[Git][ghc/ghc][wip/sand-witch/modern-STV-extension-shuffling] 28 commits: Add fused multiply-add instructions

[Ben Gamari (@bgamari)]

Ben Gamari pushed to branch wip/sand-witch/modern-STV-extension-shuffling at Glasgow Haskell Compiler / GHC


Commits:
87eebf98 by sheaf at 2023-05-11T11:55:22-04:00
Add fused multiply-add instructions

This patch adds eight new primops that fuse a multiplication and an
addition or subtraction:

  - `{fmadd,fmsub,fnmadd,fnmsub}{Float,Double}#`

fmadd x y z is x * y + z, computed with a single rounding step.

This patch implements code generation for these primops in the following
backends:

  - X86, AArch64 and PowerPC NCG,
  - LLVM
  - C

WASM uses the C implementation. The primops are unsupported in the
JavaScript backend.

The following constant folding rules are also provided:

  - compute a * b + c when a, b, c are all literals,
  - x * y + 0 ==> x * y,
  - ±1 * y + z ==> z ± y and x * ±1 + z ==> z ± x.

NB: the constant folding rules incorrectly handle signed zero.
This is a known limitation with GHC's floating-point constant folding
rules (#21227), which we hope to resolve in the future.

- - - - -
ad16a066 by Krzysztof Gogolewski at 2023-05-11T11:55:59-04:00
Add a test for #21278

- - - - -
05cea68c by Matthew Pickering at 2023-05-11T11:56:36-04:00
rts: Refine memory retention behaviour to account for pinned/compacted objects

When using the copying collector there is still a lot of data which
isn't copied (such as pinned, compacted, large objects etc). The logic
to decide how much memory to retain didn't take into account that these
wouldn't be copied. Therefore we pessimistically retained 2* the amount
of memory for these blocks even though they wouldn't be copied by the
collector.

The solution is to split up the heap into two parts, the parts which
will be copied and the parts which won't be copied. Then the appropiate
factor is applied to each part individually (2 * for copying and 1.2 *
for not copying).

The T23221 test demonstrates this improvement with a program which first
allocates many unpinned ByteArray# followed by many pinned ByteArray#
and observes the difference in the ultimate memory baseline between the
two.

There are some charts on #23221.

Fixes #23221

- - - - -
1bb24432 by Cheng Shao at 2023-05-11T11:57:15-04:00
hadrian: fix no_dynamic_libs flavour transformer

This patch fixes the no_dynamic_libs flavour transformer and make
fully_static reuse it. Previously building with no_dynamic_libs fails
since ghc program is still dynamic and transitively brings in dyn ways
of rts which are produced by no rules.

- - - - -
0ed493a3 by Josh Meredith at 2023-05-11T23:08:27-04:00
JS: refactor jsSaturate to return a saturated JStat (#23328)

- - - - -
a856d98e by Pierre Le Marre at 2023-05-11T23:09:08-04:00
Doc: Fix out-of-sync using-optimisation page

- Make explicit that default flag values correspond to their -O0 value.
- Fix -fignore-interface-pragmas, -fstg-cse, -fdo-eta-reduction,
  -fcross-module-specialise, -fsolve-constant-dicts, -fworker-wrapper.

- - - - -
c176ad18 by sheaf at 2023-05-12T06:10:57-04:00
Don't panic in mkNewTyConRhs

This function could come across invalid newtype constructors, as we
only perform validity checking of newtypes once we are outside the
knot-tied typechecking loop.
This patch changes this function to fake up a stub type in the case of
an invalid newtype, instead of panicking.

This patch also changes "checkNewDataCon" so that it reports as many
errors as possible at once.

Fixes #23308

- - - - -
ab63daac by Krzysztof Gogolewski at 2023-05-12T06:11:38-04:00
Allow Core optimizations when interpreting bytecode

Tracking ticket: #23056

MR: !10399

This adds the flag `-funoptimized-core-for-interpreter`, permitting use
of the `-O` flag to enable optimizations when compiling with the
interpreter backend, like in ghci.

- - - - -
c6cf9433 by Ben Gamari at 2023-05-12T06:12:14-04:00
hadrian: Fix mention of non-existent removeFiles function

Previously Hadrian's bindist Makefile referred to a `removeFiles`
function that was previously defined by the `make` build system. Since
the `make` build system is no longer around, this function is now
undefined. Naturally, make being make, this appears to be silently
ignored instead of producing an error.

Fix this by rewriting it to `rm -f`.

Closes #23373.

- - - - -
eb60ec18 by Bodigrim at 2023-05-12T06:12:54-04:00
Mention new implementation of GHC.IORef.atomicSwapIORef in the changelog

- - - - -
aa84cff4 by Teo Camarasu at 2023-05-12T19:27:23-04:00
rts: Ensure non-moving gc is not running when pausing

- - - - -
5ad776ab by Teo Camarasu at 2023-05-12T19:27:23-04:00
rts: Teach listAllBlocks about nonmoving heap

List all blocks on the non-moving heap.

Resolves #22627

- - - - -
d683b2e5 by Krzysztof Gogolewski at 2023-05-12T19:28:00-04:00
Fix coercion optimisation for SelCo (#23362)

setNominalRole_maybe is supposed to output a nominal coercion.
In the SelCo case, it was not updating the stored role to Nominal,
causing #23362.

- - - - -
59aa4676 by Alexis King at 2023-05-12T19:28:47-04:00
hadrian: Fix linker script flag for MergeObjects builder

This fixes what appears to have been a typo in !9530. The `-t` flag just
enables tracing on all versions of `ld` I’ve looked at, while `-T` is
used to specify a linker script. It seems that this worked anyway for
some reason on some `ld` implementations (perhaps because they
automatically detect linker scripts), but the missing `-T` argument
causes `gold` to complain.

- - - - -
4bf9fa0f by Adam Gundry at 2023-05-12T23:49:49-04:00
Less coercion optimization for non-newtype axioms

See Note [Push transitivity inside newtype axioms only] for an explanation
of the change here.  This change substantially improves the performance of
coercion optimization for programs involving transitive type family reductions.

-------------------------
Metric Decrease:
    CoOpt_Singletons
    LargeRecord
    T12227
    T12545
    T13386
    T15703
    T5030
    T8095
-------------------------

- - - - -
dc0c9574 by Adam Gundry at 2023-05-12T23:49:49-04:00
Move checkAxInstCo to GHC.Core.Lint

A consequence of the previous change is that checkAxInstCo is no longer
called during coercion optimization, so it can be moved back where it belongs.

Also includes some edits to Note [Conflict checking with AxiomInstCo] as
suggested by @simonpj.

- - - - -
8b9b7dbc by Simon Peyton Jones at 2023-05-12T23:50:25-04:00
Use the eager unifier in the constraint solver

This patch continues the refactoring of the constraint solver
described in #23070.

The Big Deal in this patch is to call the regular, eager unifier from the
constraint solver, when we want to create new equalities. This
replaces the existing, unifyWanted which amounted to
yet-another-unifier, so it reduces duplication of a rather subtle
piece of technology. See

  * Note [The eager unifier] in GHC.Tc.Utils.Unify
  * GHC.Tc.Solver.Monad.wrapUnifierTcS

I did lots of other refactoring along the way

* I simplified the treatment of right hand sides that contain CoercionHoles.
  Now, a constraint that contains a hetero-kind CoercionHole is non-canonical,
  and cannot be used for rewriting or unification alike.  This required me
  to add the ch_hertero_kind flag to CoercionHole, with consequent knock-on
  effects. See wrinkle (2) of `Note [Equalities with incompatible kinds]` in
  GHC.Tc.Solver.Equality.

* I refactored the StopOrContinue type to add StartAgain, so that after a
  fundep improvement (for example) we can simply start the pipeline again.

* I got rid of the unpleasant (and inefficient) rewriterSetFromType/Co functions.
  With Richard I concluded that they are never needed.

* I discovered Wrinkle (W1) in Note [Wanteds rewrite Wanteds] in
  GHC.Tc.Types.Constraint, and therefore now prioritise non-rewritten equalities.

Quite a few error messages change, I think always for the better.

Compiler runtime stays about the same, with one outlier: a 17% improvement in T17836

Metric Decrease:
    T17836
    T18223

- - - - -
5cad28e7 by Bartłomiej Cieślar at 2023-05-12T23:51:06-04:00
Cleanup of dynflags override in export renaming
The deprecation warnings are normally emitted whenever the name's GRE is being looked up, which calls the GHC.Rename.Env.addUsedGRE function. We do not want those warnings to be emitted when renaming export lists, so they are artificially turned off by removing all warning categories from DynFlags at the beginning of GHC.Tc.Gen.Export.rnExports. This commit removes that dependency by unifying the function used for GRE lookup in lookup_ie to lookupGreAvailRn and disabling the call to addUsedGRE in said function (the warnings are also disabled in a call to lookupSubBndrOcc_helper in lookupChildrenExport), as per #17957. This commit also changes the setting for whether to warn about deprecated names in addUsedGREs to be an explicit enum instead of a boolean.

- - - - -
d85ed900 by Alexis King at 2023-05-13T08:45:18-04:00
Use a uniform return convention in bytecode for unary results

fixes #22958

- - - - -
8a0d45f7 by Bodigrim at 2023-05-13T08:45:58-04:00
Add more instances for Compose: Enum, Bounded, Num, Real, Integral

See https://github.com/haskell/core-libraries-committee/issues/160 for discussion

- - - - -
902f0730 by Simon Peyton Jones at 2023-05-13T14:58:34-04:00
Make GHC.Types.Id.Make.shouldUnpackTy a bit more clever

As #23307, GHC.Types.Id.Make.shouldUnpackTy was leaving money on the
table, failing to unpack arguments that are perfectly unpackable.

The fix is pretty easy; see Note [Recursive unboxing]

- - - - -
a5451438 by sheaf at 2023-05-13T14:59:13-04:00
Fix bad multiplicity role in tyConAppFunCo_maybe

The function tyConAppFunCo_maybe produces a multiplicity coercion
for the multiplicity argument of the function arrow, except that
it could be at the wrong role if asked to produce a representational
coercion. We fix this by using the 'funRole' function, which computes
the right roles for arguments to the function arrow TyCon.

Fixes #23386

- - - - -
5b9e9300 by sheaf at 2023-05-15T11:26:59-04:00
Turn "ambiguous import" error into a panic

This error should never occur, as a lookup of a type or data constructor
should never be ambiguous. This is because a single module cannot export
multiple Names with the same OccName, as per item (1) of
Note [Exporting duplicate declarations] in GHC.Tc.Gen.Export.

This code path was intended to handle duplicate record fields, but the
rest of the code had since been refactored to handle those in a
different way.

We also remove the AmbiguousImport constructor of IELookupError, as
it is no longer used.

Fixes #23302

- - - - -
e305e60c by M Farkas-Dyck at 2023-05-15T11:27:41-04:00
Unbreak some tests with latest GNU grep, which now warns about stray '\'.

Confusingly, the testsuite mangled the error to say "stray /".

We also migrate some tests from grep to grep -E, as it seems the author actually wanted an "POSIX extended" (a.k.a. sane) regex.

Background: POSIX specifies 2 "regex" syntaxen: "basic" and "extended". Of these, only "extended" syntax is actually a regular expression. Furthermore, "basic" syntax is inconsistent in its use of the '\' character — sometimes it escapes a regex metacharacter, but sometimes it unescapes it, i.e. it makes an otherwise normal character become a metacharacter. This baffles me and it seems also the authors of these tests. Also, the regex(7) man page (at least on Linux) says "basic" syntax is obsolete. Nearly all modern tools and libraries are consistent in this use of the '\' character (of which many use "extended" syntax by default).

- - - - -
5ae81842 by sheaf at 2023-05-15T14:49:17-04:00
Improve "ambiguous occurrence" error messages

This error was sometimes a bit confusing, especially when data families
were involved. This commit improves the general presentation of the
"ambiguous occurrence" error, and adds a bit of extra context in the
case of data families.

Fixes #23301

- - - - -
2f571afe by Sylvain Henry at 2023-05-15T14:50:07-04:00
Fix GHCJS OS platform (fix #23346)

- - - - -
86aae570 by Oleg Grenrus at 2023-05-15T14:50:43-04:00
Split DynFlags structure into own module

This will allow to make command line parsing to depend on
diagnostic system (which depends on dynflags)

- - - - -
07ec9f6f by Ben Gamari at 2023-05-15T16:57:54-04:00
Extension shuffling (#23291)

Where introduced 4 new extensions:
  - PatternSignatures
  - ExtendedForAllScope
  - MethodTypeVariables
  - ImplicitForAll

Tasks of ScopedTypeVariables extension were distributed between
PatternSignatures, ExtendedForAllScope and MethodTypeVariables according
to the proposal. Now ScopedTypeVaribles only implies these three exntesions.

Extension ImplicitForAll saves current behavior. NoImplicitForAll
disables implicit bounding of type variables in many contexts.

Was introduced one new warning option: -Wpattern-signature-binds
It warns when pattern signature binds into scope new type variable. For
example:

  f (a :: t) = ...

- - - - -


30 changed files:

- compiler/GHC/Builtin/primops.txt.pp
- compiler/GHC/ByteCode/Asm.hs
- compiler/GHC/ByteCode/Instr.hs
- compiler/GHC/Cmm/MachOp.hs
- compiler/GHC/Cmm/Parser.y
- compiler/GHC/CmmToAsm/AArch64/CodeGen.hs
- compiler/GHC/CmmToAsm/AArch64/Instr.hs
- compiler/GHC/CmmToAsm/AArch64/Ppr.hs
- compiler/GHC/CmmToAsm/PPC/CodeGen.hs
- compiler/GHC/CmmToAsm/PPC/Instr.hs
- compiler/GHC/CmmToAsm/PPC/Ppr.hs
- compiler/GHC/CmmToAsm/Wasm/FromCmm.hs
- compiler/GHC/CmmToAsm/X86/CodeGen.hs
- compiler/GHC/CmmToAsm/X86/Instr.hs
- compiler/GHC/CmmToAsm/X86/Ppr.hs
- compiler/GHC/CmmToC.hs
- compiler/GHC/CmmToLlvm/CodeGen.hs
- compiler/GHC/Core/Coercion.hs
- compiler/GHC/Core/Coercion.hs-boot
- compiler/GHC/Core/Coercion/Opt.hs
- compiler/GHC/Core/Lint.hs
- compiler/GHC/Core/Opt/CallerCC.hs
- compiler/GHC/Core/Opt/ConstantFold.hs
- compiler/GHC/Core/Opt/Monad.hs
- compiler/GHC/Core/Predicate.hs
- compiler/GHC/Core/Reduction.hs
- compiler/GHC/Core/Rules.hs
- compiler/GHC/Core/TyCo/Compare.hs
- compiler/GHC/Core/TyCo/Rep.hs
- compiler/GHC/Core/TyCo/Rep.hs-boot


The diff was not included because it is too large.


View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/b2ca3e4195208507480496600214141542585acb...07ec9f6f5e051b342ce1a5d18da26d5df1840736

-- 
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/b2ca3e4195208507480496600214141542585acb...07ec9f6f5e051b342ce1a5d18da26d5df1840736
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/20230515/b34318b8/attachment-0001.html>