Commit 2fd29ed7 authored by Harald Sitter's avatar Harald Sitter 🏳️‍🌈
Browse files


parent d6e40ca9
Pipeline #90529 passed with stage
in 2 minutes and 4 seconds
......@@ -19,12 +19,12 @@ off_t TransferSegment::segmentSizeForFileSize(const off_t fileSize_)
// read() internally splits our read requests into multiple server
// requests and then assembles the responses into our buffer.
// The larger the chunks we request the better the performance.
// At the same time we'll want a semblence of progress reporting
// At the same time we'll want a semblance of progress reporting
// and also not eat too much RAM. It's a balancing act :|
off_t segmentSize = c_minSegmentSize;
// The segment size is largely arbitrary and sacrifices better throughput for
// greater memory use.
// This only goes up to a maxiumum because bigger transfer blobs directly
// This only goes up to a maximum because bigger transfer blobs directly
// translate to more RAM use. Mind that the effective RAM use will
// be (segmentSize * (segments + 1)). The +1 is because smbc internally will also
// allocate up to a full segment for one read() call.
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment