shader_recompiler: Add lowering pass for when 64-bit float is unsupported. (#2858)

* shader_recompiler: Add lowering pass for when 64-bit float is unsupported.

* shader_recompiler: Fix PackDouble2x32/UnpackDouble2x32 type.

* shader_recompiler: Remove extra bit cast implementations.
This commit is contained in:
squidbus 2025-04-28 00:04:16 -07:00 committed by GitHub
parent 385c5a4507
commit 81fa9b7fff
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
14 changed files with 220 additions and 33 deletions

View file

@ -181,7 +181,8 @@ public:
[[nodiscard]] U64 PackUint2x32(const Value& vector);
[[nodiscard]] Value UnpackUint2x32(const U64& value);
[[nodiscard]] F64 PackFloat2x32(const Value& vector);
[[nodiscard]] F64 PackDouble2x32(const Value& vector);
[[nodiscard]] Value UnpackDouble2x32(const F64& value);
[[nodiscard]] U32 Pack2x16(AmdGpu::NumberFormat number_format, const Value& vector);
[[nodiscard]] Value Unpack2x16(AmdGpu::NumberFormat number_format, const U32& value);