Additional paths will be inserted into the search path used for executables on systems which have a compatible CPU. Those additional paths will mirror the AMD64 / x86_64 "microarchitecture levels" supported by the glibc-hwcaps mechanism: x86-64-v2, x86-64-v3, x86_64-v4. Systemd will be modified to insert the additional directories into the $PATH environment variable (affecting all programs on the system) and the equivalent internal mechanism in systemd (affecting what executables are used by services). Individual packages can provide optimized libraries via the glibc-hwcaps mechanism and optimized executables via the extended search path. This optimized code will be used if the CPU supports it. Which packages provide the optimized code and at which level will be made by individual package maintainers based on benchmark results.
Owners, do not implement this work until the FESCo vote has explicitly ended. The Fedora Program Manager will create a tracking bug in Bugzilla for this Change, which is your indication to proceed. See the FESCo ticket policy and the Changes policy for more information.
REMINDER: This ticket is for FESCo members to vote on the proposal. Further discussion should happen in the devel list thread linked above.
There still seems to be a fair bit of discussion about how this should be done on the mailing list thread.
On the one hand, yes there is still a lot of ongoing discussion. On the other hand, I've not yet heard anyone offer an alternative that they plan to implement. Given that we're already late in the F40 cycle, I think we need to start the voting process.
So, vote +1 if you think we should give the approach described in the Change the opportunity to prove itself out and vote -1 if you think that it's an unsound approach.
As far as I can tell, even if we do it, it's still quite possible to opt-out, even on supported hardware. I'm +1 for trying it.
Well, it seemed that there was the plan in the change, but then an alternative plan was suggested ( using a wrapper script).
I'd like to know for sure what way the change owners are going. Is the path munging still the plan? Or the wrapper script?
I guess I am +1 either way, but would be good to know which direction things are going.
I would like to know what the change owners plan to do here as well, hence why I haven't voted.
Apologies, this needs more design work, and it's too late for F40 anyway. We'll rework the proposal and resubmit it again.
Metadata Update from @zbyszek: - Issue close_status updated to: Rejected - Issue status updated to: Closed (was: Open)
Login to comment on this ticket.