* Preventively switch to uninterruptible mask in withMVar'
withMVar' is used to update the shakeSession var and it's crucial that the
third argument is not interrupted.
'mask' can still be interrupted for I/O actions and, while we were careful to
ensure none was used, if it ever breaks it will lead to very hard to debug
problems.
* refactor: move to RuleTypes
* Add a TestRequest to wait for arbitrary ide actions
Closes#955
* expectCurrentDiagnostics
* Add a test suite for cancellation
* Introduce --test-no-kick to fix cancellation tests reliability
* delete unsafeClearDiagnostics (unused)
* GetModSummaryWithoutTimestamps - remove StringBuffer
Since the contents of the buffer are not tracked by the fingerprint.
* Fix diagnostics bug
Given a FOI F with non null typechecking diagnostics D, imagine the following scenario:
1. An edit notification for F is received, creating a new version
2. GetModTime is executed, producing 0 diagnostics.
2.1 updateFileDiagnostics is called
2.2 setStageDiagnostics is called
2.3 LSP.updateDiagnostics is called with a new version, resetting all the
diagnostics for F
2.4 newDiags=[] in updateFileDiagnostics, which is different from D (the last
published diagnostics), which enqueues a new publishDiagnostics [] in the
Debouncer
3. An edit notification for F is received before typechecking has a chance to
run which undoes the previous edit
4. The debouncer publishes the empty set of diagnostics after waiting 0.1s
5. GetFileContents runs and since the contents of the file haven't changed since
the last time it ran, early cutoff skips everything donwstream
Since TypeCheck is skipped, the empty set of diagnostics stays published until
another edit comes.
The goal of this change is to prevent setStageDiagnostics from losing
diagnostics from other stages. To achieve this, we recover the old diagnostics
for all stages and merge them with the new stage.
* Fix hlint
* Use Map.insert for clarity
* Fix redundant imports
* Fix "code actions after edit" experiment"
* [bench-hist] break down in rule functions
* Extract the benchmarking Shake rules to a shake-bench package
There's some room for reusing the rules used in the historic benchmarking suite
in other projects. This change makes that a bit easier and improves the
documentation and code structure.
The new structure is:
- lib:shake-bench - a Cabal library with functions to generate Shake rules
- ghcide:bench:benchHist - the ghcide instantiation of the above Shake rules
That's not to say that shake-bench is completely decoupled from ghcide -
there are still plenty of assumptions on how the benchmarks are organized, their
outputs, etc. But with a little bit of effort, it should be easy to make
these rules more reusable
* Fix nix build
* Fix license
* hlints and redundant imports
* more hlints
* Exclude shake-bench from the stack build
* Move tracing functions to own module
* Bump opentelemetry to 0.6.0
* Write Values map size to OpenTelemetry metric
* Trace all requests and notifications
Instead of doing it in `HoverDefinition`, do it in
with{Response,Notification,...}. These wrap all handlers, so this should
cover everything. It also means that the span covers the entire
processing time for the request, where before we missed the setup
happening in the with* functions.
* Add flag for OpenTelemetry profiling
Run GC regularly with --ot-profiling
* Add flag to enable OT profiling in benchmark
* Use heapsize instead of ghc-datasize
I renamed the fork to distringuish from the original.
It is still being pulled from git using stack. This will be addressed
once I can push the fork to hackage.
* Bump opentelemetry to 0.6.1 - fixes 8.6 build
* Use heapsize from hackage
* Address HLint messages
* Record size of each key independently
* Refactor `startTelemetry` function
* Remove delay between measuring memory loops
* Each key in values map gets own OT instrument
* Measure values map length more rarely
* Rename --ot-profiling to --ot-memory-profiling
* Add docs for how to use the opentelemetry output
* Add instructions to build release version of tracy
* Clarify dependencies in opentelemetry instructions
* Fix LSP traces
* otTraced: delete unused
* Extract types out of D.IDE.Core.Shake
to avoid circular module dependencies
* Extract startTelemetry out of D.IDE.Shake and upgrade to 0.2
No more segfaults
* [nix] install opentelemetry
* [nix] install tracy
* Fix merge wibble
* Measure recursive sizes with sharing
* Sort keys for cost attribution
* Remove debug traces
* Allocate less, group keys, clean up hlints
* Add -A4G to the flags used for --ot-memory-profiling
* Modularize D.IDE.Core.Tracing
I want to reuse this code more directly in the non lsp driver
* Direct driver: report closure sizes when --ot-memory-profiling
An eventlog memory analysis doesnt' seem so relevant since this mode is not
interactive, but we could easily produce both if wanted to
* Everything is reachable from GhcSessionIO, so compute it last
I suspect the ShakeExtras record is reachable from GhcSessionIO
* bound recursion and use logger
* hlint suggestions
* Fix 8.6 build
* Format imports
* Do the memory analysis with full sharing. GhcSessionIO last
* Fail fast in the memory analysis
* error handling
* runHeapsize now takes initSize as an input argument
* Trace Shake sessions
* Reduced frequency for sampling values length
* Drop the -fexternal-interpreter flag in the Windows stack build
* Produce more benchmark artifacts
* Fix stack descriptors to use heapsize-0.2 from Hackage
* Bump to heapsize-0.3.0
* Record completions snippets (#900)
* Add field for RecordSnippets to CachcedCompletion
* Initial version of local record snippets
* Supprt record snippet completion for non local declarations.
* Better integration of local completions with current implementation
* Clean up non-local completions.
* Remove commented code.
* Switch from String to Text
* Remove ununsed definition
* Treat only Records and leave other defintions as is.
* Differentiate Records from Data constructors for external declaration
* Update test to include snippet in local record completions expected list.
* Update completionTest to also compare insertText.
* Add test for record snippet completion for imported records.
* Hlint fixes
* Hlint fixes
* Hlint suggestions.
* Update type.
* Consolidate imports
* Unpack tuple with explicit names
* Idiomatic changes
* Remove unused variable
* Better variable name
* Hlint suggestions
* Handle exhaustive pattern warning
* Add _ to snippet field name suggestions
* Remove type information passed around but not used
* Update to list comprehension style
* Eliminate intermediate function
* HLint suggestions.
* Idiomatic list comprehension
Co-authored-by: Pepe Iborra <pepeiborra@me.com>
* [nix] use gitignore.nix (#920)
* Ignore import list while producing completions (#919)
* Drop any items in explicit import list
* Test if imports not included in explicit list show up in completions
* Update README.md (#924)
* Custom cradle loading (#928)
When using ghcide as a library, it may be desirable to host the hie.yaml file
in a location other than the project root, or even avoid the file system altogether
* Favor `lookupPathToId` over `pathToId` (#926)
* Favor `lookupPathToId` over `pathToId`
* Fix `typecheckParentsAction`
* Fix `needsCompilationRule`
* Return completion snippets only when client supports it (#929)
* Use the real client capabilities on completions
* Return completion snippets only when supported by the client
Restored from https://github.com/haskell/ghcide/pull/900
* Redundant import
* Fix stack windows build
Co-authored-by: Michalis Pardalos <m.pardalos@gmail.com>
Co-authored-by: Michalis Pardalos <mpardalos@gmail.com>
Co-authored-by: Guru Devanla <gdevanla@users.noreply.github.com>
Co-authored-by: Samuel Ainsworth <skainsworth@gmail.com>
* parse allocations
* WaitForShakeQueue
* Measure user time and shake time in experiments
* clean ups
* Prevent a potential crash of the shake enqueue thread
* Fix a bug that was preventing reenqueud actions from getting flushed
* Avoid running the check-project action per file
What we really want is to check the project once per cradle
* Backwards compat.
* Review feedback
* Fix typo
Co-authored-by: Neil Mitchell <ndmitchell@gmail.com>
Co-authored-by: Neil Mitchell <ndmitchell@gmail.com>
* Remove hardcoded --stack-yaml and upstream/master assumption
* support Cabal in bench suite
* add benchmark run to CI
Even if the time measurements are unreliable in a shared CI environment, the
memory usage will be an accurate indicator of space leaks
* Update bench/README
* use origin/master
* default to stack in benchmarks (for CI)
* ignore ghcide-bench and ghcide-preprocessor binaries too
* Review feedbacks
* Add the v0.3.0 tag in bench/hist.yaml
commented out to keep the CI time as tight as possible
* Add .artifactignore file to avoid publishing binaries in azure bench pipeline
* use default stack.yaml
* Faster completions
* optimize withProgressVar
We never remove elements from the map so alter is unnecesary
* [ghcide-bench] accept ghcide options
* Expand completion tests suite
* hlints
* completions for local foreign decls
* Minor improvements for local completions
* Restore completion docs in legacy code path
* Compatibility with GHC < 8.8
* fix merge issue
* address review feedback
* [bench-hist] save messages to log file
And fix the commitid rule to always rerun the git query
* Do not build benchHist by default
Hopefully avoiding the additional dependencies for charts
* Simplify with FileStdout
* Add a flag for bench-hist
Could benchHist be a benchmark instead of an executable, removing the need
for this flag?
Almost. `stack bench` fails because `benchHist` cannot find `ghcide-bench`
in the path. It seems like a bad idea to have a benchmark that fails out of
the box
* Turn benchHist into a benchmark and ghcide-bench into an exe
This works out nicely because:
1. benchHist already runs ghcide-bench,
2. benchHist has additional deps, but ghcide-bench does not.
(benchmark deps don't get built by default)
3. This is the only way I've found to get ghcide-bench in the PATH for benchHist
* Remove redundant dep on applicative-combinators
* Bump versions in stack-ghc-lib.yaml
* update lower bounds for extra in benchHist executable
* Update README guideline on benchmarks
* [benchHist] Fix the commitid rule to always rerun the git query
* fix caps
* send WorkDoneProgressEnd only when work done
* Progress reporting now spans over multiple overlapping kicks
* Repurpose benchmark experiments as tests
Fixes#650
* use stack to fetch from Hackage
* benchmark tests run with the same lsp-test config as other tests
* Fix stack cradle in benchmark
* Make stack unpack --silent
* Fix issues in "code actions after edit" experiment
- Repeated breaking edits make ghc run out of suggestions
- Diagnostics seem to come and go in-between edits, which leads to a timing
issue when asking for code actions. The fix is to wait for diagnostics to be
present before asking for code actions
* Fix stack.yaml generation in example project
* Fix getDefinition in GHC 8.4
Did it break before 0.2.0 or after?
* better naming for the progress event TVar
* stop progress reporting in shakeShut
https://github.com/digital-asset/ghcide/pull/649#discussion_r443408884
* hlint
* Write a cabal.project file
As suggested in #617. Taken fron #624
* Write a cabal.project.local
Otherwise Cabal still errors out
* Override default hie dir
Otherwise .hi and .hie files end up in different locations, which causes the getDefinition experiment to fail the second time it's run.
This is because we assume in ghcide that .hi and .hie files have the same lifetimes, which is not true when the ..hie files are wiped but the .hi files aren't.
* benchmark history script
* if HEAD no need to rebuild worktree
* add bench/README.md
* Enable all experiments
* Fix dependency tracking for git branches
* hlints
* Add stack84 extra-deps
* Identify failed experiments in graphs
* Filter our failed benchmarks from aggregate graphs
Otherwise they tend to distort the axis
* Improve graphs (more and easier to see colors)
* update cradles
* customizable output folder
* Cache the config for the duration of the script
Otherwise the script is vulnerable to config edits
* Allow omitting the git: field
* Ignore bench-hist intermediate artifacts
Handy for including bench-hist results in a branch while avoiding the
intermediate artifacts
* Add a benchmark to track startup times
* Benchmark automation
disable benchmarks easily
save GC stats to file
cradle, rts, filter and samples options
path to ghcide
configurable example
--help
more detailed CSV output
hover after edit
pause for GC
configurable timeout
upgrade extra (required to build bench)
Include max residency in BenchRun
Include all details on output
* reduce threadDelay to avoid upsetting lsp-test
* Fix startup time measurement
* Added new edit experiment
* fix doc comment
* hlints
* Upgrade to lsp-test 0.11.0.2
* Flag failed experiments
* Update ghcide.cabal
* Initial benchmark suite, reusing ideas from Neil's post
https://neilmitchell.blogspot.com/2020/05/fixing-space-leaks-in-ghcide.html
* Add an experiment for code actions without edit
* formatting
* fix code actions bench script
* error handling + options + how to run
* extract Positions and clean up imports
(Neil's review feedback)
* replace with Extra.duration
* allow ImplicitParams
* add bench to the cradle
* applied @mpickering review feedback
* clean up after benchmark
* remove TODO
