260 Commits

Author SHA1 Message Date
John Kessenich
377bccb143 Build: Fix #1665: remove __fastcall 2019-02-07 23:28:37 -07:00
baldurk
657acc0c40 Add option to reflect buffer blocks & variables separately to uniforms
* Also note the uniform indices of atomic counter buffers
2019-02-04 12:02:59 +00:00
baldurk
0af5e3e346 Reflect pipeline outputs as well as inputs, optionally from other stages
* We add an option to reflect inputs from other stages than vertex, if only a
  later subset of the stages is linked into the program.
2019-02-04 12:02:59 +00:00
baldurk
6d47785825 Add options to control how reflection information is built 2019-02-04 11:21:09 +00:00
baldurk
1dc5dcf0a5 Move TObjectReflection into public interface to clean up reflection
* Forwarding functions are left to preserve source compatibility with code using
  the old queries.
2019-02-04 11:21:07 +00:00
dan sinclair
e96fa717d3 Make sure source strings are terminated (#1641)
* Make sure source strings are terminated

The source strings may or may not have a null terminator. We need to
make sure we add one before outputting the source strings as we iterate
over the c-str looking for the null terminator.

* Review feedback
2019-01-07 23:47:37 +07:00
John Kessenich
b23d232ec5 Licensing. Fixes #958. Add licenes file and update copyrights. 2018-12-14 10:47:35 -07:00
Rémi Verschelde
ebfd91a719 Remove Unix executable permission from text files 2018-12-10 10:49:16 +01:00
Ashwin Lele
ff1783db87 Multiple features and misc fixes for final raytracing non experimental version. 2018-10-31 07:55:24 -07:00
Neil Roberts
b0f3d794c8 Add an option to set the base uniform location
This will be used to generate uniform locations with --aml
2018-10-02 12:15:22 +02:00
Neil Roberts
16f53474c8 Add a command-line option to override uniform locations 2018-10-02 12:15:22 +02:00
John Kessenich
4508a8170a
Merge pull request #1507 from sparmarNV/fix-SPV_NV_mesh_shader
Add ES 320 support and additional error checks for SPV_NV_mesh_shader
2018-09-28 06:24:02 -06:00
Sahil Parmar
95e2d4ec02 Add ES 320 support and additional error checks for SPV_NV_mesh_shader
- Add ES 320 support
- Error out use of perprimitiveNV for non mesh/fragment shaders
- Error out use of mesh/task shaders w/o use of NV_mesh_shader
- Error out use of NV_mesh_shader for non task/mesh shaders
- Error out use of perviewNV for non mesh shaders
- Error out use of taskNV for non mesh/task shaders
- Add test case for mesh shader with ES 320 profile
2018-09-25 13:45:32 -07:00
Ashwin Lele
a535bc1353 Fix missing case label for miss stage. 2018-09-25 13:19:34 -07:00
Chao Chen
b50c02ef53 Add-support-for-SPV_NVX_raytracing 2018-09-19 13:07:43 -07:00
Chao Chen
3c3669904c Add-support-for-SPV_NV_mesh_shader 2018-09-19 13:07:43 -07:00
John Kessenich
546b78854a Printing preprocessed shaders (not a supported path): Fix #1490: strings
Put quote marks around strings, due to a change in how tokenization works.
2018-09-10 11:42:16 -06:00
dmpakas
f556e5da26 Reflection exposes the Shader Stages where a Uniform is present 2018-07-10 18:25:48 +03:00
Cory Bloor
39bbad5a00 Minor cleanup in ShaderLang.cpp
Use unique_ptr to simplify memory management in ProcessDeferred.
2018-04-09 16:18:37 -06:00
Matthew Albrecht
2199356979 Added getter for uniform block bindings 2018-03-30 09:58:01 -05:00
John Kessenich
5d610ee1dc Implement SPV_GOOGLE_hlsl_functionality1.
Enabled via -fhlsl_functionality1
2018-03-16 20:21:44 -06:00
siavashserver
b502a802f4
Fix memory leak upon ProcessDeferred failure. 2018-03-08 09:13:36 +03:30
John Kessenich
66011cb2c2 SPV: Implement Vulkan 1.1 features and extensions. 2018-03-06 16:12:04 -07:00
John Kessenich
c6c80a6e48 Versioning: Address #1255: Move to semantic versioning.
Still missing is automation and final decisions on mechanisms in general.
2018-03-05 22:23:17 -07:00
John Kessenich
b5b5f918c6 Non-functional: Make the vulkan version number meaning be only 0 or not 0. 2018-03-02 18:17:42 -07:00
Arseny Kapoulkine
75cffdf92c Preprocessor: Use std::string instead of std::stringstream
std::stringstream has a measurable overhead for preprocessing - it
appears that operator<< does a tiny bit of extra work for appending
chars/strings and also can't be inlined in most cases on VS2015;
additionally, std::endl triggers a stream flush which also adds up.

Replacing this with std::string buffer gets the preprocessing time down
from 180ms to 135ms in one case, making it 1.33x faster.

Note that integer-to-string conversion is using std::to_string; in
theory this could be slower than sprintf or manual conversion, but I
haven't found these cases to affect preprocessing time in practice
(std::to_string would always use the short string buffer for
line/version numbers, and the number of calls is not too significant).
2018-02-05 16:07:39 -08:00
LoopDawg
b22c069f7a HLSL: add optional position.Y inversion
Adds command line options:

   --invert-y
   --iy

(synonyms) which invert position.Y on vertex shader output.  Handles these cases:

* Direct single variable return
* Member of direct returned struct
* Single variable output parameter
* Member of struct output parameter

API:

    // Enables position.Y output negation in vertex shader
    void TShader::setInvertY(bool invert);

Fixes #1173
2017-12-07 10:48:14 -07:00
John Kessenich
e2c15b4f6e Memory: Reference count number of clients, to support multiple independent clients in same process.
Addresses 2nd item in #976.
2017-11-16 23:02:14 -07:00
John Kessenich
5da8ead703 Memory: Remove redundant pool popAll(), which is potentially confusing. 2017-11-14 15:19:41 -07:00
John Kessenich
cb42541e51 Memory: Remove the need for per-thread tear down.
Make key objects using the memory pool own their own pool and delete it,
such that there is not generic per-thread pool to manage.
2017-11-13 01:52:11 -07:00
John Kessenich
ff8e59f510 Memory: Do process and 1st thread tear down.
Addresses #928, #389, and 1st item in #976. Overlaps #916.

This had been dropped, when moving away from the old Win32 DLL model.

Issue: per-thread tear down for other threads.
2017-11-12 15:55:22 -07:00
John Kessenich
4ceaab166c Memory: Move to a normal model of ownership of memory pools, for new/delete.
Addresses step 4 of #976, overlaps #916.

For each pool, now, it is newed, remembered, and freed by the same entity,
rather than having a mix (thread finalize freeing current pool) that could
lead to double freeing of the same pool. It is quite rational and simple now.

This will enable reinstalling process and thread tear down.
2017-11-12 15:46:40 -07:00
John Kessenich
be20905582 Memory: Non-Functional: Rationalize and improve encapsulation of TLS usage.
This will make the next (functional) commit easier to see.
2017-11-12 15:28:58 -07:00
LoopDawg
e570955cdc Reverse order of setShiftBindingForSet parameters
Per feedback on PR #1111, this reverses the order of the parameters for the setShiftBinding API.

It is now:

    void TShader::setShiftBindingForSet(TResourceType res, unsigned int base, unsigned int set);
2017-11-09 09:51:15 -07:00
John Kessenich
a372a3ed1e Versioning: Update some version strings. Still need better overall version. 2017-11-02 22:34:43 -06:00
LoopDawg
08a14422c1 Add per-descriptor-set IO mapping shift values.
This PR adds the ability to provide per-descriptor-set IO mapping shift
values.  If a particular binding does not land into a per-set value,
then it falls back to the prior behavior (global shifts per resource class).

Because there were already 6 copies of many different methods and internal
variables and functions, and this PR would have added 6 more, a new API is
introduced to cut down on replication and present a cleaner interface.
For the global (non-set-specific) API, the old entry points still exist
for backward compatibility, but are phrased internally in terms of the
following.

    // Resource type for IO resolver
    enum TResourceType {
        EResSampler,
        EResTexture,
        EResImage,
        EResUbo,
        EResSsbo,
        EResUav,
        EResCount
    };

Methods on TShader:

    void setShiftBinding(TResourceType res, unsigned int base);
    void setShiftBindingForSet(TResourceType res, unsigned int set, unsigned int base);

The first method replaces the 6 prior entry points of various spellings, which
exist now in depreciated form.  The second provides per-resource-set functionality.
Both accept an enum from the list above.

From the command line, the existing options can accept either a single shift value as
before, or a series of 1 or more [set offset] pairs.  Both can be provided, as in:

    ... --stb 20 --stb 2 25 3 30 ...

which will use the offset 20 for anything except descriptor set 2 (which uses 25) and
3 (which uses 30).
2017-10-19 11:50:30 -06:00
John Kessenich
9855bdad00 GLSL: Promote HLSL entry-point renaming code to be used by GLSL as well.
Fixes #1045.
2017-09-12 09:40:54 -06:00
John Kessenich
2a27116cae SPV reflection: Add OpModuleProcessed for compile options. 2017-08-25 11:48:02 -06:00
John Kessenich
de16e52b25 GLSL: Initiate version GLSL 460, including accept extraneous semicolons. 2017-07-31 03:00:04 -06:00
John Kessenich
53863a3a90 GLSL: Implement version 320 for ES. 2017-07-23 13:54:15 -06:00
John Kessenich
9353f1afab GLSL: Add version-number checking. 2017-07-23 11:49:42 -06:00
John Kessenich
a2b71902e2 Reflection: Fix #977: Expose getBinding(), use in new getUniformBinding(). 2017-07-20 16:44:17 -06:00
John Kessenich
4be4aebdcd Infrastructure: Non-functional: Move to rich description of environment.
This is for input languages, client APIs, code to generate, etc.
2017-06-26 14:49:11 -06:00
John Kessenich
a931366f56 Standalone: Implement -D and -U for preprocessor macros.
Works for both GLSL and HLSL.
Fixes #87.
2017-06-15 10:57:18 -06:00
John Kessenich
e00e8f45a6 GLSL: replace general missing functionality with specific messages. 2017-06-05 16:42:33 -06:00
John Kessenich
3494b4da9b HLSL: Add an Includer to handle #include. 2017-06-01 11:16:32 -06:00
John Kessenich
121853f4df SPV: Add OpSource shader source code and file name. 2017-05-31 17:14:15 -06:00
t.jung
baf570efa5 Pure Texture to Sampled Texture Transform
Adds a transformation step to the post processing step.
Two modes are available:
1) keep
- Keeps samplers, textures and sampled textures as is
2) transform pure texture into sampled texture and remove pure samplers
- removes all pure samplers
- transforms all pure textures into its sampled counter part

Change-Id: If54972e8052961db66c23f4b7e719d363cf6edbd
2017-05-29 18:29:45 +02:00
John Kessenich
71facdf435 SPV: Give error on not assigning locations to I/O.
Also, provides an option to auto-assign locations.
Existing tests use this option, to avoid the error message,
however, it is not fully implemented yet.
2017-05-18 15:07:05 -06:00
Hyangran Park
36dc82908f HLSL: Manually configure descriptor set and binding number for resources 2017-05-10 16:50:18 +09:00