fabi/Xenia-Canary
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

atomic cas use prefetchw if available

Browse Source 
remove useless memorybarrier
remove double membarrier in wait pm4 cmd
add int64 cvar
use int64 cvar for x64 feature mask
Rework some functions that were frontend bound according to vtune placing some of their code in different noinline functions, profiling after indicating l1 cache misses decreased and perf of func increased
remove long vpinsrd dep chain code for conversion.h, instead do normal load+bswap or movbe if avail
Much faster entry table via split_map, code size could be improved though
GetResolveInfo was very large and had impact on icache, mark callees as noinline + msvc pragma optimize small
use log2 shifts instead of integer divides in memory
minor optimizations in PhysicalHeap::EnableAccessCallbacks, the majority of time in the function is spent looping, NOT calling Protect! Someone should optimize this function and rework the algo completely
remove wonky scheduling log message, it was spammy and unhelpful
lock count was unnecessary for criticalsection mutex, criticalsection is already a recursive mutex
brief notes i gotta run
This commit is contained in:
chss95cs@gmail.com
2022-09-17 04:04:53 -07:00
parent 0fd4a2533b
commit eb8154908c
35 changed files with 942 additions and 738 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
src/xenia/gpu/primitive_processor.h
View File

@@ -430,7 +430,7 @@ class PrimitiveProcessor {
--count;
uint32_t index = *(source++) & low_bits_mask_guest_endian;
*(dest++) = index != reset_index_guest_endian
? xenos::GpuSwap(index, HostSwap)
? xenos::GpuSwapInline(index, HostSwap)
: UINT32_MAX;
}
if (count >= kSimdVectorU32Elements) {
@@ -442,10 +442,10 @@ class PrimitiveProcessor {
__m128i host_swap_shuffle;
if constexpr (HostSwap != xenos::Endian::kNone) {
host_swap_shuffle = _mm_set_epi32(
int32_t(xenos::GpuSwap(uint32_t(0x0F0E0D0C), HostSwap)),
int32_t(xenos::GpuSwap(uint32_t(0x0B0A0908), HostSwap)),
int32_t(xenos::GpuSwap(uint32_t(0x07060504), HostSwap)),
int32_t(xenos::GpuSwap(uint32_t(0x03020100), HostSwap)));
int32_t(xenos::GpuSwapInline(uint32_t(0x0F0E0D0C), HostSwap)),
int32_t(xenos::GpuSwapInline(uint32_t(0x0B0A0908), HostSwap)),
int32_t(xenos::GpuSwapInline(uint32_t(0x07060504), HostSwap)),
int32_t(xenos::GpuSwapInline(uint32_t(0x03020100), HostSwap)));
}
#endif // XE_ARCH_AMD64
while (count >= kSimdVectorU32Elements) {
@@ -490,7 +490,7 @@ class PrimitiveProcessor {
while (count--) {
uint32_t index = *(source++) & low_bits_mask_guest_endian;
*(dest++) = index != reset_index_guest_endian
? xenos::GpuSwap(index, HostSwap)
? xenos::GpuSwapInline(index, HostSwap)
: UINT32_MAX;
}
}
@@ -510,19 +510,19 @@ class PrimitiveProcessor {
};
struct To24Swapping8In16IndexTransform {
uint32_t operator()(uint32_t index) const {
return xenos::GpuSwap(index, xenos::Endian::k8in16) &
return xenos::GpuSwapInline(index, xenos::Endian::k8in16) &
xenos::kVertexIndexMask;
}
};
struct To24Swapping8In32IndexTransform {
uint32_t operator()(uint32_t index) const {
return xenos::GpuSwap(index, xenos::Endian::k8in32) &
return xenos::GpuSwapInline(index, xenos::Endian::k8in32) &
xenos::kVertexIndexMask;
}
};
struct To24Swapping16In32IndexTransform {
uint32_t operator()(uint32_t index) const {
return xenos::GpuSwap(index, xenos::Endian::k16in32) &
return xenos::GpuSwapInline(index, xenos::Endian::k16in32) &
xenos::kVertexIndexMask;
}
};