Skip to content

Describe IO threads that follow the HOUSEKEEPING pool - #162

Open
rucoder wants to merge 2 commits into
lf-edge:mainfrom
rucoder:rucoder/io-placement-follows-housekeeping
Open

rucoder wants to merge 2 commits into
lf-edge:mainfrom
rucoder:rucoder/io-placement-follows-housekeeping

Conversation

@rucoder

@rucoder rucoder commented Nov 9, 2026 •

Copy link
Copy Markdown
Contributor

Comments only: no field or value changes.

Under IO_PLACEMENT_HOUSEKEEPING, EVE ran a VM's non-vCPU QEMU threads on the CPUs reserved for EVE alone, with a fallback for a node that reserves none. It now runs them on the whole HOUSEKEEPING pool -- the CPUs reserved for EVE and every other CPU no workload holds, never a kernel-isolated one -- and moves them as other workloads take and release CPUs (lf-edge/eve#6335). The threads then spread over every shared CPU instead of piling onto EVE's own.

This updates the comments that describe it:

  • IO_PLACEMENT_HOUSEKEEPING (config/vm.proto): the threads run on the HOUSEKEEPING pool and follow it; the pool always holds the CPUs reserved for EVE.
  • AppCPUAssignment.io_cpu_ids (info/info.proto): the pool's CPUs, which change as other workloads take and release CPUs.
  • CPU_POOL_KIND_HOUSEKEEPING: also where those threads run, and why the pool is never empty: EVE never gives the CPUs reserved for it (the lowest eve_max_vcpus) to a workload of its own, and housekeeping workloads share them with EVE, so the DEDICATED pool can grow to every CPU but those.
  • CPU_PLACEMENT_ERROR_INSUFFICIENT: no longer names an IO_PLACEMENT_HOUSEKEEPING with no CPU to serve it, which cannot happen.

Two commits as usual: the proto change, then make proto (only the comments in the generated Go code change).

@rucoder
rucoder requested a review from eriknordmark as a code owner November 9, 2026 15:46
Under IO_PLACEMENT_HOUSEKEEPING a VM's non-vCPU threads ran on the
CPUs reserved for EVE alone, with a fallback for a node that reserves
none. EVE now runs them on the whole HOUSEKEEPING pool -- the CPUs
reserved for EVE and every other CPU no workload holds, never a
kernel-isolated one -- and moves them as other workloads take and
release CPUs, so they spread over every shared CPU instead of piling
onto EVE's.

Say so in IO_PLACEMENT_HOUSEKEEPING and AppCPUAssignment.io_cpu_ids.
CPU_POOL_KIND_HOUSEKEEPING now says why the pool is never empty: EVE
never gives the CPUs reserved for it (the lowest eve_max_vcpus) to a
workload of its own, so the pool always holds them, and housekeeping
workloads share them with EVE. That also leaves
IO_PLACEMENT_HOUSEKEEPING without a case to refuse, so
CPU_PLACEMENT_ERROR_INSUFFICIENT no longer names one. Comments only:
no field or value changes.

Signed-off-by: Mikhail Malyshev <mike.malyshev@gmail.com>
Regenerated with make proto after the comment changes to vm.proto and info.proto; only the comments in the generated Go code change.

Signed-off-by: Mikhail Malyshev <mike.malyshev@gmail.com>
@rucoder
rucoder force-pushed the rucoder/io-placement-follows-housekeeping branch from 7cf16aa to 60c64c7 Compare November 9, 2026 18:23
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant