[Git][ghc/ghc][wip/amg/T22156] 34 commits: genSym: Reimplement via CAS on 32-bit platforms

Adam Gundry (@adamgundry) gitlab at gitlab.haskell.org
Thu Jan 11 18:03:16 UTC 2024 


Adam Gundry pushed to branch wip/amg/T22156 at Glasgow Haskell Compiler / GHC


Commits:
2db11c08 by Ben Gamari at 2023-12-29T15:35:48-05:00
genSym: Reimplement via CAS on 32-bit platforms

Previously the remaining use of the C implementation on 32-bit platforms
resulted in a subtle bug, #24261. This was due to the C object (which
used the RTS's `atomic_inc64` macro) being compiled without `-threaded`
yet later being used in a threaded compiler.

Side-step this issue by using the pure Haskell `genSym` implementation on
all platforms. This required implementing `fetchAddWord64Addr#` in terms
of CAS on 64-bit platforms.

- - - - -
19328a8c by Xiaoyan Ren at 2023-12-29T15:36:30-05:00
Do not color the diagnostic code in error messages (#24172)

- - - - -
685b467c by Krzysztof Gogolewski at 2023-12-29T15:37:06-05:00
Enforce that bindings of implicit parameters are lifted

Fixes #24298

- - - - -
bc4d67b7 by Matthew Craven at 2023-12-31T06:15:42-05:00
StgToCmm: Detect some no-op case-continuations

...and generate no code for them. Fixes #24264.

- - - - -
5b603139 by Krzysztof Gogolewski at 2023-12-31T06:16:18-05:00
Revert "testsuite: mark jspace as fragile on i386."

This reverts commit 0038d052c8c80b4b430bb2aa1c66d5280be1aa95.

The atomicity bug should be fixed by !11802.

- - - - -
d55216ad by Krzysztof Gogolewski at 2024-01-01T12:05:49-05:00
Refactor: store [[PrimRep]] rather than [Type] in STG

StgConApp stored a list of types. This list was used exclusively
during unarisation of unboxed sums (mkUbxSum).
However, this is at a wrong level of abstraction:
STG shouldn't be concerned with Haskell types, only PrimReps.
Update the code to store a [[PrimRep]]. Also, there's no point in storing
this list when we're not dealing with an unboxed sum.

- - - - -
8b340bc7 by Ömer Sinan Ağacan at 2024-01-01T12:06:29-05:00
Kind signatures docs: mention that they're allowed in newtypes

- - - - -
989bf8e5 by Zubin Duggal at 2024-01-03T20:08:47-05:00
ci: Ensure we use the correct bindist name for the test artifact when generating
release ghcup metadata

Fixes #24268

- - - - -
89299a89 by Krzysztof Gogolewski at 2024-01-03T20:09:23-05:00
Refactor: remove calls to typePrimRepArgs

The function typePrimRepArgs is just a thin wrapper around
typePrimRep, adding a VoidRep if the list is empty.
However, in StgToByteCode, we were discarding that VoidRep anyway,
so there's no point in calling it.

- - - - -
c7be0c68 by mmzk1526 at 2024-01-03T20:10:07-05:00
Use "-V" for alex version check for better backward compatibility
Fixes #24302.
In recent versions of alex, "-v" is used for "--verbose" instead of "-version".

- - - - -
67dbcc0a by Krzysztof Gogolewski at 2024-01-05T02:07:18-05:00
Fix VoidRep handling in ghci debugger

'go' inside extractSubTerms was giving a bad result given a VoidRep,
attempting to round towards the next multiple of 0.
I don't understand much about the debugger but the code should be better
than it was.

Fixes #24306

- - - - -
90ea574e by Krzysztof Gogolewski at 2024-01-05T02:07:54-05:00
VoidRep-related refactor

* In GHC.StgToByteCode, replace bcIdPrimId with idPrimRep,
  bcIdArgRep with idArgRep, atomPrimRep with stgArgRep1.
  All of them were duplicates.
* In GHC.Stg.Unarise, we were converting a PrimRep to a Type and back to
  PrimRep. Remove the calls to primRepToType and typePrimRep1 which cancel out.
* In GHC.STG.Lint, GHC.StgToCmm, GHC.Types.RepType we were filtering out
  VoidRep from the result of typePrimRep. But typePrimRep never returns
  VoidRep - remove the filtering.

- - - - -
eaf72479 by brian at 2024-01-06T23:03:09-05:00
Add unaligned Addr# primops

Implements CLC proposal #154:
  https://github.com/haskell/core-libraries-committee/issues/154

* add unaligned addr primops

* add tests

* accept tests

* add documentation

* fix js primops

* uncomment in access ops

* use Word64 in tests

* apply suggestions

* remove extra file

* move docs

* remove random options

* use setByteArray# primop

* better naming

* update base-exports test

* add base-exports for other architectures

- - - - -
d471d445 by Krzysztof Gogolewski at 2024-01-06T23:03:47-05:00
Remove VoidRep from PrimRep, introduce PrimOrVoidRep

This introduces

data PrimOrVoidRep = VoidRep | NVRep PrimRep

changes typePrimRep1 to return PrimOrVoidRep, and adds a new function
typePrimRepU to be used when the argument is definitely non-void.
Details in Note [VoidRep] in GHC.Types.RepType.

Fixes #19520

- - - - -
48720a07 by Matthew Craven at 2024-01-08T18:57:36-05:00
Apply Note [Sensitivity to unique increment] to LargeRecord

- - - - -
9e2e180f by Sebastian Graf at 2024-01-08T18:58:13-05:00
Debugging: Add diffUFM for convenient diffing between UniqFMs

- - - - -
948f3e35 by Sebastian Graf at 2024-01-08T18:58:13-05:00
Rename Opt_D_dump_stranal to Opt_D_dump_dmdanal

... and Opt_D_dump_str_signatures to Opt_D_dump_dmd_signatures

- - - - -
4e217e3e by Sebastian Graf at 2024-01-08T18:58:13-05:00
Deprecate -ddump-stranal and -ddump-str-signatures

... and suggest -ddump-dmdanal and -ddump-dmd-signatures instead

- - - - -
6c613c90 by Sebastian Graf at 2024-01-08T18:58:13-05:00
Move testsuite/tests/stranal to testsuite/tests/dmdanal

A separate commit so that the rename is obvious to Git(Lab)

- - - - -
c929f02b by Sebastian Graf at 2024-01-08T18:58:13-05:00
CoreSubst: Stricten `substBndr` and `cloneBndr`

Doing so reduced allocations of `cloneBndr` by about 25%.

```
T9233(normal) ghc/alloc    672,488,656    663,083,216  -1.4% GOOD
T9675(optasm) ghc/alloc    423,029,256    415,812,200  -1.7%

    geo. mean                                          -0.1%
    minimum                                            -1.7%
    maximum                                            +0.1%
```

Metric Decrease:
    T9233

- - - - -
e3ca78f3 by Krzysztof Gogolewski at 2024-01-10T17:35:59-05:00
Deprecate -Wsemigroup

This warning was used to prepare for Semigroup becoming a superclass
of Monoid, and for (<>) being exported from Prelude. This happened in
GHC 8.4 in 8ae263ceb3566 and feac0a3bc69fd3.

The leftover logic for (<>) has been removed in GHC 9.8, 4d29ecdfcc79.
Now the warning does nothing at all and can be deprecated.

- - - - -
08d14925 by amesgen at 2024-01-10T17:36:42-05:00
WASM metadata: use correct GHC version

- - - - -
7a808419 by Xiaoyan Ren at 2024-01-10T17:37:24-05:00
Allow SCC declarations in TH (#24081)

- - - - -
28827c51 by Xiaoyan Ren at 2024-01-10T17:37:24-05:00
Fix prettyprinting of SCC pragmas

- - - - -
ae9cc1a8 by Matthew Craven at 2024-01-10T17:38:01-05:00
Fix loopification in the presence of void arguments

This also removes Note [Void arguments in self-recursive tail calls],
which was just misleading.  It's important to count void args both
in the function's arity and at the call site.

Fixes #24295.

- - - - -
b718b145 by Zubin Duggal at 2024-01-10T17:38:36-05:00
testsuite: Teach testsuite driver about c++ sources

- - - - -
09cb57ad by Zubin Duggal at 2024-01-10T17:38:36-05:00
driver: Set -DPROFILING when compiling C++ sources with profiling

Earlier, we used to pass all preprocessor flags to the c++ compiler.
This meant that -DPROFILING was passed to the c++ compiler because
it was a part of C++ flags
However, this was incorrect and the behaviour was changed in
8ff3134ed4aa323b0199ad683f72165e51a59ab6. See #21291.

But that commit exposed this bug where -DPROFILING was no longer being passed
when compiling c++ sources.

The fix is to explicitly include -DPROFILING in `opt_cxx` when profiling is
enabled to ensure we pass the correct options for the way to both C and C++
compilers

Fixes #24286

- - - - -
2cf9dd96 by Zubin Duggal at 2024-01-10T17:38:36-05:00
testsuite: rename objcpp -> objcxx

To avoid confusion with C Pre Processsor

- - - - -
af6932d6 by Simon Peyton Jones at 2024-01-10T17:39:12-05:00
Make TYPE and CONSTRAINT not-apart

Issue #24279 showed up a bug in the logic in GHC.Core.Unify.unify_ty
which is supposed to make TYPE and CONSTRAINT be not-apart.

Easily fixed.

- - - - -
4a39b5ff by Zubin Duggal at 2024-01-10T17:39:48-05:00
ci: Fix typo in mk_ghcup_metadata.py

There was a missing colon in the fix to #24268 in 989bf8e53c08eb22de716901b914b3607bc8dd08

- - - - -
13503451 by Zubin Duggal at 2024-01-10T17:40:24-05:00
release-ci: remove release-x86_64-linux-deb11-release+boot_nonmoving_gc job

There is no reason to have this release build or distribute this variation.
This configuration is for testing purposes only.

- - - - -
afca46a4 by Sebastian Graf at 2024-01-10T17:41:00-05:00
Parser: Add a Note detailing why we need happy's `error` to implement layout

- - - - -
0bb36d78 by Adam Gundry at 2024-01-11T18:02:49+00:00
Representation-polymorphic HasField (fixes #22156)

This generalises the HasField class to support representation polymorphism,
so that instead of

    type HasField :: forall {k} . k -> Type -> Type -> Constraint

we have

    type HasField :: forall {k} {r_rep} {a_rep} . k -> TYPE r_rep -> TYPE a_rep -> Constraint

- - - - -
ef383e21 by Adam Gundry at 2024-01-11T18:02:49+00:00
Mark GHC.Records as Trustworthy

The previous patch added an import of GHC.Exts, which means GHC.Records
ceases to be Safe-Inferred. Rather than breaking any user code which
relied on GHC.Records being considered Safe, we declare it Trustworthy.

- - - - -


30 changed files:

- .gitlab/generate-ci/gen_ci.hs
- .gitlab/jobs.yaml
- .gitlab/rel_eng/mk-ghcup-metadata/mk_ghcup_metadata.py
- compiler/GHC/Builtin/Names/TH.hs
- compiler/GHC/Builtin/PrimOps.hs
- compiler/GHC/ByteCode/InfoTable.hs
- compiler/GHC/Cmm/Utils.hs
- compiler/GHC/CmmToAsm/Wasm/Asm.hs
- compiler/GHC/Core/Opt/Pipeline.hs
- compiler/GHC/Core/Subst.hs
- compiler/GHC/Core/TyCon.hs
- compiler/GHC/Core/Unify.hs
- compiler/GHC/CoreToStg.hs
- compiler/GHC/Driver/Config/Core/Lint.hs
- compiler/GHC/Driver/Flags.hs
- compiler/GHC/Driver/Session.hs
- compiler/GHC/HsToCore/Foreign/Utils.hs
- compiler/GHC/HsToCore/Quote.hs
- compiler/GHC/Parser.y
- compiler/GHC/Platform/Ways.hs
- compiler/GHC/Runtime/Heap/Inspect.hs
- compiler/GHC/Stg/Lift/Analysis.hs
- compiler/GHC/Stg/Lint.hs
- compiler/GHC/Stg/Syntax.hs
- compiler/GHC/Stg/Unarise.hs
- compiler/GHC/StgToByteCode.hs
- compiler/GHC/StgToCmm.hs
- compiler/GHC/StgToCmm/ArgRep.hs
- compiler/GHC/StgToCmm/Bind.hs
- compiler/GHC/StgToCmm/Closure.hs


The diff was not included because it is too large.


View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/c04fcaba8f7d9cc32c9a1506384e664dba880c62...ef383e21ae92150d1f1aa0d0c4839ea253cc5422

-- 
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/c04fcaba8f7d9cc32c9a1506384e664dba880c62...ef383e21ae92150d1f1aa0d0c4839ea253cc5422
You're receiving this email because of your account on gitlab.haskell.org.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/ghc-commits/attachments/20240111/4f701341/attachment-0001.html>

More information about the ghc-commits mailing list