Note
This is a "meta" issue for tracking purposes, since GitHub does not support special types of issues.

Tracking issue for any perf infra improvements + performance icebergs investigation/fixes.

Open

Others:

• Optimization for string interpolation - simple scenarios: unfolding constants and lowering to concatenation #16247
• Resurrect compiler perf improvements PRs #16125
• Suboptimal code generated when pattern matching a tuple in a lambda parameter #16037
• Eliminate union case remapping common in monadic functions #15872
• Create a test project for bulk code fix benchmarking #15716
• Too many & and | patterns cause Visual Studio/Ionide to use up all available memory #15086
• Cache provided types in the compiler #14754
• Reliable FCS benchmarks which run on every release #13606
• Difference of compilation time between 13 000 "static member" and 13 000 "let binding" #13218
• dotnet fsi startup overhead surprisingly high #12636
• Bug Quadratic(?) compiler perf for huge nested expressions #12421
• Optimization inlining applied inconsistently with piping #12416
• Don't emit unnecessary closures #11898
• Reduce memory footprint and improve performance of expression-based ORM usage #10522
• The performance of Comparing and Ordering things #9348
• isOpenStatementUsed uses a lot of CPU and memory #9230
• Code perf: optimize typeof equality in inlined code #6228
• Boxing occurs when hashing a range - 200+ MB over a 90 seconds of normal IDE usage #6052
• Compiling causes OutOfMemoryException and IDE unresponsive/crash when pattern matching against record with many nested DU fields #4691

Closed

• Write up 'Compiler Perf design session' notes from the summer (hand written papers at @T-Gro's desk) #16498
• Avoid boxing on equality and comparison #16497
• Timebox: Set up FCS sync in the dotnet/performance repo #16639
• Replace active pattern with structs in the compiler code #16286
• Run benchmarks in CI #16408
• Facelift the existing benchmarks #16407
• Update the Benchmark README #16406
• Inline functions in the Result module #15833
• Reference equality optimisation commented out in implementation of (=) #15486
• FSC: Consider enabling background JIT with non-dynamic PGO for faster compilation #15434
• fsc on .NET 6/7 is spawning dotnet process #13960
• Revalidate perf of batch/server GC settings for fsc, fsi on .NET Framework and .NET Core #13739
• Investigate why build processes started consuming much more memory #13600
• Support caching assembly paths resolved by the DependencyManager #13236
• Perf improvement for primitive comparison #13098
• Perf - Lots of time being spent in HashCompare.GenericCompare due to LowerComputedListOrArrayExpr #13033
• A big part of FCS analysis time is spent in GC #13031
• F# sig data contains Too Much Information #12746
• Performance decrease when running FSharpChecker.ParseAndCheckProject  #12380
• Validate .tailcall improvements, identify further optimisations #12196
• VS 2022 Preview - Startup time for CoreClr compiler slower than desktop #11907
• FCS performance benchmark notebook #11802
• Pinning a string should prefer string.GetPinnableReference over RuntimeHelpers.GetOffsetToStringData #9251
• Equality operator causes boxing on value types #526