[Git][ghc/ghc][wip/warning-for-last-and-init] 33 commits: Extend record-selector usage ticking to all binds using a record field
by Bodigrim (＠Bodigrim) 15 Aug '25
by Bodigrim (＠Bodigrim) 15 Aug '25
15 Aug '25
Bodigrim pushed to branch wip/warning-for-last-and-init at Glasgow Haskell Compiler / GHC
Commits:
62899117 by Florian Ragwitz at 2025-08-13T21:01:34-04:00
Extend record-selector usage ticking to all binds using a record field
This extends the previous handling of ticking for RecordWildCards and
NamedFieldPuns to all var bindings that involve record selectors.
Note that certain patterns such as `Foo{foo = 42}` will currently not tick the
`foo` selector, as ticking is triggered by `HsVar`s.
Closes #26191.
- - - - -
b37b3af7 by Florian Ragwitz at 2025-08-13T21:01:34-04:00
Add release notes for 9.16.1 and move description of latest HPC changes there.
- - - - -
a5e4b7d9 by Ben Gamari at 2025-08-13T21:02:18-04:00
rts: Clarify rationale for undefined atomic wrappers
Since c06e3f46d24ef69f3a3d794f5f604cb8c2a40cbc the RTS has declared
various atomic operation wrappers defined by ghc-internal as undefined.
While the rationale for this isn't clear from the commit message, I
believe that this is necessary due to the unregisterised backend.
Specifically, the code generator will reference these symbols when
compiling RTS Cmm sources.
- - - - -
50842f83 by Andreas Klebinger at 2025-08-13T21:03:01-04:00
Make unexpected LLVM versions a warning rather than an error.
Typically a newer LLVM version *will* work so erroring out if
a user uses a newer LLVM version is too aggressive.
Fixes #25915
- - - - -
c91e2650 by fendor at 2025-08-13T21:03:43-04:00
Store `StackTrace` and `StackSnapshot` in `Backtraces`
Instead of decoding the stack traces when collecting the `Backtraces`,
defer this decoding until actually showing the `Backtraces`.
This allows users to customise how `Backtraces` are displayed by
using a custom implementation of `displayExceptionWithInfo`, overwriting
the default implementation for `Backtraces` (`displayBacktraces`).
- - - - -
dee28cdd by fendor at 2025-08-13T21:03:43-04:00
Allow users to customise the collection of exception annotations
Add a global `CollectExceptionAnnotationMechanism` which determines how
`ExceptionAnnotation`s are collected upon throwing an `Exception`.
This API is exposed via `ghc-experimental`.
By overriding how we collect `Backtraces`, we can control how the
`Backtraces` are displayed to the user by newtyping `Backtraces` and
giving a different instance for `ExceptionAnnotation`.
A concrete use-case for this feature is allowing us to experiment with
alternative stack decoders, without having to modify `base`, which take
additional information from the stack frames.
This commit does not modify how `Backtraces` are currently
collected or displayed.
- - - - -
66024722 by fendor at 2025-08-13T21:03:43-04:00
Expose Backtraces internals from ghc-experimental
Additionally, expose the same API `base:Control.Exception.Backtrace`
to make it easier to use as a drop-in replacement.
- - - - -
a766286f by Reed Mullanix at 2025-08-13T21:04:36-04:00
ghc-internal: Fix naturalAndNot for NB/NS case
When the first argument to `naturalAndNot` is larger than a `Word` and the second is `Word`-sized, `naturalAndNot` will truncate the
result:
```
>>> naturalAndNot ((2 ^ 65) .|. (2 ^ 3)) (2 ^ 3)
0
```
In contrast, `naturalAndNot` does not truncate when both arguments are larger than a `Word`, so this appears to be a bug.
Luckily, the fix is pretty easy: we just need to call `bigNatAndNotWord#` instead of truncating.
Fixes #26230
- - - - -
3506fa7d by Simon Hengel at 2025-08-13T21:05:18-04:00
Report -pgms as a deprecated flag
(instead of reporting an unspecific warning)
Before:
on the commandline: warning:
Object splitting was removed in GHC 8.8
After:
on the commandline: warning: [GHC-53692] [-Wdeprecated-flags]
-pgms is deprecated: Object splitting was removed in GHC 8.8
- - - - -
51c701fe by Zubin Duggal at 2025-08-13T21:06:00-04:00
testsuite: Be more permissive when filtering out GNU_PROPERTY_TYPE linker warnings
The warning text is slightly different with ld.bfd.
Fixes #26249
- - - - -
dfe6f464 by Simon Hengel at 2025-08-13T21:06:43-04:00
Refactoring: Don't misuse `MCDiagnostic` for lint messages
`MCDiagnostic` is meant to be used for compiler diagnostics.
Any code that creates `MCDiagnostic` directly, without going through
`GHC.Driver.Errors.printMessage`, side steps `-fdiagnostics-as-json`
(see e.g. !14475, !14492 !14548).
To avoid this in the future I want to control more narrowly who creates
`MCDiagnostic` (see #24113).
Some parts of the compiler use `MCDiagnostic` purely for formatting
purposes, without creating any real compiler diagnostics. This change
introduces a helper function, `formatDiagnostic`, that can be used in
such cases instead of constructing `MCDiagnostic`.
- - - - -
a8b2fbae by Teo Camarasu at 2025-08-13T21:07:24-04:00
rts: ensure MessageBlackHole.link is always a valid closure
We turn a MessageBlackHole into an StgInd in wakeBlockingQueue().
Therefore it's important that the link field, which becomes the
indirection field, always points to a valid closure.
It's unclear whether it's currently possible for the previous behaviour
to lead to a crash, but it's good to be consistent about this invariant nonetheless.
Co-authored-by: Andreas Klebinger <klebinger.andreas(a)gmx.at>
- - - - -
4021181e by Teo Camarasu at 2025-08-13T21:07:24-04:00
rts: spin if we see a WHITEHOLE in messageBlackHole
When a BLACKHOLE gets cancelled in raiseAsync, we indirect to a THUNK.
GC can then shortcut this, replacing our BLACKHOLE with a fresh THUNK.
This THUNK is not guaranteed to have a valid indirectee field.
If at the same time, a message intended for the previous BLACKHOLE is
processed and concurrently we BLACKHOLE the THUNK, thus temporarily
turning it into a WHITEHOLE, we can get a segfault, since we look at the
undefined indirectee field of the THUNK
The fix is simple: spin if we see a WHITEHOLE, and it will soon be
replaced with a valid BLACKHOLE.
Resolves #26205
- - - - -
1107af89 by Oleg Grenrus at 2025-08-13T21:08:06-04:00
Allow defining HasField instances for naughty fields
Resolves #26295
... as HasField solver doesn't solve for fields with "naughty"
selectors, we could as well allow defining HasField instances for these
fields.
- - - - -
020e7587 by Sylvain Henry at 2025-08-13T21:09:00-04:00
Fix Data.List unqualified import warning
- - - - -
fd811ded by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Make injecting implicit bindings into its own pass
Previously we were injecting "impliicit bindings" (data constructor
worker and wrappers etc)
- both at the end of CoreTidy,
- and at the start of CorePrep
This is unpleasant and confusing. This patch puts it it its own pass,
addImplicitBinds, which runs between the two.
The function `GHC.CoreToStg.AddImplicitBinds.addImplicitBinds` now takes /all/
TyCons, not just the ones for algebraic data types. That change ripples
through to
- corePrepPgm
- doCodeGen
- byteCodeGen
All take [TyCon] which includes all TyCons
- - - - -
9bd7fcc5 by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Implement unary classes
The big change is described exhaustively in
Note [Unary class magic] in GHC.Core.TyCon
Other changes
* We never unbox class dictionaries in worker/wrapper. This has been true for some
time now, but the logic is now centralised in functions in
GHC.Core.Opt.WorkWrap.Utils, namely `canUnboxTyCon`, and `canUnboxArg`
See Note [Do not unbox class dictionaries] in GHC.Core.Opt.WorkWrap.Utils.
* Refactored the `notWorthFloating` logic in GHc.Core.Opt.SetLevels.
I can't remember if I actually changed any behaviour here, but if so it's
only in a corner cases.
* Fixed a bug in `GHC.Core.TyCon.isEnumerationTyCon`, which was wrongly returning
True for (##).
* Remove redundant Role argument to `liftCoSubstWithEx`. It was always
Representational.
* I refactored evidence generation in the constraint solver:
* Made GHC.Tc.Types.Evidence contain better abstactions for evidence
generation.
* I deleted the file `GHC.Tc.Types.EvTerm` and merged its (small) contents
elsewhere. It wasn't paying its way.
* Made evidence for implicit parameters go via a proper abstraction.
* Fix inlineBoringOk; see (IB6) in Note [inlineBoringOk]
This fixes a slowdown in `countdownEffectfulDynLocal`
in the `effectful` library.
Smaller things
* Rename `isDataTyCon` to `isBoxedDataTyCon`.
* GHC.Core.Corecion.liftCoSubstWithEx was only called with Representational role,
so I baked that into the function and removed the argument.
* Get rid of `GHC.Core.TyCon.tyConSingleAlgDataCon_maybe` in favour of calling
`not isNewTyCon` at the call sites; more explicit.
* Refatored `GHC.Core.TyCon.isInjectiveTyCon`; but I don't think I changed its
behaviour
* Moved `decomposeIPPred` to GHC.Core.Predicate
Compile time performance changes:
geo. mean +0.1%
minimum -6.8%
maximum +14.4%
The +14% one is in T21839c, where it seems that a bit more inlining
is taking place. That seems acceptable; and the average change is small
Metric Decrease:
LargeRecord
T12227
T12707
T16577
T21839r
T5642
Metric Increase:
T15164
T21839c
T3294
T5321FD
T5321Fun
WWRec
- - - - -
b4075d71 by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Slight improvement to pre/postInlineUnconditionally
Avoids an extra simplifier iteration
- - - - -
9e443596 by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Fix a long-standing assertion error in normSplitTyConApp_maybe
- - - - -
91310ad0 by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Add comment to coercion optimiser
- - - - -
5b841d82 by Teo Camarasu at 2025-08-14T17:57:56-04:00
template-haskell: move some identifiers from ghc-internal to template-haskell
These identifiers are not used internally by the compiler. Therefore we
have no reason for them to be in ghc-internal.
By moving them to template-haskell, we benefit from it being easier to
change them and we avoid having to build them in stage0.
Resolves #26048
- - - - -
33e2c7e5 by Teo Camarasu at 2025-08-14T17:57:56-04:00
template-haskell: transfer $infix note to public module
This Haddock note should be in the public facing module
- - - - -
2a411fc4 by Sylvain Henry at 2025-08-14T17:59:09-04:00
JS: export HEAP8 symbol (#26290)
Newer Emscripten requires this.
- - - - -
248f78ca by Ben Gamari at 2025-08-14T17:59:51-04:00
users-guide: Drop the THREAD_RUNNABLE event
As of f361281c89fbce42865d8b8b27b0957205366186 it is no longer emitted.
- - - - -
706d33e3 by Recursion Ninja at 2025-08-15T04:12:12-04:00
Resolving issues #20645 and #26109
Correctly sign extending and casting smaller bit width types for LLVM operations:
- bitReverse8#
- bitReverse16#
- bitReverse32#
- byteSwap16#
- byteSwap32#
- pdep8#
- pdep16#
- pext8#
- pext16#
- - - - -
1cdc6f46 by Cheng Shao at 2025-08-15T04:12:56-04:00
hadrian: enforce have_llvm=False for wasm32/js
This patch fixes hadrian to always pass have_llvm=False to the
testsuite driver for wasm32/js targets. These targets don't really
support the LLVM backend, and the optllvm test way doesn't work. We
used to special-case wasm32/js to avoid auto-adding optllvm way in
testsuite/config/ghc, but this is still problematic if someone writes
a new LLVM-related test and uses something like when(have_llvm(),
extra_ways(["optllvm"])). So better just enforce have_llvm=False for
these targets here.
- - - - -
869a90c0 by Mike Pilgrem at 2025-08-15T19:22:34+01:00
Re CLC issue 292 Warn GHC.Internal.List.{init,last} are partial
Also corrects the warning for `tail` to refer to `Data.List.uncons` (like the existing warning for `head`).
In module `Settings.Warnings`, applies `-Wno-unrecognised-warning-flags` `-Wno-x-partial` to the `Cabal`, `filepath`, `hsc2hs`, `hpc`, `parsec`, `text` and `time` packages (outside GHC's repository).
Also bumps submodules.
- - - - -
81d591b3 by Andrew Lelechenko at 2025-08-15T19:22:34+01:00
Wibble
- - - - -
1cd09314 by Andrew Lelechenko at 2025-08-15T19:22:34+01:00
Wobble
- - - - -
d832a1bd by Andrew Lelechenko at 2025-08-15T19:22:34+01:00
Revert GHC.Internal.System.IO
- - - - -
33ce0807 by Andrew Lelechenko at 2025-08-15T19:22:34+01:00
Revert GHC.Core.Utils
- - - - -
22f5ab7e by Andrew Lelechenko at 2025-08-15T19:22:34+01:00
Revert template-haskell
- - - - -
6ed912bc by Andrew Lelechenko at 2025-08-15T19:22:34+01:00
Revert filepath
- - - - -
206 changed files:
- compiler/GHC/Builtin/Types.hs
- compiler/GHC/Builtin/primops.txt.pp
- compiler/GHC/ByteCode/InfoTable.hs
- compiler/GHC/CmmToLlvm/CodeGen.hs
- compiler/GHC/Core/Class.hs
- compiler/GHC/Core/Coercion.hs
- compiler/GHC/Core/Coercion/Opt.hs
- compiler/GHC/Core/DataCon.hs
- compiler/GHC/Core/FamInstEnv.hs
- compiler/GHC/Core/Lint.hs
- compiler/GHC/Core/Opt/Arity.hs
- compiler/GHC/Core/Opt/CprAnal.hs
- compiler/GHC/Core/Opt/DmdAnal.hs
- compiler/GHC/Core/Opt/OccurAnal.hs
- compiler/GHC/Core/Opt/SetLevels.hs
- compiler/GHC/Core/Opt/Simplify/Utils.hs
- compiler/GHC/Core/Opt/Specialise.hs
- compiler/GHC/Core/Opt/WorkWrap/Utils.hs
- compiler/GHC/Core/Predicate.hs
- compiler/GHC/Core/TyCo/Rep.hs
- compiler/GHC/Core/TyCon.hs
- compiler/GHC/Core/Type.hs
- compiler/GHC/Core/Unfold.hs
- compiler/GHC/Core/Unfold/Make.hs
- compiler/GHC/Core/Utils.hs
- compiler/GHC/CoreToStg.hs
- + compiler/GHC/CoreToStg/AddImplicitBinds.hs
- compiler/GHC/CoreToStg/Prep.hs
- compiler/GHC/Driver/Errors/Ppr.hs
- compiler/GHC/Driver/Main.hs
- compiler/GHC/Driver/Session.hs
- compiler/GHC/Driver/Session/Units.hs
- compiler/GHC/HsToCore/Binds.hs
- compiler/GHC/HsToCore/Expr.hs
- compiler/GHC/HsToCore/Foreign/Call.hs
- compiler/GHC/HsToCore/Ticks.hs
- compiler/GHC/Iface/Decl.hs
- compiler/GHC/Iface/Syntax.hs
- compiler/GHC/Iface/Tidy.hs
- compiler/GHC/IfaceToCore.hs
- compiler/GHC/Prelude/Basic.hs
- compiler/GHC/Rename/Module.hs
- compiler/GHC/Runtime/Debugger/Breakpoints.hs
- compiler/GHC/Stg/Lint.hs
- compiler/GHC/StgToByteCode.hs
- compiler/GHC/StgToCmm.hs
- compiler/GHC/Tc/Errors.hs
- compiler/GHC/Tc/Gen/Bind.hs
- compiler/GHC/Tc/Gen/Expr.hs
- compiler/GHC/Tc/Instance/Class.hs
- compiler/GHC/Tc/Instance/Family.hs
- compiler/GHC/Tc/Solver/Default.hs
- compiler/GHC/Tc/Solver/Dict.hs
- compiler/GHC/Tc/Solver/Equality.hs
- compiler/GHC/Tc/Solver/Monad.hs
- compiler/GHC/Tc/Solver/Solve.hs
- compiler/GHC/Tc/TyCl.hs
- compiler/GHC/Tc/TyCl/Build.hs
- compiler/GHC/Tc/TyCl/Instance.hs
- compiler/GHC/Tc/TyCl/PatSyn.hs
- compiler/GHC/Tc/TyCl/Utils.hs
- − compiler/GHC/Tc/Types/EvTerm.hs
- compiler/GHC/Tc/Types/Evidence.hs
- compiler/GHC/Tc/Utils/TcMType.hs
- compiler/GHC/Tc/Validity.hs
- compiler/GHC/Types/Demand.hs
- compiler/GHC/Types/Error.hs
- compiler/GHC/Types/Id.hs
- compiler/GHC/Types/Id/Make.hs
- compiler/GHC/Types/RepType.hs
- compiler/GHC/Types/TyThing.hs
- compiler/GHC/Utils/Error.hs
- compiler/ghc.cabal.in
- − docs/users_guide/9.14.1-notes.rst
- + docs/users_guide/9.16.1-notes.rst
- docs/users_guide/eventlog-formats.rst
- docs/users_guide/release-notes.rst
- ghc/GHCi/UI.hs
- ghc/Main.hs
- hadrian/src/Settings/Builders/RunTest.hs
- hadrian/src/Settings/Warnings.hs
- libraries/base/changelog.md
- libraries/filepath
- libraries/ghc-bignum/changelog.md
- libraries/ghc-boot-th/GHC/Boot/TH/Ppr.hs
- + libraries/ghc-experimental/src/GHC/Exception/Backtrace/Experimental.hs
- libraries/ghc-internal/cbits/pdep.c
- libraries/ghc-internal/cbits/pext.c
- libraries/ghc-internal/src/GHC/Internal/Bignum/Natural.hs
- libraries/ghc-internal/src/GHC/Internal/Exception.hs
- libraries/ghc-internal/src/GHC/Internal/Exception/Backtrace.hs
- libraries/ghc-internal/src/GHC/Internal/Exception/Backtrace.hs-boot
- libraries/ghc-internal/src/GHC/Internal/Float.hs
- libraries/ghc-internal/src/GHC/Internal/List.hs
- libraries/ghc-internal/src/GHC/Internal/System/IO.hs
- libraries/ghc-internal/src/GHC/Internal/TH/Lib.hs
- libraries/ghc-internal/src/GHC/Internal/TH/Lift.hs
- libraries/ghc-internal/src/GHC/Internal/TH/Syntax.hs
- + libraries/ghc-internal/tests/Makefile
- + libraries/ghc-internal/tests/all.T
- + libraries/ghc-internal/tests/backtraces/Makefile
- + libraries/ghc-internal/tests/backtraces/T14532a.hs
- + libraries/ghc-internal/tests/backtraces/T14532a.stdout
- + libraries/ghc-internal/tests/backtraces/T14532b.hs
- + libraries/ghc-internal/tests/backtraces/T14532b.stdout
- + libraries/ghc-internal/tests/backtraces/all.T
- libraries/template-haskell/Language/Haskell/TH/Lib.hs
- libraries/template-haskell/Language/Haskell/TH/Quote.hs
- libraries/template-haskell/Language/Haskell/TH/Syntax.hs
- libraries/template-haskell/tests/all.T
- libraries/template-haskell/vendored-filepath/System/FilePath/Posix.hs
- libraries/template-haskell/vendored-filepath/System/FilePath/Windows.hs
- rts/Messages.c
- rts/StgMiscClosures.cmm
- rts/Updates.h
- rts/external-symbols.list.in
- rts/js/mem.js
- rts/rts.cabal
- testsuite/config/ghc
- testsuite/driver/testlib.py
- testsuite/tests/arrows/should_compile/T21301.stderr
- testsuite/tests/core-to-stg/T24124.stderr
- testsuite/tests/deSugar/should_compile/T2431.stderr
- testsuite/tests/deSugar/should_fail/DsStrictFail.stderr
- testsuite/tests/deSugar/should_run/T20024.stderr
- testsuite/tests/deSugar/should_run/dsrun005.stderr
- testsuite/tests/deSugar/should_run/dsrun007.stderr
- testsuite/tests/deSugar/should_run/dsrun008.stderr
- testsuite/tests/deriving/should_run/T9576.stderr
- testsuite/tests/dmdanal/should_compile/T16029.stdout
- testsuite/tests/dmdanal/sigs/T21119.stderr
- testsuite/tests/dmdanal/sigs/T21888.stderr
- testsuite/tests/driver/j-space/jspace.hs
- testsuite/tests/ghci.debugger/scripts/break011.stdout
- testsuite/tests/ghci.debugger/scripts/break024.stdout
- testsuite/tests/ghci/scripts/Defer02.stderr
- testsuite/tests/ghci/scripts/T15325.stderr
- testsuite/tests/hpc/recsel/recsel.hs
- testsuite/tests/hpc/recsel/recsel.stdout
- testsuite/tests/indexed-types/should_compile/T2238.hs
- testsuite/tests/interface-stability/template-haskell-exports.stdout
- + testsuite/tests/llvm/should_run/T20645.hs
- + testsuite/tests/llvm/should_run/T20645.stdout
- testsuite/tests/llvm/should_run/all.T
- testsuite/tests/numeric/should_compile/T15547.stderr
- testsuite/tests/numeric/should_compile/T23907.stderr
- + testsuite/tests/numeric/should_run/T26230.hs
- + testsuite/tests/numeric/should_run/T26230.stdout
- testsuite/tests/numeric/should_run/all.T
- testsuite/tests/numeric/should_run/foundation.hs
- + testsuite/tests/overloadedrecflds/should_run/T26295.hs
- + testsuite/tests/overloadedrecflds/should_run/T26295.stdout
- testsuite/tests/overloadedrecflds/should_run/all.T
- testsuite/tests/patsyn/should_run/ghci.stderr
- testsuite/tests/quasiquotation/T4491/test.T
- testsuite/tests/quotes/LiftErrMsgDefer.stderr
- testsuite/tests/roles/should_compile/Roles14.stderr
- testsuite/tests/roles/should_compile/Roles3.stderr
- testsuite/tests/roles/should_compile/Roles4.stderr
- testsuite/tests/rts/KeepCafsBase.hs
- testsuite/tests/safeHaskell/safeLanguage/SafeLang15.stderr
- testsuite/tests/simplCore/should_compile/DataToTagFamilyScrut.stderr
- testsuite/tests/simplCore/should_compile/T15205.stderr
- testsuite/tests/simplCore/should_compile/T17366.stderr
- testsuite/tests/simplCore/should_compile/T17966.stderr
- testsuite/tests/simplCore/should_compile/T22309.stderr
- testsuite/tests/simplCore/should_compile/T22375DataFamily.stderr
- testsuite/tests/simplCore/should_compile/T23307.stderr
- testsuite/tests/simplCore/should_compile/T23307a.stderr
- testsuite/tests/simplCore/should_compile/T25389.stderr
- testsuite/tests/simplCore/should_compile/T25713.stderr
- testsuite/tests/simplCore/should_compile/T7360.stderr
- testsuite/tests/simplStg/should_compile/T15226b.stderr
- testsuite/tests/tcplugins/CtIdPlugin.hs
- testsuite/tests/th/Makefile
- testsuite/tests/type-data/should_run/T22332a.stderr
- testsuite/tests/typecheck/should_compile/Makefile
- testsuite/tests/typecheck/should_compile/T12763.stderr
- testsuite/tests/typecheck/should_compile/T14774.stdout
- testsuite/tests/typecheck/should_compile/T18406b.stderr
- testsuite/tests/typecheck/should_compile/T18529.stderr
- testsuite/tests/typecheck/should_compile/all.T
- testsuite/tests/typecheck/should_run/T10284.stderr
- testsuite/tests/typecheck/should_run/T13838.stderr
- testsuite/tests/typecheck/should_run/T9497a-run.stderr
- testsuite/tests/typecheck/should_run/T9497b-run.stderr
- testsuite/tests/typecheck/should_run/T9497c-run.stderr
- testsuite/tests/unboxedsums/unpack_sums_7.stdout
- testsuite/tests/unsatisfiable/T23816.stderr
- testsuite/tests/unsatisfiable/UnsatDefer.stderr
- testsuite/tests/wasm/should_run/control-flow/LoadCmmGroup.hs
- testsuite/tests/wasm/should_run/control-flow/RunWasm.hs
- utils/check-exact/Main.hs
- utils/check-exact/Transform.hs
- utils/check-exact/Utils.hs
- utils/genprimopcode/Lexer.x
- utils/genprimopcode/Main.hs
- utils/genprimopcode/Parser.y
- utils/genprimopcode/ParserM.hs
- utils/genprimopcode/Syntax.hs
- utils/ghc-pkg/Main.hs
- utils/ghc-toolchain/src/GHC/Toolchain/CheckArm.hs
- utils/haddock/haddock-api/src/Haddock/Backends/Xhtml.hs
- utils/haddock/haddock-library/src/Documentation/Haddock/Parser.hs
- utils/hpc
- utils/hsc2hs
The diff was not included because it is too large.
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/3fa7314c859a76e8d976f3d9115eae…
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/3fa7314c859a76e8d976f3d9115eae…
You're receiving this email because of your account on gitlab.haskell.org.
1
0
[Git][ghc/ghc][wip/fix-26065] 27 commits: Extend record-selector usage ticking to all binds using a record field
by recursion-ninja (＠recursion-ninja) 15 Aug '25
by recursion-ninja (＠recursion-ninja) 15 Aug '25
15 Aug '25
recursion-ninja pushed to branch wip/fix-26065 at Glasgow Haskell Compiler / GHC
Commits:
62899117 by Florian Ragwitz at 2025-08-13T21:01:34-04:00
Extend record-selector usage ticking to all binds using a record field
This extends the previous handling of ticking for RecordWildCards and
NamedFieldPuns to all var bindings that involve record selectors.
Note that certain patterns such as `Foo{foo = 42}` will currently not tick the
`foo` selector, as ticking is triggered by `HsVar`s.
Closes #26191.
- - - - -
b37b3af7 by Florian Ragwitz at 2025-08-13T21:01:34-04:00
Add release notes for 9.16.1 and move description of latest HPC changes there.
- - - - -
a5e4b7d9 by Ben Gamari at 2025-08-13T21:02:18-04:00
rts: Clarify rationale for undefined atomic wrappers
Since c06e3f46d24ef69f3a3d794f5f604cb8c2a40cbc the RTS has declared
various atomic operation wrappers defined by ghc-internal as undefined.
While the rationale for this isn't clear from the commit message, I
believe that this is necessary due to the unregisterised backend.
Specifically, the code generator will reference these symbols when
compiling RTS Cmm sources.
- - - - -
50842f83 by Andreas Klebinger at 2025-08-13T21:03:01-04:00
Make unexpected LLVM versions a warning rather than an error.
Typically a newer LLVM version *will* work so erroring out if
a user uses a newer LLVM version is too aggressive.
Fixes #25915
- - - - -
c91e2650 by fendor at 2025-08-13T21:03:43-04:00
Store `StackTrace` and `StackSnapshot` in `Backtraces`
Instead of decoding the stack traces when collecting the `Backtraces`,
defer this decoding until actually showing the `Backtraces`.
This allows users to customise how `Backtraces` are displayed by
using a custom implementation of `displayExceptionWithInfo`, overwriting
the default implementation for `Backtraces` (`displayBacktraces`).
- - - - -
dee28cdd by fendor at 2025-08-13T21:03:43-04:00
Allow users to customise the collection of exception annotations
Add a global `CollectExceptionAnnotationMechanism` which determines how
`ExceptionAnnotation`s are collected upon throwing an `Exception`.
This API is exposed via `ghc-experimental`.
By overriding how we collect `Backtraces`, we can control how the
`Backtraces` are displayed to the user by newtyping `Backtraces` and
giving a different instance for `ExceptionAnnotation`.
A concrete use-case for this feature is allowing us to experiment with
alternative stack decoders, without having to modify `base`, which take
additional information from the stack frames.
This commit does not modify how `Backtraces` are currently
collected or displayed.
- - - - -
66024722 by fendor at 2025-08-13T21:03:43-04:00
Expose Backtraces internals from ghc-experimental
Additionally, expose the same API `base:Control.Exception.Backtrace`
to make it easier to use as a drop-in replacement.
- - - - -
a766286f by Reed Mullanix at 2025-08-13T21:04:36-04:00
ghc-internal: Fix naturalAndNot for NB/NS case
When the first argument to `naturalAndNot` is larger than a `Word` and the second is `Word`-sized, `naturalAndNot` will truncate the
result:
```
>>> naturalAndNot ((2 ^ 65) .|. (2 ^ 3)) (2 ^ 3)
0
```
In contrast, `naturalAndNot` does not truncate when both arguments are larger than a `Word`, so this appears to be a bug.
Luckily, the fix is pretty easy: we just need to call `bigNatAndNotWord#` instead of truncating.
Fixes #26230
- - - - -
3506fa7d by Simon Hengel at 2025-08-13T21:05:18-04:00
Report -pgms as a deprecated flag
(instead of reporting an unspecific warning)
Before:
on the commandline: warning:
Object splitting was removed in GHC 8.8
After:
on the commandline: warning: [GHC-53692] [-Wdeprecated-flags]
-pgms is deprecated: Object splitting was removed in GHC 8.8
- - - - -
51c701fe by Zubin Duggal at 2025-08-13T21:06:00-04:00
testsuite: Be more permissive when filtering out GNU_PROPERTY_TYPE linker warnings
The warning text is slightly different with ld.bfd.
Fixes #26249
- - - - -
dfe6f464 by Simon Hengel at 2025-08-13T21:06:43-04:00
Refactoring: Don't misuse `MCDiagnostic` for lint messages
`MCDiagnostic` is meant to be used for compiler diagnostics.
Any code that creates `MCDiagnostic` directly, without going through
`GHC.Driver.Errors.printMessage`, side steps `-fdiagnostics-as-json`
(see e.g. !14475, !14492 !14548).
To avoid this in the future I want to control more narrowly who creates
`MCDiagnostic` (see #24113).
Some parts of the compiler use `MCDiagnostic` purely for formatting
purposes, without creating any real compiler diagnostics. This change
introduces a helper function, `formatDiagnostic`, that can be used in
such cases instead of constructing `MCDiagnostic`.
- - - - -
a8b2fbae by Teo Camarasu at 2025-08-13T21:07:24-04:00
rts: ensure MessageBlackHole.link is always a valid closure
We turn a MessageBlackHole into an StgInd in wakeBlockingQueue().
Therefore it's important that the link field, which becomes the
indirection field, always points to a valid closure.
It's unclear whether it's currently possible for the previous behaviour
to lead to a crash, but it's good to be consistent about this invariant nonetheless.
Co-authored-by: Andreas Klebinger <klebinger.andreas(a)gmx.at>
- - - - -
4021181e by Teo Camarasu at 2025-08-13T21:07:24-04:00
rts: spin if we see a WHITEHOLE in messageBlackHole
When a BLACKHOLE gets cancelled in raiseAsync, we indirect to a THUNK.
GC can then shortcut this, replacing our BLACKHOLE with a fresh THUNK.
This THUNK is not guaranteed to have a valid indirectee field.
If at the same time, a message intended for the previous BLACKHOLE is
processed and concurrently we BLACKHOLE the THUNK, thus temporarily
turning it into a WHITEHOLE, we can get a segfault, since we look at the
undefined indirectee field of the THUNK
The fix is simple: spin if we see a WHITEHOLE, and it will soon be
replaced with a valid BLACKHOLE.
Resolves #26205
- - - - -
1107af89 by Oleg Grenrus at 2025-08-13T21:08:06-04:00
Allow defining HasField instances for naughty fields
Resolves #26295
... as HasField solver doesn't solve for fields with "naughty"
selectors, we could as well allow defining HasField instances for these
fields.
- - - - -
020e7587 by Sylvain Henry at 2025-08-13T21:09:00-04:00
Fix Data.List unqualified import warning
- - - - -
fd811ded by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Make injecting implicit bindings into its own pass
Previously we were injecting "impliicit bindings" (data constructor
worker and wrappers etc)
- both at the end of CoreTidy,
- and at the start of CorePrep
This is unpleasant and confusing. This patch puts it it its own pass,
addImplicitBinds, which runs between the two.
The function `GHC.CoreToStg.AddImplicitBinds.addImplicitBinds` now takes /all/
TyCons, not just the ones for algebraic data types. That change ripples
through to
- corePrepPgm
- doCodeGen
- byteCodeGen
All take [TyCon] which includes all TyCons
- - - - -
9bd7fcc5 by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Implement unary classes
The big change is described exhaustively in
Note [Unary class magic] in GHC.Core.TyCon
Other changes
* We never unbox class dictionaries in worker/wrapper. This has been true for some
time now, but the logic is now centralised in functions in
GHC.Core.Opt.WorkWrap.Utils, namely `canUnboxTyCon`, and `canUnboxArg`
See Note [Do not unbox class dictionaries] in GHC.Core.Opt.WorkWrap.Utils.
* Refactored the `notWorthFloating` logic in GHc.Core.Opt.SetLevels.
I can't remember if I actually changed any behaviour here, but if so it's
only in a corner cases.
* Fixed a bug in `GHC.Core.TyCon.isEnumerationTyCon`, which was wrongly returning
True for (##).
* Remove redundant Role argument to `liftCoSubstWithEx`. It was always
Representational.
* I refactored evidence generation in the constraint solver:
* Made GHC.Tc.Types.Evidence contain better abstactions for evidence
generation.
* I deleted the file `GHC.Tc.Types.EvTerm` and merged its (small) contents
elsewhere. It wasn't paying its way.
* Made evidence for implicit parameters go via a proper abstraction.
* Fix inlineBoringOk; see (IB6) in Note [inlineBoringOk]
This fixes a slowdown in `countdownEffectfulDynLocal`
in the `effectful` library.
Smaller things
* Rename `isDataTyCon` to `isBoxedDataTyCon`.
* GHC.Core.Corecion.liftCoSubstWithEx was only called with Representational role,
so I baked that into the function and removed the argument.
* Get rid of `GHC.Core.TyCon.tyConSingleAlgDataCon_maybe` in favour of calling
`not isNewTyCon` at the call sites; more explicit.
* Refatored `GHC.Core.TyCon.isInjectiveTyCon`; but I don't think I changed its
behaviour
* Moved `decomposeIPPred` to GHC.Core.Predicate
Compile time performance changes:
geo. mean +0.1%
minimum -6.8%
maximum +14.4%
The +14% one is in T21839c, where it seems that a bit more inlining
is taking place. That seems acceptable; and the average change is small
Metric Decrease:
LargeRecord
T12227
T12707
T16577
T21839r
T5642
Metric Increase:
T15164
T21839c
T3294
T5321FD
T5321Fun
WWRec
- - - - -
b4075d71 by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Slight improvement to pre/postInlineUnconditionally
Avoids an extra simplifier iteration
- - - - -
9e443596 by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Fix a long-standing assertion error in normSplitTyConApp_maybe
- - - - -
91310ad0 by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Add comment to coercion optimiser
- - - - -
5b841d82 by Teo Camarasu at 2025-08-14T17:57:56-04:00
template-haskell: move some identifiers from ghc-internal to template-haskell
These identifiers are not used internally by the compiler. Therefore we
have no reason for them to be in ghc-internal.
By moving them to template-haskell, we benefit from it being easier to
change them and we avoid having to build them in stage0.
Resolves #26048
- - - - -
33e2c7e5 by Teo Camarasu at 2025-08-14T17:57:56-04:00
template-haskell: transfer $infix note to public module
This Haddock note should be in the public facing module
- - - - -
2a411fc4 by Sylvain Henry at 2025-08-14T17:59:09-04:00
JS: export HEAP8 symbol (#26290)
Newer Emscripten requires this.
- - - - -
248f78ca by Ben Gamari at 2025-08-14T17:59:51-04:00
users-guide: Drop the THREAD_RUNNABLE event
As of f361281c89fbce42865d8b8b27b0957205366186 it is no longer emitted.
- - - - -
706d33e3 by Recursion Ninja at 2025-08-15T04:12:12-04:00
Resolving issues #20645 and #26109
Correctly sign extending and casting smaller bit width types for LLVM operations:
- bitReverse8#
- bitReverse16#
- bitReverse32#
- byteSwap16#
- byteSwap32#
- pdep8#
- pdep16#
- pext8#
- pext16#
- - - - -
1cdc6f46 by Cheng Shao at 2025-08-15T04:12:56-04:00
hadrian: enforce have_llvm=False for wasm32/js
This patch fixes hadrian to always pass have_llvm=False to the
testsuite driver for wasm32/js targets. These targets don't really
support the LLVM backend, and the optllvm test way doesn't work. We
used to special-case wasm32/js to avoid auto-adding optllvm way in
testsuite/config/ghc, but this is still problematic if someone writes
a new LLVM-related test and uses something like when(have_llvm(),
extra_ways(["optllvm"])). So better just enforce have_llvm=False for
these targets here.
- - - - -
553650d9 by Alex Washburn at 2025-08-15T13:26:04-04:00
Correcting LLVM linking of Intel BMI intrinsics pdep{8,16} and pext{8,16}.
This patch fixes #26045.
The LLVM interface does not expose bindings to:
- llvm.x86.bmi.pdep.8
- llvm.x86.bmi.pdep.16
- llvm.x86.bmi.pext.8
- llvm.x86.bmi.pext.16
So calls are instead made to llvm.x86.bmi.{pdep,pext}.32 in these cases,
with pre/post-operation truncation to constrain the logical value range.
- - - - -
185 changed files:
- compiler/GHC/Builtin/Types.hs
- compiler/GHC/Builtin/primops.txt.pp
- compiler/GHC/ByteCode/InfoTable.hs
- compiler/GHC/CmmToLlvm/CodeGen.hs
- compiler/GHC/Core/Class.hs
- compiler/GHC/Core/Coercion.hs
- compiler/GHC/Core/Coercion/Opt.hs
- compiler/GHC/Core/DataCon.hs
- compiler/GHC/Core/FamInstEnv.hs
- compiler/GHC/Core/Lint.hs
- compiler/GHC/Core/Opt/Arity.hs
- compiler/GHC/Core/Opt/CprAnal.hs
- compiler/GHC/Core/Opt/DmdAnal.hs
- compiler/GHC/Core/Opt/OccurAnal.hs
- compiler/GHC/Core/Opt/SetLevels.hs
- compiler/GHC/Core/Opt/Simplify/Utils.hs
- compiler/GHC/Core/Opt/Specialise.hs
- compiler/GHC/Core/Opt/WorkWrap/Utils.hs
- compiler/GHC/Core/Predicate.hs
- compiler/GHC/Core/TyCo/Rep.hs
- compiler/GHC/Core/TyCon.hs
- compiler/GHC/Core/Type.hs
- compiler/GHC/Core/Unfold.hs
- compiler/GHC/Core/Unfold/Make.hs
- compiler/GHC/Core/Utils.hs
- compiler/GHC/CoreToStg.hs
- + compiler/GHC/CoreToStg/AddImplicitBinds.hs
- compiler/GHC/CoreToStg/Prep.hs
- compiler/GHC/Driver/Errors/Ppr.hs
- compiler/GHC/Driver/Main.hs
- compiler/GHC/Driver/Session.hs
- compiler/GHC/HsToCore/Binds.hs
- compiler/GHC/HsToCore/Expr.hs
- compiler/GHC/HsToCore/Foreign/Call.hs
- compiler/GHC/HsToCore/Ticks.hs
- compiler/GHC/Iface/Decl.hs
- compiler/GHC/Iface/Syntax.hs
- compiler/GHC/Iface/Tidy.hs
- compiler/GHC/IfaceToCore.hs
- compiler/GHC/Rename/Module.hs
- compiler/GHC/Runtime/Debugger/Breakpoints.hs
- compiler/GHC/Stg/Lint.hs
- compiler/GHC/StgToByteCode.hs
- compiler/GHC/StgToCmm.hs
- compiler/GHC/Tc/Errors.hs
- compiler/GHC/Tc/Gen/Bind.hs
- compiler/GHC/Tc/Gen/Expr.hs
- compiler/GHC/Tc/Instance/Class.hs
- compiler/GHC/Tc/Instance/Family.hs
- compiler/GHC/Tc/Solver/Default.hs
- compiler/GHC/Tc/Solver/Dict.hs
- compiler/GHC/Tc/Solver/Equality.hs
- compiler/GHC/Tc/Solver/Monad.hs
- compiler/GHC/Tc/Solver/Solve.hs
- compiler/GHC/Tc/TyCl.hs
- compiler/GHC/Tc/TyCl/Build.hs
- compiler/GHC/Tc/TyCl/Instance.hs
- compiler/GHC/Tc/TyCl/PatSyn.hs
- compiler/GHC/Tc/TyCl/Utils.hs
- − compiler/GHC/Tc/Types/EvTerm.hs
- compiler/GHC/Tc/Types/Evidence.hs
- compiler/GHC/Tc/Utils/TcMType.hs
- compiler/GHC/Tc/Validity.hs
- compiler/GHC/Types/Demand.hs
- compiler/GHC/Types/Error.hs
- compiler/GHC/Types/Id.hs
- compiler/GHC/Types/Id/Make.hs
- compiler/GHC/Types/RepType.hs
- compiler/GHC/Types/TyThing.hs
- compiler/GHC/Utils/Error.hs
- compiler/ghc.cabal.in
- − docs/users_guide/9.14.1-notes.rst
- + docs/users_guide/9.16.1-notes.rst
- docs/users_guide/eventlog-formats.rst
- docs/users_guide/release-notes.rst
- hadrian/src/Settings/Builders/RunTest.hs
- libraries/base/changelog.md
- libraries/ghc-bignum/changelog.md
- + libraries/ghc-experimental/src/GHC/Exception/Backtrace/Experimental.hs
- libraries/ghc-internal/cbits/pdep.c
- libraries/ghc-internal/cbits/pext.c
- libraries/ghc-internal/src/GHC/Internal/Bignum/Natural.hs
- libraries/ghc-internal/src/GHC/Internal/Exception.hs
- libraries/ghc-internal/src/GHC/Internal/Exception/Backtrace.hs
- libraries/ghc-internal/src/GHC/Internal/Exception/Backtrace.hs-boot
- libraries/ghc-internal/src/GHC/Internal/TH/Lib.hs
- libraries/ghc-internal/src/GHC/Internal/TH/Lift.hs
- libraries/ghc-internal/src/GHC/Internal/TH/Syntax.hs
- + libraries/ghc-internal/tests/Makefile
- + libraries/ghc-internal/tests/all.T
- + libraries/ghc-internal/tests/backtraces/Makefile
- + libraries/ghc-internal/tests/backtraces/T14532a.hs
- + libraries/ghc-internal/tests/backtraces/T14532a.stdout
- + libraries/ghc-internal/tests/backtraces/T14532b.hs
- + libraries/ghc-internal/tests/backtraces/T14532b.stdout
- + libraries/ghc-internal/tests/backtraces/all.T
- libraries/template-haskell/Language/Haskell/TH/Lib.hs
- libraries/template-haskell/Language/Haskell/TH/Quote.hs
- libraries/template-haskell/Language/Haskell/TH/Syntax.hs
- libraries/template-haskell/tests/all.T
- rts/Messages.c
- rts/StgMiscClosures.cmm
- rts/Updates.h
- rts/external-symbols.list.in
- rts/js/mem.js
- rts/rts.cabal
- testsuite/config/ghc
- testsuite/driver/testlib.py
- testsuite/tests/arrows/should_compile/T21301.stderr
- testsuite/tests/core-to-stg/T24124.stderr
- testsuite/tests/deSugar/should_compile/T2431.stderr
- testsuite/tests/deSugar/should_fail/DsStrictFail.stderr
- testsuite/tests/deSugar/should_run/T20024.stderr
- testsuite/tests/deSugar/should_run/dsrun005.stderr
- testsuite/tests/deSugar/should_run/dsrun007.stderr
- testsuite/tests/deSugar/should_run/dsrun008.stderr
- testsuite/tests/deriving/should_run/T9576.stderr
- testsuite/tests/dmdanal/should_compile/T16029.stdout
- testsuite/tests/dmdanal/sigs/T21119.stderr
- testsuite/tests/dmdanal/sigs/T21888.stderr
- testsuite/tests/ghci.debugger/scripts/break011.stdout
- testsuite/tests/ghci.debugger/scripts/break024.stdout
- testsuite/tests/ghci/scripts/Defer02.stderr
- testsuite/tests/ghci/scripts/T15325.stderr
- testsuite/tests/hpc/recsel/recsel.hs
- testsuite/tests/hpc/recsel/recsel.stdout
- testsuite/tests/indexed-types/should_compile/T2238.hs
- testsuite/tests/interface-stability/template-haskell-exports.stdout
- + testsuite/tests/llvm/should_run/T20645.hs
- + testsuite/tests/llvm/should_run/T20645.stdout
- + testsuite/tests/llvm/should_run/T26065.hs
- + testsuite/tests/llvm/should_run/T26065.stdout
- testsuite/tests/llvm/should_run/all.T
- testsuite/tests/numeric/should_compile/T15547.stderr
- testsuite/tests/numeric/should_compile/T23907.stderr
- + testsuite/tests/numeric/should_run/T26230.hs
- + testsuite/tests/numeric/should_run/T26230.stdout
- testsuite/tests/numeric/should_run/all.T
- testsuite/tests/numeric/should_run/foundation.hs
- + testsuite/tests/overloadedrecflds/should_run/T26295.hs
- + testsuite/tests/overloadedrecflds/should_run/T26295.stdout
- testsuite/tests/overloadedrecflds/should_run/all.T
- testsuite/tests/patsyn/should_run/ghci.stderr
- testsuite/tests/quasiquotation/T4491/test.T
- testsuite/tests/quotes/LiftErrMsgDefer.stderr
- testsuite/tests/roles/should_compile/Roles14.stderr
- testsuite/tests/roles/should_compile/Roles3.stderr
- testsuite/tests/roles/should_compile/Roles4.stderr
- testsuite/tests/safeHaskell/safeLanguage/SafeLang15.stderr
- testsuite/tests/simplCore/should_compile/DataToTagFamilyScrut.stderr
- testsuite/tests/simplCore/should_compile/T15205.stderr
- testsuite/tests/simplCore/should_compile/T17366.stderr
- testsuite/tests/simplCore/should_compile/T17966.stderr
- testsuite/tests/simplCore/should_compile/T22309.stderr
- testsuite/tests/simplCore/should_compile/T22375DataFamily.stderr
- testsuite/tests/simplCore/should_compile/T23307.stderr
- testsuite/tests/simplCore/should_compile/T23307a.stderr
- testsuite/tests/simplCore/should_compile/T25389.stderr
- testsuite/tests/simplCore/should_compile/T25713.stderr
- testsuite/tests/simplCore/should_compile/T7360.stderr
- testsuite/tests/simplStg/should_compile/T15226b.stderr
- testsuite/tests/tcplugins/CtIdPlugin.hs
- testsuite/tests/th/Makefile
- testsuite/tests/type-data/should_run/T22332a.stderr
- testsuite/tests/typecheck/should_compile/Makefile
- testsuite/tests/typecheck/should_compile/T12763.stderr
- testsuite/tests/typecheck/should_compile/T14774.stdout
- testsuite/tests/typecheck/should_compile/T18406b.stderr
- testsuite/tests/typecheck/should_compile/T18529.stderr
- testsuite/tests/typecheck/should_compile/all.T
- testsuite/tests/typecheck/should_run/T10284.stderr
- testsuite/tests/typecheck/should_run/T13838.stderr
- testsuite/tests/typecheck/should_run/T9497a-run.stderr
- testsuite/tests/typecheck/should_run/T9497b-run.stderr
- testsuite/tests/typecheck/should_run/T9497c-run.stderr
- testsuite/tests/unboxedsums/unpack_sums_7.stdout
- testsuite/tests/unsatisfiable/T23816.stderr
- testsuite/tests/unsatisfiable/UnsatDefer.stderr
- testsuite/tests/wasm/should_run/control-flow/LoadCmmGroup.hs
- testsuite/tests/wasm/should_run/control-flow/RunWasm.hs
- utils/genprimopcode/Lexer.x
- utils/genprimopcode/Main.hs
- utils/genprimopcode/Parser.y
- utils/genprimopcode/ParserM.hs
- utils/genprimopcode/Syntax.hs
The diff was not included because it is too large.
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/344decf713237110e6b1b51483292b…
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/344decf713237110e6b1b51483292b…
You're receiving this email because of your account on gitlab.haskell.org.
1
0
[Git][ghc/ghc][wip/backports-9.14] 2 commits: Fix a long-standing assertion error in normSplitTyConApp_maybe
by Ben Gamari (＠bgamari) 15 Aug '25
by Ben Gamari (＠bgamari) 15 Aug '25
15 Aug '25
Ben Gamari pushed to branch wip/backports-9.14 at Glasgow Haskell Compiler / GHC
Commits:
e09e1443 by Simon Peyton Jones at 2025-08-15T12:25:41-04:00
Fix a long-standing assertion error in normSplitTyConApp_maybe
(cherry picked from commit 60db2c66b870bac28615d272936bef6905befdbb)
- - - - -
7b814248 by Simon Peyton Jones at 2025-08-15T12:25:50-04:00
Add comment to coercion optimiser
(cherry picked from commit b2b95dce34ce7f6a8f0acdd637cbe2cafc56c922)
- - - - -
2 changed files:
- compiler/GHC/Core/Coercion/Opt.hs
- compiler/GHC/Core/Utils.hs
Changes:
=====================================
compiler/GHC/Core/Coercion/Opt.hs
=====================================
@@ -208,10 +208,12 @@ optCoercion' env co
in
warnPprTrace (not (isReflCo out_co) && isReflexiveCo out_co)
"optCoercion: reflexive but not refl" details $
--- assertPpr (substTyUnchecked env in_ty1 `eqType` out_ty1 &&
--- substTyUnchecked env in_ty2 `eqType` out_ty2 &&
--- in_role == out_role)
--- (hang (text "optCoercion changed types!") 2 details) $
+ -- The coercion optimiser should usually optimise
+ -- co:ty~ty --> Refl ty
+ -- But given a silly `newtype N = MkN N`, the axiom has type (N ~ N),
+ -- and so that can trigger this warning (e.g. test str002).
+ -- Maybe we should optimise that coercion to (Refl N), but it
+ -- just doesn't seem worth the bother
out_co
| otherwise
=====================================
compiler/GHC/Core/Utils.hs
=====================================
@@ -2730,8 +2730,10 @@ normSplitTyConApp_maybe fam_envs ty
| let Reduction co ty1 = topNormaliseType_maybe fam_envs ty
`orElse` (mkReflRedn Representational ty)
, Just (tc, tc_args) <- splitTyConApp_maybe ty1
- = assertPpr (not (isNewTyCon tc)) (ppr ty $$ ppr ty1) $ -- Check post-condition
- Just (tc, tc_args, co)
+ , not (isNewTyCon tc) -- How can tc be a newtype, after `topNormaliseType`?
+ -- Answer: if it is a recursive newtype, `topNormaliseType`
+ -- may be a no-op. Example: tc226
+ = Just (tc, tc_args, co)
normSplitTyConApp_maybe _ _ = Nothing
{-
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/779a7bc6c1efa07e0cdf8eb125936b…
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/779a7bc6c1efa07e0cdf8eb125936b…
You're receiving this email because of your account on gitlab.haskell.org.
1
0
[Git][ghc/ghc][wip/symbolizer] 4 commits: Resolving issues #20645 and #26109
by Cheng Shao (＠TerrorJack) 15 Aug '25
by Cheng Shao (＠TerrorJack) 15 Aug '25
15 Aug '25
Cheng Shao pushed to branch wip/symbolizer at Glasgow Haskell Compiler / GHC
Commits:
706d33e3 by Recursion Ninja at 2025-08-15T04:12:12-04:00
Resolving issues #20645 and #26109
Correctly sign extending and casting smaller bit width types for LLVM operations:
- bitReverse8#
- bitReverse16#
- bitReverse32#
- byteSwap16#
- byteSwap32#
- pdep8#
- pdep16#
- pext8#
- pext16#
- - - - -
1cdc6f46 by Cheng Shao at 2025-08-15T04:12:56-04:00
hadrian: enforce have_llvm=False for wasm32/js
This patch fixes hadrian to always pass have_llvm=False to the
testsuite driver for wasm32/js targets. These targets don't really
support the LLVM backend, and the optllvm test way doesn't work. We
used to special-case wasm32/js to avoid auto-adding optllvm way in
testsuite/config/ghc, but this is still problematic if someone writes
a new LLVM-related test and uses something like when(have_llvm(),
extra_ways(["optllvm"])). So better just enforce have_llvm=False for
these targets here.
- - - - -
11f9545d by Cheng Shao at 2025-08-15T17:29:06+02:00
rts: remove libbfd logic
- - - - -
2924dad4 by Cheng Shao at 2025-08-15T17:29:06+02:00
compiler/rts: add debug symbolizer
- - - - -
32 changed files:
- compiler/GHC/Builtin/primops.txt.pp
- + compiler/GHC/Cmm/GenerateDebugSymbolStub.hs
- compiler/GHC/CmmToLlvm/CodeGen.hs
- compiler/GHC/Driver/CodeOutput.hs
- compiler/GHC/Driver/Main.hs
- compiler/ghc.cabal.in
- configure.ac
- hadrian/cfg/system.config.in
- hadrian/src/Oracles/Flag.hs
- hadrian/src/Settings/Builders/RunTest.hs
- hadrian/src/Settings/Packages.hs
- libraries/ghc-internal/cbits/pdep.c
- libraries/ghc-internal/cbits/pext.c
- − m4/fp_bfd_support.m4
- rts/Printer.c
- rts/Printer.h
- rts/RtsStartup.c
- rts/configure.ac
- rts/include/Rts.h
- rts/include/rts/Config.h
- + rts/include/rts/Debug.h
- rts/rts.cabal
- testsuite/config/ghc
- + testsuite/tests/llvm/should_run/T20645.hs
- + testsuite/tests/llvm/should_run/T20645.stdout
- testsuite/tests/llvm/should_run/all.T
- testsuite/tests/numeric/should_run/foundation.hs
- utils/genprimopcode/Lexer.x
- utils/genprimopcode/Main.hs
- utils/genprimopcode/Parser.y
- utils/genprimopcode/ParserM.hs
- utils/genprimopcode/Syntax.hs
The diff was not included because it is too large.
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/1c5f14f2bab4f6fa9e84fa911bc34c…
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/1c5f14f2bab4f6fa9e84fa911bc34c…
You're receiving this email because of your account on gitlab.haskell.org.
1
0
[Git][ghc/ghc][wip/backports-9.14] 2 commits: Make injecting implicit bindings into its own pass
by Ben Gamari (＠bgamari) 15 Aug '25
by Ben Gamari (＠bgamari) 15 Aug '25
15 Aug '25
Ben Gamari pushed to branch wip/backports-9.14 at Glasgow Haskell Compiler / GHC
Commits:
5194ccbc by Simon Peyton Jones at 2025-08-15T09:07:30-04:00
Make injecting implicit bindings into its own pass
Previously we were injecting "impliicit bindings" (data constructor
worker and wrappers etc)
- both at the end of CoreTidy,
- and at the start of CorePrep
This is unpleasant and confusing. This patch puts it it its own pass,
addImplicitBinds, which runs between the two.
The function `GHC.CoreToStg.AddImplicitBinds.addImplicitBinds` now takes /all/
TyCons, not just the ones for algebraic data types. That change ripples
through to
- corePrepPgm
- doCodeGen
- byteCodeGen
All take [TyCon] which includes all TyCons
(cherry picked from commit fd811ded65bb9b19571ba525fb5eaf8f23a4533d)
- - - - -
779a7bc6 by Simon Peyton Jones at 2025-08-15T09:07:33-04:00
Implement unary classes
The big change is described exhaustively in
Note [Unary class magic] in GHC.Core.TyCon
Other changes
* We never unbox class dictionaries in worker/wrapper. This has been true for some
time now, but the logic is now centralised in functions in
GHC.Core.Opt.WorkWrap.Utils, namely `canUnboxTyCon`, and `canUnboxArg`
See Note [Do not unbox class dictionaries] in GHC.Core.Opt.WorkWrap.Utils.
* Refactored the `notWorthFloating` logic in GHc.Core.Opt.SetLevels.
I can't remember if I actually changed any behaviour here, but if so it's
only in a corner cases.
* Fixed a bug in `GHC.Core.TyCon.isEnumerationTyCon`, which was wrongly returning
True for (##).
* Remove redundant Role argument to `liftCoSubstWithEx`. It was always
Representational.
* I refactored evidence generation in the constraint solver:
* Made GHC.Tc.Types.Evidence contain better abstactions for evidence
generation.
* I deleted the file `GHC.Tc.Types.EvTerm` and merged its (small) contents
elsewhere. It wasn't paying its way.
* Made evidence for implicit parameters go via a proper abstraction.
* Fix inlineBoringOk; see (IB6) in Note [inlineBoringOk]
This fixes a slowdown in `countdownEffectfulDynLocal`
in the `effectful` library.
Smaller things
* Rename `isDataTyCon` to `isBoxedDataTyCon`.
* GHC.Core.Corecion.liftCoSubstWithEx was only called with Representational role,
so I baked that into the function and removed the argument.
* Get rid of `GHC.Core.TyCon.tyConSingleAlgDataCon_maybe` in favour of calling
`not isNewTyCon` at the call sites; more explicit.
* Refatored `GHC.Core.TyCon.isInjectiveTyCon`; but I don't think I changed its
behaviour
* Moved `decomposeIPPred` to GHC.Core.Predicate
Compile time performance changes:
geo. mean +0.1%
minimum -6.8%
maximum +14.4%
The +14% one is in T21839c, where it seems that a bit more inlining
is taking place. That seems acceptable; and the average change is small
Metric Decrease:
LargeRecord
T12227
T12707
T16577
T21839r
T5642
Metric Increase:
T15164
T21839c
T3294
T5321FD
T5321Fun
WWRec
(cherry picked from commit 9bd7fcc518111a1549c98720c222cdbabd32ed46)
- - - - -
93 changed files:
- compiler/GHC/Builtin/Types.hs
- compiler/GHC/ByteCode/InfoTable.hs
- compiler/GHC/Core/Class.hs
- compiler/GHC/Core/Coercion.hs
- compiler/GHC/Core/DataCon.hs
- compiler/GHC/Core/FamInstEnv.hs
- compiler/GHC/Core/Opt/Arity.hs
- compiler/GHC/Core/Opt/CprAnal.hs
- compiler/GHC/Core/Opt/DmdAnal.hs
- compiler/GHC/Core/Opt/OccurAnal.hs
- compiler/GHC/Core/Opt/SetLevels.hs
- compiler/GHC/Core/Opt/Specialise.hs
- compiler/GHC/Core/Opt/WorkWrap/Utils.hs
- compiler/GHC/Core/Predicate.hs
- compiler/GHC/Core/TyCo/Rep.hs
- compiler/GHC/Core/TyCon.hs
- compiler/GHC/Core/Type.hs
- compiler/GHC/Core/Unfold.hs
- compiler/GHC/Core/Unfold/Make.hs
- compiler/GHC/Core/Utils.hs
- compiler/GHC/CoreToStg.hs
- + compiler/GHC/CoreToStg/AddImplicitBinds.hs
- compiler/GHC/CoreToStg/Prep.hs
- compiler/GHC/Driver/Main.hs
- compiler/GHC/HsToCore/Binds.hs
- compiler/GHC/HsToCore/Expr.hs
- compiler/GHC/HsToCore/Foreign/Call.hs
- compiler/GHC/Iface/Decl.hs
- compiler/GHC/Iface/Syntax.hs
- compiler/GHC/Iface/Tidy.hs
- compiler/GHC/IfaceToCore.hs
- compiler/GHC/Rename/Module.hs
- compiler/GHC/StgToByteCode.hs
- compiler/GHC/StgToCmm.hs
- compiler/GHC/Tc/Errors.hs
- compiler/GHC/Tc/Gen/Bind.hs
- compiler/GHC/Tc/Gen/Expr.hs
- compiler/GHC/Tc/Instance/Class.hs
- compiler/GHC/Tc/Instance/Family.hs
- compiler/GHC/Tc/Solver/Default.hs
- compiler/GHC/Tc/Solver/Dict.hs
- compiler/GHC/Tc/Solver/Equality.hs
- compiler/GHC/Tc/Solver/Monad.hs
- compiler/GHC/Tc/Solver/Solve.hs
- compiler/GHC/Tc/TyCl.hs
- compiler/GHC/Tc/TyCl/Build.hs
- compiler/GHC/Tc/TyCl/Instance.hs
- compiler/GHC/Tc/TyCl/PatSyn.hs
- compiler/GHC/Tc/TyCl/Utils.hs
- − compiler/GHC/Tc/Types/EvTerm.hs
- compiler/GHC/Tc/Types/Evidence.hs
- compiler/GHC/Tc/Utils/TcMType.hs
- compiler/GHC/Types/Demand.hs
- compiler/GHC/Types/Id.hs
- compiler/GHC/Types/Id/Make.hs
- compiler/GHC/Types/RepType.hs
- compiler/GHC/Types/TyThing.hs
- compiler/ghc.cabal.in
- testsuite/tests/core-to-stg/T24124.stderr
- testsuite/tests/deSugar/should_compile/T2431.stderr
- testsuite/tests/dmdanal/should_compile/T16029.stdout
- testsuite/tests/dmdanal/sigs/T21119.stderr
- testsuite/tests/dmdanal/sigs/T21888.stderr
- testsuite/tests/ghci.debugger/scripts/break011.stdout
- testsuite/tests/ghci.debugger/scripts/break024.stdout
- testsuite/tests/indexed-types/should_compile/T2238.hs
- testsuite/tests/numeric/should_compile/T15547.stderr
- testsuite/tests/numeric/should_compile/T23907.stderr
- testsuite/tests/roles/should_compile/Roles14.stderr
- testsuite/tests/roles/should_compile/Roles3.stderr
- testsuite/tests/roles/should_compile/Roles4.stderr
- testsuite/tests/simplCore/should_compile/DataToTagFamilyScrut.stderr
- testsuite/tests/simplCore/should_compile/T15205.stderr
- testsuite/tests/simplCore/should_compile/T17366.stderr
- testsuite/tests/simplCore/should_compile/T17966.stderr
- testsuite/tests/simplCore/should_compile/T22309.stderr
- testsuite/tests/simplCore/should_compile/T22375DataFamily.stderr
- testsuite/tests/simplCore/should_compile/T23307.stderr
- testsuite/tests/simplCore/should_compile/T23307a.stderr
- testsuite/tests/simplCore/should_compile/T25389.stderr
- testsuite/tests/simplCore/should_compile/T25713.stderr
- testsuite/tests/simplCore/should_compile/T7360.stderr
- testsuite/tests/simplStg/should_compile/T15226b.stderr
- testsuite/tests/tcplugins/CtIdPlugin.hs
- testsuite/tests/typecheck/should_compile/Makefile
- testsuite/tests/typecheck/should_compile/T12763.stderr
- testsuite/tests/typecheck/should_compile/T14774.stdout
- testsuite/tests/typecheck/should_compile/T18406b.stderr
- testsuite/tests/typecheck/should_compile/T18529.stderr
- testsuite/tests/typecheck/should_compile/all.T
- testsuite/tests/unboxedsums/unpack_sums_7.stdout
- testsuite/tests/wasm/should_run/control-flow/LoadCmmGroup.hs
- testsuite/tests/wasm/should_run/control-flow/RunWasm.hs
The diff was not included because it is too large.
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/58d1744f720a259a82768282b71e62…
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/58d1744f720a259a82768282b71e62…
You're receiving this email because of your account on gitlab.haskell.org.
1
0
[Git][ghc/ghc][wip/backports-9.14] 2 commits: Treat tuple dictionaries uniformly; don't unbox them
by Ben Gamari (＠bgamari) 15 Aug '25
by Ben Gamari (＠bgamari) 15 Aug '25
15 Aug '25
Ben Gamari pushed to branch wip/backports-9.14 at Glasgow Haskell Compiler / GHC
Commits:
bdf66b51 by Simon Peyton Jones at 2025-08-15T06:26:00-04:00
Treat tuple dictionaries uniformly; don't unbox them
See `Note [Do not unbox class dictionaries]` in DmdAnal.hs,
sep (DNB1).
This MR reverses the plan in #23398, which suggested a special case to
unbox tuple dictionaries in worker/wrapper. But:
- This was the cause of a pile of complexity in the specialiser (#26158)
- Even with that complexity, specialision was still bad, very bad
See https://gitlab.haskell.org/ghc/ghc/-/issues/19747#note_626297
And it's entirely unnecessary! Specialision works fine without
unboxing tuple dictionaries.
(cherry picked from commit ca7a9d421f602355822b674dd01fb48cd6291108)
- - - - -
58d1744f by Simon Peyton Jones at 2025-08-15T06:27:36-04:00
Implement unary classes
The big change is described exhaustively in
Note [Unary class magic] in GHC.Core.TyCon
Other changes
* We never unbox class dictionaries in worker/wrapper. This has been true for some
time now, but the logic is now centralised in functions in
GHC.Core.Opt.WorkWrap.Utils, namely `canUnboxTyCon`, and `canUnboxArg`
See Note [Do not unbox class dictionaries] in GHC.Core.Opt.WorkWrap.Utils.
* Refactored the `notWorthFloating` logic in GHc.Core.Opt.SetLevels.
I can't remember if I actually changed any behaviour here, but if so it's
only in a corner cases.
* Fixed a bug in `GHC.Core.TyCon.isEnumerationTyCon`, which was wrongly returning
True for (##).
* Remove redundant Role argument to `liftCoSubstWithEx`. It was always
Representational.
* I refactored evidence generation in the constraint solver:
* Made GHC.Tc.Types.Evidence contain better abstactions for evidence
generation.
* I deleted the file `GHC.Tc.Types.EvTerm` and merged its (small) contents
elsewhere. It wasn't paying its way.
* Made evidence for implicit parameters go via a proper abstraction.
* Fix inlineBoringOk; see (IB6) in Note [inlineBoringOk]
This fixes a slowdown in `countdownEffectfulDynLocal`
in the `effectful` library.
Smaller things
* Rename `isDataTyCon` to `isBoxedDataTyCon`.
* GHC.Core.Corecion.liftCoSubstWithEx was only called with Representational role,
so I baked that into the function and removed the argument.
* Get rid of `GHC.Core.TyCon.tyConSingleAlgDataCon_maybe` in favour of calling
`not isNewTyCon` at the call sites; more explicit.
* Refatored `GHC.Core.TyCon.isInjectiveTyCon`; but I don't think I changed its
behaviour
* Moved `decomposeIPPred` to GHC.Core.Predicate
Compile time performance changes:
geo. mean +0.1%
minimum -6.8%
maximum +14.4%
The +14% one is in T21839c, where it seems that a bit more inlining
is taking place. That seems acceptable; and the average change is small
Metric Decrease:
LargeRecord
T12227
T12707
T16577
T21839r
T5642
Metric Increase:
T15164
T21839c
T3294
T5321FD
T5321Fun
WWRec
(cherry picked from commit 9bd7fcc518111a1549c98720c222cdbabd32ed46)
- - - - -
85 changed files:
- compiler/GHC/Builtin/Types.hs
- compiler/GHC/Core/Class.hs
- compiler/GHC/Core/Coercion.hs
- compiler/GHC/Core/DataCon.hs
- compiler/GHC/Core/FamInstEnv.hs
- compiler/GHC/Core/Opt/Arity.hs
- compiler/GHC/Core/Opt/CprAnal.hs
- compiler/GHC/Core/Opt/DmdAnal.hs
- compiler/GHC/Core/Opt/OccurAnal.hs
- compiler/GHC/Core/Opt/SetLevels.hs
- compiler/GHC/Core/Opt/Specialise.hs
- compiler/GHC/Core/Opt/WorkWrap/Utils.hs
- compiler/GHC/Core/Predicate.hs
- compiler/GHC/Core/TyCo/Rep.hs
- compiler/GHC/Core/TyCon.hs
- compiler/GHC/Core/Type.hs
- compiler/GHC/Core/Unfold.hs
- compiler/GHC/Core/Unfold/Make.hs
- compiler/GHC/Core/Utils.hs
- compiler/GHC/CoreToStg.hs
- compiler/GHC/HsToCore/Binds.hs
- compiler/GHC/HsToCore/Expr.hs
- compiler/GHC/HsToCore/Foreign/Call.hs
- compiler/GHC/Iface/Decl.hs
- compiler/GHC/Iface/Syntax.hs
- compiler/GHC/IfaceToCore.hs
- compiler/GHC/Rename/Module.hs
- compiler/GHC/StgToByteCode.hs
- compiler/GHC/Tc/Errors.hs
- compiler/GHC/Tc/Gen/Bind.hs
- compiler/GHC/Tc/Gen/Expr.hs
- compiler/GHC/Tc/Instance/Class.hs
- compiler/GHC/Tc/Instance/Family.hs
- compiler/GHC/Tc/Solver/Default.hs
- compiler/GHC/Tc/Solver/Dict.hs
- compiler/GHC/Tc/Solver/Equality.hs
- compiler/GHC/Tc/Solver/Monad.hs
- compiler/GHC/Tc/Solver/Solve.hs
- compiler/GHC/Tc/TyCl.hs
- compiler/GHC/Tc/TyCl/Build.hs
- compiler/GHC/Tc/TyCl/Instance.hs
- compiler/GHC/Tc/TyCl/PatSyn.hs
- − compiler/GHC/Tc/Types/EvTerm.hs
- compiler/GHC/Tc/Types/Evidence.hs
- compiler/GHC/Tc/Utils/TcMType.hs
- compiler/GHC/Types/Demand.hs
- compiler/GHC/Types/Id.hs
- compiler/GHC/Types/Id/Make.hs
- compiler/GHC/Types/RepType.hs
- compiler/ghc.cabal.in
- testsuite/tests/core-to-stg/T24124.stderr
- testsuite/tests/deSugar/should_compile/T2431.stderr
- testsuite/tests/dmdanal/should_compile/T16029.stdout
- testsuite/tests/dmdanal/should_compile/T23398.hs
- testsuite/tests/dmdanal/should_compile/T23398.stderr
- testsuite/tests/dmdanal/sigs/T21119.stderr
- testsuite/tests/dmdanal/sigs/T21888.stderr
- testsuite/tests/ghci.debugger/scripts/break011.stdout
- testsuite/tests/ghci.debugger/scripts/break024.stdout
- testsuite/tests/indexed-types/should_compile/T2238.hs
- testsuite/tests/numeric/should_compile/T15547.stderr
- testsuite/tests/numeric/should_compile/T23907.stderr
- testsuite/tests/roles/should_compile/Roles14.stderr
- testsuite/tests/roles/should_compile/Roles3.stderr
- testsuite/tests/roles/should_compile/Roles4.stderr
- testsuite/tests/simplCore/should_compile/DataToTagFamilyScrut.stderr
- testsuite/tests/simplCore/should_compile/T15205.stderr
- testsuite/tests/simplCore/should_compile/T17366.stderr
- testsuite/tests/simplCore/should_compile/T17966.stderr
- testsuite/tests/simplCore/should_compile/T22309.stderr
- testsuite/tests/simplCore/should_compile/T22375DataFamily.stderr
- testsuite/tests/simplCore/should_compile/T23307.stderr
- testsuite/tests/simplCore/should_compile/T23307a.stderr
- testsuite/tests/simplCore/should_compile/T25389.stderr
- testsuite/tests/simplCore/should_compile/T25713.stderr
- testsuite/tests/simplCore/should_compile/T7360.stderr
- testsuite/tests/simplStg/should_compile/T15226b.stderr
- testsuite/tests/tcplugins/CtIdPlugin.hs
- testsuite/tests/typecheck/should_compile/Makefile
- testsuite/tests/typecheck/should_compile/T12763.stderr
- testsuite/tests/typecheck/should_compile/T14774.stdout
- testsuite/tests/typecheck/should_compile/T18406b.stderr
- testsuite/tests/typecheck/should_compile/T18529.stderr
- testsuite/tests/typecheck/should_compile/all.T
- testsuite/tests/unboxedsums/unpack_sums_7.stdout
The diff was not included because it is too large.
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/1f853c2ef2173a0e4726ba56c3411e…
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/1f853c2ef2173a0e4726ba56c3411e…
You're receiving this email because of your account on gitlab.haskell.org.
1
0
[Git][ghc/ghc][wip/26296] ci: Teach ci.sh to fetch FreeBSD artifacts from ghcup unofficial bindists and...
by Zubin (＠wz1000) 15 Aug '25
by Zubin (＠wz1000) 15 Aug '25
15 Aug '25
Zubin pushed to branch wip/26296 at Glasgow Haskell Compiler / GHC
Commits:
d3c25239 by Zubin Duggal at 2025-08-15T15:06:44+05:30
ci: Teach ci.sh to fetch FreeBSD artifacts from ghcup unofficial bindists and bootstrap compiler on FreeBSD to 9.10.1
Fixes #26296
- - - - -
3 changed files:
- .gitlab/ci.sh
- .gitlab/generate-ci/gen_ci.hs
- .gitlab/jobs.yaml
Changes:
=====================================
.gitlab/ci.sh
=====================================
@@ -197,9 +197,6 @@ function set_toolchain_paths() {
CABAL="$toolchain/bin/cabal$exe"
HAPPY="$toolchain/bin/happy$exe"
ALEX="$toolchain/bin/alex$exe"
- if [ "$(uname)" = "FreeBSD" ]; then
- GHC=/usr/local/bin/ghc
- fi
;;
nix)
if [[ ! -f toolchain.sh ]]; then
@@ -275,14 +272,35 @@ function setup() {
}
function fetch_ghc() {
+ local should_fetch=false
+
if [ ! -e "$GHC" ]; then
+ should_fetch=true
+ elif [ -n "${GHC_VERSION:-}" ]; then
+ local current_version
+ if current_version=$($GHC --numeric-version 2>/dev/null); then
+ if [ "$current_version" != "$GHC_VERSION" ]; then
+ info "GHC version mismatch: found $current_version, expected $GHC_VERSION"
+ should_fetch=true
+ fi
+ fi
+ fi
+
+ if [ "$should_fetch" = true ]; then
local v="$GHC_VERSION"
if [[ -z "$v" ]]; then
fail "neither GHC nor GHC_VERSION are not set"
fi
start_section fetch-ghc "Fetch GHC"
- url="https://downloads.haskell.org/~ghc/${GHC_VERSION}/ghc-${GHC_VERSION}-${boot…"
+ case "$(uname)" in
+ FreeBSD)
+ url="https://downloads.haskell.org/ghcup/unofficial-bindists/ghc/${GHC_VERSION}/…"
+ ;;
+ *)
+ url="https://downloads.haskell.org/~ghc/${GHC_VERSION}/ghc-${GHC_VERSION}-${boot…"
+ ;;
+ esac
info "Fetching GHC binary distribution from $url..."
curl "$url" > ghc.tar.xz || fail "failed to fetch GHC binary distribution"
$TAR -xJf ghc.tar.xz || fail "failed to extract GHC binary distribution"
=====================================
.gitlab/generate-ci/gen_ci.hs
=====================================
@@ -446,7 +446,7 @@ opsysVariables _ FreeBSD14 = mconcat
-- Prefer to use the system's clang-based toolchain and not gcc
, "CC" =: "cc"
, "CXX" =: "c++"
- , "GHC_VERSION" =: "9.6.4"
+ , "GHC_VERSION" =: "9.10.1"
, "CABAL_INSTALL_VERSION" =: "3.10.3.0"
]
opsysVariables arch (Linux distro) = distroVariables arch distro
=====================================
.gitlab/jobs.yaml
=====================================
@@ -1467,7 +1467,7 @@
"CC": "cc",
"CONFIGURE_ARGS": "--with-iconv-includes=/usr/local/include --with-iconv-libraries=/usr/local/lib --with-system-libffi --with-ffi-includes=/usr/local/include --with-ffi-libraries=/usr/local/lib --with-gmp-includes=/usr/local/include --with-gmp-libraries=/usr/local/lib --enable-strict-ghc-toolchain-check",
"CXX": "c++",
- "GHC_VERSION": "9.6.4",
+ "GHC_VERSION": "9.10.1",
"INSTALL_CONFIGURE_ARGS": "--enable-strict-ghc-toolchain-check",
"RUNTEST_ARGS": "",
"TEST_ENV": "x86_64-freebsd14-validate",
@@ -4355,7 +4355,7 @@
"CC": "cc",
"CONFIGURE_ARGS": "--with-iconv-includes=/usr/local/include --with-iconv-libraries=/usr/local/lib --with-system-libffi --with-ffi-includes=/usr/local/include --with-ffi-libraries=/usr/local/lib --with-gmp-includes=/usr/local/include --with-gmp-libraries=/usr/local/lib --enable-strict-ghc-toolchain-check",
"CXX": "c++",
- "GHC_VERSION": "9.6.4",
+ "GHC_VERSION": "9.10.1",
"IGNORE_PERF_FAILURES": "all",
"INSTALL_CONFIGURE_ARGS": "--enable-strict-ghc-toolchain-check",
"RUNTEST_ARGS": "",
@@ -5782,7 +5782,7 @@
"CC": "cc",
"CONFIGURE_ARGS": "--with-iconv-includes=/usr/local/include --with-iconv-libraries=/usr/local/lib --with-system-libffi --with-ffi-includes=/usr/local/include --with-ffi-libraries=/usr/local/lib --with-gmp-includes=/usr/local/include --with-gmp-libraries=/usr/local/lib --enable-strict-ghc-toolchain-check",
"CXX": "c++",
- "GHC_VERSION": "9.6.4",
+ "GHC_VERSION": "9.10.1",
"INSTALL_CONFIGURE_ARGS": "--enable-strict-ghc-toolchain-check",
"RUNTEST_ARGS": "",
"TEST_ENV": "x86_64-freebsd14-validate"
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/d3c2523909999ef1b7733ce0c0c5717…
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/d3c2523909999ef1b7733ce0c0c5717…
You're receiving this email because of your account on gitlab.haskell.org.
1
0
[Git][ghc/ghc][master] hadrian: enforce have_llvm=False for wasm32/js
by Marge Bot (＠marge-bot) 15 Aug '25
by Marge Bot (＠marge-bot) 15 Aug '25
15 Aug '25
Marge Bot pushed to branch master at Glasgow Haskell Compiler / GHC
Commits:
1cdc6f46 by Cheng Shao at 2025-08-15T04:12:56-04:00
hadrian: enforce have_llvm=False for wasm32/js
This patch fixes hadrian to always pass have_llvm=False to the
testsuite driver for wasm32/js targets. These targets don't really
support the LLVM backend, and the optllvm test way doesn't work. We
used to special-case wasm32/js to avoid auto-adding optllvm way in
testsuite/config/ghc, but this is still problematic if someone writes
a new LLVM-related test and uses something like when(have_llvm(),
extra_ways(["optllvm"])). So better just enforce have_llvm=False for
these targets here.
- - - - -
2 changed files:
- hadrian/src/Settings/Builders/RunTest.hs
- testsuite/config/ghc
Changes:
=====================================
hadrian/src/Settings/Builders/RunTest.hs
=====================================
@@ -86,6 +86,16 @@ data TestCompilerArgs = TestCompilerArgs{
, pkgConfCacheFile :: FilePath }
deriving (Eq, Show)
+-- | Some archs like wasm32/js used to report have_llvm=True because
+-- they are based on LLVM related toolchains like wasi-sdk/emscripten,
+-- but these targets don't really support the LLVM backend, and the
+-- optllvm test way doesn't work. We used to special-case wasm32/js to
+-- avoid auto-adding optllvm way in testsuite/config/ghc, but this is
+-- still problematic if someone writes a new LLVM-related test and
+-- uses something like when(have_llvm(), extra_ways(["optllvm"])). So
+-- better just enforce have_llvm=False for these targets here.
+allowHaveLLVM :: String -> Bool
+allowHaveLLVM = not . (`elem` ["wasm32", "javascript"])
-- | If the tree is in-compiler then we already know how we will build it so
-- don't build anything in order to work out what we will build.
@@ -129,7 +139,7 @@ inTreeCompilerArgs stg = do
llc_cmd <- queryTargetTarget tgtLlc
llvm_as_cmd <- queryTargetTarget tgtLlvmAs
- let have_llvm = all isJust [llc_cmd, llvm_as_cmd]
+ let have_llvm = allowHaveLLVM arch && all isJust [llc_cmd, llvm_as_cmd]
top <- topDirectory
@@ -176,7 +186,7 @@ outOfTreeCompilerArgs = do
let debugged = "debug" `isInfixOf` rtsWay
llc_cmd <- getTestSetting TestLLC
- have_llvm <- liftIO (isJust <$> findExecutable llc_cmd)
+ have_llvm <- (allowHaveLLVM arch &&) <$> liftIO (isJust <$> findExecutable llc_cmd)
profiled <- getBooleanSetting TestGhcProfiled
pkgConfCacheFile <- getTestSetting TestGhcPackageDb <&> (</> "package.cache")
=====================================
testsuite/config/ghc
=====================================
@@ -70,7 +70,7 @@ if windows:
config.other_ways += winio_ways
# LLVM
-if not config.unregisterised and not config.arch in {"wasm32", "javascript"} and config.have_llvm:
+if not config.unregisterised and config.have_llvm:
config.compile_ways.append('optllvm')
config.run_ways.append('optllvm')
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/1cdc6f46c4168dd92a8f5ea0c398b67…
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/1cdc6f46c4168dd92a8f5ea0c398b67…
You're receiving this email because of your account on gitlab.haskell.org.
1
0
15 Aug '25
Marge Bot pushed to branch master at Glasgow Haskell Compiler / GHC
Commits:
706d33e3 by Recursion Ninja at 2025-08-15T04:12:12-04:00
Resolving issues #20645 and #26109
Correctly sign extending and casting smaller bit width types for LLVM operations:
- bitReverse8#
- bitReverse16#
- bitReverse32#
- byteSwap16#
- byteSwap32#
- pdep8#
- pdep16#
- pext8#
- pext16#
- - - - -
13 changed files:
- compiler/GHC/Builtin/primops.txt.pp
- compiler/GHC/CmmToLlvm/CodeGen.hs
- libraries/ghc-internal/cbits/pdep.c
- libraries/ghc-internal/cbits/pext.c
- + testsuite/tests/llvm/should_run/T20645.hs
- + testsuite/tests/llvm/should_run/T20645.stdout
- testsuite/tests/llvm/should_run/all.T
- testsuite/tests/numeric/should_run/foundation.hs
- utils/genprimopcode/Lexer.x
- utils/genprimopcode/Main.hs
- utils/genprimopcode/Parser.y
- utils/genprimopcode/ParserM.hs
- utils/genprimopcode/Syntax.hs
Changes:
=====================================
compiler/GHC/Builtin/primops.txt.pp
=====================================
@@ -148,6 +148,7 @@ defaults
vector = []
deprecated_msg = {} -- A non-empty message indicates deprecation
div_like = False -- Second argument expected to be non zero - used for tests
+ defined_bits = Nothing -- The number of bits the operation is defined for (if not all bits)
-- Note [When do out-of-line primops go in primops.txt.pp]
-- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -1065,8 +1066,10 @@ primop CtzOp "ctz#" GenPrimOp Word# -> Word#
primop BSwap16Op "byteSwap16#" GenPrimOp Word# -> Word#
{Swap bytes in the lower 16 bits of a word. The higher bytes are undefined. }
+ with defined_bits = 16
primop BSwap32Op "byteSwap32#" GenPrimOp Word# -> Word#
{Swap bytes in the lower 32 bits of a word. The higher bytes are undefined. }
+ with defined_bits = 32
primop BSwap64Op "byteSwap64#" GenPrimOp Word64# -> Word64#
{Swap bytes in a 64 bits of a word.}
primop BSwapOp "byteSwap#" GenPrimOp Word# -> Word#
@@ -1074,10 +1077,13 @@ primop BSwapOp "byteSwap#" GenPrimOp Word# -> Word#
primop BRev8Op "bitReverse8#" GenPrimOp Word# -> Word#
{Reverse the order of the bits in a 8-bit word.}
+ with defined_bits = 8
primop BRev16Op "bitReverse16#" GenPrimOp Word# -> Word#
{Reverse the order of the bits in a 16-bit word.}
+ with defined_bits = 16
primop BRev32Op "bitReverse32#" GenPrimOp Word# -> Word#
{Reverse the order of the bits in a 32-bit word.}
+ with defined_bits = 32
primop BRev64Op "bitReverse64#" GenPrimOp Word64# -> Word64#
{Reverse the order of the bits in a 64-bit word.}
primop BRevOp "bitReverse#" GenPrimOp Word# -> Word#
=====================================
compiler/GHC/CmmToLlvm/CodeGen.hs
=====================================
@@ -230,23 +230,22 @@ genCall t@(PrimTarget (MO_Prefetch_Data localityInt)) [] args
statement $ Expr $ Call StdCall fptr (argVars' ++ argSuffix) []
| otherwise = panic $ "prefetch locality level integer must be between 0 and 3, given: " ++ (show localityInt)
--- Handle PopCnt, Clz, Ctz, and BSwap that need to only convert arg
--- and return types
-genCall t@(PrimTarget (MO_PopCnt w)) dsts args =
- genCallSimpleCast w t dsts args
-
-genCall t@(PrimTarget (MO_Pdep w)) dsts args =
- genCallSimpleCast2 w t dsts args
-genCall t@(PrimTarget (MO_Pext w)) dsts args =
- genCallSimpleCast2 w t dsts args
-genCall t@(PrimTarget (MO_Clz w)) dsts args =
- genCallSimpleCast w t dsts args
-genCall t@(PrimTarget (MO_Ctz w)) dsts args =
- genCallSimpleCast w t dsts args
-genCall t@(PrimTarget (MO_BSwap w)) dsts args =
- genCallSimpleCast w t dsts args
-genCall t@(PrimTarget (MO_BRev w)) dsts args =
- genCallSimpleCast w t dsts args
+-- Handle Clz, Ctz, BRev, BSwap, Pdep, Pext, and PopCnt that need to only
+-- convert arg and return types
+genCall (PrimTarget op@(MO_Clz w)) [dst] args =
+ genCallSimpleCast w op dst args
+genCall (PrimTarget op@(MO_Ctz w)) [dst] args =
+ genCallSimpleCast w op dst args
+genCall (PrimTarget op@(MO_BRev w)) [dst] args =
+ genCallSimpleCast w op dst args
+genCall (PrimTarget op@(MO_BSwap w)) [dst] args =
+ genCallSimpleCast w op dst args
+genCall (PrimTarget op@(MO_Pdep w)) [dst] args =
+ genCallSimpleCast w op dst args
+genCall (PrimTarget op@(MO_Pext w)) [dst] args =
+ genCallSimpleCast w op dst args
+genCall (PrimTarget op@(MO_PopCnt w)) [dst] args =
+ genCallSimpleCast w op dst args
genCall (PrimTarget (MO_AtomicRMW width amop)) [dst] [addr, n] = runStmtsDecls $ do
addrVar <- exprToVarW addr
@@ -640,63 +639,28 @@ genCallExtract _ _ _ _ =
-- since GHC only really has i32 and i64 types and things like Word8 are backed
-- by an i32 and just present a logical i8 range. So we must handle conversions
-- from i32 to i8 explicitly as LLVM is strict about types.
-genCallSimpleCast :: Width -> ForeignTarget -> [CmmFormal] -> [CmmActual]
- -> LlvmM StmtData
-genCallSimpleCast w t@(PrimTarget op) [dst] args = do
- let width = widthToLlvmInt w
- dstTy = cmmToLlvmType $ localRegType dst
-
- fname <- cmmPrimOpFunctions op
- (fptr, _, top3) <- getInstrinct fname width [width]
-
- (dstV, _dst_ty) <- getCmmReg (CmmLocal dst)
-
- let (_, arg_hints) = foreignTargetHints t
- let args_hints = zip args arg_hints
- (argsV, stmts2, top2) <- arg_vars args_hints ([], nilOL, [])
- (argsV', stmts4) <- castVars Signed $ zip argsV [width]
- (retV, s1) <- doExpr width $ Call StdCall fptr argsV' []
- (retVs', stmts5) <- castVars (cmmPrimOpRetValSignage op) [(retV,dstTy)]
- let retV' = singletonPanic "genCallSimpleCast" retVs'
- let s2 = Store retV' dstV Nothing []
-
- let stmts = stmts2 `appOL` stmts4 `snocOL`
- s1 `appOL` stmts5 `snocOL` s2
- return (stmts, top2 ++ top3)
-genCallSimpleCast _ _ dsts _ =
- panic ("genCallSimpleCast: " ++ show (length dsts) ++ " dsts")
-
--- Handle simple function call that only need simple type casting, of the form:
--- truncate arg >>= \a -> call(a) >>= zext
---
--- since GHC only really has i32 and i64 types and things like Word8 are backed
--- by an i32 and just present a logical i8 range. So we must handle conversions
--- from i32 to i8 explicitly as LLVM is strict about types.
-genCallSimpleCast2 :: Width -> ForeignTarget -> [CmmFormal] -> [CmmActual]
- -> LlvmM StmtData
-genCallSimpleCast2 w t@(PrimTarget op) [dst] args = do
- let width = widthToLlvmInt w
- dstTy = cmmToLlvmType $ localRegType dst
-
- fname <- cmmPrimOpFunctions op
- (fptr, _, top3) <- getInstrinct fname width (const width <$> args)
-
- (dstV, _dst_ty) <- getCmmReg (CmmLocal dst)
-
- let (_, arg_hints) = foreignTargetHints t
- let args_hints = zip args arg_hints
- (argsV, stmts2, top2) <- arg_vars args_hints ([], nilOL, [])
- (argsV', stmts4) <- castVars Signed $ zip argsV (const width <$> argsV)
- (retV, s1) <- doExpr width $ Call StdCall fptr argsV' []
- (retVs', stmts5) <- castVars (cmmPrimOpRetValSignage op) [(retV,dstTy)]
- let retV' = singletonPanic "genCallSimpleCast2" retVs'
- let s2 = Store retV' dstV Nothing []
+genCallSimpleCast :: Width -> CallishMachOp -> CmmFormal -> [CmmActual]
+ -> LlvmM StmtData
+genCallSimpleCast specW op dst args = do
+ let width = widthToLlvmInt specW
+ argsW = const width <$> args
+ dstType = cmmToLlvmType $ localRegType dst
+ signage = cmmPrimOpRetValSignage op
+
+ fname <- cmmPrimOpFunctions op
+ (fptr, _, top3) <- getInstrinct fname width argsW
+ (dstV, _dst_ty) <- getCmmReg (CmmLocal dst)
+ let (_, arg_hints) = foreignTargetHints $ PrimTarget op
+ let args_hints = zip args arg_hints
+ (argsV, stmts2, top2) <- arg_vars args_hints ([], nilOL, [])
+ (argsV', stmts4) <- castVars signage $ zip argsV argsW
+ (retV, s1) <- doExpr width $ Call StdCall fptr argsV' []
+ (retV', stmts5) <- castVar signage retV dstType
+ let s2 = Store retV' dstV Nothing []
let stmts = stmts2 `appOL` stmts4 `snocOL`
- s1 `appOL` stmts5 `snocOL` s2
+ s1 `snocOL` stmts5 `snocOL` s2
return (stmts, top2 ++ top3)
-genCallSimpleCast2 _ _ dsts _ =
- panic ("genCallSimpleCast2: " ++ show (length dsts) ++ " dsts")
-- | Create a function pointer from a target.
getFunPtrW :: (LMString -> LlvmType) -> ForeignTarget
@@ -811,11 +775,47 @@ castVar signage v t | getVarType v == t
Signed -> LM_Sext
Unsigned -> LM_Zext
-
cmmPrimOpRetValSignage :: CallishMachOp -> Signage
cmmPrimOpRetValSignage mop = case mop of
- MO_Pdep _ -> Unsigned
- MO_Pext _ -> Unsigned
+ -- Some bit-wise operations /must/ always treat the input and output values
+ -- as 'Unsigned' in order to return the expected result values when pre/post-
+ -- operation bit-width truncation and/or extension occur. For example,
+ -- consider the Bit-Reverse operation:
+ --
+ -- If the result of a Bit-Reverse is treated as signed,
+ -- an positive input can result in an negative output, i.e.:
+ --
+ -- identity(0x03) = 0x03 = 00000011
+ -- breverse(0x03) = 0xC0 = 11000000
+ --
+ -- Now if an extension is performed after the operation to
+ -- promote a smaller bit-width value into a larger bit-width
+ -- type, it is expected that the /bit-wise/ operations will
+ -- not be treated /numerically/ as signed.
+ --
+ -- To illustrate the difference, consider how a signed extension
+ -- for the type i16 to i32 differs for out values above:
+ -- ext_zeroed(i32, breverse(0x03)) = 0x00C0 = 0000000011000000
+ -- ext_signed(i32, breverse(0x03)) = 0xFFC0 = 1111111111000000
+ --
+ -- Here we can see that the former output is the expected result
+ -- of a bit-wise operation which needs to be promoted to a larger
+ -- bit-width type. The latter output is not desirable when we must
+ -- constraining a value into a range of i16 within an i32 type.
+ --
+ -- Hence we always treat the "signage" as unsigned for Bit-Reverse!
+ --
+ -- The same reasoning applied to Bit-Reverse above applies to the other
+ -- bit-wise operations; do not sign extend a possibly negated number!
+ MO_BRev _ -> Unsigned
+ MO_BSwap _ -> Unsigned
+ MO_Clz _ -> Unsigned
+ MO_Ctz _ -> Unsigned
+ MO_Pdep _ -> Unsigned
+ MO_Pext _ -> Unsigned
+ MO_PopCnt _ -> Unsigned
+
+ -- All other cases, default to preserving the numeric sign when extending.
_ -> Signed
-- | Decide what C function to use to implement a CallishMachOp
=====================================
libraries/ghc-internal/cbits/pdep.c
=====================================
@@ -24,20 +24,23 @@ hs_pdep64(StgWord64 src, StgWord64 mask)
return result;
}
+// When dealing with values of bit-width shorter than uint64_t, ensure to
+// cast the return value to correctly truncate the undefined upper bits.
+// This is *VERY* important when GHC is using the LLVM backend!
StgWord
hs_pdep32(StgWord src, StgWord mask)
{
- return hs_pdep64(src, mask);
+ return (StgWord) ((StgWord32) hs_pdep64(src, mask));
}
StgWord
hs_pdep16(StgWord src, StgWord mask)
{
- return hs_pdep64(src, mask);
+ return (StgWord) ((StgWord16) hs_pdep64(src, mask));
}
StgWord
hs_pdep8(StgWord src, StgWord mask)
{
- return hs_pdep64(src, mask);
+ return (StgWord) ((StgWord8) hs_pdep64(src, mask));
}
=====================================
libraries/ghc-internal/cbits/pext.c
=====================================
@@ -1,13 +1,13 @@
#include "Rts.h"
#include "MachDeps.h"
-StgWord64
-hs_pext64(StgWord64 src, StgWord64 mask)
+static StgWord64
+hs_pext(const unsigned char bit_width, const StgWord64 src, const StgWord64 mask)
{
uint64_t result = 0;
int offset = 0;
- for (int bit = 0; bit != sizeof(uint64_t) * 8; ++bit) {
+ for (int bit = 0; bit != bit_width; ++bit) {
const uint64_t src_bit = (src >> bit) & 1;
const uint64_t mask_bit = (mask >> bit) & 1;
@@ -20,20 +20,29 @@ hs_pext64(StgWord64 src, StgWord64 mask)
return result;
}
+StgWord64
+hs_pext64(const StgWord64 src, const StgWord64 mask)
+{
+ return hs_pext(64, src, mask);
+}
+
+// When dealing with values of bit-width shorter than uint64_t, ensure to
+// cast the return value to correctly truncate the undefined upper bits.
+// This is *VERY* important when GHC is using the LLVM backend!
StgWord
-hs_pext32(StgWord src, StgWord mask)
+hs_pext32(const StgWord src, const StgWord mask)
{
- return hs_pext64(src, mask);
+ return (StgWord) ((StgWord32) hs_pext(32, src, mask));
}
StgWord
-hs_pext16(StgWord src, StgWord mask)
+hs_pext16(const StgWord src, const StgWord mask)
{
- return hs_pext64(src, mask);
+ return (StgWord) ((StgWord16) hs_pext(16, src, mask));
}
StgWord
-hs_pext8(StgWord src, StgWord mask)
+hs_pext8(const StgWord src, const StgWord mask)
{
- return hs_pext64(src, mask);
+ return (StgWord) ((StgWord8) hs_pext(8, src, mask));
}
=====================================
testsuite/tests/llvm/should_run/T20645.hs
=====================================
@@ -0,0 +1,18 @@
+-- Minimal reproducer for https://gitlab.haskell.org/ghc/ghc/-/issues/20645
+{-# LANGUAGE MagicHash #-}
+{-# LANGUAGE ExtendedLiterals #-}
+import GHC.Exts
+import GHC.Word
+import Numeric (showHex)
+
+opaqueInt8# :: Int8# -> Int8#
+opaqueInt8# x = x
+{-# OPAQUE opaqueInt8# #-}
+
+main :: IO ()
+main = let !x = opaqueInt8# 109#Int8
+ !y = opaqueInt8# 1#Int8
+ in putStrLn $ flip showHex "" (W# ( pext8#
+ (word8ToWord# (int8ToWord8# (0#Int8 `subInt8#` x )))
+ (word8ToWord# (int8ToWord8# (y `subInt8#` 4#Int8)))
+ ))
=====================================
testsuite/tests/llvm/should_run/T20645.stdout
=====================================
@@ -0,0 +1 @@
+49
=====================================
testsuite/tests/llvm/should_run/all.T
=====================================
@@ -17,3 +17,4 @@ test('T22487', [normal, normalise_errmsg_fun(ignore_llvm_and_vortex)], compile_a
test('T22033', [normal, normalise_errmsg_fun(ignore_llvm_and_vortex)], compile_and_run, [''])
test('T25730', [req_c, unless(arch('x86_64'), skip), normalise_errmsg_fun(ignore_llvm_and_vortex)], compile_and_run, ['T25730C.c'])
# T25730C.c contains Intel instrinsics, so only run this test on x86
+test('T20645', [normal, normalise_errmsg_fun(ignore_llvm_and_vortex), when(have_llvm(), extra_ways(["optllvm"]))], compile_and_run, [''])
=====================================
testsuite/tests/numeric/should_run/foundation.hs
=====================================
@@ -24,6 +24,7 @@ module Main
( main
) where
+import Data.Bits (Bits((.&.), bit))
import Data.Word
import Data.Int
import GHC.Natural
@@ -408,6 +409,33 @@ instance TestPrimop (Word# -> Int# -> Word#) where
testPrimop s l r = Property s $ \(uWord -> a1) (uInt -> a2) -> (wWord (l a1 a2)) === wWord (r a1 a2)
-}
+-- | A special data-type for representing functions where,
+-- since only some number of the lower bits are defined,
+-- testing for strict equality in the undefined upper bits is not appropriate!
+-- Without using this data-type, false-positive failures will be reported
+-- when the undefined bit regions do not match, even though the equality of bits
+-- in this undefined region has no bearing on correctness.
+data LowerBitsAreDefined =
+ LowerBitsAreDefined
+ { definedLowerWidth :: Word
+ -- ^ The (strictly-non-negative) number of least-significant bits
+ -- for which the attached function is defined.
+ , undefinedBehavior :: (Word# -> Word#)
+ -- ^ Function with undefined behavior for some of its most significant bits.
+ }
+
+instance TestPrimop LowerBitsAreDefined where
+ testPrimop s l r = Property s $ \ (uWord#-> x0) ->
+ let -- Create a mask to unset all bits in the undefined area,
+ -- leaving set bits only in the area of defined behavior.
+ -- Since the upper bits are undefined,
+ -- if the function defines behavior for the lower N bits,
+ -- then /only/ the lower N bits are preserved,
+ -- and the upper WORDSIZE - N bits are discarded.
+ mask = bit (fromEnum (definedLowerWidth r)) - 1
+ valL = wWord# (undefinedBehavior l x0) .&. mask
+ valR = wWord# (undefinedBehavior r x0) .&. mask
+ in valL === valR
twoNonZero :: (a -> a -> b) -> a -> NonZero a -> b
twoNonZero f x (NonZero y) = f x y
@@ -655,13 +683,13 @@ testPrimops = Group "primop"
, testPrimop "ctz32#" Primop.ctz32# Wrapper.ctz32#
, testPrimop "ctz64#" Primop.ctz64# Wrapper.ctz64#
, testPrimop "ctz#" Primop.ctz# Wrapper.ctz#
- , testPrimop "byteSwap16#" Primop.byteSwap16# Wrapper.byteSwap16#
- , testPrimop "byteSwap32#" Primop.byteSwap32# Wrapper.byteSwap32#
+ , testPrimop "byteSwap16#" (16 `LowerBitsAreDefined` Primop.byteSwap16#) (16 `LowerBitsAreDefined` Wrapper.byteSwap16#)
+ , testPrimop "byteSwap32#" (32 `LowerBitsAreDefined` Primop.byteSwap32#) (32 `LowerBitsAreDefined` Wrapper.byteSwap32#)
, testPrimop "byteSwap64#" Primop.byteSwap64# Wrapper.byteSwap64#
, testPrimop "byteSwap#" Primop.byteSwap# Wrapper.byteSwap#
- , testPrimop "bitReverse8#" Primop.bitReverse8# Wrapper.bitReverse8#
- , testPrimop "bitReverse16#" Primop.bitReverse16# Wrapper.bitReverse16#
- , testPrimop "bitReverse32#" Primop.bitReverse32# Wrapper.bitReverse32#
+ , testPrimop "bitReverse8#" (8 `LowerBitsAreDefined` Primop.bitReverse8#) (8 `LowerBitsAreDefined` Wrapper.bitReverse8#)
+ , testPrimop "bitReverse16#" (16 `LowerBitsAreDefined` Primop.bitReverse16#) (16 `LowerBitsAreDefined` Wrapper.bitReverse16#)
+ , testPrimop "bitReverse32#" (32 `LowerBitsAreDefined` Primop.bitReverse32#) (32 `LowerBitsAreDefined` Wrapper.bitReverse32#)
, testPrimop "bitReverse64#" Primop.bitReverse64# Wrapper.bitReverse64#
, testPrimop "bitReverse#" Primop.bitReverse# Wrapper.bitReverse#
, testPrimop "narrow8Int#" Primop.narrow8Int# Wrapper.narrow8Int#
=====================================
utils/genprimopcode/Lexer.x
=====================================
@@ -56,6 +56,7 @@ words :-
<0> "CanFail" { mkT TCanFail }
<0> "ThrowsException" { mkT TThrowsException }
<0> "ReadWriteEffect" { mkT TReadWriteEffect }
+ <0> "defined_bits" { mkT TDefinedBits }
<0> "can_fail_warning" { mkT TCanFailWarnFlag }
<0> "DoNotWarnCanFail" { mkT TDoNotWarnCanFail }
<0> "WarnIfEffectIsCanFail" { mkT TWarnIfEffectIsCanFail }
=====================================
utils/genprimopcode/Main.hs
=====================================
@@ -10,6 +10,7 @@ module Main where
import Parser
import Syntax
+import Control.Applicative (asum)
import Data.Char
import Data.List (union, intersperse, intercalate, nub, sort)
import Data.Maybe ( catMaybes, mapMaybe )
@@ -116,9 +117,15 @@ desugarVectorSpec i = case vecOptions i of
main :: IO ()
main = getArgs >>= \args ->
if length args /= 1 || head args `notElem` known_args
- then error ("usage: genprimopcode command < primops.txt > ...\n"
+ then error ("Usage: genprimopcode command < primops.txt > ...\n"
++ " where command is one of\n"
++ unlines (map (" "++) known_args)
+ ++ unlines
+ [ ""
+ , "Nota Bene: Be sure to manually run primops.txt through the C Pre-Processor"
+ , " before sending the input stream to STDIN, i.e:"
+ , ""
+ , " cpp -P -w primops.txt | genprimopcode command" ]
)
else
do hSetEncoding stdin utf8 -- The input file is in UTF-8. Set the encoding explicitly.
@@ -312,6 +319,7 @@ gen_hs_source (Info defaults entries) =
opt (OptionVector _) = ""
opt (OptionFixity mf) = "fixity = " ++ show mf
opt (OptionEffect eff) = "effect = " ++ show eff
+ opt (OptionDefinedBits bc) = "defined_bits = " ++ show bc
opt (OptionCanFailWarnFlag wf) = "can_fail_warning = " ++ show wf
hdr s@(Section {}) = sec s
@@ -638,6 +646,7 @@ gen_switch_from_attribs attrib_name fn_name (Info defaults entries)
getAltRhs (OptionVector _) = "True"
getAltRhs (OptionFixity mf) = show mf
getAltRhs (OptionEffect eff) = show eff
+ getAltRhs (OptionDefinedBits bc) = show bc
getAltRhs (OptionCanFailWarnFlag wf) = show wf
mkAlt po
@@ -753,7 +762,12 @@ gen_foundation_tests (Info _ entries)
= let testPrimOpHow = if is_divLikeOp po
then "testPrimopDivLike"
else "testPrimop"
- in Just $ intercalate " " [testPrimOpHow, "\"" ++ poName ++ "\"", wrap "Primop" poName, wrap "Wrapper" poName]
+ qualOp qualification =
+ let qName = wrap qualification poName
+ in case mb_defined_bits po of
+ Nothing -> qName
+ Just bs -> concat ["(", show bs, " `LowerBitsAreDefined` ", qName, ")"]
+ in Just $ intercalate " " [testPrimOpHow, "\"" ++ poName ++ "\"", qualOp "Primop", qualOp "Wrapper"]
| otherwise = Nothing
@@ -771,6 +785,16 @@ gen_foundation_tests (Info _ entries)
divableTyCons = ["Int#", "Word#", "Word8#", "Word16#", "Word32#", "Word64#"
,"Int8#", "Int16#", "Int32#", "Int64#"]
+ mb_defined_bits :: Entry -> Maybe Word
+ mb_defined_bits op@(PrimOpSpec{}) =
+ let opOpts = opts op
+ getDefBits :: Option -> Maybe Word
+ getDefBits (OptionDefinedBits x) = x
+ getDefBits _ = Nothing
+ in asum $ getDefBits <$> opOpts
+ mb_defined_bits _ = Nothing
+
+
------------------------------------------------------------------
-- Create PrimOpInfo text from PrimOpSpecs -----------------------
------------------------------------------------------------------
=====================================
utils/genprimopcode/Parser.y
=====================================
@@ -50,6 +50,7 @@ import AccessOps
CanFail { TCanFail }
ThrowsException { TThrowsException }
ReadWriteEffect { TReadWriteEffect }
+ defined_bits { TDefinedBits }
can_fail_warning { TCanFailWarnFlag }
DoNotWarnCanFail { TDoNotWarnCanFail }
WarnIfEffectIsCanFail { TWarnIfEffectIsCanFail }
@@ -81,13 +82,14 @@ pOptions : pOption pOptions { $1 : $2 }
| {- empty -} { [] }
pOption :: { Option }
-pOption : lowerName '=' false { OptionFalse $1 }
- | lowerName '=' true { OptionTrue $1 }
- | lowerName '=' pStuffBetweenBraces { OptionString $1 $3 }
- | lowerName '=' integer { OptionInteger $1 $3 }
- | vector '=' pVectorTemplate { OptionVector $3 }
- | fixity '=' pInfix { OptionFixity $3 }
- | effect '=' pEffect { OptionEffect $3 }
+pOption : lowerName '=' false { OptionFalse $1 }
+ | lowerName '=' true { OptionTrue $1 }
+ | lowerName '=' pStuffBetweenBraces { OptionString $1 $3 }
+ | lowerName '=' integer { OptionInteger $1 $3 }
+ | vector '=' pVectorTemplate { OptionVector $3 }
+ | fixity '=' pInfix { OptionFixity $3 }
+ | effect '=' pEffect { OptionEffect $3 }
+ | defined_bits '=' pGoodBits { OptionDefinedBits $3 }
| can_fail_warning '=' pPrimOpCanFailWarnFlag { OptionCanFailWarnFlag $3 }
pInfix :: { Maybe Fixity }
@@ -102,6 +104,10 @@ pEffect : NoEffect { NoEffect }
| ThrowsException { ThrowsException }
| ReadWriteEffect { ReadWriteEffect }
+pGoodBits :: { Maybe Word }
+pGoodBits : integer { Just $ toEnum $1 }
+ | nothing { Nothing }
+
pPrimOpCanFailWarnFlag :: { PrimOpCanFailWarnFlag }
pPrimOpCanFailWarnFlag : DoNotWarnCanFail { DoNotWarnCanFail }
| WarnIfEffectIsCanFail { WarnIfEffectIsCanFail }
=====================================
utils/genprimopcode/ParserM.hs
=====================================
@@ -116,6 +116,7 @@ data Token = TEOF
| TCanFail
| TThrowsException
| TReadWriteEffect
+ | TDefinedBits
| TCanFailWarnFlag
| TDoNotWarnCanFail
| TWarnIfEffectIsCanFail
=====================================
utils/genprimopcode/Syntax.hs
=====================================
@@ -76,6 +76,7 @@ data Option
| OptionFixity (Maybe Fixity) -- fixity = infix{,l,r} <int> | Nothing
| OptionEffect PrimOpEffect -- effect = NoEffect | DoNotSpeculate | CanFail | ThrowsException | ReadWriteEffect | FallibleReadWriteEffect
| OptionCanFailWarnFlag PrimOpCanFailWarnFlag -- can_fail_warning = DoNotWarnCanFail | WarnIfEffectIsCanFail | YesWarnCanFail
+ | OptionDefinedBits (Maybe Word) -- defined_bits = Just 16 | Nothing
deriving Show
-- categorises primops
@@ -196,6 +197,7 @@ get_attrib_name (OptionVector _) = "vector"
get_attrib_name (OptionFixity _) = "fixity"
get_attrib_name (OptionEffect _) = "effect"
get_attrib_name (OptionCanFailWarnFlag _) = "can_fail_warning"
+get_attrib_name (OptionDefinedBits _) = "defined_bits"
lookup_attrib :: String -> [Option] -> Maybe Option
lookup_attrib _ [] = Nothing
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/706d33e38fbdcea3f84c34c727c55cf…
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/706d33e38fbdcea3f84c34c727c55cf…
You're receiving this email because of your account on gitlab.haskell.org.
1
0
[Git][ghc/ghc][wip/T23162-spj] 26 commits: Extend record-selector usage ticking to all binds using a record field
by Simon Peyton Jones (＠simonpj) 15 Aug '25
by Simon Peyton Jones (＠simonpj) 15 Aug '25
15 Aug '25
Simon Peyton Jones pushed to branch wip/T23162-spj at Glasgow Haskell Compiler / GHC
Commits:
62899117 by Florian Ragwitz at 2025-08-13T21:01:34-04:00
Extend record-selector usage ticking to all binds using a record field
This extends the previous handling of ticking for RecordWildCards and
NamedFieldPuns to all var bindings that involve record selectors.
Note that certain patterns such as `Foo{foo = 42}` will currently not tick the
`foo` selector, as ticking is triggered by `HsVar`s.
Closes #26191.
- - - - -
b37b3af7 by Florian Ragwitz at 2025-08-13T21:01:34-04:00
Add release notes for 9.16.1 and move description of latest HPC changes there.
- - - - -
a5e4b7d9 by Ben Gamari at 2025-08-13T21:02:18-04:00
rts: Clarify rationale for undefined atomic wrappers
Since c06e3f46d24ef69f3a3d794f5f604cb8c2a40cbc the RTS has declared
various atomic operation wrappers defined by ghc-internal as undefined.
While the rationale for this isn't clear from the commit message, I
believe that this is necessary due to the unregisterised backend.
Specifically, the code generator will reference these symbols when
compiling RTS Cmm sources.
- - - - -
50842f83 by Andreas Klebinger at 2025-08-13T21:03:01-04:00
Make unexpected LLVM versions a warning rather than an error.
Typically a newer LLVM version *will* work so erroring out if
a user uses a newer LLVM version is too aggressive.
Fixes #25915
- - - - -
c91e2650 by fendor at 2025-08-13T21:03:43-04:00
Store `StackTrace` and `StackSnapshot` in `Backtraces`
Instead of decoding the stack traces when collecting the `Backtraces`,
defer this decoding until actually showing the `Backtraces`.
This allows users to customise how `Backtraces` are displayed by
using a custom implementation of `displayExceptionWithInfo`, overwriting
the default implementation for `Backtraces` (`displayBacktraces`).
- - - - -
dee28cdd by fendor at 2025-08-13T21:03:43-04:00
Allow users to customise the collection of exception annotations
Add a global `CollectExceptionAnnotationMechanism` which determines how
`ExceptionAnnotation`s are collected upon throwing an `Exception`.
This API is exposed via `ghc-experimental`.
By overriding how we collect `Backtraces`, we can control how the
`Backtraces` are displayed to the user by newtyping `Backtraces` and
giving a different instance for `ExceptionAnnotation`.
A concrete use-case for this feature is allowing us to experiment with
alternative stack decoders, without having to modify `base`, which take
additional information from the stack frames.
This commit does not modify how `Backtraces` are currently
collected or displayed.
- - - - -
66024722 by fendor at 2025-08-13T21:03:43-04:00
Expose Backtraces internals from ghc-experimental
Additionally, expose the same API `base:Control.Exception.Backtrace`
to make it easier to use as a drop-in replacement.
- - - - -
a766286f by Reed Mullanix at 2025-08-13T21:04:36-04:00
ghc-internal: Fix naturalAndNot for NB/NS case
When the first argument to `naturalAndNot` is larger than a `Word` and the second is `Word`-sized, `naturalAndNot` will truncate the
result:
```
>>> naturalAndNot ((2 ^ 65) .|. (2 ^ 3)) (2 ^ 3)
0
```
In contrast, `naturalAndNot` does not truncate when both arguments are larger than a `Word`, so this appears to be a bug.
Luckily, the fix is pretty easy: we just need to call `bigNatAndNotWord#` instead of truncating.
Fixes #26230
- - - - -
3506fa7d by Simon Hengel at 2025-08-13T21:05:18-04:00
Report -pgms as a deprecated flag
(instead of reporting an unspecific warning)
Before:
on the commandline: warning:
Object splitting was removed in GHC 8.8
After:
on the commandline: warning: [GHC-53692] [-Wdeprecated-flags]
-pgms is deprecated: Object splitting was removed in GHC 8.8
- - - - -
51c701fe by Zubin Duggal at 2025-08-13T21:06:00-04:00
testsuite: Be more permissive when filtering out GNU_PROPERTY_TYPE linker warnings
The warning text is slightly different with ld.bfd.
Fixes #26249
- - - - -
dfe6f464 by Simon Hengel at 2025-08-13T21:06:43-04:00
Refactoring: Don't misuse `MCDiagnostic` for lint messages
`MCDiagnostic` is meant to be used for compiler diagnostics.
Any code that creates `MCDiagnostic` directly, without going through
`GHC.Driver.Errors.printMessage`, side steps `-fdiagnostics-as-json`
(see e.g. !14475, !14492 !14548).
To avoid this in the future I want to control more narrowly who creates
`MCDiagnostic` (see #24113).
Some parts of the compiler use `MCDiagnostic` purely for formatting
purposes, without creating any real compiler diagnostics. This change
introduces a helper function, `formatDiagnostic`, that can be used in
such cases instead of constructing `MCDiagnostic`.
- - - - -
a8b2fbae by Teo Camarasu at 2025-08-13T21:07:24-04:00
rts: ensure MessageBlackHole.link is always a valid closure
We turn a MessageBlackHole into an StgInd in wakeBlockingQueue().
Therefore it's important that the link field, which becomes the
indirection field, always points to a valid closure.
It's unclear whether it's currently possible for the previous behaviour
to lead to a crash, but it's good to be consistent about this invariant nonetheless.
Co-authored-by: Andreas Klebinger <klebinger.andreas(a)gmx.at>
- - - - -
4021181e by Teo Camarasu at 2025-08-13T21:07:24-04:00
rts: spin if we see a WHITEHOLE in messageBlackHole
When a BLACKHOLE gets cancelled in raiseAsync, we indirect to a THUNK.
GC can then shortcut this, replacing our BLACKHOLE with a fresh THUNK.
This THUNK is not guaranteed to have a valid indirectee field.
If at the same time, a message intended for the previous BLACKHOLE is
processed and concurrently we BLACKHOLE the THUNK, thus temporarily
turning it into a WHITEHOLE, we can get a segfault, since we look at the
undefined indirectee field of the THUNK
The fix is simple: spin if we see a WHITEHOLE, and it will soon be
replaced with a valid BLACKHOLE.
Resolves #26205
- - - - -
1107af89 by Oleg Grenrus at 2025-08-13T21:08:06-04:00
Allow defining HasField instances for naughty fields
Resolves #26295
... as HasField solver doesn't solve for fields with "naughty"
selectors, we could as well allow defining HasField instances for these
fields.
- - - - -
020e7587 by Sylvain Henry at 2025-08-13T21:09:00-04:00
Fix Data.List unqualified import warning
- - - - -
fd811ded by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Make injecting implicit bindings into its own pass
Previously we were injecting "impliicit bindings" (data constructor
worker and wrappers etc)
- both at the end of CoreTidy,
- and at the start of CorePrep
This is unpleasant and confusing. This patch puts it it its own pass,
addImplicitBinds, which runs between the two.
The function `GHC.CoreToStg.AddImplicitBinds.addImplicitBinds` now takes /all/
TyCons, not just the ones for algebraic data types. That change ripples
through to
- corePrepPgm
- doCodeGen
- byteCodeGen
All take [TyCon] which includes all TyCons
- - - - -
9bd7fcc5 by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Implement unary classes
The big change is described exhaustively in
Note [Unary class magic] in GHC.Core.TyCon
Other changes
* We never unbox class dictionaries in worker/wrapper. This has been true for some
time now, but the logic is now centralised in functions in
GHC.Core.Opt.WorkWrap.Utils, namely `canUnboxTyCon`, and `canUnboxArg`
See Note [Do not unbox class dictionaries] in GHC.Core.Opt.WorkWrap.Utils.
* Refactored the `notWorthFloating` logic in GHc.Core.Opt.SetLevels.
I can't remember if I actually changed any behaviour here, but if so it's
only in a corner cases.
* Fixed a bug in `GHC.Core.TyCon.isEnumerationTyCon`, which was wrongly returning
True for (##).
* Remove redundant Role argument to `liftCoSubstWithEx`. It was always
Representational.
* I refactored evidence generation in the constraint solver:
* Made GHC.Tc.Types.Evidence contain better abstactions for evidence
generation.
* I deleted the file `GHC.Tc.Types.EvTerm` and merged its (small) contents
elsewhere. It wasn't paying its way.
* Made evidence for implicit parameters go via a proper abstraction.
* Fix inlineBoringOk; see (IB6) in Note [inlineBoringOk]
This fixes a slowdown in `countdownEffectfulDynLocal`
in the `effectful` library.
Smaller things
* Rename `isDataTyCon` to `isBoxedDataTyCon`.
* GHC.Core.Corecion.liftCoSubstWithEx was only called with Representational role,
so I baked that into the function and removed the argument.
* Get rid of `GHC.Core.TyCon.tyConSingleAlgDataCon_maybe` in favour of calling
`not isNewTyCon` at the call sites; more explicit.
* Refatored `GHC.Core.TyCon.isInjectiveTyCon`; but I don't think I changed its
behaviour
* Moved `decomposeIPPred` to GHC.Core.Predicate
Compile time performance changes:
geo. mean +0.1%
minimum -6.8%
maximum +14.4%
The +14% one is in T21839c, where it seems that a bit more inlining
is taking place. That seems acceptable; and the average change is small
Metric Decrease:
LargeRecord
T12227
T12707
T16577
T21839r
T5642
Metric Increase:
T15164
T21839c
T3294
T5321FD
T5321Fun
WWRec
- - - - -
b4075d71 by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Slight improvement to pre/postInlineUnconditionally
Avoids an extra simplifier iteration
- - - - -
9e443596 by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Fix a long-standing assertion error in normSplitTyConApp_maybe
- - - - -
91310ad0 by Simon Peyton Jones at 2025-08-14T17:56:47-04:00
Add comment to coercion optimiser
- - - - -
5b841d82 by Teo Camarasu at 2025-08-14T17:57:56-04:00
template-haskell: move some identifiers from ghc-internal to template-haskell
These identifiers are not used internally by the compiler. Therefore we
have no reason for them to be in ghc-internal.
By moving them to template-haskell, we benefit from it being easier to
change them and we avoid having to build them in stage0.
Resolves #26048
- - - - -
33e2c7e5 by Teo Camarasu at 2025-08-14T17:57:56-04:00
template-haskell: transfer $infix note to public module
This Haddock note should be in the public facing module
- - - - -
2a411fc4 by Sylvain Henry at 2025-08-14T17:59:09-04:00
JS: export HEAP8 symbol (#26290)
Newer Emscripten requires this.
- - - - -
248f78ca by Ben Gamari at 2025-08-14T17:59:51-04:00
users-guide: Drop the THREAD_RUNNABLE event
As of f361281c89fbce42865d8b8b27b0957205366186 it is no longer emitted.
- - - - -
8a547df2 by Richard Eisenberg at 2025-08-14T23:53:01+01:00
Move some fundep solving to new spot
Work in progress...[skip ci]
This completes moving dict fundeps to the main loop
We need wanted/wanted fundeps too
...and some other refactors
Wibbles
Make FunDeps into a new module
Solve new_eqs rather than adding them to WantedConstraints
Wibble
Import wibbles
Comments only [skip ci]
WIP on FunDeps [skip ci]
Work in progress [skip ci]
More WIP
Wibbles to fundeps [skip ci]
Kill off kickOutAfterUnification
More wibbles
Need to remove the unification-count stuff entirely
and do more tidying up -- this commit is mainly for CI
Wibbles solver
Iterate the simples more often than plugins
Start to extend to equalities
Whitespace only
Small improvements
Wibbles
- - - - -
b01a4e28 by Simon Peyton Jones at 2025-08-14T23:53:06+01:00
Improve pretty printer for HsExpr
Given a very deeply-nested application, it just kept printing
deeper and deeper. This small change makes it cut off.
- - - - -
177 changed files:
- compiler/GHC/Builtin/Types.hs
- compiler/GHC/ByteCode/InfoTable.hs
- compiler/GHC/Core/Class.hs
- compiler/GHC/Core/Coercion.hs
- compiler/GHC/Core/Coercion/Opt.hs
- compiler/GHC/Core/DataCon.hs
- compiler/GHC/Core/FamInstEnv.hs
- compiler/GHC/Core/Lint.hs
- compiler/GHC/Core/Opt/Arity.hs
- compiler/GHC/Core/Opt/CprAnal.hs
- compiler/GHC/Core/Opt/DmdAnal.hs
- compiler/GHC/Core/Opt/OccurAnal.hs
- compiler/GHC/Core/Opt/SetLevels.hs
- compiler/GHC/Core/Opt/Simplify/Utils.hs
- compiler/GHC/Core/Opt/Specialise.hs
- compiler/GHC/Core/Opt/WorkWrap/Utils.hs
- compiler/GHC/Core/Predicate.hs
- compiler/GHC/Core/TyCo/Rep.hs
- compiler/GHC/Core/TyCon.hs
- compiler/GHC/Core/Type.hs
- compiler/GHC/Core/Unfold.hs
- compiler/GHC/Core/Unfold/Make.hs
- compiler/GHC/Core/Utils.hs
- compiler/GHC/CoreToStg.hs
- + compiler/GHC/CoreToStg/AddImplicitBinds.hs
- compiler/GHC/CoreToStg/Prep.hs
- compiler/GHC/Driver/Errors/Ppr.hs
- compiler/GHC/Driver/Main.hs
- compiler/GHC/Driver/Session.hs
- compiler/GHC/Hs/Expr.hs
- compiler/GHC/HsToCore/Binds.hs
- compiler/GHC/HsToCore/Expr.hs
- compiler/GHC/HsToCore/Foreign/Call.hs
- compiler/GHC/HsToCore/Ticks.hs
- compiler/GHC/Iface/Decl.hs
- compiler/GHC/Iface/Syntax.hs
- compiler/GHC/Iface/Tidy.hs
- compiler/GHC/IfaceToCore.hs
- compiler/GHC/Rename/Module.hs
- compiler/GHC/Runtime/Debugger/Breakpoints.hs
- compiler/GHC/Runtime/Heap/Inspect.hs
- compiler/GHC/Stg/Lint.hs
- compiler/GHC/StgToByteCode.hs
- compiler/GHC/StgToCmm.hs
- compiler/GHC/Tc/Errors.hs
- compiler/GHC/Tc/Gen/Bind.hs
- compiler/GHC/Tc/Gen/Expr.hs
- compiler/GHC/Tc/Instance/Class.hs
- compiler/GHC/Tc/Instance/Family.hs
- compiler/GHC/Tc/Solver/Default.hs
- compiler/GHC/Tc/Solver/Dict.hs
- compiler/GHC/Tc/Solver/Equality.hs
- + compiler/GHC/Tc/Solver/FunDeps.hs
- compiler/GHC/Tc/Solver/InertSet.hs
- compiler/GHC/Tc/Solver/Monad.hs
- compiler/GHC/Tc/Solver/Solve.hs
- compiler/GHC/Tc/TyCl.hs
- compiler/GHC/Tc/TyCl/Build.hs
- compiler/GHC/Tc/TyCl/Instance.hs
- compiler/GHC/Tc/TyCl/PatSyn.hs
- compiler/GHC/Tc/TyCl/Utils.hs
- − compiler/GHC/Tc/Types/EvTerm.hs
- compiler/GHC/Tc/Types/Evidence.hs
- compiler/GHC/Tc/Utils/TcMType.hs
- compiler/GHC/Tc/Utils/Unify.hs
- compiler/GHC/Tc/Validity.hs
- compiler/GHC/Types/Basic.hs
- compiler/GHC/Types/Demand.hs
- compiler/GHC/Types/Error.hs
- compiler/GHC/Types/Id.hs
- compiler/GHC/Types/Id/Make.hs
- compiler/GHC/Types/RepType.hs
- compiler/GHC/Types/TyThing.hs
- compiler/GHC/Utils/Error.hs
- compiler/GHC/Utils/Outputable.hs
- compiler/ghc.cabal.in
- − docs/users_guide/9.14.1-notes.rst
- + docs/users_guide/9.16.1-notes.rst
- docs/users_guide/eventlog-formats.rst
- docs/users_guide/release-notes.rst
- libraries/base/changelog.md
- libraries/ghc-bignum/changelog.md
- + libraries/ghc-experimental/src/GHC/Exception/Backtrace/Experimental.hs
- libraries/ghc-internal/src/GHC/Internal/Bignum/Natural.hs
- libraries/ghc-internal/src/GHC/Internal/Exception.hs
- libraries/ghc-internal/src/GHC/Internal/Exception/Backtrace.hs
- libraries/ghc-internal/src/GHC/Internal/Exception/Backtrace.hs-boot
- libraries/ghc-internal/src/GHC/Internal/TH/Lib.hs
- libraries/ghc-internal/src/GHC/Internal/TH/Lift.hs
- libraries/ghc-internal/src/GHC/Internal/TH/Syntax.hs
- + libraries/ghc-internal/tests/Makefile
- + libraries/ghc-internal/tests/all.T
- + libraries/ghc-internal/tests/backtraces/Makefile
- + libraries/ghc-internal/tests/backtraces/T14532a.hs
- + libraries/ghc-internal/tests/backtraces/T14532a.stdout
- + libraries/ghc-internal/tests/backtraces/T14532b.hs
- + libraries/ghc-internal/tests/backtraces/T14532b.stdout
- + libraries/ghc-internal/tests/backtraces/all.T
- libraries/template-haskell/Language/Haskell/TH/Lib.hs
- libraries/template-haskell/Language/Haskell/TH/Quote.hs
- libraries/template-haskell/Language/Haskell/TH/Syntax.hs
- libraries/template-haskell/tests/all.T
- rts/Messages.c
- rts/StgMiscClosures.cmm
- rts/Updates.h
- rts/external-symbols.list.in
- rts/js/mem.js
- rts/rts.cabal
- testsuite/driver/testlib.py
- testsuite/tests/arrows/should_compile/T21301.stderr
- testsuite/tests/core-to-stg/T24124.stderr
- testsuite/tests/deSugar/should_compile/T2431.stderr
- testsuite/tests/deSugar/should_fail/DsStrictFail.stderr
- testsuite/tests/deSugar/should_run/T20024.stderr
- testsuite/tests/deSugar/should_run/dsrun005.stderr
- testsuite/tests/deSugar/should_run/dsrun007.stderr
- testsuite/tests/deSugar/should_run/dsrun008.stderr
- testsuite/tests/deriving/should_run/T9576.stderr
- testsuite/tests/dmdanal/should_compile/T16029.stdout
- testsuite/tests/dmdanal/sigs/T21119.stderr
- testsuite/tests/dmdanal/sigs/T21888.stderr
- testsuite/tests/ghci.debugger/scripts/break011.stdout
- testsuite/tests/ghci.debugger/scripts/break024.stdout
- testsuite/tests/ghci/scripts/Defer02.stderr
- testsuite/tests/ghci/scripts/T15325.stderr
- testsuite/tests/hpc/recsel/recsel.hs
- testsuite/tests/hpc/recsel/recsel.stdout
- testsuite/tests/indexed-types/should_compile/T2238.hs
- testsuite/tests/interface-stability/template-haskell-exports.stdout
- testsuite/tests/numeric/should_compile/T15547.stderr
- testsuite/tests/numeric/should_compile/T23907.stderr
- + testsuite/tests/numeric/should_run/T26230.hs
- + testsuite/tests/numeric/should_run/T26230.stdout
- testsuite/tests/numeric/should_run/all.T
- + testsuite/tests/overloadedrecflds/should_run/T26295.hs
- + testsuite/tests/overloadedrecflds/should_run/T26295.stdout
- testsuite/tests/overloadedrecflds/should_run/all.T
- testsuite/tests/patsyn/should_run/ghci.stderr
- testsuite/tests/quasiquotation/T4491/test.T
- testsuite/tests/quotes/LiftErrMsgDefer.stderr
- testsuite/tests/roles/should_compile/Roles14.stderr
- testsuite/tests/roles/should_compile/Roles3.stderr
- testsuite/tests/roles/should_compile/Roles4.stderr
- testsuite/tests/safeHaskell/safeLanguage/SafeLang15.stderr
- testsuite/tests/simplCore/should_compile/DataToTagFamilyScrut.stderr
- testsuite/tests/simplCore/should_compile/T15205.stderr
- testsuite/tests/simplCore/should_compile/T17366.stderr
- testsuite/tests/simplCore/should_compile/T17966.stderr
- testsuite/tests/simplCore/should_compile/T22309.stderr
- testsuite/tests/simplCore/should_compile/T22375DataFamily.stderr
- testsuite/tests/simplCore/should_compile/T23307.stderr
- testsuite/tests/simplCore/should_compile/T23307a.stderr
- testsuite/tests/simplCore/should_compile/T25389.stderr
- testsuite/tests/simplCore/should_compile/T25713.stderr
- testsuite/tests/simplCore/should_compile/T7360.stderr
- testsuite/tests/simplStg/should_compile/T15226b.stderr
- testsuite/tests/tcplugins/CtIdPlugin.hs
- testsuite/tests/th/Makefile
- testsuite/tests/type-data/should_run/T22332a.stderr
- testsuite/tests/typecheck/should_compile/Makefile
- testsuite/tests/typecheck/should_compile/T12763.stderr
- testsuite/tests/typecheck/should_compile/T14774.stdout
- testsuite/tests/typecheck/should_compile/T18406b.stderr
- testsuite/tests/typecheck/should_compile/T18529.stderr
- testsuite/tests/typecheck/should_compile/all.T
- testsuite/tests/typecheck/should_compile/tc126.hs
- testsuite/tests/typecheck/should_fail/tcfail143.stderr
- testsuite/tests/typecheck/should_run/T10284.stderr
- testsuite/tests/typecheck/should_run/T13838.stderr
- testsuite/tests/typecheck/should_run/T9497a-run.stderr
- testsuite/tests/typecheck/should_run/T9497b-run.stderr
- testsuite/tests/typecheck/should_run/T9497c-run.stderr
- testsuite/tests/unboxedsums/unpack_sums_7.stdout
- testsuite/tests/unsatisfiable/T23816.stderr
- testsuite/tests/unsatisfiable/UnsatDefer.stderr
- testsuite/tests/wasm/should_run/control-flow/LoadCmmGroup.hs
- testsuite/tests/wasm/should_run/control-flow/RunWasm.hs
The diff was not included because it is too large.
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/58d6142b4d86d4b6e157acdbf3cc68…
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/58d6142b4d86d4b6e157acdbf3cc68…
You're receiving this email because of your account on gitlab.haskell.org.
1
0