Skip to content

RADV: pin the VideoOut swapchain-size change and record its console results - #1

Closed
mpereiraesaa wants to merge 1 commit into
mihawk-99:mainfrom
mpereiraesaa:videoout-swapchain-sizes
Closed

mpereiraesaa wants to merge 1 commit into
mihawk-99:mainfrom
mpereiraesaa:videoout-swapchain-sizes

Conversation

@mpereiraesaa

Copy link
Copy Markdown

Summary

  • Goes with PS5_Mesa#1 (VideoOut swapchains smaller than the 3840x2160 mode, scaled by VideoOut).
  • tools/build-radv.sh pins that PR's head, 5d9f448. If you merge PS5_Mesa#1 by squash or rebase, the pin needs your resulting commit instead; with a merge commit, 5d9f448 stays reachable from main.
  • docs/HARDWARE_FINDINGS.md gets a 2026-09-29 finding (a 1080p framebuffer set fills the 4K output; a second set registers in the same process), and the 2026-09-27 VideoOut entry points to it. docs/RADV_PHASE.md gets the console results. The README row and docs/VULKAN_PROBE_ACTIVE.md mention the change.

Test plan

  • tools/build-radv.sh release builds that PS5_Mesa change (PROVENANCE revision recorded); bash -n tools/build-radv.sh
  • Console results as recorded in RADV_PHASE.md (FW 12.02, through Prospero Win, seen on the TV)
  • Debug variant and CTS not rerun for this change

🤖 Generated with Claude Code

https://claude.ai/code/session_01MPrDs8zJKK7tKFGsQdvVHQ

…esults

tools/build-radv.sh pins PS5_Mesa 5d9f448: VideoOut registers a
framebuffer set per swapchain size, so a 1920x1080 swapchain fills the
3840x2160 output instead of its top-left quarter. HARDWARE_FINDINGS.md and
RADV_PHASE.md record what the console showed; the README and the active
probe notes mention it.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01MPrDs8zJKK7tKFGsQdvVHQ
@mihawk-99 mihawk-99 closed this Oct 1, 2026
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.

2 participants