August 2014 - ghc-tickets

[GHC] #9515: Deprecate -XExplicitForAll
by GHC 28 Sep '14

28 Sep '14

#9515: Deprecate -XExplicitForAll -------------------------------------+------------------------------------- Reporter: dfeuer | Owner: Type: feature request | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.3 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- The problem is that it uses the same syntax as -XScopedTypeVariables to mean something a bit different, and incompatible. -XExplicitForAll is of rather limited utility, whereas -XScopedTypeVariables is a wonderful thing. Confusion is bad. {{{#!hs f :: forall a . a -> [a] f a = let g :: a -> a g x = x in g [a] }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9515> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 16

[GHC] #9326: Minor change to list comprehension structure leads to poor performance
by GHC 28 Sep '14

28 Sep '14

#9326: Minor change to list comprehension structure leads to poor performance -------------------------------------+------------------------------------- Reporter: dfeuer | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.3 Keywords: | Differential Revisions: Operating System: Unknown/Multiple | Architecture: Type of failure: Runtime | Unknown/Multiple performance bug | Difficulty: Unknown Test Case: | Blocked By: Blocking: | Related Tickets: -------------------------------------+------------------------------------- According to [http://stackoverflow.com/questions/24690406/haskell-list- comprehension-speed-inconsistencies] ''and my own testing'', the code in compspeed.hs reliably runs several times faster than the code in compspeedslow.hs with ghc 7.8.3, although this apparently does not happen with 7.6.3. The basic distinction is between a list comprehension of the form {{{ [(a,h) | ..., let h = expr] }}} and {{{ [(a,expr) | ...] }}} The core produced by -ddump-simpl is completely different in each case, so I can't figure out what's going on. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9326> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #8593: vector is being built even with `--no-dph`
by GHC 26 Sep '14

26 Sep '14

#8593: vector is being built even with `--no-dph` ------------------------------------+------------------------------------- Reporter: nomeata | Owner: Type: feature request | Status: new Priority: low | Milestone: Component: Build 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: | ------------------------------------+------------------------------------- I usually use `./sync-all --no-dph get` to speed up things a bit. But packages `vector`, `primitive` and `random` are still being pulled (by virtue of being submodules), and then built. It would be nice if the build system would build them only if `libraries/dph` is available, or have another way of making them optional. (Potentially obsolete once #8545 is done.) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8593> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #8529: Sequent reading from socket/connection on Windows fails
by GHC 24 Sep '14

24 Sep '14

#8529: Sequent reading from socket/connection on Windows fails -------------------------------------+------------------------------------- Reporter: Heimdell | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: | Version: 7.4.2 libraries/haskell2010 | Operating System: Windows Keywords: sockets, windows, | Type of failure: Incorrect result read failure | at runtime Architecture: Unknown/Multiple | Test Case: Difficulty: Unknown | Blocking: Blocked By: | Related Tickets: | -------------------------------------+------------------------------------- I wrote a simple TCP-client, which consumes and produces 8-byte packets. When I run it on linux, it works perfectly, being part of the loop (write- read-write-read-...). On windows it receives only first msg from the other side (write-read- write-write-...). The packets are still going, although. Before that, I used the sockets directly; changing to HandleStream didn't help. The code is: {{{ transmit :: Int -> HandleStream ByteString -> ByteString -> IO [Bytestring] transmit delay sock packet = do let input = timeout delay $ sock `readBlock` 8 <* putStrLn "\nData was read!" sock `writeBlock` pack strings <- whileJust input return [str | Right str <- strings] whileJust action = do result <- action case result of Just a -> (:) <$> return a <*> whileJust action Nothing -> return [] }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8529> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #9338: Some ghc options do not seem to be documented:
by GHC 24 Sep '14

24 Sep '14

#9338: Some ghc options do not seem to be documented: -------------------------------------+------------------------------------- Reporter: jrp | Owner: Type: task | Status: new Priority: normal | Milestone: Component: Documentation | Version: 7.8.3 Keywords: | Differential Revisions: Operating System: Unknown/Multiple | Architecture: Type of failure: Documentation | Unknown/Multiple bug | Difficulty: Unknown Test Case: | Blocked By: Blocking: | Related Tickets: -------------------------------------+------------------------------------- ghc --show-options comes up with options that I could not find in the manual. For example -parallel, -smp, -mavx. Would it be possible to * come up with an alphabetical list in the manual? * have an option that picked the best combination of options for the cpu on which you are running? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9338> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #8480: ghc 7.6.3 + clang-compatibility patches
by GHC 24 Sep '14

24 Sep '14

#8480: ghc 7.6.3 + clang-compatibility patches ----------------------------------------------+---------------------------- Reporter: darinmorrison | Owner: Type: task | Status: new Priority: high | Milestone: Component: Compiler | Version: 7.6.3 Keywords: | Operating System: MacOS X Architecture: Unknown/Multiple | Type of failure: Difficulty: Moderate (less than a day) | None/Unknown Blocked By: | Test Case: Related Tickets: | Blocking: ----------------------------------------------+---------------------------- I did a first pass at back-porting Austin's clang-compatibility patches from HEAD to 7.6.3 in anticipating of a 7.6.4 release, or at least some sort of update for OS X 10.9 folks. The back-ported commits are on branches in my GHC GitHub repo: https://github.com/darinmorrison/ghc ghc-7.6.3-clang (just clang stuff) ghc-7.6.3-clang+mavericks (clang stuff + 10.9 fixes) I haven't tested this fully yet, or even compiled it yet with the latest changes, but I think I caught most of the stuff that might cause problems. (Actually I just tried now and there's another error further along I still need to track down.) There are a few other rough edges still: warnings, etc. Also, if you want to use the validate script, you'll need to turn off Werror. Anyway, I'll try to get remaining stuff sorted out and get some binaries up in the next couple of days. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8480> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #9366: Trac stylesheet minor cosmetic issue
by GHC 24 Sep '14

24 Sep '14

#9366: Trac stylesheet minor cosmetic issue -------------------------------------+------------------------------------- Reporter: cheater | Owner: hvr Type: bug | Status: new Priority: lowest | Milestone: Component: Trac & Git | Version: 7.8.2 Keywords: css, stylesheet, | Operating System: navbar | Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: Other Difficulty: Easy (less than 1 | Test Case: hour) | Blocking: Blocked By: | Differential Revisions: Related Tickets: | -------------------------------------+------------------------------------- Hi, the styling of the navigation bar is somehow weird. It's not a big issue, I just thought it's worth keeping track of. Please see attachment. Tested on Firefox 31 and Chromium 34 with the same effect. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9366> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #8251: Validate submodule references during pre-receive hook
by GHC 24 Sep '14

24 Sep '14

#8251: Validate submodule references during pre-receive hook ------------------------------------+------------------------------------- Reporter: hvr | Owner: hvr Type: task | Status: new Priority: normal | Milestone: Component: None | Version: 7.6.3 Keywords: admin git | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- Right now, you can push a commit introducing a not-yet existing Git submodule commit reference. However, such commits can be rejected for submodule repositories hosted at git.haskell.org, thus forcing the user to push the submodule first, and only after that push the submodule-affecting commit in ghc.git This kind of sanity checking ghc.git commits will become more important, if more repositories are converted into Git submodules. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8251> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #8983: sync-all get should respect branches
by GHC 24 Sep '14

24 Sep '14

#8983: sync-all get should respect branches ------------------------------------+------------------------------------- Reporter: ezyang | Owner: Type: bug | Status: new Priority: low | Milestone: Component: Build System | Version: 7.9 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- If I have ghc-7.8 checked out and I do a './sync-all get', I might expect the repositories that are retrieved would also be set for ghc-7.8. However, this does not currently appear to be the case. This may be a little trickier, since if you're on a topic branch but you are mostly based off of master, then 'sync-all get' will fail, since the topic branch in question won't exist. So we should probably fall back in some way, rather than fail if the specified branch is not available. At the very least, we should print a warning if we sync-all get, the main branch is not master but we checked out a copy of master in the subrepo. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8983> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #8369: Small improvements to ./sync-all
by GHC 24 Sep '14

24 Sep '14

#8369: Small improvements to ./sync-all ------------------------------------+------------------------------------- Reporter: nwf | Owner: Type: bug | Status: new Priority: lowest | Milestone: Component: Compiler | 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: | ------------------------------------+------------------------------------- ./sync-all get in a local clone does not use local submodules. That is, if I {{{ git clone ghc ghc-tmp; cd ghc-tmp; ./sync-all get }}} the contained clones come over just fine from ghc/... into ghc-tmp/..., but the submodules are fetched over the network connection. I think the attached patch brings the "fetch from the clone source if possible" behavior over to submodules, though possibly not in an ideal way. Separately, I find that I want to run shell commands in each repository; adding a "shell" verb is trivial and a second patch is attached. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8369> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6