mirror/dusklight
1
0
Fork 0
mirror of https://github.com/TwilitRealm/dusklight synced 2026-10-28 02:16:21 -04:00
Code Issues Packages Projects Releases Wiki Activity

Use std::byteswap from C++23 instead of compiler intrinsincs for endian swapping.

Browse Source 
Just cleanup.

fvb.cpp changes were needed since the enums weren't accepted and previously were getting int casted implicitly.
This commit is contained in:
Pieter-Jan Briers
2026-10-07 20:45:36 +02:00
parent 593ecd0591
commit 29eee7ad33
2 changed files with 7 additions and 62 deletions
Download Patch File Download Diff File Expand all files Collapse all files
include/helpers/endian.h
+3 -58
View File
@@ -1,7 +1,6 @@
#pragma once
#include <bit>
#include <type_traits>
#include "dolphin/types.h"
#include "dolphin/mtx.h"
@@ -14,87 +13,33 @@
#endif
#if TARGET_LITTLE_ENDIAN
#ifdef _MSC_VER
#include <stdlib.h>
#define BSWAP16(x) _byteswap_ushort(x)
#define BSWAP32(x) _byteswap_ulong(x)
#define BSWAP64(x) _byteswap_uint64(x)
#else
#define BSWAP16(x) __builtin_bswap16(x)
#define BSWAP32(x) __builtin_bswap32(x)
#define BSWAP64(x) __builtin_bswap64(x)
#endif
#define BSWAP16(x) std::byteswap(x)
#define BSWAP32(x) std::byteswap(x)
#define BSWAP64(x) std::byteswap(x)
#else
#define BSWAP16(x) (x)
#define BSWAP32(x) (x)
#endif
// clang-format off
// yeah ofc Microsoft's _byteswap_* aren't constexpr like the GCC/Clang ones.
constexpr u16 be16_manual(u16 val)
{
return (val >> 8) |
(val << 8);
}
constexpr u32 be32_manual(u32 val)
{
return (val >> 24) |
((val << 8) & 0x00FF0000) |
((val >> 8) & 0x0000FF00) |
(val << 24);
}
constexpr u64 be64_manual(u64 val)
{
return (val >> 56) |
((val << 40) & 0x00FF000000000000) |
((val << 24) & 0x0000FF0000000000) |
((val << 8) & 0x000000FF00000000) |
((val >> 8) & 0x00000000FF000000) |
((val >> 24) & 0x0000000000FF0000) |
((val >> 40) & 0x000000000000FF00) |
(val << 56);
}
// clang-format on
// Big-Endian to Host conversion
// _manual is behind std::is_constant_evaluated() checks,
// to avoid pessimizing debug perf too much.
constexpr u16 be16(u16 val) {
if (std::is_constant_evaluated()) {
return be16_manual(val);
}
return BSWAP16(val);
}
constexpr s16 be16s(s16 val) {
if (std::is_constant_evaluated()) {
return (s16)be16_manual((u16)val);
}
return (s16)BSWAP16((u16)val);
}
constexpr u32 be32(u32 val) {
if (std::is_constant_evaluated()) {
return be32_manual(val);
}
return BSWAP32(val);
}
constexpr s32 be32s(s32 val) {
if (std::is_constant_evaluated()) {
return (s32)be32_manual((u32)val);
}
return (s32)BSWAP32((u32)val);
}
constexpr u64 be64(u64 val) {
if (std::is_constant_evaluated()) {
return be64_manual(val);
}
return BSWAP64(val);
}
constexpr s64 be64s(s64 val) {
if (std::is_constant_evaluated()) {
return (s64)be64_manual((u64)val);
}
return (s64)BSWAP64((u64)val);
}
libs/JSystem/src/JStudio/JStudio/fvb.cpp
+4 -4
View File
@@ -108,7 +108,7 @@ void TObject::prepare(data::TParse_TBlock const& rBlock, TControl* pControl) {
break;
}
TFunctionValue::TEProgress prog = (TFunctionValue::TEProgress)BSWAP32(*(TFunctionValue::TEProgress*)pContent);
TFunctionValue::TEProgress prog = (TFunctionValue::TEProgress)BSWAP32(*(u32*)pContent);
pfvaRange->range_setProgress(prog);
} break;
case 0x14: {
@@ -121,7 +121,7 @@ void TObject::prepare(data::TParse_TBlock const& rBlock, TControl* pControl) {
break;
}
TFunctionValue::TEAdjust adjust = (TFunctionValue::TEAdjust)BSWAP32(*(TFunctionValue::TEAdjust*)pContent);
TFunctionValue::TEAdjust adjust = (TFunctionValue::TEAdjust)BSWAP32(*(u32*)pContent);
pfvaRange->range_setAdjust(adjust);
} break;
case 0x15: {
@@ -148,7 +148,7 @@ void TObject::prepare(data::TParse_TBlock const& rBlock, TControl* pControl) {
break;
}
TFunctionValue::TEInterpolate interp = (TFunctionValue::TEInterpolate)BSWAP32(*(TFunctionValue::TEInterpolate*)pContent);
TFunctionValue::TEInterpolate interp = (TFunctionValue::TEInterpolate)BSWAP32(*(u32*)pContent);
pfvaInterpolate->interpolate_set(interp);
} break;
default:
@@ -224,7 +224,7 @@ void TObject_composite::prepare_data_(const data::TParse_TParagraph::TData& rDat
JGADGET_ASSERTWARN(310, u32Size== 8);
JUT_ASSERT(311, pContent!=NULL);
data::TEComposite type = (data::TEComposite)BSWAP32(pContent->composite_type);
data::TEComposite type = (data::TEComposite)BSWAP32((u32)pContent->composite_type);
const data::CompositeOperation* op = getCompositeOperation_(type);
data::CompositeDataFunc pfn = op->getCompositeData;
JUT_ASSERT(316, pfn!=NULL);