perf(release): trim AIGS engines, skip compressing TRT blobs, time installer build#11
Merged
Merged
Conversation
…staller build - manifest: VfxModelGlobs -> AIGS_*_m0* (green-screen MODE is hardcoded 0 in aigs.cpp, so only m0 engines ever load). Drops m1/m2/m3 -> ~290 MB off the multi-arch bundle. - installer: ship .trtpkg engines with nocompression + exclude them from the catch-all, so the solid lzma2/ultra64 block stops re-compressing already- compressed TensorRT blobs. Faster release build, no double-pack. - build-installer: stopwatch timing split per step (shim/dotnet/bundle/iscc) so the next release run shows where the ~7 min goes. ci.yml is untouched (already <1 min on the self-hosted runner); the slow path was release.yml's installer build. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Build/CI optimization. Measured on the self-hosted RTX runner.
maxine-manifest.psd1VfxModelGlobs->AIGS_*_m0*aigs.cpp), only m0 loads -> drops m1/m2/m3, ~290 MB off the multi-arch bundleCameraOnScreen.iss.trtpkgenginesnocompression+ excluded from catch-allbuild-installer.ps1shim / dotnet / bundle / isccsecondsWhy
.exesize + 15-min CI. Findings:cublasLt514 MB +cufft181 MB) is already excluded by the current TRT-10.9 manifest; the on-diskdist\that showed 1.5 GB was stale (Nov-2024 SDK).Verify (runner)
bundle-maxine.ps1fresh-size measure.trace_closure/bundle_probeon the trimmed bundle (both effects load,COS_*unset).Local: 21 Pester tests pass (
Bundle-Maxine.Tests,Build-Installer.Tests).🤖 Generated with Claude Code