August 2013 - ghc-tickets

[GHC] #8096: Add fudge-factor for performance tests run on non-validate builds
by GHC 26 Aug '15

26 Aug '15

#8096: Add fudge-factor for performance tests run on non-validate builds ------------------------------------+------------------------------------- Reporter: ezyang | Owner: Type: task | 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: | ------------------------------------+------------------------------------- Since I'm not going to get around to this immediately, Trac'ifying for posterity: These tests have been doing better than expected in the nightlies for some while. {{{ Unexpected failures: perf/compiler T3064 [stat too good] (normal) perf/compiler T3294 [stat too good] (normal) perf/compiler T5642 [stat too good] (normal) perf/haddock haddock.Cabal [stat too good] (normal) perf/haddock haddock.base [stat too good] (normal) }}} Unfortunately, fixing them is not a simple matter of shifting the ranges up, since the tests only exceed expectations on a /perf/ build, so on a normal build such as 'quick', these tests all pass normally. I could bump up the upper bounds so that the builder stops bleating about them; perhaps we could do something more complicated where the expected performance depends on what level of optimization GHC was built with (but I don't know how to implement this.) ---- The problem with just widening the bounds to cover 2 different types of build is that it increases the chance that performance changes won't actually be noticed by thge person responsible. Having different bounds for different build configurations is a pain, because (a) the testsuite has to work out which set of bounds to use, and (b) you now have even more wobbly values to keep up-to-date. I think perhaps the best thing would be to add some sort of (per-test?) fudge factor for non-validate builds. That way validate will still find performance regressions, like it does today, but other builds are less likely to give false positives. (Igloo) -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8096> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 1

[GHC] #8178: Need TypeRep for Symbol and numeric type literals; and Typeable instances
by GHC 08 Aug '15

08 Aug '15

#8178: Need TypeRep for Symbol and numeric type literals; and Typeable instances ------------------------------------+------------------------------------- Reporter: simonpj | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.6.3 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- Nicolas Trangez points out that we don't have a `TypeRep` for types involving literal strings or numerics. E.g. what is `(typeRep (undefined :: Proxy "foo"))`? Here is Nicolas's example: {{{ {-# LANGUAGE DataKinds, KindSignatures, DeriveFunctor, DeriveDataTypeable #-} module Main where import Data.Typeable import GHC.TypeLits data NoSymbol n a b = NoSymbol a b deriving (Typeable) data WithSymbol (n :: Symbol) a b = WithSymbol a b deriving (Typeable) data Sym deriving (Typeable) main :: IO () main = do print $ typeOf (undefined :: NoSymbol Sym Int Int) let d = undefined :: WithSymbol "sym" Int Int {- print $ typeOf d No instance for (Typeable Symbol "sym") arising from a use of ‛typeOf’ -} return () }}} Just as types contain literal strings and natural numbers, so too must `TypeRep`, in some way. Once we have a suitable `TypeRep` we can make built-in instance for `Typeable` that return the appropriate representation. At the moment we have (in `Data.Typeable.Internals`): {{{ data TypeRep = TypeRep {-# UNPACK #-} !Fingerprint TyCon [TypeRep] data TyCon = TyCon { tyConHash :: {-# UNPACK #-} !Fingerprint, tyConPackage :: String, tyConModule :: String, tyConName :: String } }}} with instances for Eq and Ord. Perhaps we need {{{ data TypeRep = TcApp {-# UNPACK #-} !Fingerprint TyCon [TypeRep] | TcSym String | TcNat Integer }}} or something like that? I'm not certain. I think Iavor is going to think about it. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8178> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #8151: ghc-7.4.2 on OpenIndiana (Solaris) createSubprocess fails
by GHC 31 Jul '15

31 Jul '15

#8151: ghc-7.4.2 on OpenIndiana (Solaris) createSubprocess fails ----------------------------+-------------------------------------------- Reporter: troydm | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.4.2 Keywords: | Operating System: Solaris Architecture: x86 | Type of failure: Runtime performance bug Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ----------------------------+-------------------------------------------- when i'm compiling this test.hs with ghc -thread ./test.hs and then execute it i get Starting Subprocess Exit code: ExitFailure 127 This error code indicates that command isn't found (i have ghc in /usr/local/bin) and it's specified in PATH variable, however it acts as if it's not even there however if i remove -threaded flag and recompile code again, it works Starting Subprocess ghc: no input files Usage: For basic information, try the `--help' option. Exit code: ExitFailure 1 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8151> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #8131: T7571 with WAY=llvm fails, but not WAY=optllvm
by GHC 19 Jul '15

19 Jul '15

#8131: T7571 with WAY=llvm fails, but not WAY=optllvm ------------------------------------+------------------------------------- Reporter: thoughtpolice | Owner: Type: bug | Status: new Priority: high | 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: | ------------------------------------+------------------------------------- Test case (#7571 just will not die, it seems) {{{ $ make TEST=T7571 WAY=llvm ... ====> Scanning ./all.T =====> T7571(llvm) 6 of 7 [0, 0, 0] cd . && '/home/a/ghc/ghc-pristine/inplace/bin/ghc-stage2' -fforce-recomp -dcore-lint -dcmm-lint -dno-debug-output -no-user-package-db -rtsopts -fno-ghci-history -c T7571.cmm -fllvm -no-hs-main >T7571.comp.stderr 2>&1 Actual stderr output differs from expected: --- /dev/null 2013-08-07 22:18:14.132934017 -0500 +++ ./T7571.comp.stderr 2013-08-14 01:26:33.942103723 -0500 @@ -0,0 +1 @@ +WARNING: Non constant alignment value given for memcpy! Please report to GHC developers *** unexpected failure for T7571(llvm) }}} However, this test case does not fail under WAY=optllvm, which I think is worth investigating probably: {{{ $ make TEST=T7571 WAY=optllvm ... ====> Scanning ./all.T =====> T7571(optllvm) 6 of 7 [0, 0, 0] cd . && '/home/a/ghc/ghc-pristine/inplace/bin/ghc-stage2' -fforce-recomp -dcore-lint -dcmm-lint -dno-debug-output -no-user-package-db -rtsopts -fno-ghci-history -c T7571.cmm -O -fllvm -no-hs-main >T7571.comp.stderr 2>&1 OVERALL SUMMARY for test run started at Wed Aug 14 01:25:42 CDT 2013 7 total tests, which gave rise to 28 test cases, of which 27 were skipped 0 had missing libraries 1 expected passes 0 expected failures 0 caused framework failures 0 unexpected passes 0 unexpected failures }}} However, there is another way to trigger this bug in *both* cases: {{{ #include "Cmm.h" testMemcpy (W_ dst, W_ src, W_ l, W_ sz) { prim %memcpy(dst, src, l, sz); return (); } }}} will fail both ways. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8131> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 17

[GHC] #8123: GHCi warns about -eventlog even though it's sometimes necessary
by GHC 19 Jul '15

19 Jul '15

#8123: GHCi warns about -eventlog even though it's sometimes necessary -------------------------+------------------------------------------------- Reporter: akio | Owner: Type: bug | Status: new Priority: | Milestone: normal | Version: 7.6.3 Component: GHCi | Operating System: Linux Keywords: | Type of failure: Incorrect warning at Architecture: | compile-time x86_64 (amd64) | Test Case: Difficulty: | Blocking: Unknown | Blocked By: | Related Tickets: | -------------------------+------------------------------------------------- If I invoke GHCi as {{{ghci -eventlog}}}, it says: {{{Warning: -debug, -threaded and -ticky are ignored by GHCi}}}. However, the flag does make a difference, in that GHCi will be able to load object files that were compiled using {{{-eventlog}}}. For this reason, I don't think {{{-eventlog}}} should trigger the warning. Also runghc has the same issue. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8123> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #8206: Add support for Portable Native Client
by GHC 05 Jul '15

05 Jul '15

#8206: Add support for Portable Native Client -------------------------------------------------+------------------------- Reporter: guest | Owner: Alex Type: feature request | Sayers Priority: normal | Status: new Component: Compiler | Milestone: Keywords: NaCl PNaCl Portable Native | Version: Client pexe | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Project (more than a week) | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: -------------------------------------------------+------------------------- == Motivation == [http://www.chromium.org/nativeclient/pnacl/building-and-testing-portable- native-client Portable Native Client] (PNaCl) is a feature in Google Chrome which (roughly) allows it to use LLVM bitcode like browsers use javascript today. The bitcode is downloaded, validated, compiled, and (safely) executed, with access to an API for manipulating the DOM of the page which loaded it. The dream: Haskell executed in the browser at near- native speed. Unfortunately, while GHC has an LLVM backend, the bitcode it generates doesn't satisfy the PNaCl validator. == Portable Native Client == The user visits a page with an <embed> tag pointing to a "pexe" - a PNaCl bitcode file. The bitcode is downloaded and streamed first into a code validator, and then into a compiler. The resulting machine code is then executed as a memory-protected process. The [https://developers.google.com /native-client/pepperc/ Pepper API], a library Google provides, provides access to the browser for things like connecting to websockets, playing audio, and communicating with javascript running on the page. PNaCl is present in Chrome 29 (go to chrome://flags to enable it), and will be enabled for arbitrary sources (the "open web") in Chrome 30. Mozilla have announced that they don't intend to support Native Client. == The Easily-Fixed == The pexe bitcode format is a frozen subset of LLVM bitcode. The PNaCl toolchain provides `pnacl-finalize`, which translates LLVM bitcode into pexe bitcode. PNaCl [http://www.chromium.org/nativeclient/pnacl/bitcode- abi##callingconventions only allows] the standard "ccc" calling convention, whereas GHC likes to use its own "cc 10" calling convention. You can make GHC use ccc by using an unregisterised build, for a slow-down factor of about 2. It seems likely that code-next-to-tables will upset the validator. If so, this can be turned off for a small performance hit. == The Not-So-Easily-Fixed == The validator requires that requests for IO are sent to Pepper. PNaCl also requires that you use newlib instead of glibc. Spawning new processes is forbidden (although creating pthreads is fine). It also seems likely that the validator with take umbrage at calls to `mmap`, which the GHC runtime uses to request memory allocation. It seems, therefore, that significant chunks of the RTS will have to be ported. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8206> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #8144: Interface hashes include time stamp of included .h file when CPP is used
by GHC 25 Jun '15

25 Jun '15

#8144: Interface hashes include time stamp of included .h file when CPP is used -------------------------+------------------------------------------------- Reporter: nh2 | Owner: Type: bug | Status: new Priority: | Milestone: normal | Version: 7.6.3 Component: | Operating System: Unknown/Multiple Compiler | Type of failure: Compile-time performance bug Keywords: | Test Case: Architecture: | Blocking: Unknown/Multiple | Difficulty: | Unknown | Blocked By: | Related Tickets: | -------------------------+------------------------------------------------- See https://github.com/nh2/ghc-bug-time-dependent-interface-hashes for a fast test case. Having '''Test.hs''': {{{ {-# LANGUAGE CPP #-} module Test () where }}} and an empty '''cabal_macros.h''', {{{ touch cabal_macros.h ghc -c Test.hs -optP-include -optPcabal_macros.h }}} will generate you a different interface hash ''most of the time'' in '''ghc --show-iface Test.hi'''. It looks like the modification time of the .h file makes it into the interface hash, but only at second resolution, and you get ''compilation IS NOT required'' if you manage to run it twice per second. I think the interface hash should be independent from the time stamp of the header file. This bug can trigger unnecessary recompilation and makes compilations with identical inputs non-deterministic. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8144> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 34

[GHC] #8172: Expose CWD and import search paths in GHCi via new `:show paths` command
by GHC 11 Jun '15

11 Jun '15

#8172: Expose CWD and import search paths in GHCi via new `:show paths` command -------------------------------------------+------------------------------- Reporter: hvr | Owner: hvr Type: feature request | Status: new Priority: normal | Milestone: Component: GHCi | Version: 7.7 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Easy (less than 1 hour) | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: -------------------------------------------+------------------------------- - Currently, GHCi provides a `:cd` command but no respective builtin command to query the currently set path. - Moreover, there's no obvious way to query the currently active `importPaths` dynamic flags field, to find out which `-i` options are currently in effect. The attached patch extends `:show` with a `:show paths` subcommand, printing the CWD and the current value of `importPaths`: {{{ Ok, modules loaded: Main, InteractiveUI, Paths_ghci_ng, GhciMonad, GhciTags. λ> :show paths current working directory: /home/hvr/Haskell/My/ghci-ng module import search paths: dist/build/ghci-ng/ghci-ng-tmp ghc dist/build/autogen λ> :set -i λ> :show paths current working directory: /home/hvr/Haskell/My/ghci-ng module import search paths: none > :cd /tmp Warning: changing directory causes all loaded modules to be unloaded, because the search path has changed. > :show paths current working directory: /tmp module import search paths: none }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8172> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #8122: make binary-dist broken on OS X in HEAD
by GHC 26 May '15

26 May '15

#8122: make binary-dist broken on OS X in HEAD --------------------------------+------------------------------------------ Reporter: danharaj | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Build System | Version: 7.7 Keywords: | Operating System: MacOS X Architecture: x86_64 | Type of failure: Installing GHC failed (amd64) | Test Case: Difficulty: Unknown | Blocking: Blocked By: | Related Tickets: | --------------------------------+------------------------------------------ After a successful configure and make, make binary-dist fails: {{{ ➜ ghc git:(master) make ===--- building phase 0 make -r --no-print-directory -f ghc.mk phase=0 phase_0_builds make[1]: Nothing to be done for `phase_0_builds'. ===--- building phase 1 make -r --no-print-directory -f ghc.mk phase=1 phase_1_builds make[1]: Nothing to be done for `phase_1_builds'. ===--- building final phase make -r --no-print-directory -f ghc.mk phase=final all make[1]: Nothing to be done for `all'. ➜ ghc git:(master) make binary-dist rm -f bindist-list make -r --no-print-directory -f ghc.mk bindist BINDIST=YES "cp" -p ghc/stage1/build/tmp/ inplace/lib/bin/ghc-stage1 cp: ghc/stage1/build/tmp/ is a directory (not copied). make[1]: *** [inplace/lib/bin/ghc-stage1] Error 1 make: *** [binary-dist-prep] Error 2 }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8122> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

[GHC] #8086: Minimal install for GHC
by GHC 04 May '15

04 May '15

#8086: Minimal install for GHC ------------------------------------+--------------------------------- Reporter: gidyn | Owner: Type: feature request | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.6.3 Keywords: | Operating System: Linux Architecture: Unknown/Multiple | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+--------------------------------- Installing GHC on a shared server requires a minimal footprint, without GHCi, profiling, debugging, documentation, or anything else not needed for regular building and installing. And anything stripable should be striped, if it isn't already. It would be great if the configuration script would support a --minimal option. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8086> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3