ghc-hetmet.git
16 years agoGenerate conversion from PRepr to original type
Roman Leshchinskiy [Wed, 22 Aug 2007 06:18:25 +0000 (06:18 +0000)]
Generate conversion from PRepr to original type

16 years agoRefactoring
Roman Leshchinskiy [Wed, 22 Aug 2007 05:41:31 +0000 (05:41 +0000)]
Refactoring

16 years agoGenerate conversion to PRepr during vectorisation
Roman Leshchinskiy [Wed, 22 Aug 2007 05:09:27 +0000 (05:09 +0000)]
Generate conversion to PRepr during vectorisation

16 years agoUse NDP's own (:*:) and (:+:)
Roman Leshchinskiy [Wed, 22 Aug 2007 04:46:34 +0000 (04:46 +0000)]
Use NDP's own (:*:) and (:+:)

16 years agoUtility functions for vectorisation
Roman Leshchinskiy [Wed, 22 Aug 2007 03:44:11 +0000 (03:44 +0000)]
Utility functions for vectorisation

16 years agoMore refactoring
Roman Leshchinskiy [Wed, 22 Aug 2007 02:49:23 +0000 (02:49 +0000)]
More refactoring

16 years agoRefactoring
Roman Leshchinskiy [Wed, 22 Aug 2007 02:32:30 +0000 (02:32 +0000)]
Refactoring

16 years agoAdd built-ins to vectorisation monad
Roman Leshchinskiy [Wed, 22 Aug 2007 02:11:33 +0000 (02:11 +0000)]
Add built-ins to vectorisation monad

16 years agoRefactoring
Roman Leshchinskiy [Wed, 22 Aug 2007 02:08:14 +0000 (02:08 +0000)]
Refactoring

16 years agoVectorise nullary constructors correctly
Roman Leshchinskiy [Wed, 22 Aug 2007 01:59:30 +0000 (01:59 +0000)]
Vectorise nullary constructors correctly

16 years agoRename vectorisation-related stuff
Roman Leshchinskiy [Wed, 22 Aug 2007 01:29:07 +0000 (01:29 +0000)]
Rename vectorisation-related stuff

16 years agoGenerate PRepr instances during vectorisation
Roman Leshchinskiy [Wed, 22 Aug 2007 01:22:36 +0000 (01:22 +0000)]
Generate PRepr instances during vectorisation

16 years agoUse functions from BuildTyCl for vectorisation
Roman Leshchinskiy [Tue, 21 Aug 2007 07:17:09 +0000 (07:17 +0000)]
Use functions from BuildTyCl for vectorisation

16 years agoAdd a couple of built-ins to the vectorisation monad
Roman Leshchinskiy [Tue, 21 Aug 2007 04:00:34 +0000 (04:00 +0000)]
Add a couple of built-ins to the vectorisation monad

16 years agoGet the PA dictionary for closures from the right module
Roman Leshchinskiy [Tue, 21 Aug 2007 02:45:29 +0000 (02:45 +0000)]
Get the PA dictionary for closures from the right module

16 years agoRemove packPA and combinePA from vectorisation built-ins for now
Roman Leshchinskiy [Tue, 21 Aug 2007 02:44:46 +0000 (02:44 +0000)]
Remove packPA and combinePA from vectorisation built-ins for now

16 years agoAdd NDP-related PrelNames
Roman Leshchinskiy [Tue, 21 Aug 2007 02:26:45 +0000 (02:26 +0000)]
Add NDP-related PrelNames

16 years agoPrint infix function definitions correctly in HsSyn
simonpj@microsoft.com [Wed, 22 Aug 2007 23:03:24 +0000 (23:03 +0000)]
Print infix function definitions correctly in HsSyn

16 years agoTrim imports
simonpj@microsoft.com [Wed, 22 Aug 2007 23:02:55 +0000 (23:02 +0000)]
Trim imports

16 years agoAdd a bindisttest directory
Ian Lynagh [Wed, 22 Aug 2007 20:20:40 +0000 (20:20 +0000)]
Add a bindisttest directory
After
    make binary-dist stage=2
you can run make in bindisttest/ to test it

16 years agoInclude package.mk in target.mk, in the bindists's mk/
Ian Lynagh [Wed, 22 Aug 2007 20:10:49 +0000 (20:10 +0000)]
Include package.mk in target.mk, in the bindists's mk/

16 years agoBuild fix: modBreaks -> modBreaks_decls modbreaks
Michael D. Adams [Wed, 22 Aug 2007 18:36:32 +0000 (18:36 +0000)]
Build fix: modBreaks -> modBreaks_decls modbreaks

16 years agoUse unsafeDupableInterleaveIO in UniqSupply
Ian Lynagh [Wed, 22 Aug 2007 14:34:22 +0000 (14:34 +0000)]
Use unsafeDupableInterleaveIO in UniqSupply
We'll need to change this if we ever use threads in GHC, but for now it
drops mkSplitUniqSupply from 35% time down below 0.1% when compiling
J.hs from #1136.

16 years agoLink ghc-pkg and hpc with containers if building ghc 6.7 or newer
Esa Ilari Vuokko [Tue, 21 Aug 2007 21:04:59 +0000 (21:04 +0000)]
Link ghc-pkg and hpc with containers if building ghc 6.7 or newer

16 years agomakeRelativeRefTo checks tablesNextToCode, so we don't have to before using it
Ian Lynagh [Mon, 20 Aug 2007 10:30:58 +0000 (10:30 +0000)]
makeRelativeRefTo checks tablesNextToCode, so we don't have to before using it

16 years agoAllow redirection of -ddump-* to file
Ben.Lippmeier@anu.edu.au [Tue, 21 Aug 2007 16:31:10 +0000 (16:31 +0000)]
Allow redirection of -ddump-* to file
Whilst compiling Main.hs with -ddump-stg, ddump-asm and friends
you can how add -ddump-to-file and you'll get the dumps redirected
to Main.dump-stg, Main.dump-asm etc.

16 years agoInstrument linear register allocator.
Ben.Lippmeier@anu.edu.au [Tue, 21 Aug 2007 12:08:55 +0000 (12:08 +0000)]
Instrument linear register allocator.

Linear register allocator now keeps track of why it spilled certain
registers. It had to be refactored somewhat to thread the information
through, but the algorithm is unchanged.

16 years agoNCG debugging cleanup
Ben.Lippmeier@anu.edu.au [Mon, 20 Aug 2007 10:18:59 +0000 (10:18 +0000)]
NCG debugging cleanup

16 years agoAdd vreg-conflicts and vreg-conflict-lifetimes to drop-asm-stats
Ben.Lippmeier@anu.edu.au [Mon, 20 Aug 2007 09:39:19 +0000 (09:39 +0000)]
Add vreg-conflicts and vreg-conflict-lifetimes to drop-asm-stats

16 years agoAdd vreg-population-lifetimes to drop-asm-stats
Ben.Lippmeier@anu.edu.au [Fri, 17 Aug 2007 16:22:54 +0000 (16:22 +0000)]
Add vreg-population-lifetimes to drop-asm-stats

16 years agoAdd dumping of native code gen stats to file.
Ben.Lippmeier@anu.edu.au [Fri, 17 Aug 2007 14:37:30 +0000 (14:37 +0000)]
Add dumping of native code gen stats to file.

Compiling module Foo with -ddrop-asm-stats produces a file
called Foo.dump-asm-stats which will contain increasingly
interesting information.

16 years agoRefactor dumping of register allocator statistics.
Ben.Lippmeier@anu.edu.au [Fri, 17 Aug 2007 12:15:57 +0000 (12:15 +0000)]
Refactor dumping of register allocator statistics.

16 years agoLinker.lhs to prefer DSO when it's build as dynamic library
Clemens Fruhwirth [Tue, 21 Aug 2007 10:34:51 +0000 (10:34 +0000)]
Linker.lhs to prefer DSO when it's build as dynamic library

16 years agoA partial attempt to improve :stepover
Pepe Iborra [Wed, 22 Aug 2007 09:08:44 +0000 (09:08 +0000)]
A partial attempt to improve :stepover

  With this patch, :stepover can effectively appear to step over recursive calls and
  calls to locally bound functions (in a where clause).

  However, when we run out of ticks in the current expression,
  the illusion vanishes and laziness brings us to the body of the last function
  we "stepped over".
  This is not desired at all, it is potentially very confusing.
  As a countermeasure, when this happens :stepover emits a warning

     "Warning: no more breakpoints in this function body, switching to :step"

16 years agoBetter document :stepover and its limitations
Pepe Iborra [Wed, 22 Aug 2007 08:42:54 +0000 (08:42 +0000)]
Better document :stepover and its limitations

:stepover only works lexically locally, in the context of the
current expression. I have tried to make this point clear
in the users guide with an example.

16 years agoSmall rearrangements
Pepe Iborra [Tue, 21 Aug 2007 12:15:16 +0000 (12:15 +0000)]
Small rearrangements

16 years agoExtend ModBreaks with the srcspan's of the enclosing expressions
Pepe Iborra [Tue, 21 Aug 2007 12:03:50 +0000 (12:03 +0000)]
Extend ModBreaks with the srcspan's of the enclosing expressions

16 years agoDoc for new manifest-related flags
Simon Marlow [Wed, 22 Aug 2007 08:40:26 +0000 (08:40 +0000)]
Doc for new manifest-related flags

16 years agoFIX #1271: create manifests, and embed them in executables on Windows
Simon Marlow [Fri, 10 Aug 2007 08:47:53 +0000 (08:47 +0000)]
FIX #1271: create manifests, and embed them in executables on Windows

We have 4 new flags:

  -fno-gen-manifest
     suppresses creation of foo.exe.manifest

  -fno-embed-manifest
     suppresses embedding of the manifest in the executable

  -pgmwindres
     specify a program to use instead of windres

  -optwindres
     extra options to pass to windres

"windres" is now copied from mingw and included in a binary
distribution.

16 years agoFIX #1427, #1569: gcc 4.2.x needs -fno-toplevel-reorder
Simon Marlow [Tue, 21 Aug 2007 15:15:53 +0000 (15:15 +0000)]
FIX #1427, #1569: gcc 4.2.x needs -fno-toplevel-reorder
We now have various flags that need to be passed to gcc, but only if
the local gcc version supports them.  So instead of wiring this
knowledge into ghc when it is built, we now put these "extra gcc
flags" into a text file in $libdir, extra-gcc-flags, which is created
by configure.  It is also created by the configure script of a binary
distribution, so a binary dist is independent of the gcc version used
to build it.

Hopefully I got the binary-dist stuff right, but binary dists aren't
currently working so we'll need to come back and check this.

16 years agowe need to add utils/pwd to a binary distribution
Simon Marlow [Tue, 21 Aug 2007 15:12:05 +0000 (15:12 +0000)]
we need to add utils/pwd to a binary distribution

16 years agobinary-dist: remove comment from the middle of a shell command
Simon Marlow [Tue, 21 Aug 2007 14:46:41 +0000 (14:46 +0000)]
binary-dist: remove comment from the middle of a shell command
This appears to cause trouble in at least some shells

16 years agoAdd R_X86_64_PC64 relocation support to the Linker
Clemens Fruhwirth [Wed, 8 Aug 2007 14:57:00 +0000 (14:57 +0000)]
Add R_X86_64_PC64 relocation support to the Linker

16 years agoAlso build gmp shared if neccessary
Clemens Fruhwirth [Fri, 10 Aug 2007 06:30:40 +0000 (06:30 +0000)]
Also build gmp shared if neccessary

16 years agoAdd __PIC__ define for -fPIC compiled C code
Clemens Fruhwirth [Tue, 7 Aug 2007 16:06:16 +0000 (16:06 +0000)]
Add __PIC__ define for -fPIC compiled C code

16 years agoRemove unused -optdll option.
Clemens Fruhwirth [Tue, 7 Aug 2007 14:41:24 +0000 (14:41 +0000)]
Remove unused -optdll option.

16 years agoLink Windows DLLs and OSX dylibs correctly in linkDynLib
Clemens Fruhwirth [Tue, 7 Aug 2007 14:40:18 +0000 (14:40 +0000)]
Link Windows DLLs and OSX dylibs correctly in linkDynLib

16 years agoAdd graph coloring register allocator.
Ben.Lippmeier@anu.edu.au [Tue, 14 Aug 2007 10:36:23 +0000 (10:36 +0000)]
Add graph coloring register allocator.

Refactored linear allocator into separate liveness annotation and allocation stages.
Added graph coloring allocator, use -fregs-graph to enable.
  New dump flags are
    -ddump-asm-native          -- output of cmm -> native transform.
    -ddump-asm-liveness        -- code annotated with register liveness info
    -ddump-asm-coalesce        -- output of register move coalescing
                                    (this is a separate pass when using the coloring allocator)
                                    (this could change in the future)
    -ddump-asm-regalloc        -- code after register allocation
    -ddump-asm-regalloc-stages -- blocks after each build/spill stage of coloring allocator
    -ddump-asm-conflicts       -- a global register liveness graph in graphviz format

The new register allocator will allocate some registers, but it's not
quite ready for prime-time yet. The spill code generator needs some work...

16 years agoMake ghc-inplace return GHC's exitcode on Windows
Ian Lynagh [Tue, 21 Aug 2007 10:10:28 +0000 (10:10 +0000)]
Make ghc-inplace return GHC's exitcode on Windows

16 years agoRemove odd recursive call in the driver/ghci Makefile
Ian Lynagh [Tue, 21 Aug 2007 08:59:37 +0000 (08:59 +0000)]
Remove odd recursive call in the driver/ghci Makefile
I'm not sure why this was here, but it makes ghci.o get built twice on
Windows, which can cause the build to fail if it tries to link it
in the middle.

16 years agoFix quoting of double quotes in the Makefile on Windows
Ian Lynagh [Mon, 20 Aug 2007 23:52:21 +0000 (23:52 +0000)]
Fix quoting of double quotes in the Makefile on Windows

16 years agoFix ghc-inplace.c on Windows; new code based on ghci.c
Ian Lynagh [Mon, 20 Aug 2007 22:27:01 +0000 (22:27 +0000)]
Fix ghc-inplace.c on Windows; new code based on ghci.c

16 years agoFIX part of #1372, improvements to the recompilation checker
Simon Marlow [Fri, 3 Aug 2007 10:20:45 +0000 (10:20 +0000)]
FIX part of #1372, improvements to the recompilation checker
This patch adds a check that each imported module is listed in
the dependencies from the previous interface.  It catches the
following cases and forces recompilation:
   - an exposed package has been upgraded
   - we are compiling with different package flags
   - a home module that was shadowing a package module has been removed
   - a new home module has been added that shadows a package module

I haven't yet added the package timestamping as described in #1372.

16 years agoput CmmReturnInfo into a CmmCall (and related types)
Norman Ramsey [Mon, 20 Aug 2007 19:54:41 +0000 (19:54 +0000)]
put CmmReturnInfo into a CmmCall (and related types)

16 years agoAdd a paragraph break
simonpj@microsoft.com [Mon, 20 Aug 2007 21:02:22 +0000 (21:02 +0000)]
Add a paragraph break

16 years agoannotate C-- calls that do not return
Norman Ramsey [Mon, 20 Aug 2007 16:41:05 +0000 (16:41 +0000)]
annotate C-- calls that do not return
  * The correct definition of C-- requires that a procedure not
    'fall off the end'.  The 'never returns' annotation tells us
    if a (foreign) call is not going to return.

    Validated!

16 years ago(short) Documentation for :stepover in the Users Guide
Pepe Iborra [Mon, 20 Aug 2007 14:20:35 +0000 (14:20 +0000)]
(short) Documentation for :stepover in the Users Guide

16 years agoMake :stepover <expr> work like :step
Pepe Iborra [Mon, 20 Aug 2007 13:35:46 +0000 (13:35 +0000)]
Make :stepover <expr> work like :step

16 years agoCheck for correct number of parameters in type instances
Manuel M T Chakravarty [Sun, 19 Aug 2007 11:38:25 +0000 (11:38 +0000)]
Check for correct number of parameters in type instances

16 years agoFIX: family instances for infix type constructors
Manuel M T Chakravarty [Sun, 19 Aug 2007 11:36:47 +0000 (11:36 +0000)]
FIX: family instances for infix type constructors
- Fixed test "Infix" in the testsuite, which Roman added

16 years agoATs inherit kind annotations from class partameters
Manuel M T Chakravarty [Sun, 19 Aug 2007 07:23:40 +0000 (07:23 +0000)]
ATs inherit kind annotations from class partameters
- Fixes the test "Kind" in the testsuite, which Roman added

16 years agoUpdated comments
Manuel M T Chakravarty [Sun, 19 Aug 2007 05:56:18 +0000 (05:56 +0000)]
Updated comments

16 years agoDocument the compiler --info flag
Ian Lynagh [Sun, 19 Aug 2007 00:05:10 +0000 (00:05 +0000)]
Document the compiler --info flag

16 years agoImprove runghc's argument handling
Ian Lynagh [Sun, 19 Aug 2007 00:04:36 +0000 (00:04 +0000)]
Improve runghc's argument handling

16 years agoAdd some runghc docs to the users guide
Ian Lynagh [Sun, 19 Aug 2007 00:02:58 +0000 (00:02 +0000)]
Add some runghc docs to the users guide

16 years agoMake runghc warning-free and turn -Wall on in the Makefile
Ian Lynagh [Sat, 18 Aug 2007 22:42:46 +0000 (22:42 +0000)]
Make runghc warning-free and turn -Wall on in the Makefile

16 years agoMake the runghc code prettier and fix some warnings
Ian Lynagh [Sat, 18 Aug 2007 22:40:21 +0000 (22:40 +0000)]
Make the runghc code prettier and fix some warnings

16 years agoImplement --info, and change how can_split is defined
Ian Lynagh [Sat, 18 Aug 2007 22:37:58 +0000 (22:37 +0000)]
Implement --info, and change how can_split is defined
We now get the value of can_split from the SplitObjs setting in the
build system, rather than working out whether it should be possible
based on the platform.

16 years agoMake "runghc -f path-to-ghc Main.hs" work
Magnus Jonsson [Sun, 29 Jul 2007 21:02:30 +0000 (21:02 +0000)]
Make "runghc -f path-to-ghc Main.hs" work

16 years agoUpdate user guide bug list
Ian Lynagh [Sat, 18 Aug 2007 20:13:47 +0000 (20:13 +0000)]
Update user guide bug list
Remove the bug that claims GHC treats source files as ISO-8859-1 rather
than UTF8.

16 years agoUse the qualified name for catch
Ian Lynagh [Sat, 18 Aug 2007 19:44:36 +0000 (19:44 +0000)]
Use the qualified name for catch
Unqualified it's ambiguous with some configurations.

16 years agoAdd some invariants to OrdList
Ian Lynagh [Sat, 18 Aug 2007 13:21:02 +0000 (13:21 +0000)]
Add some invariants to OrdList
isNilOL is now constant time, rather than possibly having to walk a
tree of Two's. Compiling J.hs from trac #1136 now makes 10302 isNilOL
calls rather than 50050152. It's gone from 10.8% time to being unlisted
(i.e. <= 0.1%).

16 years agoFix trac #1565: :r DoesNotExist needs to remove any let-bound values
Ian Lynagh [Sat, 18 Aug 2007 15:22:50 +0000 (15:22 +0000)]
Fix trac #1565: :r DoesNotExist needs to remove any let-bound values

16 years agoFix passing double quotes to GHC on Windows
Ian Lynagh [Sat, 18 Aug 2007 13:54:31 +0000 (13:54 +0000)]
Fix passing double quotes to GHC on Windows

16 years agoMake :i (->) work; fixes trac #1587
Ian Lynagh [Fri, 17 Aug 2007 22:17:33 +0000 (22:17 +0000)]
Make :i (->) work; fixes trac #1587

16 years agoRemove dupliate "io (revertCAFs)" and discardActiveBreakPoints calls
Ian Lynagh [Fri, 17 Aug 2007 21:23:20 +0000 (21:23 +0000)]
Remove dupliate "io (revertCAFs)" and discardActiveBreakPoints calls

16 years agoAdd a comment that -Onot is deprecated
Ian Lynagh [Fri, 17 Aug 2007 20:36:31 +0000 (20:36 +0000)]
Add a comment that -Onot is deprecated

16 years agoAdd a GhcDebugged build setting, for whether GHC is linked with -debug
Ian Lynagh [Fri, 17 Aug 2007 20:36:07 +0000 (20:36 +0000)]
Add a GhcDebugged build setting, for whether GHC is linked with -debug

16 years agoMake ghc-inplace understand -v
Ian Lynagh [Fri, 17 Aug 2007 17:59:44 +0000 (17:59 +0000)]
Make ghc-inplace understand -v
If the first arg is -v, make ghc-inplace print out the real GHC path
and extra flags it adds

16 years agoDon't freeProfiling1 until after calling reportCCSProfiling
Ian Lynagh [Fri, 17 Aug 2007 15:02:05 +0000 (15:02 +0000)]
Don't freeProfiling1 until after calling reportCCSProfiling
Fixes usage of free'd memory

16 years agoFix description and type profiling
Ian Lynagh [Fri, 17 Aug 2007 14:45:31 +0000 (14:45 +0000)]
Fix description and type profiling
Consistently make the type and description in the info table an offset
or a pointer, depending on whether tables are next to code or not.

16 years agoCompute max_era more simply
Ian Lynagh [Thu, 16 Aug 2007 23:05:50 +0000 (23:05 +0000)]
Compute max_era more simply

16 years agoDocument +RTS --info, and make it a Read'able Haskell value
Ian Lynagh [Thu, 16 Aug 2007 18:32:59 +0000 (18:32 +0000)]
Document +RTS --info, and make it a Read'able Haskell value

16 years agoFix the threaded RTS on Windows
Ian Lynagh [Thu, 16 Aug 2007 15:45:16 +0000 (15:45 +0000)]
Fix the threaded RTS on Windows
When calling EnterCriticalSection and LeaveCriticalSection from C--
code, we go via wrappers which use ccall (rather than stdcall).

16 years agoAdd newline to end of Hpc.hs
Ian Lynagh [Thu, 16 Aug 2007 15:09:00 +0000 (15:09 +0000)]
Add newline to end of Hpc.hs

16 years agoConvert the remaining _scc_s in the GHC source to pragmas
Ian Lynagh [Thu, 16 Aug 2007 15:07:08 +0000 (15:07 +0000)]
Convert the remaining _scc_s in the GHC source to pragmas

16 years agoOptimise Digraph.postOrd, used when finding strongly connected components
Ian Lynagh [Thu, 16 Aug 2007 02:15:14 +0000 (02:15 +0000)]
Optimise Digraph.postOrd, used when finding strongly connected components

16 years agoReplace _scc_'s in the GHC source with SCC pragmas
Ian Lynagh [Thu, 16 Aug 2007 01:22:24 +0000 (01:22 +0000)]
Replace _scc_'s in the GHC source with SCC pragmas

16 years agoAdd a deprecated warning for _scc_
Ian Lynagh [Thu, 16 Aug 2007 01:08:40 +0000 (01:08 +0000)]
Add a deprecated warning for _scc_

16 years agoMake dependencies of source files on Makefiles disableable
Ian Lynagh [Thu, 16 Aug 2007 01:07:40 +0000 (01:07 +0000)]
Make dependencies of source files on Makefiles disableable
Avoids lots of unnecessary recmopilation when you're fiddling with
Makefiles.

16 years agoCosmetic changes only
Ian Lynagh [Wed, 15 Aug 2007 23:28:21 +0000 (23:28 +0000)]
Cosmetic changes only

16 years agoStop ghc-6.7 linking in ghc-inplace.o
Ian Lynagh [Wed, 15 Aug 2007 17:36:49 +0000 (17:36 +0000)]
Stop ghc-6.7 linking in ghc-inplace.o

16 years agoUse a C program for ghc-inplace
Ian Lynagh [Wed, 15 Aug 2007 17:27:07 +0000 (17:27 +0000)]
Use a C program for ghc-inplace

16 years agoDocument :stepover in ghci help
Pepe Iborra [Thu, 16 Aug 2007 09:49:44 +0000 (09:49 +0000)]
Document :stepover in ghci help

16 years agoSteal the bold trick from :list for :history
Pepe Iborra [Wed, 15 Aug 2007 16:11:35 +0000 (16:11 +0000)]
Steal the bold trick from :list for :history

16 years agoTeach :history to show the name of the enclosing declaration
Pepe Iborra [Wed, 15 Aug 2007 16:11:30 +0000 (16:11 +0000)]
Teach :history to show the name of the enclosing declaration
 together with src locs

Purely for convenience and user friendliness

16 years agoGeneralize some code dealing with SrcSpan sorting
Pepe Iborra [Tue, 14 Aug 2007 16:19:53 +0000 (16:19 +0000)]
Generalize some code dealing with SrcSpan sorting
and the subspan relation, and move it to the SrcLoc module

16 years agoA new :stepover command for the debugger
Pepe Iborra [Thu, 9 Aug 2007 20:11:49 +0000 (20:11 +0000)]
A new :stepover command for the debugger

  Step from statement to statement without leaving the block.
  Tries to do the sensible thing when used on expressions.

      The idea is to:
      1 - Step to the next breakpoint and examine the srcloc
      2 - If it is contained in the same statement block as we were,
          then stop and give control to the user,
        else continue to the next breakpoint
      3 - Repeat from 1. If we reach the end of the statement block,
          i.e. no more ticks in this expression after the current one,
          then step normally.

  Replace statement block with 'declaration block' (of an expression) in the
  pseudo algo. above.
  Let's see how well this idea works in practice...

16 years ago:step does not delete the :history anymore, and now it logs like :trace
Pepe Iborra [Thu, 9 Aug 2007 20:11:21 +0000 (20:11 +0000)]
:step does not delete the :history anymore, and now it logs like :trace

16 years agoFunctor instance for the GHCi monad
Pepe Iborra [Thu, 9 Aug 2007 16:09:33 +0000 (16:09 +0000)]
Functor instance for the GHCi monad