Re: Build failure under Fedora 21

2015-01-11 Thread Dominick Samperi
Thank you Edward. I managed to build from HEAD under Fedora 21 (after 'cabal install containers'). The errors I reported may be related to the way I tried to clone a particular branch: git clone -b ghc-7.8 git://github.com/ghc/ghc.git It is not clear how to checkout/build/package a particular v

RE: Clarification of HsBang and isBanged

2015-01-11 Thread Simon Peyton Jones
Shouldn't the second case look at whether -funbox-strict-fields or -funbox-small-strict-fields is set and use unpackedName instead of isStrictName if so? What is repBangTy for? No, we are generating code that, when run, will generate the TH data structure for a data type declaration. That is, s

RE: Clarification of HsBang and isBanged

2015-01-11 Thread Simon Peyton Jones
No it shouldn’t. This is TH so we are trying to reify source code. If we have what the user wrote (a HsSrcBang) we just follow it. If we don’t (i.e. HsStrict/HsUnpack) then we just have to do the best we can Simon From: Johan Tibell [mailto:johan.tib...@gmail.com] Sent: 11 January 2015 19:12

RE: Clarification of HsBang and isBanged

2015-01-11 Thread Simon Peyton Jones
Correct. And when dealing with an imported DataCon, it must slavishly follow the decisions taken in the defining module S From: Johan Tibell [mailto:johan.tib...@gmail.com] Sent: 11 January 2015 22:27 To: Simon Peyton Jones Cc: ghc-devs@haskell.org Subject: Re: Clarification of HsBang and isBan

RE: Clarification of HsBang and isBanged

2015-01-11 Thread Simon Peyton Jones
OK, so one thing I failed to explain in the comment is this: for imported DataCons, the dcSrcBangs field is precisely the [HsImplBang] decisions computed when compiling the defining module. So if the defining module was compiled with –O –funbox-strict-fields, GHC will make one set of choices.

Re: Clarification of HsBang and isBanged

2015-01-11 Thread Johan Tibell
Yet more questions. I think I'm on the wrong track. I was trying to change MkId.dataConArgRep in order to make user-defined fields get the right strictness. However, some debug tracing suggests that this function isn't used (or isn't only used) to compute the strictness and "unpackedness" of a dat

Re: Clarification of HsBang and isBanged

2015-01-11 Thread Johan Tibell
Yet another one. TcSplice.reifyStrict doesn't take the unboxing flags into account either. Should it? reifyStrict :: DataCon.HsSrcBang -> TH.Strict reifyStrict HsNoBang = TH.NotStrict reifyStrict (HsSrcBang _ False) = TH.NotStrict reifyStrict (HsSrcBang

Re: Clarification of HsBang and isBanged

2015-01-11 Thread Johan Tibell
Those comments and the renaming really help. Here are a couple of more questions I got after exploring some more: DsMeta.repBangTy look wrong to me: repBangTy :: LBangType Name -> DsM (Core (TH.StrictTypeQ)) repBangTy ty= do MkC s <- rep2 str [] MkC t <- repLTy ty' rep2

Re: seq#: do we actually need it as a primitive?

2015-01-11 Thread Kim-Ee Yeoh
On Sun, Jan 11, 2015 at 9:25 PM, Roman Cheplyaka wrote: > I made an attempt at a better documentation for evaluate. > See here: https://phabricator.haskell.org/D615 > Wunderbar. I especially liked the prescription at the end on when to use evaluate and when to prefer (return $!). -- Kim-Ee ___

Re: seq#: do we actually need it as a primitive?

2015-01-11 Thread Roman Cheplyaka
I made an attempt at a better documentation for evaluate. See here: https://phabricator.haskell.org/D615 ___ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

Re: Build failure under Fedora 21

2015-01-11 Thread Roman Cheplyaka
Do you know you can simply install the ghc binary distribution without having to compile anything? https://www.haskell.org/ghc/download_ghc_7_8_4 On 11/01/15 08:07, Dominick Samperi wrote: > Hello, > > I'm trying to build ghc-7.8 branch under Fedora 21 and I get the > failure diagnostics appended