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).
TGlslangToSpvTraverser::getExtBuiltins is only used when AMD_EXTENSIONS
is defined, so only define it in that case to avoid an unused function
warning.
- correct inheritence (or not) of the right XFB buffer
- compute implicit stride (fixes#1212)
- semantic check block-member redeclarations
- inherit stride from a member
- fixes#1209, addresses most of #1187
- only query feature availability on seeing the feature
(was doing it for every single token)
- correct case-sensitive checks for multi-character suffixes
- partially addresses #1209 and #1187
- only query 64-bit extension on seeing use of a 64-bit literal
(was doing it for every single token)
- correct HLSL acceptance of 64-bit literal syntax (still an int though)
- error on overflow of 32-bit literal type
The grammar for no semicolon and no object name for cbuffer/tbuffer
was correct, but the production still skipped the anonymous declarations
if an identifier followed.
Set type to r-value resulting from indexing vector, to prevent
float->uint conversion when source is already uint. Resulting
OpConvertFToU would otherwise fail validation because source is
already uint.
For LoadN, incorrect uint->float->uint can be avoided; fixing
potential truncation of big integer values.
Also, only emit this XFB information where the SPIR-V spec says
it should be emitted: essentially, on objects.
This and the previous commit together fix#1185.
In DX10/DX11 you can only output RT/Viewport indices from GS; however,
DX11.4/DX12 add support for outputting these from VS as well.
This is supported by Vulkan if the relevant extension is available, and
by MSL and by MSL (which you can cross-compile to via SPIRV-Cross).
Some stage (e.g, hull shaders) have arrayed builtin outputs (e.g, position).
When copying from the internal structure to the split form, it is necessary
to propagate that indirection to the actual arrayed outputs. This was not
happening.
Addresses #1181
