April 2016 - ghc-tickets

[GHC] #11383: CAFs lose sharing due to implicit call stacks
by GHC 05 Apr '16

05 Apr '16

#11383: CAFs lose sharing due to implicit call stacks -------------------------------------+------------------------------------- Reporter: simonmar | Owner: gridaphobe Type: bug | Status: new Priority: normal | Milestone: 8.0.1 Component: Compiler | Version: 8.0.1-rc1 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 implicit call stack machinery adds a constraint to CAFs, which loses sharing in some cases. The regression is fixed by -O (actually `-ffull- laziness`), but it is surprising nonetheless, and might cause problems for people using GHCi or other places where -O is turned off. For example: {{{ {-# LANGUAGE NoMonomorphismRestriction #-} module Main where import System.Environment fib :: Integer -> Integer fib n = if n < 2 then 1 else fib (n-1) + fib (n-2) x = if fib 3 > 20 then error "x" else fib 30 main = do [n] <- getArgs case n of "a" -> print (x + x) "b" -> print x }}} Try it as follows (requires 8.0+): {{{ $ ghc imp.hs -fforce-recomp -O -fno-full-laziness $ ./imp a +RTS -t 2692538 <<ghc: 430859568 bytes, 822 GCs, 36580/44384 avg/max bytes residency (2 samples), 1M in use, 0.000 INIT (0.000 elapsed), 0.156 MUT (0.203 elapsed), 0.018 GC (0.009 elapsed) :ghc>> $ ./imp b +RTS -t 1346269 <<ghc: 215456192 bytes, 411 GCs, 36580/44384 avg/max bytes residency (2 samples), 1M in use, 0.000 INIT (0.000 elapsed), 0.073 MUT (0.119 elapsed), 0.008 GC (0.004 elapsed) :ghc>> }}} With GHC 7.10 and earlier both commands perform the same. Note that this only uses `error`, and doesn't require `ImplicitParams`. It does require `NoMonomorphismRestriction`, however. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11383> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9

Re: [GHC] #3711: Bad error reporting when calling a function in a module which depends on a DLL on Windows
by GHC 04 Apr '16

04 Apr '16

#3711: Bad error reporting when calling a function in a module which depends on a DLL on Windows -------------------------------------+------------------------------------- Reporter: fasta | Owner: Type: bug | Status: new Priority: lowest | Milestone: Component: Runtime System | Version: 6.10.4 Resolution: | Keywords: Operating System: Windows | Architecture: Type of failure: Incorrect | Unknown/Multiple warning at compile-time | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by Phyx-): * cc: simonmar (added) Comment: At the very least we do call `GetLastError()` now. {{{ sysErrorBelch("addDLL: %" PATH_FMT " (Win32 error %lu)", dll_name, GetLastError()); }}} Also I don't think we can get more information aside from filling in the format specifiers in the error we get back from Windows. Aside from that, such errors are quite easy to debug from the outside. Just enable [https://msdn.microsoft.com/en- us/library/windows/hardware/ff556886(v=vs.85).aspx loader snaps] and attach a debugger (e.g. gdb or windbg) and you'll see why the load failed. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/3711#comment:17> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #11231: GHC's configure script does not honour CC env var
by GHC 04 Apr '16

04 Apr '16

#11231: GHC's configure script does not honour CC env var -------------------------------------+------------------------------------- Reporter: hvr | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Build System | Version: 7.11 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: -------------------------------------+------------------------------------- `./configure --help` tells us that: {{{ ... Some influential environment variables: CC C compiler command CFLAGS C compiler flags LDFLAGS linker flags, e.g. -L<lib dir> if you have libraries in a nonstandard directory <lib dir> LIBS libraries to pass to the linker, e.g. -l<library> CPPFLAGS (Objective) C/C++ preprocessor flags, e.g. -I<include dir> if you have headers in a nonstandard directory <include dir> CPP C preprocessor Use these variables to override the choices made by `configure' or to help it to find libraries and programs with nonstandard names/locations. Report bugs to <glasgow-haskell-bugs(a)haskell.org>. }}} However, neither - `CC=clang-3.8 ./configure`, nor - `./configure CC=clang-3.8` has any effect. `./configure` keeps using the `/usr/bin/gcc` in `$PATH`. The only invocation that has the desired effect is - `./configure --with-gcc=clang-3.8` However, this is utterly surprising and confusing to experienced developers who are used to Autoconf `./configure` scripts adhering to the conventions regarding `CC`, and contradicts the `--help` message. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11231> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #10970: Built in MIN_VERSION macro support
by GHC 04 Apr '16

04 Apr '16

#10970: Built in MIN_VERSION macro support -------------------------------------+------------------------------------- Reporter: ezyang | Owner: Type: feature | Status: new request | Priority: normal | Milestone: Component: Compiler | Version: 7.11 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: -------------------------------------+------------------------------------- Pursuant to https://mail.haskell.org/pipermail/ghc- devs/2015-September/010015.html Recapped here: when preprocessing Haskell source code, we generate a temporary stub header file which defines a number of MIN_VERSION macros, ala what Cabal generates today. Specifically, for every explicitly specified package dependency (via a `-package` or `-package-id` argument), we generate a `MIN_VERSION_pkgname` macro which tests for the version recorded for the package database. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10970> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 17

[GHC] #11751: Parse error with case in where
by GHC 03 Apr '16

03 Apr '16

#11751: Parse error with case in where ----------------------------------------+--------------------------------- Reporter: samuela | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.2 Keywords: | Operating System: MacOS X Architecture: Unknown/Multiple | Type of failure: None/Unknown Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: ----------------------------------------+--------------------------------- The following code produces a peculiar parse error: {{{#!hs subtype = 0 where (x, y) = case (1, 2) of (a, b) -> (3, 4) }}} `parse_error.hs:3:5: parse error on input ‘(’` -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11751> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #11763: Implement `-f(no-)version-macros` flag for controlling version macro generation
by GHC 01 Apr '16

01 Apr '16

#11763: Implement `-f(no-)version-macros` flag for controlling version macro generation -------------------------------------+------------------------------------- Reporter: hvr | Owner: Type: feature | Status: new request | Priority: highest | Milestone: 8.0.1 Component: Compiler | Version: 7.11 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: #10970 Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Since a9c93bdd8b027d6de09a3eada7721e7fd2d3e050 / #10970 GHC is able to generate `MIN_VERSION_...` and `VERSION_...` natively without Cabal's help. Currently, this feature is currently triggered by `-hide-all-packages`. However, additionally, it seems reasonable to be able to control this more explicitly. E.g. either to be able to explicitly disable even if `-hide- all-packages` implicitly enabled the macros, as well as be able to explicitly request macro generation for currently visible packages even without any `-hide-all-package`/`-package` flags. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11763> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7