Skip to content

Adding robustness to the hostname indexing #466

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
TApplencourt merged 3 commits into from
Feb 16, 2026
Merged

Adding robustness to the hostname indexing #466

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
456f6c4
adding robustness to the hostname indexing
colleeneb Feb 16, 2026
d94c2a9
removing split from hostfile read
colleeneb Feb 16, 2026
eeb9db9
removing split from hostfile read
colleeneb Feb 16, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 8 additions & 4 deletions xprof/xprof.rb.in
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -251,11 +251,15 @@ module MPITopo
number_hostname, hostname_index = if hostfile.nil?
[10_000, rank_id / local_size]
else
# splitting by "." in case the hostfile contains full hostnames with address
# (e.g. x4117c4s4b0n0.hsn.cm.aurora.alcf.anl.gov) and the Socket.gethostname is just x4117c4s4b0n0
hostnames = File.readlines(hostfile).map { |el| el.split('.', 2).first }
hostnames = File.readlines(hostfile)
# find index of hostname_string in list_hostnames
hostname_id = hostnames.find_index(Socket.gethostname)
hostname_id = hostnames.find_index{ |host| host.start_with?(Socket.gethostname) }
Copy link
Collaborator

@TApplencourt TApplencourt Feb 16, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

so we can do it in online now :D
File.readlines(hostfile).find_index{ |hostname| hostname.start_with?(Socket.gethostname) }

Copy link
Contributor Author

@colleeneb colleeneb Feb 16, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ah, we need the hostnames to get the number_hostname. We could make this change but then we'd need to read the hostfile again to get the number of hosts.

Copy link
Collaborator

@TApplencourt TApplencourt Feb 16, 2026 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Oh true, my bad then! Didn't read enough of the diff :D

# without "start_with" this could fail if something happens like on sunspot
# where Socket.gethostname returns x1921c1s2b0n0 but the hostfile contains
# x1921c1s2b0n0-hsn0.hsn.cm.sunspot.alcf.anl.gov.
if hostname_id.nil?
raise "Trying to find the index of each hostname in the hostfile list did not work."
end
# compute the offset, based on splitting MAX_UINT64_VALUE by the number of hosts
[hostnames.length, hostname_id]
end
Expand Down