eglantine-cirakg/src/video_core/shader/decode/arithmetic_half.cpp, branch nx2 Mirror of github.com/cira0100/Eglantine-public nx2, a fork of Djihads80/Eglantine. https://git.cirakg.xyz/eglantine-cirakg/atom?h=nx2 2021-07-23T01:51:22+00:00 shader: Remove old shader management 2021-07-23T01:51:22+00:00 ReinUsesLisp reinuseslisp@airmail.cc 2021-02-16T23:52:12+00:00 urn:sha1:65069df8aa3975b055d70c3c615f9c99d60c72e0 video_core: Enforce -Werror=switch 2020-09-16T20:48:01+00:00 ReinUsesLisp reinuseslisp@airmail.cc 2020-09-16T20:48:01+00:00 urn:sha1:ee7e70cfbc5d60e507aaa19202ad8192dc0fa2a7 This forces us to fix all -Wswitch warnings in video_core. decode/arithmetic_half: Fix HADD2 and HMUL2 absolute and negation bits 2020-04-23T21:29:38+00:00 ReinUsesLisp reinuseslisp@airmail.cc 2020-04-20T06:24:00+00:00 urn:sha1:3e808936a86c795b1ed35858cf001869240bdd0b The encoding for negation and absolute value was wrong. Extracting is now done manually. Similar instructions having different encodings is the rule, not the exception. To keep sanity and readability I preferred to extract the desired bit manually. This is implemented against nxas: https://github.com/ReinUsesLisp/nxas/blob/8dbc38995711cc12206aa370145a3a02665fd989/table.h#L68 That is itself tested against nvdisasm (Nvidia's official disassembler). shader/decode: Reduce severity of arithmetic rounding warnings 2019-11-07T04:43:38+00:00 ReinUsesLisp reinuseslisp@airmail.cc 2019-11-04T05:24:58+00:00 urn:sha1:93d1d8713bfc99034e28ee91b06d4a88d8247dd1 shader: Use shared_ptr to store nodes and move initialization to file 2019-06-05T23:41:52+00:00 ReinUsesLisp reinuseslisp@airmail.cc 2019-06-05T01:44:06+00:00 urn:sha1:fd392543e80944f4ff4dba770b139573fa5e1b4e Instead of having a vector of unique_ptr stored in a vector and returning star pointers to this, use shared_ptr. While changing initialization code, move it to a separate file when possible. This is a first step to allow code analysis and node generation beyond the ShaderIR class. shader/decode/*: Eliminate indirect inclusions 2019-05-23T05:55:52+00:00 Lioncash mathew1800@gmail.com 2019-05-22T21:17:19+00:00 urn:sha1:bf42020f25c6fbff0c0b0db2f276fbc6934aea53 Amends cases where we were using things that were indirectly being satisfied through other headers. This way, if those headers change and eliminate dependencies on other headers in the future, we don't have cascading compilation errors. Corrections Half Float operations on const buffers and implement saturation. 2019-04-21T01:11:33+00:00 Fernando Sahmkow fsahmkow27@gmail.com 2019-04-20T03:10:19+00:00 urn:sha1:b17f7da309a826d56fec822d938bccfa7f29b75e shader_ir/decode: Fix half float pre-operations and remove MetaHalfArithmetic 2019-04-16T00:16:10+00:00 ReinUsesLisp reinuseslisp@airmail.cc 2019-04-15T22:48:11+00:00 urn:sha1:6ea1afc2bcde6de965c292499c3d353c66749717 Operations done before the main half float operation (like HAdd) were managing a packed value instead of the unpacked one. Adding an unpacked operation allows us to drop the per-operand MetaHalfArithmetic entry, simplifying the code overall. shader_ir/decode: Reduce severity of unimplemented half-float FTZ 2019-04-16T00:16:09+00:00 ReinUsesLisp reinuseslisp@airmail.cc 2019-04-09T20:34:11+00:00 urn:sha1:9c4449696a1810d2426969ed92f92dd6cd0bd5d9 shader_ir: Rename BasicBlock to NodeBlock 2019-02-03T20:21:20+00:00 ReinUsesLisp reinuseslisp@airmail.cc 2019-01-30T05:09:40+00:00 urn:sha1:92c948999b3c354891a22b230adbb1393f5084a3 It's not always used as a basic block. Rename it for consistency.