straightforwardly implement this functionality. This works exclusively for mremap() invocations which specify MREMAP_FIXED. It is not compatible with VMAs which use userfaultfd, as the notification of the userland fault handler would require us to drop the mmap lock. It is also not compatible with file-backed mappings with customised get_unmapped_area() handlers as these may not honour MREMAP_FIXED. The input and output addresses ranges must not overlap. We carefully account for moves which would result in VMA iterator invalidation. While there can be gaps between VMAs in the input range, there can be no gap before the first VMA in the range. v3: * Disallowed move operation except for MREMAP_FIXED. * Disallow gap at start of aggregate range to avoid confusion. * Disallow any file-baked VMAs with custom get_unmapped_area. * Renamed multi_vma to seen_vma to be clearer. Stop reusing new_addr, use separate target_addr var to track next target address. * Check if first VMA fails multi VMA check, if so we'll allow one VMA but not multiple. * Updated the commit message for patch 9 to be clearer about gap behaviour. * Removed accidentally included debug goto statement in test (doh!). Test was and is passing regardless. * Unmap target range in test, previously we ended up moving additional VMAs unintentionally. This still all passed :) but was not what was intended. * Removed self-merge check - there is absolutely no way this can happen across multiple VMAs, as there is no means of moving VMAs such that a VMA merges with itself. v2: * Squashed uffd stub fix into series. * Propagated tags, thanks! * Fixed param naming in patch 4 as per Vlastimil. * Renamed vma_reset to vmi_needs_reset + dropped reset on unmap as per Liam. * Correctly return -EFAULT if no VMAs in input range. * Account for get_unmapped_area() disregarding MAP_FIXED and returning an altered address. * Added additional explanatatory comment to the remap_move() function. https://lore.kernel.org/all/cover.1751865330.git.lorenzo.stoakes@oracle.com/ v1: https://lore.kernel.org/all/cover.1751865330.git.lorenzo.stoakes@oracle.com/ Lorenzo Stoakes (10): mm/mremap: perform some simple cleanups mm/mremap: refactor initial parameter sanity checks mm/mremap: put VMA check and prep logic into helper function mm/mremap: cleanup post-processing stage of mremap mm/mremap: use an explicit uffd failure path for mremap mm/mremap: check remap conditions earlier mm/mremap: move remap_is_valid() into check_prep_vma() mm/mremap: clean up mlock populate behaviour mm/mremap: permit mremap() move of multiple VMAs tools/testing/selftests: extend mremap_test to test multi-VMA mremap fs/userfaultfd.c | 15 +- include/linux/userfaultfd_k.h | 5 + mm/mremap.c | 553 +++++++++++++++-------- tools/testing/selftests/mm/mremap_test.c | 146 +++++- 4 files changed, 518 insertions(+), 201 deletions(-) -- 2.50.0

4 weeks, 1 day

3
19
0 0

[PATCH v19 4/5] binder: add transaction_report feature entry

by Carlos Llamas

From: Li Li <dualli(a)google.com> Add "transaction_report" to the binderfs feature list, to help userspace determine if the "BINDER_CMD_REPORT" generic netlink api is supported by the binder driver. Signed-off-by: Li Li <dualli(a)google.com> Signed-off-by: Carlos Llamas <cmllamas(a)google.com> --- drivers/android/binderfs.c | 8 ++++++++ .../selftests/filesystems/binderfs/binderfs_test.c | 1 + 2 files changed, 9 insertions(+) diff --git a/drivers/android/binderfs.c b/drivers/android/binderfs.c index 4f827152d18e..f74a7e380261 100644 --- a/drivers/android/binderfs.c +++ b/drivers/android/binderfs.c @@ -59,6 +59,7 @@ struct binder_features { bool oneway_spam_detection; bool extended_error; bool freeze_notification; + bool transaction_report; }; static const struct constant_table binderfs_param_stats[] = { @@ -76,6 +77,7 @@ static struct binder_features binder_features = { .oneway_spam_detection = true, .extended_error = true, .freeze_notification = true, + .transaction_report = true, }; static inline struct binderfs_info *BINDERFS_SB(const struct super_block *sb) @@ -616,6 +618,12 @@ static int init_binder_features(struct super_block *sb) if (IS_ERR(dentry)) return PTR_ERR(dentry); + dentry = binderfs_create_file(dir, "transaction_report", + &binder_features_fops, + &binder_features.transaction_report); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + return 0; } diff --git a/tools/testing/selftests/filesystems/binderfs/binderfs_test.c b/tools/testing/selftests/filesystems/binderfs/binderfs_test.c index 81db85a5cc16..39a68078a79b 100644 --- a/tools/testing/selftests/filesystems/binderfs/binderfs_test.c +++ b/tools/testing/selftests/filesystems/binderfs/binderfs_test.c @@ -65,6 +65,7 @@ static int __do_binderfs_test(struct __test_metadata *_metadata) "oneway_spam_detection", "extended_error", "freeze_notification", + "transaction_report", }; change_mountns(_metadata); -- 2.50.1.470.g6ba607880d-goog

4 weeks, 1 day

1
0
0 0

[PATCH net v2 1/2] macsec: set IFF_UNICAST_FLT priv flag

by Stanislav Fomichev

Cosmin reports the following locking issue: # BUG: sleeping function called from invalid context at kernel/locking/mutex.c:275 # dump_stack_lvl+0x4f/0x60 # __might_resched+0xeb/0x140 # mutex_lock+0x1a/0x40 # dev_set_promiscuity+0x26/0x90 # __dev_set_promiscuity+0x85/0x170 # __dev_set_rx_mode+0x69/0xa0 # dev_uc_add+0x6d/0x80 # vlan_dev_open+0x5f/0x120 [8021q] # __dev_open+0x10c/0x2a0 # __dev_change_flags+0x1a4/0x210 # netif_change_flags+0x22/0x60 # do_setlink.isra.0+0xdb0/0x10f0 # rtnl_newlink+0x797/0xb00 # rtnetlink_rcv_msg+0x1cb/0x3f0 # netlink_rcv_skb+0x53/0x100 # netlink_unicast+0x273/0x3b0 # netlink_sendmsg+0x1f2/0x430 Which is similar to recent syzkaller reports in [0] and [1] and triggers because macsec does not advertise IFF_UNICAST_FLT although it has proper ndo_set_rx_mode callback that takes care of pushing uc/mc addresses down to the real device. In general, dev_uc_add call path is problematic for stacking non-IFF_UNICAST_FLT because we might grab netdev instance lock under addr_list_lock spinlock, so this is not a systemic fix. 0: https://lore.kernel.org/netdev/686d55b4.050a0220.1ffab7.0014.GAE@google.com 1: https://lore.kernel.org/netdev/68712acf.a00a0220.26a83e.0051.GAE@google.com/ Reviewed-by: Simon Horman <horms(a)kernel.org> Tested-by: Simon Horman <horms(a)kernel.org> Link: https://lore.kernel.org/netdev/2aff4342b0f5b1539c02ffd8df4c7e58dd9746e7.cam… Fixes: 7e4d784f5810 ("net: hold netdev instance lock during rtnetlink operations") Reported-by: Cosmin Ratiu <cratiu(a)nvidia.com> Tested-by: Cosmin Ratiu <cratiu(a)nvidia.com> Signed-off-by: Stanislav Fomichev <sdf(a)fomichev.me> --- drivers/net/macsec.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 7edbe76b5455..4c75d1fea552 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -3868,7 +3868,7 @@ static void macsec_setup(struct net_device *dev) ether_setup(dev); dev->min_mtu = 0; dev->max_mtu = ETH_MAX_MTU; - dev->priv_flags |= IFF_NO_QUEUE; + dev->priv_flags |= IFF_NO_QUEUE | IFF_UNICAST_FLT; dev->netdev_ops = &macsec_netdev_ops; dev->needs_free_netdev = true; dev->priv_destructor = macsec_free_netdev; -- 2.50.1

4 weeks, 1 day

2
2
0 0

[PATCH net 0/2] bonding: fix LACP negotiation issues in passive mode

by Hangbin Liu

This patchset fixes an issue where bonding fails to establish a stable LACP negotiation when operating in passive mode (lacp_active=off). In passive mode, the current implementation only replies when the partner's state changes, which results in LACP timeout and unstable aggregator formation. With this change, the bond responds to each received LACPDU in passive mode by setting ntt = true, ensuring timely replies and stable LACP negotiation. Hangbin Liu (2): bonding: update ntt to true in passive mode selftests: bonding: add test for passive LACP mode drivers/net/bonding/bond_3ad.c | 6 ++ .../drivers/net/bonding/bond_passive_lacp.sh | 21 +++++ .../drivers/net/bonding/bond_topo_lacp.sh | 77 +++++++++++++++++++ 3 files changed, 104 insertions(+) create mode 100755 tools/testing/selftests/drivers/net/bonding/bond_passive_lacp.sh create mode 100644 tools/testing/selftests/drivers/net/bonding/bond_topo_lacp.sh -- 2.46.0

4 weeks, 1 day

4
14
0 0

[RFC PATCH v2 0/9] KVM: Enable Nested Virt selftests

by Ganapatrao Kulkarni

This patch series makes the selftest work with NV enabled. The guest code is run in vEL2 instead of EL1. We add a command line option to enable testing of NV. The NV tests are disabled by default. Modified around 12 selftests in this series. Changes since v1: - Updated NV helper functions as per comments [1]. - Modified existing testscases to run guest code in vEL2. [1] https://lkml.iu.edu/hypermail/linux/kernel/2502.0/07001.html Ganapatrao Kulkarni (9): KVM: arm64: nv: selftests: Add support to run guest code in vEL2. KVM: arm64: nv: selftests: Add simple test to run guest code in vEL2 KVM: arm64: nv: selftests: Enable hypervisor timer tests to run in vEL2 KVM: arm64: nv: selftests: enable aarch32_id_regs test to run in vEL2 KVM: arm64: nv: selftests: Enable vgic tests to run in vEL2 KVM: arm64: nv: selftests: Enable set_id_regs test to run in vEL2 KVM: arm64: nv: selftests: Enable test to run in vEL2 KVM: selftests: arm64: Extend kvm_page_table_test to run guest code in vEL2 KVM: arm64: nv: selftests: Enable page_fault_test test to run in vEL2 tools/testing/selftests/kvm/Makefile.kvm | 2 + tools/testing/selftests/kvm/arch_timer.c | 8 +- .../selftests/kvm/arm64/aarch32_id_regs.c | 34 ++++- .../testing/selftests/kvm/arm64/arch_timer.c | 118 +++++++++++++++--- .../selftests/kvm/arm64/nv_guest_hypervisor.c | 68 ++++++++++ .../selftests/kvm/arm64/page_fault_test.c | 35 +++++- .../testing/selftests/kvm/arm64/set_id_regs.c | 57 ++++++++- tools/testing/selftests/kvm/arm64/vgic_init.c | 54 +++++++- tools/testing/selftests/kvm/arm64/vgic_irq.c | 27 ++-- .../selftests/kvm/arm64/vgic_lpi_stress.c | 19 ++- .../testing/selftests/kvm/guest_print_test.c | 32 +++++ .../selftests/kvm/include/arm64/arch_timer.h | 16 +++ .../kvm/include/arm64/kvm_util_arch.h | 3 + .../selftests/kvm/include/arm64/nv_util.h | 45 +++++++ .../selftests/kvm/include/arm64/vgic.h | 1 + .../testing/selftests/kvm/include/kvm_util.h | 3 + .../selftests/kvm/include/timer_test.h | 1 + .../selftests/kvm/kvm_page_table_test.c | 30 ++++- tools/testing/selftests/kvm/lib/arm64/nv.c | 46 +++++++ .../selftests/kvm/lib/arm64/processor.c | 61 ++++++--- tools/testing/selftests/kvm/lib/arm64/vgic.c | 8 ++ 21 files changed, 604 insertions(+), 64 deletions(-) create mode 100644 tools/testing/selftests/kvm/arm64/nv_guest_hypervisor.c create mode 100644 tools/testing/selftests/kvm/include/arm64/nv_util.h create mode 100644 tools/testing/selftests/kvm/lib/arm64/nv.c -- 2.48.1

4 weeks, 1 day

6
28
0 0

[PATCH RFC 00/14] sparc64: vdso: Switch to generic vDSO library

by Thomas Weißschuh

The generic vDSO provides a lot common functionality shared between different architectures. SPARC is the last architecture not using it, preventing some necessary code cleanup. Make use of the generic infrastructure. Follow-up to and replacement for Arnd's SPARC vDSO removal patches: https://lore.kernel.org/lkml/20250707144726.4008707-1-arnd@kernel.org/ Only tested on QEMU. Based von v6.16-rc1. Marked as RFC for testing and review only. Will be properly resubmitted after v6.17-rc1. Signed-off-by: Thomas Weißschuh <thomas.weissschuh(a)linutronix.de> --- Arnd Bergmann (1): clocksource: remove ARCH_CLOCKSOURCE_DATA Thomas Weißschuh (13): vdso: add struct __kernel_old_timeval forward declaration to gettime.h sparc64: time: Remove architecture-specific clocksource data sparc64: vdso: Link with -z noexecstack sparc64: vdso: Remove obsolete "fake section table" reservation sparc64: vdso: Replace code patching with runtime conditional sparc64: vdso: Move hardware counter read into header sparc64: vdso: Move syscall fallbacks into header sparc64: vdso: Introduce vdso/processor.h sparc64: vdso: Switch to the generic vDSO library sparc64: vdso2c: Drop sym_vvar_start handling sparc64: vdso2c: Remove symbol handling sparc64: vdso: Implement clock_gettime64() sparc64: vdso: Implement clock_getres() arch/sparc/Kconfig | 5 +- arch/sparc/include/asm/clocksource.h | 9 - arch/sparc/include/asm/processor.h | 3 + arch/sparc/include/asm/processor_32.h | 2 - arch/sparc/include/asm/processor_64.h | 25 -- arch/sparc/include/asm/vdso.h | 2 - arch/sparc/include/asm/vdso/clocksource.h | 10 + arch/sparc/include/asm/vdso/gettimeofday.h | 208 ++++++++++++++++ arch/sparc/include/asm/vdso/processor.h | 41 ++++ arch/sparc/include/asm/vdso/vsyscall.h | 10 + arch/sparc/include/asm/vvar.h | 75 ------ arch/sparc/kernel/Makefile | 1 - arch/sparc/kernel/time_64.c | 6 +- arch/sparc/kernel/vdso.c | 69 ------ arch/sparc/vdso/Makefile | 8 +- arch/sparc/vdso/vclock_gettime.c | 382 +++-------------------------- arch/sparc/vdso/vdso-layout.lds.S | 26 +- arch/sparc/vdso/vdso.lds.S | 4 +- arch/sparc/vdso/vdso2c.c | 24 -- arch/sparc/vdso/vdso2c.h | 45 +--- arch/sparc/vdso/vdso32/vdso32.lds.S | 6 +- arch/sparc/vdso/vma.c | 274 ++------------------- include/linux/clocksource.h | 6 +- include/vdso/gettime.h | 1 + kernel/time/Kconfig | 4 - 25 files changed, 344 insertions(+), 902 deletions(-) --- base-commit: eaa6313d2ceb2a3f1c870866621058ad6081f028 change-id: 20250722-vdso-sparc64-generic-2-25f2e058e92c Best regards, -- Thomas Weißschuh <thomas.weissschuh(a)linutronix.de>

4 weeks, 1 day

2
15
0 0

[PATCH net-next 0/2] selftests: drv-net: Fix and improve command requirement checking

by Gal Pressman

This series fixes remote command checking and cleans up command requirement calls across tests. The first patch fixes require_cmd() incorrectly checking commands locally even when remote=True was specified due to a missing host parameter. The second patch makes require_cmd() usage explicit about local/remote requirements, avoiding unnecessary test failures and consolidating duplicate calls. Gal Pressman (2): selftests: drv-net: Fix remote command checking in require_cmd() selftests: drv-net: Make command requirements explicit tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py | 3 +-- tools/testing/selftests/drivers/net/hw/rss_input_xfrm.py | 2 +- tools/testing/selftests/drivers/net/hw/tso.py | 2 +- tools/testing/selftests/drivers/net/lib/py/env.py | 2 +- tools/testing/selftests/drivers/net/lib/py/load.py | 2 +- tools/testing/selftests/drivers/net/ping.py | 2 +- 6 files changed, 6 insertions(+), 7 deletions(-) -- 2.40.1

4 weeks, 1 day

2
3
0 0

[PATCH net 0/3] selftests: drv-net: tso: fix issues with tso selftest

by Daniel Zahka

There are a couple issues with the tso selftest. - Features required for test cases are detected by searching the set of active features at test start, so if a feature is supported by hw, but disabled, the test will report that the feature under test is not available and fail. - The vxlan test cases do not use the correct ip link flags based on the gso feature under test - The non-tunneled tso6 test case is showing up with the wrong name. With all patches applied test output is: # Detected qstat for LSO wire-packets TAP version 13 1..14 ok 1 tso.ipv4 # Testing with mangleid enabled ok 2 tso.vxlan4_ipv4 ok 3 tso.vxlan4_ipv6 # Testing with mangleid enabled ok 4 tso.vxlan_csum4_ipv4 ok 5 tso.vxlan_csum4_ipv6 # Testing with mangleid enabled ok 6 tso.gre4_ipv4 ok 7 tso.gre4_ipv6 ok 8 tso.ipv6 # Testing with mangleid enabled ok 9 tso.vxlan6_ipv4 ok 10 tso.vxlan6_ipv6 # Testing with mangleid enabled ok 11 tso.vxlan_csum6_ipv4 ok 12 tso.vxlan_csum6_ipv6 # Testing with mangleid enabled ok 13 tso.gre6_ipv4 ok 14 tso.gre6_ipv6 # Totals: pass:14 fail:0 xfail:0 xpass:0 skip:0 error:0 Daniel Zahka (3): selftests: drv-net: tso: enable test cases based on hw_features selftests: drv-net: tso: fix vxlan tunnel flags to get correct gso_type selftests: drv-net: tso: fix non-tunneled tso6 test case name tools/testing/selftests/drivers/net/hw/tso.py | 99 +++++++++++-------- 1 file changed, 59 insertions(+), 40 deletions(-) -- 2.47.1

4 weeks, 1 day

2
4
0 0

[PATCH RFC] selftests/pidfd: Fix duplicate-symbol warnings for SCHED_ CPP symbols

by Paul E. McKenney

The pidfd selftests run in userspace and include both userspace and kernel header files. On some distros (for example, CentOS), this results in duplicate-symbol warnings in allmodconfig builds, while on other distros (for example, Ubuntu) it does not. (This happens in recent -next trees, including next-20250714.) Therefore, use #undef to get rid of the userspace definitions in favor of the kernel definitions. Other ways of handling this include splitting up the selftest code so that the userspace definitions go into one translation unit and the kernel definitions into another (which might or might not be feasible) or to adjust compiler command-line options to suppress the warnings (which might or might not be desirable). Signed-off-by: Paul E. McKenney <paulmck(a)kernel.org> Cc: Christian Brauner <brauner(a)kernel.org> Cc: Shuah Khan <shuah(a)kernel.org> Cc: <linux-kselftest(a)vger.kernel.org> --- pidfd.h | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/tools/testing/selftests/pidfd/pidfd.h b/tools/testing/selftests/pidfd/pidfd.h index efd74063126eb..6ff495398e872 100644 --- a/tools/testing/selftests/pidfd/pidfd.h +++ b/tools/testing/selftests/pidfd/pidfd.h @@ -16,6 +16,10 @@ #include <sys/types.h> #include <sys/wait.h> +#undef SCHED_NORMAL +#undef SCHED_FLAG_KEEP_ALL +#undef SCHED_FLAG_UTIL_CLAMP + #include "../kselftest.h" #include "../clone3/clone3_selftests.h"

4 weeks, 1 day

2
6
0 0

[PATCH] selftests/bpf: Install test modules into $INSTALL_PATH

by Ricardo B. Marlière

The tests expect the modules to be in the same working directory, but when using a different $INSTALL_PATH they are not copied over. Signed-off-by: Ricardo B. Marlière <rbm(a)suse.com> --- # cd tools/testing/selftests/kselftest_install && ./run_kselftest.sh -t bpf:test_verifier TAP version 13 1..1 # timeout set to 0 # selftests: bpf: test_verifier # Can't find bpf_testmod.ko kernel module: -2 not ok 1 selftests: bpf: test_verifier # exit=1 --- tools/testing/selftests/bpf/Makefile | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index 4863106034dfbcd35f830432322f054d897bb406..56b0565af8a76a9e784836a836935dd22e814fc0 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -877,5 +877,7 @@ override define INSTALL_RULE @for DIR in $(TEST_INST_SUBDIRS); do \ mkdir -p $(INSTALL_PATH)/$$DIR; \ rsync -a $(OUTPUT)/$$DIR/*.bpf.o $(INSTALL_PATH)/$$DIR;\ + rsync -a $(OUTPUT)/$$DIR/*.ko $(INSTALL_PATH)/$$DIR;\ + rsync -a $(OUTPUT)/*.ko $(INSTALL_PATH);\ done endef --- base-commit: f227e9ed4fe4f2fed40e4725d6c10860d30c2ea2 change-id: 20250724-bpf-next_for-next-f1de3e4becc8 Best regards, -- Ricardo B. Marlière <rbm(a)suse.com>

4 weeks, 1 day

1
0
0 0

[PATCH] selftests/tracing: Fix false failure of subsystem event test

by Steven Rostedt

From: Steven Rostedt <rostedt(a)goodmis.org> The subsystem event test enables all "sched" events and makes sure there's at least 3 different events in the output. It used to cat the entire trace file to | wc -l, but on slow machines, that could last a very long time. To solve that, it was changed to just read the first 100 lines of the trace file. This can cause false failures as some events repeat so often, that the 100 lines that are examined could possibly be of only one event. Instead, create an awk script that looks for 3 different events and will exit out after it finds them. This will find the 3 events the test looks for (eventually if it works), and still exit out after the test is satisfied and not cause slower machines to run forever. Reported-by: Tengda Wu <wutengda(a)huaweicloud.com> Closes: https://lore.kernel.org/all/20250710130134.591066-1-wutengda@huaweicloud.co… Fixes: 1a4ea83a6e67 ("selftests/ftrace: Limit length in subsystem-enable tests") Signed-off-by: Steven Rostedt (Google) <rostedt(a)goodmis.org> --- .../ftrace/test.d/event/subsystem-enable.tc | 28 +++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/ftrace/test.d/event/subsystem-enable.tc b/tools/testing/selftests/ftrace/test.d/event/subsystem-enable.tc index b7c8f29c09a9..65916bb55dfb 100644 --- a/tools/testing/selftests/ftrace/test.d/event/subsystem-enable.tc +++ b/tools/testing/selftests/ftrace/test.d/event/subsystem-enable.tc @@ -14,11 +14,35 @@ fail() { #msg exit_fail } +# As reading trace can last forever, simply look for 3 different +# events then exit out of reading the file. If there's not 3 different +# events, then the test has failed. +check_unique() { + cat trace | grep -v '^#' | awk ' + BEGIN { cnt = 0; } + { + for (i = 0; i < cnt; i++) { + if (event[i] == $5) { + break; + } + } + if (i == cnt) { + event[cnt++] = $5; + if (cnt > 2) { + exit; + } + } + } + END { + printf "%d", cnt; + }' +} + echo 'sched:*' > set_event yield -count=`head -n 100 trace | grep -v ^# | awk '{ print $5 }' | sort -u | wc -l` +count=`check_unique` if [ $count -lt 3 ]; then fail "at least fork, exec and exit events should be recorded" fi @@ -29,7 +53,7 @@ echo 1 > events/sched/enable yield -count=`head -n 100 trace | grep -v ^# | awk '{ print $5 }' | sort -u | wc -l` +count=`check_unique` if [ $count -lt 3 ]; then fail "at least fork, exec and exit events should be recorded" fi -- 2.47.2

4 weeks, 1 day

2
2
0 0

[PATCH v2] selftests: firmware: Add details in error logging

by Harshal

Specify details in logs of failed cases Signed-off-by: Harshal <embedkari167(a)gmail.com> --- v2: - revert back to exit() instead of die() to avoid modifying system behaviour v1: https://lore.kernel.org/all/c7c071ed-6a4e-4a9c-ba9d-c745fd42c22f@linuxfound… tools/testing/selftests/firmware/fw_namespace.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/tools/testing/selftests/firmware/fw_namespace.c b/tools/testing/selftests/firmware/fw_namespace.c index 04757dc7e546..5b0032498ede 100644 --- a/tools/testing/selftests/firmware/fw_namespace.c +++ b/tools/testing/selftests/firmware/fw_namespace.c @@ -38,7 +38,7 @@ static void trigger_fw(const char *fw_name, const char *sys_path) fd = open(sys_path, O_WRONLY); if (fd < 0) - die("open failed: %s\n", + die("open of sys_path failed: %s\n", strerror(errno)); if (write(fd, fw_name, strlen(fw_name)) != strlen(fw_name)) exit(EXIT_FAILURE); @@ -52,10 +52,10 @@ static void setup_fw(const char *fw_path) fd = open(fw_path, O_WRONLY | O_CREAT, 0600); if (fd < 0) - die("open failed: %s\n", + die("open of firmware file failed: %s\n", strerror(errno)); if (write(fd, fw, sizeof(fw) -1) != sizeof(fw) -1) - die("write failed: %s\n", + die("write to firmware file failed: %s\n", strerror(errno)); close(fd); } @@ -66,7 +66,7 @@ static bool test_fw_in_ns(const char *fw_name, const char *sys_path, bool block_ if (block_fw_in_parent_ns) if (mount("test", "/lib/firmware", "tmpfs", MS_RDONLY, NULL) == -1) - die("blocking firmware in parent ns failed\n"); + die("blocking firmware in parent namespace failed\n"); child = fork(); if (child == -1) { @@ -99,11 +99,11 @@ static bool test_fw_in_ns(const char *fw_name, const char *sys_path, bool block_ strerror(errno)); } if (mount(NULL, "/", NULL, MS_SLAVE|MS_REC, NULL) == -1) - die("remount root in child ns failed\n"); + die("remount root in child namespace failed\n"); if (!block_fw_in_parent_ns) { if (mount("test", "/lib/firmware", "tmpfs", MS_RDONLY, NULL) == -1) - die("blocking firmware in child ns failed\n"); + die("blocking firmware in child namespace failed\n"); } else umount("/lib/firmware"); @@ -129,8 +129,8 @@ int main(int argc, char **argv) die("error: failed to build full fw_path\n"); setup_fw(fw_path); - setvbuf(stdout, NULL, _IONBF, 0); + /* Positive case: firmware in PID1 mount namespace */ printf("Testing with firmware in parent namespace (assumed to be same file system as PID1)\n"); if (!test_fw_in_ns(fw_name, sys_path, false)) -- 2.43.0

1 month

3
3
0 0

[PATCH v18 4/5] binder: add transaction_report feature entry

by Carlos Llamas

From: Li Li <dualli(a)google.com> Add "transaction_report" to the binderfs feature list, to help userspace determine if the "BINDER_CMD_REPORT" generic netlink api is supported by the binder driver. Signed-off-by: Li Li <dualli(a)google.com> Signed-off-by: Carlos Llamas <cmllamas(a)google.com> --- drivers/android/binderfs.c | 8 ++++++++ .../selftests/filesystems/binderfs/binderfs_test.c | 1 + 2 files changed, 9 insertions(+) diff --git a/drivers/android/binderfs.c b/drivers/android/binderfs.c index 4f827152d18e..f74a7e380261 100644 --- a/drivers/android/binderfs.c +++ b/drivers/android/binderfs.c @@ -59,6 +59,7 @@ struct binder_features { bool oneway_spam_detection; bool extended_error; bool freeze_notification; + bool transaction_report; }; static const struct constant_table binderfs_param_stats[] = { @@ -76,6 +77,7 @@ static struct binder_features binder_features = { .oneway_spam_detection = true, .extended_error = true, .freeze_notification = true, + .transaction_report = true, }; static inline struct binderfs_info *BINDERFS_SB(const struct super_block *sb) @@ -616,6 +618,12 @@ static int init_binder_features(struct super_block *sb) if (IS_ERR(dentry)) return PTR_ERR(dentry); + dentry = binderfs_create_file(dir, "transaction_report", + &binder_features_fops, + &binder_features.transaction_report); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + return 0; } diff --git a/tools/testing/selftests/filesystems/binderfs/binderfs_test.c b/tools/testing/selftests/filesystems/binderfs/binderfs_test.c index 81db85a5cc16..39a68078a79b 100644 --- a/tools/testing/selftests/filesystems/binderfs/binderfs_test.c +++ b/tools/testing/selftests/filesystems/binderfs/binderfs_test.c @@ -65,6 +65,7 @@ static int __do_binderfs_test(struct __test_metadata *_metadata) "oneway_spam_detection", "extended_error", "freeze_notification", + "transaction_report", }; change_mountns(_metadata); -- 2.50.1.470.g6ba607880d-goog

1 month

1
0
0 0

Re: [PATCH net] selftests: rtnetlink.sh: remove esp4_offload after test

by Shannon Nelson

(yeah, I made the same non-ASCII mistake...) On 7/24/25 10:49 AM, Shannon Nelson wrote: > On 7/24/25 1:20 AM, Xiumei Mu wrote: >> resent the reply again with "plain text mode" >> >> On Thu, Jul 24, 2025 at 2:25 PM Hangbin Liu<liuhangbin(a)gmail.com> wrote: >>> Hi Xiumei, >>> On Thu, Jul 24, 2025 at 12:55:02PM +0800, Xiumei Mu wrote: >>>> The esp4_offload module, loaded during IPsec offload tests, should >>>> be reset to its default settings after testing. >>>> Otherwise, leaving it enabled could unintentionally affect subsequence >>>> test cases by keeping offload active. >>> Would you please show which subsequence test will be affected? >> Any general ipsec case, which expects to be tested by default >> behavior(without offload). >> esp4_offload will affect the performance. >> >>>> Fixes: 2766a11161cc ("selftests: rtnetlink: add ipsec offload API test") >>> It would be good to Cc the fix commit author. You can use >>> `./scripts/get_maintainer.pl your_patch_file` to get the contacts you >>> need to Cc. >> I used the script to generate the cc list. >> and I double checked the old email of the author is invalid >> added his personal email in the cc list: >> >> Shannon Nelson<shannon.nelson(a)oracle.com>. -----> Shannon Nelson >> <sln(a)onemain.com> >> >> get the information from here: >> https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git/commit/?id=a… > > Yep, that was me a couple of corporate email addresses ago. Thanks for > digging up the new email address. Luckily I have a couple of mail > filters watching for old email addresses. > >>>> Signed-off-by: Xiumei Mu<xmu(a)redhat.com> >>>> --- >>>> tools/testing/selftests/net/rtnetlink.sh | 6 ++++++ >>>> 1 file changed, 6 insertions(+) >>>> >>>> diff --git a/tools/testing/selftests/net/rtnetlink.sh b/tools/testing/selftests/net/rtnetlink.sh >>>> index 2e8243a65b50..5cc1b5340a1a 100755 >>>> --- a/tools/testing/selftests/net/rtnetlink.sh >>>> +++ b/tools/testing/selftests/net/rtnetlink.sh >>>> @@ -673,6 +673,11 @@ kci_test_ipsec_offload() >>>> sysfsf=$sysfsd/ipsec >>>> sysfsnet=/sys/bus/netdevsim/devices/netdevsim0/net/ >>>> probed=false >>>> + esp4_offload_probed_default=false >>>> + >>>> + if lsmod | grep -q esp4_offload; then >>>> + esp4_offload_probed_default=true >>>> + fi >>> If the mode is loaded by default, how to avoid the subsequence test to be >>> failed? >> The module is not loaded by default, but some users or testers may >> need to load esp4_offload in their own environments. >> Therefore, resetting it to the default configuration is the best >> practice to prevent this self-test case from impacting subsequent >> tests > > Seems reasonable to me. > >>>> if ! mount | grep -q debugfs; then >>>> mount -t debugfs none /sys/kernel/debug/ &> /dev/null >>>> @@ -766,6 +771,7 @@ EOF >>>> fi >>>> >>>> # clean up any leftovers >>>> + [ $esp4_offload_probed_default == false ] && rmmod esp4_offload >>> The new patch need to pass shellcheck. We need to double quote the variable. >> Thanks your comment, I will add double quote in patchv2 > > Or you keep with the existing style as done a line or two later: > $esp4_offload_probed_default && rmmod esp4_offload Either way, > Reviewed-by: Shannon Nelson <sln(a)onemain.com> Cheers, sln >>> Thanks >>> Hangbin >>>> echo 0 > /sys/bus/netdevsim/del_device >>>> $probed && rmmod netdevsim >>>> >>>> -- >>>> 2.50.1 >>>> >

1 month

1
0
0 0

[PATCH bpf-next v3 0/4] bpf: Show precise rejected function when attaching to __noreturn and deny list functions

by KaFai Wan

Show precise rejected function when attaching fexit/fmod_ret to __noreturn functions. Add log for attaching tracing programs to functions in deny list. Add selftest for attaching tracing programs to functions in deny list. Migrate fexit_noreturns case into tracing_failure test suite. changes: v3: - add tracing_deny case into existing files (Alexei) - migrate fexit_noreturns into tracing_failure - change SOB v2: - change verifier log message (Alexei) - add missing Suggested-by https://lore.kernel.org/bpf/20250714120408.1627128-1-mannkafai@gmail.com/ v1: https://lore.kernel.org/all/20250710162717.3808020-1-mannkafai@gmail.com/ --- KaFai Wan (4): bpf: Show precise rejected function when attaching fexit/fmod_ret to __noreturn functions bpf: Add log for attaching tracing programs to functions in deny list selftests/bpf: Add selftest for attaching tracing programs to functions in deny list selftests/bpf: Migrate fexit_noreturns case into tracing_failure test suite kernel/bpf/verifier.c | 5 +- .../bpf/prog_tests/fexit_noreturns.c | 9 ---- .../bpf/prog_tests/tracing_failure.c | 52 +++++++++++++++++++ .../selftests/bpf/progs/fexit_noreturns.c | 15 ------ .../selftests/bpf/progs/tracing_failure.c | 12 +++++ 5 files changed, 68 insertions(+), 25 deletions(-) delete mode 100644 tools/testing/selftests/bpf/prog_tests/fexit_noreturns.c delete mode 100644 tools/testing/selftests/bpf/progs/fexit_noreturns.c -- 2.43.0

1 month

3
11
0 0

[PATCH net-next V6 0/5] selftests: drv-net: Test XDP native support

by Mohsin Bashir

This patch series add tests to validate XDP native support for PASS, DROP, ABORT, and TX actions, as well as headroom and tailroom adjustment. For adjustment tests, validate support for both the extension and shrinking cases across various packet sizes and offset values. The pass criteria for head/tail adjustment tests require that at-least one adjustment value works for at-least one packet size. This ensure that the variability in maximum supported head/tail adjustment offset across different drivers is being incorporated. The results reported in this series are based on netdevsim. However, the series is tested against multiple other drivers including fbnic. Note: The XDP support for fbnic will be added later. --- Change-log: - Force checksum computation in netdevsim xdp hook - Update checksum when updating packet for head/tail adjustment cases - Use 1 as the minimum value for the data growth while adjusting tail V5: https://lore.kernel.org/netdev/20250715210553.1568963-6-mohsin.bashr@gmail.… V4: https://lore.kernel.org/netdev/20250714210352.1115230-1-mohsin.bashr@gmail.… V3: https://lore.kernel.org/netdev/20250712002648.2385849-1-mohsin.bashr@gmail.… V2: https://lore.kernel.org/netdev/20250710184351.63797-1-mohsin.bashr@gmail.com V1: https://lore.kernel.org/netdev/20250709173707.3177206-1-mohsin.bashr@gmail.… Jakub Kicinski (1): net: netdevsim: hook in XDP handling Mohsin Bashir (4): selftests: drv-net: Test XDP_PASS/DROP support selftests: drv-net: Test XDP_TX support selftests: drv-net: Test tail-adjustment support selftests: drv-net: Test head-adjustment support drivers/net/netdevsim/netdev.c | 21 +- tools/testing/selftests/drivers/net/Makefile | 1 + tools/testing/selftests/drivers/net/xdp.py | 656 ++++++++++++++++++ .../selftests/net/lib/xdp_native.bpf.c | 621 +++++++++++++++++ 4 files changed, 1298 insertions(+), 1 deletion(-) create mode 100755 tools/testing/selftests/drivers/net/xdp.py create mode 100644 tools/testing/selftests/net/lib/xdp_native.bpf.c -- 2.47.1

1 month

8
20
0 0

[PATCH v3 0/4] procfs: make reference pidns more user-visible

by Aleksa Sarai

Ever since the introduction of pid namespaces, procfs has had very implicit behaviour surrounding them (the pidns used by a procfs mount is auto-selected based on the mounting process's active pidns, and the pidns itself is basically hidden once the mount has been constructed). /* pidns mount option for procfs */ This implicit behaviour has historically meant that userspace was required to do some special dances in order to configure the pidns of a procfs mount as desired. Examples include: * In order to bypass the mnt_too_revealing() check, Kubernetes creates a procfs mount from an empty pidns so that user namespaced containers can be nested (without this, the nested containers would fail to mount procfs). But this requires forking off a helper process because you cannot just one-shot this using mount(2). * Container runtimes in general need to fork into a container before configuring its mounts, which can lead to security issues in the case of shared-pidns containers (a privileged process in the pidns can interact with your container runtime process). While SUID_DUMP_DISABLE and user namespaces make this less of an issue, the strict need for this due to a minor uAPI wart is kind of unfortunate. Things would be much easier if there was a way for userspace to just specify the pidns they want. Patch 1 implements a new "pidns" argument which can be set using fsconfig(2): fsconfig(procfd, FSCONFIG_SET_FD, "pidns", NULL, nsfd); fsconfig(procfd, FSCONFIG_SET_STRING, "pidns", "/proc/self/ns/pid", 0); or classic mount(2) / mount(8): // mount -t proc -o pidns=/proc/self/ns/pid proc /tmp/proc mount("proc", "/tmp/proc", "proc", MS_..., "pidns=/proc/self/ns/pid"); The initial security model I have in this RFC is to be as conservative as possible and just mirror the security model for setns(2) -- which means that you can only set pidns=... to pid namespaces that your current pid namespace is a direct ancestor of and you have CAP_SYS_ADMIN privileges over the pid namespace. This fulfils the requirements of container runtimes, but I suspect that this may be too strict for some usecases. The pidns argument is not displayed in mountinfo -- it's not clear to me what value it would make sense to show (maybe we could just use ns_dname to provide an identifier for the namespace, but this number would be fairly useless to userspace). I'm open to suggestions. Note that PROCFS_GET_PID_NAMESPACE (see below) does at least let userspace get information about this outside of mountinfo. Note that you cannot change the pidns of an already-created procfs instance. The primary reason is that allowing this to be changed would require RCU-protecting proc_pid_ns(sb) and thus auditing all of fs/proc/* and some of the users in fs/* to make sure they wouldn't UAF the pid namespace. Since creating procfs instances is very cheap, it seems unnecessary to overcomplicate this upfront. Trying to reconfigure procfs this way errors out with -EBUSY. /* ioctl(PROCFS_GET_PID_NAMESPACE) */ In addition, being able to figure out what pid namespace is being used by a procfs mount is quite useful when you have an administrative process (such as a container runtime) which wants to figure out the correct way of mapping PIDs between its own namespace and the namespace for procfs (using NS_GET_{PID,TGID}_{IN,FROM}_PIDNS). There are alternative ways to do this, but they all rely on ancillary information that third-party libraries and tools do not necessarily have access to. To make this easier, add a new ioctl (PROCFS_GET_PID_NAMESPACE) which can be used to get a reference to the pidns that a procfs is using. It's not quite clear what is the correct security model for this API, but the current approach I've taken is to: * Make the ioctl only valid on the root (meaning that a process without access to the procfs root -- such as only having an fd to a procfs file or some open_tree(2)-like subset -- cannot use this API). * Require that the process requesting either has access to /proc/1/ns/pid anyway (i.e. has ptrace-read access to the pidns pid1), has CAP_SYS_ADMIN access to the pidns (i.e. has administrative access to it and can join it if they had a handle), or is in a pidns that is a direct ancestor of the target pidns (i.e. all of the pids are already visible in the procfs for the current process's pidns). The security model for this is a little loose, as it seems to me that all of the cases mentioned are valid cases to allow access, but I'm open to suggestions for whether we need to make this stricter or looser. Signed-off-by: Aleksa Sarai <cyphar(a)cyphar.com> --- Changes in v3: - Disallow changing pidns for existing procfs instances, as we'd probably have to RCU-protect everything that touches the pinned pidns reference. - Improve tests with slightly nicer ASSERT_ERRNO* macros. - v2: <https://lore.kernel.org/r/20250723-procfs-pidns-api-v2-0-621e7edd8e40@cypha…> Changes in v2: - #ifdef CONFIG_PID_NS - Improve cover letter wording to make it clear we're talking about two separate features with different permission models. [Andy Lutomirski] - Fix build warnings in pidns_is_ancestor() patch. [kernel test robot] - v1: <https://lore.kernel.org/r/20250721-procfs-pidns-api-v1-0-5cd9007e512d@cypha…> --- Aleksa Sarai (4): pidns: move is-ancestor logic to helper procfs: add "pidns" mount option procfs: add PROCFS_GET_PID_NAMESPACE ioctl selftests/proc: add tests for new pidns APIs Documentation/filesystems/proc.rst | 12 ++ fs/proc/root.c | 156 +++++++++++++++++- include/linux/pid_namespace.h | 9 ++ include/uapi/linux/fs.h | 3 + kernel/pid_namespace.c | 23 ++- tools/testing/selftests/proc/.gitignore | 1 + tools/testing/selftests/proc/Makefile | 1 + tools/testing/selftests/proc/proc-pidns.c | 252 ++++++++++++++++++++++++++++++ 8 files changed, 441 insertions(+), 16 deletions(-) --- base-commit: 66639db858112bf6b0f76677f7517643d586e575 change-id: 20250717-procfs-pidns-api-8ed1583431f0 Best regards, -- Aleksa Sarai <cyphar(a)cyphar.com>

1 month

1
4
0 0

[PATCH net] selftests: rtnetlink.sh: remove esp4_offload after test

by Xiumei Mu

The esp4_offload module, loaded during IPsec offload tests, should be reset to its default settings after testing. Otherwise, leaving it enabled could unintentionally affect subsequence test cases by keeping offload active. Fixes: 2766a11161cc ("selftests: rtnetlink: add ipsec offload API test") Signed-off-by: Xiumei Mu <xmu(a)redhat.com> --- tools/testing/selftests/net/rtnetlink.sh | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/tools/testing/selftests/net/rtnetlink.sh b/tools/testing/selftests/net/rtnetlink.sh index 2e8243a65b50..5cc1b5340a1a 100755 --- a/tools/testing/selftests/net/rtnetlink.sh +++ b/tools/testing/selftests/net/rtnetlink.sh @@ -673,6 +673,11 @@ kci_test_ipsec_offload() sysfsf=$sysfsd/ipsec sysfsnet=/sys/bus/netdevsim/devices/netdevsim0/net/ probed=false + esp4_offload_probed_default=false + + if lsmod | grep -q esp4_offload; then + esp4_offload_probed_default=true + fi if ! mount | grep -q debugfs; then mount -t debugfs none /sys/kernel/debug/ &> /dev/null @@ -766,6 +771,7 @@ EOF fi # clean up any leftovers + [ $esp4_offload_probed_default == false ] && rmmod esp4_offload echo 0 > /sys/bus/netdevsim/del_device $probed && rmmod netdevsim -- 2.50.1

1 month

2
2
0 0

[PATCH] selftests: netfilter: ipvs.sh: Explicity disable rp_filter on interface tunl0

by Yi Chen

Although setup_ns() set net.ipv4.conf.default.rp_filter=0, loading certain module such as ipip will automatically create a tunl0 interface in all netns including new created ones, this in script is before than default.rp_filter=0 applied, as a result tunl0.rp_filter remains set to 1 which causes the test report FAIL when ipip module is preloaded. Before fix: Testing DR mode... Testing NAT mode... Testing Tunnel mode... ipvs.sh: FAIL After fix: Testing DR mode... Testing NAT mode... Testing Tunnel mode... ipvs.sh: PASS Fixes: ("7c8b89ec5 selftests: netfilter: remove rp_filter configuration") Signed-off-by: Yi Chen <yiche(a)redhat.com> --- tools/testing/selftests/net/netfilter/ipvs.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/netfilter/ipvs.sh b/tools/testing/selftests/net/netfilter/ipvs.sh index 6af2ea3ad6b8..9c9d5b38ab71 100755 --- a/tools/testing/selftests/net/netfilter/ipvs.sh +++ b/tools/testing/selftests/net/netfilter/ipvs.sh @@ -151,7 +151,7 @@ test_nat() { test_tun() { ip netns exec "${ns0}" ip route add "${vip_v4}" via "${gip_v4}" dev br0 - ip netns exec "${ns1}" modprobe -q ipip + modprobe -q ipip ip netns exec "${ns1}" ip link set tunl0 up ip netns exec "${ns1}" sysctl -qw net.ipv4.ip_forward=0 ip netns exec "${ns1}" sysctl -qw net.ipv4.conf.all.send_redirects=0 @@ -160,10 +160,10 @@ test_tun() { ip netns exec "${ns1}" ipvsadm -a -i -t "${vip_v4}:${port}" -r ${rip_v4}:${port} ip netns exec "${ns1}" ip addr add ${vip_v4}/32 dev lo:1 - ip netns exec "${ns2}" modprobe -q ipip ip netns exec "${ns2}" ip link set tunl0 up ip netns exec "${ns2}" sysctl -qw net.ipv4.conf.all.arp_ignore=1 ip netns exec "${ns2}" sysctl -qw net.ipv4.conf.all.arp_announce=2 + ip netns exec "${ns2}" sysctl -qw net.ipv4.conf.tunl0.rp_filter=0 ip netns exec "${ns2}" ip addr add "${vip_v4}/32" dev lo:1 test_service -- 2.50.1

1 month

2
1
0 0

[PATCH v26 net-next 0/6] DUALPI2 patch

by chia-yu.chang＠nokia-bell-labs.com

From: Chia-Yu Chang <chia-yu.chang(a)nokia-bell-labs.com> Hello, Please find the DualPI2 patch v26. This patch serise adds DualPI Improved with a Square (DualPI2) with following features: * Supports congestion controls that comply with the Prague requirements in RFC9331 (e.g. TCP-Prague) * Coupled dual-queue that separates the L4S traffic in a low latency queue (L-queue), without harming remaining traffic that is scheduled in classic queue (C-queue) due to congestion-coupling using PI2 as defined in RFC9332 * Configurable overload strategies * Use of sojourn time to reliably estimate queue delay * Supports ECN L4S-identifier (IP.ECN==0b*1) to classify traffic into respective queues For more details of DualPI2, please refer IETF RFC9332 (https://datatracker.ietf.org/doc/html/rfc9332). Best regards, Chia-Yu --- v25 (19-Jul-2025) and v26 (22-Jul-2025) - Restruct to avoid using lock and unlock when both step_thresh are provided (Jakub Kicinski <kuba(a)kernel.org>) v24 (18-Jul-2025) - Replace TCA_DUALPI2 prefix with TC_DUALPI2 for enums in pkt_sched.h (Jakub Kicinski <kuba(a)kernel.org>) - Report error if both packet and time step thresholds are provided (Jakub Kicinski <kuba(a)kernel.org>) v22 (11-Jul-2025) and v23 (13-Jul-2025) - Fix issue when user would like to change DualPI2 but provides an empty TCA_OPTIONS with no nested attributes (Paolo Abeni <pabeni(a)redhat.com>, Jakub Kicinski <kuba(a)kernel.org>) v21 (02-Jul-2025) - Replace STEP_THRESH and STEP_PACKETS with STEP_THRESH_PKTS and STEP_THRESH_US (Jakub Kicinski <kuba(a)kernel.org>) - Move READ_ONCE and WRITE_ONCE to later DualPI2 patches (Jakub Kicinski <kuba(a)kernel.org>) - Replace NLA_POLICY_FULL_RANGE with NLA_POLICY_RANGE (Jakub Kicinski <kuba(a)kernel.org>) - Set extra error message for dualpi2_change (Jakub Kicinski <kuba(a)kernel.org>) - Drop redundant else for better readability (Paolo Abeni <pabeni(a)redhat.com>) - Replace step-thresh and step-packets with step-thresh-pkts and step-thresh-us (Jakub Kicinski <kuba(a)kernel.org>) - Remove redundant name-prefix and simplify entries of dualpi2 enums (Jakub Kicinski <kuba(a)kernel.org>) - Fix some typos and format issues of dualpi2 attributes v20 (21-Jun-2025) - Add one more commit to fix warning and style check on tdc.sh reported by shellcheck - Remove double-prefixed of "tc_tc_dualpi2_attrs" in tc-user.h (Donald Hunter <donald.hunter(a)gmail.com>) v19 (14-Jun-2025) - Fix one typo in the comment of #1 (ALOK TIWARI <alok.a.tiwari(a)oracle.com>) - Update commit message of #4 (ALOK TIWARI <alok.a.tiwari(a)oracle.com>) - Wrap long lines of Documentation/netlink/specs/tc.yaml to within 80 characters (Jakub Kicinski <kuba(a)kernel.org>) v18 (13-Jun-2025) - Add the num of enum used by DualPI2 and fix name and name-prefix of DualPI2 enum and attribute - Replace from_timer() with timer_container_of() (Pedro Tammela <pctammela(a)mojatatu.com>) v17 (25-May-2025, Resent at 11-Jun-2025) - Replace 0xffffffff with U32_MAX (Paolo Abeni <pabeni(a)redhat.com>) - Use helper function qdisc_dequeue_internal() and add new helper function skb_apply_step() (Paolo Abeni <pabeni(a)redhat.com>) - Add s64 casting when calculating the delta of the PI controller (Paolo Abeni <pabeni(a)redhat.com>) - Change the drop reason into SKB_DROP_REASON_QDISC_CONGESTED for drop_early (Paolo Abeni <pabeni(a)redhat.com>) - Modify the condition to remove the original skb when enqueuing multiple GSO segments (Paolo Abeni <pabeni(a)redhat.com>) - Add READ_ONCE() in dualpi2_dump_stat() (Paolo Abeni <pabeni(a)redhat.com>) - Add comments, brackets, and brackets for readability (Paolo Abeni <pabeni(a)redhat.com>) v16 (16-MAy-2025) - Add qdisc_lock() to dualpi2_timer() in dualpi2_timer (Paolo Abeni <pabeni(a)redhat.com>) - Introduce convert_ns_to_usec() to convert usec to nsec without overflow in #1 (Paolo Abeni <pabeni(a)redhat.com>) - Update convert_us_tonsec() to convert nsec to usec without overflow in #2 (Paolo Abeni <pabeni(a)redhat.com>) - Add more descriptions with respect to DualPI2 in the cover ltter and add changelog in each patch (Paolo Abeni <pabeni(a)redhat.com>) v15 (09-May-2025) - Add enum of TCA_DUALPI2_ECN_MASK_CLA_ECT to remove potential leakeage in #1 (Simon Horman <horms(a)kernel.org>) - Fix one typo in comment of #2 - Update tc.yaml in #5 to aligh with the updated enum of pkt_sched.h v14 (05-May-2025) - Modify tc.yaml: (1) Replace flags with enum and remove enum-as-flags, (2) Remove credit-queue in xstats, and (3) Change attribute types (Donald Hunter <donald.hun - Add enum and fix the ordering of variables in pkt_sched.h to align with the modified tc.yaml (Donald Hunter <donald.hunter(a)gmail.com>) - Add validators for DROP_OVERLOAD, DROP_EARLY, ECN_MASK, and SPLIT_GSO in sch_dualpi2.c (Donald Hunter <donald.hunter(a)gmail.com>) - Update dualpi2.json to align with the updated variable order in pkt_sched.h - Reorder patches (Donald Hunter <donald.hunter(a)gmail.com>) v13 (26-Apr-2025) - Use dashes in member names to follow YNL conventions in tc.yaml (Donald Hunter <donald.hunter(a)gmail.com>) - Define enumerations separately for flags of drop-early, drop-overload, ecn-mask, credit-queue in tc.yaml (Donald Hunter <donald.hunter(a)gmail.com>) - Change the types of split-gso and step-packets into flag in tc.yaml (Donald Hunter <donald.hunter(a)gmail.com>) - Revert to u32/u8 types for tc-dualpi2-xstats members in tc.yaml (Donald Hunter <donald.hunter(a)gmail.com>) - Add new test cases in tc-tests/qdiscs/dualpi2.json to cover all dualpi2 parameters (Donald Hunter <donald.hunter(a)gmail.com>) - Change the type of TCA_DUALPI2_STEP_PACKETS into NLA_FLAG (Donald Hunter <donald.hunter(a)gmail.com>) v12 (22-Apr-2025) - Remove anonymous struct in sch_dualpi2.c (Paolo Abeni <pabeni(a)redhat.com>) - Replace u32/u8 with uint and s32 with int in tc spec document (Paolo Abeni <pabeni(a)redhat.com>) - Introduce get_memory_limit function to handle potential overflow when multipling limit with MTU (Paolo Abeni <pabeni(a)redhat.com>) - Double the packet length to further include packet overhead in memory_limit (Paolo Abeni <pabeni(a)redhat.com>) - Remove the check of qdisc_qlen(sch) when calling qdisc_tree_reduce_backlog (Paolo Abeni <pabeni(a)redhat.com>) v11 (15-Apr-2025) - Replace hstimer_init with hstimer_setup in sch_dualpi2.c v10 (25-Mar-2025) - Remove leftover include in include/linux/netdevice.h and anonymous struct in sch_dualpi2.c (Paolo Abeni <pabeni(a)redhat.com>) - Use kfree_skb_reason() and add SKB_DROP_REASON_DUALPI2_STEP_DROP drop reason (Paolo Abeni <pabeni(a)redhat.com>) - Split sch_dualpi2.c into 3 patches (and overall 5 patches): Struct definition & parsing, Dump stats & configuration, Enqueue/Dequeue (Paolo Abeni <pabeni(a)redhat.com>) v9 (16-Mar-2025) - Fix mem_usage error in previous version - Add min_qlen_step to the dualpi2 attribute as the minimum queue length in number of packets in the L-queue to start step threshold marking. In previous versions, this value was fixed to 2, so the step threshold was applied to mark packets in the L queue only when the queue length of the L queue was greater than or equal to 2 packets. This will cause larger queuing delays for L4S traffic at low rates (<20Mbps). So we parameterize it and change the default value to 0. Comparison of tcp_1down run 'HTB 20Mbit + DUALPI2 + 10ms base delay' Old versions: avg median # data pts Ping (ms) ICMP : 11.55 11.70 ms 350 TCP upload avg : 18.96 N/A Mbits/s 350 TCP upload sum : 18.96 N/A Mbits/s 350 New version (v9): avg median # data pts Ping (ms) ICMP : 10.81 10.70 ms 350 TCP upload avg : 18.91 N/A Mbits/s 350 TCP upload sum : 18.91 N/A Mbits/s 350 Comparison of tcp_1down run 'HTB 10Mbit + DUALPI2 + 10ms base delay' Old versions: avg median # data pts Ping (ms) ICMP : 12.61 12.80 ms 350 TCP upload avg : 9.48 N/A Mbits/s 350 TCP upload sum : 9.48 N/A Mbits/s 350 New version (v9): avg median # data pts Ping (ms) ICMP : 11.06 10.80 ms 350 TCP upload avg : 9.43 N/A Mbits/s 350 TCP upload sum : 9.43 N/A Mbits/s 350 Comparison of tcp_1down run 'HTB 10Mbit + DUALPI2 + 10ms base delay' Old versions: avg median # data pts Ping (ms) ICMP : 40.86 37.45 ms 350 TCP upload avg : 0.88 N/A Mbits/s 350 TCP upload sum : 0.88 N/A Mbits/s 350 TCP upload::1 : 0.88 0.97 Mbits/s 350 New version (v9): avg median # data pts Ping (ms) ICMP : 11.07 10.40 ms 350 TCP upload avg : 0.55 N/A Mbits/s 350 TCP upload sum : 0.55 N/A Mbits/s 350 TCP upload::1 : 0.55 0.59 Mbits/s 350 v8 (11-Mar-2025) - Fix warning messages in v7 v7 (07-Mar-2025) - Separate into 3 patches to avoid mixing changes of documentation, selftest, and code. (Cong Wang <xiyou.wangcong(a)gmail.com>) v6 (04-Mar-2025) - Add modprobe for dulapi2 in tc-testing script tc-testing/tdc.sh (Jakub Kicinski <kuba(a)kernel.org>) - Update test cases in dualpi2.json - Update commit message v5 (22-Feb-2025) - A comparison was done between MQ + DUALPI2, MQ + FQ_PIE, MQ + FQ_CODEL: Unshaped 1gigE with 4 download streams test: - Summary of tcp_4down run 'MQ + FQ_CODEL': avg median # data pts Ping (ms) ICMP : 1.19 1.34 ms 349 TCP download avg : 235.42 N/A Mbits/s 349 TCP download sum : 941.68 N/A Mbits/s 349 TCP download::1 : 235.19 235.39 Mbits/s 349 TCP download::2 : 235.03 235.35 Mbits/s 349 TCP download::3 : 236.89 235.44 Mbits/s 349 TCP download::4 : 234.57 235.19 Mbits/s 349 - Summary of tcp_4down run 'MQ + FQ_PIE' avg median # data pts Ping (ms) ICMP : 1.21 1.37 ms 350 TCP download avg : 235.42 N/A Mbits/s 350 TCP download sum : 941.61 N/A Mbits/s 350 TCP download::1 : 232.54 233.13 Mbits/s 350 TCP download::2 : 232.52 232.80 Mbits/s 350 TCP download::3 : 233.14 233.78 Mbits/s 350 TCP download::4 : 243.41 241.48 Mbits/s 350 - Summary of tcp_4down run 'MQ + DUALPI2' avg median # data pts Ping (ms) ICMP : 1.19 1.34 ms 349 TCP download avg : 235.42 N/A Mbits/s 349 TCP download sum : 941.68 N/A Mbits/s 349 TCP download::1 : 235.19 235.39 Mbits/s 349 TCP download::2 : 235.03 235.35 Mbits/s 349 TCP download::3 : 236.89 235.44 Mbits/s 349 TCP download::4 : 234.57 235.19 Mbits/s 349 Unshaped 1gigE with 128 download streams test: - Summary of tcp_128down run 'MQ + FQ_CODEL': avg median # data pts Ping (ms) ICMP : 1.88 1.86 ms 350 TCP download avg : 7.39 N/A Mbits/s 350 TCP download sum : 946.47 N/A Mbits/s 350 - Summary of tcp_128down run 'MQ + FQ_PIE': avg median # data pts Ping (ms) ICMP : 1.88 1.86 ms 350 TCP download avg : 7.39 N/A Mbits/s 350 TCP download sum : 946.47 N/A Mbits/s 350 - Summary of tcp_128down run 'MQ + DUALPI2': avg median # data pts Ping (ms) ICMP : 1.88 1.86 ms 350 TCP download avg : 7.39 N/A Mbits/s 350 TCP download sum : 946.47 N/A Mbits/s 350 Unshaped 10gigE with 4 download streams test: - Summary of tcp_4down run 'MQ + FQ_CODEL': avg median # data pts Ping (ms) ICMP : 0.22 0.23 ms 350 TCP download avg : 2354.08 N/A Mbits/s 350 TCP download sum : 9416.31 N/A Mbits/s 350 TCP download::1 : 2353.65 2352.81 Mbits/s 350 TCP download::2 : 2354.54 2354.21 Mbits/s 350 TCP download::3 : 2353.56 2353.78 Mbits/s 350 TCP download::4 : 2354.56 2354.45 Mbits/s 350 - Summary of tcp_4down run 'MQ + FQ_PIE': avg median # data pts Ping (ms) ICMP : 0.20 0.19 ms 350 TCP download avg : 2354.76 N/A Mbits/s 350 TCP download sum : 9419.04 N/A Mbits/s 350 TCP download::1 : 2354.77 2353.89 Mbits/s 350 TCP download::2 : 2353.41 2354.29 Mbits/s 350 TCP download::3 : 2356.18 2354.19 Mbits/s 350 TCP download::4 : 2354.68 2353.15 Mbits/s 350 - Summary of tcp_4down run 'MQ + DUALPI2': avg median # data pts Ping (ms) ICMP : 0.24 0.24 ms 350 TCP download avg : 2354.11 N/A Mbits/s 350 TCP download sum : 9416.43 N/A Mbits/s 350 TCP download::1 : 2354.75 2353.93 Mbits/s 350 TCP download::2 : 2353.15 2353.75 Mbits/s 350 TCP download::3 : 2353.49 2353.72 Mbits/s 350 TCP download::4 : 2355.04 2353.73 Mbits/s 350 Unshaped 10gigE with 128 download streams test: - Summary of tcp_128down run 'MQ + FQ_CODEL': avg median # data pts Ping (ms) ICMP : 7.57 8.69 ms 350 TCP download avg : 73.97 N/A Mbits/s 350 TCP download sum : 9467.82 N/A Mbits/s 350 - Summary of tcp_128down run 'MQ + FQ_PIE': avg median # data pts Ping (ms) ICMP : 7.82 8.91 ms 350 TCP download avg : 73.97 N/A Mbits/s 350 TCP download sum : 9468.42 N/A Mbits/s 350 - Summary of tcp_128down run 'MQ + DUALPI2': avg median # data pts Ping (ms) ICMP : 6.87 7.93 ms 350 TCP download avg : 73.95 N/A Mbits/s 350 TCP download sum : 9465.87 N/A Mbits/s 350 From the results shown above, we see small differences between combinations. - Update commit message to include results of no_split_gso and split_gso (Dave Taht <dave.taht(a)gmail.com> and Paolo Abeni <pabeni(a)redhat.com>) - Add memlimit in the dualpi2 attribute, and add memory_used, max_memory_used, memory_limit in dualpi2 stats (Dave Taht <dave.taht(a)gmail.com>) - Update note in sch_dualpi2.c related to BBRv3 status (Dave Taht <dave.taht(a)gmail.com>) - Update license identifier (Dave Taht <dave.taht(a)gmail.com>) - Add selftest in tools/testing/selftests/tc-testing (Cong Wang <xiyou.wangcong(a)gmail.com>) - Use netlink policies for parameter checks (Jamal Hadi Salim <jhs(a)mojatatu.com>) - Modify texts & fix typos in Documentation/netlink/specs/tc.yaml (Dave Taht <dave.taht(a)gmail.com>) - Add descriptions of packet counter statistics and the reset function of sch_dualpi2.c - Fix step_thresh in packets - Update code comments in sch_dualpi2.c v4 (22-Oct-2024) - Update statement in Kconfig for DualPI2 (Stephen Hemminger <stephen(a)networkplumber.org>) - Put a blank line after #define in sch_dualpi2.c (Stephen Hemminger <stephen(a)networkplumber.org>) - Fix line length warning. v3 (19-Oct-2024) - Fix compilaiton error - Update Documentation/netlink/specs/tc.yaml (Jakub Kicinski <kuba(a)kernel.org>) v2 (18-Oct-2024) - Add Documentation/netlink/specs/tc.yaml (Jakub Kicinski <kuba(a)kernel.org>) - Use dualpi2 instead of skb prefix (Jamal Hadi Salim <jhs(a)mojatatu.com>) - Replace nla_parse_nested_deprecated with nla_parse_nested (Jamal Hadi Salim <jhs(a)mojatatu.com>) - Fix line length warning --- Chia-Yu Chang (5): sched: Struct definition and parsing of dualpi2 qdisc sched: Dump configuration and statistics of dualpi2 qdisc selftests/tc-testing: Fix warning and style check on tdc.sh selftests/tc-testing: Add selftests for qdisc DualPI2 Documentation: netlink: specs: tc: Add DualPI2 specification Koen De Schepper (1): sched: Add enqueue/dequeue of dualpi2 qdisc Documentation/netlink/specs/tc.yaml | 151 ++- include/net/dropreason-core.h | 6 + include/uapi/linux/pkt_sched.h | 68 + net/sched/Kconfig | 12 + net/sched/Makefile | 1 + net/sched/sch_dualpi2.c | 1175 +++++++++++++++++ tools/testing/selftests/tc-testing/config | 1 + .../tc-testing/tc-tests/qdiscs/dualpi2.json | 254 ++++ tools/testing/selftests/tc-testing/tdc.sh | 6 +- 9 files changed, 1669 insertions(+), 5 deletions(-) create mode 100644 net/sched/sch_dualpi2.c create mode 100644 tools/testing/selftests/tc-testing/tc-tests/qdiscs/dualpi2.json -- 2.34.1

1 month

2
7
0 0

[PATCH net-next] devlink: Fix excessive stack usage in rate TC bandwidth parsing

by Tariq Toukan

From: Carolina Jubran <cjubran(a)nvidia.com> The devlink_nl_rate_tc_bw_parse function uses a large stack array for devlink attributes, which triggers a warning about excessive stack usage: net/devlink/rate.c: In function 'devlink_nl_rate_tc_bw_parse': net/devlink/rate.c:382:1: error: the frame size of 1648 bytes is larger than 1536 bytes [-Werror=frame-larger-than=] Introduce a separate attribute set specifically for rate TC bandwidth parsing that only contains the two attributes actually used: index and bandwidth. This reduces the stack array from DEVLINK_ATTR_MAX entries to just 2 entries, solving the stack usage issue. Update devlink selftest to use the new 'index' and 'bw' attribute names consistent with the YAML spec. Example usage with ynl with the new spec: ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/devlink.yaml \ --do rate-set --json '{ "bus-name": "pci", "dev-name": "0000:08:00.0", "port-index": 1, "rate-tc-bws": [ {"index": 0, "bw": 50}, {"index": 1, "bw": 50}, {"index": 2, "bw": 0}, {"index": 3, "bw": 0}, {"index": 4, "bw": 0}, {"index": 5, "bw": 0}, {"index": 6, "bw": 0}, {"index": 7, "bw": 0} ] }' ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/devlink.yaml \ --do rate-get --json '{ "bus-name": "pci", "dev-name": "0000:08:00.0", "port-index": 1 }' output for rate-get: {'bus-name': 'pci', 'dev-name': '0000:08:00.0', 'port-index': 1, 'rate-tc-bws': [{'bw': 50, 'index': 0}, {'bw': 50, 'index': 1}, {'bw': 0, 'index': 2}, {'bw': 0, 'index': 3}, {'bw': 0, 'index': 4}, {'bw': 0, 'index': 5}, {'bw': 0, 'index': 6}, {'bw': 0, 'index': 7}], 'rate-tx-max': 0, 'rate-tx-priority': 0, 'rate-tx-share': 0, 'rate-tx-weight': 0, 'rate-type': 'leaf'} Fixes: 566e8f108fc7 ("devlink: Extend devlink rate API with traffic classes bandwidth management") Reported-by: Arnd Bergmann <arnd(a)arndb.de> Closes: https://lore.kernel.org/netdev/20250708160652.1810573-1-arnd@kernel.org/ Reported-by: kernel test robot <lkp(a)intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202507171943.W7DJcs6Y-lkp@intel.com/ Suggested-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Carolina Jubran <cjubran(a)nvidia.com> Tested-by: Carolina Jubran <cjubran(a)nvidia.com> Signed-off-by: Tariq Toukan <tariqt(a)nvidia.com> --- Documentation/netlink/specs/devlink.yaml | 26 ++++++++----------- include/uapi/linux/devlink.h | 11 ++++++-- net/devlink/netlink_gen.c | 6 ++--- net/devlink/netlink_gen.h | 2 +- net/devlink/rate.c | 20 +++++++------- .../drivers/net/hw/devlink_rate_tc_bw.py | 16 ++++++------ 6 files changed, 42 insertions(+), 39 deletions(-) diff --git a/Documentation/netlink/specs/devlink.yaml b/Documentation/netlink/specs/devlink.yaml index 1c4bb0cbe5f0..bb87111d5e16 100644 --- a/Documentation/netlink/specs/devlink.yaml +++ b/Documentation/netlink/specs/devlink.yaml @@ -853,18 +853,6 @@ attribute-sets: type: nest multi-attr: true nested-attributes: dl-rate-tc-bws - - - name: rate-tc-index - type: u8 - checks: - max: rate-tc-index-max - - - name: rate-tc-bw - type: u32 - doc: | - Specifies the bandwidth share assigned to the Traffic Class. - The bandwidth for the traffic class is determined - in proportion to the sum of the shares of all configured classes. - name: dl-dev-stats subset-of: devlink @@ -1271,12 +1259,20 @@ attribute-sets: type: flag - name: dl-rate-tc-bws - subset-of: devlink + name-prefix: devlink-rate-tc-attr- attributes: - - name: rate-tc-index + name: index + type: u8 + checks: + max: rate-tc-index-max - - name: rate-tc-bw + name: bw + type: u32 + doc: | + Specifies the bandwidth share assigned to the Traffic Class. + The bandwidth for the traffic class is determined + in proportion to the sum of the shares of all configured classes. operations: enum-model: directional diff --git a/include/uapi/linux/devlink.h b/include/uapi/linux/devlink.h index e72bcc239afd..9fcb25a0f447 100644 --- a/include/uapi/linux/devlink.h +++ b/include/uapi/linux/devlink.h @@ -635,8 +635,6 @@ enum devlink_attr { DEVLINK_ATTR_REGION_DIRECT, /* flag */ DEVLINK_ATTR_RATE_TC_BWS, /* nested */ - DEVLINK_ATTR_RATE_TC_INDEX, /* u8 */ - DEVLINK_ATTR_RATE_TC_BW, /* u32 */ /* Add new attributes above here, update the spec in * Documentation/netlink/specs/devlink.yaml and re-generate @@ -647,6 +645,15 @@ enum devlink_attr { DEVLINK_ATTR_MAX = __DEVLINK_ATTR_MAX - 1 }; +enum devlink_rate_tc_attr { + DEVLINK_RATE_TC_ATTR_UNSPEC, + DEVLINK_RATE_TC_ATTR_INDEX, /* u8 */ + DEVLINK_RATE_TC_ATTR_BW, /* u32 */ + + __DEVLINK_RATE_TC_ATTR_MAX, + DEVLINK_RATE_TC_ATTR_MAX = __DEVLINK_RATE_TC_ATTR_MAX - 1 +}; + /* Mapping between internal resource described by the field and system * structure */ diff --git a/net/devlink/netlink_gen.c b/net/devlink/netlink_gen.c index c50436433c18..d97c326a9045 100644 --- a/net/devlink/netlink_gen.c +++ b/net/devlink/netlink_gen.c @@ -45,9 +45,9 @@ const struct nla_policy devlink_dl_port_function_nl_policy[DEVLINK_PORT_FN_ATTR_ [DEVLINK_PORT_FN_ATTR_CAPS] = NLA_POLICY_BITFIELD32(15), }; -const struct nla_policy devlink_dl_rate_tc_bws_nl_policy[DEVLINK_ATTR_RATE_TC_BW + 1] = { - [DEVLINK_ATTR_RATE_TC_INDEX] = NLA_POLICY_MAX(NLA_U8, DEVLINK_RATE_TC_INDEX_MAX), - [DEVLINK_ATTR_RATE_TC_BW] = { .type = NLA_U32, }, +const struct nla_policy devlink_dl_rate_tc_bws_nl_policy[DEVLINK_RATE_TC_ATTR_BW + 1] = { + [DEVLINK_RATE_TC_ATTR_INDEX] = NLA_POLICY_MAX(NLA_U8, DEVLINK_RATE_TC_INDEX_MAX), + [DEVLINK_RATE_TC_ATTR_BW] = { .type = NLA_U32, }, }; const struct nla_policy devlink_dl_selftest_id_nl_policy[DEVLINK_ATTR_SELFTEST_ID_FLASH + 1] = { diff --git a/net/devlink/netlink_gen.h b/net/devlink/netlink_gen.h index fb733b5d4ff1..09cc6f264ccf 100644 --- a/net/devlink/netlink_gen.h +++ b/net/devlink/netlink_gen.h @@ -13,7 +13,7 @@ /* Common nested types */ extern const struct nla_policy devlink_dl_port_function_nl_policy[DEVLINK_PORT_FN_ATTR_CAPS + 1]; -extern const struct nla_policy devlink_dl_rate_tc_bws_nl_policy[DEVLINK_ATTR_RATE_TC_BW + 1]; +extern const struct nla_policy devlink_dl_rate_tc_bws_nl_policy[DEVLINK_RATE_TC_ATTR_BW + 1]; extern const struct nla_policy devlink_dl_selftest_id_nl_policy[DEVLINK_ATTR_SELFTEST_ID_FLASH + 1]; /* Ops table for devlink */ diff --git a/net/devlink/rate.c b/net/devlink/rate.c index d39300a9b3d4..110b3fa8a0b1 100644 --- a/net/devlink/rate.c +++ b/net/devlink/rate.c @@ -90,8 +90,8 @@ static int devlink_rate_put_tc_bws(struct sk_buff *msg, u32 *tc_bw) if (!nla_tc_bw) return -EMSGSIZE; - if (nla_put_u8(msg, DEVLINK_ATTR_RATE_TC_INDEX, i) || - nla_put_u32(msg, DEVLINK_ATTR_RATE_TC_BW, tc_bw[i])) + if (nla_put_u8(msg, DEVLINK_RATE_TC_ATTR_INDEX, i) || + nla_put_u32(msg, DEVLINK_RATE_TC_ATTR_BW, tc_bw[i])) goto nla_put_failure; nla_nest_end(msg, nla_tc_bw); @@ -346,26 +346,26 @@ static int devlink_nl_rate_tc_bw_parse(struct nlattr *parent_nest, u32 *tc_bw, unsigned long *bitmap, struct netlink_ext_ack *extack) { - struct nlattr *tb[DEVLINK_ATTR_MAX + 1]; + struct nlattr *tb[DEVLINK_RATE_TC_ATTR_MAX + 1]; u8 tc_index; int err; - err = nla_parse_nested(tb, DEVLINK_ATTR_MAX, parent_nest, + err = nla_parse_nested(tb, DEVLINK_RATE_TC_ATTR_MAX, parent_nest, devlink_dl_rate_tc_bws_nl_policy, extack); if (err) return err; - if (!tb[DEVLINK_ATTR_RATE_TC_INDEX]) { + if (!tb[DEVLINK_RATE_TC_ATTR_INDEX]) { NL_SET_ERR_ATTR_MISS(extack, parent_nest, - DEVLINK_ATTR_RATE_TC_INDEX); + DEVLINK_RATE_TC_ATTR_INDEX); return -EINVAL; } - tc_index = nla_get_u8(tb[DEVLINK_ATTR_RATE_TC_INDEX]); + tc_index = nla_get_u8(tb[DEVLINK_RATE_TC_ATTR_INDEX]); - if (!tb[DEVLINK_ATTR_RATE_TC_BW]) { + if (!tb[DEVLINK_RATE_TC_ATTR_BW]) { NL_SET_ERR_ATTR_MISS(extack, parent_nest, - DEVLINK_ATTR_RATE_TC_BW); + DEVLINK_RATE_TC_ATTR_BW); return -EINVAL; } @@ -376,7 +376,7 @@ static int devlink_nl_rate_tc_bw_parse(struct nlattr *parent_nest, u32 *tc_bw, return -EINVAL; } - tc_bw[tc_index] = nla_get_u32(tb[DEVLINK_ATTR_RATE_TC_BW]); + tc_bw[tc_index] = nla_get_u32(tb[DEVLINK_RATE_TC_ATTR_BW]); return 0; } diff --git a/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py b/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py index 820d8a03becc..835c357919a8 100755 --- a/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py +++ b/tools/testing/selftests/drivers/net/hw/devlink_rate_tc_bw.py @@ -208,14 +208,14 @@ def setup_devlink_rate(cfg): "port-index": port_index, "rate-tx-max": 125000000, "rate-tc-bws": [ - {"rate-tc-index": 0, "rate-tc-bw": 0}, - {"rate-tc-index": 1, "rate-tc-bw": 0}, - {"rate-tc-index": 2, "rate-tc-bw": 0}, - {"rate-tc-index": 3, "rate-tc-bw": 20}, - {"rate-tc-index": 4, "rate-tc-bw": 80}, - {"rate-tc-index": 5, "rate-tc-bw": 0}, - {"rate-tc-index": 6, "rate-tc-bw": 0}, - {"rate-tc-index": 7, "rate-tc-bw": 0}, + {"index": 0, "bw": 0}, + {"index": 1, "bw": 0}, + {"index": 2, "bw": 0}, + {"index": 3, "bw": 20}, + {"index": 4, "bw": 80}, + {"index": 5, "bw": 0}, + {"index": 6, "bw": 0}, + {"index": 7, "bw": 0}, ] }) except NlError as exc: base-commit: 3fc894728fb3a0d9282e81247b68c07468fe2985 -- 2.31.1

1 month

4
3
0 0

[PATCH net v2] selftests: drv-net: wait for iperf client to stop sending

by Nimrod Oren

A few packets may still be sent out during the termination of iperf processes. These late packets cause failures in rss_ctx.py when they arrive on queues expected to be empty. Example failure observed: Check failed 2 != 0 traffic on inactive queues (context 1): [0, 0, 1, 1, 386385, 397196, 0, 0, 0, 0, ...] Check failed 4 != 0 traffic on inactive queues (context 2): [0, 0, 0, 0, 2, 2, 247152, 253013, 0, 0, ...] Check failed 2 != 0 traffic on inactive queues (context 3): [0, 0, 0, 0, 0, 0, 1, 1, 282434, 283070, ...] To avoid such failures, wait until all client sockets for the requested port are either closed or in the TIME_WAIT state. Fixes: 847aa551fa78 ("selftests: drv-net: rss_ctx: factor out send traffic and check") Signed-off-by: Nimrod Oren <noren(a)nvidia.com> Reviewed-by: Gal Pressman <gal(a)nvidia.com> Reviewed-by: Carolina Jubran <cjubran(a)nvidia.com> --- Changelog: v2: - Replace fixed sleep with logic that waits for client sockets to close. - Update commit title and message to reflect new approach. v1: https://lore.kernel.org/all/20250629111812.644282-1-noren@nvidia.com/ --- .../selftests/drivers/net/lib/py/load.py | 23 +++++++++++++++---- 1 file changed, 18 insertions(+), 5 deletions(-) diff --git a/tools/testing/selftests/drivers/net/lib/py/load.py b/tools/testing/selftests/drivers/net/lib/py/load.py index d9c10613ae67..44151b7b1a24 100644 --- a/tools/testing/selftests/drivers/net/lib/py/load.py +++ b/tools/testing/selftests/drivers/net/lib/py/load.py @@ -1,5 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 +import re import time from lib.py import ksft_pr, cmd, ip, rand_port, wait_port_listen @@ -10,12 +11,11 @@ class GenerateTraffic: self.env = env - if port is None: - port = rand_port() - self._iperf_server = cmd(f"iperf3 -s -1 -p {port}", background=True) - wait_port_listen(port) + self.port = rand_port() if port is None else port + self._iperf_server = cmd(f"iperf3 -s -1 -p {self.port}", background=True) + wait_port_listen(self.port) time.sleep(0.1) - self._iperf_client = cmd(f"iperf3 -c {env.addr} -P 16 -p {port} -t 86400", + self._iperf_client = cmd(f"iperf3 -c {env.addr} -P 16 -p {self.port} -t 86400", background=True, host=env.remote) # Wait for traffic to ramp up @@ -56,3 +56,16 @@ class GenerateTraffic: ksft_pr(">> Server:") ksft_pr(self._iperf_server.stdout) ksft_pr(self._iperf_server.stderr) + self._wait_client_stopped() + + def _wait_client_stopped(self, sleep=0.005, timeout=5): + end = time.monotonic() + timeout + + live_port_pattern = re.compile(fr":{self.port:04X} 0[^6] ") + + while time.monotonic() < end: + data = cmd("cat /proc/net/tcp*", host=self.env.remote).stdout + if not live_port_pattern.search(data): + return + time.sleep(sleep) + raise Exception(f"Waiting for client to stop timed out after {timeout}s") -- 2.40.1

1 month

3
2
0 0

[PATCH net-next v3 4/4] selftests: drv-net: add test for RSS on flow label

by Jakub Kicinski

Add a simple test for checking that RSS on flow label works, and that its rejected for IPv4 flows. # ./tools/testing/selftests/drivers/net/hw/rss_flow_label.py TAP version 13 1..2 ok 1 rss_flow_label.test_rss_flow_label ok 2 rss_flow_label.test_rss_flow_label_6only # Totals: pass:2 fail:0 xfail:0 xpass:0 skip:0 error:0 Reviewed-by: Willem de Bruijn <willemb(a)google.com> Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> --- v2: - check for RPS / RFS v1: https://lore.kernel.org/20250722014915.3365370-5-kuba@kernel.org CC: shuah(a)kernel.org CC: sdf(a)fomichev.me CC: linux-kselftest(a)vger.kernel.org --- .../testing/selftests/drivers/net/hw/Makefile | 1 + .../drivers/net/hw/rss_flow_label.py | 167 ++++++++++++++++++ 2 files changed, 168 insertions(+) create mode 100755 tools/testing/selftests/drivers/net/hw/rss_flow_label.py diff --git a/tools/testing/selftests/drivers/net/hw/Makefile b/tools/testing/selftests/drivers/net/hw/Makefile index fdc97355588c..5159fd34cb33 100644 --- a/tools/testing/selftests/drivers/net/hw/Makefile +++ b/tools/testing/selftests/drivers/net/hw/Makefile @@ -18,6 +18,7 @@ TEST_PROGS = \ pp_alloc_fail.py \ rss_api.py \ rss_ctx.py \ + rss_flow_label.py \ rss_input_xfrm.py \ tso.py \ xsk_reconfig.py \ diff --git a/tools/testing/selftests/drivers/net/hw/rss_flow_label.py b/tools/testing/selftests/drivers/net/hw/rss_flow_label.py new file mode 100755 index 000000000000..6fa95fe27c47 --- /dev/null +++ b/tools/testing/selftests/drivers/net/hw/rss_flow_label.py @@ -0,0 +1,167 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 + +""" +Tests for RSS hashing on IPv6 Flow Label. +""" + +import glob +import os +import socket +from lib.py import CmdExitFailure +from lib.py import ksft_run, ksft_exit, ksft_eq, ksft_ge, ksft_in, \ + ksft_not_in, ksft_raises, KsftSkipEx +from lib.py import bkg, cmd, defer, fd_read_timeout, rand_port +from lib.py import NetDrvEpEnv + + +def _check_system(cfg): + if not hasattr(socket, "SO_INCOMING_CPU"): + raise KsftSkipEx("socket.SO_INCOMING_CPU was added in Python 3.11") + + qcnt = len(glob.glob(f"/sys/class/net/{cfg.ifname}/queues/rx-*")) + if qcnt < 2: + raise KsftSkipEx(f"Local has only {qcnt} queues") + + for f in [f"/sys/class/net/{cfg.ifname}/queues/rx-0/rps_flow_cnt", + f"/sys/class/net/{cfg.ifname}/queues/rx-0/rps_cpus"]: + try: + with open(f, 'r') as fp: + setting = fp.read().strip() + # CPU mask will be zeros and commas + if setting.replace("0", "").replace(",", ""): + raise KsftSkipEx(f"RPS/RFS is configured: {f}: {setting}") + except FileNotFoundError: + pass + + # 1 is the default, if someone changed it we probably shouldn"t mess with it + af = cmd("cat /proc/sys/net/ipv6/auto_flowlabels", host=cfg.remote).stdout + if af.strip() != "1": + raise KsftSkipEx("Remote does not have auto_flowlabels enabled") + + +def _ethtool_get_cfg(cfg, fl_type): + descr = cmd(f"ethtool -n {cfg.ifname} rx-flow-hash {fl_type}").stdout + + converter = { + "IP SA": "s", + "IP DA": "d", + "L3 proto": "t", + "L4 bytes 0 & 1 [TCP/UDP src port]": "f", + "L4 bytes 2 & 3 [TCP/UDP dst port]": "n", + "IPv6 Flow Label": "l", + } + + ret = "" + for line in descr.split("\n")[1:-2]: + # if this raises we probably need to add more keys to converter above + ret += converter[line] + return ret + + +def _traffic(cfg, one_sock, one_cpu): + local_port = rand_port(socket.SOCK_DGRAM) + remote_port = rand_port(socket.SOCK_DGRAM) + + sock = socket.socket(socket.AF_INET6, socket.SOCK_DGRAM) + sock.bind(("", local_port)) + sock.connect((cfg.remote_addr_v["6"], 0)) + if one_sock: + send = f"exec 5<>/dev/udp/{cfg.addr_v['6']}/{local_port}; " \ + "for i in `seq 20`; do echo a >&5; sleep 0.02; done; exec 5>&-" + else: + send = "for i in `seq 20`; do echo a | socat -t0.02 - UDP6:" \ + f"[{cfg.addr_v['6']}]:{local_port},sourceport={remote_port}; done" + + cpus = set() + with bkg(send, shell=True, host=cfg.remote, exit_wait=True): + for _ in range(20): + fd_read_timeout(sock.fileno(), 1) + cpu = sock.getsockopt(socket.SOL_SOCKET, socket.SO_INCOMING_CPU) + cpus.add(cpu) + + if one_cpu: + ksft_eq(len(cpus), 1, + f"{one_sock=} - expected one CPU, got traffic on: {cpus=}") + else: + ksft_ge(len(cpus), 2, + f"{one_sock=} - expected many CPUs, got traffic on: {cpus=}") + + +def test_rss_flow_label(cfg): + """ + Test hashing on IPv6 flow label. Send traffic over a single socket + and over multiple sockets. Depend on the remote having auto-label + enabled so that it randomizes the label per socket. + """ + + cfg.require_ipver("6") + cfg.require_cmd("socat", remote=True) + _check_system(cfg) + + # Enable flow label hashing for UDP6 + initial = _ethtool_get_cfg(cfg, "udp6") + no_lbl = initial.replace("l", "") + if "l" not in initial: + try: + cmd(f"ethtool -N {cfg.ifname} rx-flow-hash udp6 l{no_lbl}") + except CmdExitFailure as exc: + raise KsftSkipEx("Device doesn't support Flow Label for UDP6") from exc + + defer(cmd, f"ethtool -N {cfg.ifname} rx-flow-hash udp6 {initial}") + + _traffic(cfg, one_sock=True, one_cpu=True) + _traffic(cfg, one_sock=False, one_cpu=False) + + # Disable it, we should see no hashing (reset was already defer()ed) + cmd(f"ethtool -N {cfg.ifname} rx-flow-hash udp6 {no_lbl}") + + _traffic(cfg, one_sock=False, one_cpu=True) + + +def _check_v4_flow_types(cfg): + for fl_type in ["tcp4", "udp4", "ah4", "esp4", "sctp4"]: + try: + cur = cmd(f"ethtool -n {cfg.ifname} rx-flow-hash {fl_type}").stdout + ksft_not_in("Flow Label", cur, + comment=f"{fl_type=} has Flow Label:" + cur) + except CmdExitFailure: + # Probably does not support this flow type + pass + + +def test_rss_flow_label_6only(cfg): + """ + Test interactions with IPv4 flow types. It should not be possible to set + IPv6 Flow Label hashing for an IPv4 flow type. The Flow Label should also + not appear in the IPv4 "current config". + """ + + with ksft_raises(CmdExitFailure) as cm: + cmd(f"ethtool -N {cfg.ifname} rx-flow-hash tcp4 sdfnl") + ksft_in("Invalid argument", cm.exception.cmd.stderr) + + _check_v4_flow_types(cfg) + + # Try to enable Flow Labels and check again, in case it leaks thru + initial = _ethtool_get_cfg(cfg, "udp6") + changed = initial.replace("l", "") if "l" in initial else initial + "l" + + cmd(f"ethtool -N {cfg.ifname} rx-flow-hash udp6 {changed}") + restore = defer(cmd, f"ethtool -N {cfg.ifname} rx-flow-hash udp6 {initial}") + + _check_v4_flow_types(cfg) + restore.exec() + _check_v4_flow_types(cfg) + + +def main() -> None: + with NetDrvEpEnv(__file__, nsim_test=False) as cfg: + ksft_run([test_rss_flow_label, + test_rss_flow_label_6only], + args=(cfg, )) + ksft_exit() + + +if __name__ == "__main__": + main() -- 2.50.1

1 month

1
0
0 0

[PATCH RFC 0/4] procfs: make reference pidns more user-visible

by Aleksa Sarai

Ever since the introduction of pid namespaces, procfs has had very implicit behaviour surrounding them (the pidns used by a procfs mount is auto-selected based on the mounting process's active pidns, and the pidns itself is basically hidden once the mount has been constructed). This has historically meant that userspace was required to do some special dances in order to configure the pidns of a procfs mount as desired. Examples include: * In order to bypass the mnt_too_revealing() check, Kubernetes creates a procfs mount from an empty pidns so that user namespaced containers can be nested (without this, the nested containers would fail to mount procfs). But this requires forking off a helper process because you cannot just one-shot this using mount(2). * Container runtimes in general need to fork into a container before configuring its mounts, which can lead to security issues in the case of shared-pidns containers (a privileged process in the pidns can interact with your container runtime process). While SUID_DUMP_DISABLE and user namespaces make this less of an issue, the strict need for this due to a minor uAPI wart is kind of unfortunate. Things would be much easier if there was a way for userspace to just specify the pidns they want. Patch 1 implements a new "pidns" argument which can be set using fsconfig(2): fsconfig(procfd, FSCONFIG_SET_FD, "pidns", NULL, nsfd); fsconfig(procfd, FSCONFIG_SET_STRING, "pidns", "/proc/self/ns/pid", 0); or classic mount(2) / mount(8): // mount -t proc -o pidns=/proc/self/ns/pid proc /tmp/proc mount("proc", "/tmp/proc", "proc", MS_..., "pidns=/proc/self/ns/pid"); The initial security model I have in this RFC is to be as conservative as possible and just mirror the security model for setns(2) -- which means that you can only set pidns=... to pid namespaces that your current pid namespace is a direct ancestor of. This fulfils the requirements of container runtimes, but I suspect that this may be too strict for some usecases. The pidns argument is not displayed in mountinfo -- it's not clear to me what value it would make sense to show (maybe we could just use ns_dname to provide an identifier for the namespace, but this number would be fairly useless to userspace). I'm open to suggestions. In addition, being able to figure out what pid namespace is being used by a procfs mount is quite useful when you have an administrative process (such as a container runtime) which wants to figure out the correct way of mapping PIDs between its own namespace and the namespace for procfs (using NS_GET_{PID,TGID}_{IN,FROM}_PIDNS). There are alternative ways to do this, but they all rely on ancillary information that third-party libraries and tools do not necessarily have access to. To make this easier, add a new ioctl (PROCFS_GET_PID_NAMESPACE) which can be used to get a reference to the pidns that a procfs is using. It's not quite clear what is the correct security model for this API, but the current approach I've taken is to: * Make the ioctl only valid on the root (meaning that a process without access to the procfs root -- such as only having an fd to a procfs file or some open_tree(2)-like subset -- cannot use this API). * Require that the process requesting either has access to /proc/1/ns/pid anyway (i.e. has ptrace-read access to the pidns pid1), has CAP_SYS_ADMIN access to the pidns (i.e. has administrative access to it and can join it if they had a handle), or is in a pidns that is a direct ancestor of the target pidns (i.e. all of the pids are already visible in the procfs for the current process's pidns). The security model for this is a little loose, as it seems to me that all of the cases mentioned are valid cases to allow access, but I'm open to suggestions for whether we need to make this stricter or looser. Signed-off-by: Aleksa Sarai <cyphar(a)cyphar.com> --- Aleksa Sarai (4): pidns: move is-ancestor logic to helper procfs: add pidns= mount option procfs: add PROCFS_GET_PID_NAMESPACE ioctl selftests/proc: add tests for new pidns APIs Documentation/filesystems/proc.rst | 10 ++ fs/proc/root.c | 132 +++++++++++++- include/linux/pid_namespace.h | 9 + include/uapi/linux/fs.h | 3 + kernel/pid_namespace.c | 21 ++- tools/testing/selftests/proc/.gitignore | 1 + tools/testing/selftests/proc/Makefile | 1 + tools/testing/selftests/proc/proc-pidns.c | 286 ++++++++++++++++++++++++++++++ 8 files changed, 448 insertions(+), 15 deletions(-) --- base-commit: 4c838c7672c39ec6ec48456c6ce22d14a68f4cda change-id: 20250717-procfs-pidns-api-8ed1583431f0 Best regards, -- Aleksa Sarai <cyphar(a)cyphar.com>

1 month

2
7
0 0

[PATCH v2 0/5] KVM: Improve VMware guest support

by Zack Rusin

This is the second version of a series that lets us run VMware Workstation on Linux on top of KVM. The most significant change in this series is the introduction of CONFIG_KVM_VMWARE which is, in general, a nice cleanup for various bits of VMware compatibility code that have been scattered around KVM. (first patch) The rest of the series builds upon the VMware platform to implement features that are needed to run VMware guests without any modifications on top of KVM: - ability to turn on the VMware backdoor at runtime on a per-vm basis (used to be a kernel boot argument only) - support for VMware hypercalls - VMware products have a huge collection of hypercalls, all of which are handled in userspace, - support for handling legacy VMware backdoor in L0 in nested configs - in cases where we have WS running a Windows VBS guest, the L0 would be KVM, L1 Hyper-V so by default VMware Tools backdoor calls endup in Hyper-V which can not handle them, so introduce a cap to let L0 handle those. The final change in the series is a kselftest of the VMware hypercall functionality. Cc: Paolo Bonzini <pbonzini(a)redhat.com> Cc: Jonathan Corbet <corbet(a)lwn.net> Cc: Sean Christopherson <seanjc(a)google.com> Cc: Thomas Gleixner <tglx(a)linutronix.de> Cc: Ingo Molnar <mingo(a)redhat.com> Cc: Borislav Petkov <bp(a)alien8.de> Cc: Dave Hansen <dave.hansen(a)linux.intel.com> Cc: x86(a)kernel.org Cc: "H. Peter Anvin" <hpa(a)zytor.com> Cc: Zack Rusin <zack.rusin(a)broadcom.com> Cc: Doug Covelli <doug.covelli(a)broadcom.com> Cc: Shuah Khan <shuah(a)kernel.org> Cc: Namhyung Kim <namhyung(a)kernel.org> Cc: Arnaldo Carvalho de Melo <acme(a)redhat.com> Cc: Michael Ellerman <mpe(a)ellerman.id.au> Cc: Joel Stanley <joel(a)jms.id.au> Cc: Isaku Yamahata <isaku.yamahata(a)intel.com> Cc: kvm(a)vger.kernel.org Cc: linux-doc(a)vger.kernel.org Cc: linux-kernel(a)vger.kernel.org Cc: linux-kselftest(a)vger.kernel.org Zack Rusin (5): KVM: x86: Centralize KVM's VMware code KVM: x86: Allow enabling of the vmware backdoor via a cap KVM: x86: Add support for VMware guest specific hypercalls KVM: x86: Add support for legacy VMware backdoors in nested setups KVM: selftests: x86: Add a test for KVM_CAP_X86_VMWARE_HYPERCALL Documentation/virt/kvm/api.rst | 86 +++++++- MAINTAINERS | 9 + arch/x86/include/asm/kvm_host.h | 13 ++ arch/x86/kvm/Kconfig | 16 ++ arch/x86/kvm/Makefile | 1 + arch/x86/kvm/emulate.c | 11 +- arch/x86/kvm/kvm_vmware.c | 85 ++++++++ arch/x86/kvm/kvm_vmware.h | 189 ++++++++++++++++++ arch/x86/kvm/pmu.c | 39 +--- arch/x86/kvm/pmu.h | 4 - arch/x86/kvm/svm/nested.c | 6 + arch/x86/kvm/svm/svm.c | 10 +- arch/x86/kvm/vmx/nested.c | 6 + arch/x86/kvm/vmx/vmx.c | 5 +- arch/x86/kvm/x86.c | 74 +++---- arch/x86/kvm/x86.h | 2 - include/uapi/linux/kvm.h | 27 +++ tools/include/uapi/linux/kvm.h | 3 + tools/testing/selftests/kvm/Makefile.kvm | 1 + .../selftests/kvm/x86/vmware_hypercall_test.c | 121 +++++++++++ 20 files changed, 614 insertions(+), 94 deletions(-) create mode 100644 arch/x86/kvm/kvm_vmware.c create mode 100644 arch/x86/kvm/kvm_vmware.h create mode 100644 tools/testing/selftests/kvm/x86/vmware_hypercall_test.c -- 2.48.1

1 month

2
2
0 0

[PATCH v2] rtc: Rename lib_test to test_rtc_lib

by Geert Uytterhoeven

When compiling the RTC library functions test as a module, the module has the non-descriptive name "lib_test.ko". Fix this by renaming it to "test_rtc_lib.ko". Signed-off-by: Geert Uytterhoeven <geert(a)linux-m68k.org> --- v2: - s/rtc_lib_test/test_rtc_lib/. --- drivers/rtc/Makefile | 2 +- drivers/rtc/{lib_test.c => test_rtc_lib.c} | 0 2 files changed, 1 insertion(+), 1 deletion(-) rename drivers/rtc/{lib_test.c => test_rtc_lib.c} (100%) diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile index 4619aa2ac4697591..789bddfea99d8fcd 100644 --- a/drivers/rtc/Makefile +++ b/drivers/rtc/Makefile @@ -15,7 +15,7 @@ rtc-core-$(CONFIG_RTC_INTF_DEV) += dev.o rtc-core-$(CONFIG_RTC_INTF_PROC) += proc.o rtc-core-$(CONFIG_RTC_INTF_SYSFS) += sysfs.o -obj-$(CONFIG_RTC_LIB_KUNIT_TEST) += lib_test.o +obj-$(CONFIG_RTC_LIB_KUNIT_TEST) += test_rtc_lib.o # Keep the list ordered. diff --git a/drivers/rtc/lib_test.c b/drivers/rtc/test_rtc_lib.c similarity index 100% rename from drivers/rtc/lib_test.c rename to drivers/rtc/test_rtc_lib.c -- 2.43.0

1 month

2
1
0 0

[PATCH][next] tools/testing/selftests: Fix spelling mistake "unnmap" -> "unmap"

by Colin Ian King

There is a spelling mistake in ksft_test_result_fail messages. Fix them. Signed-off-by: Colin Ian King <colin.i.king(a)gmail.com> --- tools/testing/selftests/mm/mremap_test.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tools/testing/selftests/mm/mremap_test.c b/tools/testing/selftests/mm/mremap_test.c index fccf9e797a0c..774cdba102fc 100644 --- a/tools/testing/selftests/mm/mremap_test.c +++ b/tools/testing/selftests/mm/mremap_test.c @@ -525,10 +525,10 @@ static void mremap_move_multiple_vmas(unsigned int pattern_seed, out: if (success) ksft_test_result_pass("%s%s\n", test_name, - dont_unmap ? " [dontunnmap]" : ""); + dont_unmap ? " [dontunmap]" : ""); else ksft_test_result_fail("%s%s\n", test_name, - dont_unmap ? " [dontunnmap]" : ""); + dont_unmap ? " [dontunmap]" : ""); } static void mremap_shrink_multiple_vmas(unsigned long page_size, @@ -727,10 +727,10 @@ static void mremap_move_multiple_vmas_split(unsigned int pattern_seed, out: if (success) ksft_test_result_pass("%s%s\n", test_name, - dont_unmap ? " [dontunnmap]" : ""); + dont_unmap ? " [dontunmap]" : ""); else ksft_test_result_fail("%s%s\n", test_name, - dont_unmap ? " [dontunnmap]" : ""); + dont_unmap ? " [dontunmap]" : ""); } /* Returns the time taken for the remap on success else returns -1. */ -- 2.50.0

1 month

5
4
0 0

[PATCH/RFC] kunit/rtc: Add real support for very slow tests

by Geert Uytterhoeven

When running rtc_lib_test ("lib_test" before my "[PATCH] rtc: Rename lib_test to rtc_lib_test") on m68k/ARAnyM: KTAP version 1 1..1 KTAP version 1 # Subtest: rtc_lib_test_cases # module: rtc_lib_test 1..2 # rtc_time64_to_tm_test_date_range_1000: Test should be marked slow (runtime: 3.222371420s) ok 1 rtc_time64_to_tm_test_date_range_1000 # rtc_time64_to_tm_test_date_range_160000: try timed out # rtc_time64_to_tm_test_date_range_160000: test case timed out # rtc_time64_to_tm_test_date_range_160000.speed: slow not ok 2 rtc_time64_to_tm_test_date_range_160000 # rtc_lib_test_cases: pass:1 fail:1 skip:0 total:2 # Totals: pass:1 fail:1 skip:0 total:2 not ok 1 rtc_lib_test_cases Commit 02c2d0c2a84172c3 ("kunit: Add speed attribute") added the notion of "very slow" tests, but this is further unused and unhandled. Hence: 1. Introduce KUNIT_CASE_VERY_SLOW(), 2. Increase timeout by ten; ideally this should only be done for very slow tests, but I couldn't find how to access kunit_case.attr.case from kunit_try_catch_run(), 3. Mark rtc_time64_to_tm_test_date_range_1000 slow, 4. Mark rtc_time64_to_tm_test_date_range_160000 very slow. Afterwards: KTAP version 1 1..1 KTAP version 1 # Subtest: rtc_lib_test_cases # module: rtc_lib_test 1..2 # rtc_time64_to_tm_test_date_range_1000.speed: slow ok 1 rtc_time64_to_tm_test_date_range_1000 # rtc_time64_to_tm_test_date_range_160000.speed: very_slow ok 2 rtc_time64_to_tm_test_date_range_160000 # rtc_lib_test_cases: pass:2 fail:0 skip:0 total:2 # Totals: pass:2 fail:0 skip:0 total:2 ok 1 rtc_lib_test_cases Signed-off-by: Geert Uytterhoeven <geert(a)linux-m68k.org> --- drivers/rtc/rtc_lib_test.c | 4 ++-- include/kunit/test.h | 11 +++++++++++ lib/kunit/try-catch.c | 3 ++- 3 files changed, 15 insertions(+), 3 deletions(-) diff --git a/drivers/rtc/rtc_lib_test.c b/drivers/rtc/rtc_lib_test.c index c30c759662e39b48..fd3210e39d37dbc6 100644 --- a/drivers/rtc/rtc_lib_test.c +++ b/drivers/rtc/rtc_lib_test.c @@ -85,8 +85,8 @@ static void rtc_time64_to_tm_test_date_range_1000(struct kunit *test) } static struct kunit_case rtc_lib_test_cases[] = { - KUNIT_CASE(rtc_time64_to_tm_test_date_range_1000), - KUNIT_CASE_SLOW(rtc_time64_to_tm_test_date_range_160000), + KUNIT_CASE_SLOW(rtc_time64_to_tm_test_date_range_1000), + KUNIT_CASE_VERY_SLOW(rtc_time64_to_tm_test_date_range_160000), {} }; diff --git a/include/kunit/test.h b/include/kunit/test.h index 9b773406e01f3c43..4e3c1cae5b41466e 100644 --- a/include/kunit/test.h +++ b/include/kunit/test.h @@ -183,6 +183,17 @@ static inline char *kunit_status_to_ok_not_ok(enum kunit_status status) { .run_case = test_name, .name = #test_name, \ .attr.speed = KUNIT_SPEED_SLOW, .module_name = KBUILD_MODNAME} +/** + * KUNIT_CASE_VERY_SLOW - A helper for creating a &struct kunit_case + * with the very slow attribute + * + * @test_name: a reference to a test case function. + */ + +#define KUNIT_CASE_VERY_SLOW(test_name) \ + { .run_case = test_name, .name = #test_name, \ + .attr.speed = KUNIT_SPEED_VERY_SLOW, .module_name = KBUILD_MODNAME} + /** * KUNIT_CASE_PARAM - A helper for creation a parameterized &struct kunit_case * diff --git a/lib/kunit/try-catch.c b/lib/kunit/try-catch.c index 6bbe0025b0790bd2..92099c67bb21d0a4 100644 --- a/lib/kunit/try-catch.c +++ b/lib/kunit/try-catch.c @@ -56,7 +56,8 @@ static unsigned long kunit_test_timeout(void) * If tests timeout due to exceeding sysctl_hung_task_timeout_secs, * the task will be killed and an oops generated. */ - return 300 * msecs_to_jiffies(MSEC_PER_SEC); /* 5 min */ + // FIXME times ten for KUNIT_SPEED_VERY_SLOW? + return 10 * 300 * msecs_to_jiffies(MSEC_PER_SEC); /* 5 min */ } void kunit_try_catch_run(struct kunit_try_catch *try_catch, void *context) -- 2.43.0

1 month

3
3
0 0

[PATCH net-next v2 4/4] selftests: drv-net: add test for RSS on flow label

by Jakub Kicinski

Add a simple test for checking that RSS on flow label works, and that its rejected for IPv4 flows. # ./tools/testing/selftests/drivers/net/hw/rss_flow_label.py TAP version 13 1..2 ok 1 rss_flow_label.test_rss_flow_label ok 2 rss_flow_label.test_rss_flow_label_6only # Totals: pass:2 fail:0 xfail:0 xpass:0 skip:0 error:0 Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> --- CC: shuah(a)kernel.org CC: sdf(a)fomichev.me CC: linux-kselftest(a)vger.kernel.org --- .../testing/selftests/drivers/net/hw/Makefile | 1 + .../drivers/net/hw/rss_flow_label.py | 151 ++++++++++++++++++ 2 files changed, 152 insertions(+) create mode 100755 tools/testing/selftests/drivers/net/hw/rss_flow_label.py diff --git a/tools/testing/selftests/drivers/net/hw/Makefile b/tools/testing/selftests/drivers/net/hw/Makefile index fdc97355588c..5159fd34cb33 100644 --- a/tools/testing/selftests/drivers/net/hw/Makefile +++ b/tools/testing/selftests/drivers/net/hw/Makefile @@ -18,6 +18,7 @@ TEST_PROGS = \ pp_alloc_fail.py \ rss_api.py \ rss_ctx.py \ + rss_flow_label.py \ rss_input_xfrm.py \ tso.py \ xsk_reconfig.py \ diff --git a/tools/testing/selftests/drivers/net/hw/rss_flow_label.py b/tools/testing/selftests/drivers/net/hw/rss_flow_label.py new file mode 100755 index 000000000000..e471e13160ae --- /dev/null +++ b/tools/testing/selftests/drivers/net/hw/rss_flow_label.py @@ -0,0 +1,151 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 + +""" +Tests for RSS hashing on IPv6 Flow Label. +""" + +import glob +import socket +from lib.py import CmdExitFailure +from lib.py import ksft_run, ksft_exit, ksft_eq, ksft_ge, ksft_in, \ + ksft_not_in, ksft_raises, KsftSkipEx +from lib.py import bkg, cmd, defer, fd_read_timeout, rand_port +from lib.py import NetDrvEpEnv + + +def _ethtool_get_cfg(cfg, fl_type): + descr = cmd(f"ethtool -n {cfg.ifname} rx-flow-hash {fl_type}").stdout + + converter = { + "IP SA": "s", + "IP DA": "d", + "L3 proto": "t", + "L4 bytes 0 & 1 [TCP/UDP src port]": "f", + "L4 bytes 2 & 3 [TCP/UDP dst port]": "n", + "IPv6 Flow Label": "l", + } + + ret = "" + for line in descr.split("\n")[1:-2]: + # if this raises we probably need to add more keys to converter above + ret += converter[line] + return ret + + +def _traffic(cfg, one_sock, one_cpu): + local_port = rand_port(socket.SOCK_DGRAM) + remote_port = rand_port(socket.SOCK_DGRAM) + + sock = socket.socket(socket.AF_INET6, socket.SOCK_DGRAM) + sock.bind(("", local_port)) + sock.connect((cfg.remote_addr_v["6"], 0)) + if one_sock: + send = f"exec 5<>/dev/udp/{cfg.addr_v['6']}/{local_port}; " \ + "for i in `seq 20`; do echo a >&5; sleep 0.02; done; exec 5>&-" + else: + send = "for i in `seq 20`; do echo a | socat -t0.02 - UDP6:" \ + f"[{cfg.addr_v['6']}]:{local_port},sourceport={remote_port}; done" + + cpus = set() + with bkg(send, shell=True, host=cfg.remote, exit_wait=True): + for _ in range(20): + fd_read_timeout(sock.fileno(), 1) + cpu = sock.getsockopt(socket.SOL_SOCKET, socket.SO_INCOMING_CPU) + cpus.add(cpu) + + if one_cpu: + ksft_eq(len(cpus), 1, + f"{one_sock=} - expected one CPU, got traffic on: {cpus=}") + else: + ksft_ge(len(cpus), 2, + f"{one_sock=} - expected many CPUs, got traffic on: {cpus=}") + + +def test_rss_flow_label(cfg): + """ + Test hashing on IPv6 flow label. Send traffic over a single socket + and over multiple sockets. Depend on the remote having auto-label + enabled so that it randomizes the label per socket. + """ + + cfg.require_ipver("6") + cfg.require_cmd("socat", remote=True) + if not hasattr(socket, "SO_INCOMING_CPU"): + raise KsftSkipEx("socket.SO_INCOMING_CPU was added in Python 3.11") + + # 1 is the default, if someone changed it we probably shouldn"t mess with it + af = cmd("cat /proc/sys/net/ipv6/auto_flowlabels", host=cfg.remote).stdout + if af.strip() != "1": + raise KsftSkipEx("Remote does not have auto_flowlabels enabled") + + qcnt = len(glob.glob(f"/sys/class/net/{cfg.ifname}/queues/rx-*")) + if qcnt < 2: + raise KsftSkipEx(f"Local has only {qcnt} queues") + + # Enable flow label hashing for UDP6 + initial = _ethtool_get_cfg(cfg, "udp6") + no_lbl = initial.replace("l", "") + if "l" not in initial: + try: + cmd(f"ethtool -N {cfg.ifname} rx-flow-hash udp6 l{no_lbl}") + except CmdExitFailure as exc: + raise KsftSkipEx("Device doesn't support Flow Label for UDP6") from exc + + defer(cmd, f"ethtool -N {cfg.ifname} rx-flow-hash udp6 {initial}") + + _traffic(cfg, one_sock=True, one_cpu=True) + _traffic(cfg, one_sock=False, one_cpu=False) + + # Disable it, we should see no hashing (reset was already defer()ed) + cmd(f"ethtool -N {cfg.ifname} rx-flow-hash udp6 {no_lbl}") + + _traffic(cfg, one_sock=False, one_cpu=True) + + +def _check_v4_flow_types(cfg): + for fl_type in ["tcp4", "udp4", "ah4", "esp4", "sctp4"]: + try: + cur = cmd(f"ethtool -n {cfg.ifname} rx-flow-hash {fl_type}").stdout + ksft_not_in("Flow Label", cur, + comment=f"{fl_type=} has Flow Label:" + cur) + except CmdExitFailure: + # Probably does not support this flow type + pass + + +def test_rss_flow_label_6only(cfg): + """ + Test interactions with IPv4 flow types. It should not be possible to set + IPv6 Flow Label hashing for an IPv4 flow type. The Flow Label should also + not appear in the IPv4 "current config". + """ + + with ksft_raises(CmdExitFailure) as cm: + cmd(f"ethtool -N {cfg.ifname} rx-flow-hash tcp4 sdfnl") + ksft_in("Invalid argument", cm.exception.cmd.stderr) + + _check_v4_flow_types(cfg) + + # Try to enable Flow Labels and check again, in case it leaks thru + initial = _ethtool_get_cfg(cfg, "udp6") + changed = initial.replace("l", "") if "l" in initial else initial + "l" + + cmd(f"ethtool -N {cfg.ifname} rx-flow-hash udp6 {changed}") + restore = defer(cmd, f"ethtool -N {cfg.ifname} rx-flow-hash udp6 {initial}") + + _check_v4_flow_types(cfg) + restore.exec() + _check_v4_flow_types(cfg) + + +def main() -> None: + with NetDrvEpEnv(__file__, nsim_test=False) as cfg: + ksft_run([test_rss_flow_label, + test_rss_flow_label_6only], + args=(cfg, )) + ksft_exit() + + +if __name__ == "__main__": + main() -- 2.50.1

1 month

2
1
0 0

[PATCH v6] selftests/mm: add process_madvise() tests

by wang lian

Add tests for 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> Suggested-by: Zi Yan <ziy(a)nvidia.com> Suggested-by: Mark Brown <broonie(a)kernel.org> Acked-by: SeongJae Park <sj(a)kernel.org> --- Changelog v6: - Refactor child process and pidfd management to use the kselftest fixture's setup and teardown mechanism. This ensures that child processes are reliably terminated and file descriptors are closed, even when a test is aborted by an ASSERT or SKIP macro. This resolves the issue where a failed assertion could lead to a leaked child process. Changelog v5: https://lore.kernel.org/lkml/20250714122533.3135-1-lianux.mm@gmail.com/ - Refactor the remote_collapse test to concentrate on its primary goal confirming the successful remote invocation of process_madvise() on a child process. - Split the validation logic for invalid pidfds out of the remote test and into two new (`exited_process_pidfd` and `bad_pidfd`). - Based mm-new branch, can ensure clean application Changelog v4: https://lore.kernel.org/lkml/20250710112249.58722-1-lianux.mm@gmail.com/ - Refine resource cleanup logic in test teardown to be more robust. - Improve remote_collapse test to correctly handle different THP (Transparent Huge Page) policies ('always', 'madvise', 'never'), including handling race conditions with khugepaged. - Resolve build errors Changelog v3: https://lore.kernel.org/lkml/20250703044326.65061-1-lianux.mm@gmail.com/ - Rebased onto the latest mm-stable branch to ensure clean application. - Refactor common signal handling logic into vm_util to reduce code duplication. - Improve test robustness and diagnostics based on community feedback. - Address minor code style and script corrections. Changelog v2: https://lore.kernel.org/lkml/20250630140957.4000-1-lianux.mm@gmail.com/ - 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. -V1: https://lore.kernel.org/lkml/20250621133003.4733-1-lianux.mm@gmail.com/ tools/testing/selftests/mm/.gitignore | 1 + tools/testing/selftests/mm/Makefile | 1 + tools/testing/selftests/mm/process_madv.c | 302 ++++++++++++++++++++++ tools/testing/selftests/mm/run_vmtests.sh | 5 + 4 files changed, 309 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 f2dafa0b700b..e7b23a8a05fe --- a/tools/testing/selftests/mm/.gitignore +++ b/tools/testing/selftests/mm/.gitignore @@ -21,6 +21,7 @@ on-fault-limit transhuge-stress pagemap_ioctl pfnmap +process_madv *.tmp* protection_keys protection_keys_32 diff --git a/tools/testing/selftests/mm/Makefile b/tools/testing/selftests/mm/Makefile index ae6f994d3add..d13b3cef2a2b 100644 --- a/tools/testing/selftests/mm/Makefile +++ b/tools/testing/selftests/mm/Makefile @@ -85,6 +85,7 @@ 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..8a83eac3bfab --- /dev/null +++ b/tools/testing/selftests/mm/process_madv.c @@ -0,0 +1,302 @@ +// 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 <linux/mman.h> +#include <sys/syscall.h> +#include <unistd.h> +#include <sched.h> +#include "vm_util.h" + +#include "../pidfd/pidfd.h" + +FIXTURE(process_madvise) +{ + unsigned long page_size; + pid_t child_pid; + int remote_pidfd; + int pidfd; +}; + +FIXTURE_SETUP(process_madvise) +{ + self->page_size = (unsigned long)sysconf(_SC_PAGESIZE); + self->pidfd = PIDFD_SELF; + self->remote_pidfd = -1; + self->child_pid = -1; +}; + +FIXTURE_TEARDOWN_PARENT(process_madvise) +{ + /* This teardown is guaranteed to run, even if tests SKIP or ASSERT */ + if (self->child_pid > 0) { + kill(self->child_pid, SIGKILL); + waitpid(self->child_pid, NULL, 0); + } + + if (self->remote_pidfd >= 0) + close(self->remote_pidfd); +} + +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); +} + +/* + * This test uses PIDFD_SELF to target the current process. The main + * goal is to verify the basic behavior of process_madvise() with + * a vector of non-contiguous memory ranges, not its cross-process + * capabilities. + */ +TEST_F(process_madvise, basic) +{ + const unsigned long pagesize = self->page_size; + const int madvise_pages = 4; + struct iovec vec[madvise_pages]; + int pidfd = self->pidfd; + ssize_t ret; + char *map; + + /* + * 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); + if (map == MAP_FAILED) + SKIP(return, "mmap failed, not enough memory.\n"); + + /* 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, vec, madvise_pages, MADV_DONTNEED, 0); + if (ret == -1 && errno == EPERM) + SKIP(return, + "process_madvise() unsupported or permission denied, try running as root.\n"); + else if (errno == EINVAL) + SKIP(return, + "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; + + /* 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]; + + for (int i = 0; i < pagesize; i++) + ASSERT_EQ(unadvised_page[i], 'A'); + + /* Cleanup. */ + ASSERT_EQ(munmap(map, pagesize * 10), 0); +} + +/* + * 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, + * focus on process_madv remote result, only check addresses and lengths. + * The correctness of the MADV_COLLAPSE can be found in the relevant test examples in khugepaged. + */ +TEST_F(process_madvise, remote_collapse) +{ + const unsigned long pagesize = self->page_size; + 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) + SKIP(return, "Could not determine a valid huge page size.\n"); + + ASSERT_EQ(pipe(pipe_info), 0); + + self->child_pid = fork(); + ASSERT_NE(self->child_pid, -1); + + if (self->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(self->child_pid, NULL, 0); + SKIP(return, "Failed to read child info from pipe.\n"); + } + ASSERT_EQ(ret, sizeof(info)); + close(pipe_info[0]); + self->child_pid = info.pid; + + self->remote_pidfd = syscall(__NR_pidfd_open, self->child_pid, 0); + ASSERT_GE(self->remote_pidfd, 0); + + vec.iov_base = info.map_addr; + vec.iov_len = huge_page_size; + + ret = sys_process_madvise(self->remote_pidfd, &vec, 1, MADV_COLLAPSE, 0); + if (ret == -1) { + if (errno == EINVAL) + SKIP(return, "PROCESS_MADV_ADVISE is not supported.\n"); + else if (errno == EPERM) + SKIP(return, + "No process_madvise() permissions, try running as root.\n"); + return; + } + + ASSERT_EQ(ret, huge_page_size); +} + +/* + * Test process_madvise() with a pidfd for a process that has already + * exited to ensure correct error handling. + */ +TEST_F(process_madvise, exited_process_pidfd) +{ + struct iovec vec; + ssize_t ret; + int pidfd; + + vec.iov_base = (void *)0x1234; + vec.iov_len = 4096; + + /* + * Using a pidfd for a process that has already exited should fail + * with ESRCH. + */ + self->child_pid = fork(); + ASSERT_NE(self->child_pid, -1); + + if (self->child_pid == 0) + exit(0); + + pidfd = syscall(__NR_pidfd_open, self->child_pid, 0); + ASSERT_GE(pidfd, 0); + + /* Wait for the child to ensure it has terminated. */ + waitpid(self->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 bad pidfds to ensure correct error + * handling. + */ +TEST_F(process_madvise, bad_pidfd) +{ + struct iovec vec; + ssize_t ret; + + 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); +} + +/* + * Test process_madvise() with an invalid flag value. Currently, only a flag + * value of 0 is supported. This test is reserved for the future, e.g., if + * synchronous flags are added. + */ +TEST_F(process_madvise, flag) +{ + const unsigned long pagesize = self->page_size; + unsigned int invalid_flag; + int pidfd = self->pidfd; + struct iovec vec; + char *map; + ssize_t ret; + + map = mmap(NULL, pagesize, PROT_READ, MAP_PRIVATE | MAP_ANONYMOUS, -1, + 0); + if (map == MAP_FAILED) + SKIP(return, "mmap failed, not enough memory.\n"); + + vec.iov_base = map; + vec.iov_len = pagesize; + + invalid_flag = 0x80000000; + + ret = sys_process_madvise(pidfd, &vec, 1, MADV_DONTNEED, invalid_flag); + ASSERT_EQ(ret, -1); + ASSERT_EQ(errno, EINVAL); + + /* Cleanup. */ + ASSERT_EQ(munmap(map, pagesize), 0); +} + +TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/mm/run_vmtests.sh b/tools/testing/selftests/mm/run_vmtests.sh index a38c984103ce..471e539d82b8 100755 --- a/tools/testing/selftests/mm/run_vmtests.sh +++ b/tools/testing/selftests/mm/run_vmtests.sh @@ -65,6 +65,8 @@ separated by spaces: test pagemap_scan IOCTL - pfnmap tests for VM_PFNMAP handling +- process_madv + test for process_madv - cow test copy-on-write semantics - thp @@ -425,6 +427,9 @@ CATEGORY="madv_guard" run_test ./guard-regions # MADV_POPULATE_READ and MADV_POPULATE_WRITE tests CATEGORY="madv_populate" run_test ./madv_populate +# PROCESS_MADV test +CATEGORY="process_madv" run_test ./process_madv + CATEGORY="vma_merge" run_test ./merge if [ -x ./memfd_secret ] -- 2.43.0

1 month

2
2
0 0

[PATCH RFC 0/4] landlock: add LANDLOCK_SCOPE_MEMFD_EXEC execution

by Abhinav Saxena

This patch series introduces LANDLOCK_SCOPE_MEMFD_EXEC, a new Landlock scoping mechanism that restricts execution of anonymous memory file descriptors (memfd) created via memfd_create(2). This addresses security gaps where processes can bypass W^X policies and execute arbitrary code through anonymous memory objects. Fixes: https://github.com/landlock-lsm/linux/issues/37 SECURITY PROBLEM ================ Current Landlock filesystem restrictions do not cover memfd objects, allowing processes to: 1. Read-to-execute bypass: Create writable memfd, inject code, then execute via mmap(PROT_EXEC) or direct execve() 2. Anonymous execution: Execute code without touching the filesystem via execve("/proc/self/fd/N") where N is a memfd descriptor 3. Cross-domain access violations: Pass memfd between processes to bypass domain restrictions These scenarios can occur in sandboxed environments where filesystem access is restricted but memfd creation remains possible. IMPLEMENTATION ============== The implementation adds hierarchical execution control through domain scoping: Core Components: - is_memfd_file(): Reliable memfd detection via "memfd:" dentry prefix - domain_is_scoped(): Cross-domain hierarchy checking (moved to domain.c) - LSM hooks: mmap_file, file_mprotect, bprm_creds_for_exec - Creation-time restrictions: hook_file_alloc_security Security Matrix: Execution decisions follow domain hierarchy rules preventing both same-domain bypass attempts and cross-domain access violations while preserving legitimate hierarchical access patterns. Domain Hierarchy with LANDLOCK_SCOPE_MEMFD_EXEC: =============================================== Root (no domain) - No restrictions | +-- Domain A [SCOPE_MEMFD_EXEC] Layer 1 | +-- memfd_A (tagged with Domain A as creator) | | | +-- Domain A1 (child) [NO SCOPE] Layer 2 | | +-- Inherits Layer 1 restrictions from parent | | +-- memfd_A1 (can create, inherits restrictions) | | +-- Domain A1a [SCOPE_MEMFD_EXEC] Layer 3 | | +-- memfd_A1a (tagged with Domain A1a) | | | +-- Domain A2 (child) [SCOPE_MEMFD_EXEC] Layer 2 | +-- memfd_A2 (tagged with Domain A2 as creator) | +-- CANNOT access memfd_A1 (different subtree) | +-- Domain B [SCOPE_MEMFD_EXEC] Layer 1 +-- memfd_B (tagged with Domain B as creator) +-- CANNOT access ANY memfd from Domain A subtree Execution Decision Matrix: ======================== Executor-> | A | A1 | A1a | A2 | B | Root Creator | | | | | | ------------|-----|----|-----|----|----|----- Domain A | X | X | X | X | X | Y Domain A1 | Y | X | X | X | X | Y Domain A1a | Y | Y | X | X | X | Y Domain A2 | Y | X | X | X | X | Y Domain B | X | X | X | X | X | Y Root | Y | Y | Y | Y | Y | Y Legend: Y = Execution allowed, X = Execution denied Scenarios Covered: - Direct mmap(PROT_EXEC) on memfd files - Two-stage mmap(PROT_READ) + mprotect(PROT_EXEC) bypass attempts - execve("/proc/self/fd/N") anonymous execution - execveat() and fexecve() file descriptor execution - Cross-process memfd inheritance and IPC passing TESTING ======= All patches have been validated with: - scripts/checkpatch.pl --strict (clean) - Selftests covering same-domain restrictions, cross-domain hierarchy enforcement, and regular file isolation - KUnit tests for memfd detection edge cases DISCLAIMER ========== My understanding of Landlock scoping semantics may be limited, but this implementation reflects my current understanding based on available documentation and code analysis. I welcome feedback and corrections regarding the scoping logic and domain hierarchy enforcement. Signed-off-by: Abhinav Saxena <xandfury(a)gmail.com> --- Abhinav Saxena (4): landlock: add LANDLOCK_SCOPE_MEMFD_EXEC scope landlock: implement memfd detection landlock: add memfd exec LSM hooks and scoping selftests/landlock: add memfd execution tests include/uapi/linux/landlock.h | 5 + security/landlock/.kunitconfig | 1 + security/landlock/audit.c | 4 + security/landlock/audit.h | 1 + security/landlock/cred.c | 14 - security/landlock/domain.c | 67 ++++ security/landlock/domain.h | 4 + security/landlock/fs.c | 405 ++++++++++++++++++++- security/landlock/limits.h | 2 +- security/landlock/task.c | 67 ---- .../selftests/landlock/scoped_memfd_exec_test.c | 325 +++++++++++++++++ 11 files changed, 812 insertions(+), 83 deletions(-) --- base-commit: 5b74b2eff1eeefe43584e5b7b348c8cd3b723d38 change-id: 20250716-memfd-exec-ac0d582018c3 Best regards, -- Abhinav Saxena <xandfury(a)gmail.com>

1 month

2
6
0 0

[PATCH net 1/2] macsec: set IFF_UNICAST_FLT priv flag

by Stanislav Fomichev

Cosmin reports the following locking issue: # BUG: sleeping function called from invalid context at kernel/locking/mutex.c:275 # dump_stack_lvl+0x4f/0x60 # __might_resched+0xeb/0x140 # mutex_lock+0x1a/0x40 # dev_set_promiscuity+0x26/0x90 # __dev_set_promiscuity+0x85/0x170 # __dev_set_rx_mode+0x69/0xa0 # dev_uc_add+0x6d/0x80 # vlan_dev_open+0x5f/0x120 [8021q] # __dev_open+0x10c/0x2a0 # __dev_change_flags+0x1a4/0x210 # netif_change_flags+0x22/0x60 # do_setlink.isra.0+0xdb0/0x10f0 # rtnl_newlink+0x797/0xb00 # rtnetlink_rcv_msg+0x1cb/0x3f0 # netlink_rcv_skb+0x53/0x100 # netlink_unicast+0x273/0x3b0 # netlink_sendmsg+0x1f2/0x430 Which is similar to recent syzkaller reports in [0] and [1] and triggers because macsec does not advertise IFF_UNICAST_FLT although it has proper ndo_set_rx_mode callback that takes care of pushing uc/mc addresses down to the real device. In general, dev_uc_add call path is problematic for stacking non-IFF_UNICAST_FLT because we might grab netdev instance lock under addr_list_lock spinlock, so this is not a systemic fix. 0: https://lore.kernel.org/netdev/686d55b4.050a0220.1ffab7.0014.GAE@google.com 1: https://lore.kernel.org/netdev/68712acf.a00a0220.26a83e.0051.GAE@google.com/ Link: 2aff4342b0f5b1539c02ffd8df4c7e58dd9746e7.camel(a)nvidia.com Fixes: 7e4d784f5810 ("net: hold netdev instance lock during rtnetlink operations") Reported-by: Cosmin Ratiu <cratiu(a)nvidia.com> Tested-by: Cosmin Ratiu <cratiu(a)nvidia.com> Signed-off-by: Stanislav Fomichev <sdf(a)fomichev.me> --- drivers/net/macsec.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 7edbe76b5455..4c75d1fea552 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -3868,7 +3868,7 @@ static void macsec_setup(struct net_device *dev) ether_setup(dev); dev->min_mtu = 0; dev->max_mtu = ETH_MAX_MTU; - dev->priv_flags |= IFF_NO_QUEUE; + dev->priv_flags |= IFF_NO_QUEUE | IFF_UNICAST_FLT; dev->netdev_ops = &macsec_netdev_ops; dev->needs_free_netdev = true; dev->priv_destructor = macsec_free_netdev; -- 2.50.1

1 month

3
4
0 0

[PATCH v2 0/7] Replace "__auto_type" with "auto"

by H. Peter Anvin

"auto" was defined as a keyword back in the K&R days, but as a storage type specifier. No one ever used it, since it was and is the default storage type for local variables. C++11 recycled the keyword to allow a type to be declared based on the type of an initializer. This was finally adopted into standard C in C23. gcc and clang provide the "__auto_type" alias keyword as an extension for pre-C23, however, there is no reason to pollute the bulk of the source base with this temporary keyword; instead define "auto" as a macro unless the compiler is running in C23+ mode. This macro is added in <linux/compiler_types.h> because that header is included in some of the tools headers, wheres <linux/compiler.h> is not as it has a bunch of very kernel-specific things in it. Changes in v2: - Restore indentation of macro backslashes (David Laight) - arch/nios2: Replace an adjacent typeof() with a similar "auto" construct (Linus Torvalds) - fs/proc/inode.c: change "__auto_type" to "const auto" (Alexey Dobriyan) --- arch/nios2/include/asm/uaccess.h | 8 ++++---- arch/x86/include/asm/bug.h | 2 +- arch/x86/include/asm/string_64.h | 6 +++--- arch/x86/include/asm/uaccess_64.h | 2 +- fs/proc/inode.c | 16 ++++++++-------- include/linux/cleanup.h | 6 +++--- include/linux/compiler.h | 2 +- include/linux/compiler_types.h | 13 +++++++++++++ include/linux/minmax.h | 6 +++--- tools/testing/selftests/bpf/prog_tests/socket_helpers.h | 9 +++++++-- tools/virtio/linux/compiler.h | 2 +- 11 files changed, 45 insertions(+), 27 deletions(-)

1 month

5
12
0 0

[PATCH] kselftest/arm64: Test FPSIMD format data writes via NT_ARM_SVE in fp-ptrace

by Mark Brown

The NT_ARM_SVE register set supports two data formats, the native SVE one and an alternative format where we embed a copy of user_fpsimd_data as used for NT_PRFPREG in the SVE register set. The register data is set as for a write to NT_PRFPREG and changes in vector length and streaming mode are handled as for any NT_ARM_SVE write. This has not previously been tested by fp-ptrace, add coverage of it. We do not support writes in FPSIMD format for NT_ARM_SSVE so we skip the test for anything that would leave us in streaming mode. Signed-off-by: Mark Brown <broonie(a)kernel.org> --- tools/testing/selftests/arm64/fp/fp-ptrace.c | 66 +++++++++++++++++++++++++++- 1 file changed, 64 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/arm64/fp/fp-ptrace.c b/tools/testing/selftests/arm64/fp/fp-ptrace.c index 191c47ca0ed8..c479c97dea1a 100644 --- a/tools/testing/selftests/arm64/fp/fp-ptrace.c +++ b/tools/testing/selftests/arm64/fp/fp-ptrace.c @@ -1066,6 +1066,23 @@ static bool sve_write_supported(struct test_config *config) return true; } +static bool sve_write_fpsimd_supported(struct test_config *config) +{ + if (!sve_supported()) + return false; + + if ((config->svcr_in & SVCR_ZA) != (config->svcr_expected & SVCR_ZA)) + return false; + + if (config->svcr_expected & SVCR_SM) + return false; + + if (config->sme_vl_in != config->sme_vl_expected) + return false; + + return true; +} + static void fpsimd_write_expected(struct test_config *config) { int vl; @@ -1152,7 +1169,7 @@ static void sve_write_expected(struct test_config *config) } } -static void sve_write(pid_t child, struct test_config *config) +static void sve_write_sve(pid_t child, struct test_config *config) { struct user_sve_header *sve; struct iovec iov; @@ -1195,6 +1212,45 @@ static void sve_write(pid_t child, struct test_config *config) free(iov.iov_base); } +static void sve_write_fpsimd(pid_t child, struct test_config *config) +{ + struct user_sve_header *sve; + struct user_fpsimd_state *fpsimd; + struct iovec iov; + int ret, vl, vq; + + vl = vl_expected(config); + vq = __sve_vq_from_vl(vl); + + if (!vl) + return; + + iov.iov_len = SVE_PT_SVE_OFFSET + SVE_PT_SVE_SIZE(vq, + SVE_PT_REGS_FPSIMD); + iov.iov_base = malloc(iov.iov_len); + if (!iov.iov_base) { + ksft_print_msg("Failed allocating %lu byte SVE write buffer\n", + iov.iov_len); + return; + } + memset(iov.iov_base, 0, iov.iov_len); + + sve = iov.iov_base; + sve->size = iov.iov_len; + sve->flags = SVE_PT_REGS_FPSIMD; + sve->vl = vl; + + fpsimd = iov.iov_base + SVE_PT_REGS_OFFSET; + memcpy(&fpsimd->vregs, v_expected, sizeof(v_expected)); + + ret = ptrace(PTRACE_SETREGSET, child, NT_ARM_SVE, &iov); + if (ret != 0) + ksft_print_msg("Failed to write SVE: %s (%d)\n", + strerror(errno), errno); + + free(iov.iov_base); +} + static bool za_write_supported(struct test_config *config) { if ((config->svcr_in & SVCR_SM) != (config->svcr_expected & SVCR_SM)) @@ -1386,7 +1442,13 @@ static struct test_definition sve_test_defs[] = { .name = "SVE write", .supported = sve_write_supported, .set_expected_values = sve_write_expected, - .modify_values = sve_write, + .modify_values = sve_write_sve, + }, + { + .name = "SVE write FPSIMD format", + .supported = sve_write_fpsimd_supported, + .set_expected_values = fpsimd_write_expected, + .modify_values = sve_write_fpsimd, }, }; --- base-commit: 86731a2a651e58953fc949573895f2fa6d456841 change-id: 20250718-arm64-fp-ptrace-sve-fpsimd-ea20bdd9138b Best regards, -- Mark Brown <broonie(a)kernel.org>

1 month

2
1
0 0

[PATCH] kselftest/arm64: Allow sve-ptrace to run on SME only systems

by Mark Brown

Currently the sve-ptrace test program only runs if the system supports SVE but since SME includes streaming SVE the tests it offers are valid even on a system that only supports SME. Since the tests already have individual hwcap checks just remove the top level test and rely on those. Signed-off-by: Mark Brown <broonie(a)kernel.org> --- tools/testing/selftests/arm64/fp/sve-ptrace.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/tools/testing/selftests/arm64/fp/sve-ptrace.c b/tools/testing/selftests/arm64/fp/sve-ptrace.c index 7f9b6a61d369..b22303778fb0 100644 --- a/tools/testing/selftests/arm64/fp/sve-ptrace.c +++ b/tools/testing/selftests/arm64/fp/sve-ptrace.c @@ -753,9 +753,6 @@ int main(void) ksft_print_header(); ksft_set_plan(EXPECTED_TESTS); - if (!(getauxval(AT_HWCAP) & HWCAP_SVE)) - ksft_exit_skip("SVE not available\n"); - child = fork(); if (!child) return do_child(); --- base-commit: 9e8ebfe677f9101bbfe1f75d548a5aec581e8213 change-id: 20250718-arm64-sve-ptrace-sme-only-4ab49d037295 Best regards, -- Mark Brown <broonie(a)kernel.org>

1 month

2
1
0 0

[PATCH 0/3] kselftest/arm64: Fixes for fp-ptrace on SME only systems

by Mark Brown

When testing SME only systems I noticed that fp-ptrace does not cope at all well with them, this series fixes the major issues so that the test program completes successfully. The reason I was looking at this is that following the recent round of fixes to ptrace we do not currently offer any mechanism for disabling streaming mode via ptrace, this series brings the program to a point where it tests the currently implemented ABI. A further series allowing the disabling of streaming mode via ptrace will follow. Signed-off-by: Mark Brown <broonie(a)kernel.org> --- Mark Brown (3): kselftest/arm64: Test SME on SME only systems in fp-ptrace kselftest/arm64: Fix SVE write data generation for SME only systems kselftest/arm64: Handle attempts to disable SM on SME only systems tools/testing/selftests/arm64/fp/fp-ptrace.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) --- base-commit: 86731a2a651e58953fc949573895f2fa6d456841 change-id: 20250718-arm64-fp-ptrace-sme-only-ab327d7f0d32 Best regards, -- Mark Brown <broonie(a)kernel.org>

1 month

2
4
0 0

[PATCH v25 net-next 0/6] DUALPI2 patch

by chia-yu.chang＠nokia-bell-labs.com

From: Chia-Yu Chang <chia-yu.chang(a)nokia-bell-labs.com> Hello, Please find the DualPI2 patch v25. This patch serise adds DualPI Improved with a Square (DualPI2) with following features: * Supports congestion controls that comply with the Prague requirements in RFC9331 (e.g. TCP-Prague) * Coupled dual-queue that separates the L4S traffic in a low latency queue (L-queue), without harming remaining traffic that is scheduled in classic queue (C-queue) due to congestion-coupling using PI2 as defined in RFC9332 * Configurable overload strategies * Use of sojourn time to reliably estimate queue delay * Supports ECN L4S-identifier (IP.ECN==0b*1) to classify traffic into respective queues For more details of DualPI2, please refer IETF RFC9332 (https://datatracker.ietf.org/doc/html/rfc9332). Best regards, Chia-Yu --- v25 (19-Jul-2025) - Fix the missing sch_tree_unlock() in v24 (Jakub Kicinski <kuba(a)kernel.org>) v24 (18-Jul-2025) - Replace TCA_DUALPI2 prefix with TC_DUALPI2 for enums in pkt_sched.h (Jakub Kicinski <kuba(a)kernel.org>) - Report error if both packet and time step thresholds are provided (Jakub Kicinski <kuba(a)kernel.org>) v23 (13-Jul-2025) and v22 (11-Jul-2025) - Fix issue when user would like to change DualPI2 but provides an empty TCA_OPTIONS with no nested attributes (Paolo Abeni <pabeni(a)redhat.com>, Jakub Kicinski <kuba(a)kernel.org>) v21 (02-Jul-2025) - Replace STEP_THRESH and STEP_PACKETS with STEP_THRESH_PKTS and STEP_THRESH_US (Jakub Kicinski <kuba(a)kernel.org>) - Move READ_ONCE and WRITE_ONCE to later DualPI2 patches (Jakub Kicinski <kuba(a)kernel.org>) - Replace NLA_POLICY_FULL_RANGE with NLA_POLICY_RANGE (Jakub Kicinski <kuba(a)kernel.org>) - Set extra error message for dualpi2_change (Jakub Kicinski <kuba(a)kernel.org>) - Drop redundant else for better readability (Paolo Abeni <pabeni(a)redhat.com>) - Replace step-thresh and step-packets with step-thresh-pkts and step-thresh-us (Jakub Kicinski <kuba(a)kernel.org>) - Remove redundant name-prefix and simplify entries of dualpi2 enums (Jakub Kicinski <kuba(a)kernel.org>) - Fix some typos and format issues of dualpi2 attributes v20 (21-Jun-2025) - Add one more commit to fix warning and style check on tdc.sh reported by shellcheck - Remove double-prefixed of "tc_tc_dualpi2_attrs" in tc-user.h (Donald Hunter <donald.hunter(a)gmail.com>) v19 (14-Jun-2025) - Fix one typo in the comment of #1 (ALOK TIWARI <alok.a.tiwari(a)oracle.com>) - Update commit message of #4 (ALOK TIWARI <alok.a.tiwari(a)oracle.com>) - Wrap long lines of Documentation/netlink/specs/tc.yaml to within 80 characters (Jakub Kicinski <kuba(a)kernel.org>) v18 (13-Jun-2025) - Add the num of enum used by DualPI2 and fix name and name-prefix of DualPI2 enum and attribute - Replace from_timer() with timer_container_of() (Pedro Tammela <pctammela(a)mojatatu.com>) v17 (25-May-2025, Resent at 11-Jun-2025) - Replace 0xffffffff with U32_MAX (Paolo Abeni <pabeni(a)redhat.com>) - Use helper function qdisc_dequeue_internal() and add new helper function skb_apply_step() (Paolo Abeni <pabeni(a)redhat.com>) - Add s64 casting when calculating the delta of the PI controller (Paolo Abeni <pabeni(a)redhat.com>) - Change the drop reason into SKB_DROP_REASON_QDISC_CONGESTED for drop_early (Paolo Abeni <pabeni(a)redhat.com>) - Modify the condition to remove the original skb when enqueuing multiple GSO segments (Paolo Abeni <pabeni(a)redhat.com>) - Add READ_ONCE() in dualpi2_dump_stat() (Paolo Abeni <pabeni(a)redhat.com>) - Add comments, brackets, and brackets for readability (Paolo Abeni <pabeni(a)redhat.com>) v16 (16-MAy-2025) - Add qdisc_lock() to dualpi2_timer() in dualpi2_timer (Paolo Abeni <pabeni(a)redhat.com>) - Introduce convert_ns_to_usec() to convert usec to nsec without overflow in #1 (Paolo Abeni <pabeni(a)redhat.com>) - Update convert_us_tonsec() to convert nsec to usec without overflow in #2 (Paolo Abeni <pabeni(a)redhat.com>) - Add more descriptions with respect to DualPI2 in the cover ltter and add changelog in each patch (Paolo Abeni <pabeni(a)redhat.com>) v15 (09-May-2025) - Add enum of TCA_DUALPI2_ECN_MASK_CLA_ECT to remove potential leakeage in #1 (Simon Horman <horms(a)kernel.org>) - Fix one typo in comment of #2 - Update tc.yaml in #5 to aligh with the updated enum of pkt_sched.h v14 (05-May-2025) - Modify tc.yaml: (1) Replace flags with enum and remove enum-as-flags, (2) Remove credit-queue in xstats, and (3) Change attribute types (Donald Hunter <donald.hun - Add enum and fix the ordering of variables in pkt_sched.h to align with the modified tc.yaml (Donald Hunter <donald.hunter(a)gmail.com>) - Add validators for DROP_OVERLOAD, DROP_EARLY, ECN_MASK, and SPLIT_GSO in sch_dualpi2.c (Donald Hunter <donald.hunter(a)gmail.com>) - Update dualpi2.json to align with the updated variable order in pkt_sched.h - Reorder patches (Donald Hunter <donald.hunter(a)gmail.com>) v13 (26-Apr-2025) - Use dashes in member names to follow YNL conventions in tc.yaml (Donald Hunter <donald.hunter(a)gmail.com>) - Define enumerations separately for flags of drop-early, drop-overload, ecn-mask, credit-queue in tc.yaml (Donald Hunter <donald.hunter(a)gmail.com>) - Change the types of split-gso and step-packets into flag in tc.yaml (Donald Hunter <donald.hunter(a)gmail.com>) - Revert to u32/u8 types for tc-dualpi2-xstats members in tc.yaml (Donald Hunter <donald.hunter(a)gmail.com>) - Add new test cases in tc-tests/qdiscs/dualpi2.json to cover all dualpi2 parameters (Donald Hunter <donald.hunter(a)gmail.com>) - Change the type of TCA_DUALPI2_STEP_PACKETS into NLA_FLAG (Donald Hunter <donald.hunter(a)gmail.com>) v12 (22-Apr-2025) - Remove anonymous struct in sch_dualpi2.c (Paolo Abeni <pabeni(a)redhat.com>) - Replace u32/u8 with uint and s32 with int in tc spec document (Paolo Abeni <pabeni(a)redhat.com>) - Introduce get_memory_limit function to handle potential overflow when multipling limit with MTU (Paolo Abeni <pabeni(a)redhat.com>) - Double the packet length to further include packet overhead in memory_limit (Paolo Abeni <pabeni(a)redhat.com>) - Remove the check of qdisc_qlen(sch) when calling qdisc_tree_reduce_backlog (Paolo Abeni <pabeni(a)redhat.com>) v11 (15-Apr-2025) - Replace hstimer_init with hstimer_setup in sch_dualpi2.c v10 (25-Mar-2025) - Remove leftover include in include/linux/netdevice.h and anonymous struct in sch_dualpi2.c (Paolo Abeni <pabeni(a)redhat.com>) - Use kfree_skb_reason() and add SKB_DROP_REASON_DUALPI2_STEP_DROP drop reason (Paolo Abeni <pabeni(a)redhat.com>) - Split sch_dualpi2.c into 3 patches (and overall 5 patches): Struct definition & parsing, Dump stats & configuration, Enqueue/Dequeue (Paolo Abeni <pabeni(a)redhat.com>) v9 (16-Mar-2025) - Fix mem_usage error in previous version - Add min_qlen_step to the dualpi2 attribute as the minimum queue length in number of packets in the L-queue to start step threshold marking. In previous versions, this value was fixed to 2, so the step threshold was applied to mark packets in the L queue only when the queue length of the L queue was greater than or equal to 2 packets. This will cause larger queuing delays for L4S traffic at low rates (<20Mbps). So we parameterize it and change the default value to 0. Comparison of tcp_1down run 'HTB 20Mbit + DUALPI2 + 10ms base delay' Old versions: avg median # data pts Ping (ms) ICMP : 11.55 11.70 ms 350 TCP upload avg : 18.96 N/A Mbits/s 350 TCP upload sum : 18.96 N/A Mbits/s 350 New version (v9): avg median # data pts Ping (ms) ICMP : 10.81 10.70 ms 350 TCP upload avg : 18.91 N/A Mbits/s 350 TCP upload sum : 18.91 N/A Mbits/s 350 Comparison of tcp_1down run 'HTB 10Mbit + DUALPI2 + 10ms base delay' Old versions: avg median # data pts Ping (ms) ICMP : 12.61 12.80 ms 350 TCP upload avg : 9.48 N/A Mbits/s 350 TCP upload sum : 9.48 N/A Mbits/s 350 New version (v9): avg median # data pts Ping (ms) ICMP : 11.06 10.80 ms 350 TCP upload avg : 9.43 N/A Mbits/s 350 TCP upload sum : 9.43 N/A Mbits/s 350 Comparison of tcp_1down run 'HTB 10Mbit + DUALPI2 + 10ms base delay' Old versions: avg median # data pts Ping (ms) ICMP : 40.86 37.45 ms 350 TCP upload avg : 0.88 N/A Mbits/s 350 TCP upload sum : 0.88 N/A Mbits/s 350 TCP upload::1 : 0.88 0.97 Mbits/s 350 New version (v9): avg median # data pts Ping (ms) ICMP : 11.07 10.40 ms 350 TCP upload avg : 0.55 N/A Mbits/s 350 TCP upload sum : 0.55 N/A Mbits/s 350 TCP upload::1 : 0.55 0.59 Mbits/s 350 v8 (11-Mar-2025) - Fix warning messages in v7 v7 (07-Mar-2025) - Separate into 3 patches to avoid mixing changes of documentation, selftest, and code. (Cong Wang <xiyou.wangcong(a)gmail.com>) v6 (04-Mar-2025) - Add modprobe for dulapi2 in tc-testing script tc-testing/tdc.sh (Jakub Kicinski <kuba(a)kernel.org>) - Update test cases in dualpi2.json - Update commit message v5 (22-Feb-2025) - A comparison was done between MQ + DUALPI2, MQ + FQ_PIE, MQ + FQ_CODEL: Unshaped 1gigE with 4 download streams test: - Summary of tcp_4down run 'MQ + FQ_CODEL': avg median # data pts Ping (ms) ICMP : 1.19 1.34 ms 349 TCP download avg : 235.42 N/A Mbits/s 349 TCP download sum : 941.68 N/A Mbits/s 349 TCP download::1 : 235.19 235.39 Mbits/s 349 TCP download::2 : 235.03 235.35 Mbits/s 349 TCP download::3 : 236.89 235.44 Mbits/s 349 TCP download::4 : 234.57 235.19 Mbits/s 349 - Summary of tcp_4down run 'MQ + FQ_PIE' avg median # data pts Ping (ms) ICMP : 1.21 1.37 ms 350 TCP download avg : 235.42 N/A Mbits/s 350 TCP download sum : 941.61 N/A Mbits/s 350 TCP download::1 : 232.54 233.13 Mbits/s 350 TCP download::2 : 232.52 232.80 Mbits/s 350 TCP download::3 : 233.14 233.78 Mbits/s 350 TCP download::4 : 243.41 241.48 Mbits/s 350 - Summary of tcp_4down run 'MQ + DUALPI2' avg median # data pts Ping (ms) ICMP : 1.19 1.34 ms 349 TCP download avg : 235.42 N/A Mbits/s 349 TCP download sum : 941.68 N/A Mbits/s 349 TCP download::1 : 235.19 235.39 Mbits/s 349 TCP download::2 : 235.03 235.35 Mbits/s 349 TCP download::3 : 236.89 235.44 Mbits/s 349 TCP download::4 : 234.57 235.19 Mbits/s 349 Unshaped 1gigE with 128 download streams test: - Summary of tcp_128down run 'MQ + FQ_CODEL': avg median # data pts Ping (ms) ICMP : 1.88 1.86 ms 350 TCP download avg : 7.39 N/A Mbits/s 350 TCP download sum : 946.47 N/A Mbits/s 350 - Summary of tcp_128down run 'MQ + FQ_PIE': avg median # data pts Ping (ms) ICMP : 1.88 1.86 ms 350 TCP download avg : 7.39 N/A Mbits/s 350 TCP download sum : 946.47 N/A Mbits/s 350 - Summary of tcp_128down run 'MQ + DUALPI2': avg median # data pts Ping (ms) ICMP : 1.88 1.86 ms 350 TCP download avg : 7.39 N/A Mbits/s 350 TCP download sum : 946.47 N/A Mbits/s 350 Unshaped 10gigE with 4 download streams test: - Summary of tcp_4down run 'MQ + FQ_CODEL': avg median # data pts Ping (ms) ICMP : 0.22 0.23 ms 350 TCP download avg : 2354.08 N/A Mbits/s 350 TCP download sum : 9416.31 N/A Mbits/s 350 TCP download::1 : 2353.65 2352.81 Mbits/s 350 TCP download::2 : 2354.54 2354.21 Mbits/s 350 TCP download::3 : 2353.56 2353.78 Mbits/s 350 TCP download::4 : 2354.56 2354.45 Mbits/s 350 - Summary of tcp_4down run 'MQ + FQ_PIE': avg median # data pts Ping (ms) ICMP : 0.20 0.19 ms 350 TCP download avg : 2354.76 N/A Mbits/s 350 TCP download sum : 9419.04 N/A Mbits/s 350 TCP download::1 : 2354.77 2353.89 Mbits/s 350 TCP download::2 : 2353.41 2354.29 Mbits/s 350 TCP download::3 : 2356.18 2354.19 Mbits/s 350 TCP download::4 : 2354.68 2353.15 Mbits/s 350 - Summary of tcp_4down run 'MQ + DUALPI2': avg median # data pts Ping (ms) ICMP : 0.24 0.24 ms 350 TCP download avg : 2354.11 N/A Mbits/s 350 TCP download sum : 9416.43 N/A Mbits/s 350 TCP download::1 : 2354.75 2353.93 Mbits/s 350 TCP download::2 : 2353.15 2353.75 Mbits/s 350 TCP download::3 : 2353.49 2353.72 Mbits/s 350 TCP download::4 : 2355.04 2353.73 Mbits/s 350 Unshaped 10gigE with 128 download streams test: - Summary of tcp_128down run 'MQ + FQ_CODEL': avg median # data pts Ping (ms) ICMP : 7.57 8.69 ms 350 TCP download avg : 73.97 N/A Mbits/s 350 TCP download sum : 9467.82 N/A Mbits/s 350 - Summary of tcp_128down run 'MQ + FQ_PIE': avg median # data pts Ping (ms) ICMP : 7.82 8.91 ms 350 TCP download avg : 73.97 N/A Mbits/s 350 TCP download sum : 9468.42 N/A Mbits/s 350 - Summary of tcp_128down run 'MQ + DUALPI2': avg median # data pts Ping (ms) ICMP : 6.87 7.93 ms 350 TCP download avg : 73.95 N/A Mbits/s 350 TCP download sum : 9465.87 N/A Mbits/s 350 From the results shown above, we see small differences between combinations. - Update commit message to include results of no_split_gso and split_gso (Dave Taht <dave.taht(a)gmail.com> and Paolo Abeni <pabeni(a)redhat.com>) - Add memlimit in the dualpi2 attribute, and add memory_used, max_memory_used, memory_limit in dualpi2 stats (Dave Taht <dave.taht(a)gmail.com>) - Update note in sch_dualpi2.c related to BBRv3 status (Dave Taht <dave.taht(a)gmail.com>) - Update license identifier (Dave Taht <dave.taht(a)gmail.com>) - Add selftest in tools/testing/selftests/tc-testing (Cong Wang <xiyou.wangcong(a)gmail.com>) - Use netlink policies for parameter checks (Jamal Hadi Salim <jhs(a)mojatatu.com>) - Modify texts & fix typos in Documentation/netlink/specs/tc.yaml (Dave Taht <dave.taht(a)gmail.com>) - Add descriptions of packet counter statistics and the reset function of sch_dualpi2.c - Fix step_thresh in packets - Update code comments in sch_dualpi2.c v4 (22-Oct-2024) - Update statement in Kconfig for DualPI2 (Stephen Hemminger <stephen(a)networkplumber.org>) - Put a blank line after #define in sch_dualpi2.c (Stephen Hemminger <stephen(a)networkplumber.org>) - Fix line length warning. v3 (19-Oct-2024) - Fix compilaiton error - Update Documentation/netlink/specs/tc.yaml (Jakub Kicinski <kuba(a)kernel.org>) v2 (18-Oct-2024) - Add Documentation/netlink/specs/tc.yaml (Jakub Kicinski <kuba(a)kernel.org>) - Use dualpi2 instead of skb prefix (Jamal Hadi Salim <jhs(a)mojatatu.com>) - Replace nla_parse_nested_deprecated with nla_parse_nested (Jamal Hadi Salim <jhs(a)mojatatu.com>) - Fix line length warning --- Chia-Yu Chang (5): sched: Struct definition and parsing of dualpi2 qdisc sched: Dump configuration and statistics of dualpi2 qdisc selftests/tc-testing: Fix warning and style check on tdc.sh selftests/tc-testing: Add selftests for qdisc DualPI2 Documentation: netlink: specs: tc: Add DualPI2 specification Koen De Schepper (1): sched: Add enqueue/dequeue of dualpi2 qdisc Documentation/netlink/specs/tc.yaml | 151 ++- include/net/dropreason-core.h | 6 + include/uapi/linux/pkt_sched.h | 68 + net/sched/Kconfig | 12 + net/sched/Makefile | 1 + net/sched/sch_dualpi2.c | 1177 +++++++++++++++++ tools/testing/selftests/tc-testing/config | 1 + .../tc-testing/tc-tests/qdiscs/dualpi2.json | 254 ++++ tools/testing/selftests/tc-testing/tdc.sh | 6 +- 9 files changed, 1671 insertions(+), 5 deletions(-) create mode 100644 net/sched/sch_dualpi2.c create mode 100644 tools/testing/selftests/tc-testing/tc-tests/qdiscs/dualpi2.json -- 2.34.1

1 month

3
8
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-kselftest-mirror