- 07 May, 2019 7 commits
-
-
Martin Schwidefsky authored
Use trap4 as the guard instruction for the restartable sequence abort handler. Signed-off-by:
Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by:
Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Signed-off-by:
Shuah Khan <skhan@linuxfoundation.org>
-
Mathieu Desnoyers authored
Use ud1 as the guard instruction for the restartable sequence abort handler. Its benefit compared to nopl is to trap execution if the program ends up trying to execute it by mistake, which makes debugging easier. The 4-byte signature per se is unchanged (it is the instruction operand). Only the opcode is changed from nopl to ud1. Signed-off-by:
Peter Zijlstra <peterz@infradead.org> CC: Peter Zijlstra <peterz@infradead.org> CC: Thomas Gleixner <tglx@linutronix.de> CC: Joel Fernandes <joelaf@google.com> CC: Catalin Marinas <catalin.marinas@arm.com> CC: Dave Watson <davejwatson@fb.com> CC: Will Deacon <will.deacon@arm.com> CC: Shuah Khan <shuah@kernel.org> CC: Andi Kleen <andi@firstfloor.org> CC: linux-kselftest@vger.kernel.org CC: "H . Peter Anvin" <hpa@zytor.com> CC: Chris Lameter <cl@linux.com> CC: Russell King <linux@arm.linux.org.uk> CC: Michael Kerrisk <mtk.manpages@gmail.com> CC: "Paul E . McKenney" <paulmck@linux.vnet.ibm.com> CC: Paul Turner <pjt@google.com> CC: Boqun Feng <boqun.feng@gmail.com> CC: Josh Triplett <josh@joshtriplett.org> CC: Steven Rostedt <rostedt@goodmis.org> CC: Ben Maurer <bmaurer@fb.com> CC: linux-api@vger.kernel.org CC: Andy Lutomirski <luto@amacapital.net> CC: Andrew Morton <akpm@linux-foundation.org> CC: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
-
Mathieu Desnoyers authored
The branch target range of the "j" instruction is 64K, which is not enough for the general case. Suggested-by:
-
Mathieu Desnoyers authored
In order to integrate rseq into user-space applications, expose a __rseq_handled symbol so many rseq users can be linked into the same application (e.g. librseq and glibc). The __rseq_refcount TLS variable is static to the librseq library. It ensures that rseq syscall registration/unregistration happens only for the most early/late caller to rseq_{,un}register_current_thread for each thread, thus ensuring that rseq is registered across the lifetime of all rseq users for a given thread. Signed-off-by: -
Mathieu Desnoyers authored
The entries within __rseq_table are aligned on 32 bytes due to linux/rseq.h struct rseq_cs uapi requirements, but the start of the __rseq_table section is not guaranteed to be 32-byte aligned. It can cause padding to be added at the start of the section, which makes it hard to use as an array of items by debuggers. Considering that __rseq_table does not really consist of a table due to the presence of padding, rename this section to __rseq_cs. Create a new __rseq_cs_ptr_array section which contains 64-bit packed pointers to entries within the __rseq_cs section. Signed-off-by:
-
Mathieu Desnoyers authored
Knowing all exit points is useful to assist debuggers stepping over the rseq critical sections without requiring them to disassemble the content of the critical section to figure out the exit points. Signed-off-by:
-
Mathieu Desnoyers authored
gcc-8 version 8.1.0, 8.2.0, and 8.3.0 generate broken assembler with asm goto that have a thread-local storage "m" input operand on both x86-32 and x86-64. For instance: __thread int var; static int fct(void) { asm goto ( "jmp %l[testlabel]\n\t" : : [var] "m" (var) : : testlabel); return 0; testlabel: return 1; } int main() { return fct(); } % gcc-8 -O2 -o test-asm-goto test-asm-goto.c /tmp/ccAdHJbe.o: In function `main': test-asm-goto.c:(.text.startup+0x1): undefined reference to `.L2' collect2: error: ld returned 1 exit status % gcc-8 -m32 -O2 -o test-asm-goto test-asm-goto.c /tmp/ccREsVXA.o: In function `main': test-asm-goto.c:(.text.startup+0x1): undefined reference to `.L2' collect2: error: ld returned 1 exit status Work-around this compiler bug in the rseq-x86.h header by passing the address of the __rseq_abi TLS as a register operand rather than using the "m" input operand. Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90193Signed-off-by:
-
- 25 Apr, 2019 8 commits
-
-
Kees Cook authored
The test plan for TAP needs to be declared immediately after the header. This adds the test plan API to kselftest.h and updates all callers to declare their expected test counts. Signed-off-by:
Kees Cook <keescook@chromium.org> Signed-off-by:
-
Kees Cook authored
Since sub-testing can now be detected by indentation level, this removes KSFT_TAP_LEVEL so that subtests report their TAP header for later parsing. Signed-off-by:
-
Kees Cook authored
This changes the selftest output so that each test's output is prefixed with "# " as a TAP "diagnostic line". This creates a bit of a kernel-specific TAP dialect where the diagnostics precede the results. The TAP spec isn't entirely clear about this, though, so I think it's the correct solution so as to keep interactive runs making sense. If the output _followed_ the result line in the spec-suggested YAML form, each test would dump all of its output at once instead of as it went, making debugging harder. This does, however, solve the recursive TAP output problem, as sub-tests will simply be prefixed by "# ". Parsing sub-tests becomes a simple problem of just removing the first two characters of a given top-level test's diagnostic output, and parsing the results. Note that the shell construct needed to both get an exit code from the first command in a pipe and still filter the pipe (to add the "# " prefix) uses a POSIX solution rather than the bash "pipefail" option which is not supported by dash. Since some test environments may have a very minimal set of utilities available, the new prefixing code will fall back to doing line-at-a-time prefixing if perl and/or stdbuf are not available. Signed-off-by:
-
Kees Cook authored
If a test was missing (e.g. wrong architecture, etc), the test runner would incorrectly claim the test was non-executable. This adds an existence check to report correctly. Signed-off-by:
-
Kees Cook authored
The TAP version 13 spec requires a "plan" line, which has been missing. Since we always know how many tests we're going to run, emit the count on the plan line. This also fixes the result lines to remove the "1.." prefix which is against spec, and to mark skips with the correct "# SKIP" suffix. Signed-off-by:
-
Kees Cook authored
This moves the logic for running multiple tests into a single "run_many" function of runner.sh. Both "run_tests" and "emit_tests" are modified to use it. Summary handling is now controlled by the "per_test_logging" shell flag. Signed-off-by:
-
Kees Cook authored
This reuses the new runner.sh for the emit targets instead of manually running each test via run_kselftest.sh. Signed-off-by:
-
Kees Cook authored
In order to improve the reusability of the kselftest test running logic, this extracts the single-test logic from lib.mk into kselftest/runner.sh which lib.mk can call directly. No changes in output. As part of the change, this moves the "summary" Makefile logic around to set a new "logfile" output. This will be used again in the future "emit_tests" target as well. Signed-off-by:
-
- 22 Apr, 2019 1 commit
-
-
Shuah Khan authored
Build and run gpio when output directory is the src dir. gpio has dependency on tools/gpio and builds tools/gpio objects in the src directory in all cases making the src repo dirty even when object relocation is specified. This fixes the following commands from generating gpio objects in the source repository: make O=dir kselftest export KBUILD_OUTPUT=dir; make kselftest make O=dir -C tools/testing/selftests expoert KBUILD_OUTPUT=dir; make -C tools/testing/selftests The following commands still build gpio objects in the source repo (gpio Makefile needs to fixed): make O=dir kselftest TARGETS="gpio" export KBUILD_OUTPUT=dir; make kselftest TARGETS="gpio" make O=dir -C tools/testing/selftests TARGETS="gpio" expoert KBUILD_OUTPUT=dir; make -C tools/testing/selftests TARGETS="gpio" Signed-off-by:
-
- 19 Apr, 2019 3 commits
-
-
Kees Cook authored
This fixes the various compiler warnings when building the msgque selftest. The primary change is using sys/msg.h instead of linux/msg.h directly to gain the API declarations. Fixes: 3a665531 ("selftests: IPC message queue copy feature test") Signed-off-by:
-
Po-Hsu Lin authored
Test files created by test_create() and test_create_empty() tests will stay in the $efivarfs_mount directory until the system was rebooted. When the tester tries to run this efivarfs test again on the same system, the immutable characteristics in that directory will cause some "Operation not permitted" noises, and a false-positve test result as the file was created in previous run. -------------------- running test_create -------------------- ./efivarfs.sh: line 59: /sys/firmware/efi/efivars/test_create-210be57c-9849-4fc7-a635-e6382d1aec27: Operation not permitted [PASS] -------------------- running test_create_empty -------------------- ./efivarfs.sh: line 78: /sys/firmware/efi/efivars/test_create_empty-210be57c-9849-4fc7-a635-e6382d1aec27: Operation not permitted [PASS] -------------------- Create a file_cleanup() to remove those test files in the end of each test to solve this issue. For the test_create_read, we can move the clean up task to the end of the test to ensure the system is clean. Also, use this function to replace the existing file removal code. Signed-off-by:Po-Hsu Lin <po-hsu.lin@canonical.com> Signed-off-by:
-
Shuah Khan authored
"make kselftest" fails with "Circular Makefile.o <- prepare dependency dropped." error, when lib.mk invokes "make headers_install". Make level 0: Main make calls selftests run_tests target ... Make level n: selftests lib.mk invokes main make's headers_install The secondary level make inherits builtin-rules which will use the rule to generate Makefile.o and runs into "Circular Makefile.o <- prepare dependency dropped." error, and kselftest compile fails. Invoke headers_install target with --no-builtin-rules to avoid circular error. In addition, lib.mk installs headers in the default HDR_PATH, even when build relocation is requested with O= or export KBUILD_OUTPUT. Fix the problem by passing in INSTALL_HDR_PATH. The headers are installed under the specified output "dir/usr". Signed-off-by:
-
- 17 Apr, 2019 10 commits
-
-
Mimi Zohar authored
The get_secureboot_mode() function unnecessarily requires both CONFIG_EFIVAR_FS and CONFIG_EFI_VARS to be enabled to determine if the system is booted in secure boot mode. On some systems the old EFI variable support is not enabled or, possibly, even implemented. This patch first checks the efivars filesystem for the SecureBoot and SetupMode flags, but falls back to using the old EFI variable support. The "secure_boot_file" and "setup_mode_file" couldn't be quoted due to globbing. This patch also removes the globbing. Signed-off-by:
Mimi Zohar <zohar@linux.ibm.com> Reviewed-by:
Petr Vorel <pvorel@suse.cz> Signed-off-by:
-
Mimi Zohar authored
Verify IMA is enabled before failing tests or emitting irrelevant messages. Suggested-by:
Dave Young <dyoung@redhat.com> Signed-off-by:
-
Mimi Zohar authored
Skip the kexec_load and kexec_file_load tests, if they aren't configured in the kernel. This change adds a new requirement that ikconfig is configured in the kexec_load test. Suggested-by:
-
Petr Vorel authored
so the file can be used as kernel config snippet. Signed-off-by:
-
Mimi Zohar authored
The kernel can be configured to verify PE signed kernel images, IMA kernel image signatures, both types of signatures, or none. This test verifies only properly signed kernel images are loaded into memory, based on the kernel configuration and runtime policies. Signed-off-by:
-
Mimi Zohar authored
Many tests require root privileges. Define a common function. Suggested-by:
-
Mimi Zohar authored
Define log_info, log_pass, log_fail, and log_skip functions. Suggested-by:
-
Mimi Zohar authored
Define, update and move get_secureboot_mode() to a common file for use by other tests. Updated to check both the efivar SecureBoot-$(UUID) and SetupMode-$(UUID), based on Dave Young's review. Signed-off-by:
-
Mimi Zohar authored
Remove the few bashisms and use the complete option name for clarity. Signed-off-by:
-
Mimi Zohar authored
As requested move the existing kexec_load selftest and subsequent kexec tests to the selftests/kexec directory. Suggested-by:
-
- 16 Apr, 2019 2 commits
-
-
Kees Cook authored
In order to keep tests from hanging forever, this adds an alarm signal to each test run. This assumes an individual test doesn't take longer than 30 seconds. Signed-off-by:
-
Kees Cook authored
When running without USERNS or PIDNS the seccomp test would hang since it was waiting forever for the child to trigger the user notification since it seems the glibc() abort handler makes a call to getpid(), which would trap again. This changes the getpid filter to getppid, and makes sure ASSERTs execute to stop from spawning the listener. Reported-by:
Shuah Khan <shuah@kernel.org> Fixes: 6a21cc50 ("seccomp: add a return code to trap to userspace") Cc: stable@vger.kernel.org # > 5.0 Signed-off-by:
Tycho Andersen <tycho@tycho.ws> Signed-off-by:
-
- 08 Apr, 2019 9 commits
-
-
Roman Gushchin authored
Dan reported, that cleanup path in test_memcg_subtree_control() triggers a static checker warning: ./tools/testing/selftests/cgroup/test_memcontrol.c:76 \ test_memcg_subtree_control() error: uninitialized symbol 'child2'. Fix this by initializing child2 and parent2 variables and split the cleanup path into few stages. Signed-off-by:
Roman Gushchin <guro@fb.com> Fixes: 84092dbc ("selftests: cgroup: add memory controller self-tests") Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Shuah Khan (Samsung OSG) <shuah@kernel.org> Cc: Mike Rapoport <rppt@linux.vnet.ibm.com> Signed-off-by:
-
ZhangXiaoxu authored
After the first run, the test case 'test_create_read' will always fail because the file is exist and file's attr is 'S_IMMUTABLE', open with 'O_RDWR' will always return -EPERM. Signed-off-by:
ZhangXiaoxu <zhangxiaoxu5@huawei.com> Signed-off-by:
-
Mathieu Desnoyers authored
On smaller systems, running a test with 200 threads can take a long time on machines with smaller number of CPUs. Detect the number of online cpus at test runtime, and multiply that by 6 to have 6 rseq threads per cpu preempting each other. Signed-off-by:
-
Tobin C. Harding authored
Add a test module for the new strscpy_pad() function. Tie it into the kselftest infrastructure for lib/ tests. Acked-by:
Tobin C. Harding <tobin@kernel.org> Signed-off-by:
-
Tobin C. Harding authored
We have a function to copy strings safely and we have a function to copy strings and zero the tail of the destination (if source string is shorter than destination buffer) but we do not have a function to do both at once. This means developers must write this themselves if they desire this functionality. This is a chore, and also leaves us open to off by one errors unnecessarily. Add a function that calls strscpy() then memset()s the tail to zero if the source string is shorter than the destination buffer. Acked-by:
-
Tobin C. Harding authored
We just added a new C header file for use with test modules that are intended to be run with kselftest. We can reduce code duplication by using this header. Use new kselftest header to reduce code duplication in test_printf and test_bitmap test modules. Acked-by:
-
Tobin C. Harding authored
kselftest runs as a userspace process. Sometimes we need to test things from kernel space. One way of doing this is by creating a test module. Currently doing so requires developers to write a bunch of boiler plate in the module if kselftest is to be used to run the tests. This means we currently have a load of duplicate code to achieve these ends. If we have a uniform method for implementing test modules then we can reduce code duplication, ensure uniformity in the test framework, ease code maintenance, and reduce the work required to create tests. This all helps to encourage developers to write and run tests. Add a C header file that can be included in test modules. This provides a single point for common test functions/macros. Implement a few macros that make up the start of the test framework. Add documentation for new kselftest header to kselftest documentation. Acked-by:
-
Tobin C. Harding authored
Currently if we wish to use kselftest to run tests within a kernel module we write a small script to load/unload and do error reporting. There are a bunch of these under tools/testing/selftests/lib/ that are all identical except for the test name. We can reduce code duplication and improve maintainability if we have one version of this. However kselftest requires an executable for each test. We can move all the script logic to a central script then have each individual test script call the main script. Oneliner to call kselftest_module.sh courtesy of Kees, thanks! Add test runner creation script. Convert tools/testing/selftests/lib/*.sh to use new test creation script. Testing ------- Configure kselftests for lib/ then build and boot kernel. Then run kselftests as follows: $ cd /path/to/kernel/tree $ sudo make O=$output_path -C tools/testing/selftests TARGETS="lib" run_tests and also $ cd /path/to/kernel/tree $ cd tools/testing/selftests $ sudo make O=$output_path TARGETS="lib" run_tests and also $ cd /path/to/kernel/tree $ cd tools/testing/selftests $ sudo make TARGETS="lib" run_tests Acked-by:
-
Tobin C. Harding authored
Currently the test_printf module does not have an exit function, this prevents the module from being unloaded. If we cannot unload the module we cannot run the tests a second time. Add an empty exit function. Acked-by:
-
