September 2015 - ghc-tickets

[GHC] #10884: Indentation requirements for parsing seem inconsistent
by GHC 15 Sep '15

15 Sep '15

#10884: Indentation requirements for parsing seem inconsistent -------------------------------------+------------------------------------- Reporter: Samvh | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.2 (Parser) | Keywords: parsing, | Operating System: Unknown/Multiple indentation, do-notation | Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- The "golden rule of indentation" does not seem to be strictly adhered to in all cases: there is an interaction between let/where clauses and do- notation that makes indentation requirements different in different situations: Example 1, parsed successfully: {{{#!hs f = let y = do Just 1 in y }}} Example 2, causes an error: {{{#!hs f = let y = Just 1 in y }}} Example 3, causes an error: {{{#!hs f = do Just 1 }}} This behavior seems inconsistent to me. See also [http://stackoverflow.com/questions/32531508/haskell-parse-error- from-incorrect-indent]. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10884> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

Re: [GHC] #7830: Error: operand out of range
by GHC 15 Sep '15

15 Sep '15

#7830: Error: operand out of range -------------------------------------+------------------------------------- Reporter: erikd | Owner: Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Compiler | Version: 7.8.1-rc1 Resolution: | Keywords: Operating System: Linux | Architecture: powerpc Type of failure: Installing GHC | Test Case: failed | Blocked By: | Blocking: Related Tickets: | Differential Revisions: -------------------------------------+------------------------------------- Comment (by trommler): Replying to [comment:31 erikd]: > Replying to [comment:29 erikd]: > > At @rwbarton's suggestion added `NOINLINE` pragmas to `above_` and `beside_` functions in `compiler/utils/Pretty.hs` and now the assembler file for `compiler/cmm/PprC.hs` is 158390 lines. > > That worked when building the stage1 compiler using the bootstrap compiler, but it fails again when build the stage2 compiler with the stage1 compiler. The assembler file for `compiler/cmm/PprC.hs` compiled with the stage1 compiler is 220293 lines. It looks like we have to support 32-bit loads from/stores to the stack too. Fortunately, we are looking at constant offset from the C stack pointer (r1) and the offsets are known at compilation time, so performance of the object code produced by GHC is not going to suffer much if at all. I could prepare a patch next week and verify on powerpc64. @erikd could you help out building the patch on your machine? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7830#comment:34> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7830: Error: operand out of range
by GHC 15 Sep '15

15 Sep '15

#7830: Error: operand out of range -------------------------------------+------------------------------------- Reporter: erikd | Owner: Type: bug | Status: new Priority: highest | Milestone: 8.0.1 Component: Compiler | Version: 7.8.1-rc1 Resolution: | Keywords: Operating System: Linux | Architecture: powerpc Type of failure: Installing GHC | Test Case: failed | Blocked By: | Blocking: Related Tickets: | Differential Revisions: -------------------------------------+------------------------------------- Comment (by trommler): Replying to [comment:26 trommler]: > This looks like code spilling registers onto the stack. But why are the offsets so large? Could you attach the entire assembler file? This looks suspicious: {{{ .Lc138b: addi 1, 1, -28336 }}} I wonder why GHC allocates ~3500 spill slots (of size 8 bytes each) on top of the roughly 2000 that are allocated in the RTS before we enter Haskell land in `StgCRun`. > > I tried to reproduce on powerpc64 HEAD but compile failed to typecheck in package uualib. I'll try to reproduce with the commit you mentioned in comment:24. I cannot reproduce the issue on powerpc64 even though it has the same 16-bit displacement restriction. I did a `perf` build, perhaps that makes a difference?! Did you use a `build.mk` file? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7830#comment:33> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #8552: Rename StgArrWords to StgArrBytes
by GHC 15 Sep '15

15 Sep '15

#8552: Rename StgArrWords to StgArrBytes ------------------------------------+------------------------------------- Reporter: Tarrasch | Owner: Type: task | Status: new Priority: lowest | Milestone: Component: Compiler | 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: | ------------------------------------+------------------------------------- Since #3800, `StgArrWords` contains a certain number of words and not bytes. One can easily forget this and make programming mistakes. For example if you try to store the number of words in the `bytes` field, the garbage collector won't honor your decision. (I learned the hard way :)) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8552> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #10787: Incorrect code example for pattern synonyms
by GHC 15 Sep '15

15 Sep '15

#10787: Incorrect code example for pattern synonyms -------------------------------------+------------------------------------- Reporter: mpickering | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: | Version: 7.10.2 Documentation | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Documentation Unknown/Multiple | bug Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- MkT is not the name of a type in the section "Typing of pattern synonyms" https://downloads.haskell.org/~ghc/latest/docs/html/users_guide/syntax- extns.html -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10787> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #10883: Error messages referring to arrow operators use wrong symbols when outputting Unicode
by GHC 15 Sep '15

15 Sep '15

#10883: Error messages referring to arrow operators use wrong symbols when outputting Unicode -------------------------------------+------------------------------------- Reporter: zardoz | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: | Version: 7.10.2 libraries/pretty | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Other Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: 10509 Differential Revisions: | -------------------------------------+------------------------------------- As per #10509, GHC documentation gave the wrong glyphs for Unicode alternatives to the -< and >- arrow operators (the codepoints were correct, but the glyphs were not). The incorrect glyphs have also made it into the error output. So, in GHC error messages, we see something like the following: … In the command: arr (+ 2) ↢ x … Where ↢ should actually be ⤙. The alternative to >- is also incorrectly printed. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10883> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #10509: UnicodeSyntax documentation lists wrong symbols
by GHC 15 Sep '15

15 Sep '15

#10509: UnicodeSyntax documentation lists wrong symbols -------------------------------------+------------------------------------- Reporter: zardoz | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: | Version: 7.10.1 Documentation | Operating System: Unknown/Multiple Keywords: | Type of failure: None/Unknown Architecture: | Blocked By: Unknown/Multiple | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- I’ve noticed that the documentation on UnicodeSyntax lists erroneous Unicode variants for (-<) and (>-). Notably, it gives ↢ and ↣, while the lexer actually uses ⤙ and ⤚. The numeric Unicode codepoints are accurate, but glyphs displayed in the table are wrong. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10509> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9

Re: [GHC] #4012: Compilation results are not deterministic
by GHC 15 Sep '15

15 Sep '15

#4012: Compilation results are not deterministic -------------------------------------+------------------------------------- Reporter: kili | Owner: niteria Type: bug | Status: patch Priority: high | Milestone: 8.0.1 Component: Compiler | Version: 6.12.2 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: Other | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Revisions: Phab:D910, | Phab:D1073, Phab:D1133, Phab:D1192 -------------------------------------+------------------------------------- Comment (by simonmar): FYI, @niteria posted a good explanation of why the remaining problems with determinism are hard: https://mail.haskell.org/pipermail/ghc- devs/2015-September/009964.html (long, but all relevant). We'd welcome more input on this. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4012#comment:118> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #4823: Loop strength reduction for array indexing
by GHC 15 Sep '15

15 Sep '15

#4823: Loop strength reduction for array indexing -------------------------------------+------------------------------------- Reporter: tibbe | Owner: Type: feature request | Status: new Priority: low | Milestone: 8.0.1 Component: Compiler | Version: 7.1 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Runtime | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: #7116 | Differential Revisions: -------------------------------------+------------------------------------- Changes (by WrenThornton): * cc: wren@… (added) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4823#comment:8> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #4413: (^^) is not correct for Double and Float
by GHC 15 Sep '15

15 Sep '15

#4413: (^^) is not correct for Double and Float -------------------------------------+------------------------------------- Reporter: | Owner: tcsavage daniel.is.fischer | Type: bug | Status: new Priority: low | Milestone: 8.0.1 Component: Core Libraries | Version: 7.1 Resolution: | Keywords: Double, | Float, exponentiation Operating System: Unknown/Multiple | Architecture: Type of failure: Incorrect result | Unknown/Multiple at runtime | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Revisions: -------------------------------------+------------------------------------- Changes (by WrenThornton): * cc: wren@… (added) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/4413#comment:12> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0