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

[PATCH net-next v2 0/7] netpoll: Factor out functions from netpoll_send_udp() and add ipv6 selftest

by Breno Leitao

Refactors the netpoll UDP transmit path to improve code clarity, maintainability, and protocol-layer encapsulation. Function netpoll_send_udp() has more than 100 LoC, which is hard to understand and review. After this patchset, it has only 32 LoC, which is more manageable. The series systematically moves the construction of protocol headers (UDP, IPv4, IPv6, Ethernet) out of the core `netpoll_send_udp()` function into dedicated static helpers: - `push_udp()` for UDP header setup - `push_ipv4()` and `push_ipv6()` for IP header setup - `push_eth()` for Ethernet header setup This results in a clean, layered abstraction that mirrors the protocol stack, reduces code duplication, and improves readability. Also, to make sure this is not breaking anything, add IPv6 selftest to netconsole tests, which will exercise this code. This test would also pick problems similiar to the one fixed by f599020702698 ("net: netpoll: Initialize UDP checksum field before checksumming"), which was embarrassin we didn't have a selftest catch it. Anyway, there are **no functional changes** intended in this patchset. Signed-off-by: Breno Leitao <leitao(a)debian.org> --- Changes in v2: - Move ethernet data from push_ipv{4,6} into push_eth() function, making the slice up even more clearer (Jakub). - Fixed some long lines to get checkpatch happier. - Link to v1: https://lore.kernel.org/r/20250627-netpoll_untagle_ip-v1-0-61a21692f84a@deb… --- Breno Leitao (7): netpoll: Improve code clarity with explicit struct size calculations netpoll: factor out UDP checksum calculation into helper netpoll: factor out IPv6 header setup into push_ipv6() helper netpoll: factor out IPv4 header setup into push_ipv4() helper netpoll: factor out UDP header setup into push_udp() helper netpoll: move Ethernet setup to push_eth() helper selftests: net: Add IPv6 support to netconsole basic tests net/core/netpoll.c | 192 +++++++++++++-------- .../selftests/drivers/net/lib/sh/lib_netcons.sh | 76 +++++++- .../testing/selftests/drivers/net/netcons_basic.sh | 53 +++--- 3 files changed, 215 insertions(+), 106 deletions(-) --- base-commit: 8efa26fcbf8a7f783fd1ce7dd2a409e9b7758df0 change-id: 20250620-netpoll_untagle_ip-e37c799a6925 Best regards, -- Breno Leitao <leitao(a)debian.org>

3 months, 3 weeks

3
16
0 0

[PATCH 0/2] tracing: Fixes for filter

by Masami Hiramatsu (Google)

Hi, Here is a patch series to fix some issues on the trace event and function filters. The first patch fixes an issue that the event filter can not handle the string pointer with BTF attribute tag. This happens with CONFIG_DEBUG_INFO_BTF=y and PAHOLE_HAS_BTF_TAG=y. The second patch fixes a selftest issue on the function glob filter. Since mutex_trylock() can be an inline function, it is not a good example for ftrace. This replaces it with mutex_unlock(). Thank you, --- Masami Hiramatsu (Google) (2): tracing: Handle "(const) char __attribute() *" as string ptr type selftests: tracing: Use mutex_unlock for testing glob filter kernel/trace/trace_events_filter.c | 5 +++++ tools/testing/selftests/ftrace/test.d/ftrace/func-filter-glob.tc | 2 +- 2 files changed, 6 insertions(+), 1 deletion(-)

3 months, 3 weeks

4
7
0 0

[PATCH v2 1/2] selftests: cachestat: add tests for mmap

by Suresh K C

From: Suresh K C <suresh.k.chandrappa(a)gmail.com> Add a test case to verify cachestat behavior with memory-mapped files using mmap(). This ensures that pages accessed via mmap are correctly accounted for in the page cache. Tested on x86_64 with default kernel config Signed-off-by: Suresh K C <suresh.k.chandrappa(a)gmail.com> --- .../selftests/cachestat/test_cachestat.c | 39 ++++++++++++++++--- 1 file changed, 33 insertions(+), 6 deletions(-) diff --git a/tools/testing/selftests/cachestat/test_cachestat.c b/tools/testing/selftests/cachestat/test_cachestat.c index 632ab44737ec..b6452978dae0 100644 --- a/tools/testing/selftests/cachestat/test_cachestat.c +++ b/tools/testing/selftests/cachestat/test_cachestat.c @@ -33,6 +33,11 @@ void print_cachestat(struct cachestat *cs) cs->nr_evicted, cs->nr_recently_evicted); } +enum file_type { + FILE_MMAP, + FILE_SHMEM +}; + bool write_exactly(int fd, size_t filesize) { int random_fd = open("/dev/urandom", O_RDONLY); @@ -202,7 +207,7 @@ static int test_cachestat(const char *filename, bool write_random, bool create, return ret; } -bool test_cachestat_shmem(void) +bool run_cachestat_test(enum file_type type) { size_t PS = sysconf(_SC_PAGESIZE); size_t filesize = PS * 512 * 2; /* 2 2MB huge pages */ @@ -212,27 +217,43 @@ bool test_cachestat_shmem(void) char *filename = "tmpshmcstat"; struct cachestat cs; bool ret = true; + int fd; unsigned long num_pages = compute_len / PS; - int fd = shm_open(filename, O_CREAT | O_RDWR, 0600); + if (type == FILE_SHMEM) + fd = shm_open(filename, O_CREAT | O_RDWR, 0600); + else + fd = open(filename, O_RDWR | O_CREAT | O_TRUNC, 0666); if (fd < 0) { - ksft_print_msg("Unable to create shmem file.\n"); + ksft_print_msg("Unable to create file.\n"); ret = false; goto out; } if (ftruncate(fd, filesize)) { - ksft_print_msg("Unable to truncate shmem file.\n"); + ksft_print_msg("Unable to truncate file.\n"); ret = false; goto close_fd; } if (!write_exactly(fd, filesize)) { - ksft_print_msg("Unable to write to shmem file.\n"); + ksft_print_msg("Unable to write to file.\n"); ret = false; goto close_fd; } + if (type == FILE_MMAP){ + char *map = mmap(NULL, filesize, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0); + if (map == MAP_FAILED) { + ksft_print_msg("mmap failed.\n"); + ret = false; + goto close_fd; + } + for (int i = 0; i < filesize; i++) { + map[i] = 'A'; + } + map[filesize - 1] = 'X'; + } syscall_ret = syscall(__NR_cachestat, fd, &cs_range, &cs, 0); if (syscall_ret) { @@ -308,12 +329,18 @@ int main(void) break; } - if (test_cachestat_shmem()) + if (run_cachestat_test(FILE_SHMEM)) ksft_test_result_pass("cachestat works with a shmem file\n"); else { ksft_test_result_fail("cachestat fails with a shmem file\n"); ret = 1; } + if (run_cachestat_test(FILE_MMAP)) + ksft_test_result_pass("cachestat works with a mmap file\n"); + else { + ksft_test_result_fail("cachestat fails with a mmap file\n"); + ret = 1; + } return ret; } -- 2.43.0

3 months, 3 weeks

2
2
0 0

[PATCH v2 net-next] selftests: devmem: configure HDS threshold

by Taehee Yoo

The devmem TCP requires the hds-thresh value to be 0, but it doesn't change it automatically. Therefore, make configure_headersplit() sets hds-thresh value to 0. Signed-off-by: Taehee Yoo <ap420073(a)gmail.com> --- v2: - Do not implement configure_hds_thresh(). - Make configure_headersplit() sets hds-thresh to 0. tools/testing/selftests/drivers/net/hw/ncdevmem.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/drivers/net/hw/ncdevmem.c b/tools/testing/selftests/drivers/net/hw/ncdevmem.c index cc9b40d9c5d5..52b72de11e3b 100644 --- a/tools/testing/selftests/drivers/net/hw/ncdevmem.c +++ b/tools/testing/selftests/drivers/net/hw/ncdevmem.c @@ -331,6 +331,12 @@ static int configure_headersplit(bool on) ret = ethtool_rings_set(ys, req); if (ret < 0) fprintf(stderr, "YNL failed: %s\n", ys->err.msg); + if (on) { + ethtool_rings_set_req_set_hds_thresh(req, 0); + ret = ethtool_rings_set(ys, req); + if (ret < 0) + fprintf(stderr, "YNL failed: %s\n", ys->err.msg); + } ethtool_rings_set_req_free(req); if (ret == 0) { @@ -338,9 +344,12 @@ static int configure_headersplit(bool on) ethtool_rings_get_req_set_header_dev_index(get_req, ifindex); get_rsp = ethtool_rings_get(ys, get_req); ethtool_rings_get_req_free(get_req); - if (get_rsp) + if (get_rsp) { fprintf(stderr, "TCP header split: %s\n", tcp_data_split_str(get_rsp->tcp_data_split)); + fprintf(stderr, "HDS threshold: %u\n", + get_rsp->hds_thresh); + } ethtool_rings_get_rsp_free(get_rsp); } -- 2.34.1

3 months, 3 weeks

4
7
0 0

[kvm-unit-tests PATCH] riscv: lib: Add sbi-exit-code to configure and environment

by Jesse Taube

Add --[enable|disable]-sbi-exit-code to configure script. With the default value disabled. Add a check for SBI_PASS_EXIT_CODE in the environment, so that passing of the test status is configurable from both the environment and the configure script Signed-off-by: Jesse Taube <jesse(a)rivosinc.com> --- configure | 11 +++++++++++ lib/riscv/io.c | 12 +++++++++++- 2 files changed, 22 insertions(+), 1 deletion(-) diff --git a/configure b/configure index 20bf5042..7c949bdc 100755 --- a/configure +++ b/configure @@ -67,6 +67,7 @@ earlycon= console= efi= efi_direct= +sbi_exit_code=0 target_cpu= # Enable -Werror by default for git repositories only (i.e. developer builds) @@ -141,6 +142,9 @@ usage() { system and run from the UEFI shell. Ignored when efi isn't enabled and defaults to enabled when efi is enabled for riscv64. (arm64 and riscv64 only) + --[enable|disable]-sbi-exit-code + Enable or disable sending pass/fail exit code to SBI SRST. + (disabled by default, riscv only) EOF exit 1 } @@ -236,6 +240,12 @@ while [[ $optno -le $argc ]]; do --disable-efi-direct) efi_direct=n ;; + --enable-sbi-exit-code) + sbi_exit_code=1 + ;; + --disable-sbi-exit-code) + sbi_exit_code=0 + ;; --enable-werror) werror=-Werror ;; @@ -551,6 +561,7 @@ EOF elif [ "$arch" = "riscv32" ] || [ "$arch" = "riscv64" ]; then echo "#define CONFIG_UART_EARLY_BASE ${uart_early_addr}" >> lib/config.h [ "$console" = "sbi" ] && echo "#define CONFIG_SBI_CONSOLE" >> lib/config.h + echo "#define CONFIG_SBI_EXIT_CODE ${sbi_exit_code}" >> lib/config.h echo >> lib/config.h fi echo "#endif" >> lib/config.h diff --git a/lib/riscv/io.c b/lib/riscv/io.c index b1163404..0e666009 100644 --- a/lib/riscv/io.c +++ b/lib/riscv/io.c @@ -162,8 +162,18 @@ void halt(int code); void exit(int code) { + char *s = getenv("SBI_PASS_EXIT_CODE"); + bool pass_exit = CONFIG_SBI_EXIT_CODE; + printf("\nEXIT: STATUS=%d\n", ((code) << 1) | 1); - sbi_shutdown(code == 0); + + if (s) + pass_exit = (*s == '1' || *s == 'y' || *s == 'Y'); + + if (pass_exit) + sbi_shutdown(code == 0); + else + sbi_shutdown(true); halt(code); __builtin_unreachable(); } -- 2.43.0

3 months, 4 weeks

2
3
0 0

[RFC PATCH 00/33] vfio: Introduce selftests for VFIO

by David Matlack

This series introduces VFIO selftests, located in tools/testing/selftests/vfio/. VFIO selftests aim to enable kernel developers to write and run tests that take the form of userspace programs that interact with VFIO and IOMMUFD uAPIs. VFIO selftests can be used to write functional tests for new features, regression tests for bugs, and performance tests for optimizations. These tests are designed to interact with real PCI devices, i.e. they do not rely on mocking out or faking any behavior in the kernel. This allows the tests to exercise not only VFIO but also IOMMUFD, the IOMMU driver, interrupt remapping, IRQ handling, etc. We chose selftests to host these tests primarily to enable integration with the existing KVM selftests. As explained in the next section, enabling KVM developers to test the interaction between VFIO and KVM is one of the motivators of this series. Motivation ----------------------------------------------------------------------- The main motivation for this series is upcoming development in the kernel to support Hypervisor Live Updates [1][2]. Live Update is a specialized reboot process where selected devices are kept operational and their kernel state is preserved and recreated across a kexec. For devices, DMA and interrupts may continue during the reboot. VFIO-bound devices are the main target, since the first usecase of Live Updates is to enable host kernel upgrades in a Cloud Computing environment without disrupting running customer VMs. To prepare for upcoming support for Live Updates in VFIO, IOMMUFD, IOMMU drivers, the PCI layer, etc., we'd like to first lay the ground work for exercising and testing VFIO from kernel selftests. This way when we eventually upstream support for Live Updates, we can also upstream tests for those changes, rather than purely relying on Live Update integration tests which would be hard to share and reproduce upstream. But even without Live Updates, VFIO and IOMMUFD are becoming an increasingly critical component of running KVM-based VMs in cloud environments. Virtualized networking and storage are increasingly being offloaded to smart NICs/cards, and demand for high performance networking, storage, and AI are also leading to NICs, SSDs, and GPUs being directly attached to VMs via VFIO. VFIO selftests increases our ability to test in several ways. - It enables developers sending VFIO, IOMMUFD, etc. commits upstream to test their changes against all existing VFIO selftests, reducing the probability of regressions. - It enables developers sending VFIO, IOMMUFD, etc. commits upstream to include tests alongside their changes, increasing the quality of the code that is merged. - It enables testing the interaction between VFIO and KVM. There are some paths in KVM that are only exercised through VFIO, such as IRQ bypass. VFIO selftests provides a helper library to enable KVM developers to write KVM selftests to test those interactions [3]. Design ----------------------------------------------------------------------- VFIO selftests are designed around interacting with with VFIO-managed PCI devices. As such, the core data struture is struct vfio_pci_device, which represents a single PCI device. struct vfio_pci_device *device; device = vfio_pci_device_init("0000:6a:01.0", iommu_mode); ... vfio_pci_device_cleanup(device); vfio_pci_device_init() sets up a container or iommufd, depending on the iommu_mode argument, to manage DMA mappings, fetches information about the device and what interrupts it supports from VFIO and caches it, and mmap()s all mappable BARs for the test to use. There are helper methods that operate on struct vfio_pci_device to do things like read and write to PCI config space, enable/disable IRQs, and map memory for DMA, struct vfio_pci_device and its methods do not care about what device they are actually interacting with. It can be a GPU, a NIC, an SSD, etc. To keep things simple initially, VFIO selftests only support a single device per group and per container/iommufd. But it should be possible to relax those restrictions in the future, e.g. to enable testing with multiple devices in the same container/iommufd. Driver Framework ----------------------------------------------------------------------- In order to support VFIO selftests where a device is generating DMA and interrupts on command, the VFIO selftests supports a driver framework. This framework abstracts away device-specific details allowing VFIO selftests to be written in a generic way, and then run against different devices depending on what hardware developers have access to. The framework also aims to support carrying drivers out-of-tree, e.g. so that companies can run VFIO selftests with custom/test hardware. Drivers must implement the following methods: - probe(): Check if the driver supports a given device. - init(): Initialize the driver. - remove(): Deinitialize the driver and reset the device. - memcpy_start(): Kick off a series of repeated memcpys (DMA reads and DMA writes). - memcpy_wait(): Wait for a memcpy operation to complete. - send_msi(): Make the device send an MSI interrupt. memcpy_start/wait() are for generating DMA. We separate the operation into 2 steps so that tests can trigger a long-running DMA operation. We expect to use this to stress test Live Updates by kicking off a long-running mempcy operation and then performing a Live Update. These methods are required to not generate any interrupts. send_msi() is used for testing MSI and MSI-x interrupts. The driver tells the test which MSI it will be using via device->driver.msi. It's the responsibility of the test to set up a region of memory and map it into the device for use by the driver, e.g. for in-memory descriptors, before calling init(). A demo of the driver framework can be found in tools/testing/selftests/vfio/vfio_pci_driver_test.c. In addition, this series introduces a new KVM selftest to demonstrate delivering a device MSI directly into a guest, which can be found in tools/testing/selftests/kvm/vfio_pci_device_irq_test.c. Tests ----------------------------------------------------------------------- There are 5 tests in this series, mostly to demonstrate as a proof-of-concept: - tools/testing/selftests/vfio/vfio_pci_device_test.c - tools/testing/selftests/vfio/vfio_pci_driver_test.c - tools/testing/selftests/vfio/vfio_iommufd_setup_test.c - tools/testing/selftests/vfio/vfio_dma_mapping_test.c - tools/testing/selftests/kvm/vfio_pci_device_irq_test.c Integrating with KVM selftests ----------------------------------------------------------------------- To support testing the interactions between VFIO and KVM, the VFIO selftests support sharing its library with the KVM selftest. The patches at the end of this series demonstrate how that works. Essentially, we allow the KVM selftests to build their own copy of tools/testing/selftests/vfio/lib/ and link it into KVM selftests binaries. This requires minimal changes to the KVM selftests Makefile. Future Areas of Development ----------------------------------------------------------------------- Library: - Driver support for devices that can be used on AMD, ARM, and other platforms. - Driver support for a device available in QEMU VMs. - Support for tests that use multiple devices. - Support for IOMMU groups with multiple devices. - Support for multiple devices sharing the same container/iommufd. - Sharing TEST_ASSERT() macros and other common code between KVM and VFIO selftests. Tests: - DMA mapping performance tests for BARs/HugeTLB/etc. - Live Update selftests. - Porting Sean's KVM selftest for posted interrupts to use the VFIO selftests library [3] This series can also be found on GitHub: https://github.com/dmatlack/linux/tree/vfio/selftests/rfc Cc: Alex Williamson <alex.williamson(a)redhat.com> Cc: Jason Gunthorpe <jgg(a)nvidia.com> Cc: Kevin Tian <kevin.tian(a)intel.com> Cc: Paolo Bonzini <pbonzini(a)redhat.com> Cc: Sean Christopherson <seanjc(a)google.com> Cc: Vipin Sharma <vipinsh(a)google.com> Cc: Josh Hilke <jrhilke(a)google.com> Cc: Pasha Tatashin <pasha.tatashin(a)soleen.com> Cc: Saeed Mahameed <saeedm(a)nvidia.com> Cc: Saeed Mahameed <saeedm(a)nvidia.com> Cc: Adithya Jayachandran <ajayachandra(a)nvidia.com> Cc: Parav Pandit <parav(a)nvidia.com> Cc: Leon Romanovsky <leonro(a)nvidia.com> Cc: Vinicius Costa Gomes <vinicius.gomes(a)intel.com> Cc: Dave Jiang <dave.jiang(a)intel.com> Cc: Dan Williams <dan.j.williams(a)intel.com> [1] https://lore.kernel.org/all/f35359d5-63e1-8390-619f-67961443bfe1@google.com/ [2] https://lore.kernel.org/all/20250515182322.117840-1-pasha.tatashin@soleen.c… [3] https://lore.kernel.org/kvm/20250404193923.1413163-68-seanjc@google.com/ David Matlack (28): selftests: Create tools/testing/selftests/vfio vfio: selftests: Add a helper library for VFIO selftests vfio: selftests: Introduce vfio_pci_device_test tools headers: Add stub definition for __iomem tools headers: Import asm-generic MMIO helpers tools headers: Import x86 MMIO helper overrides tools headers: Import iosubmit_cmds512() tools headers: Import drivers/dma/ioat/{hw.h,registers.h} tools headers: Import drivers/dma/idxd/registers.h tools headers: Import linux/pci_ids.h vfio: selftests: Keep track of DMA regions mapped into the device vfio: selftests: Enable asserting MSI eventfds not firing vfio: selftests: Add a helper for matching vendor+device IDs vfio: selftests: Add driver framework vfio: sefltests: Add vfio_pci_driver_test vfio: selftests: Add driver for Intel CBDMA vfio: selftests: Add driver for Intel DSA vfio: selftests: Move helper to get cdev path to libvfio vfio: selftests: Encapsulate IOMMU mode vfio: selftests: Add [-i iommu_mode] option to all tests vfio: selftests: Add vfio_type1v2_mode vfio: selftests: Add iommufd_compat_type1{,v2} modes vfio: selftests: Add iommufd mode vfio: selftests: Make iommufd the default iommu_mode vfio: selftests: Add a script to help with running VFIO selftests KVM: selftests: Build and link sefltests/vfio/lib into KVM selftests KVM: selftests: Test sending a vfio-pci device IRQ to a VM KVM: selftests: Use real device MSIs in vfio_pci_device_irq_test Josh Hilke (5): vfio: selftests: Test basic VFIO and IOMMUFD integration vfio: selftests: Move vfio dma mapping test to their own file vfio: selftests: Add test to reset vfio device. vfio: selftests: Use command line to set hugepage size for DMA mapping test vfio: selftests: Validate 2M/1G HugeTLB are mapped as 2M/1G in IOMMU MAINTAINERS | 7 + tools/arch/x86/include/asm/io.h | 101 + tools/arch/x86/include/asm/special_insns.h | 27 + tools/include/asm-generic/io.h | 482 +++ tools/include/asm/io.h | 11 + tools/include/drivers/dma/idxd/registers.h | 601 +++ tools/include/drivers/dma/ioat/hw.h | 270 ++ tools/include/drivers/dma/ioat/registers.h | 251 ++ tools/include/linux/compiler.h | 4 + tools/include/linux/io.h | 4 +- tools/include/linux/pci_ids.h | 3212 +++++++++++++++++ tools/testing/selftests/Makefile | 1 + tools/testing/selftests/kvm/Makefile.kvm | 6 +- .../testing/selftests/kvm/include/kvm_util.h | 4 + tools/testing/selftests/kvm/lib/kvm_util.c | 21 + .../selftests/kvm/vfio_pci_device_irq_test.c | 173 + tools/testing/selftests/vfio/.gitignore | 7 + tools/testing/selftests/vfio/Makefile | 20 + .../testing/selftests/vfio/lib/drivers/dsa.c | 416 +++ .../testing/selftests/vfio/lib/drivers/ioat.c | 235 ++ .../selftests/vfio/lib/include/vfio_util.h | 271 ++ tools/testing/selftests/vfio/lib/libvfio.mk | 26 + .../selftests/vfio/lib/vfio_pci_device.c | 573 +++ .../selftests/vfio/lib/vfio_pci_driver.c | 126 + tools/testing/selftests/vfio/run.sh | 110 + .../selftests/vfio/vfio_dma_mapping_test.c | 239 ++ .../selftests/vfio/vfio_iommufd_setup_test.c | 133 + .../selftests/vfio/vfio_pci_device_test.c | 195 + .../selftests/vfio/vfio_pci_driver_test.c | 256 ++ 29 files changed, 7780 insertions(+), 2 deletions(-) create mode 100644 tools/arch/x86/include/asm/io.h create mode 100644 tools/arch/x86/include/asm/special_insns.h create mode 100644 tools/include/asm-generic/io.h create mode 100644 tools/include/asm/io.h create mode 100644 tools/include/drivers/dma/idxd/registers.h create mode 100644 tools/include/drivers/dma/ioat/hw.h create mode 100644 tools/include/drivers/dma/ioat/registers.h create mode 100644 tools/include/linux/pci_ids.h create mode 100644 tools/testing/selftests/kvm/vfio_pci_device_irq_test.c create mode 100644 tools/testing/selftests/vfio/.gitignore create mode 100644 tools/testing/selftests/vfio/Makefile create mode 100644 tools/testing/selftests/vfio/lib/drivers/dsa.c create mode 100644 tools/testing/selftests/vfio/lib/drivers/ioat.c create mode 100644 tools/testing/selftests/vfio/lib/include/vfio_util.h create mode 100644 tools/testing/selftests/vfio/lib/libvfio.mk create mode 100644 tools/testing/selftests/vfio/lib/vfio_pci_device.c create mode 100644 tools/testing/selftests/vfio/lib/vfio_pci_driver.c create mode 100755 tools/testing/selftests/vfio/run.sh create mode 100644 tools/testing/selftests/vfio/vfio_dma_mapping_test.c create mode 100644 tools/testing/selftests/vfio/vfio_iommufd_setup_test.c create mode 100644 tools/testing/selftests/vfio/vfio_pci_device_test.c create mode 100644 tools/testing/selftests/vfio/vfio_pci_driver_test.c base-commit: a11a72229881d8ac1d52ea727101bc9c744189c1 prerequisite-patch-id: 3bae97c9e1093148763235f47a84fa040b512d04 -- 2.49.0.1151.ga128411c76-goog

3 months, 4 weeks

10
63
0 0

[PATCH bpf-next v3 0/2] bpf: Fix and test aux usage after do_check_insn()

by Luis Gerhorst

Fix cur_aux()->nospec_result test after do_check_insn() referring to the to-be-analyzed (potentially unsafe) instruction, not the already-analyzed (safe) instruction. This might allow a unsafe insn to slip through on a speculative path. Create some tests from the reproducer [1]. Commit d6f1c85f2253 ("bpf: Fall back to nospec for Spectre v1") should not be in any stable kernel yet, therefore bpf-next should suffice. [1] https://lore.kernel.org/bpf/685b3c1b.050a0220.2303ee.0010.GAE@google.com/ Changes since v2: - Use insn_aux variable instead of introducing prev_aux() as suggested by Eduard (and therefore also drop patch 1) - v2: https://lore.kernel.org/bpf/20250628145016.784256-1-luis.gerhorst@fau.de/ Changes since v1: - Fix compiler error due to missed rename of prev_insn_idx in first patch - v1: https://lore.kernel.org/bpf/20250628125927.763088-1-luis.gerhorst@fau.de/ Changes since RFC: - Introduce prev_aux() as suggested by Alexei. For this, we must move the env->prev_insn_idx assignment to happen directly after do_check_insn(), for which I have created a separate commit. This patch could be simplified by using a local prev_aux variable as sugested by Eduard, but I figured one might find the new assignment-strategy easier to understand (before, prev_insn_idx and env->prev_insn_idx were out-of-sync for the latter part of the loop). Also, like this we do not have an additional prev_* variable that must be kept in-sync and the local variable's usage (old prev_insn_idx, new tmp) is much more local. If you think it would be better to not take the risk and keep the fix simple by just introducing the prev_aux variable, let me know. - Change WARN_ON_ONCE() to verifier_bug_if() as suggested by Alexei - Change assertion to check instruction is BPF_JMP[32] as suggested by Eduard - RFC: https://lore.kernel.org/bpf/8734bmoemx.fsf@fau.de/ Luis Gerhorst (2): bpf: Fix aux usage after do_check_insn() selftests/bpf: Add Spectre v4 tests kernel/bpf/verifier.c | 19 ++- tools/testing/selftests/bpf/progs/bpf_misc.h | 4 + .../selftests/bpf/progs/verifier_unpriv.c | 149 ++++++++++++++++++ 3 files changed, 167 insertions(+), 5 deletions(-) base-commit: 03fe01ddd1d8be7799419ea5e5f228a0186ae8c2 -- 2.49.0

3 months, 4 weeks

3
4
0 0

[PATCH] selftests/bpf: Set CONFIG_PACKET=y for selftests

by Saket Kumar Bhaskar

BPF selftest fails to build with below error: CLNG-BPF [test_progs] lsm_cgroup.bpf.o progs/lsm_cgroup.c:105:21: error: variable has incomplete type 'struct sockaddr_ll' 105 | struct sockaddr_ll sa = {}; | ^ progs/lsm_cgroup.c:105:9: note: forward declaration of 'struct sockaddr_ll' 105 | struct sockaddr_ll sa = {}; | ^ 1 error generated. lsm_cgroup selftest requires sockaddr_ll structure which is not there in vmlinux.h when the kernel is built with CONFIG_PACKET=m. Enabling CONFIG_PACKET=y ensures that sockaddr_ll is available in vmlinux, allowing it to be captured in the generated vmlinux.h for bpf selftests. Reported-by: Sachin P Bappalige <sachinpb(a)linux.ibm.com> Signed-off-by: Saket Kumar Bhaskar <skb99(a)linux.ibm.com> --- tools/testing/selftests/bpf/config | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/bpf/config b/tools/testing/selftests/bpf/config index f74e1ea0ad3b..7247833fe623 100644 --- a/tools/testing/selftests/bpf/config +++ b/tools/testing/selftests/bpf/config @@ -105,6 +105,7 @@ CONFIG_IP_NF_IPTABLES=y CONFIG_IP6_NF_IPTABLES=y CONFIG_IP6_NF_FILTER=y CONFIG_NF_NAT=y +CONFIG_PACKET=y CONFIG_RC_CORE=y CONFIG_SECURITY=y CONFIG_SECURITYFS=y -- 2.43.5

3 months, 4 weeks

3
2
0 0

[RFC v2] tools/nolibc: add sigaction()

by Benjamin Berg

From: Benjamin Berg <benjamin.berg(a)intel.com> In preparation to add tests that use it. Note that some architectures do not have a usable linux/signal.h include file. However, in those cases we can use asm-generic/signal.h instead. Signed-off-by: Benjamin Berg <benjamin.berg(a)intel.com> --- Another attempt at signal handling for nolibc which should actually be working. Some trickery is needed to get the right definition, but I feel it is sufficiently clean this way. Submitting this as RFC mostly because I do not yet have a proper patch to add a test that uses the feature. Benjamin --- tools/include/nolibc/arch-aarch64.h | 3 + tools/include/nolibc/arch-arm.h | 7 ++ tools/include/nolibc/arch-i386.h | 13 +++ tools/include/nolibc/arch-loongarch.h | 3 + tools/include/nolibc/arch-m68k.h | 10 ++ tools/include/nolibc/arch-mips.h | 3 + tools/include/nolibc/arch-powerpc.h | 8 ++ tools/include/nolibc/arch-riscv.h | 3 + tools/include/nolibc/arch-s390.h | 8 +- tools/include/nolibc/arch-sparc.h | 43 +++++++++ tools/include/nolibc/arch-x86_64.h | 10 ++ tools/include/nolibc/signal.h | 97 ++++++++++++++++++++ tools/include/nolibc/sys.h | 2 +- tools/include/nolibc/time.h | 3 +- tools/testing/selftests/nolibc/nolibc-test.c | 52 +++++++++++ 15 files changed, 261 insertions(+), 4 deletions(-) diff --git a/tools/include/nolibc/arch-aarch64.h b/tools/include/nolibc/arch-aarch64.h index 937a348da42e..736aae6dbd47 100644 --- a/tools/include/nolibc/arch-aarch64.h +++ b/tools/include/nolibc/arch-aarch64.h @@ -10,6 +10,9 @@ #include "compiler.h" #include "crt.h" +/* Architecture has a usable linux/signal.h */ +#include <linux/signal.h> + /* Syscalls for AARCH64 : * - registers are 64-bit * - stack is 16-byte aligned diff --git a/tools/include/nolibc/arch-arm.h b/tools/include/nolibc/arch-arm.h index 1f66e7e5a444..1faf6c2dbeb8 100644 --- a/tools/include/nolibc/arch-arm.h +++ b/tools/include/nolibc/arch-arm.h @@ -10,6 +10,13 @@ #include "compiler.h" #include "crt.h" +/* Needed to get the correct struct sigaction definition */ +#define SA_RESTORER 0x04000000 + +/* Avoid linux/signal.h, it has an incorrect _NSIG and sigset_t */ +#include <asm-generic/signal.h> +#include <asm-generic/siginfo.h> + /* Syscalls for ARM in ARM or Thumb modes : * - registers are 32-bit * - stack is 8-byte aligned diff --git a/tools/include/nolibc/arch-i386.h b/tools/include/nolibc/arch-i386.h index 7c9b38e96418..fbec7490a92c 100644 --- a/tools/include/nolibc/arch-i386.h +++ b/tools/include/nolibc/arch-i386.h @@ -10,6 +10,19 @@ #include "compiler.h" #include "crt.h" +/* Needed to get the correct struct sigaction definition */ +#define SA_RESTORER 0x04000000 + +/* Restorer must be set on i386 */ +#define _NOLIBC_ARCH_NEEDS_SA_RESTORER + +/* Otherwise we would need to use sigreturn instead of rt_sigreturn */ +#define _NOLIBC_ARCH_FORCE_SIG_FLAGS SA_SIGINFO + +/* Avoid linux/signal.h, it has an incorrect _NSIG and sigset_t */ +#include <asm-generic/signal.h> +#include <asm-generic/siginfo.h> + /* Syscalls for i386 : * - mostly similar to x86_64 * - registers are 32-bit diff --git a/tools/include/nolibc/arch-loongarch.h b/tools/include/nolibc/arch-loongarch.h index 5511705303ea..68d60d04ef59 100644 --- a/tools/include/nolibc/arch-loongarch.h +++ b/tools/include/nolibc/arch-loongarch.h @@ -10,6 +10,9 @@ #include "compiler.h" #include "crt.h" +/* Architecture has a usable linux/signal.h */ +#include <linux/signal.h> + /* Syscalls for LoongArch : * - stack is 16-byte aligned * - syscall number is passed in a7 diff --git a/tools/include/nolibc/arch-m68k.h b/tools/include/nolibc/arch-m68k.h index 6dac1845f298..981b4cc55a69 100644 --- a/tools/include/nolibc/arch-m68k.h +++ b/tools/include/nolibc/arch-m68k.h @@ -13,6 +13,16 @@ #include "compiler.h" #include "crt.h" +/* + * Needed to get the correct struct sigaction definition. m68k does not use + * sa_restorer, but it is included in the structure. + */ +#define SA_RESTORER 0x04000000 + +/* Avoid linux/signal.h, it has an incorrect _NSIG and sigset_t */ +#include <asm-generic/signal.h> +#include <asm-generic/siginfo.h> + #define _NOLIBC_SYSCALL_CLOBBERLIST "memory" #define my_syscall0(num) \ diff --git a/tools/include/nolibc/arch-mips.h b/tools/include/nolibc/arch-mips.h index 753a8ed2cf69..a8837452e744 100644 --- a/tools/include/nolibc/arch-mips.h +++ b/tools/include/nolibc/arch-mips.h @@ -14,6 +14,9 @@ #error Unsupported MIPS ABI #endif +/* Architecture has a usable linux/signal.h */ +#include <linux/signal.h> + /* Syscalls for MIPS ABI O32 : * - WARNING! there's always a delayed slot! * - WARNING again, the syntax is different, registers take a '$' and numbers diff --git a/tools/include/nolibc/arch-powerpc.h b/tools/include/nolibc/arch-powerpc.h index 204564bbcd32..c846a7ddcf3c 100644 --- a/tools/include/nolibc/arch-powerpc.h +++ b/tools/include/nolibc/arch-powerpc.h @@ -10,6 +10,14 @@ #include "compiler.h" #include "crt.h" +/* Needed to get the correct struct sigaction definition */ +#define SA_RESTORER 0x04000000 +#define _NOLIBC_ARCH_NEEDS_SA_RESTORER + +/* Avoid linux/signal.h, it has an incorrect _NSIG and sigset_t */ +#include <asm-generic/signal.h> +#include <asm-generic/siginfo.h> + /* Syscalls for PowerPC : * - stack is 16-byte aligned * - syscall number is passed in r0 diff --git a/tools/include/nolibc/arch-riscv.h b/tools/include/nolibc/arch-riscv.h index 885383a86c38..709e6a262d9a 100644 --- a/tools/include/nolibc/arch-riscv.h +++ b/tools/include/nolibc/arch-riscv.h @@ -10,6 +10,9 @@ #include "compiler.h" #include "crt.h" +/* Architecture has a usable linux/signal.h */ +#include <linux/signal.h> + /* Syscalls for RISCV : * - stack is 16-byte aligned * - syscall number is passed in a7 diff --git a/tools/include/nolibc/arch-s390.h b/tools/include/nolibc/arch-s390.h index df4c3cc713ac..0dccb6d1ad64 100644 --- a/tools/include/nolibc/arch-s390.h +++ b/tools/include/nolibc/arch-s390.h @@ -5,13 +5,19 @@ #ifndef _NOLIBC_ARCH_S390_H #define _NOLIBC_ARCH_S390_H -#include <linux/signal.h> #include <linux/unistd.h> #include "compiler.h" #include "crt.h" #include "std.h" +/* Needed to get the correct struct sigaction definition */ +#define SA_RESTORER 0x04000000 + +/* Avoid linux/signal.h, it has an incorrect _NSIG and sigset_t */ +#include <asm-generic/signal.h> +#include <asm-generic/siginfo.h> + /* Syscalls for s390: * - registers are 64-bit * - syscall number is passed in r1 diff --git a/tools/include/nolibc/arch-sparc.h b/tools/include/nolibc/arch-sparc.h index 1435172f3dfe..303291d4b8fb 100644 --- a/tools/include/nolibc/arch-sparc.h +++ b/tools/include/nolibc/arch-sparc.h @@ -12,6 +12,19 @@ #include "compiler.h" #include "crt.h" +/* Otherwise we would need to use sigreturn instead of rt_sigreturn */ +#define _NOLIBC_ARCH_FORCE_SIG_FLAGS SA_SIGINFO + +/* The includes are sane, if one sets __WANT_POSIX1B_SIGNALS__ */ +#define __WANT_POSIX1B_SIGNALS__ +#include <linux/signal.h> + +/* + * sparc has ODD_RT_SIGACTION, we always pass our restorer as an argument + * to rt_sigaction. The restorer is implemented in this file. + */ +#define _NOLIBC_RT_SIGACTION_PASSES_RESTORER + /* * Syscalls for SPARC: * - registers are native word size @@ -188,4 +201,34 @@ pid_t sys_fork(void) } #define sys_fork sys_fork +#define __nolibc_stringify_1(x...) #x +#define __nolibc_stringify(x...) __stringify_1(x) + +/* The compiler insists on adding a SAVE call to the start of every function */ +#define __nolibc_sa_restorer __nolibc_sa_restorer +void __nolibc_sa_restorer (void); +#ifdef __arch64__ +__asm__( \ + ".section .text\n" \ + ".align 4 \n" \ + "__nolibc_sa_restorer:\n" \ + "nop\n" \ + "nop\n" \ + "mov " __stringify(__NR_rt_sigreturn) ", %g1 \n" \ + "t 0x6d \n"); +#else +__asm__( \ + ".section .text\n" \ + ".align 4 \n" \ + "__nolibc_sa_restorer:\n" \ + "nop\n" \ + "nop\n" \ + "mov " __stringify(__NR_rt_sigreturn) ", %g1 \n" \ + "t 0x10 \n" \ + ); +#endif + +#undef __nolibc_stringify_1(x...) +#undef __nolibc_stringify + #endif /* _NOLIBC_ARCH_SPARC_H */ diff --git a/tools/include/nolibc/arch-x86_64.h b/tools/include/nolibc/arch-x86_64.h index 67305e24dbef..9f13a2205876 100644 --- a/tools/include/nolibc/arch-x86_64.h +++ b/tools/include/nolibc/arch-x86_64.h @@ -10,6 +10,16 @@ #include "compiler.h" #include "crt.h" +/* Needed to get the correct struct sigaction definition */ +#define SA_RESTORER 0x04000000 + +/* Restorer must be set on i386 */ +#define _NOLIBC_ARCH_NEEDS_SA_RESTORER + +/* Avoid linux/signal.h, it has an incorrect _NSIG and sigset_t */ +#include <asm-generic/signal.h> +#include <asm-generic/siginfo.h> + /* Syscalls for x86_64 : * - registers are 64-bit * - syscall number is passed in rax diff --git a/tools/include/nolibc/signal.h b/tools/include/nolibc/signal.h index ac13e53ac31d..829672250ede 100644 --- a/tools/include/nolibc/signal.h +++ b/tools/include/nolibc/signal.h @@ -14,6 +14,8 @@ #include "arch.h" #include "types.h" #include "sys.h" +#include "string.h" +/* signal definitions are included by arch.h */ /* This one is not marked static as it's needed by libgcc for divide by zero */ int raise(int signal); @@ -23,4 +25,99 @@ int raise(int signal) return sys_kill(sys_getpid(), signal); } +/* + * sigaction(int signum, const struct sigaction *act, struct sigaction *oldact) + */ +#if defined(_NOLIBC_ARCH_NEEDS_SA_RESTORER) && !defined(__nolibc_sa_restorer) +static __attribute__((noreturn)) __nolibc_entrypoint __no_stack_protector +void __nolibc_sa_restorer(void) +{ + my_syscall0(__NR_rt_sigreturn); + __nolibc_entrypoint_epilogue(); +} +#endif + +static __attribute__((unused)) +int sys_rt_sigaction(int signum, const struct sigaction *act, struct sigaction *oldact) +{ + struct sigaction real_act = *act; +#if defined(SA_RESTORER) && defined(_NOLIBC_ARCH_NEEDS_SA_RESTORER) + if (!(real_act.sa_flags & SA_RESTORER)) { + real_act.sa_flags |= SA_RESTORER; + real_act.sa_restorer = __nolibc_sa_restorer; + } +#endif +#ifdef _NOLIBC_ARCH_FORCE_SIG_FLAGS + real_act.sa_flags |= _NOLIBC_ARCH_FORCE_SIG_FLAGS; +#endif + +#ifndef _NOLIBC_RT_SIGACTION_PASSES_RESTORER + return my_syscall4(__NR_rt_sigaction, signum, &real_act, oldact, + sizeof(act->sa_mask)); +#else + return my_syscall5(__NR_rt_sigaction, signum, &real_act, oldact, + __nolibc_sa_restorer, sizeof(act->sa_mask)); +#endif +} + +static __attribute__((unused)) +int sigaction(int signum, const struct sigaction *act, struct sigaction *oldact) +{ + return __sysret(sys_rt_sigaction(signum, act, oldact)); +} + +/* + * int sigemptyset(sigset_t *set) + */ +static __attribute__((unused)) +int sigemptyset(sigset_t *set) +{ + memset(set, 0, sizeof(*set)); + return 0; +} + +/* + * int sigfillset(sigset_t *set) + */ +static __attribute__((unused)) +int sigfillset(sigset_t *set) +{ + memset(set, 0xff, sizeof(*set)); + return 0; +} + +/* + * int sigaddset(sigset_t *set, int signum) + */ +static __attribute__((unused)) +int sigaddset(sigset_t *set, int signum) +{ + set->sig[(signum - 1) / (8 * sizeof(set->sig[0]))] |= + 1UL << ((signum - 1) % (8 * sizeof(set->sig[0]))); + return 0; +} + +/* + * int sigdelset(sigset_t *set, int signum) + */ +static __attribute__((unused)) +int sigdelset(sigset_t *set, int signum) +{ + set->sig[(signum - 1) / (8 * sizeof(set->sig[0]))] &= + ~(1UL << ((signum - 1) % (8 * sizeof(set->sig[0])))); + return 0; +} + +/* + * int sigismember(sigset_t *set, int signum) + */ +static __attribute__((unused)) +int sigismember(sigset_t *set, int signum) +{ + unsigned long res = + set->sig[(signum - 1) / (8 * sizeof(set->sig[0]))] & + (1UL << ((signum - 1) % (8 * sizeof(set->sig[0])))); + return !!res; +} + #endif /* _NOLIBC_SIGNAL_H */ diff --git a/tools/include/nolibc/sys.h b/tools/include/nolibc/sys.h index 9556c69a6ae1..fb9a1ccf2440 100644 --- a/tools/include/nolibc/sys.h +++ b/tools/include/nolibc/sys.h @@ -14,7 +14,6 @@ /* system includes */ #include <linux/unistd.h> -#include <linux/signal.h> /* for SIGCHLD */ #include <linux/termios.h> #include <linux/mman.h> #include <linux/fs.h> @@ -23,6 +22,7 @@ #include <linux/auxvec.h> #include <linux/fcntl.h> /* for O_* and AT_* */ #include <linux/stat.h> /* for statx() */ +/* signal definitions are included by arch.h */ #include "errno.h" #include "stdarg.h" diff --git a/tools/include/nolibc/time.h b/tools/include/nolibc/time.h index fc387940d51f..70ef31d4117f 100644 --- a/tools/include/nolibc/time.h +++ b/tools/include/nolibc/time.h @@ -14,9 +14,8 @@ #include "arch.h" #include "types.h" #include "sys.h" - -#include <linux/signal.h> #include <linux/time.h> +/* signal definitions are included by arch.h */ static __inline__ void __nolibc_timespec_user_to_kernel(const struct timespec *ts, struct __kernel_timespec *kts) diff --git a/tools/testing/selftests/nolibc/nolibc-test.c b/tools/testing/selftests/nolibc/nolibc-test.c index dbe13000fb1a..af66b739ea18 100644 --- a/tools/testing/selftests/nolibc/nolibc-test.c +++ b/tools/testing/selftests/nolibc/nolibc-test.c @@ -1750,6 +1750,57 @@ static int run_protection(int min __attribute__((unused)), } } +volatile int signal_check; + +void test_sighandler(int signum) +{ + if (signum == SIGUSR1) { + kill(getpid(), SIGUSR2); + signal_check = 1; + } else { + signal_check++; + } +} + +int run_signal(int min, int max) +{ + struct sigaction sa = { + .sa_flags = 0, + .sa_handler = test_sighandler, + }; + int llen; /* line length */ + int ret = 0; + int res; + + (void)min; + (void)max; + + signal_check = 0; + + sigemptyset(&sa.sa_mask); + sigaddset(&sa.sa_mask, SIGUSR2); + + res = sigaction(SIGUSR1, &sa, NULL); + llen = printf("register SIGUSR1: %d", res); + EXPECT_EQ(1, 0, res); + res = sigaction(SIGUSR2, &sa, NULL); + llen = printf("register SIGUSR2: %d", res); + EXPECT_EQ(1, 0, res); + + /* Trigger the first signal. */ + kill(getpid(), SIGUSR1); + + /* If signal_check is 1 or higher, then signal emission worked */ + llen = printf("signal emission: 1 <= signal_check"); + EXPECT_GE(1, signal_check, 1); + + /* If it is 2, then signal masking worked */ + llen = printf("signal masking: 2 == signal_check"); + EXPECT_EQ(1, signal_check, 2); + + return ret; +} + /* prepare what needs to be prepared for pid 1 (stdio, /dev, /proc, etc) */ int prepare(void) { @@ -1815,6 +1866,7 @@ static const struct test test_names[] = { { .name = "stdlib", .func = run_stdlib }, { .name = "printf", .func = run_printf }, { .name = "protection", .func = run_protection }, + { .name = "signal", .func = run_signal }, { 0 } }; -- 2.50.0

3 months, 4 weeks

3
5
0 0

[PATCH v14 3/3] selftests/rseq: Add test for mm_cid compaction

by Gabriele Monaco

A task in the kernel (task_mm_cid_work) runs somewhat periodically to compact the mm_cid for each process. Add a test to validate that it runs correctly and timely. The test spawns 1 thread pinned to each CPU, then each thread, including the main one, runs in short bursts for some time. During this period, the mm_cids should be spanning all numbers between 0 and nproc. At the end of this phase, a thread with high enough mm_cid (>= nproc/2) is selected to be the new leader, all other threads terminate. After some time, the only running thread should see 0 as mm_cid, if that doesn't happen, the compaction mechanism didn't work and the test fails. The test never fails if only 1 core is available, in which case, we cannot test anything as the only available mm_cid is 0. Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers(a)efficios.com> Acked-by: Shuah Khan <skhan(a)linuxfoundation.org> Signed-off-by: Gabriele Monaco <gmonaco(a)redhat.com> --- tools/testing/selftests/rseq/.gitignore | 1 + tools/testing/selftests/rseq/Makefile | 2 +- .../selftests/rseq/mm_cid_compaction_test.c | 200 ++++++++++++++++++ 3 files changed, 202 insertions(+), 1 deletion(-) create mode 100644 tools/testing/selftests/rseq/mm_cid_compaction_test.c diff --git a/tools/testing/selftests/rseq/.gitignore b/tools/testing/selftests/rseq/.gitignore index 0fda241fa62b0..b3920c59bf401 100644 --- a/tools/testing/selftests/rseq/.gitignore +++ b/tools/testing/selftests/rseq/.gitignore @@ -3,6 +3,7 @@ basic_percpu_ops_test basic_percpu_ops_mm_cid_test basic_test basic_rseq_op_test +mm_cid_compaction_test param_test param_test_benchmark param_test_compare_twice diff --git a/tools/testing/selftests/rseq/Makefile b/tools/testing/selftests/rseq/Makefile index 0d0a5fae59547..bc4d940f66d40 100644 --- a/tools/testing/selftests/rseq/Makefile +++ b/tools/testing/selftests/rseq/Makefile @@ -17,7 +17,7 @@ OVERRIDE_TARGETS = 1 TEST_GEN_PROGS = basic_test basic_percpu_ops_test basic_percpu_ops_mm_cid_test param_test \ param_test_benchmark param_test_compare_twice param_test_mm_cid \ param_test_mm_cid_benchmark param_test_mm_cid_compare_twice \ - syscall_errors_test + syscall_errors_test mm_cid_compaction_test TEST_GEN_PROGS_EXTENDED = librseq.so diff --git a/tools/testing/selftests/rseq/mm_cid_compaction_test.c b/tools/testing/selftests/rseq/mm_cid_compaction_test.c new file mode 100644 index 0000000000000..7ddde3b657dd6 --- /dev/null +++ b/tools/testing/selftests/rseq/mm_cid_compaction_test.c @@ -0,0 +1,200 @@ +// SPDX-License-Identifier: LGPL-2.1 +#define _GNU_SOURCE +#include <assert.h> +#include <pthread.h> +#include <sched.h> +#include <stdint.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <stddef.h> + +#include "../kselftest.h" +#include "rseq.h" + +#define VERBOSE 0 +#define printf_verbose(fmt, ...) \ + do { \ + if (VERBOSE) \ + printf(fmt, ##__VA_ARGS__); \ + } while (0) + +/* 0.5 s */ +#define RUNNER_PERIOD 500000 +/* Number of runs before we terminate or get the token */ +#define THREAD_RUNS 5 + +/* + * Number of times we check that the mm_cid were compacted. + * Checks are repeated every RUNNER_PERIOD. + */ +#define MM_CID_COMPACT_TIMEOUT 10 + +struct thread_args { + int cpu; + int num_cpus; + pthread_mutex_t *token; + pthread_barrier_t *barrier; + pthread_t *tinfo; + struct thread_args *args_head; +}; + +static void __noreturn *thread_runner(void *arg) +{ + struct thread_args *args = arg; + int i, ret, curr_mm_cid; + cpu_set_t cpumask; + + CPU_ZERO(&cpumask); + CPU_SET(args->cpu, &cpumask); + ret = pthread_setaffinity_np(pthread_self(), sizeof(cpumask), &cpumask); + if (ret) { + errno = ret; + perror("Error: failed to set affinity"); + abort(); + } + pthread_barrier_wait(args->barrier); + + for (i = 0; i < THREAD_RUNS; i++) + usleep(RUNNER_PERIOD); + curr_mm_cid = rseq_current_mm_cid(); + /* + * We select one thread with high enough mm_cid to be the new leader. + * All other threads (including the main thread) will terminate. + * After some time, the mm_cid of the only remaining thread should + * converge to 0, if not, the test fails. + */ + if (curr_mm_cid >= args->num_cpus / 2 && + !pthread_mutex_trylock(args->token)) { + printf_verbose( + "cpu%d has mm_cid=%d and will be the new leader.\n", + sched_getcpu(), curr_mm_cid); + for (i = 0; i < args->num_cpus; i++) { + if (args->tinfo[i] == pthread_self()) + continue; + ret = pthread_join(args->tinfo[i], NULL); + if (ret) { + errno = ret; + perror("Error: failed to join thread"); + abort(); + } + } + pthread_barrier_destroy(args->barrier); + free(args->tinfo); + free(args->token); + free(args->barrier); + free(args->args_head); + + for (i = 0; i < MM_CID_COMPACT_TIMEOUT; i++) { + curr_mm_cid = rseq_current_mm_cid(); + printf_verbose("run %d: mm_cid=%d on cpu%d.\n", i, + curr_mm_cid, sched_getcpu()); + if (curr_mm_cid == 0) + exit(EXIT_SUCCESS); + usleep(RUNNER_PERIOD); + } + exit(EXIT_FAILURE); + } + printf_verbose("cpu%d has mm_cid=%d and is going to terminate.\n", + sched_getcpu(), curr_mm_cid); + pthread_exit(NULL); +} + +int test_mm_cid_compaction(void) +{ + cpu_set_t affinity; + int i, j, ret = 0, num_threads; + pthread_t *tinfo; + pthread_mutex_t *token; + pthread_barrier_t *barrier; + struct thread_args *args; + + sched_getaffinity(0, sizeof(affinity), &affinity); + num_threads = CPU_COUNT(&affinity); + tinfo = calloc(num_threads, sizeof(*tinfo)); + if (!tinfo) { + perror("Error: failed to allocate tinfo"); + return -1; + } + args = calloc(num_threads, sizeof(*args)); + if (!args) { + perror("Error: failed to allocate args"); + ret = -1; + goto out_free_tinfo; + } + token = malloc(sizeof(*token)); + if (!token) { + perror("Error: failed to allocate token"); + ret = -1; + goto out_free_args; + } + barrier = malloc(sizeof(*barrier)); + if (!barrier) { + perror("Error: failed to allocate barrier"); + ret = -1; + goto out_free_token; + } + if (num_threads == 1) { + fprintf(stderr, "Cannot test on a single cpu. " + "Skipping mm_cid_compaction test.\n"); + /* only skipping the test, this is not a failure */ + goto out_free_barrier; + } + pthread_mutex_init(token, NULL); + ret = pthread_barrier_init(barrier, NULL, num_threads); + if (ret) { + errno = ret; + perror("Error: failed to initialise barrier"); + goto out_free_barrier; + } + for (i = 0, j = 0; i < CPU_SETSIZE && j < num_threads; i++) { + if (!CPU_ISSET(i, &affinity)) + continue; + args[j].num_cpus = num_threads; + args[j].tinfo = tinfo; + args[j].token = token; + args[j].barrier = barrier; + args[j].cpu = i; + args[j].args_head = args; + if (!j) { + /* The first thread is the main one */ + tinfo[0] = pthread_self(); + ++j; + continue; + } + ret = pthread_create(&tinfo[j], NULL, thread_runner, &args[j]); + if (ret) { + errno = ret; + perror("Error: failed to create thread"); + abort(); + } + ++j; + } + printf_verbose("Started %d threads.\n", num_threads); + + /* Also main thread will terminate if it is not selected as leader */ + thread_runner(&args[0]); + + /* only reached in case of errors */ +out_free_barrier: + free(barrier); +out_free_token: + free(token); +out_free_args: + free(args); +out_free_tinfo: + free(tinfo); + + return ret; +} + +int main(int argc, char **argv) +{ + if (!rseq_mm_cid_available()) { + fprintf(stderr, "Error: rseq_mm_cid unavailable\n"); + return -1; + } + if (test_mm_cid_compaction()) + return -1; + return 0; +} -- 2.50.0

3 months, 4 weeks

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-kselftest-mirror