DiscIO: Decrease RAM usage during zstd compression

By calling ZSTD_CCtx_setPledgedSrcSize, we can let zstd know
how large a chunk is going to be before which start compressing
it, which lets zstd avoid allocating more memory than needed
for various internal buffers. This greatly reduces the RAM usage
when using a high compression level with a small chunk size,
and doesn't have much of an effect in other circumstances.

A side effect of calling ZSTD_CCtx_setPledgedSrcSize is that
zstd by default will write the uncompressed size into the
compressed data stream as metadata. In order to save space,
and since the decompressed size can be figured out through
the structure of the RVZ format anyway, we disable writing
the uncompressed size by setting ZSTD_c_contentSizeFlag to 0.
This commit is contained in:
JosJuice 2020-10-04 00:09:55 +02:00
parent ebdcddfcd0
commit 3feea108db
3 changed files with 25 additions and 13 deletions

View File

@ -1040,7 +1040,7 @@ std::optional<std::vector<u8>> WIARVZFileReader<RVZ>::Compress(Compressor* compr
{ {
if (compressor) if (compressor)
{ {
if (!compressor->Start() || !compressor->Compress(data, size) || !compressor->End()) if (!compressor->Start(size) || !compressor->Compress(data, size) || !compressor->End())
return std::nullopt; return std::nullopt;
data = compressor->GetData(); data = compressor->GetData();
@ -1564,7 +1564,7 @@ WIARVZFileReader<RVZ>::ProcessAndCompress(CompressThreadState* state, CompressPa
if (state->compressor) if (state->compressor)
{ {
if (!state->compressor->Start()) if (!state->compressor->Start(entry.exception_lists.size() + entry.main_data.size()))
return ConversionResultCode::InternalError; return ConversionResultCode::InternalError;
} }

View File

@ -446,7 +446,7 @@ PurgeCompressor::PurgeCompressor()
PurgeCompressor::~PurgeCompressor() = default; PurgeCompressor::~PurgeCompressor() = default;
bool PurgeCompressor::Start() bool PurgeCompressor::Start(std::optional<u64> size)
{ {
m_buffer.clear(); m_buffer.clear();
m_bytes_written = 0; m_bytes_written = 0;
@ -550,7 +550,7 @@ Bzip2Compressor::~Bzip2Compressor()
BZ2_bzCompressEnd(&m_stream); BZ2_bzCompressEnd(&m_stream);
} }
bool Bzip2Compressor::Start() bool Bzip2Compressor::Start(std::optional<u64> size)
{ {
ASSERT_MSG(DISCIO, m_stream.state == nullptr, ASSERT_MSG(DISCIO, m_stream.state == nullptr,
"Called Bzip2Compressor::Start() twice without calling Bzip2Compressor::End()"); "Called Bzip2Compressor::Start() twice without calling Bzip2Compressor::End()");
@ -674,7 +674,7 @@ LZMACompressor::~LZMACompressor()
lzma_end(&m_stream); lzma_end(&m_stream);
} }
bool LZMACompressor::Start() bool LZMACompressor::Start(std::optional<u64> size)
{ {
if (m_initialization_failed) if (m_initialization_failed)
return false; return false;
@ -745,8 +745,11 @@ ZstdCompressor::ZstdCompressor(int compression_level)
{ {
m_stream = ZSTD_createCStream(); m_stream = ZSTD_createCStream();
if (ZSTD_isError(ZSTD_CCtx_setParameter(m_stream, ZSTD_c_compressionLevel, compression_level))) if (ZSTD_isError(ZSTD_CCtx_setParameter(m_stream, ZSTD_c_compressionLevel, compression_level)) ||
ZSTD_isError(ZSTD_CCtx_setParameter(m_stream, ZSTD_c_contentSizeFlag, 0)))
{
m_stream = nullptr; m_stream = nullptr;
}
} }
ZstdCompressor::~ZstdCompressor() ZstdCompressor::~ZstdCompressor()
@ -754,7 +757,7 @@ ZstdCompressor::~ZstdCompressor()
ZSTD_freeCStream(m_stream); ZSTD_freeCStream(m_stream);
} }
bool ZstdCompressor::Start() bool ZstdCompressor::Start(std::optional<u64> size)
{ {
if (!m_stream) if (!m_stream)
return false; return false;
@ -762,7 +765,16 @@ bool ZstdCompressor::Start()
m_buffer.clear(); m_buffer.clear();
m_out_buffer = {}; m_out_buffer = {};
return !ZSTD_isError(ZSTD_CCtx_reset(m_stream, ZSTD_reset_session_only)); if (ZSTD_isError(ZSTD_CCtx_reset(m_stream, ZSTD_reset_session_only)))
return false;
if (size)
{
if (ZSTD_isError(ZSTD_CCtx_setPledgedSrcSize(m_stream, *size)))
return false;
}
return true;
} }
bool ZstdCompressor::Compress(const u8* data, size_t size) bool ZstdCompressor::Compress(const u8* data, size_t size)

View File

@ -154,7 +154,7 @@ public:
// First call Start, then AddDataOnlyForPurgeHashing/Compress any number of times, // First call Start, then AddDataOnlyForPurgeHashing/Compress any number of times,
// then End, then GetData/GetSize any number of times. // then End, then GetData/GetSize any number of times.
virtual bool Start() = 0; virtual bool Start(std::optional<u64> size) = 0;
virtual bool AddPrecedingDataOnlyForPurgeHashing(const u8* data, size_t size) { return true; } virtual bool AddPrecedingDataOnlyForPurgeHashing(const u8* data, size_t size) { return true; }
virtual bool Compress(const u8* data, size_t size) = 0; virtual bool Compress(const u8* data, size_t size) = 0;
virtual bool End() = 0; virtual bool End() = 0;
@ -169,7 +169,7 @@ public:
PurgeCompressor(); PurgeCompressor();
~PurgeCompressor(); ~PurgeCompressor();
bool Start() override; bool Start(std::optional<u64> size) override;
bool AddPrecedingDataOnlyForPurgeHashing(const u8* data, size_t size) override; bool AddPrecedingDataOnlyForPurgeHashing(const u8* data, size_t size) override;
bool Compress(const u8* data, size_t size) override; bool Compress(const u8* data, size_t size) override;
bool End() override; bool End() override;
@ -189,7 +189,7 @@ public:
Bzip2Compressor(int compression_level); Bzip2Compressor(int compression_level);
~Bzip2Compressor(); ~Bzip2Compressor();
bool Start() override; bool Start(std::optional<u64> size) override;
bool Compress(const u8* data, size_t size) override; bool Compress(const u8* data, size_t size) override;
bool End() override; bool End() override;
@ -211,7 +211,7 @@ public:
u8* compressor_data_size_out); u8* compressor_data_size_out);
~LZMACompressor(); ~LZMACompressor();
bool Start() override; bool Start(std::optional<u64> size) override;
bool Compress(const u8* data, size_t size) override; bool Compress(const u8* data, size_t size) override;
bool End() override; bool End() override;
@ -234,7 +234,7 @@ public:
ZstdCompressor(int compression_level); ZstdCompressor(int compression_level);
~ZstdCompressor(); ~ZstdCompressor();
bool Start() override; bool Start(std::optional<u64> size) override;
bool Compress(const u8* data, size_t size) override; bool Compress(const u8* data, size_t size) override;
bool End() override; bool End() override;