Skip to content

kernel: no loop devices at boot, one 8250 port (experiment) - #284

Draft
aledbf wants to merge 1 commit into
mainfrom
kernel-trim
Draft

aledbf wants to merge 1 commit into
mainfrom
kernel-trim

Conversation

@aledbf

@aledbf aledbf commented Nov 19, 2026

Copy link
Copy Markdown
Member

An experiment, measured with boot:initcalls against the release's kernel (kernel_run).

  • BLK_DEV_LOOP_MIN_COUNT 8 → 0: loop devices are made on demand through /dev/loop-control; loop_init was 1.08 ms nested.
  • SERIAL_8250_RUNTIME_UARTS 4 → 1: the machine has one port; serial8250_init was 0.75 ms nested.

Both numbers are from lab 44563496308. Not taken: RTC_DRV_CMOS off (cmos_init 1.63 ms nested), because the image ships hwclock on purpose.

🤖 Generated with Claude Code

BLK_DEV_LOOP_MIN_COUNT 8 -> 0: loop devices are made on demand through
/dev/loop-control, as losetup and docker ask for them; loop_init was
1.08 ms nested. SERIAL_8250_RUNTIME_UARTS 4 -> 1: the machine has one
port, ttyS0, and serial8250_init probing four was 0.75 ms nested
(boot:initcalls, lab 44563496308).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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