January 2015 - ghc-tickets

[GHC] #8379: sync-all broken when using the GitHub mirror
by GHC 21 Mar '15

21 Mar '15

#8379: sync-all broken when using the GitHub mirror ------------------------------------+------------------------------------- Reporter: tibbe | Owner: Type: bug | Status: new Priority: normal | 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: | ------------------------------------+------------------------------------- Making a fresh checkout using the GitHub mirror: {{{ $ git clone git@github.com:tibbe/ghc.git $ ./sync-all get == running git config core.ignorecase true == running git clone git@github.com:tibbe/ghc-tarballs.git ghc-tarballs Cloning into 'ghc-tarballs'... ERROR: Repository not found. fatal: The remote end hung up unexpectedly git failed: 32768 at ./sync-all line 178. == Checking for old haddock repo == Checking for old binary repo == Checking for old mtl repo == Checking for old Cabal repo == Checking for old time from tarball ============================ ATTENTION! You have an old time package in your GHC tree! Please remove it (e.g. "rm -r libraries/time"), and then run "./sync-all get" to get the new repository. ============================ }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8379> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 13

[GHC] #8150: Foreign export requires redundant type signature
by GHC 20 Mar '15

20 Mar '15

#8150: Foreign export requires redundant type signature -------------------------+------------------------------------------------- Reporter: | Owner: evincarofautumn | Status: new Type: bug | Milestone: Priority: | Version: 7.6.3 normal | Operating System: Unknown/Multiple Component: | Type of failure: Incorrect warning at Compiler | compile-time Keywords: | Test Case: Architecture: | Blocking: Unknown/Multiple | Difficulty: | Unknown | Blocked By: | Related Tickets: | -------------------------+------------------------------------------------- With a source file such as: {{{ module FFITest where foreign export ccall "function" function :: IO () function :: IO () function = putStrLn "Hello, world!" }}} The second `function :: IO ()` signature is required to satisfy `-fwarn- missing-signatures`, even though a signature is already present in the `foreign export` declaration, and the types are required to be the same. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8150> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #10011: The Data instance for Ratio violates internal invariants.
by GHC 20 Mar '15

20 Mar '15

#10011: The Data instance for Ratio violates internal invariants. -------------------------------------+------------------------------------- Reporter: ekmett | Owner: ekmett Type: bug | Status: new Priority: normal | Milestone: Component: Core | Version: 7.10.1-rc1 Libraries | Operating System: Unknown/Multiple Keywords: | Type of failure: None/Unknown Architecture: | Blocked By: Unknown/Multiple | Related Tickets: Test Case: | Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- I found this when Simon was cleaning up unused dependencies in {{{ https://phabricator.haskell.org/rGHCc409b6f30373535b6eed92e55d4695688d32be9… }}} The Data instance for Ratio just uses the raw `(:%)` constructor and doesn't check that the result is reduced to normal form. It strikes me that the fix is to add back the Integral constraint on the Data instance and to use `(%)` rather than `(:%)` in the `gfoldl` and `gunfold` code. This restores the invariant and matches the behavior of "virtual constructors" we've used to patch up such problems elsewhere. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10011> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 17

[GHC] #9572: nofib target for just building should be part of validate
by GHC 20 Mar '15

20 Mar '15

#9572: nofib target for just building should be part of validate -------------------------------------+------------------------------------- Reporter: ezyang | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: NoFib benchmark | Version: 7.8.2 suite | Operating System: Keywords: | Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: Difficulty: Unknown | None/Unknown Blocked By: | Test Case: Related Tickets: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- nofib is never run during validates, which makes it easy to bitrot (c.f. AMP changes). It would be better if: 1. We had a build target for just building nofib tests, not running them, (maybe optionally with optimizations turned off, or just `-fno-code`) and 2. This build target was executed during validates This should strike a balance between wanting to avoid having to spend a lot of time running the benchmarks on validate, while still exercising the nofib tests to some degree. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9572> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #9609: GHCi gives overly specific error message for unknown constructor
by GHC 20 Mar '15

20 Mar '15

#9609: GHCi gives overly specific error message for unknown constructor -------------------------------------+------------------------------------- Reporter: dfeuer | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.2 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- What happens: {{{ Prelude> :i Fish Top level: Not in scope: data constructor ‘Fish’ }}} This is potentially confusing because `:i` can be used for type constructors and classes, as well as data constructors. What I think should happen: {{{ Prelude> :i Fish Top level: Not in scope: constructor or class ‘Fish’ }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9609> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #9956: Command line flag deprecated warning could be annoying for -Werror users
by GHC 20 Mar '15

20 Mar '15

#9956: Command line flag deprecated warning could be annoying for -Werror users -------------------------------------+------------------------------------- Reporter: ezyang | Owner: ezyang Type: bug | Status: new Priority: high | Milestone: 7.10.1 Component: Compiler | Version: 7.11 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Blocked By: Test Case: | Related Tickets: Blocking: | Differential Revisions: | -------------------------------------+------------------------------------- In 7.10 we deprecated the poorly named `-package-name` for `-this-package- key`. However, I was working with GHC's validate script and I noticed that this behavior actually might be pretty annoying, because it causes GHC to fail on `-Werror`, and there doesn't seem to be any way of turning off this behavior. So maybe we should add a -fno-warn flag which controls flag warnings. But old GHCs wouldn't accept that flag, so it doesn't make it any more convenient for people to keep their old build scripts running while squelching the warning. Or we could drop the deprecation warning for now and start bleating about it next release. Setting priority high because whatever we decide to do, we need to decide it before 7.10 release. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9956> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #9392: "\n" is displayed weirdly in error messages
by GHC 20 Mar '15

20 Mar '15

#9392: "\n" is displayed weirdly in error messages -------------------------------------+------------------------------------- Reporter: Artyom.Kazak | Owner: Type: bug | Status: new Priority: low | Milestone: Component: Compiler | Version: 7.8.3 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Other Blocked By: | Test Case: Related Tickets: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- Observe: {{{ Prelude> "a\nb" + () <interactive>:2:10: Couldn't match expected type ‘[Char]’ with actual type ‘()’ In the second argument of ‘(+)’, namely ‘()’ In the expression: "a\n\ \b" + () }}} Moreover, if “\n” is at the end of the string, it simply gets eaten: {{{ Prelude> "a\n" + () <interactive>:3:9: Couldn't match expected type ‘[Char]’ with actual type ‘()’ In the second argument of ‘(+)’, namely ‘()’ In the expression: "a" + () }}} It happens in GHC as well as GHCi. Tested on 7.8.3. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9392> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #9873: Use CONF_GCC_LINKER_OPTS_STAGE2 only in stage2
by GHC 19 Mar '15

19 Mar '15

#9873: Use CONF_GCC_LINKER_OPTS_STAGE2 only in stage2 -------------------------------------+------------------------------------- Reporter: nomeata | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Build System | Version: 7.9 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- On ARM, we want to make sure that GHC uses the gold linker. In order to achieve that, we need to get `-fuse-ld=gold` into SettingsCCompilerLinkFlags in the settings. This field is filled with only CONF_GCC_LINKER_OPTS_STAGE2. So we want that flag to show up there. But this variable is used in other places as well. For example, the configuration of tinfo fails: {{{ $ make libraries/terminfo/dist-boot/package-data.mk ===--- building phase 0 make -r --no-print-directory -f ghc.mk phase=0 phase_0_builds libraries/terminfo/ghc.mk:3: libraries/terminfo/dist-boot/package-data.mk: Datei oder Verzeichnis nicht gefunden "inplace/bin/ghc-cabal" configure libraries/terminfo dist-boot "" --with- ghc="/usr/bin/ghc" --with-ghc-pkg="/usr/bin/ghc-pkg" --package- db=/home/jojo/build/haskell/ghc/libraries/bootstrapping.conf --disable- library-for-ghci --enable-library-vanilla --disable-library-profiling --disable-shared --with-hscolour="/usr/bin/HsColour" --configure- option=CFLAGS=" -fno-stack-protector " --configure-option=LDFLAGS=" -fuse-ld=gold " --configure-option=CPPFLAGS=" " --gcc-options=" -fno- stack-protector -fuse-ld=gold " --constraint "binary == 0.7.1.0" --constraint "Cabal == 1.21.1.0" --constraint "hpc == 0.6.0.2" --constraint "bin-package-db == 0.0.0.0" --constraint "hoopl == 3.10.0.2" --constraint "transformers == 0.4.2.0" --constraint "terminfo == 0.4.0.0" --with-gcc="/usr/bin/gcc" --configure-option=--with- cc="/usr/bin/gcc" --with-ar="/usr/bin/ar" --with-alex="/usr/bin/alex" --with-happy="/usr/bin/happy" Configuring terminfo-0.4.0.0... configure: WARNING: unrecognized options: --with-compiler, --with-gcc checking for gcc... /usr/bin/gcc checking whether the C compiler works... yes checking for C compiler default output file name... a.out checking for suffix of executables... checking whether we are cross compiling... no checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether /usr/bin/gcc accepts -g... yes checking for /usr/bin/gcc option to accept ISO C89... none needed checking how to run the C preprocessor... /usr/bin/gcc -E checking for grep that handles long lines and -e... /bin/grep checking for egrep... /bin/grep -E checking for ANSI C header files... yes checking for sys/types.h... yes checking for sys/stat.h... yes checking for stdlib.h... yes checking for string.h... yes checking for memory.h... yes checking for strings.h... yes checking for inttypes.h... yes checking for stdint.h... yes checking for unistd.h... yes checking ncurses.h usability... yes checking ncurses.h presence... yes checking for ncurses.h... yes checking for setupterm in -ltinfo... yes configure: creating ./config.status config.status: creating terminfo.buildinfo configure: WARNING: unrecognized options: --with-compiler, --with-gcc ghc-cabal: Missing dependency on a foreign library: * Missing C library: tinfo This problem can usually be solved by installing the system package that provides this library (you may need the "-dev" version). If the library is already installed but in a non-standard location then you can use the flags --extra-include-dirs= and --extra-lib-dirs= to specify where it is. libraries/terminfo/ghc.mk:3: recipe for target 'libraries/terminfo/dist- boot/package-data.mk' failed make[1]: *** [libraries/terminfo/dist-boot/package-data.mk] Error 1 Makefile:71: recipe for target 'libraries/terminfo/dist-boot/package- data.mk' failed make: *** [libraries/terminfo/dist-boot/package-data.mk] Error 2 }}} The problem (unfortunately, not very visible) is that `ghc-cabal` calls `gcc` to test if the C libaries are available. It uses the linker flags that it’s being told to use from the host GHC (here: 7.6.3), which contains the usual: {{{ $ ghc --info |fgrep 'C compiler flags' ,("C compiler flags"," -fno-stack-protector -Wl,--hash-size=31 -Wl ,--reduce-memory-overheads") }}} But to that it adds `-fuse-ld-gold` because `ghc-cabal` is gets `" --gcc- options=" -fno-stack-protector -fuse-ld=gold "` passed. So the build system should not pass flags from `CONF_GCC_LINKER_OPTS_STAGE2` anywhere besides gcc when used *by the stage 2 ghc* during linking. The chain of variables leading to this can be seen here: {{{ $ git grep CONFIGURE_LDFLAGS rules/build-package-data.mk:$1_$2_CONFIGURE_LDFLAGS = $$(SRC_LD_OPTS) $$(CONF_GCC_LINKER_OPTS_STAGE$3) $$($1_LD_OPTS) $$($1_$2_LD_OPTS) rules/build-package-data.mk:$1_$2_CONFIGURE_OPTS += --configure- option=LDFLAGS="$$($1_$2_CONFIGURE_LDFLAGS)" rules/build-package-data.mk:$1_$2_CONFIGURE_OPTS += --gcc- options="$$($1_$2_CONFIGURE_CFLAGS) $$($1_$2_CONFIGURE_LDFLAGS)" }}} I’m currently experimenting with a patch that simply does not use `CONF_GCC_LINKER_OPTS_STAGE` anywhere besides when defining `SettingsCCompilerLinkFlags`, but I’m afraid that this might introduce new unwanted behaviour. But since `ghc-cabal` reads gcc-flags from `ghc --info`, maybe that channel of information is sufficient. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9873> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #9921: Building Haddocks with Hoogle output results in an error
by GHC 18 Mar '15

18 Mar '15

#9921: Building Haddocks with Hoogle output results in an error -------------------------------------+------------------------------------- Reporter: snoyberg | Owner: Type: bug | Status: new Priority: normal | Milestone: 7.10.1 Component: Compiler | Version: 7.9 Keywords: | Operating System: Architecture: x86_64 (amd64) | Unknown/Multiple Difficulty: Unknown | Type of failure: Runtime Blocked By: | crash Related Tickets: | Test Case: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- With the GHC 7.10 release candidate 1 and cabal-install 1.22 (via Herbert's PPA), I unpacked BoundedChan, configured, and then ran the following, with the given output: {{{ $ cabal haddock --hoogle Running Haddock for BoundedChan-1.0.3.0... Preprocessing library BoundedChan-1.0.3.0... Haddock coverage: 100% ( 10 / 10) in 'Control.Concurrent.BoundedChan' haddock: internal error: expectJust getPackageDetails }}} I can reopen against the Haddock repo instead, but I thought this should be on the 7.10 milestone. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9921> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #8315: Improve specialized Hoopl module
by GHC 18 Mar '15

18 Mar '15

#8315: Improve specialized Hoopl module ------------------------------------+------------------------------------- Reporter: jstolarek | Owner: Type: task | Status: new Priority: low | 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: | ------------------------------------+------------------------------------- GHC uses Hoopl to perform some of the Cmm-to-Cmm optimization passes, but instead of relying directly on Hoopl library GHC uses its own specialized module (see [[GhcFile(compiler/cmm/Hoopl/Dataflow.hs)]]). Interface of that module, as well as its internal implementation can be improved. Right now interface places responsibility of maintaining some of the invariants on user. It also ignores differences between forward and backward analysis. [http://ghc.haskell.org/trac/ghc/wiki/Hoopl/Cleanup This wiki page] gives full detail about proposed design changes. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8315> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1