Skip to content
This repository was archived by the owner on Mar 16, 2026. It is now read-only.

Smarter CPU affinity selection#10

Merged
il-steffen merged 2 commits intoIntelLabs:masterfrom
il-steffen:flex_affinity
Sep 23, 2022
Merged

Smarter CPU affinity selection#10
il-steffen merged 2 commits intoIntelLabs:masterfrom
il-steffen:flex_affinity

Conversation

@il-steffen
Copy link
Copy Markdown
Contributor

Individual workers now pick the Nth item out of currently configured CPU set rather than overriding an existing config. This allows to limit the cpu set more flexibly by the parent, e.g. using taskset.

Additionally, before starting the workers, the manager now performs a scan for other Qemu instances in the system and removes their respective pinned CPUs from the own available set.

Together, this should avoid the need for manually checking available CPUs and defining --cpu-offset for kafl_fuzz.py

Individual workers now pick the Nth item out of currently configured CPU set
rather than overriding an existing config. This allows to limit the
cpu set more flexibly by the parent, e.g. using taskset.

Additionally, before starting the workers, the manager now performs
a scan for other Qemu instances in the system and removes their
respective pinned CPUs from the own available set.

Together, this should avoid the need for manually checking available
CPUs and defining --cpu-offset for kafl_fuzz.py
@il-steffen il-steffen merged commit 42f77b0 into IntelLabs:master Sep 23, 2022
@il-steffen il-steffen deleted the flex_affinity branch October 4, 2022 14:39
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant