Created attachment 1452068 [details] Log file from kernel tests. Description of problem: Kernel 4.16.15-300.fc28.x86_64 on F28 failed the PAX test. This is on a bare metal installation of F28 that is fully up to date. The hardware is a Lenovo H50-55 with an AMD A10-7800 processor, 12Gb ram, and 2 Tb hard drive. How reproducible: Reproduces everytime the kernel tests are run. This bug has been persistent through F27 (Bug #1546440)and now is occurring in F28 Steps to Reproduce: 1. CD kernel-tests 2. sudo ./runtests.sh 3. The failure is recorded both on the terminal and in the log file. Actual results: PAX test fails Expected results: PAX test pass Additional info: The log file is attached.
This bug persists in kernel 4.16.16-300. Starting test ./default/paxtest Main executable randomisation (PIE) randomness is less than 28 bits (26)
This bug persists in kernel 4.17.2-200.fc28.x86_64. Starting test ./default/paxtest Main executable randomisation (PIE) randomness is less than 28 bits (25)
This bug persists in kernel 4.17.3-200.fc28.x86_64. Starting test ./default/paxtest Main executable randomisation (PIE) randomness is less than 28 bits (25)
This bug persists in kernel 4.17.5-200.fc28.x86_64. Starting test ./default/paxtest Main executable randomisation (PIE) randomness is less than 28 bits (25)
This bug persists in kernel 4.17.6-200.fc28.x86_64. Starting test ./default/paxtest Main executable randomisation (PIE) randomness is less than 28 bits (25)
This bug persists in kernel 4.17.7-200.fc28.x86_64. Starting test ./default/paxtest Main executable randomisation (PIE) randomness is less than 28 bits (26)
*** Bug 1546440 has been marked as a duplicate of this bug. ***
This is actually an issue with the test, and how AMD is working after the spectre/meltdown updates. It needs to be fixed in the test. The issue should be tracked on https://pagure.io/kernel-tests/issues