March 2013 - ghc-tickets

[GHC] #7731: Add "source" button for instances in Hackage
by GHC 24 Mar '13

24 Mar '13

#7731: Add "source" button for instances in Hackage -----------------------------+---------------------------------------------- Reporter: polkovnikov.ph | Owner: Type: feature request | Status: new Priority: normal | Component: Documentation Version: | Keywords: hackage, instance Os: Unknown/Multiple | Architecture: Unknown/Multiple Failure: None/Unknown | Blockedby: Blocking: | Related: -----------------------------+---------------------------------------------- There's no "Source" button to the right of the TC instance documentation in Hackage. I can see there're some structural problems with source navigation in Hackage (i.e. wrong files shown when dealing with preprocessor). I suppose it happens, because the only file it looks into is the one currently opened. This may be a real problem due to the scoping of instances. I'm rather unsure whether I should file an another ticket for this. -- Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/7731> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #7749: hsc2hs generated executable does not reflect additional library directories
by GHC 24 Mar '13

24 Mar '13

#7749: hsc2hs generated executable does not reflect additional library directories ---------------------+------------------------------------------------------ Reporter: ijmorlan | Owner: Type: bug | Status: new Priority: normal | Component: hsc2hs Version: 7.4.2 | Keywords: Os: MacOS X | Architecture: Unknown/Multiple Failure: Other | Blockedby: Blocking: | Related: ---------------------+------------------------------------------------------ This came up attempting to "cabal install postgresql-libpq", with the following result: {{{ 14:36 ijmorlan(a)27imac.local$ cabal install postgresql-libpq Resolving dependencies... [1 of 1] Compiling Main ( /var/folders/Lh/LhVs9ls4Hme1jR+t1DJbm++++TQ/-Tmp-/postgresql- libpq-0.8.2.2-60148/postgresql-libpq-0.8.2.2/Setup.hs, /var/folders/Lh/LhVs9ls4Hme1jR+t1DJbm++++TQ/-Tmp-/postgresql- libpq-0.8.2.2-60148/postgresql-libpq-0.8.2.2/dist/setup/Main.o ) Linking /var/folders/Lh/LhVs9ls4Hme1jR+t1DJbm++++TQ/-Tmp-/postgresql- libpq-0.8.2.2-60148/postgresql-libpq-0.8.2.2/dist/setup/setup ... Configuring postgresql-libpq-0.8.2.2... Building postgresql-libpq-0.8.2.2... Preprocessing library postgresql-libpq-0.8.2.2... dyld: Library not loaded: libpq.5.dylib Referenced from: /private/var/folders/Lh/LhVs9ls4Hme1jR+t1DJbm++++TQ/-Tmp-/postgresql- libpq-0.8.2.2-60148/postgresql- libpq-0.8.2.2/dist/build/Database/PostgreSQL/LibPQ_hsc_make Reason: image not found running dist/build/Database/PostgreSQL/LibPQ_hsc_make failed (exit code 5) command was: dist/build/Database/PostgreSQL/LibPQ_hsc_make >dist/build/Database/PostgreSQL/LibPQ.hs cabal: Error: some packages failed to install: postgresql-libpq-0.8.2.2 failed during the building phase. The exception was: ExitFailure 1 14:36 ijmorlan(a)27imac.local$ }}} Some background: My PostgreSQL installation was made using the one-click installer available from PostgreSQL:http://www.postgresql.org/download/macosx/. Libraries are in /Library/PostgreSQL/9.2/lib. The postgresql-libpq Setup.hs uses pg_config to pick up the proper include and library directories, and these are passed to hsc2hs with a "-- lflag=-L/Library/PostgreSQL/9.2/lib" option. hsc2hs successfully builds its temporary intermediate executable. However, running this executable gives the dyld error seen above. I managed to use install_name_tool to fix the intermediate executable, as follows: {{{ 14:16 ijmorlan(a)27imac.local$ install_name_tool -add_rpath /Library/PostgreSQL/9.2/lib dist/build/Database/PostgreSQL/LibPQ_hsc_make 14:17 ijmorlan(a)27imac.local$ install_name_tool -change libiconv.2.dylib @rpath/libiconv.2.dylib dist/build/Database/PostgreSQL/LibPQ_hsc_make 14:17 ijmorlan(a)27imac.local$ install_name_tool -change libpq.5.dylib @rpath/libpq.5.dylib dist/build/Database/PostgreSQL/LibPQ_hsc_make }}} That is, add an "rpath" to the executable, and change both libraries that need it to use it, with the following result: {{{ 14:41 ijmorlan(a)27imac.local$ otool -L dist/build/Database/PostgreSQL/LibPQ_hsc_make dist/build/Database/PostgreSQL/LibPQ_hsc_make: @rpath/libpq.5.dylib (compatibility version 5.0.0, current version 5.5.0) @rpath/libiconv.2.dylib (compatibility version 8.0.0, current version 8.1.0) /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 125.2.11) }}} The modified executable runs and dumps out a bunch of .hs code, as expected. So, I'm not sure if this is really best characterized as a bug in hsc2hs, but it seems like hsc2hs might be the right place to fix the problem. I've been trying to figure out extra parameters to hsc2hs to tell it to give extra parameters to the linker to install the rpath, but there are too many layers and I'm not confident the gcc documentation re: the -rpath option I'm reading is even appropriate on Mac OS X. So what I'm really hoping is that somebody familiar with hsc2hs and Mac OS X library management can give me some hints on how this might be fixed, and how it might be that I'm the only one who can't install postresql- libpq on Mac OS X. As I mentioned, my PostgreSQL installation is pretty stock, and I re-installed Haskell Platform just the other day, so I can't see how the problem is being caused by anything unusual that I'm doing. -- Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/7749> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

Re: [GHC] #2555: Template Haskell does not respect -package and -hide constraints
by GHC 24 Mar '13

24 Mar '13

#2555: Template Haskell does not respect -package and -hide constraints ---------------------------------+------------------------------------------ Reporter: guest | Type: bug Status: closed | Priority: lowest Milestone: 7.6.2 | Component: Template Haskell Version: 6.8.2 | Resolution: worksforme Keywords: | Os: Linux Architecture: Unknown/Multiple | Failure: None/Unknown Difficulty: Unknown | Testcase: Blockedby: | Blocking: Related: | ---------------------------------+------------------------------------------ Changes (by igloo): * status: new => closed * resolution: => worksforme Comment: Agreed; we don't have a way to reproduce it, so closing. -- Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/2555#comment:17> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #4962: Dead code fed to CorePrep because RULEs keep it alive spuriously
by GHC 24 Mar '13

24 Mar '13

#4962: Dead code fed to CorePrep because RULEs keep it alive spuriously --------------------------------------+------------------------------------- Reporter: batterseapower | Owner: Type: bug | Status: closed Priority: normal | Milestone: 7.2.1 Component: Compiler | Version: 7.0.1 Resolution: fixed | Keywords: Os: Unknown/Multiple | Architecture: Unknown/Multiple Failure: Runtime performance bug | Difficulty: Unknown Testcase: | Blockedby: Blocking: | Related: --------------------------------------+------------------------------------- Changes (by nfrisby): * cc: nfrisby (added) -- Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4962#comment:12> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #3333: GHCi doesn't load weak symbols
by GHC 24 Mar '13

24 Mar '13

#3333: GHCi doesn't load weak symbols --------------------------------------+------------------------------------- Reporter: heatsink | Owner: akio Type: bug | Status: patch Priority: normal | Milestone: 7.6.2 Component: GHCi | Version: 6.10.4 Keywords: weak, dynamic loading | Os: Linux Architecture: x86 | Failure: None/Unknown Difficulty: Unknown | Testcase: Blockedby: 3658 | Blocking: Related: | --------------------------------------+------------------------------------- Changes (by batterseapower): * cc: batterseapower@… (added) Comment: I got bit by this as well. My charsetdetect library (which links against a C++ library) won't load into ghci because it can't find __dso_handle. -- Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/3333#comment:25> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #7717: Libffi errors while cross-compiling GHC 7.6.2 (Stage 1)
by GHC 23 Mar '13

23 Mar '13

#7717: Libffi errors while cross-compiling GHC 7.6.2 (Stage 1) -----------------------------+---------------------------------------------- Reporter: guest | Owner: Type: bug | Status: new Priority: normal | Component: Compiler (FFI) Version: 7.6.2 | Keywords: Os: Unknown/Multiple | Architecture: mips Failure: None/Unknown | Blockedby: Blocking: | Related: -----------------------------+---------------------------------------------- I'm trying to cross-build GHC 7.6.2. I got stuck when I was trying to build a Stage 1 compiler [1]. (There is a chance that it's not a bug; maybe I misunderstood the process.) Steps to reproduce: $ ./configure --target=mips64el-linux --with-gcc=/usr/bin/gcc \ --with-ld=/usr/bin/ld --with-nm=/usr/bin/nm Did I specify the wrong binaries? These are the same I used for Stage 0 and libs boot phases. Maybe I should use the ones for mips64el... But how? ---------------------------------------------------------------------- Configure completed successfully. Building GHC version : 7.6.2 Build platform : i386-unknown-linux Host platform : i386-unknown-linux Target platform : mips-unknown-linux Bootstrapping using : /usr/local/bin/ghc which is version : 7.6.2 Using GCC : /usr/bin/gcc which is version : 4.4.5 Building a cross compiler : YES Porting to foreign arch : NO Alien script : ld : /usr/bin/ld Happy : /usr/bin/happy (1.18.4) Alex : /usr/bin/alex (2.3.3) Python : /usr/bin/python Perl : /usr/bin/perl dblatex : xsltproc : /usr/bin/xsltproc HsColour was not found; documentation will not contain source links Building DocBook HTML documentation : NO Building DocBook PS documentation : NO Building DocBook PDF documentation : NO ---------------------------------------------------------------------- $ make make[5]: Entering directory `/ghc-build2/ghc-7.6.2/libffi/build/mips64el- unknown-linux-gnu' depbase=`echo src/prep_cif.lo | sed 's|[^/]*$|.deps/&|;s|\.lo$||'`;\ /bin/bash ./libtool --tag=CC --mode=compile /usr/bin/gcc -DHAVE_CONFIG_H -I. -I.. -I. -I../include -Iinclude -I../src -DFFI_BUILDING -g -fno-stack-protector -w -Wall -fexceptions -MT src/prep_cif.lo -MMD -MP -MF $depbase.Tpo -c -o src/prep_cif.lo ../src/prep_cif.c &&\ mv -f $depbase.Tpo $depbase.Plo libtool: compile: /usr/bin/gcc -DHAVE_CONFIG_H -I. -I.. -I. -I../include -Iinclude -I../src -DFFI_BUILDING -g -fno-stack-protector -w -Wall -fexceptions -MT src/prep_cif.lo -MMD -MP -MF src/.deps/prep_cif.Tpo -c ../src/prep_cif.c -o src/prep_cif.o In file included from include/ffi.h:67, from ../src/prep_cif.c:26: include/ffitarget.h:59:3: error: #error -- something is very wrong -- make[5]: *** [src/prep_cif.lo] Error 1 make[5]: Leaving directory `/ghc-build2/ghc-7.6.2/libffi/build/mips64el- unknown-linux-gnu' make[4]: *** [all-recursive] Error 1 make[4]: Leaving directory `/ghc-build2/ghc-7.6.2/libffi/build/mips64el- unknown-linux-gnu' make[3]: *** [all] Error 2 make[3]: Leaving directory `/ghc-build2/ghc-7.6.2/libffi/build/mips64el- unknown-linux-gnu' make[2]: *** [all-all] Error 2 make[1]: *** [libffi/stamp.ffi.static.build] Error 2 make: *** [all] Error 2 Similar report: [2]. I added these lines before the error: #ifndef _MIPS_SIM # define _MIPS_SIM _ABIN32 #endif And continued: $ make ../src/mips/ffi.c: In function 'ffi_prep_cif_machdep': ../src/mips/ffi.c:468: error: duplicate case value ../src/mips/ffi.c:463: error: previously used here ../src/mips/ffi.c:548: error: duplicate case value ../src/mips/ffi.c:541: error: previously used here make[5]: *** [src/mips/ffi.lo] Error 1 make[5]: Leaving directory `/ghc-build2/ghc-7.6.2/libffi/build/mips64el- unknown-linux-gnu' make[4]: *** [all-recursive] Error 1 make[4]: Leaving directory `/ghc-build2/ghc-7.6.2/libffi/build/mips64el- unknown-linux-gnu' make[3]: *** [all] Error 2 make[3]: Leaving directory `/ghc-build2/ghc-7.6.2/libffi/build/mips64el- unknown-linux-gnu' make[2]: *** [all-all] Error 2 make[1]: *** [libffi/stamp.ffi.static.build] Error 2 make: *** [all] Error 2 Similar report: [3]. All my attempts to fix the above failed with these errors: ../src/mips/n32.S: Assembler messages: ../src/mips/n32.S:44: Error: unknown pseudo-op: `.abicalls' ../src/mips/n32.S:46: Error: expected comma after "mips4" ../src/mips/n32.S:50: Error: unknown pseudo-op: `.ent' ../src/mips/n32.S:53: Error: unknown pseudo-op: `.frame' ../src/mips/n32.S:54: Error: unknown pseudo-op: `.mask' ../src/mips/n32.S:55: Error: unknown pseudo-op: `.fmask' ../src/mips/n32.S:58: Error: no such instruction: `dsubu $sp,( 8 * 8)' [...] make[5]: *** [src/mips/n32.lo] Error 1 make[5]: Leaving directory `/ghc-build2/ghc-7.6.2/libffi/build/mips64el- unknown-linux-gnu' make[4]: *** [all-recursive] Error 1 make[4]: Leaving directory `/ghc-build2/ghc-7.6.2/libffi/build/mips64el- unknown-linux-gnu' make[3]: *** [all] Error 2 make[3]: Leaving directory `/ghc-build2/ghc-7.6.2/libffi/build/mips64el- unknown-linux-gnu' make[2]: *** [all-all] Error 2 make[1]: *** [libffi/stamp.ffi.static.build] Error 2 make: *** [all] Error 2 [1] http://hackage.haskell.org/trac/ghc/wiki/CrossCompilation [2] http://mail-index.netbsd.org/netbsd-bugs/2012/11/16/msg030949.html [3] http://sourceware.org/ml/libffi-discuss/2012/msg00201.html -- Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/7717> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

Re: [GHC] #4364: Template Haskell: Cycle in type synonym declarations
by GHC 22 Mar '13

22 Mar '13

#4364: Template Haskell: Cycle in type synonym declarations ----------------------------------------+----------------------------------- Reporter: igloo | Owner: simonpj Type: bug | Status: new Priority: high | Milestone: 7.8.1 Component: Compiler (Type checker) | Version: 7.1 Keywords: | Os: Unknown/Multiple Architecture: Unknown/Multiple | Failure: None/Unknown Difficulty: Unknown | Testcase: Blockedby: | Blocking: Related: | ----------------------------------------+----------------------------------- Changes (by awson): * cc: awson (added) -- Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4364#comment:8> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #3658: Dynamically link GHCi (and use system linker) on platforms that support it
by GHC 22 Mar '13

22 Mar '13

#3658: Dynamically link GHCi (and use system linker) on platforms that support it ---------------------------------+------------------------------------------ Reporter: simonmar | Owner: igloo Type: task | Status: new Priority: high | Milestone: 7.8.1 Component: GHCi | Version: 6.10.4 Keywords: | Os: Unknown/Multiple Architecture: Unknown/Multiple | Failure: None/Unknown Difficulty: Unknown | Testcase: Blockedby: 5987 | Blocking: 781, 1883, 2283, 3242, 3333, 3372, 3654, 4244, 5062, 5197, 5435, 6107, 7043, 7056, 7072, 7097, 7103, 7134, 7475 Related: | ---------------------------------+------------------------------------------ Changes (by igloo): * blocking: 781, 1883, 2283, 3242, 3372, 3654, 4244, 5062, 5197, 5435, 6107, 7043, 7056, 7072, 7097, 7103, 7134, 7475 => 781, 1883, 2283, 3242, 3333, 3372, 3654, 4244, 5062, 5197, 5435, 6107, 7043, 7056, 7072, 7097, 7103, 7134, 7475 -- Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/3658#comment:29> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #5642: Deriving Generic of a big type takes a long time and lots of space
by GHC 22 Mar '13

22 Mar '13

#5642: Deriving Generic of a big type takes a long time and lots of space ---------------------------------+------------------------------------------ Reporter: basvandijk | Owner: dimitris Type: bug | Status: new Priority: high | Milestone: 7.6.2 Component: Compiler | Version: 7.3 Keywords: | Os: Unknown/Multiple Architecture: Unknown/Multiple | Failure: Compile-time performance bug Difficulty: Unknown | Testcase: T5642 Blockedby: | Blocking: Related: | ---------------------------------+------------------------------------------ Changes (by liyang): * cc: hackage.haskell.org@… (added) -- Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5642#comment:16> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #5539: GHC panic - Simplifier ticks exhausted
by GHC 22 Mar '13

22 Mar '13

#5539: GHC panic - Simplifier ticks exhausted ---------------------------------+------------------------------------------ Reporter: hvr | Owner: simonpj Type: bug | Status: new Priority: high | Milestone: 7.6.2 Component: Compiler | Version: 7.3 Resolution: | Keywords: Os: Linux | Architecture: x86_64 (amd64) Failure: Compile-time crash | Difficulty: Unknown Testcase: | Blockedby: Blocking: | Related: ---------------------------------+------------------------------------------ Changes (by liyang): * cc: hackage.haskell.org@… (added) -- Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5539#comment:49> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0