160e52ec7e
Resolves #2054 **Context**: branch creation needs to wait for GC to acquire `gc_cs` lock, which prevents creating new timelines during GC. However, because individual timeline GC iteration also requires `compaction_cs` lock, branch creation may also need to wait for compactions of multiple timelines. This results in large latency when creating a new branch, which we advertised as *"instantly"*. This PR optimizes the latency of branch creation by separating GC into two phases: 1. Collect GC data (branching points, cutoff LSNs, etc) 2. Perform GC for each timeline The GC bottleneck comes from step 2, which must wait for compaction of multiple timelines. This PR modifies the branch creation and GC functions to allow GC to hold the GC lock only in step 1. As a result, branch creation doesn't need to wait for compaction to finish but only needs to wait for GC data collection step, which is fast.
This module contains utilities for working with PostgreSQL file formats. It's a collection of structs that are auto-generated from the PostgreSQL header files using bindgen, and Rust functions to read and manipulate them.
There are also a bunch of constants in pg_constants.rs that are copied
from various PostgreSQL headers, rather than auto-generated. They mostly
should be auto-generated too, but that's a TODO.
The PostgreSQL on-disk file format is not portable across different CPU architectures and operating systems. It is also subject to change in each major PostgreSQL version. Currently, this module is based on PostgreSQL v14, but in the future we will probably need a separate copy for each PostgreSQL version.
TODO: Currently, there is also some code that deals with WAL records in pageserver/src/waldecoder.rs. That should be moved into this module. The rest of the codebase should not have intimate knowledge of PostgreSQL file formats or WAL layout, that knowledge should be encapsulated in this module.