March 2016 - ghc-tickets

[GHC] #8668: SPECIALIZE silently fails to apply
by GHC 07 Dec '16

07 Dec '16

#8668: SPECIALIZE silently fails to apply ----------------------------------+------------------------------------- Reporter: crockeea | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.6.2 Keywords: | Operating System: Unknown/Multiple Architecture: x86_64 (amd64) | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ----------------------------------+------------------------------------- I have a small example where GHC refuses to specialize a call to `(+)`, compiling with -O2. The two files are Foo.hs (http://lpaste.net/98464) and Main.hs (http://lpaste.net/98342) There seem to be two problems: 1. The active `SPECIALIZE` pragma should be applied, but isn't. This can be seen by comparing the core and runtimes of `fcTest` (slow) vs `vtTest` (fast). I need this version of the pragma in my real code as the phantom type `m` is reified, so I need to specialize the vector code without specifying the phantom type. 2. I can get `fcTest` to run fast if I use the commented-out `SPECIALIZE` pragma instead. However, that pragma seems very straightforward to me (all types are concrete). The docs indicate that GHC should automatically specialize in most cases, why does it not specialize to the commented-out signature automatically? This problem is also posted here: http://stackoverflow.com/questions/21071706/specialization-with- constraints -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8668> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 26

[GHC] #11703: Segmentation fault/internal error: evacuate: strange closure type 15 with GHC.Generics
by GHC 07 Dec '16

07 Dec '16

#11703: Segmentation fault/internal error: evacuate: strange closure type 15 with GHC.Generics -------------------------------------+------------------------------------- Reporter: NathanWaivio | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Runtime | Version: 7.10.3 System | Keywords: Segmentation | Operating System: Unknown/Multiple Fault, Generics, Runtime error | Architecture: x86_64 | Type of failure: Runtime crash (amd64) | Test Case: attached | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Reproduced issue on both Windows-7/GHC-7.10.3 and Linux-4.1.15/GHC-7.8.4 Seems to probably be a problem with GHC.Generics and the runtime. Windows/GHC-7.10.3 states: internal error: evacuate: strange closure type 15 (GHC version 7.10.3 for x86_64_unkown_mingw32) Please report this as a GHC bug: httpL//www.haskell.org/ghc/reportabug This application has requested the Runtime to terminate it in an unusual way. Please contact the application's support team for more information. Linux/GHC-7.8.4 states: Segmentation fault Behavior varies based on CSV input file size, a CSV file with header with 1 to 4 rows works. A CSV with a header and more than 10 rows segmentation faults imediatly. Inbetween 4 to 10 sometimes runs forever, works, or segmentation faults. The test case uses Cassava and GHC.Generics and some derived instances to read a CSV file with a record data type with many fields. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11703> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #10840: Periodic alarm signals can cause a retry loop to get stuck
by GHC 07 Dec '16

07 Dec '16

#10840: Periodic alarm signals can cause a retry loop to get stuck -------------------------------------+------------------------------------- Reporter: Rufflewind | Owner: Type: bug | Status: new Priority: normal | 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: | -------------------------------------+------------------------------------- The [https://ghc.haskell.org/trac/ghc/wiki/Commentary/Rts/Signals periodic alarm signals emitted by the GHC RTS] can [https://github.com/haskell/directory/issues/35 hang the program under certain circumstances]. In particular, any retry-loop of the form {{{#!c while (interruptible_syscall() == FAILED && errno == EINTR); }}} can cause a hang if the syscall takes a long time and cannot be resumed, as it will forced to restart from the beginning each time it gets interrupted. If the syscall takes longer than the interval between successive alarm signals, it will be stuck in this loop forever. This was found to occur with `statfs64` (indirectly called by `realpath`) and `open` on SSHFS on Mac OS X. [https://mail.haskell.org/pipermail/ghc-devs/2015-September/009770.html Using a safe foreign import seems to mitigate the issue], but [https://mail.haskell.org/pipermail/ghc-devs/2015-September/009793.html according to Simon Marlow this is probably an accident]. So far there seems to be no way to guarantee the suspension of signals except through low-level tools like `pthread_setmask`. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10840> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 23

[GHC] #9314: Huge space leak of GHC API 7.8.x
by GHC 07 Dec '16

07 Dec '16

#9314: Huge space leak of GHC API 7.8.x ------------------------------------+------------------------------------- Reporter: kazu-yamamoto | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: GHC API | Version: 7.8.3 Keywords: | Operating System: Unknown/Multiple Architecture: Unknown/Multiple | Type of failure: Other Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | ------------------------------------+------------------------------------- GHC API 7.8.x uses much more memory than GHC API 7.6.x. Attached two files demonstrate this: - A.hs -- Simple program using GHC API (copied from Wiki) - B.hs -- A target file, just hello world You can compile A.hs as follows: {{{ % ghc A.hs -package ghc -package ghc-paths }}} A.hs stays in 10 seconds. So, we can investigate its memory usage with the "top" command.The following is the result: {{{ Mac (64bit) Linux (64bit) GHC 7.6.3: 20MB 4MB GHC 7.8.3: 106MB 13MB }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9314> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 25

[GHC] #8460: Annotation reification with types in TH
by GHC 07 Dec '16

07 Dec '16

#8460: Annotation reification with types in TH -------------------------------------------+------------------------------- Reporter: errge | Owner: Type: feature request | Status: new Priority: normal | Milestone: 7.8.1 Component: Compiler | 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: #8397 | Test Case: | Blocking: -------------------------------------------+------------------------------- The attached patch adds two simple helper functions to support typed reification of annotations. This was not in the original #8397 patch only because the typed branch was not merged at that time. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8460> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #9813: Error when reifying type constructor
by GHC 07 Dec '16

07 Dec '16

#9813: Error when reifying type constructor -------------------------------------+------------------------------------- Reporter: owst | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.8.3 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- The following code works under ghc-7.6.3 and template-haskell-2.8.0.0: {{{#!hs {-# LANGUAGE TemplateHaskell #-} data Huh = ThisDefinitely | UsedToWork constructorNames :: String constructorNames = $(do ty <- reify ''Huh let showCon (NormalC n _) = nameBase n strs = case ty of (TyConI (DataD _ _ _ cons _)) -> map showCon cons return . LitE . StringL $ concat strs) main = putStrLn constructorNames }}} Printing the following at compile time: {{{ [1 of 1] Compiling Main ( Main.hs, Main.o ) Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Loading package array-0.4.0.1 ... linking ... done. Loading package deepseq-1.3.0.1 ... linking ... done. Loading package containers-0.5.0.0 ... linking ... done. Loading package pretty-1.1.1.0 ... linking ... done. Loading package template-haskell ... linking ... done. Linking Main ... }}} and then successfully executing: {{{ $ ./Main ThisDefinitelyUsedToWork }}} However, using ghc 7.8.3/template-haskell-2.9.0.0 I get the following compile error: {{{ [1 of 1] Compiling Main ( Main.hs, Main.o ) Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Loading package array-0.5.0.0 ... linking ... done. Loading package deepseq-1.3.0.2 ... linking ... done. Loading package containers-0.5.5.1 ... linking ... done. Loading package pretty-1.1.1.1 ... linking ... done. Loading package template-haskell ... linking ... done. Main.hs:8:22: ‘Huh’ is not in the type environment at a reify In the splice: $(do { ty <- reify ''Huh; let showCon (NormalC n _) = ... ....; return . LitE . StringL $ concat strs }) }}} Is this expected? I couldn't see anything in the GHC release notes to suggest this should no longer work. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9813> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 46

[GHC] #11757: Turn on (and require) C99/C11 mode by default
by GHC 07 Dec '16

07 Dec '16

#11757: Turn on (and require) C99/C11 mode by default -------------------------------------+------------------------------------- Reporter: hvr | Owner: Type: task | Status: new Priority: normal | Milestone: 8.2.1 Component: Build System | Version: 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: -------------------------------------+------------------------------------- GCC 5.x started defaulting to C99 (or rather gnu99 mode), while Clang has been defaulting to C99 (or even C11) for a bit longer, as has Apple's patched GCC. So we should try to consistently turn on (if needed) C99 (or C11) support for all compilers supported by GHC, and adapt the code-base to expect C99 compliance (which probably allows us to get rid of a couple of compat-CPP and make the C code more idiomatic and readable in the process...). GNU Autoconf provides a `AC_PROG_CC_C99` which figures out what's needed to have the compiler turn on C99 support, but we need to integrate that with out Requiring C99 will effectively rule out GCC 3.x (which is old enough that we can drop support for that), and possibly also GCC < 4.3 at some point (for reference, even Debian 6 Squeeze didn't have GCC 4.2 anymore), see also https://gcc.gnu.org/c99status.html -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11757> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #11493: Merge compact normal forms
by GHC 07 Dec '16

07 Dec '16

#11493: Merge compact normal forms -------------------------------------+------------------------------------- Reporter: bgamari | Owner: ezyang Type: task | Status: new Priority: normal | Milestone: 8.2.1 Component: | Version: 7.10.3 libraries/compact | 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: -------------------------------------+------------------------------------- Merge Yang and Campagna's [[Compact Normal Forms work|http://ezyang.com/compact.html]] There are several pieces to this, * Types and construction logic for compact regions (see Phab:D1264) * Serialization and fixup logic (currently also in Phab:D1264) * Striped allocator for heap sharing between processes (allocator in Phab:D1434, CNF support in Phab:D1435). -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11493> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 3

[GHC] #11557: Unbundle Cabal from GHC
by GHC 07 Dec '16

07 Dec '16

#11557: Unbundle Cabal from GHC -------------------------------------+------------------------------------- Reporter: ezyang | Owner: Type: task | Status: new Priority: normal | Milestone: 8.2.1 Component: libraries | Version: 8.1 (other) | 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: -------------------------------------+------------------------------------- Recently, Duncan made it so that GHC proper does not depend on Cabal (so it is just ghc-pkg that is the user-facing executable which links against Cabal). We should now seriously consider unbundling Cabal from GHC, so that the default global database we provide does NOT include Cabal. Pros: * Distributions will be more likely to take point version releases to Cabal, as they no longer have to finesse updating Cabal without updating GHC as they have now * Stack is (improperly) coupling the version of Cabal they build with the release of GHC; while they should fix this, unbundling Cabal would also give them more flexibility with picking LTS packages. Cons: * Bootstrapping Cabal/cabal-install becomes modestly harder. Fortunately, cabal-install is already pretty obnoxious to bootstrap, so SOP is to just distribute binaries for this, in which case things are as easy as before. * We wouldn't be strictly adhering to the Cabal spec, which requires that the compiler always be able to build the Setup executable. * ghc-pkg would have to be statically linked -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11557> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #10383: AArch64: get GHC Calling convention working
by GHC 07 Dec '16

07 Dec '16

#10383: AArch64: get GHC Calling convention working -------------------------------------+------------------------------------- Reporter: erikd | Owner: erikd Type: feature | Status: new request | Milestone: 7.12.1 Priority: normal | Version: 7.11 Component: Compiler | Operating System: Unknown/Multiple Keywords: | Type of failure: Building GHC Architecture: aarch64 | failed Test Case: | Blocked By: Blocking: | Related Tickets: Differential Revisions: | -------------------------------------+------------------------------------- Building on AArch64 via the LLVM abckend has been working for some time. Its now time to get `("Unregisterised", "NO")` working as well. Enabling GHC calling convention can be done with this patch to `configure.ac`: {{{ diff --git a/configure.ac b/configure.ac index d5d9ab3..a11e5af 100644 --- a/configure.ac +++ b/configure.ac @@ -241,7 +241,7 @@ AC_SUBST(SOLARIS_BROKEN_SHLD) dnl ** Do an unregisterised build? dnl -------------------------------------------------------------- case "$HostArch" in - i386|x86_64|powerpc|arm) + i386|x86_64|powerpc|arm|aarch64) UnregisterisedDefault=NO ;; *) }}} but when building it this way, the stage2 compiler dies will `Illegal instructon`. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10383> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 41