rooty/dxvk
1
0
Fork 0
mirror of https://github.com/doitsujin/dxvk.git synced 2025-01-07 07:46:19 +01:00
Code Issues Projects Releases Wiki Activity
4,126 Commits 28 Branches 84 Tags 24 MiB
1d1d8adee6
Commit Graph

48 Commits

Author SHA1 Message Date
Robin Kertels
40444c1f50 [d3d9] Unify texture uploads 2022-02-21 00:08:41 +00:00
Robin Kertels
97ae14b6a0 [d3d9] Synchronize only to given sequence number in WaitForResources 2022-02-20 03:29:31 +00:00
Robin Kertels
917a8d00a2 [d3d9] Track last staging resource usage with a sequence number 2022-02-20 03:29:31 +00:00
Robin Kertels
0061139652 [d3d9] Copy shader defined constants into SWVP UBO 
Co-authored-by: Christopher Egert <cme3000@gmail.com>
Co-authored-by: Georg Lehmann <dadschoorse@gmail.com>
 2021-11-30 01:57:22 +00:00
Robin Kertels
530f87f158 [d3d9+dxso] Rely on robustness for unset HW VS & PS constants 2021-10-15 10:54:18 +01:00
Robin Kertels
791d533a1d [d3d9+dxso] Rely on robustness for unset SWVP constants 2021-10-15 10:54:18 +01:00
Georg Lehmann
82d40f5bba [d3d9] Return a pointer in GetUnsupportedFormatInfo 2021-08-23 17:38:50 +01:00
Joshua Ashton
3718cee9eb [d3d9] Clean up texture type tracking 
Additionally stops leaky pipelines on PS 1.x shaders from partial mask updates.
 2021-08-17 07:02:56 -07:00
Joshua Ashton
64852f05c3 [d3d9] Track fixed function sampler bitfields 
Cleans some stuff up a little bit and makes FF draws faster when this gets incorrectly invalidated.
 2021-08-17 07:02:56 -07:00
Joshua Ashton
c6a4c41d20 [d3d9] Batch texture unbinding 2021-08-17 07:02:56 -07:00
Joshua Ashton
ab7fc42f2f
[d3d9] Fix stateblock refcounting 2021-08-11 01:38:50 +01:00
Joshua Ashton
a1f0bb99ee
[d3d9] Cleanup D3D9DeviceEx members 2021-08-11 01:38:34 +01:00
Joshua Ashton
5e48035e99
[d3d9] Only bind textures if they are used by the shader 2021-08-11 01:08:13 +01:00
Joshua Ashton
c0c1565cba [d3d9, dxso] Alias color and depth samplers and improve tracking 
Takes me from 340 -> 460fps in A Hat in Time's main menu when CPU bound.
 2021-08-10 23:46:03 +00:00
Georg Lehmann
09e5939502 [d3d9] Reject 2 BeginScene + EndScene without Begin 2021-08-08 08:50:00 +00:00
Georg Lehmann
64c4eea922 [d3d9] Mark texture dirty on SetLod 2021-08-08 02:28:44 +00:00
Joshua Ashton
072d3a039e [d3d9] Track bound RTs 
This also improves active RT hazard tracking by also accounting for NULL
 2021-08-08 02:14:11 +00:00
Joshua Ashton
d3112c320b [d3d9] Add dirty texture tracking 
Reduces overhead from re-binding and unnecessary binding (srgb changes) in L4D2.

I go from about 750 -> 850-900 fps in c1m2_streets with this change.
 2021-07-14 20:38:49 +00:00
Philip Rebohle
96f5641a7e [d3d9] Ensure that the bound UP vertex buffer region is large enough 
If the buffer size is less than (vertexCount * stride), the last vertex
may be considered out-of-bounds, even if all attributes are included in
the vertex.

Fixes #2131.
 2021-06-29 18:35:47 -07:00
Joshua Ashton
0f52c85d21 [d3d9] Account for vertex declaration size for UP draws 
The stride may not give us the full picture here as the stride
may not encompass the vertex declaration entirely.

Consider a vertex declaration of size 20, and a stride of 12,
we may not have covered the whole range of space the draw wants with
VertexCount * Stride.

Some games such as FF13 Lightning Returns have two float3s in the vertex decl
and draw two triangles with the last float being out of bounds. This causes
the whole vertex element to be set to 0 on NVIDIA which breaks their fullscreen passes.

Instead, take (VertexCount - 1) * Stride + VertexDeclSize for the buffer size
and pad with 0s outside of the VertexCount * Stride range.

Closes: #2046
Closes: #1908
 2021-05-05 19:25:46 +01:00
Robin Kertels
bb11d7dee8 [d3d9] Refactor AllocUpBuffer so it can be used for managed uploads 2021-04-22 17:44:58 +01:00
Georg Lehmann
1ed6edf096 [d3d9] respect Vector4 alignment in GetShaderConstants 2021-04-15 22:25:11 +01:00
Joshua Ashton
9525f53d00 [d3d9] Only enable FETCH4 when POINT sampling 
Closes #1660
 2020-06-14 00:38:36 +01:00
Joshua Ashton
e6ed8dab63 [d3d9] Perform tracking for preloaded managed resources 2020-06-11 05:53:33 +01:00
Joshua Ashton
b1bd3597a4 [d3d9] Implement PreLoad for buffers + textures 2020-06-11 05:53:33 +01:00
Joshua Ashton
1a920f44ea [d3d9] Rename GenerateMips to EmitGenerateMips 2020-05-26 18:50:34 +01:00
Joshua Ashton
6c030afc95 [d3d9] Update device tracking when manually autogenning mips 
Avoids some needless generations potentially.
 2020-05-26 18:48:42 +01:00
Joshua Ashton
e57aea5749 [d3d9] Track and defer mipmap generations 
Turns out we need to track this crap and generate at draw time and handle a bunch of other wacky cases for it...

Bah.

Fixes #1642
(The funky water.)
 2020-05-26 18:42:53 +01:00
Joshua Ashton
8fdf9e67d3 [d3d9, dxso] Implement FETCH4 2020-05-25 06:13:21 +01:00
Joshua Ashton
6e5e50c359 [d3d9] Optimize hazard tracking in the SetTexture case 
We don't need to perform DS/RT hazard tracking updates if the texture we replaced and ourselves do not have those usages.
 2020-04-09 15:29:41 +01:00
Joshua Ashton
9b4cd8aa87 [d3d9] Remove texMask from UpdateActiveHazardsRT 
This is an invalid optimization and leads to render hazards being dropped

Our hazard tracking for render targets is done in render target indices as opposed to texture indices for depth stencil tracking so `texMask` doesn't work for that reason, and the fact that even if it did, there is no relationship to an individual texture and a render target index that has a hazard.
 2020-04-09 15:29:41 +01:00
Joshua Ashton
e33627cfdc [d3d9] Optimize unnecessary loops in hazard tracking 
We only want to update specific indices of bitfields at a time, so pass through that information and mask it off.
 2020-03-09 01:03:18 +00:00
Joshua Ashton
f6b26b302d [d3d9] Handle depth stencil hazards 
Track depth stencil textures being used and whether we have any active hazards.

Rebind the framebuffer with VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL if we encounter a depth stencil hazard.

Fixes black squares in Mass Effect and validation errors in some titles.

Closes #1484
 2020-03-09 00:53:19 +00:00
Joshua Ashton
a432befa8d [d3d9] Add RT suffix to current hazard tracking 
We can have DS hazards...
 2020-03-09 00:23:22 +00:00
Joshua Ashton
722520a9f7 [d3d9] Don't rebind user's vertex buffer/indices in DrawPrimitive[Indexed]UP 
D3D9 doesn't do this, it instead sets them to NULL so we can simplify this logic a fair bit.

Found via a Wine test.
 2020-02-28 21:37:00 +00:00
Joshua Ashton
528cacca5f [d3d9] Only have a single implicit swapchain 
I am naive to think that this API is in any way sane, and that iSwapChains actually does what you would expect.

Swapchains created by CreateAdditionalSwapchain[Ex] are not tracked in that figure... There can only ever be one.

Great API. Very cool.

This also adds and consolidates some missing thread synchronization on the swapchain functions (that should have already had it).

(note: m_implicitSwapchain always exists, so we don't need to lock if that's all we call.)
 2020-02-28 01:29:38 +00:00
Joshua Ashton
2408b881a2 [dxso] Use SSBOs for vertex constant buffers if we'd be outside of uniform range 
Impacts #1375
 2020-02-20 08:14:58 +01:00
Joshua Ashton
512393e469 [d3d9] Refactor constant buffer creation 2020-02-20 08:14:58 +01:00
Joshua Ashton
7d3ec74b40 [d3d9] Don't perform clipping in FF if disabled 
Avoids unnecessary matrix multiplications in the shader, given this isn't as cheap as it is for programmable.
 2020-02-14 19:19:22 +00:00
Joshua Ashton
ae68e3a5bc [d3d9] Defer managed texture uploads until PrepareDraw and when needed 
This also caches shader masks used for hazard tracking.
 2020-02-14 00:51:58 +00:00
Joshua Ashton
acb3cda1dc [d3d9] Hold a private reference to FVF-generated vertex decls 
Otherwise the device will never be able to be destroyed as there will be a circular dependency.

Closes #1417
 2020-02-13 18:42:49 +00:00
Joshua Ashton
720cdf383e [d3d9] Use spec constants for bool constants 2020-01-27 14:21:21 +01:00
Joshua Ashton
7d0ddc4b3b [d3d9] Remap texture stage state types onto our own enum 
Fits us nicely into a dword for captures while not exclusing D3DTSS_CONSTANT
 2020-01-23 02:05:09 +00:00
Joshua Ashton
b738c4220b [d3d9] Scale depth bias value based on current format's r value 2020-01-08 19:38:18 +00:00
Joshua Ashton
2d7f4b1a2c [d3d9] Set depth bias values separately, optimize redundant rasterizer state changes 2020-01-08 19:38:18 +00:00
Joshua Ashton
90ce37c8c9 [d3d9] Make ChangeReportedMemory actually atomic 
Massive edge case that will never happen, but better to be safe than sorry.
Reduces the atomic ops too.
 2019-12-20 18:07:35 +00:00
Joshua Ashton
009e772fe8 [d3d9] Remove initial device reset outside of constructor 
Allows us to funnel hresults from that to the response of CreateDevice
 2019-12-17 23:59:37 +00:00
Joshie
54ed8f0bb0 [d3d9] Implement Direct3D9 Frontend (#1275) 
Co-authored-by: Philip Rebohle <philip.rebohle@tu-dortmund.de>
Co-authored-by: Robin Kertels <robin.kertels@gmail.com>
Co-authored-by: pchome <pchome@users.noreply.github.com>
Co-authored-by: Christopher Egert <cme3000@gmail.com>
Co-authored-by: Derek Lesho <dereklesho52@Gmail.com>
Co-authored-by: Luis Cáceres <lacaceres97@gmail.com>
Co-authored-by: Nelson Chen <crazysim@gmail.com>
Co-authored-by: Edmondo Tommasina <edmondo.tommasina@gmail.com>
Co-authored-by: Riesi <riesi@opentrash.com>
Co-authored-by: gbMichelle <gbmichelle.dev@gmail.com>
 2019-12-16 04:28:01 +01:00