Commits on Source (64)
-
By introducing a wrapped message type we can control how certain messages are printed in GHCi (to add extra information for example)
f8ced241 -
parent diagnostic options. * Split default diagnostic options from Diagnostic class into HasDefaultDiagnosticOpts class. * Generalise UnknownDiagnostic to allow embedded diagnostics to access options. The principle idea here is that when wrapping an error message (such as GHCMessage to make GHCiMessage) then we need to also be able to lift the configuration when overriding how messages are printed (see load' for an example).
58e554c1 -
This allows API users to configure how messages are rendered when they are emitted from the load function. For an example see how 'loadWithCache' is used in GHCi.
b112546a -
* cantFindError is abstracted so that the parts which mention specific things about ghc/ghci are parameters. The intention being that GHC/GHCi can specify the right values to put here but otherwise display the same error message. * The BuildingCabalPackage argument from GenericMissing is removed and turned into a print-time option. The reason for the error is not dependent on whether `-fbuilding-cabal-package` is passed, so we don't want to store that in the error message.
2e4cf0ee -
Tickets like #22884 suggest that it is confusing that GHC used on the command line can suggest options which only work in GHCi. This ticket uses the error message infrastructure to override certain error messages which displayed GHCi specific information so that this information is only showed when using GHCi. The main annoyance is that we mostly want to display errors in the same way as before, but with some additional information. This means that the error rendering code has to be exported from the Iface/Errors/Ppr.hs module. I am unsure about whether the approach taken here is the best or most maintainable solution. Fixes #22884
34b44f7d -
If a nightly pipeline runs twice for some reason for the same version then we really don't want to override an existing entry with new bindists. This could cause ABI compatability issues for users or break ghcup's caching logic.
05a1b626 -
Previously we were using links from the web interface, but it's more robust and future-proof to use the documented links to the artifacts. https://docs.gitlab.com/ee/api/job_artifacts.html
fcbcb3cc -
The latest nightly release needs the LatestNightly tag, and all other nightly releases need the Nightly tag. Therefore when the metadata is updated we need to replace all LatestNightly with Nightly.`
5b59c8fe -
The metadata now lives in https://gitlab.haskell.org/ghc/ghcup-metadata with one metadata file per year. When we update the metadata we download and update the right file for the current year.
914e1468 -
something about pipeline date
16cf7d2e -
On a push we now have a CI job which updates gitlab pages with the metadata files.
14792c4b -
The ghcup-metadata now has a viReleaseDay field which needs to be populated with the day of the release.
1121bdd8 -
ghcup now requires us to add this field which specifies where it should download the bindist to. See ghc/ghcup-metadata#1 for some more discussion.
bc478bee -
2bdbd9da
-
This patch adds the WasmControlFlow test to test the wasm backend's relooper component.
abd9e37c -
-
Also move targetPlatform selector Plenty of GHC needs just DynFlags. Even more can be made to use .DynFlags if more selectors is migrated. This is a low hanging fruit.
fad9d092 -
The original fix for #22919 simply removed the ability to match up prior comments with the first declaration in the file. Restore it, but add a check that the comment is on a single line, by ensuring that it comes immediately prior to the next thing (comment or start of declaration), and that the token preceding it is not on the same line. closes #22919
69fdbece -
Josh Meredith authored0350b186
-
Setting a larger stack chunk size gives a greater protection from stack thrashing (where the repeated overflow/underflow allocates a lot of stack chunks which sigificantly impact allocations). This stabilises some tests against differences cause by more things being pushed onto the stack. The performance tests are generally testing work done by the compiler, using allocation as a proxy, so removing/stabilising the allocations due to the stack gives us more stable tests which are also more sensitive to actual changes in compiler performance. The tests which increase are ones where we compile a lot of modules, and for each module we spawn a thread to compile the module in. Therefore increasing these numbers has a multiplying effect on these tests because there are many more stacks which we can increase in size. The most significant improvements though are cases such as T8095 which reduce significantly in allocations (30%). This isn't a performance improvement really but just helps stabilise the test against this threshold set by the defaults. Fixes #23439 ------------------------- Metric Decrease: InstanceMatching T14683 T8095 T9872b_defer T9872d T9961 hie002 T19695 T3064 Metric Increase: MultiLayerModules T13701 T14697 -------------------------
b4816919 -
These were previously hardcoded in configure (with no option for overriding them) and simply passed onto ghc through the settings file. Since configure already guarantees gcc supports those flags, we simply move them into GHC.
6629f1c5 -
Per the new `Note [Linting linearity]`, we want optimisations over trying to preserve linearity. This will allow CPR to handle `Ur`, in particular.
981e5e11 -
Per the new `Note [Linting linearity]`, we want optimisations over trying to preserve linearity. This will avoid preventing inlinings and reductions and make linear programs more efficient.
bf9344d2 -
-
-
-
-
This commit splits up the zonker into a few separate components, described in Note [The structure of the zonker] in `GHC.Tc.Zonk.Type`. 1. `GHC.Tc.Zonk.Monad` introduces a pared-down `TcM` monad, `ZonkM`, which has enough information for zonking types. This allows us to refactor `ErrCtxt` to use `ZonkM` instead of `TcM`, which guarantees we don't throw an error while reporting an error. 2. `GHC.Tc.Zonk.Env` is the new home of `ZonkEnv`, and also defines two zonking monad transformers, `ZonkT` and `ZonkBndrT`. `ZonkT` is a reader monad transformer over `ZonkEnv`. `ZonkBndrT m` is the codensity monad over `ZonkT m`. `ZonkBndrT` is used for computations that accumulate binders in the `ZonkEnv`. 3. `GHC.Tc.Zonk.TcType` contains the code for zonking types, for use in the typechecker. It uses the `ZonkM` monad. 4. `GHC.Tc.Zonk.Type` contains the code for final zonking to `Type`, which has been refactored to use `ZonkTcM = ZonkT TcM` and `ZonkBndrTcM = ZonkBndrT TcM`. Allocations slightly decrease on the whole due to using continuation-passing style instead of manual state passing of ZonkEnv in the final zonking to Type. ------------------------- Metric Decrease: T4029 T8095 T14766 T15304 hard_hole_fits RecordUpdPerf Metric Increase: T10421 -------------------------f62d8195 -
-
Reference ticket #21766 This commit restructures IPE buffer list entries to not contain references to their corresponding info tables. IPE buffer list nodes now point to two lists of equal length, one holding the list of info table pointers and one holding the corresponding entries for each info table. This will allow the entry data to be compressed without losing the references to the info tables.
f3556d6c -
Reference ticket #21766 Adds an `--enable-ipe-data-compreesion` flag to the configure script which will check for libzstd and set the appropriate flags to allow for IPE data compression in the compiler
5d1f2411 -
Reference ticket #21766 When IPE data compression is enabled, compress the emitted IPE buffer entries and decompress them in the RTS.
b7a640ac -
Ensure that `HAVE_LIBZSTD` gets defined to either 0 or 1 in all cases and properly check that before IPE data decompression in the RTS. See ticket #21766.
5aef5658 -
See ticket #21766
69563c97 -
Make sure byte order of written IPE buffer entries matches target. Make sure the IPE-related tests properly access the fields of IPE buffer entry nodes with the new IPE layout. This commit also introduces checks to avoid importing modules if IPE compression is not enabled. See ticket #21766.
7872e2b6 -
Capacity of buffers allocated for decompressed IPE data was incorrect due to a misuse of the `ZSTD_findFrameCompressedSize` function. Fix by always storing decompressed size of IPE data in IPE buffer list nodes and using `ZSTD_findFrameCompressedSize` to determine the size of the compressed data. See ticket #21766
0e85099b -
Changes the configure script to indicate whether libnuma, libzstd, or libdw are being used as dependencies due to their optional features being enabled.
a0048866 -
- Adds an IPE job to the CI pipeline which is triggered by the ~IPE label - Introduces CI logic to enable IPE data compression - Enables uncompressed IPE data on debug CI job - Regenerates jobs.yaml MR ci-images!112 on the images repository is meant to ensure that the proper images have libzstd-dev installed.
09d93bd0 -
Add mention of IPE data compression to user's guide and the release notes for 9.8.1. Also note the impact compression has on binary size in both places. Change IpeBufferListNode compression check so only the value `1` indicates compression. See ticket #21766
3ded9a1c -
We don't need to explicitly specify the +ipe transformer to test IPE data since there are tests which manually enable IPE information. This commit does leave zstd IPE data compression enabled on the debian CI jobs.
41b41577 -
GHC.Tc.Zonk.Type uses an equality constraint. ghc.nix currently provides 9.2.
982bef3a -
-
This refactors the panoply of renamer lookup functions relating to lookupExactOrOrig to more graciously handle Exact and Orig names. In particular, we avoid the situation in which we would add Exact/Orig GREs to the tcg_used_gres field, which could cause a panic in bestImport like in #23240. Fixes #23428
8e81f140 -
As described in Note [The one-shot state monad trick], we shouldn't use derived Functor instances for monads using one-shot. This was done for most of them, but UM and RewriteM were missed.
5d415bfd -
setTestOpts() is used to modify the test options for an entire .T file, rather than a single test. If there was a test using collect_compiler_stats, all of the tests in the same file had lint disabled. Fixes #21247
2c38551e -
-
It is no longer needed since Note [Extra dependencies from .hs-boot files] was deleted in 69987720. I've also added tildes to Note headers, otherwise they're not detected by the linter.
f6e060cc -
-
For reasons similar to those described in Note [Depend on GHC.Num.Integer]. Fixes #23411.
4a4eb761 -
Flag -ddisable-js-minimizer was producing invalid code. Fix that and also a few other things to generate nicer JS code for debugging. The added test checks that we don't regress when using the flag.
f53ac0ae -
f7744e8e
-
-
Matthew Pickering authored
The overall goal of this refactoring is to reduce the dependency footprint of the parser and syntax tree. Good reasons include: - Better module graph parallelisability - Make it easier to migrate error messages without introducing module loops - Philosophically, there's not reason for the AST to depend on half the compiler. One of the key edges which added this dependency was > GHC.Hs.Expr -> GHC.Tc.Types (TcLclEnv) As this in turn depending on TcM which depends on HscEnv and so on. Therefore the goal of this patch is to move `TcLclEnv` out of `GHC.Tc.Types` so that `GHC.Hs.Expr` can import TcLclEnv without incurring a huge dependency chain. The changes in this patch are: * Move TcLclEnv from GHC.Tc.Types to GHC.Tc.Types.LclEnv * Create new smaller modules for the types used in TcLclEnv New Modules: - GHC.Tc.Types.ErrCtxt - GHC.Tc.Types.BasicTypes - GHC.Tc.Types.TH - GHC.Tc.Types.LclEnv - GHC.Tc.Types.CtLocEnv - GHC.Tc.Errors.Types.PromotionErr Removed Boot File: - {-# SOURCE #-} GHC.Tc.Types * Introduce TcLclCtxt, the part of the TcLclEnv which doesn't participate in restoreLclEnv. * Replace TcLclEnv in CtLoc with specific CtLocEnv which is defined in GHC.Tc.Types.CtLocEnv. Use CtLocEnv in Implic and CtLoc to record the location of the implication and constraint. By splitting up TcLclEnv from GHC.Tc.Types we allow GHC.Hs.Expr to no longer depend on the TcM monad and all that entails. Fixes #23389 #234097cd8a61e -
Matthew Pickering authored
This removes the usage of DynFlags from Tc.Utils.TcType so that it no longer depends on GHC.Driver.Session. In general we don't want anything which is a dependency of Language.Haskell.Syntax to depend on GHC.Driver.Session and removing this edge gets us closer to that goal.
3d8d39d1 -
Matthew Pickering authored
This removes edge from GHC.Hs.Pat to GHC.Driver.Session, which makes Language.Haskell.Syntax end up depending on GHC.Driver.Session.
18db5ada -
Matthew Pickering authored12919dd5
-
Matthew Pickering authored
The hole fit plugins are defined in terms of TcM, a type we want to avoid depending on from `GHC.Tc.Errors.Types`. By moving it into its own module we can remove this dependency. It also simplifies the necessary boot file.
eb852371 -
Matthew Pickering authored
This allows `GHC.Driver.DynFlags` to depend on these types without depending on CoreM and hence the entire simplifier pipeline. We can also remove a hs-boot file with this change.
9e5246d7 -
Matthew Pickering authored52d6a7d7
-
Matthew Pickering authored
These are in a separate commit as the improvement to these tests is the cumulative effect of the previous set of patches rather than just the responsibility of the last one in the patchset.
698d160c -
The VarBind constructor of HsBind is only used at the GhcTc stage. This commit makes that explicit by setting the extension field of VarBind to be DataConCantHappen at all other stages. This allows us to delete a dead code path in GHC.HsToCore.Quote.rep_bind, and remove some panics.
58ccf02e -
The existing utils/genprimopcode/gen_bytearray_ops.py was relocated and extended for this purpose. Additionally, hadrian now knows about this script and uses it when generating primops.txt
54b83253 -
Previously we were always adding the Nightly tag, but this led to all the previous builds getting an increasing number of nightly tags over time. Now we just add it once, when we remove the LatestNightly tag.
ecadbc7e -
Simon Peyton Jones authored
This fixes #23323 See (RC4) in Note [Tracking redundant constraints]
57f47497
Showing
- .gitlab-ci.yml 8 additions, 6 deletions.gitlab-ci.yml
- .gitlab/gen_ci.hs 31 additions, 17 deletions.gitlab/gen_ci.hs
- .gitlab/jobs.yaml 135 additions, 77 deletions.gitlab/jobs.yaml
- .gitlab/rel_eng/mk-ghcup-metadata/README.mkd 1 addition, 0 deletions.gitlab/rel_eng/mk-ghcup-metadata/README.mkd
- .gitlab/rel_eng/mk-ghcup-metadata/mk_ghcup_metadata.py 46 additions, 11 deletions.gitlab/rel_eng/mk-ghcup-metadata/mk_ghcup_metadata.py
- compiler/GHC/Builtin/Types/Prim.hs 5 additions, 0 deletionscompiler/GHC/Builtin/Types/Prim.hs
- compiler/GHC/Builtin/bytearray-ops.txt.pp 0 additions, 551 deletionscompiler/GHC/Builtin/bytearray-ops.txt.pp
- compiler/GHC/Builtin/gen_bytearray_addr_access_ops.py 201 additions, 0 deletionscompiler/GHC/Builtin/gen_bytearray_addr_access_ops.py
- compiler/GHC/Builtin/primops.txt.pp 2 additions, 228 deletionscompiler/GHC/Builtin/primops.txt.pp
- compiler/GHC/CmmToAsm.hs 1 addition, 1 deletioncompiler/GHC/CmmToAsm.hs
- compiler/GHC/CmmToLlvm.hs 1 addition, 1 deletioncompiler/GHC/CmmToLlvm.hs
- compiler/GHC/CmmToLlvm/Base.hs 1 addition, 1 deletioncompiler/GHC/CmmToLlvm/Base.hs
- compiler/GHC/Core/LateCC.hs 1 addition, 1 deletioncompiler/GHC/Core/LateCC.hs
- compiler/GHC/Core/Lint.hs 7 additions, 5 deletionscompiler/GHC/Core/Lint.hs
- compiler/GHC/Core/Opt/Arity.hs 2 additions, 12 deletionscompiler/GHC/Core/Opt/Arity.hs
- compiler/GHC/Core/Opt/CallerCC.hs 1 addition, 94 deletionscompiler/GHC/Core/Opt/CallerCC.hs
- compiler/GHC/Core/Opt/CallerCC.hs-boot 0 additions, 8 deletionscompiler/GHC/Core/Opt/CallerCC.hs-boot
- compiler/GHC/Core/Opt/CallerCC/Types.hs 113 additions, 0 deletionscompiler/GHC/Core/Opt/CallerCC/Types.hs
- compiler/GHC/Core/Opt/Pipeline.hs 1 addition, 1 deletioncompiler/GHC/Core/Opt/Pipeline.hs
- compiler/GHC/Core/Opt/SpecConstr.hs 1 addition, 1 deletioncompiler/GHC/Core/Opt/SpecConstr.hs
compiler/GHC/Core/Opt/CallerCC/Types.hs
0 → 100644