optimize startup time (#175)

* use tsl::robin_set for BlockIndexSet
don't calculate full block location when only block index is needed
* don't copy visitor function
* reserve space for attachments

4 files changed, 26 insertions, 35 deletions

diff --git a/src/zenstore/blockstore.cpp b/src/zenstore/blockstore.cpp
index 7c6677052..592d1c7fb 100644
--- a/src/zenstore/blockstore.cpp
+++ b/src/zenstore/blockstore.cpp
@@ -331,17 +331,7 @@ BlockStore::Initialize(const std::filesystem::path& BlocksBasePath, uint64_t Max
 }
 
 void
-BlockStore::BlockIndexSet::Add(uint32_t BlockIndex)
-{
-	if (!std::binary_search(begin(BlockIndexes), end(BlockIndexes), BlockIndex))
-	{
-		auto It = std::lower_bound(begin(BlockIndexes), end(BlockIndexes), BlockIndex);
-		BlockIndexes.insert(It, BlockIndex);
-	}
-}
-
-void
-BlockStore::SyncExistingBlocksOnDisk(const BlockIndexSet& KnownLocations)
+BlockStore::SyncExistingBlocksOnDisk(const BlockIndexSet& KnownBlocks)
 {
 	ZEN_TRACE_CPU("BlockStore::SyncExistingBlocksOnDisk");
 
@@ -355,7 +345,7 @@ BlockStore::SyncExistingBlocksOnDisk(const BlockIndexSet& KnownLocations)
 		DeleteBlocks.insert(It.first);
 	}
 
-	for (const uint32_t BlockIndex : KnownLocations.GetBlockIndices())
+	for (const uint32_t BlockIndex : KnownBlocks)
 	{
 		DeleteBlocks.erase(BlockIndex);
 		if (auto It = m_ChunkBlocks.find(BlockIndex); It != m_ChunkBlocks.end() && !It->second.IsNull())
@@ -1968,9 +1958,9 @@ TEST_CASE("blockstore.clean.stray.blocks")
 
 	// Recreate a fake block for a missing chunk location
 	BlockStore::BlockIndexSet KnownBlocks;
-	KnownBlocks.Add(FirstChunkLocation.BlockIndex);
-	KnownBlocks.Add(SecondChunkLocation.BlockIndex);
-	KnownBlocks.Add(ThirdChunkLocation.BlockIndex);
+	KnownBlocks.insert(FirstChunkLocation.BlockIndex);
+	KnownBlocks.insert(SecondChunkLocation.BlockIndex);
+	KnownBlocks.insert(ThirdChunkLocation.BlockIndex);
 	Store.SyncExistingBlocksOnDisk(KnownBlocks);
 
 	// We create a fake block for the location - we should still not be able to get the chunk
diff --git a/src/zenstore/cache/cachedisklayer.cpp b/src/zenstore/cache/cachedisklayer.cpp
index a4aa497c4..3fc4eca03 100644
--- a/src/zenstore/cache/cachedisklayer.cpp
+++ b/src/zenstore/cache/cachedisklayer.cpp
@@ -1136,8 +1136,8 @@ ZenCacheDiskLayer::CacheBucket::InitializeIndexFromDisk(RwLock::ExclusiveLockSco
 		}
 		else
 		{
-			const BlockStoreLocation& BlockLocation = Location.GetBlockLocation(m_Configuration.PayloadAlignment);
-			KnownBlocks.Add(BlockLocation.BlockIndex);
+			uint32_t BlockIndex = Location.Location.BlockLocation.GetBlockIndex();
+			KnownBlocks.insert(BlockIndex);
 		}
 	}
 	m_BlockStore.SyncExistingBlocksOnDisk(KnownBlocks);
@@ -3649,6 +3649,7 @@ ZenCacheDiskLayer::CacheBucket::GetReferences(GcCtx& Ctx, bool StateIsAlreadyLoc
 				{
 					continue;
 				}
+
 				const IoHash& Key = Entry.first;
 				if (Loc.IsFlagSet(DiskLocation::kStandaloneFile))
 				{
@@ -3672,7 +3673,10 @@ ZenCacheDiskLayer::CacheBucket::GetReferences(GcCtx& Ctx, bool StateIsAlreadyLoc
 			}
 		}
 
-		for (std::vector<std::size_t> ChunkIndexes : InlineBlockChunkIndexes)
+		OutReferences.reserve(OutReferences.size() + InlineKeys.size() +
+							  StandaloneKeys.size());  // Make space for at least one attachment per record
+
+		for (const std::vector<std::size_t>& ChunkIndexes : InlineBlockChunkIndexes)
 		{
 			ZEN_ASSERT(!ChunkIndexes.empty());
 
@@ -3694,12 +3698,15 @@ ZenCacheDiskLayer::CacheBucket::GetReferences(GcCtx& Ctx, bool StateIsAlreadyLoc
 				auto CaptureAttachments = [&](size_t ChunkIndex, MemoryView Data) {
 					if (GetAttachments(Data))
 					{
-						size_t AttachmentCount = OutReferences.size() - NextPrecachedReferencesStart;
-						if (WriteMetaData && AttachmentCount > 0)
+						if (WriteMetaData)
 						{
-							Keys.push_back(InlineKeys[ChunkIndex]);
-							AttachmentCounts.push_back(gsl::narrow<uint32_t>(AttachmentCount));
-							NextPrecachedReferencesStart += AttachmentCount;
+							size_t AttachmentCount = OutReferences.size() - NextPrecachedReferencesStart;
+							if (AttachmentCount > 0)
+							{
+								Keys.push_back(InlineKeys[ChunkIndex]);
+								AttachmentCounts.push_back(gsl::narrow<uint32_t>(AttachmentCount));
+								NextPrecachedReferencesStart += AttachmentCount;
+							}
 						}
 					}
 				};
diff --git a/src/zenstore/compactcas.cpp b/src/zenstore/compactcas.cpp
index 6af07a606..5fca3046c 100644
--- a/src/zenstore/compactcas.cpp
+++ b/src/zenstore/compactcas.cpp
@@ -1293,9 +1293,9 @@ CasContainerStrategy::OpenContainer(bool IsNewStore)
 
 	for (const auto& Entry : m_LocationMap)
 	{
-		const BlockStoreDiskLocation& DiskLocation	= m_Locations[Entry.second];
-		BlockStoreLocation			  BlockLocation = DiskLocation.Get(m_PayloadAlignment);
-		KnownBlocks.Add(BlockLocation.BlockIndex);
+		const BlockStoreDiskLocation& DiskLocation = m_Locations[Entry.second];
+		uint32_t					  BlockIndex   = DiskLocation.GetBlockIndex();
+		KnownBlocks.insert(BlockIndex);
 	}
 
 	m_BlockStore.SyncExistingBlocksOnDisk(KnownBlocks);
diff --git a/src/zenstore/include/zenstore/blockstore.h b/src/zenstore/include/zenstore/blockstore.h
index 29c53cf22..5af9aeb10 100644
--- a/src/zenstore/include/zenstore/blockstore.h
+++ b/src/zenstore/include/zenstore/blockstore.h
@@ -8,6 +8,7 @@
 
 ZEN_THIRD_PARTY_INCLUDES_START
 #include <tsl/robin_map.h>
+#include <tsl/robin_set.h>
 ZEN_THIRD_PARTY_INCLUDES_END
 
 #include <unordered_set>
@@ -144,18 +145,11 @@ public:
 
 	void Initialize(const std::filesystem::path& BlocksBasePath, uint64_t MaxBlockSize, uint64_t MaxBlockCount);
 
-	struct BlockIndexSet
-	{
-		void					  Add(uint32_t BlockIndex);
-		std::span<const uint32_t> GetBlockIndices() const { return BlockIndexes; }
-
-	private:
-		std::vector<uint32_t> BlockIndexes;
-	};
+	typedef tsl::robin_set<uint32_t> BlockIndexSet;
 
 	// Ask the store to create empty blocks for all locations that does not have a block
 	// Remove any block that is not referenced
-	void			   SyncExistingBlocksOnDisk(const BlockIndexSet& KnownLocations);
+	void			   SyncExistingBlocksOnDisk(const BlockIndexSet& KnownBlocks);
 	BlockEntryCountMap GetBlocksToCompact(const BlockUsageMap& BlockUsage, uint32_t BlockUsageThresholdPercent);
 
 	void Close();