March 2016 - ghc-tickets

[GHC] #10332: AArch64 : divbyzero test fails
by GHC 06 Dec '16

06 Dec '16

#10332: AArch64 : divbyzero test fails -------------------------------------+------------------------------------- Reporter: erikd | Owner: Type: bug | Status: new Priority: normal | Milestone: 7.12.1 Component: Compiler | Version: 7.11 Keywords: | Operating System: Unknown/Multiple Architecture: aarch64 | Type of failure: Incorrect result Test Case: | at runtime testsuite/tests/rts/divbyzero.hs | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- On AArch64/Linux the test in `testsuite/tests/rts/divbyzero.hs` which is: {{{ main :: IO () main = print (5 `divInt` 0) }}} just prints "0" and exits with a zero status code. From the "ARMv8 Instriction Set Overview" document I found (wasn't able to find anything more recent): https://www.element14.com/community/servlet/JiveServlet/downloadBody/41836-… Section 3.6 says "There is no hardware check for “divide by zero”, but this check can be performed in the shadow of a long latency division. A divide by zero writes zero to the destination register." Looks like we need extra code to check for this, but not sure how to report it. Should probably look at what GCC and Clang do. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10332> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #10174: AArch64 : ghc-stage2 segfaults compiling libraries/parallel
by GHC 06 Dec '16

06 Dec '16

#10174: AArch64 : ghc-stage2 segfaults compiling libraries/parallel -------------------------------------+------------------------------------- Reporter: erikd | Owner: Type: bug | Status: new Priority: normal | Milestone: 7.12.1 Component: Compiler | Version: 7.11 Keywords: | Operating System: Unknown/Multiple Architecture: aarch64 | Type of failure: Building GHC Test Case: | failed Blocking: | Blocked By: Differential Revisions: | Related Tickets: -------------------------------------+------------------------------------- Compiling git HEAD (e02ef0e6d4eefa5f0) on AArch64 hardware using ghc-7.6.3 from Debian as the bootstrap compiler. LLVM is version 3.6. It fails with: {{{ "inplace/bin/ghc-stage2" -hisuf hi -osuf o -hcsuf hc -static -H64m -O0 -fllvm \ -this-package-key paral_2NTMt5X9WUG6LNupBFIZti -hide-all-packages -i \ -ilibraries/parallel/. -ilibraries/parallel/dist-install/build \ -ilibraries/parallel/dist-install/build/autogen \ -Ilibraries/parallel/dist-install/build \ -Ilibraries/parallel/dist-install/build/autogen -Ilibraries/parallel/. \ -optP-include -optPlibraries/parallel/dist- install/build/autogen/cabal_macros.h \ -package-key array_BLJREAlFJ4zJ2kSDphVieY -package-key base_8gvrDSBdaidLA14EDtK6ja \ -package-key conta_1uqbEcrmZiO1C91Z8ePoyI -package-key deeps_Bw45clVBNVT4S1LLyUOPfh \ -Wall -feager-blackholing -XHaskell2010 -O -fllvm -no-user-package- db -rtsopts \ -odir libraries/parallel/dist-install/build \ -hidir libraries/parallel/dist-install/build \ -stubdir libraries/parallel/dist-install/build \ -dynamic-too -c libraries/parallel/./Control/Parallel/Strategies.hs \ -o libraries/parallel/dist- install/build/Control/Parallel/Strategies.o \ -dyno libraries/parallel/dist- install/build/Control/Parallel/Strategies.dyn_o libraries/parallel/ghc.mk:5: recipe for target 'libraries/parallel/dist-install/build/Control/Parallel/Strategies.o' failed make[1]: *** [libraries/parallel/dist- install/build/Control/Parallel/Strategies.o] Segmentation fault }}} It seems that `inplace/lib/bin/ghc-stage1` (which is working fine) is statically linked to all the Haskell libraries where as `inplace/lib/bin /ghc-stage2` is dynamically linked which would explain why stage1 works fine and stage2 segfaults. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10174> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #11578: Fix GHC on AArch64/Arm64
by GHC 06 Dec '16

06 Dec '16

#11578: Fix GHC on AArch64/Arm64 --------------------------------+------------------------------------- Reporter: erikd | Owner: Type: bug | Status: new Priority: normal | Milestone: 8.2.1 Component: Compiler | Version: 8.1 Keywords: | Operating System: Unknown/Multiple Architecture: aarch64 | Type of failure: None/Unknown Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: --------------------------------+------------------------------------- Raising this to track discussion about GHC on Arm64. There is already some discussion in ticket #11206 (ARM is generally a disaster), but that ticket is about 32 bit Arm which is vastly different and is now mostly fixed. Information worth saving from #11206: * GHC 7.8.4 from Debian can build the ghc-8.0 branch, but cannot build HEAD because HEAD requires ghc-7.10 to build. * GHC 7.10 from Debian is broken because of an unfixed bug in llvm-3.5.2. * User @mfox reports that ghc-7.10 does build against llvm-3.7 and results in a compiler that can build HEAD. * That GHC 7.10 can be build against llvm-3.7 is a happy accident. Versions of GHC are usually locked to a specific version of LLVM. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11578> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

[GHC] #10962: Improved arithmetic primops
by GHC 06 Dec '16

06 Dec '16

#10962: Improved arithmetic primops -------------------------------------+------------------------------------- Reporter: nkaretnikov | Owner: nkaretnikov Type: task | Status: new 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: | https://ghc.haskell.org/trac/ghc/wiki/ImprovedArithmeticPrimops -------------------------------------+------------------------------------- See the todo list on the wiki page. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10962> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9

[GHC] #8224: Excessive system time -- new IO manager problem?
by GHC 06 Dec '16

06 Dec '16

#8224: Excessive system time -- new IO manager problem? -------------------------------------+------------------------------------- Reporter: rrnewton | Owner: Type: bug | Status: new Priority: high | Milestone: 7.8.1 Component: Runtime System | Version: 7.7 Keywords: IO Manager, | Operating System: Linux System Time | Type of failure: Runtime Architecture: x86_64 (amd64) | performance bug Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | -------------------------------------+------------------------------------- This is an issue that came to light when testing the patches on #910. You can see some of the numbers there. Basically, recent GHC HEAD builds, running a large number of threads blocked on MVars will result in burning a lot of system time. The attached file provides a mediocre reproducer. With it, you can see that building with HEAD as of Aug 31st and running with `-RTS -N32` will result in around 200ms system time, whereas GHC 7.6.3 spends about 30ms in system time. This shows the disparity, but the result is not that egregious. A more noticeable example is on ticket #910, where when running on 31 threads, there is an 8 minutes of system time for 17 minutes of user time, and yet at one thread that system time drops to under two seconds! {{{ 1 thread: real 1m20.028s user 1m17.921s sys 0m1.768s 31 threads: real 1m27.445s user 17m0.314s sys 8m0.175s }}} It needs to be determined whether this system time is a result of the parallel compilation patches on #910, or whether it is a new problem with the runtime system, and in particular with the parallel IO manager. I am inclined to believe that compiling in parallel involves extra IO (repeatedly reading interface files?), but not eight minutes of it!! -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8224> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 40

[GHC] #11548: Absolutely misleading error message on kind error
by GHC 06 Dec '16

06 Dec '16

#11548: Absolutely misleading error message on kind error -------------------------------------+------------------------------------- Reporter: mniip | Owner: Type: bug | Status: new Priority: high | Milestone: Component: Compiler | Version: 8.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Incorrect Unknown/Multiple | warning at compile-time Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- {{{#!hs module Bug where import Data.Proxy fun :: Proxy a -> () fun Proxy = () bug :: () bug = fun (Proxy :: Proxy Maybe) }}} {{{#!hs [1 of 1] Compiling Bug ( Bug.hs, Bug.o ) Bug.hs:9:12: error: • Expected kind ‘Proxy Maybe’, but ‘Data.Proxy.Proxy :: Proxy Maybe’ has kind ‘Proxy Maybe’ • In the first argument of ‘fun’, namely ‘(Proxy :: Proxy Maybe)’ In the expression: fun (Proxy :: Proxy Maybe) In an equation for ‘bug’: bug = fun (Proxy :: Proxy Maybe) }}} 7.10 gives the appropriate `Couldn't match kind ‘* -> *’ with ‘*’` message. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11548> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #11485: Very unhelpful message resulting from kind mismatch
by GHC 06 Dec '16

06 Dec '16

#11485: Very unhelpful message resulting from kind mismatch -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1-rc1 (Type checker) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Other Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The following code: {{{#!hs module Foo where import Data.Typeable tyConOf :: Typeable a => Proxy a -> TyCon tyConOf = typeRepTyCon . typeRep tcList :: TyCon tcList = tyConOf (Proxy :: Proxy []) }}} fails because `-XPolyKinds` is not enabled. But the error message that you get is quite different on GHC 7.10 and 8.0. On GHC 7.10.3: {{{ [1 of 1] Compiling Foo ( Foo.hs, Foo.o ) Foo.hs:9:19: Couldn't match kind ‘* -> *’ with ‘*’ Expected type: Proxy a0 Actual type: Proxy [] In the first argument of ‘tyConOf’, namely ‘(Proxy :: Proxy [])’ In the expression: tyConOf (Proxy :: Proxy []) In an equation for ‘tcList’: tcList = tyConOf (Proxy :: Proxy []) }}} But on GHC 8.0.1-rc1: {{{ Foo.hs:9:19: error: • Expected kind ‘Proxy []’, but ‘Data.Proxy.Proxy :: Proxy []’ has kind ‘Proxy []’ • In the first argument of ‘tyConOf’, namely ‘(Proxy :: Proxy [])’ In the expression: tyConOf (Proxy :: Proxy []) In an equation for ‘tcList’: tcList = tyConOf (Proxy :: Proxy []) }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11485> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #10812: High memory usage
by GHC 02 Dec '16

02 Dec '16

#10812: High memory usage -------------------------------------+------------------------------------- Reporter: danilo2 | Owner: Type: bug | Status: new Priority: high | Milestone: Component: Compiler | Version: 7.10.2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- Hello, lets consider following program: {{{ {-# LANGUAGE NoMonomorphismRestriction #-} import System.Mem.Weak import Control.Concurrent import System.Mem data Tst a = Tst a deriving (Show, Eq) tst a = do let arr = [0 .. a*a*a] v = Tst (seq arr arr) ptr <- mkWeakPtr v Nothing return ptr main = do ptrs <- mapM tst [1..100000000] --performGC --performMajorGC --performMinorGC threadDelay 1000000 xr <- mapM deRefWeak ptrs print $ length $ filter (/= Nothing) xr threadDelay 5000000 return () }}} It simply creates 10 million of weak references to values of {{{Tst}}}. These weak pointers are returned in the main function. After that we sleep a second and ask how meany references are alive. I get the {{{0}}} as a result here - so everything seems ok - garbage collection worked. There is a problem though - somehow the memory was not released, because on the second 5-second sleep the program uses over 14 Gb of RAM on my computer. I'm compiling it simply with {{{ghc -O2 Main.hs}}}. I'm pretty sure this is a bug, because such behaviour is not expected I think. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10812> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 11

[GHC] #11744: Latest Xcode update violates POSIX compliance of `nm -P`
by GHC 01 Dec '16

01 Dec '16

#11744: Latest Xcode update violates POSIX compliance of `nm -P` ----------------------------------------+--------------------------------- Reporter: hvr | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Build System | Version: 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: ----------------------------------------+--------------------------------- TLDR: Apple messed up again and broke POSIX compliance luite ran into problems when he installed the latest Xcode update because the derived platform constants were off. He provided me the output of `nm -P ./includes/dist- derivedconstants/header/tmp.o` which turned out have numbers which didn't make sense: {{{ _derivedConstantAP_STACK_SPLIM C 1025 0 _derivedConstantBITMAP_BITS_SHIFT C 7 0 _derivedConstantBLOCKS_PER_MBLOCK C 253 0 _derivedConstantBLOCK_SIZE C 4097 0 _derivedConstantCINT_SIZE C 5 0 _derivedConstantCLONG_LONG_SIZE C 9 0 _derivedConstantCLONG_SIZE C 9 0 _derivedConstantDOUBLE_SIZE C 9 0 _derivedConstantDYNAMIC_BY_DEFAULT C 2 0 ... }}} Which with an older Xcode version would have looked like {{{ _derivedConstantAP_STACK_SPLIM C 401 0 _derivedConstantBITMAP_BITS_SHIFT C 7 0 _derivedConstantBLOCKS_PER_MBLOCK C fd 0 _derivedConstantBLOCK_SIZE C 1001 0 _derivedConstantCINT_SIZE C 5 0 _derivedConstantCLONG_LONG_SIZE C 9 0 _derivedConstantCLONG_SIZE C 9 0 _derivedConstantDOUBLE_SIZE C 9 0 _derivedConstantDYNAMIC_BY_DEFAULT C 2 0 ... }}} So obviously `nm -P`'s output changed from hexadecimal to (non-compliant) decimal formatting of numbers. For reference, here's how POSIX specifies the semantics of `nm`: http://pubs.opengroup.org/onlinepubs/9699919799/utilities/nm.html > `-P` > Write information in a portable output format, as specified in the STDOUT section. ... > If the `-P` option is specified, the previous information shall be displayed using the following portable format. The three versions differ depending on whether `-t d`, `-t o`, or `-t x` was specified, respectively: > > {{{"%s%s %s %d %d\n", <library/object name>, <name>, <type>, <value>, <size>}}} > > {{{"%s%s %s %o %o\n", <library/object name>, <name>, <type>, <value>, <size>}}} > > {{{"%s%s %s %x %x\n", <library/object name>, <name>, <type>, <value>, <size>}}} In the event that Apple can't be bothered to fix this bug (I've had bad experience with Apple and standard-compliance in the past, hence my pessimism) we could try to workaround by using `nm -P -t x` when building on OSX... -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11744> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 22

[GHC] #11724: Must check for representation polymorphism in datatype declarations
by GHC 01 Dec '16

01 Dec '16

#11724: Must check for representation polymorphism in datatype declarations -------------------------------------+------------------------------------- Reporter: goldfire | Owner: goldfire Type: bug | Status: new Priority: normal | Milestone: 8.0.1 Component: Compiler | Version: 8.1 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: -------------------------------------+------------------------------------- {{{#!hs data Foo (r :: RuntimeRep) (a :: TYPE r) = Foo a }}} panics. It should error instead. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11724> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4