Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

AAA fixes #4973

Merged
merged 21 commits into from
Aug 18, 2018
Merged

AAA fixes #4973

merged 21 commits into from
Aug 18, 2018

Commits on Aug 18, 2018

  1. rsx: Synchronization rewritten 

    - Do not do a full sync on a texture read barrier
- Avoid calling zcull sync in FIFO spin wait
- Do not flush memory to cache from the renderer side; this method is now obsolete
    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    f5f40bd View commit details
    Browse the repository at this point in the history

  2. zcull synchronization tweaks 

    - Implement forced reading when calling update method to sync partial lists
- Defer conditional render evaluation and use a read barrier to avoid extra work
- Fix HLE gcm library when binding tiles & zcull RAM
    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    0f1a0b2 View commit details
    Browse the repository at this point in the history

  3. ZCULL: lower notice severity to avoid spam

    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    8b760b5 View commit details
    Browse the repository at this point in the history

  4. Minor optimizations and fixes 

    - FIFO: avoid multiline spam
- VK: Fix program setup counter
- FS: Precalculate fragment constants buffer size during analysis step
    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    75b4afa View commit details
    Browse the repository at this point in the history

  5. rsx: Refactor and fix framebuffer layout checks 

    - Refactors shared code back into rsx core
- Adds extra check to avoid contest confusion
    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    98f105a View commit details
    Browse the repository at this point in the history

  6. rsx: Optimize hash_struct; vk cleanup

    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    32bc2b4 View commit details
    Browse the repository at this point in the history

  7. rsx: Timestamp accuracy workaround

    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    bbd060f View commit details
    Browse the repository at this point in the history

  8. vk: Fixups for type b surfaces

    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    95b79c9 View commit details
    Browse the repository at this point in the history

  9. vulkan: Fix blit engine transfer to ARGB8 render target memory

    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    6c99e6d View commit details
    Browse the repository at this point in the history

  10. rsx: Allow linear filtering when reading back GPU-resident memory

    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    d60d1f4 View commit details
    Browse the repository at this point in the history

  11. rsx: Avoid acquiring the vm lock; deadlock evasion 

    - A possible deadlock is still present if rsx is trying to get a super_ptr whilst the vm lock holder is in an access violation
  This patch makes this scenario very unlikely since each block need only be touched once
    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    504c7fd View commit details
    Browse the repository at this point in the history

  12. zcull: Improve the delay algorithm to be more consistent 

    - Use proper time checking; depending on what is being done one 'tick' can
  be almost a millisecond long or several nanoseconds
- Avoid spamming the system timer unless necessary
    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    5f39672 View commit details
    Browse the repository at this point in the history

  13. gl: Do not create secondary context if async is disabled 

    - Some third party programs fall apart when multiple contexts are created
    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    6ab2353 View commit details
    Browse the repository at this point in the history

  14. [WIP] rsx: Improve memory inheritance hierachy 

    - Cascade memory writes by invalidating 'downstream' subsurfaces
- Fixup; always resolve for overlapping surfaces before sampling (force
  atlas gather test)
    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    5a860d2 View commit details
    Browse the repository at this point in the history

  15. rsx: Enable swizzled decode for all formats unless proven otherwise 

    - Some formats are proven to ignore swizzle flag
  - DXT compressed textures
  - COMPRESSED_BG_GB class textures
- Some applications are using swizzled wide integer formats so those are confirmed to swizzle
    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    29eea93 View commit details
    Browse the repository at this point in the history

  16. gl: Reuse framebuffer resources 

    - WIP optimizations for GL backend
    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    5818f16 View commit details
    Browse the repository at this point in the history

  17. gl: Avoid unnecessary scissor state change every draw call.

    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    9b81c68 View commit details
    Browse the repository at this point in the history

  18. rsx: Followup to the memory inheritance hierachy patch 

    - Tags framebuffer resources on first use (when on_write is called to verify memory)
- Texture cache now selects the best match and even sorts atlas writes with memory write order to avoid older data showing over newer one
    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    3b243be View commit details
    Browse the repository at this point in the history

  19. fix gcc build

    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    7cc1f42 View commit details
    Browse the repository at this point in the history

  20. rsx: Force disable draw reordering when capturing a frame

    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    69dbc0b View commit details
    Browse the repository at this point in the history

  21. rsx: Do not overflow the program buffer! 

    - Some games overflow the program buffer e.g Resistance games
  The observed overflow is one instruction longer, likely an engine bug
with counting instructions
    @kd-11
    kd-11 committed Aug 18, 2018
    Configuration menu
    Copy the full SHA
    ebcb3ed View commit details
    Browse the repository at this point in the history