Buffer_Cache: Optimize and track written areas.

This commit is contained in:
Fernando Sahmkow 2019-07-20 11:42:53 -04:00 committed by FernandoS27
parent 5f4b746a1e
commit 286f4c446a
2 changed files with 105 additions and 13 deletions

View file

@ -43,8 +43,24 @@ public:
}
const auto cache_addr = ToCacheAddr(host_ptr);
// Cache management is a big overhead, so only cache entries with a given size.
// TODO: Figure out which size is the best for given games.
constexpr std::size_t max_stream_size = 0x800;
if (size < max_stream_size) {
if (!is_written && !IsRegionWritten(cache_addr, cache_addr + size - 1)) {
return StreamBufferUpload(host_ptr, size, alignment);
}
}
auto block = GetBlock(cache_addr, size);
MapAddress(block, gpu_addr, cache_addr, size);
auto map = MapAddress(block, gpu_addr, cache_addr, size);
if (is_written) {
map->MarkAsModified(true, GetModifiedTicks());
if (!map->IsWritten()) {
map->MarkAsWritten(true);
MarkRegionAsWritten(map->GetStart(), map->GetEnd() - 1);
}
}
const u64 offset = static_cast<u64>(block->GetOffset(cache_addr));
@ -124,7 +140,7 @@ protected:
std::size_t dst_offset, std::size_t size) = 0;
/// Register an object into the cache
void Register(const MapInterval& new_map) {
void Register(const MapInterval& new_map, bool inherit_written = false) {
const CacheAddr cache_ptr = new_map->GetStart();
const std::optional<VAddr> cpu_addr =
system.GPU().MemoryManager().GpuToCpuAddress(new_map->GetGpuAddress());
@ -139,6 +155,10 @@ protected:
const IntervalType interval{new_map->GetStart(), new_map->GetEnd()};
mapped_addresses.insert({interval, new_map});
rasterizer.UpdatePagesCachedCount(*cpu_addr, size, 1);
if (inherit_written) {
MarkRegionAsWritten(new_map->GetStart(), new_map->GetEnd() - 1);
new_map->MarkAsWritten(true);
}
}
/// Unregisters an object from the cache
@ -146,6 +166,9 @@ protected:
const std::size_t size = map->GetEnd() - map->GetStart();
rasterizer.UpdatePagesCachedCount(map->GetCpuAddress(), size, -1);
map->MarkAsRegistered(false);
if (map->IsWritten()) {
UnmarkRegionAsWritten(map->GetStart(), map->GetEnd() - 1);
}
const IntervalType delete_interval{map->GetStart(), map->GetEnd()};
mapped_addresses.erase(delete_interval);
}
@ -155,8 +178,8 @@ private:
return std::make_shared<MapIntervalBase>(start, end, gpu_addr);
}
void MapAddress(const TBuffer& block, const GPUVAddr gpu_addr, const CacheAddr cache_addr,
const std::size_t size) {
MapInterval MapAddress(const TBuffer& block, const GPUVAddr gpu_addr,
const CacheAddr cache_addr, const std::size_t size) {
std::vector<MapInterval> overlaps = GetMapsInRange(cache_addr, size);
if (overlaps.empty()) {
@ -165,22 +188,24 @@ private:
u8* host_ptr = FromCacheAddr(cache_addr);
UploadBlockData(block, block->GetOffset(cache_addr), size, host_ptr);
Register(new_map);
return;
return new_map;
}
const CacheAddr cache_addr_end = cache_addr + size;
if (overlaps.size() == 1) {
const MapInterval& current_map = overlaps[0];
MapInterval& current_map = overlaps[0];
if (current_map->IsInside(cache_addr, cache_addr_end)) {
return;
return current_map;
}
}
CacheAddr new_start = cache_addr;
CacheAddr new_end = cache_addr_end;
bool write_inheritance = false;
// Calculate new buffer parameters
for (auto& overlap : overlaps) {
new_start = std::min(overlap->GetStart(), new_start);
new_end = std::max(overlap->GetEnd(), new_end);
write_inheritance |= overlap->IsWritten();
}
GPUVAddr new_gpu_addr = gpu_addr + new_start - cache_addr;
for (auto& overlap : overlaps) {
@ -188,7 +213,8 @@ private:
}
UpdateBlock(block, new_start, new_end, overlaps);
MapInterval new_map = CreateMap(new_start, new_end, new_gpu_addr);
Register(new_map);
Register(new_map, write_inheritance);
return new_map;
}
void UpdateBlock(const TBuffer& block, CacheAddr start, CacheAddr end,
@ -320,6 +346,48 @@ private:
return found;
}
void MarkRegionAsWritten(const CacheAddr start, const CacheAddr end) {
u64 page_start = start >> write_page_bit;
const u64 page_end = end >> write_page_bit;
while (page_start <= page_end) {
auto it = written_pages.find(page_start);
if (it != written_pages.end()) {
it->second = it->second + 1;
} else {
written_pages[page_start] = 1;
}
page_start++;
}
}
void UnmarkRegionAsWritten(const CacheAddr start, const CacheAddr end) {
u64 page_start = start >> write_page_bit;
const u64 page_end = end >> write_page_bit;
while (page_start <= page_end) {
auto it = written_pages.find(page_start);
if (it != written_pages.end()) {
if (it->second > 1) {
it->second = it->second - 1;
} else {
written_pages.erase(it);
}
}
page_start++;
}
}
bool IsRegionWritten(const CacheAddr start, const CacheAddr end) const {
u64 page_start = start >> write_page_bit;
const u64 page_end = end >> write_page_bit;
while (page_start <= page_end) {
if (written_pages.count(page_start) > 0) {
return true;
}
page_start++;
}
return false;
}
std::unique_ptr<StreamBuffer> stream_buffer;
TBufferType stream_buffer_handle{};
@ -334,11 +402,14 @@ private:
using IntervalType = typename IntervalCache::interval_type;
IntervalCache mapped_addresses{};
static constexpr u64 block_page_bits{24};
static constexpr u64 block_page_size{1 << block_page_bits};
std::unordered_map<u64, TBuffer> blocks;
static constexpr u64 write_page_bit{11};
std::unordered_map<u64, u32> written_pages{};
std::list<TBuffer> pending_destruction;
static constexpr u64 block_page_bits{21};
static constexpr u64 block_page_size{1 << block_page_bits};
std::unordered_map<u64, TBuffer> blocks{};
std::list<TBuffer> pending_destruction{};
u64 epoch{};
u64 modified_ticks{};
VideoCore::RasterizerInterface& rasterizer;

View file

@ -54,12 +54,33 @@ public:
return end;
}
void MarkAsModified(const bool is_modified_, const u64 tick) {
is_modified = is_modified_;
ticks = tick;
}
bool IsModified() const {
return is_modified;
}
u64 GetModificationTick() const {
return ticks;
}
void MarkAsWritten(const bool is_written_) {
is_written = is_written_;
}
bool IsWritten() const {
return is_written;
}
private:
CacheAddr start;
CacheAddr end;
GPUVAddr gpu_addr;
VAddr cpu_addr{};
bool is_write{};
bool is_written{};
bool is_modified{};
bool is_registered{};
u64 ticks{};