- ghc-tickets - Haskell.org

Re: [GHC] #7828: RebindableSyntax and Arrow
by GHC 03 Jun '14

03 Jun '14

#7828: RebindableSyntax and Arrow ----------------------------------------------+---------------------------- Reporter: AlessandroVermeulen | Owner: Type: bug | jstolarek Priority: normal | Status: new Component: Compiler (Type checker) | Milestone: 7.10.1 Resolution: | Version: 7.6.2 Operating System: Unknown/Multiple | Keywords: Type of failure: GHC rejects valid program | Architecture: Test Case: | Unknown/Multiple Blocking: | Difficulty: Unknown | Blocked By: | Related Tickets: ----------------------------------------------+---------------------------- Comment (by ross): Replying to [comment:15 jstolarek]: > Thanks Ross. I think I still don't understand the stk thing. What are the "anonymous inputs to the arrow"? Is that described in any of the arrow papers? They are command arguments, the arrow analogue of function arguments. But with arrows they have to be tupled up with the local environment as the input to the arrow. In the notation paper (section 3.2) these arguments are attached to the environment with a series of pairings, but before 7.8 I had to change the representation to fit with GHC's constraint-based typechecker. Now the environment and stack are separate tuples, combined in a pair. > > The simplifier reduces that to an identity using rules in Control.Arrow > > How can I observe that? Using `-ddump-simpl` only shows that this huge expression is split into many smaller functions but it doesn't look like anything gets simplified. Using `-ddump-rules` and `-ddump-rule-rewrites` I see that some rules fire but none of them seems to optimize the above expression. I used -O as well (this is with GHC 7.8.2). -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7828#comment:16> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7730: :info and polykinds
by GHC 03 Jun '14

03 Jun '14

#7730: :info and polykinds --------------------------------------------+------------------------------ Reporter: monoidal | Owner: archblob Type: bug | Status: patch Priority: normal | Milestone: 7.10.1 Component: Compiler (Type checker) | Version: 7.6.2 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: None/Unknown | Unknown/Multiple Test Case: | Difficulty: Unknown Blocking: | Blocked By: | Related Tickets: #8776 --------------------------------------------+------------------------------ Comment (by archblob): Oh, thank you. I'm glad that I could help and this is all over. I think the ticked can be marked as fixed now. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7730#comment:54> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7730: :info and polykinds
by GHC 03 Jun '14

03 Jun '14

#7730: :info and polykinds --------------------------------------------+------------------------------ Reporter: monoidal | Owner: archblob Type: bug | Status: patch Priority: normal | Milestone: 7.10.1 Component: Compiler (Type checker) | Version: 7.6.2 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: None/Unknown | Unknown/Multiple Test Case: | Difficulty: Unknown Blocking: | Blocked By: | Related Tickets: #8776 --------------------------------------------+------------------------------ Comment (by Simon Peyton Jones <simonpj@…>): In [changeset:"dd9943472d385b55f9baa56780f333a190794bcf/ghc"]: {{{ #!CommitTicketReference repository="ghc" revision="dd9943472d385b55f9baa56780f333a190794bcf" Comments only (related to Trac #7730) }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7730#comment:53> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7828: RebindableSyntax and Arrow
by GHC 03 Jun '14

03 Jun '14

#7828: RebindableSyntax and Arrow ----------------------------------------------+---------------------------- Reporter: AlessandroVermeulen | Owner: Type: bug | jstolarek Priority: normal | Status: new Component: Compiler (Type checker) | Milestone: 7.10.1 Resolution: | Version: 7.6.2 Operating System: Unknown/Multiple | Keywords: Type of failure: GHC rejects valid program | Architecture: Test Case: | Unknown/Multiple Blocking: | Difficulty: Unknown | Blocked By: | Related Tickets: ----------------------------------------------+---------------------------- Comment (by jstolarek): Thanks Ross. I think I still don't understand the stk thing. What are the "anonymous inputs to the arrow"? Is that described in any of the arrow papers? > The simplifier reduces that to an identity using rules in Control.Arrow How can I observe that? Using `-ddump-simpl` only shows that this huge expression is split into many smaller functions but it doesn't look like anything gets simplified. Using `-ddump-rules` and `-ddump-rule-rewrites` I see that some rules fire but none of them seems to optimize the above expression. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7828#comment:15> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7730: :info and polykinds
by GHC 03 Jun '14

03 Jun '14

#7730: :info and polykinds --------------------------------------------+------------------------------ Reporter: monoidal | Owner: archblob Type: bug | Status: patch Priority: normal | Milestone: 7.10.1 Component: Compiler (Type checker) | Version: 7.6.2 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: None/Unknown | Unknown/Multiple Test Case: | Difficulty: Unknown Blocking: | Blocked By: | Related Tickets: #8776 --------------------------------------------+------------------------------ Comment (by simonpj): Well that was a struggle, but I finally pushed this through. You'd done the heavy lifting but there were some dark corners I had to peer into. Thanks for making this happen Simon -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7730#comment:52> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7730: :info and polykinds
by GHC 03 Jun '14

03 Jun '14

#7730: :info and polykinds --------------------------------------------+------------------------------ Reporter: monoidal | Owner: archblob Type: bug | Status: patch Priority: normal | Milestone: 7.10.1 Component: Compiler (Type checker) | Version: 7.6.2 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: None/Unknown | Unknown/Multiple Test Case: | Difficulty: Unknown Blocking: | Blocked By: | Related Tickets: #8776 --------------------------------------------+------------------------------ Comment (by Simon Peyton Jones <simonpj@…>): In [changeset:"b4856f9f4f0fb3db473901b247d3fa94a11c25a0/ghc"]: {{{ #!CommitTicketReference repository="ghc" revision="b4856f9f4f0fb3db473901b247d3fa94a11c25a0" Do pretty-printing of TyThings via IfaceDecl (Trac #7730) All the initial work on this was done fy 'archblob' (fcsernik(a)gmail.com) thank you! I reviewed the patch, started some tidying, up and then ended up in a huge swamp of changes, not all of which I can remember now. But: * To suppress kind arguments when we have -fno-print-explicit-kinds, - IfaceTyConApp argument types are in a tagged list IfaceTcArgs * To allow overloaded types to be printed with =>, add IfaceDFunTy to IfaceType. * When printing data/type family instances for the user, I've made them print out an informative RHS, which is a new feature. Thus ghci> info T data family T a data instance T Int = T1 Int Int data instance T Bool = T2 * In implementation terms, pprIfaceDecl has just one "context" argument, of type IfaceSyn.ShowSub, which says - How to print the binders of the decl see note [Printing IfaceDecl binders] in IfaceSyn - Which sub-comoponents (eg constructors) to print * Moved FastStringEnv from RnEnv to OccName It all took a ridiculously long time to do. But it's done! }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7730#comment:51> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7828: RebindableSyntax and Arrow
by GHC 03 Jun '14

03 Jun '14

#7828: RebindableSyntax and Arrow ----------------------------------------------+---------------------------- Reporter: AlessandroVermeulen | Owner: Type: bug | jstolarek Priority: normal | Status: new Component: Compiler (Type checker) | Milestone: 7.10.1 Resolution: | Version: 7.6.2 Operating System: Unknown/Multiple | Keywords: Type of failure: GHC rejects valid program | Architecture: Test Case: | Unknown/Multiple Blocking: | Difficulty: Unknown | Blocked By: | Related Tickets: ----------------------------------------------+---------------------------- Comment (by ross): Replying to [comment:12 jstolarek]: > After spending some time on this today I have to admit that arrow desugaring is still a mystery to me. For example this code: > > {{{ > proc n -> do > x <- returnA -< n > returnA -< x > }}} > > Is desugared to: > > {{{ > arr (\n -> (n, ())) >>> > arr (\ds -> case ds of { (ds2, ds3) -> ds2 }) >>> > arr (\ds -> ((ds, ()), ())) >>> > ((arr (\ds -> case ds of { (ds2, ds3) -> ds2 }) >>> id) *** id >>> > arr (\ds -> case ds of { (x, ds2) -> case ds2 of { () -> x } }) ) >>> > arr (\ds -> (ds, ())) >>> > arr (\ds -> case ds of { (ds2, ds3) -> ds2 }) >>> > id > }}} > This can be simplified to an identity (as expected), but GHC doesn't do it. This definitely seems like an opportunity for improvement. The simplifier reduces that to an identity using rules in Control.Arrow (though more rules are probably needed to handle slightly more complex cases). -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7828#comment:14> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7828: RebindableSyntax and Arrow
by GHC 03 Jun '14

03 Jun '14

#7828: RebindableSyntax and Arrow ----------------------------------------------+---------------------------- Reporter: AlessandroVermeulen | Owner: Type: bug | jstolarek Priority: normal | Status: new Component: Compiler (Type checker) | Milestone: 7.10.1 Resolution: | Version: 7.6.2 Operating System: Unknown/Multiple | Keywords: Type of failure: GHC rejects valid program | Architecture: Test Case: | Unknown/Multiple Blocking: | Difficulty: Unknown | Blocked By: | Related Tickets: ----------------------------------------------+---------------------------- Comment (by ross): Replying to [comment:11 jstolarek]: > I have a few questions about implementation in `DsArrows` - I think these go to Ross. > > 1. What is the purpose of the stack created in the desugaring process? > 2. How do I read the notation describing each case of `dsCmd`? Eg. the `HsCmdApp` case has this comment: > > {{{ > -- D; ys |-a cmd : (t,stk) --> t' > -- D, xs |- exp :: t > -- ------------------------ > -- D; xs |-a cmd exp : stk --> t' > -- > -- ---> premap (\ ((xs),stk) -> ((ys),(e,stk))) cmd > }}} > What is `|-a`, `xs` and `ys`? What is the difference between `:` and `::`? Could you tell me how to read this particular comment? I should then be able to understand all the others. EDIT: Is there a semantic difference between `,` and `;` as in `D; ys` vs. `D, xs`? Yes, those differences in punctuation are significant. A type judgement for an expression (ignoring constraints) looks like {{{ D |- exp :: t }}} but one for a command looks like {{{ D; xs |-a cmd : stk --> t }}} Here * a is the type of arrow for this command * D is the environment of the surrounding proc expression * xs are variables defined within the proc expression, which will be inputs to the arrow * stk is a list of anonymous inputs to the arrow * t is the output type of the arrow So the translation of this command should be an expression with type {{{ D |- [[proc (xs) -> cmd]] : a ((xs),stk) t }}} In this particular rule, exp is typed using an ordinary expression judgement. It can use both external valuables (D) and those local to the proc expression (xs), so these are combined in a single environment. (Commands, in contrast, need to keep these separate.) The value of that expression is then pushed onto stk, which, together with xs, is input to cmd. (The lambda rule for commands lets you take a value from the stack and put it in the local environment, giving it a name.) The remaining wrinkle here is that cmd may not need all the variables in xs, but just the subset ys of xs, so the function inside the premap may also trim the environment from xs to ys. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7828#comment:13> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #8768: Internal compiler error on haskell-src-exts 1.14.01 (profiling)
by GHC 03 Jun '14

03 Jun '14

#8768: Internal compiler error on haskell-src-exts 1.14.01 (profiling) ----------------------------------+--------------------------------------- Reporter: MagnusTherning | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.1-rc1 Keywords: | Operating System: Linux Architecture: x86_64 (amd64) | Type of failure: Compile-time crash Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ----------------------------------+--------------------------------------- I'm configuring and building {{{haskell-src-exts}}} like this: {{{ runhaskell Setup configure -O -p --enable-shared \ -fbase4 runhaskell Setup build }}} It builds the ordinary bits just fine, but when it gets to compiling for profiling: {{{ [ 1 of 22] Compiling Language.Haskell.Exts.Annotated.Syntax ( src/Language/Haskell/Exts/Annotated/Syntax.hs, dist/build/Language/Haskell/Exts/Annotated/Syntax.p_o ) /tmp/ghc340_0/ghc340_8.s:1:0: internal compiler error: Segmentation fault .section .rodata ^ Please submit a full bug report, with preprocessed source if appropriate. }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8768> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 16

[GHC] #8475: Library docs: broken links to Foreign.ForeignPtr
by GHC 03 Jun '14

03 Jun '14

#8475: Library docs: broken links to Foreign.ForeignPtr ------------------------------------+------------------------------------- Reporter: simonmar | Owner: Type: bug | Status: new Priority: high | Milestone: 7.8.1 Component: Documentation | Version: 7.6.3 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- In the library docs at `http://www.haskell.org/ghc/docs/latest/html/libraries/index.html`, the `Foreign.ForeignPtr` module in the contents is the one from `haskell2010`, not `base`. This causes various problems, it seems. For instance: * the link from `ForeignPtr` to `Foreign.ForeignPtr` is broken * the link from `Foreign.Concurrent` to `Foreign.ForeignPtr` is broken there are probably other broken links too. We need to figure out what to do about overlapping modules properly. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8475> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11