May 2017 - ghc-tickets

Re: [GHC] #1290: ghc runs preprocessor too much
by GHC 09 May '17

09 May '17

#1290: ghc runs preprocessor too much -------------------------------------+------------------------------------- Reporter: guest | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Driver | Version: 6.7 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Compile-time | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by bgamari): * failure: None/Unknown => Compile-time performance bug -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/1290#comment:11> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #13150: unknown warning is not reported by GHC
by GHC 09 May '17

09 May '17

#13150: unknown warning is not reported by GHC -------------------------------------+------------------------------------- Reporter: jonaprieto | Owner: Type: bug | Status: new Priority: high | Milestone: Component: Compiler | Version: 8.0.1 Keywords: | Operating System: MacOS X Architecture: | Type of failure: Incorrect Unknown/Multiple | error/warning at compile-time Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The documented and official flag for the warning when a module, function or type with a WARNING or DEPRECATED pragma is used is `-fwarn-warnings- deprecations`. (See for instance https://downloads.haskell.org/~ghc/7.8.2/docs/html/users_guide/options- sanity.html). If the goal is turn off the warning I use `-fno-warn- warnings-deprecations` flag. By default `-fwarn-warnings-deprecations` on. In the next example shows a warning by using the flag mention above. {{{ $ cat Foo.hs {-# OPTIONS_GHC -fwarn-warnings-deprecations #-} import Data.Version (Version(Version, versionBranch, versionTags)) main :: IO () main = do let v = Version { versionBranch = [1,0,0] , versionTags = ["beta"] } return () }}} and GHC outputs: {{{ $ ghc Foo.hs [1 of 1] Compiling Main ( Foo.hs, Foo.o ) Foo.hs:9:11: warning: [-Wdeprecations] In the use of ‘versionTags’ (imported from Data.Version): Deprecated: "See GHC ticket #2496" Linking Foo ... }}} The first unexcepted behavior is the warning message showed above. Following the convention the warning message should be: {{{ Foo.hs:9:11: warning: [-Wwarnings-deprecations] }}} The second unexcepted behavior happened when I used a pragma to silenced the warning `warnings-deprecations`, I committed the mistake to misspelled it, but surprisingly GHC didn't complaint about the unknown pragma, it seems treated again like a synonymous of `-fno-warn-warnings- deprecations` in this case. {{{ {-# OPTIONS_GHC -fno-warn-deprecations #-} import Data.Version (Version(Version, versionBranch, versionTags)) main :: IO () main = do let v = Version { versionBranch = [1,0,0] , versionTags = ["beta"] } return () }}} Searching a little, I found it is not documented the flag -Wdeprecations in recent versions of GHC ( I tested with 7.8.2 too) but it does for instance available in GHC 6.3.0 (See https://manned.org/ghc-cvs/42a4c961) but it actually continues appearing in the source code of the recent GHC. See https://github.com/mlen/ghc/search?utf8=%E2%9C%93&q=deprecations. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13150> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #13239: Phabricator upgrade broke the ticket custom field
by GHC 09 May '17

09 May '17

#13239: Phabricator upgrade broke the ticket custom field -------------------------------------+------------------------------------- Reporter: mpickering | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 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: -------------------------------------+------------------------------------- We have one custom field which should also appear in commit messages when running "arc diff". Since the upgrade on 07/02/2017, the field no longer appears when running "arc diff". The upgrade path is detailed in this upstream task. https://secure.phabricator.com/T12085 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13239> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

Re: [GHC] #314: #line pragmas not respected inside nested comments
by GHC 09 May '17

09 May '17

#314: #line pragmas not respected inside nested comments -------------------------------------+------------------------------------- Reporter: nobody | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: 8.4.1 Component: Compiler | Version: 6.4 (Parser) | Resolution: None | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: read032 Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by bgamari): * priority: high => normal Comment: This clearly isn't particularly hurtful to anyone given it has been open for 12 years. Happy birthday, bug! -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/314#comment:17> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #13676: Command line written to stats file by the rts is not executable
by GHC 09 May '17

09 May '17

#13676: Command line written to stats file by the rts is not executable -------------------------------------+------------------------------------- Reporter: duog | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Runtime | Version: 8.0.1 System | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Debugging Unknown/Multiple | information is incorrect Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The first line of the stats file output by the rts is the invocation of the program. However the "s around arguments are not present in that line. For example: {{{ $ ghc +RTS -trts-test -RTS -e "do {putStrLn\$ show [1,2] ++ (\"(\\\\haskell)\")}" &>/dev/null && cat rts-test | head -n 1 /opt/ghc/8.0.2/lib/ghc-8.0.2/bin/ghc -B/opt/ghc/8.0.2/lib/ghc-8.0.2 -e do {putStrLn$ show [1,2] ++ ("(\\haskell)")} +RTS -trts-test }}} I would expect that this line could be pasted into a shell to invoke the command, however this is not the case as evidenced above. There are several issues evidenced by the above example: * the quoted argument "do {...}" does not have quotes in the output. I believe this is a true bug. * the $,"s and \s inside the quoted argument are not escaped. I'm fairly sure that at least the "s and \s should be escaped. Not so sure that $ should be, since a shell isn't the only way to run a program. * The +RTS options are moved to the end. This is seems intentional and doesn't change semantics, but it seems better to be exactly the same. The relevant code is in initStatsFile in RtsFlags.c I'd suggest quoting each argument, escaping at least "s and \s inside arguments and thinking about whether escaping other shell characters like $ is a good idea. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13676> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #13673: Core lint error with defer-typed-holes
by GHC 09 May '17

09 May '17

#13673: Core lint error with defer-typed-holes -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 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: -------------------------------------+------------------------------------- Another Core Lint error {{{#!hs {-# Options_GHC -dcore-lint -fdefer-typed-holes #-} {-# Language ViewPatterns, PatternSynonyms, GeneralizedNewtypeDeriving, DeriveTraversable #-} import Data.Bifunctor.Fix import Data.Bifunctor.Tannen import Test.QuickCheck import Test.QuickCheck.Function import Control.Applicative newtype ZL a = ZL { getZL :: Fix (Tannen Maybe (,)) a } deriving (Eq, Ord, Read, Show, Functor, Applicative, Foldable, Traversable) fromList :: [a] -> ZL a fromList [] = Nil fromList (x:xs) = x:::fromList xs pattern Nil :: ZL a pattern Nil = ZL (In (Tannen Nothing)) foo :: ZL a -> Maybe (a, ZL a) foo (ZL (In (Tannen (Just (as, a))))) = Just (a, ZL as) infixr 5 ::: pattern (:::) :: a -> ZL a -> ZL a pattern a:::as <- ZL (In (Tannen (Just (ZL -> as, a)))) where a:::ZL as = ZL (In (Tannen (Just (as, a)))) u (Fn f) as bs = let (xs, ys) = (ZipList as, ZipList bs) in getZipList (liftA2 f xs ys) == toList (liftA2 f (fromList as) (fromList bs)) }}} unfortunately I don't have time to reduce it -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13673> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #13672: Pattern match on LHS of pattern synonym declaration
by GHC 09 May '17

09 May '17

#13672: Pattern match on LHS of pattern synonym declaration -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: (none) Type: feature | Status: new request | Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple PatternSynonyms | Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Changing `ZipList` to a newtype {{{#!hs import Data.Bifunctor.Fix import Data.Bifunctor.Tannen type ZipList = Fix (Tannen Maybe (,)) pattern Nil :: ZipList a pattern Nil = In (Tannen Nothing) infixr 5 ::: pattern (:::) :: a -> ZipList a -> ZipList a pattern a:::as = In (Tannen (Just (as, a))) }}} works fine for the `Nil` pattern synonym (it remains implicitly bidirectional) {{{#!hs newtype ZipList a = ZL (Fix (Tannen Maybe (,)) a) pattern Nil :: ZipList a pattern Nil = ZL (In (Tannen Nothing)) }}} but `(:::)` needs to be explicitly bidirectional {{{#!hs infixr 5 ::: pattern (:::) :: a -> ZipList a -> ZipList a pattern a:::as <- ZL (In (Tannen (Just (ZL -> as, a)))) where a:::ZL as = ZL (In (Tannen (Just (as, a)))) }}} ---- I would like to write it as {{{#!hs pattern (:::) :: a -> ZipList a -> ZipList a pattern a:::ZL as = ZL (In (Tannen (Just (as, a)))) }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13672> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #13671: Core lint error with PatternSynonyms and undefined
by GHC 09 May '17

09 May '17

#13671: Core lint error with PatternSynonyms and undefined -------------------------------------+------------------------------------- Reporter: Iceland_jack | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 Keywords: | Operating System: Unknown/Multiple PatternSynonyms | Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- I've submitted quite a few Core Lint errors lately, I don't have access to GHC HEAD so ignore this ticket if this has already been resolved {{{#!hs {-# Options_GHC -dcore-lint #-} {-# Language ViewPatterns, PatternSynonyms #-} import Data.Bifunctor.Fix import Data.Bifunctor.Tannen newtype ZipList a = ZL { getZipList :: Fix (Tannen Maybe (,)) a } pattern Nil :: ZipList a pattern Nil = ZL (In (Tannen Nothing)) infixr 5 ::: pattern (:::) :: a -> ZipList a -> ZipList a pattern a:::as <- ZL (In (Tannen (undefined -> (a, as)))) where a:::as = undefined }}} `↑` works fine without `-dcore-lint` but fails with it. Works fine if `undefined -> (a, as)` is replaced with `let x = x in x -> (a, as)`. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13671> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #13668: Space leak re-introduced
by GHC 09 May '17

09 May '17

#13668: Space leak re-introduced -------------------------------------+------------------------------------- Reporter: simonpj | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 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: -------------------------------------+------------------------------------- Test `perf/compiler/T9675` is failing for me on a clean HEAD {{{ =====> T9675(optasm) 1 of 1 [0, 0, 0] cd "./T9675.run" && "/5playpen/simonpj/HEAD-5/inplace/test spaces/ghc- stage2" -c T9675.hs -no-user-package-db -rtsopts -fno-warn-missed- specialisations -fshow-warning-groups -fdiagnostics-color=never -fno- diagnostics-show-caret -dno-debug-output +RTS -G1 -RTS -O -fasm +RTS -V0 -tT9675.comp.stats --machine-readable -RTS"/5playpen/simonpj/HEAD-5/inplace/test spaces/ghc-stage2" -c T9675.hs -no-user-package-db -rtsopts -fno-warn-missed-specialisations -fshow-warning-groups -fdiagnostics-color=never -fno-diagnostics-show- caret -dno-debug-output +RTS -G1 -RTS -O -fasm +RTS -V0 -tT9675.comp.stats --machine-readable -RTS max_bytes_used value is too high: Expected T9675(optasm) max_bytes_used: 17675240 +/-15% Lower bound T9675(optasm) max_bytes_used: 15023954 Upper bound T9675(optasm) max_bytes_used: 20326526 Actual T9675(optasm) max_bytes_used: 25571248 Deviation T9675(optasm) max_bytes_used: 44.7 % peak_megabytes_allocated value is too high: Expected T9675(optasm) peak_megabytes_allocated: 63 +/-15% Lower bound T9675(optasm) peak_megabytes_allocated: 53 Upper bound T9675(optasm) peak_megabytes_allocated: 73 Actual T9675(optasm) peak_megabytes_allocated: 93 Deviation T9675(optasm) peak_megabytes_allocated: 47.6 % *** unexpected stat test failure for T9675(optasm) }}} Compiling with `-RTS -S` gives {{{ Alloc Copied Live GC GC TOT TOT Page Flts bytes bytes bytes user elap user elap 1139048 56288 158880 0.028 0.028 0.032 0.032 0 0 (Gen: 0) 1036832 99080 193448 0.018 0.018 0.052 0.052 0 0 (Gen: 0) 1036784 137552 223696 0.018 0.018 0.073 0.073 0 0 (Gen: 0) 1015208 145088 231232 0.018 0.018 0.094 0.094 0 0 (Gen: 0) 1084448 505864 636944 0.019 0.019 0.116 0.116 0 0 (Gen: 0) 1039312 660432 783288 0.020 0.020 0.137 0.137 0 0 (Gen: 0) 1377128 845976 1001624 0.020 0.020 0.159 0.159 0 0 (Gen: 0) 1707720 1168392 1352664 0.024 0.024 0.186 0.186 0 0 (Gen: 0) 2411008 1554360 1764288 0.023 0.023 0.214 0.214 0 0 (Gen: 0) 3128632 1885104 2164224 0.021 0.021 0.239 0.240 0 0 (Gen: 0) 3891928 3029560 3501264 0.025 0.025 0.271 0.272 0 0 (Gen: 0) 6618768 4554248 5412752 0.031 0.031 0.312 0.313 0 0 (Gen: 0) 8966136 5509176 6359368 0.033 0.033 0.358 0.359 0 0 (Gen: 0) 10948736 6299736 7158072 0.034 0.034 0.409 0.410 0 0 (Gen: 0) 12489360 8415312 9277728 0.043 0.044 0.466 0.468 0 0 (Gen: 0) 16486944 7800096 8662512 0.043 0.043 0.528 0.529 0 0 (Gen: 0) 15450408 9812088 10674504 0.044 0.044 0.590 0.591 0 0 (Gen: 0) 19459168 11429320 12291736 0.054 0.054 0.671 0.673 0 0 (Gen: 0) 22684384 8611624 9474040 0.046 0.046 0.745 0.747 0 0 (Gen: 0) 17026576 10661064 11523480 0.045 0.045 0.810 0.812 0 0 (Gen: 0) 21146552 9790592 10653008 0.044 0.044 0.882 0.884 0 0 (Gen: 0) 19421144 9891256 10753672 0.045 0.045 0.955 0.957 0 0 (Gen: 0) 19622512 8678360 9540776 0.042 0.042 1.026 1.028 0 0 (Gen: 0) 17212344 9845688 10708104 0.042 0.042 1.088 1.091 0 0 (Gen: 0) 19568256 9928304 10790720 0.045 0.045 1.159 1.162 0 0 (Gen: 0) 19673096 8159672 9022088 0.040 0.040 1.223 1.226 0 0 (Gen: 0) 16172824 9362616 10225032 0.039 0.039 1.277 1.280 0 0 (Gen: 0) 18536544 10809776 11672192 0.051 0.051 1.347 1.350 0 0 (Gen: 0) 21412128 15177144 16039560 0.065 0.065 1.434 1.437 0 0 (Gen: 0) 30181520 16438872 17301288 0.071 0.071 1.541 1.544 0 0 (Gen: 0) 32568088 11742016 12604432 0.053 0.053 1.633 1.636 0 0 (Gen: 0) 23285000 13139928 14002344 0.054 0.054 1.714 1.718 0 0 (Gen: 0) 26054072 15188040 16050456 0.060 0.060 1.803 1.807 0 0 (Gen: 0) 30184528 10990656 11853072 0.050 0.050 1.890 1.894 0 0 (Gen: 0) 21814064 12636584 13499000 0.050 0.050 1.961 1.965 0 0 (Gen: 0) 25295544 13754616 14812696 0.059 0.059 2.048 2.052 0 0 (Gen: 0) 27448728 16160496 17210352 0.066 0.066 2.147 2.151 0 0 (Gen: 0) 30295624 22372880 24471344 0.090 0.090 2.279 2.283 0 0 (Gen: 0) 44477824 24497712 25572240 0.092 0.092 2.431 2.436 0 0 (Gen: 0) 48692880 2937512 3436608 0.017 0.017 2.548 2.571 0 0 (Gen: 0) 1840 0.000 0.000 662,063,640 bytes allocated in the heap 334,683,080 bytes copied during GC 25,572,240 bytes maximum residency (40 sample(s)) 1,194,192 bytes maximum slop 93 MB total memory in use (0 MB lost due to fragmentation) Tot time (elapsed) Avg pause Max pause Gen 0 40 colls, 0 par 1.684s 1.685s 0.0421s 0.0924s TASKS: 4 (1 bound, 3 peak workers (3 total), using -N1) SPARKS: 0 (0 converted, 0 overflowed, 0 dud, 0 GC'd, 0 fizzled) INIT time 0.001s ( 0.001s elapsed) MUT time 0.862s ( 0.885s elapsed) GC time 1.684s ( 1.685s elapsed) EXIT time 0.001s ( 0.009s elapsed) Total time 2.549s ( 2.581s elapsed) Alloc rate 767,692,191 bytes per MUT second Productivity 33.9% of total user, 34.6% of total elapsed }}} It looks as if a space leak has come back; space usage on this benchmark dropped by half when Reid fixed the `CoreTidy` leak. I don't know which commit re-introduced it. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13668> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #13655: Spurious untouchable type variable in connection with rank-2 type and constraint family
by GHC 09 May '17

09 May '17

#13655: Spurious untouchable type variable in connection with rank-2 type and constraint family -------------------------------------+------------------------------------- Reporter: jeltsch | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.1-rc2 (Type checker) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The following code cannot be compiled: {{{#!hs {-# LANGUAGE ConstraintKinds, Rank2Types, TypeFamilies #-} import GHC.Exts (Constraint) type family F a b :: Constraint data T b c = T f :: (forall b . F a b => T b c) -> a f _ = undefined }}} GHC outputs the following error message: {{{ Untouchable.hs:9:6: error: • Couldn't match type ‘c0’ with ‘c’ ‘c0’ is untouchable inside the constraints: F a b bound by the type signature for: f :: F a b => T b c0 at Untouchable.hs:9:6-37 ‘c’ is a rigid type variable bound by the type signature for: f :: forall a c. (forall b. F a b => T b c) -> a at Untouchable.hs:9:6 Expected type: T b c0 Actual type: T b c • In the ambiguity check for ‘f’ To defer the ambiguity check to use sites, enable AllowAmbiguousTypes In the type signature: f :: (forall b. F a b => T b c) -> a }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13655> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1