Move texture upload preparation off the main thread

[LisaScheers]

Description

Moves fetched-texture alpha classification and pick-mask generation off the frame-critical path when shared-context texture uploads are disabled.

The existing General worker pool computes the CPU-only upload metadata, then the main thread consumes it immediately before the GL upload. Explicit texture formats, unavailable queues, and shared-context uploads retain their existing paths. Prepared metadata is discarded if an upload becomes redundant.

A native Apple Silicon profile of the closely related Firestorm viewer placed LLViewerTextureList::updateImagesCreateTextures() on roughly 9% of main-thread samples, with LLImageGL::analyzeAlpha() and pick-mask generation running inside setImage(). The official develop branch contains the same code path.

Related Issues

• Please link to a relevant GitHub issue for additional context.

Issue Link: None; opened as a draft for upstream performance evaluation.

---

Checklist

• I have provided a clear title and detailed description for this pull request.
• If useful, I have included media such as screenshots and video to show off my changes.
• The PR is linked to a relevant issue with sufficient context.
• I have tested the changes locally and verified they work as intended.
• All new and existing tests pass.
• Code follows the project's style guidelines.
• Documentation has been updated if needed.
• Any dependent changes have been merged and published in downstream modules.
• I have reviewed the contributing guidelines.

---

Additional Notes

Validation performed:

• clean git diff --check
• exact stable patch-ID match with the Firestorm implementation
• code-path review confirming the worker fallback and stale-result behavior

A full official viewer build was not available because this host lacks the autobuild dependency bundle and full Xcode toolchain. This remains a draft pending native build and benchmark coverage.

[github-actions]

All contributors have signed the CLA ✍️ ✅
_{Posted by the CLA Assistant Lite bot.}

[LisaScheers]

I have read the CLA Document and I hereby sign the CLA

[Copilot]

Pull request overview

Note

Copilot was unable to run its full agentic suite in this review.

This PR introduces an asynchronous “upload preparation” step for textures, precomputing alpha-mask analysis and pick-mask generation off the main thread and applying the results during GL upload.

Changes:

• Added LLImageGL::TextureUploadPreparation and APIs to prepare/apply/discard upload precomputation state.
• Updated LLImageGL::setImage() to consume prepared results and skip redundant alpha/pick-mask work.
• Updated LLViewerFetchedTexture::scheduleCreateTexture() to enqueue background preparation for eligible textures.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 4 comments.

File	Description
indra/newview/llviewertexture.cpp	Schedules background upload preparation and applies it before enqueuing texture creation.
indra/llrender/llimagegl.h	Adds the upload-preparation struct, methods, and new members to store preparation state.
indra/llrender/llimagegl.cpp	Implements preparation logic, integrates it into setImage(), and refactors alpha mask analysis into a helper.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 5 out of 5 changed files in this pull request and generated 1 comment.

[akleshchev]

Thank you for the contribution! Merging.

[akleshchev]

I think you forgot mNeedsAlphaAndPickMask=false somewhere, a single image keeps restarting prepareForUpload job. Might also need a 'job created flag'.

Or forgot to hook up mAlphaAnalyzed and mPickMaskPrepared

[akleshchev]

Sorry, but I'm going to revert this. I checked how this works in more detail and not only it doesn't work right, this is a very focused uploads-only solution for a general, upload-independent problem.