Lup Yuen Lee lupyuen

testbuild.sh: Does this code perform retry of builds correctly?

Mostly yes: it does retry the whole build sequence, but there are a couple of correctness/robustness problems that can make retries behave differently than intended (and in one case can break parsing).

What it retries (and what “success” means)

retrytest runs dotest $line up to 6 attempts (for i in {1..6}), despite the comment saying “Retry 5 times”.
Each attempt starts with fail=0, then dotest runs:
1. distclean
2. (if not skipped) configure, build, optional run, then refresh
If any of those set global fail=1, the attempt is considered failed and it sleeps with exponential backoff and tries again.

NuttX Apps: Test Cases for new implementation of NuttX PR Labeling

See the PR: apache/nuttx-apps#3408

Read the Article: https://lupyuen.org/articles/prtarget

Discussion: apache/nuttx#18359

Arch Labeling

Test Cases for new implementation of NuttX PR Labeling

See the PR: apache/nuttx#18422

Read the Article: https://lupyuen.org/articles/prtarget

Discussion: apache/nuttx#18359

Arch Labeling

$ git clone https://github.com/lupyuen8/nuttx

$ zizmor nuttx/.github/workflows/labeler.yml
🌈 zizmor v1.22.0
 INFO audit: zizmor: 🌈 completed nuttx/.github/workflows/labeler.yml
No findings to report. Good job! (4 suppressed)

Test Cases for new implementation of NuttX PR Labeling

See the PR: apache/nuttx#18404

Discussion: apache/nuttx#18359

Note: The Test PRs below look kinda messed up, showing 48 Commits instead of One Single Commit. That's because I squashed the 47 commits for labeler.yml and pr_labeler.yml, after running the tests. Please ignore all commits for labeler.yml and pr_labeler.yml in the PRs below, they are not inspected by our new PR Labeler.

	Script started on 2026-03-30 14:43:57+08:00 [COMMAND="/home/luppy/nuttx-release/release-starpro64.sh" TERM="xterm-256color" TTY="/dev/pts/21" COLUMNS="82" LINES="63"]
	----- Validate NuttX Release for StarPro64
	release=12.13.0
	candidate=RC0
	hash=b9b5d9afbf
	Now running https://github.com/lupyuen/nuttx-release/blob/main/release-starpro64.sh
	+ export STARPRO64_SERVER=thinkcentre
	+ STARPRO64_SERVER=thinkcentre
	+ export TFTP_SERVER=tftpserver
	+ TFTP_SERVER=tftpserver

	#!/bin/bash
	## Adjust the audio to a consistent perceived loudness level without clipping

	input_file=$1
	output_file=${input_file%.*}a.mov
	echo "Transforming $input_file to $output_file"

	## Previously: -filter:a 'loudnorm=I=-16:LRA=11:tp=-2'
	## I=-16 sets the target integrated loudness to -16 LUFS (Loudness Units Full Scale), a common target for podcasts/music.
	## LRA=11 sets the Loudness Range adjustment.

	$ git clone https://github.com/lupyuen6/nuttx

	$ zizmor nuttx/.github/workflows/pr_labeler.yml
	🌈 zizmor v1.22.0
	INFO audit: zizmor: 🌈 completed nuttx/.github/workflows/pr_labeler.yml
	error[dangerous-triggers]: use of fundamentally insecure workflow trigger
	--> nuttx/.github/workflows/pr_labeler.yml:20:1
	\|
	20 \| / on:
	21 \| \| workflow_run:

	$ brew install zizmor
	$ git clone https://github.com/apache/nuttx
	$ zizmor nuttx
	🌈 zizmor v1.22.0
	INFO audit: zizmor: 🌈 completed nuttx/.github/actions/ci-container/action.yaml
	INFO audit: zizmor: 🌈 completed nuttx/.github/actions/free-disk-space/action.yaml
	INFO audit: zizmor: 🌈 completed nuttx/.github/dependabot.yml
	INFO audit: zizmor: 🌈 completed nuttx/.github/workflows/arch.yml
	INFO audit: zizmor: 🌈 completed nuttx/.github/workflows/build.yml
	INFO audit: zizmor: 🌈 completed nuttx/.github/workflows/check.yml

	luppy@nuttx-dashboard-vm:~/ingest-nuttx-builds$ cd $HOME/ingest-nuttx-builds
	luppy@nuttx-dashboard-vm:~/ingest-nuttx-builds$ ./run.sh
	+ (( 1 ))
	+ (( 1 ))
	+ find /home/luppy/riscv/nuttx -name defconfig
	+ set +x
	+ cargo run -- --user lupyuen --repo nuttx-build-log --defconfig /tmp/defconfig.txt
	+ tr -d '\033\007'
	Finished `dev` profile [unoptimized + debuginfo] target(s) in 0.24s
	Running `target/debug/ingest-nuttx-builds --user lupyuen --repo nuttx-build-log --defconfig /tmp/defconfig.txt`