Skip to content

kselftest/arm64: Floating point stress test harness

Mark Brown linux-arm-kernel linux-kselftest 2 versions – In review

Get series with b4
b4 am [email protected]

Versions newest first · linked by author and title

  1. v2 Latest4 patches 1 replyIn review
  2. v14 patches 0 repliesSuperseded

Changes in v2 from the cover letter

- Rebase onto v6.0-rc3.
- Announce the results of enumeration before we start everything.

Patches in v2 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/4kselftest/arm64: Always encourage preemption for za-testtools/testing/selftests/arm64/fp/za-test.S+1 -60 replies
2/4kselftest/arm64: Count SIGUSR2 deliveries in FP stress teststools/testing/selftests/arm64/fp/asm-offsets.h tools/testing/selftests/arm64/fp/fpsimd-test.S tools/testing/selftests/arm64/fp/sve-test.S +1 more+46 -00 replies
3/4kselftest/arm64: Install signal handlers before output in FP stress teststools/testing/selftests/arm64/fp/fpsimd-test.S tools/testing/selftests/arm64/fp/sve-test.S tools/testing/selftests/arm64/fp/za-test.S+72 -720 replies
4/4kselftest/arm64: kselftest harness for FP stress teststools/testing/selftests/arm64/fp/.gitignore tools/testing/selftests/arm64/fp/Makefile tools/testing/selftests/arm64/fp/fp-stress.c+540 -10 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v2 yet.

Read the v2 thread

Cover letter [PATCH v2 0/4] kselftest/arm64: Floating point stress test harness

This series has a few small enhancements for the existing standalone floating point stress tests and then builds on those with a kselftest integrated program which gives those a very quick spin from within kselftest, as well as having an option to set a custom timeout to allow for use with longer soak testing. This makes it much easier to get thorough testing of the floating point state management logic, rather than requiring custom setup for coverage of the various vector lengths in the system as is needed at present.

It might be nice in future to extend this to attach to some or all of the test programs with ptrace and read/write their registers as another means of potentially triggering race conditions or corruption but that's definitely another step.

v2:
- Rebase onto v6.0-rc3.
- Announce the results of enumeration before we start everything.

Mark Brown (4):
kselftest/arm64: Always encourage preemption for za-test kselftest/arm64: Count SIGUSR2 deliveries in FP stress tests kselftest/arm64: Install signal handlers before output in FP stress
tests
kselftest/arm64: kselftest harness for FP stress tests

 tools/testing/selftests/arm64/fp/.gitignore   |   1 +
 tools/testing/selftests/arm64/fp/Makefile     |   5 +-
 .../testing/selftests/arm64/fp/asm-offsets.h  |   1 +
 tools/testing/selftests/arm64/fp/fp-stress.c  | 535 ++++++++++++++++++
 .../testing/selftests/arm64/fp/fpsimd-test.S  |  51 +-
 tools/testing/selftests/arm64/fp/sve-test.S   |  51 +-
 tools/testing/selftests/arm64/fp/za-test.S    |  58 +-
 7 files changed, 641 insertions(+), 61 deletions(-)
 create mode 100644 tools/testing/selftests/arm64/fp/fp-stress.c

base-commit: b90cb1053190353cc30f0fef0ef1f378ccc063c5

Signature · 1 line
2.30.2