Commits (95)
  • Sebastian Graf's avatar
    Much simpler language for PmCheck · 30f5ac07
    Sebastian Graf authored
    Simon realised that the simple language composed of let bindings, bang
    patterns and flat constructor patterns is enough to capture the
    semantics of the source pattern language that are important for
    pattern-match checking. Well, given that the Oracle is smart enough to
    connect the dots in this less informationally dense form, which it is
    So we transform `translatePat` to return a list of `PmGrd`s relative to
    an incoming match variable. `pmCheck` then trivially translates each of
    the `PmGrd`s into constraints that the oracle understands.
    Since we pass in the match variable, we incidentally fix #15884
    (coverage checks for view patterns) through an interaction with !1746.
  • Stefan Schulze Frielinghaus's avatar
    Hadrian: Take care of assembler source files · 166e1c2a
    Stefan Schulze Frielinghaus authored
    Fixes #17286.
  • John Ericson's avatar
    Simplify Configure in a few ways · c2290596
    John Ericson authored
     - No need to distinguish between gcc-llvm and clang. First of all,
       gcc-llvm is quite old and surely unmaintained by now. Second of all,
       none of the code actually care about that distinction!
       Now, it does make sense to consider C multiple frontends for LLVMs in
       the form of clang vs clang-cl (same clang, yes, but tweaked
       interface). But this is better handled in terms of "gccish vs
       mvscish" and "is LLVM", yielding 4 combinations. Therefore, I don't
       think it is useful saving the existing code for that.
     - Get the remaining CC_LLVM_BACKEND, and also TABLES_NEXT_TO_CODE in
       mk/config.h the normal way, rather than hacking it post-hoc. No point
       keeping these special cases around for now reason.
     - Get rid of hand-rolled `die` function and just use `AC_MSG_ERROR`.
     - Abstract check + flag override for unregisterised and tables next to
    Oh, and as part of the above I also renamed/combined some variables
    where it felt appropriate.
     - GccIsClang -> CcLlvmBackend. This is for `AC_SUBST`, like the other
     Camal case ones. It was never about gcc-llvm, or Apple's renamed clang,
     to be clear.
     - llvm_CC_FLAVOR -> CC_LLVM_BACKEND. This is for `AC_DEFINE`, like the
     other all-caps snake case ones. llvm_CC_FLAVOR was just silly
     indirection *and* an odd name to boot.
  • Vladislav Zavialov's avatar
    Escape stats file command (#13676) · f1ce3535
    Vladislav Zavialov authored
  • Vladislav Zavialov's avatar
    Skip T13767 on Darwin · cd1a8808
    Vladislav Zavialov authored
    The CI job fails with:
    	+++ rts/T13676.run/T13676.run.stderr.normalised	2019-10-09 12:27:56.000000000 -0700
    	@@ -0,0 +1,4 @@
    	+dyld: Library not loaded: @rpath/libHShaskeline-
    	+  Referenced from: /Users/builder/builds/ewzE5N2p/0/ghc/ghc/inplace/lib/bin/ghc
    	+  Reason: image not found
    	+*** Exception: readCreateProcess: '/Users/builder/builds/ewzE5N2p/0/ghc/ghc/inplace/lib/bin/ghc' '-B/Users/builder/builds/ewzE5N2p/0/ghc/ghc/inplace/lib' '-e' ''/''$'/'' == '/''/x0024'/''' +RTS '-tT13676.t'  (exit -6): failed
    Unable to reproduce locally.
  • Ryan Scott's avatar
    Use newDFunName for both manual and derived instances (#17339) · 0a338264
    Ryan Scott authored
    Issue #17339 was caused by using a slightly different version of
    `newDFunName` for derived instances that, confusingly enough, did not
    take all arguments to the class into account when generating the
    `DFun` name. I cannot think of any good reason for doing this, so
    this patch uses `newDFunName` uniformly for both derived instances
    and manually written instances alike.
    Fixes #17339.
  • Simon Peyton Jones's avatar
    Fix validity checking for inferred types · c50e4c92
    Simon Peyton Jones authored
    GHC is suposed to uphold the principle that an /inferred/ type
    for a let-binding should obey the rules for that module.  E.g.
    we should only accept an inferred higher rank type if we have
    RankNTypes on.
    But we were failing to check this: TcValidity.checkValidType
    allowed arbitrary rank for inferred types.
    This patch fixes the bug.  It might in principle cause some breakage,
    but if so that's good: the user should add RankNTypes and/or a
    manual signature.  (And almost every package has explicit user
    signatures for all top-level things anyway.)  Let's see.
    Fixes #17213.
    Metric Decrease:
  • Simon Peyton Jones's avatar
    Do not add a 'solved dict' for quantified constraints · 226d86d2
    Simon Peyton Jones authored
    GHC has a wonderful-but-delicate mechanism for building recursive
    dictionaries by adding a goal to the "solved dictionaries" before
    solving the sub-goals.  See Note [Solved dictionaries] in TcSMonad
    Ticket #17267 showed that if you use this mechanism for local
    /quantified/ constraints you can get a loop -- or even unsafe
    coerce.   This patch fixes the bug.
    * Make TcSMonad.addSolvedDict be conditional on using a
      /top level/ instance, not a quantified one.
    * Moreover, we /also/ don't want to add a solved dict
      for equalities (a~b).
    * Add lots more comments to Note [Solved dictionaries]
      to explain the above cryptic stuff.
    * Extend InstanceWhat to identify those strange built-in
      equality instances.
    A couple of other things along the way
    * Delete the unused Type.isIPPred_maybe.
    * Stop making addSolvedDict conditional on not being an
      impolicit parameter.  This comes from way back. But
      it's irrelevant now because IP dicts are never solved
      via an instance.
  • Alex D's avatar
  • Andreas Klebinger's avatar
    Fix #17334 where NCG did not properly update the CFG. · c1bd07cd
    Andreas Klebinger authored
    Statements can change the basic block in which instructions
    are placed during instruction selection.
    We have to keep track of this switch of the current basic block
    as we need this information in order to properly update the CFG.
    This commit implements this change and fixes #17334.
    We do so by having stmtToInstr return the new block id
    if a statement changed the basic block.
  • Takenobu Tani's avatar
    users-guide: Add GHCi's ::<builtin-command> form · 1eda9f28
    Takenobu Tani authored
    This commit explicitly adds description about double colon command
    of GHCi.
    [skip ci]
  • Takenobu Tani's avatar
  • Ryan Scott's avatar
    Add docs/users_guide/.log to .gitignore · 78463fc5
    Ryan Scott authored
    When the users guide fails to build (as in #17346), a
    `docs/users_guide/.log` file will be generated with contents that
    look something like this:
    WARNING: unknown config value 'latex_paper_size' in override, ignoring
    /home/rgscott/Software/ghc5/docs/users_guide/ghci.rst:3410: WARNING: u'ghc-flag' reference target not found: -pgmo ?option?
    /home/rgscott/Software/ghc5/docs/users_guide/ghci.rst:3410: WARNING: u'ghc-flag' reference target not found: -pgmo ?port?
    Encoding error:
    'ascii' codec can't encode character u'\u27e8' in position 132: ordinal not in range(128)
    The full traceback has been saved in /tmp/sphinx-err-rDF2LX.log, if you want to report the issue to the developers.
    This definitely should not be checked in to version control, so let's
    add this to `.gitignore`.
  • Ryan Scott's avatar
    Mention changes from #16980, #17213 in 8.10.1 release notes · 4aba72d6
    Ryan Scott authored
    The fixes for these issues both have user-facing consequences, so it
    would be good to mention them in the release notes for GHC 8.10.1.
    While I'm in town, also mention `UnboxedSums` in the release notes
    entry related to `-fobject-code`.
  • Ben Gamari's avatar
    gitlab-ci: Move hadrian-ghc-in-ghci job first · 0ca044fd
    Ben Gamari authored
    This is a very cheap job and can catch a number of "easy" failure modes
    (e.g. missing imports in the compiler). Let's run it first.
  • Ryan Scott's avatar
    Refactor some cruft in TcDerivInfer.inferConstraints · a2d3594c
    Ryan Scott authored
    The latest installment in my quest to clean up the code in
    `TcDeriv*`. This time, my sights are set on
    `TcDerivInfer.inferConstraints`, which infers the context for derived
    instances. This function is a wee bit awkward at the moment:
    * It's not terribly obvious from a quick glance, but
      `inferConstraints` is only ever invoked when using the `stock` or
      `anyclass` deriving strategies, as the code for inferring the
      context for `newtype`- or `via`-derived instances is located
      separately in `mk_coerce_based_eqn`. But there's no good reason
      for things to be this way, so I moved this code from
      `mk_coerce_based_eqn` to `inferConstraints` so that everything
      related to inferring instance contexts is located in one place.
    * In this process, I discovered that the Haddocks for the auxiliary
      function `inferConstraintsDataConArgs` are completely wrong. It
      claims that it handles both `stock` and `newtype` deriving, but
      this is completely wrong, as discussed above—it only handles
      `stock`. To rectify this, I renamed this function to
      `inferConstraintsStock` to reflect its actual purpose and created
      a new `inferConstraintsCoerceBased` function to specifically
      handle `newtype` (and `via`) deriving.
    Doing this revealed some opportunities for further simplification:
    * Removing the context-inference–related code from
      `mk_coerce_based_eqn` made me realize that the overall structure
      of the function is basically identical to `mk_originative_eqn`.
      In fact, I was easily able to combine the two functions into a
      single `mk_eqn_from_mechanism` function.
      As part of this merger, I now invoke
      `atf_coerce_based_error_checks` from `doDerivInstErrorChecks1`.
    * I discovered that GHC defined this function:
      typeToTypeKind = liftedTypeKind `mkVisFunTy` liftedTypeKind
      No fewer than four times in different modules. I consolidated all
      of these definitions in a single location in `TysWiredIn`.
  • Ryan Scott's avatar
    Don't skip validity checks for built-in classes (#17355) · 426b0ddc
    Ryan Scott authored
    Issue #17355 occurred because the control flow for
    `TcValidity.check_valid_inst_head` was structured in such a way that
    whenever it checked a special, built-in class (like `Generic` or
    `HasField`), it would skip the most important check of all:
    `checkValidTypePats`, which rejects nonsense like this:
    instance Generic (forall a. a)
    This fixes the issue by carving out `checkValidTypePats` from
    `check_valid_inst_head` so that `checkValidTypePats` is always
    invoked. `check_valid_inst_head` has also been renamed to
    `check_special_inst_head` to reflect its new purpose of _only_
    checking for instances headed by special classes.
    Fixes #17355.
  • Alp Mestanogullari's avatar
  • Ben Gamari's avatar
    hadrian: Add support for bindist compressors other than Xz · 9c11f817
    Ben Gamari authored
    Fixes #17351.
  • Andreas Klebinger's avatar
    Add loop level analysis to the NCG backend. · 535a88e1
    Andreas Klebinger authored
    For backends maintaining the CFG during codegen
    we can now find loops and their nesting level.
    This is based on the Cmm CFG and dominator analysis.
    As a result we can estimate edge frequencies a lot better
    for methods, resulting in far better code layout.
    Speedup on nofib: ~1.5%
    Increase in compile times: ~1.9%
    To make this feasible this commit adds:
    * Dominator analysis based on the Lengauer-Tarjan Algorithm.
    * An algorithm estimating global edge frequences from branch
    probabilities - In CFG.hs
    A few static branch prediction heuristics:
    * Expect to take the backedge in loops.
    * Expect to take the branch NOT exiting a loop.
    * Expect integer vs constant comparisons to be false.
    We also treat heap/stack checks special for branch prediction
    to avoid them being treated as loops.
  • Adithya Kumar's avatar
  • Takenobu Tani's avatar
    testsuite: Add test for #8305 · 19641957
    Takenobu Tani authored
    This is a test for the current algorithm of GHCi command name resolution.
    I add this test in preparation for updating GHCi command name resolution.
    For the current algorithm, see https://downloads.haskell.org/ghc/latest/docs/html/users_guide/ghci.html#the-ghci-files
  • Sebastian Graf's avatar
    Infer rho-types instead of sigma-types in guard BindStmts and TransStmts · 6ede3554
    Sebastian Graf authored
    In #17343 we saw that we didn't handle the pattern guard `!_ <-
    undefined` correctly: The `undefined` was never evaluated. Indeed,
    elaboration failed to insert the invisible type aruments to `undefined`.
    So `undefined` was trivially a normal-form and in turn never entered.
    The problem is that we used to infer a sigma-type for the RHS of the
    guard, the leading qualifiers of which will never be useful in a pattern
    match situation. Hence we infer a rho-type now.
    Fixes #17343.
  • John Ericson's avatar
    Delete ghctags cabal file · 798037a1
    John Ericson authored
    It came back to life in 381c3ae3 by
  • Richard Eisenberg's avatar
    Break up TcRnTypes, among other modules. · 51fad9e6
    Richard Eisenberg authored
    This introduces three new modules:
     - basicTypes/Predicate.hs describes predicates, moving
       this logic out of Type. Predicates don't really exist
       in Core, and so don't belong in Type.
     - typecheck/TcOrigin.hs describes the origin of constraints
       and types. It was easy to remove from other modules and
       can often be imported instead of other, scarier modules.
     - typecheck/Constraint.hs describes constraints as used in
       the solver. It is taken from TcRnTypes.
    No work other than module splitting is in this patch.
    This is the first step toward homogeneous equality, which will
    rely more strongly on predicates. And homogeneous equality is the
    next step toward a dependently typed core language.
  • Ben Gamari's avatar
    hadrian: Introduce enableDebugInfo flavour transformer · 11d4fc50
    Ben Gamari authored
    Also refactor things a bit to eliminate repetition.
  • Ryan Scott's avatar
    Make Coverage.TM a newtype · deb96399
    Ryan Scott authored
  • Brian Wignall's avatar
  • Ben Gamari's avatar
    testsuite: Ensure that makefile tests get run · b15a7fb8
    Ben Gamari authored
    Previously `makefile_test` and `run_command` tests could easily end up
    in a situation where they wouldn't be run if the user used the
    `only_ways` modifier. The reason is to build the set of a ways to run
    the test in we first start with a candidate set determined by the test
    type (e.g. `makefile_test`, `compile_run`, etc.) and then filter that
    set with the constraints given by the test's modifiers.
    `makefile_test` and `run_command` tests' candidate sets were simply
    `{normal}`, and consequently most uses of `only_ways` would result in
    the test being never run.
    To avoid this we rather use all ways as the candidate sets for these
    test types. This may result in a few more testcases than we would like
    (given that some `run_command` tests are insensitive to way) but this
    can be fixed by adding modifiers and we would much rather run too many
    tests than too few.
    This fixes #16042 and a number of other tests afflicted by the same issue.
    However, there were a few cases that required special attention:
     * `T14028` is currently failing and is therefore marked as broken due
       to #17300
     * `T-signals-child` is fragile in the `threaded1` and `threaded2` ways
       (tracked in #17307)
  • Richard Eisenberg's avatar
  • Ben Gamari's avatar
    testsuite: Assert that testsuite ways are known · c4c9904b
    Ben Gamari authored
    This ensures that all testsuite way names given to `omit_ways`,
    `only_ways`, etc. are known ways.
  • Ömer Sinan Ağacan's avatar
    rts/GC: Add an obvious assertion during block initialization · 697be2b6
    Ömer Sinan Ağacan authored
    Namely ensure that block descriptors are initialized with valid
    generation numbers.
    Co-Authored-By: Ben Gamari's avatarBen Gamari <ben@well-typed.com>
  • Ben Gamari's avatar
    rts: Add Note explaining applicability of selector optimisation depth limit · 61d2ed42
    Ben Gamari authored
    This was slightly non-obvious so a note seems deserved.
  • Ben Gamari's avatar
    rts/Capability: A few documentation comments · 11395037
    Ben Gamari authored
  • Ben Gamari's avatar
    rts: Give stack flags proper macros · 206f782a
    Ben Gamari authored
    This were previously quite unclear and will change a bit under the
    non-moving collector so let's clear this up now.
  • Ben Gamari's avatar
    rts/GC: Refactor gcCAFs · 81d4675e
    Ben Gamari authored
  • Ben Gamari's avatar
    rts: Fix macro parenthesisation · 4d674c4e
    Ben Gamari authored
  • Ben Gamari's avatar
    rts/Schedule: Allow synchronization without holding a capability · bfcafd39
    Ben Gamari authored
    The concurrent mark-and-sweep will be performed by a GHC task which will
    not hold a capability. This is necessary to avoid a concurrent mark from
    interfering with minor generation collections.
    However, the major collector must synchronize with the mutators at the
    end of marking to flush their update remembered sets. This patch extends
    the `requestSync` mechanism used to synchronize garbage collectors to
    allow synchronization without holding a capability.
    This change is fairly straightforward as the capability was previously
    only required for two reasons:
     1. to ensure that we don't try to re-acquire a capability that we
        the sync requestor already holds.
     2. to provide a way to suspend and later resume the sync request if
        there is already a sync pending.
    When synchronizing without holding a capability we needn't worry about
    consideration (1) at all.
    (2) is slightly trickier and may happen, for instance, when a capability
    requests a minor collection and shortly thereafter the non-moving mark
    thread requests a post-mark synchronization. In this case we need to
    ensure that the non-moving mark thread suspends his request until after
    the minor GC has concluded to avoid dead-locking. For this we introduce
    a condition variable, `sync_finished_cond`, which a
    non-capability-bearing requestor will wait on and which is signalled
    after a synchronization or GC has finished.
  • Ömer Sinan Ağacan's avatar
    rts/BlockAlloc: Allow aligned allocation requests · 921e4e36
    Ömer Sinan Ağacan authored
    This implements support for block group allocations which are aligned to
    an integral number of blocks.
    This will be used by the nonmoving garbage collector, which uses the
    block allocator to allocate the segments which back its heap. These
    segments are a fixed number of blocks in size, with each segment being
    aligned to the segment size boundary. This allows us to easily find the
    segment metadata stored at the beginning of the segment.
  • Ben Gamari's avatar
    Merge branches 'wip/gc/sync-without-capability' and... · 1db1f413
    Ben Gamari authored
    Merge branches 'wip/gc/sync-without-capability' and 'wip/gc/aligned-block-allocation' into wip/gc/preparation
  • Ömer Sinan Ağacan's avatar
    rts/StableName: Expose FOR_EACH_STABLE_NAME, freeSnEntry, SNT_size · 4b261381
    Ömer Sinan Ağacan authored
    These will be needed when we implement sweeping in the nonmoving
  • Ben Gamari's avatar
    rts: Disable aggregate-return warnings from gcc · ac650082
    Ben Gamari authored
    This warning is a bit of a relic; there is little reason to avoid
    aggregate return values in 2019.
  • Ömer Sinan Ağacan's avatar
    rts/Scav: Expose scavenging functions · a244d137
    Ömer Sinan Ağacan authored
    To keep the non-moving collector nicely separated from the moving
    collector its scavenging phase will live in another file,
    `NonMovingScav.c`. However, it will need to use these functions so
    let's expose them.
  • Ben Gamari's avatar
    rts: Introduce flag to enable the nonmoving old generation · 4a72fb4a
    Ben Gamari authored
    This flag will enable the use of a non-moving oldest generation.
  • Ben Gamari's avatar
    rts: Introduce debug flag for non-moving GC · 19fa0e8f
    Ben Gamari authored
  • Ömer Sinan Ağacan's avatar
    rts: Non-concurrent mark and sweep · 2309789a
    Ömer Sinan Ağacan authored
    This implements the core heap structure and a serial mark/sweep
    collector which can be used to manage the oldest-generation heap.
    This is the first step towards a concurrent mark-and-sweep collector
    aimed at low-latency applications.
    The full design of the collector implemented here is described in detail
    in a technical note
        B. Gamari. "A Concurrent Garbage Collector For the Glasgow Haskell
        Compiler" (2018)
    The basic heap structure used in this design is heavily inspired by
        K. Ueno & A. Ohori. "A fully concurrent garbage collector for
        functional programs on multicore processors." /ACM SIGPLAN Notices/
        Vol. 51. No. 9 (presented by ICFP 2016)
    This design is intended to allow both marking and sweeping
    concurrent to execution of a multi-core mutator. Unlike the Ueno design,
    which requires no global synchronization pauses, the collector
    introduced here requires a stop-the-world pause at the beginning and end
    of the mark phase.
    To avoid heap fragmentation, the allocator consists of a number of
    fixed-size /sub-allocators/. Each of these sub-allocators allocators into
    its own set of /segments/, themselves allocated from the block
    allocator. Each segment is broken into a set of fixed-size allocation
    blocks (which back allocations) in addition to a bitmap (used to track
    the liveness of blocks) and some additional metadata (used also used
    to track liveness).
    This heap structure enables collection via mark-and-sweep, which can be
    performed concurrently via a snapshot-at-the-beginning scheme (although
    concurrent collection is not implemented in this patch).
    The mark queue is a fairly straightforward chunked-array structure.
    The representation is a bit more verbose than a typical mark queue to
    accomodate a combination of two features:
     * a mark FIFO, which improves the locality of marking, reducing one of
       the major overheads seen in mark/sweep allocators (see [1] for
     * the selector optimization and indirection shortcutting, which
       requires that we track where we found each reference to an object
       in case we need to update the reference at a later point (e.g. when
       we find that it is an indirection). See Note [Origin references in
       the nonmoving collector] (in `NonMovingMark.h`) for details.
    Beyond this the mark/sweep is fairly run-of-the-mill.
    [1] R. Garner, S.M. Blackburn, D. Frampton. "Effective Prefetch for
        Mark-Sweep Garbage Collection." ISMM 2007.
    Co-Authored-By: Ben Gamari's avatarBen Gamari <ben@well-typed.com>
  • Ben Gamari's avatar
    testsuite: Add nonmoving WAY · 49dc09db
    Ben Gamari authored
    This simply runs the compile_and_run tests with `-xn`, enabling the
    nonmoving oldest generation.
  • Ben Gamari's avatar
    rts: Mark binder as const · 26ffa5a2
    Ben Gamari authored
  • Ben Gamari's avatar
    rts: Implement concurrent collection in the nonmoving collector · d7017446
    Ben Gamari authored
    This extends the non-moving collector to allow concurrent collection.
    The full design of the collector implemented here is described in detail
    in a technical note
        B. Gamari. "A Concurrent Garbage Collector For the Glasgow Haskell
        Compiler" (2018)
    This extension involves the introduction of a capability-local
    remembered set, known as the /update remembered set/, which tracks
    objects which may no longer be visible to the collector due to mutation.
    To maintain this remembered set we introduce a write barrier on
    mutations which is enabled while a concurrent mark is underway.
    The update remembered set representation is similar to that of the
    nonmoving mark queue, being a chunked array of `MarkEntry`s. Each
    `Capability` maintains a single accumulator chunk, which it flushed
    when it (a) is filled, or (b) when the nonmoving collector enters its
    post-mark synchronization phase.
    While the write barrier touches a significant amount of code it is
    conceptually straightforward: the mutator must ensure that the referee
    of any pointer it overwrites is added to the update remembered set.
    However, there are a few details:
     * In the case of objects with a dirty flag (e.g. `MVar`s) we can
       exploit the fact that only the *first* mutation requires a write
     * Weak references, as usual, complicate things. In particular, we must
       ensure that the referee of a weak object is marked if dereferenced by
       the mutator. For this we (unfortunately) must introduce a read
       barrier, as described in Note [Concurrent read barrier on deRefWeak#]
       (in `NonMovingMark.c`).
     * Stable names are also a bit tricky as described in Note [Sweeping
       stable names in the concurrent collector] (`NonMovingSweep.c`).
    We take quite some pains to ensure that the high thread count often seen
    in parallel Haskell applications doesn't affect pause times. To this end
    we allow thread stacks to be marked either by the thread itself (when it
    is executed or stack-underflows) or the concurrent mark thread (if the
    thread owning the stack is never scheduled). There is a non-trivial
    handshake to ensure that this happens without racing which is described
    in Note [StgStack dirtiness flags and concurrent marking].
    Co-Authored-by: Ömer Sinan Ağacan's avatarÖmer Sinan Ağacan <omer@well-typed.com>
  • Ben Gamari's avatar
  • Ben Gamari's avatar
  • Ben Gamari's avatar
    Don't cleanup until we've stopped the collector · 47d6b1a2
    Ben Gamari authored
    This requires that we break nonmovingExit into two pieces since we need
    to first stop the collector to relinquish any capabilities, then we need
    to shutdown the scheduler, then we need to free the nonmoving
  • Ben Gamari's avatar
    nonmoving: Flush update rememb. set before disabling capabilities · 431e9bf4
    Ben Gamari authored
    Disabled capabilities do not participate in the nonmoving collector's
    final mark synchronization. Consequently it is very important that we
    flush a capability's update remembered set before disabling it since
    otherwise the nonmoving collector may not see live references.
    Fixes setnumcapabilities001.
  • Ben Gamari's avatar
    rts: Tracing support for nonmoving collection events · 6db4be95
    Ben Gamari authored
    This introduces a few events to mark key points in the nonmoving
    garbage collection cycle. These include:
     * `EVENT_CONC_MARK_BEGIN`, denoting the beginning of a round of
       marking. This may happen more than once in a single major collection
       since we the major collector iterates until it hits a fixed point.
     * `EVENT_CONC_MARK_END`, denoting the end of a round of marking.
     * `EVENT_CONC_SYNC_BEGIN`, denoting the beginning of the post-mark
       synchronization phase
     * `EVENT_CONC_UPD_REM_SET_FLUSH`, indicating that a capability has
       flushed its update remembered set.
     * `EVENT_CONC_SYNC_END`, denoting that all mutators have flushed their
       update remembered sets.
     * `EVENT_CONC_SWEEP_BEGIN`, denoting the beginning of the sweep portion
       of the major collection.
     * `EVENT_CONC_SWEEP_END`, denoting the end of the sweep portion of the
       major collection.
  • Ben Gamari's avatar
    rts: Introduce non-moving heap census · 2c93c6fd
    Ben Gamari authored
    This introduces a simple census of the non-moving heap (not to be
    confused with the heap census used by the heap profiler). This
    collects basic heap usage information (number of allocated and free
    blocks) which is useful when characterising fragmentation of the
    nonmoving heap.
  • Ben Gamari's avatar
    rts/Eventlog: More descriptive error message · b8498bf0
    Ben Gamari authored
  • Ben Gamari's avatar
    Allow census without live word count · 02bf6a69
    Ben Gamari authored
    Otherwise the census is unsafe when mutators are running due to
    concurrent mutation.
  • Ben Gamari's avatar
    NonmovingCensus: Emit samples to eventlog · 6b5fd7b0
    Ben Gamari authored
  • Ben Gamari's avatar
    rts: Add GetMyThreadCPUTime helper · 3748326f
    Ben Gamari authored
  • Ben Gamari's avatar
  • Ben Gamari's avatar
    Fix it · e5c5f612
    Ben Gamari authored
  • Ben Gamari's avatar
    Stats: Add sync pauses to +RTS -S output · e21b7a87
    Ben Gamari authored
  • Ben Gamari's avatar
  • Ben Gamari's avatar
    Nonmoving: Allow aging and refactor static objects logic · 7ff503fd
    Ben Gamari authored
    This commit does two things:
     * Allow aging of objects during the preparatory minor GC
     * Refactor handling of static objects to avoid the use of a hashtable
  • Ben Gamari's avatar
  • Ben Gamari's avatar
    More comments for aging · 495ec475
    Ben Gamari authored
  • Ben Gamari's avatar
    NonMoving: Eliminate integer division in nonmovingBlockCount · 9ef76b05
    Ben Gamari authored
    Perf showed that the this single div was capturing up to 10% of samples
    in nonmovingMark. However, the overwhelming majority of cases is looking
    at small block sizes. These cases we can easily compute explicitly,
    allowing the compiler to turn the division into a significantly more
    efficient division-by-constant.
    While the increase in source code looks scary, this all optimises down
    to very nice looking assembler. At this point the only remaining
    hotspots in nonmovingBlockCount are due to memory access.
  • Ben Gamari's avatar
  • Ben Gamari's avatar
    NonMovingMark: Optimize representation of mark queue · 3c8ad7e0
    Ben Gamari authored
    This shortens MarkQueueEntry by 30% (one word)
  • Ben Gamari's avatar
    NonMoving: Optimize bitmap search during allocation · b023a0e7
    Ben Gamari authored
    Use memchr instead of a open-coded loop. This is nearly twice as fast in
    a synthetic benchmark.
  • Ben Gamari's avatar
    rts: Add prefetch macros · 32b17d50
    Ben Gamari authored
  • Ben Gamari's avatar
    NonMoving: Prefetch when clearing bitmaps · d5b62460
    Ben Gamari authored
    Ensure that the bitmap of the segmentt that we will clear next is in
    cache by the time we reach it.
  • Ben Gamari's avatar
    NonMoving: Inline nonmovingClearAllBitmaps · 67905e83
    Ben Gamari authored
  • Ben Gamari's avatar
  • Ben Gamari's avatar
    NonMoving: Pre-fetch during mark · f44e108d
    Ben Gamari authored
    This improved overall runtime on nofib's constraints test by nearly 10%.
  • Ben Gamari's avatar
    NonMoving: Prefetch segment header · 662bffb0
    Ben Gamari authored
  • Ben Gamari's avatar
    NonMoving: Optimise allocator cache behavior · f7584d48
    Ben Gamari authored
    Previously we would look at the segment header to determine the block
    size despite the fact that we already had the block size at hand.
  • Ben Gamari's avatar
  • Ben Gamari's avatar
    NonMoving: Don't do major GC if one is already running · 1d7b6fac
    Ben Gamari authored
    Previously we would perform a preparatory moving collection, resulting
    in many things being added to the mark queue. When we finished with this
    we would realize in nonmovingCollect that there was already a collection
    running, in which case we would simply not run the nonmoving collector.
    However, it was very easy to end up in a "treadmilling" situation: all
    subsequent GC following the first failed major GC would be scheduled as
    major GCs. Consequently we would continuously feed the concurrent
    collector with more mark queue entries and it would never finish.
    This patch aborts the major collection far earlier, meaning that we
    avoid adding nonmoving objects to the mark queue and allowing the
    concurrent collector to finish.
  • Ben Gamari's avatar
  • Ben Gamari's avatar
    NonMoving: Clean mut_list · 0ab6dd70
    Ben Gamari authored
  • Ben Gamari's avatar
    Unconditionally flush update remembered set during minor GC · 278f7f22
    Ben Gamari authored
    Flush the update remembered set. The goal here is to flush periodically to
    ensure that we don't end up with a thread who marks their stack on their
    local update remembered set and doesn't flush until the nonmoving sync
    period as this would result in a large fraction of the heap being marked
    during the sync pause.
  • Ben Gamari's avatar
    testsuite: Add nonmoving_thr way · 764c6307
    Ben Gamari authored
  • Ben Gamari's avatar
    testsuite: Add nonmoving_thr_ghc way · f14100eb
    Ben Gamari authored
    This uses the nonmoving collector when compiling the testcases.
  • Ben Gamari's avatar
    testsuite: Don't run T15892 in nonmoving ways · 5ab3e332
    Ben Gamari authored
    The nonmoving GC doesn't support `+RTS -G1`, which this test insists on.
  • Ben Gamari's avatar
  • Ben Gamari's avatar
  • Ben Gamari's avatar
  • Ben Gamari's avatar
    testsuite: Skip T15892 in nonmoving_thr_ghc · 1bac727f
    Ben Gamari authored
  • Ben Gamari's avatar
    ghc-heap: Skip heap_all test with debugged RTS · 0fb0b744
    Ben Gamari authored
    The debugged RTS initializes the heap with 0xaa, which breaks the
    (admittedly rather fragile) assumption that uninitialized fields are set
    to 0x00:
    Wrong exit code for heap_all(nonmoving)(expected 0 , actual 1 )
    Stderr ( heap_all ):
    heap_all: user error (assertClosuresEq: Closures do not match
    Expected: FunClosure {info = StgInfoTable {entry = Nothing, ptrs = 0, nptrs = 1, tipe = FUN_0_1, srtlen = 0, code = Nothing}, ptrArgs = [], dataArgs = [0]}
    Actual:   FunClosure {info = StgInfoTable {entry = Nothing, ptrs = 0, nptrs = 1, tipe = FUN_0_1, srtlen = 1032832, code = Nothing}, ptrArgs = [], dataArgs = [12297829382473034410]}
    CallStack (from HasCallStack):
      assertClosuresEq, called at heap_all.hs:230:9 in main:Main
  • Ben Gamari's avatar
    Skip ghc_heap_all test in nonmoving ways · d0fffae3
    Ben Gamari authored
  • Ben Gamari's avatar
    testsuite: Don't run T9630 in nonmoving ways · 1ae81a86
    Ben Gamari authored
    The nonmoving collector doesn't support -G1
  • Ben Gamari's avatar
    testsuite: Don't run T7160 in nonmoving_thr ways · 763a94fc
    Ben Gamari authored
    The nonmoving way finalizes things in a different order.
  • Ben Gamari's avatar
    testsuite: Mark length001 as failing under nonmoving ways · 4d059588
    Ben Gamari authored
    This is consistent with the other unoptimized ways.
  • Ben Gamari's avatar
......@@ -107,6 +107,7 @@ _darcs/
......@@ -227,7 +228,7 @@ ghc.nix/
# Tooling - direnv
# Tooling - vscode
......@@ -18,13 +18,14 @@ before_script:
- "git fetch https://gitlab.haskell.org/ghc/ghc-performance-notes.git refs/notes/perf:refs/notes/perf || true"
- lint # Source linting
- build # A quick smoke-test to weed out broken commits
- full-build # Build all the things
- cleanup # See Note [Cleanup after the shell executor]
- packaging # Source distribution, etc.
- testing # head.hackage correctness and compiler performance testing
- deploy # push documentation
- lint # Source linting
- quick-build # A very quick smoke-test to weed out broken commits
- build # A quick smoke-test to weed out broken commits
- full-build # Build all the things
- cleanup # See Note [Cleanup after the shell executor]
- packaging # Source distribution, etc.
- testing # head.hackage correctness and compiler performance testing
- deploy # push documentation
# N.B.Don't run on wip/ branches, instead on run on merge requests.
.only-default: &only-default
......@@ -229,7 +230,7 @@ validate-x86_64-linux-deb9-hadrian:
<<: *only-default
stage: build
stage: quick-build
image: "registry.gitlab.haskell.org/ghc/ci-images/x86_64-linux-deb9:$DOCKER_REV"
# workaround for docker permissions
......@@ -237,6 +238,8 @@ hadrian-ghc-in-ghci:
- git submodule sync --recursive
- git submodule update --init --recursive
- git checkout .gitmodules
GHC_FLAGS: -Werror
- x86_64-linux
......@@ -842,6 +842,48 @@ AC_DEFUN([FP_CHECK_SIZEOF_AND_ALIGNMENT],
# flag, name, anti name, var name, help string,
# [var true val], [var false val], [flag true val])
# ---------------------------------------------------
# Helper for when there is a automatic detection and an explicit flag for the
# user to override disable a feature, but not override enable a feature.
# $1 = flag of feature
# $2 = name of feature
# $3 = name of anti feature
# $4 = name of variable
# $5 = help string
# $6 = when true
# $7 = when false
# $8 = default explicit case (yes/no). Used for handle "backwards" legacy
# options where enabling makes fewer assumptions than disabling.
[test x"$enableval" = x"m4_default([$8],yes)"],
[AC_MSG_NOTICE([user chose $2 matching default for platform])],
[AC_MSG_ERROR([user chose $2 overriding only supported option for platform])],
[AC_MSG_ERROR([invalid default])])
[AC_MSG_NOTICE([user chose $3 overriding for platform])],
[AC_MSG_NOTICE([user chose $3 matching default for platform])],
[AC_MSG_ERROR([invalid default])])
# ---------------------
......@@ -1293,30 +1335,19 @@ AC_SUBST(GccLT46)
dnl Check to see if the C compiler is clang or llvm-gcc
AC_MSG_CHECKING([whether C compiler is clang])
AC_MSG_CHECKING([whether C compiler has an LLVM back end])
$CC -x c /dev/null -dM -E > conftest.txt 2>&1
if grep "__clang__" conftest.txt >/dev/null 2>&1; then
if grep "__llvm__" conftest.txt >/dev/null 2>&1; then
AC_DEFINE([CC_LLVM_BACKEND], [1], [Define (to 1) if C compiler has an LLVM back end])
AC_MSG_CHECKING([whether C compiler has an LLVM back end])
if grep "__llvm__" conftest.txt >/dev/null 2>&1; then
rm -f conftest.txt
This diff is collapsed.
......@@ -93,7 +93,7 @@ tracePm herald doc = do
-- | Generate a fresh `Id` of a given type
mkPmId :: Type -> DsM Id
mkPmId ty = getUniqueM >>= \unique ->
let occname = mkVarOccFS $ fsLit "$pm"
let occname = mkVarOccFS $ fsLit "pm"
name = mkInternalName unique occname noSrcSpan
in return (mkLocalId name ty)
......@@ -1576,8 +1576,8 @@ addVarCoreCt delta x e = runMaybeT (execStateT (core_expr x e) delta)
= do { arg_ids <- traverse bind_expr args
; data_con_app x dc arg_ids }
-- See Note [Detecting pattern synonym applications in expressions]
| Var y <- e, not (isDataConWorkId x)
-- We don't consider (unsaturated!) DataCons as flexible variables
| Var y <- e, Nothing <- isDataConId_maybe x
-- We don't consider DataCons flexible variables
= modifyT (\delta -> addVarVarCt delta (x, y))
| otherwise
-- Any other expression. Try to find other uses of a semantically
......@@ -1635,9 +1635,9 @@ Compared to the situation where P and Q are DataCons, the lack of generativity
means we could never flag Q as redundant.
(also see Note [Undecidable Equality for PmAltCons] in PmTypes.)
On the other hand, if we fail to recognise the pattern synonym, we flag the
pattern match as incomplete. That wouldn't happen if had knowledge about the
scrutinee, in which case the oracle basically knows "If it's a P, then its field
is 15".
pattern match as inexhaustive. That wouldn't happen if we had knowledge about
the scrutinee, in which case the oracle basically knows "If it's a P, then its
field is 15".
This is a pretty narrow use case and I don't think we should to try to fix it
until a user complains energetically.
......@@ -631,6 +631,7 @@ emitBlackHoleCode node = do
-- work with profiling.
when eager_blackholing $ do
whenUpdRemSetEnabled dflags $ emitUpdRemSetPushThunk node
emitStore (cmmOffsetW dflags node (fixedHdrSizeW dflags)) currentTSOExpr
-- See Note [Heap memory barriers] in SMP.h.
emitPrimCall [] MO_WriteBarrier []
......@@ -42,6 +42,7 @@ import BlockId
import MkGraph
import StgSyn
import Cmm
import Module ( rtsUnitId )
import Type ( Type, tyConAppTyCon )
import TyCon
import CLabel
......@@ -339,14 +340,20 @@ dispatchPrimop dflags = \case
emitAssign (CmmLocal res) (cmmLoadIndexW dflags mutv (fixedHdrSizeW dflags) (gcWord dflags))
WriteMutVarOp -> \[mutv, var] -> OpDest_AllDone $ \res@[] -> do
old_val <- CmmLocal <$> newTemp (cmmExprType dflags var)
emitAssign old_val (cmmLoadIndexW dflags mutv (fixedHdrSizeW dflags) (gcWord dflags))
-- Without this write barrier, other CPUs may see this pointer before
-- the writes for the closure it points to have occurred.
-- Note that this also must come after we read the old value to ensure
-- that the read of old_val comes before another core's write to the
-- MutVar's value.
emitPrimCall res MO_WriteBarrier []
emitStore (cmmOffsetW dflags mutv (fixedHdrSizeW dflags)) var
[{-no results-}]
(CmmLit (CmmLabel mkDirty_MUT_VAR_Label))
[(baseExpr, AddrHint), (mutv,AddrHint)]
[(baseExpr, AddrHint), (mutv, AddrHint), (CmmReg old_val, AddrHint)]
-- #define sizzeofByteArrayzh(r,a) \
-- r = ((StgArrBytes *)(a))->bytes
......@@ -1983,17 +1990,21 @@ doWritePtrArrayOp :: CmmExpr
doWritePtrArrayOp addr idx val
= do dflags <- getDynFlags
let ty = cmmExprType dflags val
hdr_size = arrPtrsHdrSize dflags
-- Update remembered set for non-moving collector
whenUpdRemSetEnabled dflags
$ emitUpdRemSetPush (cmmLoadIndexOffExpr dflags hdr_size ty addr ty idx)
-- This write barrier is to ensure that the heap writes to the object
-- referred to by val have happened before we write val into the array.
-- See #12469 for details.
emitPrimCall [] MO_WriteBarrier []
mkBasicIndexedWrite (arrPtrsHdrSize dflags) Nothing addr ty idx val
mkBasicIndexedWrite hdr_size Nothing addr ty idx val
emit (setInfo addr (CmmLit (CmmLabel mkMAP_DIRTY_infoLabel)))
-- the write barrier. We must write a byte into the mark table:
-- bits8[a + header_size + StgMutArrPtrs_size(a) + x >> N]
-- the write barrier. We must write a byte into the mark table:
-- bits8[a + header_size + StgMutArrPtrs_size(a) + x >> N]
emit $ mkStore (
cmmOffsetExpr dflags
(cmmOffsetExprW dflags (cmmOffsetB dflags addr (arrPtrsHdrSize dflags))
(cmmOffsetExprW dflags (cmmOffsetB dflags addr hdr_size)
(loadArrPtrsSize dflags addr))
(CmmMachOp (mo_wordUShr dflags) [idx,
mkIntExpr dflags (mUT_ARR_PTRS_CARD_BITS dflags)])
......@@ -2584,6 +2595,9 @@ emitCopyArray copy src0 src_off dst0 dst_off0 n =
dst <- assignTempE dst0
dst_off <- assignTempE dst_off0
-- Nonmoving collector write barrier
emitCopyUpdRemSetPush dflags (arrPtrsHdrSizeW dflags) dst dst_off n
-- Set the dirty bit in the header.
emit (setInfo dst (CmmLit (CmmLabel mkMAP_DIRTY_infoLabel)))
......@@ -2646,6 +2660,9 @@ emitCopySmallArray copy src0 src_off dst0 dst_off n =
src <- assignTempE src0
dst <- assignTempE dst0
-- Nonmoving collector write barrier
emitCopyUpdRemSetPush dflags (smallArrPtrsHdrSizeW dflags) dst dst_off n
-- Set the dirty bit in the header.
emit (setInfo dst (CmmLit (CmmLabel mkSMAP_DIRTY_infoLabel)))
......@@ -2774,6 +2791,12 @@ doWriteSmallPtrArrayOp :: CmmExpr
doWriteSmallPtrArrayOp addr idx val = do
dflags <- getDynFlags
let ty = cmmExprType dflags val
-- Update remembered set for non-moving collector
tmp <- newTemp ty
mkBasicIndexedRead (smallArrPtrsHdrSize dflags) Nothing ty tmp addr ty idx
whenUpdRemSetEnabled dflags $ emitUpdRemSetPush (CmmReg (CmmLocal tmp))
emitPrimCall [] MO_WriteBarrier [] -- #12469
mkBasicIndexedWrite (smallArrPtrsHdrSize dflags) Nothing addr ty idx val
emit (setInfo addr (CmmLit (CmmLabel mkSMAP_DIRTY_infoLabel)))
......@@ -2953,3 +2976,31 @@ emitCtzCall res x width = do
[ res ]
(MO_Ctz width)
[ x ]
-- Pushing to the update remembered set
-- | Push a range of pointer-array elements that are about to be copied over to
-- the update remembered set.
emitCopyUpdRemSetPush :: DynFlags
-> WordOff -- ^ array header size
-> CmmExpr -- ^ destination array
-> CmmExpr -- ^ offset in destination array (in words)
-> Int -- ^ number of elements to copy
-> FCode ()
emitCopyUpdRemSetPush _dflags _hdr_size _dst _dst_off 0 = return ()
emitCopyUpdRemSetPush dflags hdr_size dst dst_off n =
whenUpdRemSetEnabled dflags $ do
updfr_off <- getUpdFrameOff
graph <- mkCall lbl (NativeNodeCall,NativeReturn) [] args updfr_off []
emit graph
lbl = mkLblExpr $ mkPrimCallLabel
$ PrimCall (fsLit "stg_copyArray_barrier") rtsUnitId
args =
[ mkIntExpr dflags hdr_size
, dst
, dst_off
, mkIntExpr dflags n
......@@ -131,8 +131,8 @@ import DynFlags
-- Turgid imports for showTypeCategory
import PrelNames
import TcType
import Type
import TyCon
import Predicate
import Data.Maybe
import qualified Data.Char
......@@ -39,6 +39,11 @@ module GHC.StgToCmm.Utils (
newStringCLit, newByteStringCLit,
-- * Update remembered set operations
) where
#include "HsVersions.h"
......@@ -576,3 +581,40 @@ assignTemp' e
let reg = CmmLocal lreg
emitAssign reg e
return (CmmReg reg)
-- Pushing to the update remembered set
whenUpdRemSetEnabled :: DynFlags -> FCode a -> FCode ()
whenUpdRemSetEnabled dflags code = do
do_it <- getCode code
the_if <- mkCmmIfThenElse' is_enabled do_it mkNop (Just False)
emit the_if
enabled = CmmLoad (CmmLit $ CmmLabel mkNonmovingWriteBarrierEnabledLabel) (bWord dflags)
zero = zeroExpr dflags
is_enabled = cmmNeWord dflags enabled zero
-- | Emit code to add an entry to a now-overwritten pointer to the update
-- remembered set.
emitUpdRemSetPush :: CmmExpr -- ^ value of pointer which was overwritten
-> FCode ()
emitUpdRemSetPush ptr = do
(fsLit "updateRemembSetPushClosure_")
[(CmmReg (CmmGlobal BaseReg), AddrHint),
(ptr, AddrHint)]
emitUpdRemSetPushThunk :: CmmExpr -- ^ the thunk
-> FCode ()
emitUpdRemSetPushThunk ptr = do
(fsLit "updateRemembSetPushThunk_")
[(CmmReg (CmmGlobal BaseReg), AddrHint),
(ptr, AddrHint)]
......@@ -1368,12 +1368,7 @@ cvtTypeKind ty_str ty
TupleT n
| Just normals <- m_normals
, normals `lengthIs` n -- Saturated
-> if n==1 then return (head normals) -- Singleton tuples treated
-- like nothing (ie just parens)
else returnL (HsTupleTy noExtField
HsBoxedOrConstraintTuple normals)
| n == 1
-> failWith (ptext (sLit ("Illegal 1-tuple " ++ ty_str ++ " constructor")))
-> returnL (HsTupleTy noExtField HsBoxedOrConstraintTuple normals)
| otherwise
-> mk_apps
(HsTyVar noExtField NotPromoted (noLoc (getRdrName (tupleTyCon Boxed n))))
......@@ -1491,8 +1486,6 @@ cvtTypeKind ty_str ty
-- Promoted data constructor; hence cName
PromotedTupleT n
| n == 1
-> failWith (ptext (sLit ("Illegal promoted 1-tuple " ++ ty_str)))
| Just normals <- m_normals
, normals `lengthIs` n -- Saturated
-> returnL (HsExplicitTupleTy noExtField normals)
......@@ -106,7 +106,9 @@ module BasicTypes(
IntWithInf, infinity, treatZeroAsInf, mkIntWithInf, intGtLimit,
TypeOrKind(..), isTypeLevel, isKindLevel
) where
import GhcPrelude
......@@ -1644,3 +1646,25 @@ data SpliceExplicitFlag
= ExplicitSplice | -- ^ <=> $(f x y)
ImplicitSplice -- ^ <=> f x y, i.e. a naked top level expression
deriving Data
{- *********************************************************************
* *
Types vs Kinds
* *
********************************************************************* -}
-- | Flag to see whether we're type-checking terms or kind-checking types
data TypeOrKind = TypeLevel | KindLevel
deriving Eq
instance Outputable TypeOrKind where
ppr TypeLevel = text "TypeLevel"
ppr KindLevel = text "KindLevel"
isTypeLevel :: TypeOrKind -> Bool
isTypeLevel TypeLevel = True
isTypeLevel KindLevel = False
isKindLevel :: TypeOrKind -> Bool
isKindLevel TypeLevel = False
isKindLevel KindLevel = True
......@@ -73,6 +73,7 @@ import FieldLabel
import Class
import Name
import PrelNames
import Predicate
import Var
import VarSet( emptyVarSet )
import Outputable
......@@ -73,9 +73,6 @@ module Id (
isConLikeId, isBottomingId, idIsFrom,
-- ** Evidence variables
DictId, isDictId, isEvVar,
-- ** Join variables
JoinId, isJoinId, isJoinId_maybe, idJoinArity,
asJoinId, asJoinId_maybe, zapJoinId,
......@@ -129,7 +126,7 @@ import IdInfo
import BasicTypes
-- Imported and re-exported
import Var( Id, CoVar, DictId, JoinId,
import Var( Id, CoVar, JoinId,
InId, InVar,
OutId, OutVar,
idInfo, idDetails, setIdDetails, globaliseId, varType,
......@@ -584,20 +581,6 @@ isDeadBinder :: Id -> Bool
isDeadBinder bndr | isId bndr = isDeadOcc (idOccInfo bndr)
| otherwise = False -- TyVars count as not dead
* *
Evidence variables
* *
isEvVar :: Var -> Bool
isEvVar var = isEvVarType (varType var)
isDictId :: Id -> Bool
isDictId id = isDictTy (idType id)
* *
Describes predicates as they are considered by the solver.
module Predicate (
Pred(..), classifyPredType,
isPredTy, isEvVarType,
-- Equality predicates
EqRel(..), eqRelRole,
isEqPrimPred, isEqPred,
getEqPredTys, getEqPredTys_maybe, getEqPredRole,
mkPrimEqPred, mkReprPrimEqPred, mkPrimEqPredRole,
mkHeteroPrimEqPred, mkHeteroReprPrimEqPred,
-- Class predicates
mkClassPred, isDictTy,
isClassPred, isEqPredClass, isCTupleClass,
getClassPredTys, getClassPredTys_maybe,
-- Implicit parameters
isIPPred, isIPPred_maybe, isIPTyCon, isIPClass, hasIPPred,
-- Evidence variables
DictId, isEvVar, isDictId
) where
import GhcPrelude
import Type
import Class
import TyCon
import Var
import Coercion
import PrelNames
import FastString
import Outputable
import Util
import Control.Monad ( guard )
-- | A predicate in the solver. The solver tries to prove Wanted predicates
-- from Given ones.
data Pred
= ClassPred Class [Type]
| EqPred EqRel Type Type
| IrredPred PredType
| ForAllPred [TyCoVarBinder] [PredType] PredType
-- ForAllPred: see Note [Quantified constraints] in TcCanonical
-- NB: There is no TuplePred case
-- Tuple predicates like (Eq a, Ord b) are just treated
-- as ClassPred, as if we had a tuple class with two superclasses
-- class (c1, c2) => (%,%) c1 c2
classifyPredType :: PredType -> Pred
classifyPredType ev_ty = case splitTyConApp_maybe ev_ty of
Just (tc, [_, _, ty1, ty2])
| tc `hasKey` eqReprPrimTyConKey -> EqPred ReprEq ty1 ty2
| tc `hasKey` eqPrimTyConKey -> EqPred NomEq ty1 ty2
Just (tc, tys)
| Just clas <- tyConClass_maybe tc
-> ClassPred clas tys
_ | (tvs, rho) <- splitForAllVarBndrs ev_ty
, (theta, pred) <- splitFunTys rho
, not (null tvs && null theta)
-> ForAllPred tvs theta pred
| otherwise
-> IrredPred ev_ty
-- --------------------- Dictionary types ---------------------------------
mkClassPred :: Class -> [Type] -> PredType
mkClassPred clas tys = mkTyConApp (classTyCon clas) tys
isDictTy :: Type -> Bool
isDictTy = isClassPred
getClassPredTys :: HasDebugCallStack => PredType -> (Class, [Type])
getClassPredTys ty = case getClassPredTys_maybe ty of
Just (clas, tys) -> (clas, tys)
Nothing -> pprPanic "getClassPredTys" (ppr ty)
getClassPredTys_maybe :: PredType -> Maybe (Class, [Type])
getClassPredTys_maybe ty = case splitTyConApp_maybe ty of
Just (tc, tys) | Just clas <- tyConClass_maybe tc -> Just (clas, tys)
_ -> Nothing
-- --------------------- Equality predicates ---------------------------------
-- | A choice of equality relation. This is separate from the type 'Role'
-- because 'Phantom' does not define a (non-trivial) equality relation.
data EqRel = NomEq | ReprEq
deriving (Eq, Ord)
instance Outputable EqRel where
ppr NomEq = text "nominal equality"
ppr ReprEq = text "representational equality"
eqRelRole :: EqRel -> Role
eqRelRole NomEq = Nominal
eqRelRole ReprEq = Representational
getEqPredTys :: PredType -> (Type, Type)
getEqPredTys ty
= case splitTyConApp_maybe ty of
Just (tc, [_, _, ty1, ty2])
| tc `hasKey` eqPrimTyConKey
|| tc `hasKey` eqReprPrimTyConKey
-> (ty1, ty2)
_ -> pprPanic "getEqPredTys" (ppr ty)
getEqPredTys_maybe :: PredType -> Maybe (Role, Type, Type)
getEqPredTys_maybe ty
= case splitTyConApp_maybe ty of
Just (tc, [_, _, ty1, ty2])
| tc `hasKey` eqPrimTyConKey -> Just (Nominal, ty1, ty2)
| tc `hasKey` eqReprPrimTyConKey -> Just (Representational, ty1, ty2)
_ -> Nothing
getEqPredRole :: PredType -> Role
getEqPredRole ty = eqRelRole (predTypeEqRel ty)
-- | Get the equality relation relevant for a pred type.
predTypeEqRel :: PredType -> EqRel
predTypeEqRel ty
| Just (tc, _) <- splitTyConApp_maybe ty
, tc `hasKey` eqReprPrimTyConKey
= ReprEq
| otherwise
= NomEq
Predicates on PredType
Note [Evidence for quantified constraints]
The superclass mechanism in TcCanonical.makeSuperClasses risks
taking a quantified constraint like
(forall a. C a => a ~ b)
and generate superclass evidence
(forall a. C a => a ~# b)
This is a funny thing: neither isPredTy nor isCoVarType are true
of it. So we are careful not to generate it in the first place:
see Note [Equality superclasses in quantified constraints]
in TcCanonical.
isEvVarType :: Type -> Bool
-- True of (a) predicates, of kind Constraint, such as (Eq a), and (a ~ b)
-- (b) coercion types, such as (t1 ~# t2) or (t1 ~R# t2)
-- See Note [Types for coercions, predicates, and evidence] in TyCoRep
-- See Note [Evidence for quantified constraints]
isEvVarType ty = isCoVarType ty || isPredTy ty
isEqPredClass :: Class -> Bool
-- True of (~) and (~~)
isEqPredClass cls = cls `hasKey` eqTyConKey
|| cls `hasKey` heqTyConKey
isClassPred, isEqPred, isEqPrimPred, isIPPred :: PredType -> Bool
isClassPred ty = case tyConAppTyCon_maybe ty of
Just tyCon | isClassTyCon tyCon -> True
_ -> False
isEqPred ty -- True of (a ~ b) and (a ~~ b)
-- ToDo: should we check saturation?
| Just tc <- tyConAppTyCon_maybe ty
, Just cls <- tyConClass_maybe tc
= isEqPredClass cls
| otherwise
= False
isEqPrimPred ty = isCoVarType ty
-- True of (a ~# b) (a ~R# b)
isIPPred ty = case tyConAppTyCon_maybe ty of
Just tc -> isIPTyCon tc
_ -> False
isIPTyCon :: TyCon -> Bool
isIPTyCon tc = tc `hasKey` ipClassKey
-- Class and its corresponding TyCon have the same Unique
isIPClass :: Class -> Bool
isIPClass cls = cls `hasKey` ipClassKey
isCTupleClass :: Class -> Bool
isCTupleClass cls = isTupleTyCon (classTyCon cls)
isIPPred_maybe :: Type -> Maybe (FastString, Type)
isIPPred_maybe ty =
do (tc,[t1,t2]) <- splitTyConApp_maybe ty
guard (isIPTyCon tc)
x <- isStrLitTy t1
return (x,t2)
hasIPPred :: PredType -> Bool
hasIPPred pred
= case classifyPredType pred of
ClassPred cls tys
| isIPClass cls -> True
| isCTupleClass cls -> any hasIPPred tys
_other -> False
* *
Evidence variables
* *
isEvVar :: Var -> Bool
isEvVar var = isEvVarType (varType var)
isDictId :: Id -> Bool
isDictId id = isDictTy (varType id)
......@@ -40,6 +40,7 @@ module CLabel (
......@@ -484,7 +485,9 @@ mkBlockInfoTableLabel name c = IdLabel name c BlockInfoTable
-- See Note [Proc-point local block entry-point].
-- Constructing Cmm Labels
mkDirty_MUT_VAR_Label, mkUpdInfoLabel,
mkBHUpdInfoLabel, mkIndStaticInfoLabel, mkMainCapabilityLabel,
......@@ -494,6 +497,8 @@ mkDirty_MUT_VAR_Label, mkUpdInfoLabel,
mkSMAP_DIRTY_infoLabel, mkBadAlignmentLabel :: CLabel
mkDirty_MUT_VAR_Label = mkForeignLabel (fsLit "dirty_MUT_VAR") Nothing ForeignLabelInExternalPackage IsFunction
= CmmLabel rtsUnitId (fsLit "nonmoving_write_barrier_enabled") CmmData
mkUpdInfoLabel = CmmLabel rtsUnitId (fsLit "stg_upd_frame") CmmInfo
mkBHUpdInfoLabel = CmmLabel rtsUnitId (fsLit "stg_bh_upd_frame" ) CmmInfo
mkIndStaticInfoLabel = CmmLabel rtsUnitId (fsLit "stg_IND_STATIC") CmmInfo
......@@ -6,8 +6,6 @@
{-# LANGUAGE ScopedTypeVariables #-}
{-# LANGUAGE TypeFamilies #-}
{-# OPTIONS_GHC -fprof-auto-top #-}
-- Copyright (c) 2010, João Dias, Simon Marlow, Simon Peyton Jones,
-- and Norman Ramsey
......@@ -108,6 +106,7 @@ analyzeCmm
-> FactBase f
-> FactBase f
analyzeCmm dir lattice transfer cmmGraph initFact =
{-# SCC analyzeCmm #-}
let entry = g_entry cmmGraph
hooplGraph = g_graph cmmGraph
blockMap =
......@@ -169,7 +168,7 @@ rewriteCmm
-> CmmGraph
-> FactBase f
-> UniqSM (CmmGraph, FactBase f)
rewriteCmm dir lattice rwFun cmmGraph initFact = do
rewriteCmm dir lattice rwFun cmmGraph initFact = {-# SCC rewriteCmm #-} do
let entry = g_entry cmmGraph
hooplGraph = g_graph cmmGraph
blockMap1 =
......@@ -30,6 +30,7 @@ import VarEnv
import Id
import Type
import TyCon ( initRecTc, checkRecTc )
import Predicate ( isDictTy )
import Coercion
import BasicTypes
import Unique
......@@ -517,7 +518,7 @@ mk_cheap_fn dflags cheap_app
= \e mb_ty -> exprIsCheapX cheap_app e
|| case mb_ty of
Nothing -> False
Just ty -> isDictLikeTy ty
Just ty -> isDictTy ty
......@@ -624,9 +625,6 @@ The (foo DInt) is floated out, and makes ineffective a RULE
One could go further and make exprIsCheap reply True to any
dictionary-typed expression, but that's more work.
See Note [Dictionary-like types] in TcType.hs for why we use
isDictLikeTy here rather than isDictTy
Note [Eta expanding thunks]
We don't eta-expand
......@@ -77,6 +77,7 @@ import Id
import IdInfo
import PrelNames( absentErrorIdKey )
import Type
import Predicate
import TyCoRep( TyCoBinder(..), TyBinder )
import Coercion
import TyCon
......@@ -1071,7 +1071,7 @@ noFVs = emptyOccEnv
-- to filter additions to the latter. This gives us complete control
-- over what free variables we track.
data TM a = TM { unTM :: TickTransEnv -> TickTransState -> (a,FreeVars,TickTransState) }
newtype TM a = TM { unTM :: TickTransEnv -> TickTransState -> (a,FreeVars,TickTransState) }
deriving (Functor)
-- a combination of a state monad (TickTransState) and a writer
-- monad (FreeVars).
......@@ -327,7 +327,7 @@ dsProcExpr pat (dL->L _ (HsCmdTop (CmdTopTc _unitTy cmd_ty ids) cmd)) = do
fail_expr <- mkFailExpr ProcExpr env_stk_ty
var <- selectSimpleMatchVarL pat
match_code <- matchSimply (Var var) ProcExpr pat env_stk_expr fail_expr
let pat_ty = hsLPatType pat
let pat_ty = hsPatType pat
let proc_code = do_premap meth_ids pat_ty env_stk_ty cmd_ty
(Lam var match_code)
......@@ -868,7 +868,7 @@ dsCmdStmt ids local_vars out_ids (BodyStmt c_ty cmd _ _) env_ids = do
-- but that's likely to be defined in terms of first.
dsCmdStmt ids local_vars out_ids (BindStmt _ pat cmd _ _) env_ids = do
let pat_ty = hsLPatType pat
let pat_ty = hsPatType pat
(core_cmd, fv_cmd, env_ids1) <- dsfixCmd ids local_vars unitTy pat_ty cmd
let pat_vars = mkVarSet (collectPatBinders pat)
......@@ -41,6 +41,7 @@ import CoreArity ( etaExpand )
import CoreUnfold
import CoreFVs
import Digraph
import Predicate
import PrelNames
import TyCon
module DsBinds where
import DsMonad ( DsM )
import CoreSyn ( CoreExpr )
import TcEvidence (HsWrapper)
dsHsWrapper :: HsWrapper -> DsM (CoreExpr -> CoreExpr)
......@@ -930,7 +930,7 @@ dsDo stmts
(pat, dsDo (stmts ++ [noLoc $ mkLastStmt (noLoc ret)]))
do_arg (XApplicativeArg nec) = noExtCon nec
arg_tys = map hsLPatType pats
arg_tys = map hsPatType pats
; rhss' <- sequence rhss
......@@ -279,7 +279,7 @@ deBindComp pat core_list1 quals core_list2 = do
let u3_ty@u1_ty = exprType core_list1 -- two names, same thing
-- u1_ty is a [alpha] type, and u2_ty = alpha
let u2_ty = hsLPatType pat
let u2_ty = hsPatType pat
let res_ty = exprType core_list2
h_ty = u1_ty `mkVisFunTy` res_ty
......@@ -373,7 +373,7 @@ dfBindComp :: Id -> Id -- 'c' and 'n'
-> DsM CoreExpr
dfBindComp c_id n_id (pat, core_list1) quals = do
-- find the required type
let x_ty = hsLPatType pat
let x_ty = hsPatType pat
let b_ty = idType n_id
-- create some new local id's
......@@ -672,7 +672,7 @@ mkSelectorBinds ticks pat val_expr
= return (v, [(v, val_expr)])
| is_flat_prod_lpat pat' -- Special case (B)
= do { let pat_ty = hsLPatType pat'
= do { let pat_ty = hsPatType pat'
; val_var <- newSysLocalDsNoLP pat_ty
; let mk_bind tick bndr_var
......@@ -758,7 +758,7 @@ mkLHsPatTup lpats = cL (getLoc (head lpats)) $
mkVanillaTuplePat :: [OutPat GhcTc] -> Boxity -> Pat GhcTc
-- A vanilla tuple pattern simply gets its type from its sub-patterns
mkVanillaTuplePat pats box = TuplePat (map hsLPatType pats) pats box
mkVanillaTuplePat pats box = TuplePat (map hsPatType pats) pats box
-- The Big equivalents for the source tuple expressions
mkBigLHsVarTupId :: [Id] -> LHsExpr GhcTc
......@@ -214,6 +214,7 @@ Library
......@@ -505,6 +506,8 @@ Library
......@@ -593,6 +596,7 @@ Library
......@@ -10,6 +10,7 @@
-- | Binary interface file support.
module BinIface (
-- * Public API for interface file serialisation
......@@ -17,7 +18,16 @@ module BinIface (
-- * Internal serialisation functions
) where
......@@ -311,6 +311,7 @@ import GhcMonad
import TcRnMonad ( finalSafeMode, fixSafeInstances, initIfaceTcRn )
import LoadIface ( loadSysInterface )
import TcRnTypes
import Predicate
import Packages
import NameSet
import RdrName
......@@ -63,6 +63,9 @@ import TyCon
import Type hiding( typeKind )
import RepType
import TcType
import Constraint
import TcOrigin
import Predicate
import Var
import Id
import Name hiding ( varName )
......@@ -558,12 +558,11 @@ cmmNativeGen dflags this_mod modLoc ncgImpl us fileIds dbgMap cmm count
(cmmTopCodeGen ncgImpl)
fileIds dbgMap opt_cmm cmmCfg
dumpIfSet_dyn dflags
Opt_D_dump_asm_native "Native code"
(vcat $ map (pprNatCmmDecl ncgImpl) native)
dumpIfSet_dyn dflags
when (not $ null nativeCfgWeights) $ dumpIfSet_dyn dflags
Opt_D_dump_cfg_weights "CFG Weights"
(pprEdgeWeights nativeCfgWeights)
......@@ -679,19 +678,20 @@ cmmNativeGen dflags this_mod modLoc ncgImpl us fileIds dbgMap cmm count
cfgRegAllocUpdates = (concatMap Linear.ra_fixupList raStats)
let cfgWithFixupBlks =
addNodesBetween nativeCfgWeights cfgRegAllocUpdates
(\cfg -> addNodesBetween cfg cfgRegAllocUpdates) <$> livenessCfg
-- Insert stack update blocks
let postRegCFG =
foldl' (\m (from,to) -> addImmediateSuccessor from to m )
cfgWithFixupBlks stack_updt_blks
pure (foldl' (\m (from,to) -> addImmediat