rseq/selftests: Add support for OpenRISC

Add support for OpenRISC in the rseq selftests. OpenRISC is 32-bit
only.

Tested this with:

Compiler: gcc version 14.2.0 (GCC)
Binutils: GNU asse

rseq/selftests: Add support for OpenRISC

Add support for OpenRISC in the rseq selftests. OpenRISC is 32-bit
only.

Tested this with:

Compiler: gcc version 14.2.0 (GCC)
Binutils: GNU assembler version 2.43.1 (or1k-smh-linux-gnu) using BFD version (GNU Binutils) 2.43.1.20241207
Linux: Linux buildroot 6.13.0-rc2-00005-g1fa73dd6c2d3-dirty #213 SMP Sat Dec 28 22:18:39 GMT 2024 openrisc GNU/Linux
Glibc: 2024-12-13 e4e49583d9 Stafford Horne or1k: Update libm-test-ulps

Signed-off-by: Stafford Horne <[email protected]>
Reviewed-by: Mathieu Desnoyers <[email protected]>
Acked-by: Shuah Khan <[email protected]>

show more ...

selftests/rseq: Do not skip !allowed_cpus for mm_cid

Indexing with mm_cid is incompatible with skipping disallowed cpumask,
because concurrency IDs are based on a virtual ID allocation which is
unre

selftests/rseq: Do not skip !allowed_cpus for mm_cid

Indexing with mm_cid is incompatible with skipping disallowed cpumask,
because concurrency IDs are based on a virtual ID allocation which is
unrelated to the physical CPU mask.

These issues can be reproduced by running the rseq selftests under a
taskset which excludes CPU 0, e.g.

taskset -c 10-20 ./run_param_test.sh

Signed-off-by: Mathieu Desnoyers <[email protected]>
Cc: Shuah Khan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: "Paul E. McKenney" <[email protected]>
Cc: Boqun Feng <[email protected]>
Signed-off-by: Shuah Khan <[email protected]>

show more ...

selftests/rseq: fix kselftest Clang build warnings

When building with Clang, I am getting many warnings from the selftests/rseq tree.

Here's one such example from rseq tree:
| param_test.c:1234:10

selftests/rseq: fix kselftest Clang build warnings

When building with Clang, I am getting many warnings from the selftests/rseq tree.

Here's one such example from rseq tree:
| param_test.c:1234:10: error: address argument to atomic operation must be a pointer to _Atomic type ('intptr_t *' (aka 'long *') invalid)
| 1234 | while (!atomic_load(&args->percpu_list_ptr)) {}
| | ^ ~~~~~~~~~~~~~~~~~~~~~~
| /usr/local/google/home/justinstitt/repos/tc-build/build/llvm/final/lib/clang/18/include/stdatomic.h:140:29: note: expanded from macro 'atomic_load'
| 140 | #define atomic_load(object) __c11_atomic_load(object, __ATOMIC_SEQ_CST)
| | ^ ~~~~~~

Use compiler builtins `__atomic_load_n()` and `__atomic_store_n()` with
accompanying __ATOMIC_ACQUIRE and __ATOMIC_RELEASE, respectively. This
will fix the warnings because the compiler builtins do not expect their
arguments to have _Atomic type. This should also make TSAN happier.

Link: https://github.com/ClangBuiltLinux/linux/issues/1698
Link: https://github.com/ClangBuiltLinux/continuous-integration2/issues/61
Suggested-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Justin Stitt <[email protected]>
Reviewed-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Shuah Khan <[email protected]>

show more ...

selftests/rseq: parametrized test: Report/abort on negative concurrency ID

Report and abort when a negative concurrency ID value is observed by the
spinlock test.

Signed-off-by: Mathieu Desnoyers <

selftests/rseq: parametrized test: Report/abort on negative concurrency ID

Report and abort when a negative concurrency ID value is observed by the
spinlock test.

Signed-off-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

selftests/rseq: Implement parametrized mm_cid test

Adapt to the rseq.h API changes introduced by commits
"selftests/rseq: <arch>: Template memory ordering and percpu access mode".

Build a new param

selftests/rseq: Implement parametrized mm_cid test

Adapt to the rseq.h API changes introduced by commits
"selftests/rseq: <arch>: Template memory ordering and percpu access mode".

Build a new param_test_mm_cid, param_test_mm_cid_benchmark, and
param_test_mm_cid_compare_twice executables to test the new "mm_cid"
rseq field.

Signed-off-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

selftests/rseq: Remove RSEQ_SKIP_FASTPATH code

This code is not currently build by the test Makefile, adds complexity,
and is not overall useful considering that the abort handling loops to
retry th

selftests/rseq: Remove RSEQ_SKIP_FASTPATH code

This code is not currently build by the test Makefile, adds complexity,
and is not overall useful considering that the abort handling loops to
retry the fast-path.

Signed-off-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

rseq/selftests: Add support for RISC-V

Add support for RISC-V in the rseq selftests, which covers both
64-bit and 32-bit ISA with little endian mode.

Signed-off-by: Vincent Chen <vincent.chen@sifiv

rseq/selftests: Add support for RISC-V

Add support for RISC-V in the rseq selftests, which covers both
64-bit and 32-bit ISA with little endian mode.

Signed-off-by: Vincent Chen <[email protected]>
Tested-by: Eric Lin <[email protected]>
Reviewed-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Palmer Dabbelt <[email protected]>

show more ...

selftests/rseq: Fix warnings about #if checks of undefined tokens

Signed-off-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link:

selftests/rseq: Fix warnings about #if checks of undefined tokens

Signed-off-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]

show more ...

selftests/rseq: Fix ppc32 offsets by using long rather than off_t

The semantic of off_t is for file offsets. We mean to use it as an
offset from a pointer. We really expect it to fit in a single reg

selftests/rseq: Fix ppc32 offsets by using long rather than off_t

The semantic of off_t is for file offsets. We mean to use it as an
offset from a pointer. We really expect it to fit in a single register,
and not use a 64-bit type on 32-bit architectures.

Fix runtime issues on ppc32 where the offset is always 0 due to
inconsistency between the argument type (off_t -> 64-bit) and type
expected by the inline assembler (32-bit).

Signed-off-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]

show more ...

selftests/rseq: Remove useless assignment to cpu variable

Signed-off-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https:/

selftests/rseq: Remove useless assignment to cpu variable

Signed-off-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]

show more ...

rseq/selftests: Fix MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ build error under other arch.

Except arch x86, the function rseq_offset_deref_addv is not defined.
The function test_membarrier_manager_thre

rseq/selftests: Fix MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ build error under other arch.

Except arch x86, the function rseq_offset_deref_addv is not defined.
The function test_membarrier_manager_thread call rseq_offset_deref_addv
produces a build error.

The RSEQ_ARCH_HAS_OFFSET_DEREF_ADD should contain all the code
for the MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ.
If the other Arch implements this feature,
defined RSEQ_ARCH_HAS_OFFSET_DEREF_ADD in the header file
to ensure that this feature is available.

Following build errors:

param_test.c: In function ‘test_membarrier_worker_thread’:
param_test.c:1164:10: warning: implicit declaration of function ‘rseq_offset_deref_addv’
ret = rseq_offset_deref_addv(&args->percpu_list_ptr,
^~~~~~~~~~~~~~~~~~~~~~
/tmp/ccMj9yHJ.o: In function `test_membarrier_worker_thread':
param_test.c:1164: undefined reference to `rseq_offset_deref_addv'
param_test.c:1164: undefined reference to `rseq_offset_deref_addv'
collect2: error: ld returned 1 exit status
make: *** [/selftests/rseq/param_test_benchmark] Error 1

Signed-off-by: Xingxing Su <[email protected]>
Acked-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Shuah Khan <[email protected]>

show more ...

rseq/selftests: Test MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ

Based on Google-internal RSEQ work done by Paul Turner and Andrew
Hunter.

This patch adds a selftest for MEMBARRIER_CMD_PRIVATE_EXPEDITED_

rseq/selftests: Test MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ

Based on Google-internal RSEQ work done by Paul Turner and Andrew
Hunter.

This patch adds a selftest for MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ.
The test quite often fails without the previous patch in this
patchset, but consistently passes with it.

Signed-off-by: Peter Oskolkov <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Acked-by: Mathieu Desnoyers <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]

show more ...

rseq/selftests: Fix: Namespace gettid() for compatibility with glibc 2.30

glibc 2.30 introduces gettid() in public headers, which clashes with
the internal static definition within rseq selftests.

rseq/selftests: Fix: Namespace gettid() for compatibility with glibc 2.30

glibc 2.30 introduces gettid() in public headers, which clashes with
the internal static definition within rseq selftests.

Rename gettid() to rseq_gettid() to eliminate this symbol name clash.

Reported-by: Tommi T. Rantala <[email protected]>
Signed-off-by: Mathieu Desnoyers <[email protected]>
Cc: Shuah Khan <[email protected]>
Cc: Tommi T. Rantala <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Peter Zijlstra (Intel) <[email protected]>
Cc: "Paul E. McKenney" <[email protected]>
Cc: Boqun Feng <[email protected]>
Cc: "H . Peter Anvin" <[email protected]>
Cc: Paul Turner <[email protected]>
Cc: Dmitry Vyukov <[email protected]>
Cc: <[email protected]> # v4.18+
Signed-off-by: Shuah Khan <[email protected]>

show more ...

rseq/selftests: fix parametrized test with -fpie

On x86-64, the parametrized selftest code for rseq crashes with a
segmentation fault when compiled with -fpie. This happens when the
param_test binar

rseq/selftests: fix parametrized test with -fpie

On x86-64, the parametrized selftest code for rseq crashes with a
segmentation fault when compiled with -fpie. This happens when the
param_test binary is loaded at an address beyond 32-bit on x86-64.

The issue is caused by use of a 32-bit register to hold the address
of the loop counter variable.

Fix this by using a 64-bit register to calculate the address of the
loop counter variables as an offset from rip.

Signed-off-by: Mathieu Desnoyers <[email protected]>
Acked-by: "Paul E . McKenney" <[email protected]>
Cc: <[email protected]> # v4.18
Cc: Shuah Khan <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Joel Fernandes <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Catalin Marinas <[email protected]>
Cc: Dave Watson <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: [email protected]
Cc: "H . Peter Anvin" <[email protected]>
Cc: Chris Lameter <[email protected]>
Cc: Russell King <[email protected]>
Cc: Michael Kerrisk <[email protected]>
Cc: "Paul E . McKenney" <[email protected]>
Cc: Paul Turner <[email protected]>
Cc: Boqun Feng <[email protected]>
Cc: Josh Triplett <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Ben Maurer <[email protected]>
Cc: Andy Lutomirski <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Linus Torvalds <[email protected]>
Signed-off-by: Shuah Khan (Samsung OSG) <[email protected]>

show more ...

rseq/selftests: add s390 support

Implement support for s390 in the rseq selftests, in order to sanity
check the recently enabled rseq syscall. The Implementation covers both
64-bit and 31-bit mode.

rseq/selftests: add s390 support

Implement support for s390 in the rseq selftests, in order to sanity
check the recently enabled rseq syscall. The Implementation covers both
64-bit and 31-bit mode.

Acked-by: Heiko Carstens <[email protected]>
Signed-off-by: Vasily Gorbik <[email protected]>
Signed-off-by: Martin Schwidefsky <[email protected]>

show more ...

rseq/selftests: Add support for arm64

Hook up arm64 support to the rseq selftests.

Acked-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Will Deacon <[email protected]>

rseq/selftests: Implement MIPS support

Implement support for both MIPS32 & MIPS64 in the rseq selftests, in
order to sanity check the recently enabled rseq syscall.

The tests all pass on a MIPS Bos

rseq/selftests: Implement MIPS support

Implement support for both MIPS32 & MIPS64 in the rseq selftests, in
order to sanity check the recently enabled rseq syscall.

The tests all pass on a MIPS Boston development board running either a
MIPS32r2 interAptiv CPU & a MIPS64r6 I6500 CPU, both of which were
configured with 2 cores each of which have 2 hardware threads (VP(E)s) -
ie. 4 CPUs.

Signed-off-by: Paul Burton <[email protected]>
Reviewed-by: James Hogan <[email protected]>
Patchwork: https://patchwork.linux-mips.org/patch/19524/
Cc: Ralf Baechle <[email protected]>
Cc: Mathieu Desnoyers <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Paul E. McKenney <[email protected]>
Cc: Boqun Feng <[email protected]>
Cc: [email protected]
Cc: [email protected]

show more ...

rseq/selftests: Provide parametrized tests

"param_test" is a parametrizable restartable sequences test. See
the "--help" output for usage.

"param_test_benchmark" is the same as "param_test", but it

rseq/selftests: Provide parametrized tests

"param_test" is a parametrizable restartable sequences test. See
the "--help" output for usage.

"param_test_benchmark" is the same as "param_test", but it removes
testing book-keeping code to allow accurate benchmarks.

"param_test_compare_twice" is the same as "param_test", but it performs
each comparison within rseq critical section twice, thus validating
invariants. If any of the second comparisons fails, an error message
is printed and the test aborts.

Signed-off-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Cc: Joel Fernandes <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Catalin Marinas <[email protected]>
Cc: Dave Watson <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Shuah Khan <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: [email protected]
Cc: "H . Peter Anvin" <[email protected]>
Cc: Chris Lameter <[email protected]>
Cc: Russell King <[email protected]>
Cc: Andrew Hunter <[email protected]>
Cc: Michael Kerrisk <[email protected]>
Cc: "Paul E . McKenney" <[email protected]>
Cc: Paul Turner <[email protected]>
Cc: Boqun Feng <[email protected]>
Cc: Josh Triplett <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Ben Maurer <[email protected]>
Cc: [email protected]
Cc: Andy Lutomirski <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Linus Torvalds <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]

show more ...