August 2017 - ghc-tickets

[GHC] #9052: Support a "stable heap" which doesn't get garbage collected
by GHC 13 Nov '17

13 Nov '17

#9052: Support a "stable heap" which doesn't get garbage collected ----------------------------------------------+---------------------------- Reporter: ezyang | Owner: Type: feature request | simonmar Priority: normal | Status: new Component: Runtime System | Milestone: Keywords: | Version: 7.9 Architecture: Unknown/Multiple | Operating System: Difficulty: Moderate (less than a day) | Unknown/Multiple Blocked By: | Type of failure: Related Tickets: | None/Unknown | Test Case: | Blocking: ----------------------------------------------+---------------------------- Simon Marlow was explaining the concept on IRC. The problem we're trying to solve is you might have some very long-lived pointer data-structures, and you don't want to waste time traversing this structure on a major GC when you know that it is still going to be alive. Here's the basic idea: 1. We have a new generation on the heap called the 'stable heap' (along- side the nursery, gen0, gen1, etc). By default, no objects are automatically promoted from the old gen to the stable heap. 2. A user can request an object be moved to the stable heap, creating a stable pointer which points to the object. 3. During garbage collection, all blocks in the stable heap are always considered reachable. We don't include the stable pointers in the root set (so we don't trace the stable heap), however, write barriers are processed as usual, and we promote any objects pointed to by the stable heap to the stable heap. 4. Users manually request the stable heap to be garbage-collected, usually after deallocating stable pointers. In short, the stable heap is just another generation, but one which (1) is only collected by user request, and (2) never receives objects from automatic promotion. This project fits in very nicely with the existing GC infrastructure, so if you understand how everything is laid out this should be reasonably easy to implement. Could be a good ticket for someone who wants to get their hands dirty with the GC. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9052> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #13990: Core Lint error on empty case
by GHC 11 Nov '17

11 Nov '17

#13990: Core Lint error on empty case -------------------------------------+------------------------------------- Reporter: mbieleck | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.1-rc3 Keywords: core-lint | Operating System: Unknown/Multiple case | Architecture: | Type of failure: Compile-time Unknown/Multiple | crash or panic Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- This module: {{{#!hs {-# LANGUAGE EmptyCase #-} module Bug where data Void absurd :: Void -> a absurd v = case v of {} data Foo = Foo !Void absurdFoo :: Foo -> a absurdFoo (Foo x) = absurd x }}} Compiled using `ghc-8.2.0.20170704 -O -dcore-lint Bug.hs` Gives the following error: {{{ [1 of 1] Compiling Bug ( Bug.hs, Bug.o ) *** Core Lint errors : in result of Simplifier *** <no location info>: warning: In a case alternative: (Foo x_ap6 :: Void) No alternatives for a case scrutinee in head-normal form: x_ap6 *** Offending Program *** absurd :: forall a. Void -> a [LclIdX, Arity=1, Unf=Unf{Src=<vanilla>, TopLvl=True, Value=True, ConLike=True, WorkFree=True, Expandable=True, Guidance=ALWAYS_IF(arity=1,unsat_ok=True,boring_ok=True)}] absurd = \ (@ a_apU) (v_ap5 :: Void) -> case v_ap5 of { } absurdFoo :: forall a. Foo -> a [LclIdX, Arity=1, Unf=Unf{Src=<vanilla>, TopLvl=True, Value=True, ConLike=True, WorkFree=True, Expandable=True, Guidance=ALWAYS_IF(arity=1,unsat_ok=True,boring_ok=True)}] absurdFoo = \ (@ a_apY) (ds_dUn :: Foo) -> case ds_dUn of { Foo x_ap6 -> case x_ap6 of { } } -- irrelevant stuff omitted }}} When I manually inline `absurd` or remove the strictness annotation on `Foo`, the error goes away. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13990> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 18

[GHC] #9420: Impredicative type instantiation without -XImpredicativeTypes
by GHC 11 Nov '17

11 Nov '17

#9420: Impredicative type instantiation without -XImpredicativeTypes -------------------------------------+------------------------------------- Reporter: goldfire | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.9 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- Consider this module: {{{ {-# LANGUAGE RankNTypes #-} module Bug where rank2 :: ((forall a. a -> a) -> b) -> () rank2 _ = () foo :: () -> () foo x = x quux :: (forall a. a -> a) -> Int quux _ = 5 bar = foo . rank2 $ quux }}} The `(.)` in the definition of `bar` requires an impredicative instantiation -- that is, one of its type variables is instantiated to a forall-type. Yet, the module compiles without `-XImpredicativeTypes`. To confirm this behavior, the following is an excerpt from `-ddump-simpl`: {{{ Bug.bar = break<7>() (break<6>() GHC.Base.. @ () @ () @ ((forall a_a1Tj. a_a1Tj -> a_a1Tj) -> GHC.Types.Int) Bug.foo (Bug.rank2 @ GHC.Types.Int)) Bug.quux }}} Note the third type argument to `(.)`. Fixing this would be a breaking change that could affect users. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9420> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #9642: LANGUAGE pragma synonyms
by GHC 10 Nov '17

10 Nov '17

#9642: LANGUAGE pragma synonyms -------------------------------------+------------------------------------- Reporter: dreixel | Owner: Type: feature request | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.9 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- Here are the first 20 lines of a typical Haskell file of mine: {{{ {-# LANGUAGE EmptyDataDecls #-} {-# LANGUAGE FlexibleContexts #-} {-# LANGUAGE FlexibleInstances #-} {-# LANGUAGE LiberalTypeSynonyms #-} {-# LANGUAGE ScopedTypeVariables #-} {-# LANGUAGE RankNTypes #-} {-# LANGUAGE StandaloneDeriving #-} {-# LANGUAGE AutoDeriveTypeable #-} {-# LANGUAGE DeriveDataTypeable #-} {-# LANGUAGE DeriveGeneric #-} {-# LANGUAGE DefaultSignatures #-} {-# LANGUAGE OverlappingInstances #-} {-# LANGUAGE UndecidableInstances #-} {-# LANGUAGE MultiParamTypeClasses #-} {-# LANGUAGE TypeOperators #-} {-# LANGUAGE TypeFamilies #-} {-# LANGUAGE GADTs #-} {-# LANGUAGE DataKinds #-} {-# LANGUAGE PolyKinds #-} {-# LANGUAGE ConstraintKinds #-} }}} Over a couple of modules in a small package, I find myself copy-pasting these 20 lines, as either I will use these extensions, or they won't really cause any harm to the code I'm writing. As a way to simplify this, and reduce code duplication, I propose introducing LANGUAGE pragma synonyms: {{{ {-# LANGUAGE PedrosHaskell = EmptyDataDecls, FlexibleContexts, FlexibleInstances , LiberalTypeSynonyms, ScopedTypeVariables , RankNTypes, StandaloneDeriving, AutoDeriveTypeable , DeriveDataTypeable, DeriveGeneric, DefaultSignatures , OverlappingInstances, UndecidableInstances , MultiParamTypeClasses, TypeOperators, TypeFamilies , GADTs, DataKinds, PolyKinds, ConstraintKinds #-} }}} After writing this, in that same module or any module that imports it, I could just write: {{{ {-# LANGUAGE PedrosHaskell #-} }}} And all the pragmas above would be enabled. Besides making my life easier, I suspect this will also help increase modularity of certain packages. For example, packages could define a package-specific language pragma that should be enabled in order to use that package. Then, if the package maintainer upgrades one datatype to become a GADT, the pragma would be updated to include `GADTs`, and user code (that could now be requiring that pragma) would automatically enable `GADTs` too. Furthermore, it can also make the language standardisation process simpler. After all, {{{ {-# LANGUAGE Haskell2010 = PatternGuards, NoNPlusKPatterns, RelaxedPolyRec , EmptyDataDecls, ForeignFunctionInterface #-} }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9642> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #12623: Make Harbormaster less terrible in the face of submodule updates
by GHC 10 Nov '17

10 Nov '17

#12623: Make Harbormaster less terrible in the face of submodule updates -------------------------------------+------------------------------------- Reporter: bgamari | Owner: bgamarii Type: task | Status: new Priority: highest | Milestone: Component: Compiler | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Currently submodule updates appear to break Harbormaster builds. This is bad. Fix it. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12623> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #13122: Investigate reporting build errors with harbormaster.sendmessage
by GHC 10 Nov '17

10 Nov '17

#13122: Investigate reporting build errors with harbormaster.sendmessage -------------------------------------+------------------------------------- Reporter: mpickering | Owner: Type: task | Status: new Priority: normal | Milestone: Component: Compiler | Version: Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The phabricator API provides a method `harbormaster.sendmessage` which allows builds to report information back so that it is displayed on the interface. The goal is to avoid having to delve into the build logs in order to investigate failures. Information about the API method can be found in [https://phabricator.haskell.org/conduit/method/harbormaster.sendmessage/ conduit]. The API provides a rich set of options to report different information about a build. Some ideas: * Reporting the precise location of all terminal build errors * Reporting the outcome of all tests (this might be quite noisy but the only way to see is to try!) I don't know whether this is a good idea or how it works precisely but integrating harbourmaster with differential would be an efficiency gain. This is potentially related to #8809. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13122> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #13121: Investigate whether uploading build artifacts from harbormaster would be usful
by GHC 10 Nov '17

10 Nov '17

#13121: Investigate whether uploading build artifacts from harbormaster would be usful -------------------------------------+------------------------------------- Reporter: mpickering | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: Keywords: newcomer | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- In harbormaster it is possible to upload build artifacts and associate them with each build, for example we could upload the user manual for each build or even binaries. I am not sure what would be useful. One use case might be checking the rendering of the user guide when it is modified. The idea is to write a small utility program which uploads suitable artifacts and then add an extra build step to the relevant build plans which calls this program. The utility doesn't have to be in-tree as it can be installed in the environment. Some more information: * There are some bindings to the phabricator API in [https://github.com /haskell-infra/arc-lite arc-lite] * The relevant API methods are `file.allocate`, `file.upload` and `harbormaster.createartifact` * `file.allocate` allows you to specify a time to delete a file, this would be good to avoid consuming lots of storage for builds. * Artifacts appear in the "artifacts" tab [https://phabricator.haskell.org/harbormaster/build/18367/ here] -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13121> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #13716: Move CI to Jenkins
by GHC 10 Nov '17

10 Nov '17

#13716: Move CI to Jenkins -------------------------------------+------------------------------------- Reporter: bgamari | Owner: (none) Type: task | Status: new Priority: normal | Milestone: Component: None | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Currently we use Harbormaster to build Differentials and commits. While this works, it leaves much to be desired, * job scheduling is (literally) random: this means that some patch authors end up waiting literally ways for their patches to be built * there is little support for automatic provisioning of builders: this means we can't scale to meet demand and make poor use of our computational resources * periodic builds (e.g. nightlies) are not supported. Ideally we would not only provide * it doesn't have a sensible interface for integration with external tools: this means that efforts like CI-before-merge have been pushed off * status reporting is poor: even answering the question of what a given builder is currently working on is surprisingly difficult * maintenance is a byzantine: Adding a new builder requires adding at least six different objects (a Harbormaster Build Plan, a Drydock Working Copy, a Drydock Blueprint, an Almanac Service, an Almanac Device, and an Almanac Binding) in various Phabricator applications. None of this configuration can be tracked under version control nor can most of it be cloned from an existing builder's configuration. * Design assumptions don't match GHC's constraints: Harbormaster was designed under the assumption that builds are cheap and computation plentiful. The maintainers have stated that they have little interest in supporting environments where this doesn't hold. Jenkins, while far from perfect, seems a bit better suited to our needs, more mature, and far more flexible. This ticket will serve as the checklist for our move to Jenkins. In the end we want, * Builders (static or dynamically-provisioned, as appropriate) for * x86-64, i386 Linux * x86-64, i386 Windows * x86-64 Darwin * x86-64 OpenBSD * Cross compile from x86-64 to ARM * Native ARM * Differential builds with sensible scheduling (e.g. first build on 64-bit Linux where machines are cheap, then build on the others) * Per-commit builds on all * Nightly builds on all, including * Collection of binary distribution for user download * Update of `master` documentation mirror on downloads.haskell.org * Slow validation (including tests requiring Hackage packages) * nofib run on some platforms? * Test-before-merge-to-master -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13716> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10

[GHC] #12363: Type application for infix
by GHC 10 Nov '17

10 Nov '17

#12363: Type application for infix -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: Type: feature | Status: new request | Priority: lowest | Milestone: Component: Compiler | Version: 8.0.1 (Parser) | Keywords: | Operating System: Unknown/Multiple TypeApplications | Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- `1 + @Int 10` is a parse error, I would like it to mean the same as {{{#!hs (+) @Int 1 10 }}} Thoughts? ---- Also {{{#!hs >>> pure `foldMap` @Maybe @[_] Just 'a' "a" >>> foldMap @Maybe @[_] pure (Just 'a') "a" }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12363> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 25

[GHC] #8238: Implement unloading of shared libraries
by GHC 09 Nov '17

09 Nov '17

#8238: Implement unloading of shared libraries ------------------------------------+------------------------------------- Reporter: simonmar | Owner: Type: task | Status: new Priority: normal | Milestone: 7.10.1 Component: Runtime System | Version: 7.7 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: 8039 | ------------------------------------+------------------------------------- In #8039 we added support for unloading static objects from the runtime linker, with the GC detecting when there are no references left. We could add this functionality for shared libraries too, using `dl_iterate_phdr`. @heisenbug's comment from #8039 with the relevant pointers: [http://eli.thegreenplace.net/2011/08/25/load-time-relocation-of-shared- libraries Eli Bendersky's article] suggests to use the [http://linux.die.net/man/3/dl_iterate_phdr dl_iterate_phdr] function for finding information about loaded libraries. Seems to be linux only. There is a [http://stackoverflow.com/questions/10009043/dl-iterate-phdr- equivalent-on-mac workaround on OSX], though, on !StackOverflow. And here is how Böhm-Demers-Weiser's GC implement a [https://github.com/ivmai/bdwgc/blob/master/dyn_load.c#L451 callback function for dl_iterate_phdr]. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8238> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6