Disable debug preemption in x86_64

[rlmenge]

Merge Checklist

All boxes should be checked before merging the PR (just tick any boxes which don't apply to this PR)

• The toolchain has been rebuilt successfully (or no changes were made to it)
• The toolchain/worker package manifests are up-to-date
• Any updated packages successfully build (or no packages were changed)
• Packages depending on static components modified in this PR (Golang, *-static subpackages, etc.) have had their Release tag incremented.
• Package tests (%check section) have been verified with RUN_CHECK=y for existing SPEC files, or added to new SPEC files
• All package sources are available
• cgmanifest files are up-to-date and sorted (./cgmanifest.json, ./toolkit/scripts/toolchain/cgmanifest.json, .github/workflows/cgmanifest.json)
• LICENSE-MAP files are up-to-date (./LICENSES-AND-NOTICES/SPECS/data/licenses.json, ./LICENSES-AND-NOTICES/SPECS/LICENSES-MAP.md, ./LICENSES-AND-NOTICES/SPECS/LICENSE-EXCEPTIONS.PHOTON)
• All source files have up-to-date hashes in the *.signatures.json files
• sudo make go-tidy-all and sudo make go-test-coverage pass
• Documentation has been updated to match any changes to the build system
• Ready to merge

---

Summary

Performance improvement.

In 6.1, the recommended default for DEBUG_PREEMPT was y. However, this led to performance hits and overhead due to frequent calls to __this_cpu_preempt_check().

The defualt was later changed in "[cc60039] lib/Kconfig.debug: do not enable DEBUG_PREEMPT by default".

Therefore, go with new recommeded default to improve performance.
Kconfig

Change Log

• Disable debug preemption in x86_64

Does this affect the toolchain?

YES

Associated issues

• https://microsoft.visualstudio.com/OS/_workitems/edit/55530233

Test Methodology

• Pipeline build id: https://dev.azure.com/mariner-org/mariner/_build/results?buildId=712227&view=results

Tested it booted on an Azure Standard_D96ps_v6 (ARM64) and Standard_D8s_v4 (x86_64)

Additionally, saw improved perf results without DEBUG_PREEPMT

Results with 6.6.57.1-6 (currently published)

azureuser@mariner-vm-StandardD8sv4-1 [ ~ ]$ perf bench sched messaging -g 16 -l 1000
# Running 'sched/messaging' benchmark:
# 20 sender and receiver processes per group
# 16 groups == 640 processes run

     Total time: 3.285 [sec]
azureuser@mariner-vm-StandardD8sv4-1 [ ~ ]$ perf bench sched pipe
# Running 'sched/pipe' benchmark:
# Executed 1000000 pipe operations between two processes

     Total time: 21.453 [sec]

      21.453281 usecs/op
          46612 ops/sec

With 6.6.64.2-2 with DEBUG_PREEMPT

azureuser@mariner-vm-StandardD8sv4-1 [ ~ ]$ uname -a
Linux mariner-vm-StandardD8sv4-1 6.6.64.2-2.azl3 #1 SMP PREEMPT_DYNAMIC Thu Jan 16 05:20:03 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
azureuser@mariner-vm-StandardD8sv4-1 [ ~ ]$ perf bench sched messaging -g 16 -l 1000
# Running 'sched/messaging' benchmark:
# 20 sender and receiver processes per group
# 16 groups == 640 processes run

     Total time: 3.315 [sec]
azureuser@mariner-vm-StandardD8sv4-1 [ ~ ]$ perf bench sched pipe
# Running 'sched/pipe' benchmark:
# Executed 1000000 pipe operations between two processes

     Total time: 21.731 [sec]

      21.731123 usecs/op
          46016 ops/sec

With 6.6.64.2 without DEBUG_PREEMPT - improvement by ~10%

azureuser@mariner-vm-StandardD8sv4-1 [ ~ ]$ perf bench sched messaging -g 16 -l 1000
# Running 'sched/messaging' benchmark:
# 20 sender and receiver processes per group
# 16 groups == 640 processes run

     Total time: 2.997 [sec]

azureuser@mariner-vm-StandardD8sv4-1 [ ~ ]$ perf bench sched pipe
# Running 'sched/pipe' benchmark:
# Executed 1000000 pipe operations between two processes

     Total time: 20.782 [sec]

      20.782934 usecs/op
          48116 ops/sec