On 10/2/24 15:01, Jason A. Donenfeld wrote:
On Wed, Oct 02, 2024 at 01:45:57PM -0600, Shuah Khan wrote:
This is not different from other kernel APIs and enhancements and correspo0nding updates to the existing selftests.
The vdso_test_getrandom test a user-space program exists in 6.11.
Use should be able to run vdso_test_getrandom compiled on 6.12 repo on a 6.11 kernel. vdso_test_getrandom test a user-space program exists in 6.11. Users should be able to run vdso_test_getrandom compiled on 6.12 repo on a 6.11 kernel. This is what several CIs do.
The x86 test from 6.12 works just fine on 6.11.
Yes x86 test is a good example to look at that handles 32-bit and 640-bit issues you brought up in your email.
That is reason why I asked you to refer to x86 Makefile to solve the architecture differences related to this test you mentioned in your email.
I really don't follow you at all or what you're getting at. I think if you actually look at the code, you'll be mostly okay with it. And if there's something that looks awry to you, send a patch or describe to me clearly what looks wrong and I'll send a patch.
To be very clear about the selftest guidelines (covered in kselftest.rst document)
1. Ideally selftests should compile on all architectures. Exception to this are few architecture specific features which can be selectively compiled either with ifdef statements in the code or Makefile arch checks.
The goal is to keep these to a minimum so we can a wide range of tests in CIs and other test systems.
2. Selftest from mainline should run on stable releases handling missing features and missing config options with skip so we don't have to deal false failures.
The goal is to minimize false negatives and false positives.
3. Reported results are clear to the users and testers.
Thank you for the patches. I will review your patches and give you feedback.
thanks, -- Shuah