[commit: ghc] master: spelling in doc (a38b832)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/a38b8327e969f8e0e6b07c36be5c47709871f158 --- commit a38b8327e969f8e0e6b07c36be5c47709871f158 Author: Gabor Greif ggr...@gmail.com Date: Wed Dec 19 13:37:08 2012 +0100 spelling in doc --- docs/users_guide/ghci.xml |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/users_guide/ghci.xml b/docs/users_guide/ghci.xml index 3b1f00e..f3e93a3 100644 --- a/docs/users_guide/ghci.xml +++ b/docs/users_guide/ghci.xml @@ -2410,8 +2410,8 @@ Prelude :. cmds.ghci and (b) all the other things mentioned in the instance are in scope (either qualified or otherwise) as a result of a literal:load/literal or literal:module/literal commands. /para - The command literal:info!/literal works in a similar fasion - but it removes restriction (b), showing all instancds that are in + The command literal:info!/literal works in a similar fashion + but it removes restriction (b), showing all instances that are in scope and mention replaceablename/replaceable in their head. /listitem /varlistentry ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typo (e3aa165)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/e3aa16595f6240bff60fdde6579c89e469f25b4f --- commit e3aa16595f6240bff60fdde6579c89e469f25b4f Author: Gabor Greif ggr...@gmail.com Date: Fri Dec 14 10:00:55 2012 +0100 typo --- includes/rts/storage/ClosureMacros.h |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/includes/rts/storage/ClosureMacros.h b/includes/rts/storage/ClosureMacros.h index dd5f428..73725e1 100644 --- a/includes/rts/storage/ClosureMacros.h +++ b/includes/rts/storage/ClosureMacros.h @@ -1,6 +1,6 @@ /* * - * (c) The GHC Team, 1998-2004 + * (c) The GHC Team, 1998-2012 * * Macros for building and manipulating closures * @@ -26,7 +26,7 @@ info pointerThe first word of the closure. Might point to either the end or the beginning of the info table, depending on whether we're using -the mini interpretter or not. GET_INFO(c) +the mini interpreter or not. GET_INFO(c) retrieves the info pointer of a closure. info table The info table structure associated with a ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: pgj (x86 FreeBSD HEAD), build 887, Failure
The latter looks like being emitted by the PowerPC backend. Cheers, Gabor On 12/13/12, Ian Lynagh i...@well-typed.com wrote: On Wed, Dec 12, 2012 at 07:27:43PM -0800, Builder wrote: rts/AdjustorAsm.S:127:0: Error: no such instruction: `load r3,HPTR_OFF(r2)' rts/AdjustorAsm.S:129:0: Error: no such instruction: `load r12,WPTR_OFF(r2)' rts/AdjustorAsm.S:131:0: Error: no such instruction: `mtctr r12' Does anyone know what's going on here? Thanks Ian ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typo (5d0d1d9)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/5d0d1d9cbe2fd0c0a5e4f4a979c8e46ff51ea6fa
---
commit 5d0d1d9cbe2fd0c0a5e4f4a979c8e46ff51ea6fa
Author: Gabor Greif ggr...@gmail.com
Date: Wed Dec 12 10:26:49 2012 +0100
typo
---
compiler/nativeGen/AsmCodeGen.lhs |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/compiler/nativeGen/AsmCodeGen.lhs
b/compiler/nativeGen/AsmCodeGen.lhs
index 05f7c3a..1c0b0a1 100644
--- a/compiler/nativeGen/AsmCodeGen.lhs
+++ b/compiler/nativeGen/AsmCodeGen.lhs
@@ -388,7 +388,7 @@ cmmNativeGens' ncgImpl us (cmm : cmms) (h, dflags, (impAcc,
profAcc)) count
let !count' = count + 1
--- force evaulation all this stuff to avoid space leaks
+-- force evaluation all this stuff to avoid space leaks
{-# SCC seqString #-} evaluate $ seqString (showSDoc dflags $ vcat $
map ppr imports)
cmmNativeGens' ncgImpl
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
RFC: Singleton equality witnesses
Hi all! After encouragement from Iavor on G+, here is a patch that implements a class method for singleton type equality witnesses in a generic way. Please comment on two things: - is this a good approach? - how can we avoid abuse of SingEq (as it is type polymorphic, can this harm?) - (possibly) bikeshedding on names. Cheers and thanks, Gabor TypeLits.hs.patch Description: Binary data ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typo (d53e222)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/d53e222b16c5dc6971e6173f8d9e5dbced68265d --- commit d53e222b16c5dc6971e6173f8d9e5dbced68265d Author: Gabor Greif ggr...@gmail.com Date: Fri Nov 30 15:18:43 2012 +0100 typo --- compiler/iface/IfaceSyn.lhs |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/compiler/iface/IfaceSyn.lhs b/compiler/iface/IfaceSyn.lhs index 06c7b67..2f827ca 100644 --- a/compiler/iface/IfaceSyn.lhs +++ b/compiler/iface/IfaceSyn.lhs @@ -213,7 +213,7 @@ data IfaceIdInfo -- (In earlier GHCs we used to drop IdInfo immediately on reading, -- but we do not do that now. Instead it's discarded when the -- ModIface is read into the various decl pools.) --- * The version comparsion sees that new (=NoInfo) differs from old (=HasInfo *) +-- * The version comparison sees that new (=NoInfo) differs from old (=HasInfo *) -- and so gives a new version. data IfaceInfoItem ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: Proposal: New procedures for handling packages in a GHC tree
Will this work with the github mirror (https://github.com/ghc), as the module names seem to be different from those on darcs.haskell.org? Just a quick check before I destroy my local repo with a ./sync-all pull :-) Cheers, Gabor On 11/25/12, Ian Lynagh i...@well-typed.com wrote: On Fri, Nov 02, 2012 at 08:10:04PM +, Ian Lynagh wrote: We have been working on a new way in which we handle the repositories that make up a GHC tree, in order to makes the process smoother for all parties. Our proposal is here: http://hackage.haskell.org/trac/ghc/wiki/WorkingConventions/Repositories We've now gone ahead with this proposal. We've stuck with only converting repositories with upstreams to submodules for now, but we may expand them to cover all repositories in the future. Thanks Ian ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typos in note (3a82618)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/3a8261827fd9e66251a4a8bf91a22ae075bfdcb9
---
commit 3a8261827fd9e66251a4a8bf91a22ae075bfdcb9
Author: Gabor Greif ggr...@gmail.com
Date: Mon Nov 26 16:04:33 2012 +0100
typos in note
---
compiler/types/Kind.lhs |6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/compiler/types/Kind.lhs b/compiler/types/Kind.lhs
index 6ce2dd9..2041508 100644
--- a/compiler/types/Kind.lhs
+++ b/compiler/types/Kind.lhs
@@ -1,5 +1,5 @@
%
-% (c) The University of Glasgow 2006
+% (c) The University of Glasgow 2006-2012
%
\begin{code}
@@ -80,11 +80,11 @@ The special thing about types of kind Constraint is that
engine inserts an extra argument of type (Ord a) at every call site
to f.
-Howver, once type inference is over, there is *no* distinction between
+However, once type inference is over, there is *no* distinction between
Constraint and *. Indeed we can have coercions between the two. Consider
class C a where
op :: a - a
-For this single-method class we may genreate a newtype, which in turn
+For this single-method class we may generate a newtype, which in turn
generates an axiom witnessing
Ord a ~ (a - a)
so on the left we have Constraint, and on the right we have *.
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: add embedded vendor (d94250e)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/d94250e68bc85e80e3eb0eb04bcbfac68b4423d2
---
commit d94250e68bc85e80e3eb0eb04bcbfac68b4423d2
Author: Gabor Greif ggr...@gmail.com
Date: Fri Nov 23 17:05:48 2012 +0100
add embedded vendor
---
aclocal.m4 |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/aclocal.m4 b/aclocal.m4
index d014b3a..78e758e 100644
--- a/aclocal.m4
+++ b/aclocal.m4
@@ -194,7 +194,7 @@ AC_DEFUN([FPTOOLS_SET_HASKELL_PLATFORM_VARS],
checkVendor() {
case [$]1 in
-dec|unknown|hp|apple|next|sun|sgi|ibm)
+dec|unknown|hp|apple|next|sun|sgi|ibm|montavista)
;;
*)
echo Unknown vendor [$]1
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typo (359efbf)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/359efbf0fc714528f95207fbe9fc03383b380228
---
commit 359efbf0fc714528f95207fbe9fc03383b380228
Author: Gabor Greif ggr...@gmail.com
Date: Fri Nov 23 17:07:14 2012 +0100
typo
---
rts/Linker.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/rts/Linker.c b/rts/Linker.c
index 066d070..3e169fd 100644
--- a/rts/Linker.c
+++ b/rts/Linker.c
@@ -1,6 +1,6 @@
/*
-
*
- * (c) The GHC Team, 2000-2004
+ * (c) The GHC Team, 2000-2012
*
* RTS Object Linker
*
@@ -1475,7 +1475,7 @@ initLinker( void )
IF_DEBUG(linker, debugBelch(initLinker: start\n));
/* Make initLinker idempotent, so we can call it
- before evey relevant operation; that means we
+ before every relevant operation; that means we
don't need to initialise the linker separately */
if (linker_init_done == 1) {
IF_DEBUG(linker, debugBelch(initLinker: idempotent return\n));
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: fix some typos (704f802)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/704f8024c75ab147c6f9558b1c76525f132e4f14
---
commit 704f8024c75ab147c6f9558b1c76525f132e4f14
Author: Gabor Greif ggr...@gmail.com
Date: Wed Nov 21 18:25:50 2012 +0100
fix some typos
---
compiler/main/GHC.hs |4 ++--
compiler/typecheck/TcRnTypes.lhs |4 ++--
compiler/utils/Outputable.lhs|8
3 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/compiler/main/GHC.hs b/compiler/main/GHC.hs
index 051edcd..82b822b 100644
--- a/compiler/main/GHC.hs
+++ b/compiler/main/GHC.hs
@@ -1,6 +1,6 @@
--
-
--
--- (c) The University of Glasgow, 2005
+-- (c) The University of Glasgow, 2005-2012
--
-- The GHC API
--
@@ -369,7 +369,7 @@ defaultErrorHandler fm (FlushOut flushOut) inner =
inner
-- | Install a default cleanup handler to remove temporary files deposited by
--- a GHC run. This is seperate from 'defaultErrorHandler', because you might
+-- a GHC run. This is separate from 'defaultErrorHandler', because you might
-- want to override the error handling, but still get the ordinary cleanup
-- behaviour.
defaultCleanupHandler :: (ExceptionMonad m, MonadIO m) =
diff --git a/compiler/typecheck/TcRnTypes.lhs b/compiler/typecheck/TcRnTypes.lhs
index 99b9a07..c269882 100644
--- a/compiler/typecheck/TcRnTypes.lhs
+++ b/compiler/typecheck/TcRnTypes.lhs
@@ -1,5 +1,5 @@
-% (c) The University of Glasgow 2006
+% (c) The University of Glasgow 2006-2012
% (c) The GRASP Project, Glasgow University, 1992-2002
%
@@ -784,7 +784,7 @@ emptyImportAvails = ImportAvails { imp_mods =
emptyModuleEnv,
-- | Union two ImportAvails
--
-- This function is a key part of Import handling, basically
--- for each import we create a seperate ImportAvails structure
+-- for each import we create a separate ImportAvails structure
-- and then union them all together with this function.
plusImportAvails :: ImportAvails - ImportAvails - ImportAvails
plusImportAvails
diff --git a/compiler/utils/Outputable.lhs b/compiler/utils/Outputable.lhs
index 7e5c180..ad0b9d7 100644
--- a/compiler/utils/Outputable.lhs
+++ b/compiler/utils/Outputable.lhs
@@ -1,5 +1,5 @@
%
-% (c) The University of Glasgow 2006
+% (c) The University of Glasgow 2006-2012
% (c) The GRASP Project, Glasgow University, 1992-1998
%
@@ -779,15 +779,15 @@ pprWithCommas :: (a - SDoc) -- ^ The pretty printing
function to use
-- comma-separated and finally packed into a
paragraph.
pprWithCommas pp xs = fsep (punctuate comma (map pp xs))
--- | Returns the seperated concatenation of the pretty printed things.
+-- | Returns the separated concatenation of the pretty printed things.
interppSP :: Outputable a = [a] - SDoc
interppSP xs = sep (map ppr xs)
--- | Returns the comma-seperated concatenation of the pretty printed things.
+-- | Returns the comma-separated concatenation of the pretty printed things.
interpp'SP :: Outputable a = [a] - SDoc
interpp'SP xs = sep (punctuate comma (map ppr xs))
--- | Returns the comma-seperated concatenation of the quoted pretty printed
things.
+-- | Returns the comma-separated concatenation of the quoted pretty printed
things.
--
-- [x,y,z] == `x', `y', `z'
pprQuotedList :: Outputable a = [a] - SDoc
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typos in comments (6e4f6c6)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/6e4f6c624fcd72dca11d7856f1a1914a01c22718 --- commit 6e4f6c624fcd72dca11d7856f1a1914a01c22718 Author: Gabor Greif ggr...@gmail.com Date: Fri Nov 16 11:26:50 2012 +0100 typos in comments --- compiler/coreSyn/CoreSyn.lhs |2 +- compiler/typecheck/TcInstDcls.lhs |8 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/compiler/coreSyn/CoreSyn.lhs b/compiler/coreSyn/CoreSyn.lhs index 2fb5aaf..711c148 100644 --- a/compiler/coreSyn/CoreSyn.lhs +++ b/compiler/coreSyn/CoreSyn.lhs @@ -464,7 +464,7 @@ data Tickish id = -- cannot discard a tick, and the compiler should preserve the number -- of ticks as far as possible. -- --- Hwever, we stil allow the simplifier to increase or decrease +-- However, we still allow the simplifier to increase or decrease -- sharing, so in practice the actual number of ticks may vary, except -- that we never change the value from zero to non-zero or vice versa. -- diff --git a/compiler/typecheck/TcInstDcls.lhs b/compiler/typecheck/TcInstDcls.lhs index b4a27b5..236b834 100644 --- a/compiler/typecheck/TcInstDcls.lhs +++ b/compiler/typecheck/TcInstDcls.lhs @@ -189,7 +189,7 @@ Instead we use a cunning trick. iff its argument satisfies exprIsConApp_maybe. This is done in MkId mkDictSelId - * We make 'df' CONLIKE, so that shared uses stil match; eg + * We make 'df' CONLIKE, so that shared uses still match; eg let d = df d1 d2 in ...(op2 d)...(op1 d)... @@ -246,7 +246,7 @@ careful when we have op = ... then we'll get an INLINE pragma on $cop_list but it's important that $cop_list only inlines when it's applied to *two* arguments (the -dictionary and the list argument). So we nust not eta-expand $df +dictionary and the list argument). So we must not eta-expand $df above. We ensure that this doesn't happen by putting an INLINE pragma on the dfun itself; after all, it ends up being just a cast. @@ -309,7 +309,7 @@ instance. Why is this justified? Because we generate a (C [a]) constraint in a context in which 'a' cannot be instantiated to anything that matches -other overlapping instances, or else we would not be excecuting this +other overlapping instances, or else we would not be executing this version of op1 in the first place. It might even be a bit disguised: @@ -323,7 +323,7 @@ It might even be a bit disguised: Precisely this is used in package 'regex-base', module Context.hs. See the overlapping instances for RegexContext, and the fact that they call 'nullFail' just like the example above. The DoCon package also -does the same thing; it shows up in module Fraction.hs +does the same thing; it shows up in module Fraction.hs. Conclusion: when typechecking the methods in a C [a] instance, we want to treat the 'a' as an *existential* type variable, in the sense described ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: fix typo, courtesy of Jon Cave, thanks! (3862220)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/386222001e6dda180cbb05e9e6aebbab04d12aae
---
commit 386222001e6dda180cbb05e9e6aebbab04d12aae
Author: Gabor Greif ggr...@gmail.com
Date: Thu Nov 8 10:44:06 2012 +0100
fix typo, courtesy of Jon Cave, thanks!
---
utils/ghc-pkg/Main.hs |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/utils/ghc-pkg/Main.hs b/utils/ghc-pkg/Main.hs
index ca278e9..e727f4d 100644
--- a/utils/ghc-pkg/Main.hs
+++ b/utils/ghc-pkg/Main.hs
@@ -239,7 +239,7 @@ usageHeader prog = substProg prog $
Prints the highest registered version of a package.\n ++
\n ++
$p check\n ++
- Check the consistency of package depenencies and list broken
packages.\n ++
+ Check the consistency of package dependencies and list broken
packages.\n ++
Accepts the --simple-output flag.\n ++
\n ++
$p describe {pkg}\n ++
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typos (3ca7260)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/3ca7260ae35adf70d31003008af67576d6b6c78c --- commit 3ca7260ae35adf70d31003008af67576d6b6c78c Author: Gabor Greif ggr...@gmail.com Date: Wed Nov 7 12:10:22 2012 +0100 typos --- compiler/ghci/RtClosureInspect.hs |2 +- compiler/typecheck/TcSMonad.lhs |2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/compiler/ghci/RtClosureInspect.hs b/compiler/ghci/RtClosureInspect.hs index 24b7e02..49e943c 100644 --- a/compiler/ghci/RtClosureInspect.hs +++ b/compiler/ghci/RtClosureInspect.hs @@ -983,7 +983,7 @@ the outside type, D. So we can't straightforwardly decompose the outside type, and we end up in the _ branch of the case. Then we match the dataConOrigResTy of the data constructor against the -outside type, hoping to get a substituion that tells how to instantiate +outside type, hoping to get a substitution that tells how to instantiate the *representation* type constructor. This looks a bit delicate to me, but it seems to work. -} diff --git a/compiler/typecheck/TcSMonad.lhs b/compiler/typecheck/TcSMonad.lhs index c0ea95c..7541cd7 100644 --- a/compiler/typecheck/TcSMonad.lhs +++ b/compiler/typecheck/TcSMonad.lhs @@ -487,7 +487,7 @@ The InertCans represents a collection of constraints with the following properti constraint solver that we rewrite types and constraints wrt to the inert substitution is TcCanonical/flattenTyVar. - - In the past we did try to have the inert substituion as + - In the past we did try to have the inert substitution as idempotent as possible but this would only be true for constraints of the same flavor, so in total the inert substitution could not be idempotent, due to flavor-related ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: another typo (50728d0)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/50728d09bed322cdbfdda00f38d0503f9aed728b
---
commit 50728d09bed322cdbfdda00f38d0503f9aed728b
Author: Gabor Greif ggr...@gmail.com
Date: Wed Nov 7 16:05:20 2012 +0100
another typo
---
compiler/typecheck/TcRnTypes.lhs |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/compiler/typecheck/TcRnTypes.lhs b/compiler/typecheck/TcRnTypes.lhs
index e6d2013..4b2ea8f 100644
--- a/compiler/typecheck/TcRnTypes.lhs
+++ b/compiler/typecheck/TcRnTypes.lhs
@@ -915,7 +915,7 @@ built (in TcCanonical).
In contrast, the type of the evidence *term* (ccev_evtm or ctev_evar) in
the evidence may *not* be fully zonked; we are careful not to look at it
-during constraint solving. Seee Note [Evidence field of CtEvidence]
+during constraint solving. See Note [Evidence field of CtEvidence]
\begin{code}
mkNonCanonical :: CtLoc - CtEvidence - Ct
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [PATCH] define own version of PRIdPTR on platform where its not available
On 11/5/12, Simon Marlow marlo...@gmail.com wrote: On 03/11/2012 00:16, Ian Lynagh wrote: Hi Karel, On Fri, Nov 02, 2012 at 07:39:45PM +0100, Karel Gardas wrote: I'm curious if this patch is kind of controversial so you are reluctant to merge it. If you find any issue with it please let me know so I can change it as per your suggestion. I've done this patch to un-break GHC HEAD build failure on my solaris 11 box. If we're going to go this route, then I think we should do it by comparing SIZEOF_VOID_P to SIZEOF_INT etc, rather than assuming that either 'ld' or 'd' is right based on _LP64. However, I'm hoping to replace mkDerivedConstants.c with probably a configure script soon, so that it can produce the right results when cross-compiling. Yes we need to replace mkDerivedConstants.c with something else for cross compilation, but I really hope it doesn't have to be a configure script :-) Surely it's possible to convince the C compiler to emit struct offsets without having to run the program, somehow? +1. A configure script will run (almost) till the heat death of the universe. I have a solution on the cross-compiler-alienless branch which does exactly what you propose: feed the relevant headers into the cross-gcc, obtain the offsets/sizes back by 'nm', and massage stuff to obtain all as macros then feed those into mkDerivedConstants.c. So, unless we want a very generic solution, where more than one cross target is possible, there is no need to abolish mkDerivedConstants.c. The current architecture would do with a few tweaks. Cheers, Gabor Cheers, Simon On 10/30/12 06:59 PM, Karel Gardas wrote: Note that PRIdPTR is considered as linux-ism so it's not available on platforms like Solaris, although some other free Unix(-like) OSes apparently supports it too. --- includes/mkDerivedConstants.c | 11 +++ 1 files changed, 11 insertions(+), 0 deletions(-) diff --git a/includes/mkDerivedConstants.c b/includes/mkDerivedConstants.c index 1393112..953b2b5 100644 --- a/includes/mkDerivedConstants.c +++ b/includes/mkDerivedConstants.c @@ -30,6 +30,17 @@ #includestdio.h #includestring.h + +#if !defined(PRIdPTR) +#if defined(_LP64) +/* compiling for 64bit target */ +#define PRIdPTR ld +#else +/* compiling for 32bit target */ +#define PRIdPTR d +#endif +#endif Thanks Ian ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typo (680dca1)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/680dca1145a08a82a83950bdbca891742fbaf526 --- commit 680dca1145a08a82a83950bdbca891742fbaf526 Author: Gabor Greif ggr...@gmail.com Date: Mon Nov 5 12:56:43 2012 +0100 typo --- docs/users_guide/ghci.xml |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/docs/users_guide/ghci.xml b/docs/users_guide/ghci.xml index c59f4b3..9e8ea2f 100644 --- a/docs/users_guide/ghci.xml +++ b/docs/users_guide/ghci.xml @@ -662,7 +662,7 @@ Prelude an attempt to distinguish it from the new literalT/literal, which is displayed as simply literalT/literal./para -paraClass and type-family instance declarations are simply added to the list of available isntances, with one +paraClass and type-family instance declarations are simply added to the list of available instances, with one exception. Since type-family instances are not permitted to overlap, but you might want to re-define one, a type-family instance emphasisreplaces/emphasis any earlier type instance with an identical left hand side. (See xref linkend=type-families/.)/para ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [PATCH] define own version of PRIdPTR on platform where its not available
Of course I use 'offsetof' in the first part of the tool pipeline. The problem is, though getting the result of it out of the cross-gcc. Here we cannot simply say printf(offsXYZ = %zu, offsetof(XY, Z)); because there is no way to run the cross-gcc output. Instead I do: char offsXYZ[offsetof(XY, Z)]; and the inspect the resulting object file with 'nm'. In the second (i.e. host) gcc invocation I am not allowed to use 'offsetof' to obtain target values, so I include the transformed 'nm' output of the form #define offsXYZ 0x12 and compile mkDerivedConstants.c with these constants. Hope this clears it up. Cheers, Gabor On 11/5/12, Erik de Castro Lopo mle...@mega-nerd.com wrote: Gabor Greif wrote: +1. A configure script will run (almost) till the heat death of the universe. I have a solution on the cross-compiler-alienless branch which does exactly what you propose: feed the relevant headers into the cross-gcc, obtain the offsets/sizes back by 'nm', and massage stuff to obtain all as macros then feed those into mkDerivedConstants.c. Doesn't the offsetof macro do the right thing for get the offset of fields within a struct? http://en.wikipedia.org/wiki/Offsetof Erik -- -- Erik de Castro Lopo http://www.mega-nerd.com/ ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: mkExprStmt not in scope in Parser.hs issue.
IIRC, I had to manually remove a 'Parser.y' file to get things rolling. Cheers, Gabor On 10/29/12, Karel Gardas karel.gar...@centrum.cz wrote: Hello, I'm trying to get GHC HEAD compilation working on my freshly installed Solaris 11/x86 box and I'm trying to solve following error now: /opt/ghc-7.4.2/bin/ghc -static -H32m -O -package-conf libraries/bootstrapping.conf -package-name ghc-7.7 -hide-all-packages -i -icompiler/basicTypes -icompiler/cmm -icompiler/codeGen -icompiler/coreSyn -icompiler/deSugar -icompiler/ghci -icompiler/hsSyn -icompiler/iface -icompiler/llvmGen -icompiler/main -icompiler/nativeGen -icompiler/parser -icompiler/prelude -icompiler/profiling -icompiler/rename -icompiler/simplCore -icompiler/simplStg -icompiler/specialise -icompiler/stgSyn -icompiler/stranal -icompiler/typecheck -icompiler/types -icompiler/utils -icompiler/vectorise -icompiler/stage1/build -icompiler/stage1/build/autogen -Icompiler/stage1/build -Icompiler/stage1/build/autogen -Icompiler/stage1 -Icompiler/. -Icompiler/parser -Icompiler/utils-optP-include -optPcompiler/stage1/build/autogen/cabal_macros.h -package Cabal-1.17.0 -package array-0.4.0.0 -package base-4.5.1.0 -package bin-package-db-0.0.0.0 -package bytestring-0.9.2.1 -package containers-0.4.2.1 -package directory-1.1.0.2 -package filepath-1.3.0.0 -package hoopl-3.9.0.0 -package hpc-0.6.0.0 -package old-time-1.1.0.0 -package process-1.1.0.1 -package time-1.4 -package unix-2.5.1.1 -Wall -fno-warn-name-shadowing -XHaskell98 -XNondecreasingIndentation -XCPP -XMagicHash -XUnboxedTuples -XPatternGuards -XForeignFunctionInterface -XEmptyDataDecls -XTypeSynonymInstances -XMultiParamTypeClasses -XFlexibleInstances -XRankNTypes -XScopedTypeVariables -XDeriveDataTypeable -XBangPatterns -DSTAGE=1 -no-user-package-conf -rtsopts -odir compiler/stage1/build -hidir compiler/stage1/build -stubdir compiler/stage1/build -hisuf hi -osuf o -hcsuf hc -c compiler/stage1/build/Parser.hs -o compiler/stage1/build/Parser.o compiler/stage1/build/Parser.hs:6705:45: Not in scope: `mkExprStmt' gmake[1]: *** [compiler/stage1/build/Parser.o] Error 1 gmake: *** [all] Error 2 I'm bootstraping with GHC 7.4.2 and happy is 1.18.6 version and alex is 2.3.5 if this is important. Is there any idea what causes this failure? The HEAD is as of today. Thanks! Karel ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: extend LD_LIBRARY_PATH instead of replacing it (fb41ac5)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/fb41ac5285895c3b5304b1c267197faaea89e8a6 --- commit fb41ac5285895c3b5304b1c267197faaea89e8a6 Author: Gabor Greif ggr...@gmail.com Date: Fri Oct 12 11:25:29 2012 +0200 extend LD_LIBRARY_PATH instead of replacing it --- rules/shell-wrapper.mk |6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/rules/shell-wrapper.mk b/rules/shell-wrapper.mk index 395968a..86af3ac 100644 --- a/rules/shell-wrapper.mk +++ b/rules/shell-wrapper.mk @@ -1,6 +1,6 @@ # - # -# (c) 2009 The University of Glasgow +# (c) 2009-2012 The University of Glasgow # # This file is part of the GHC build system. # @@ -46,9 +46,9 @@ $$(INPLACE_WRAPPER): $$($1_$2_INPLACE) $$($1_$2_INPLACE_SHELL_WRAPPER_EXTRA) ifeq $$(DYNAMIC_BY_DEFAULT) YES ifeq $$(TargetOS_CPP) linux - echo 'export LD_LIBRARY_PATH=$$($1_$2_DEP_LIB_DIRS_SEARCHPATH)' $$@ + echo 'export LD_LIBRARY_PATH=$$($1_$2_DEP_LIB_DIRS_SEARCHPATH):LD_LIBRARY_PATH' $$@ else ifeq $$(TargetOS_CPP) darwin - echo 'export DYLD_LIBRARY_PATH=$$($1_$2_DEP_LIB_DIRS_SEARCHPATH)' $$@ + echo 'export DYLD_LIBRARY_PATH=$$($1_$2_DEP_LIB_DIRS_SEARCHPATH):DYLD_LIBRARY_PATH' $$@ endif endif ifeq $$($1_$2_SHELL_WRAPPER) YES ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typo (0b3811c)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/0b3811c093736950c1d2757fb12dba60f9bf97ca
---
commit 0b3811c093736950c1d2757fb12dba60f9bf97ca
Author: Gabor Greif ggr...@gmail.com
Date: Wed Oct 24 13:45:55 2012 +0200
typo
---
compiler/ghci/Linker.lhs |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/compiler/ghci/Linker.lhs b/compiler/ghci/Linker.lhs
index 0a9de80..3ba9c3c 100644
--- a/compiler/ghci/Linker.lhs
+++ b/compiler/ghci/Linker.lhs
@@ -1,5 +1,5 @@
%
-% (c) The University of Glasgow 2005-2006
+% (c) The University of Glasgow 2005-2012
%
\begin{code}
-- | The dynamic linker for GHCi.
@@ -1239,7 +1239,7 @@ searchForLibUsingGcc dflags so dirs = do
else return (Just file)
--
--- Loading a dyanmic library (dlopen()-ish on Unix, LoadLibrary-ish on Win32)
+-- Loading a dynamic library (dlopen()-ish on Unix, LoadLibrary-ish on Win32)
-- Darwin / MacOS X only: load a framework
-- a framework is a dynamic library packaged inside a directory of the same
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: October 2012 stats (82ab12a)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/82ab12afefd6f115d2efb37924227ef2f43adb19
---
commit 82ab12afefd6f115d2efb37924227ef2f43adb19
Author: Gabor Greif ggr...@gmail.com
Date: Tue Oct 23 09:50:31 2012 +0200
October 2012 stats
---
compiler/parser/Parser.y.pp |8 +++-
1 files changed, 7 insertions(+), 1 deletions(-)
diff --git a/compiler/parser/Parser.y.pp b/compiler/parser/Parser.y.pp
index 966d4e3..3c18498 100644
--- a/compiler/parser/Parser.y.pp
+++ b/compiler/parser/Parser.y.pp
@@ -72,7 +72,13 @@ import Control.Monad( mplus )
{-
-
-24 Februar 2006
+12 October 2012
+
+Conflicts: 43 shift/reduce
+ 1 reduce/reduce
+
+-
+24 February 2006
Conflicts: 33 shift/reduce
1 reduce/reduce
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: comments only (2c2be63)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/2c2be637206cb1d58f4fc8a4a2e717e419c9fa5d --- commit 2c2be637206cb1d58f4fc8a4a2e717e419c9fa5d Author: Gabor Greif ggr...@gmail.com Date: Tue Oct 23 09:52:30 2012 +0200 comments only --- includes/stg/MiscClosures.h |6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/includes/stg/MiscClosures.h b/includes/stg/MiscClosures.h index 760a59d..c94fdb3 100644 --- a/includes/stg/MiscClosures.h +++ b/includes/stg/MiscClosures.h @@ -213,7 +213,7 @@ RTS_THUNK(stg_ap_5_upd); RTS_THUNK(stg_ap_6_upd); RTS_THUNK(stg_ap_7_upd); -/* standard application routines (see also rts/gen_apply.py, +/* standard application routines (see also utils/genapply, * and compiler/codeGen/CgStackery.lhs). */ RTS_RET(stg_ap_v); @@ -470,10 +470,10 @@ extern StgWord RTS_VAR(stable_ptr_table); // Profiling.c extern unsigned int RTS_VAR(era); extern unsigned int RTS_VAR(entering_PAP); -extern StgWord RTS_VAR(CC_LIST); /* registered CC list */ +extern StgWord RTS_VAR(CC_LIST); /* registered CC list */ extern StgWord RTS_VAR(CCS_LIST); /* registered CCS list */ extern StgWord CCS_SYSTEM[]; -extern unsigned int RTS_VAR(CC_ID);/* global ids */ +extern unsigned int RTS_VAR(CC_ID);/* global ids */ extern unsigned int RTS_VAR(CCS_ID); #endif ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Make warnings in dph
These 'make' diagnostics appeared today with a fresh pull: make[1]: Circular libraries/dph/dph-lifted-copy/dist-install/build/Data/Array/Parallel/Lifted/PArray.dyn_o - libraries/dph/dph-lifted-copy/dist-install/build/Data/Array/Parallel/Lifted/PArray.o dependency dropped. make[1]: Circular libraries/dph/dph-lifted-copy/dist-install/build/Data/Array/Parallel/Lifted/TH/Repr.dyn_o - libraries/dph/dph-lifted-copy/dist-install/build/Data/Array/Parallel/Lifted/TH/Repr.o dependency dropped. Ist it a reason for concern? Cheers, Gabor ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: Determining live registers on Cmm procedure entry?
David, this reminds me, I have added a feature bugzilla to LLVM: http://llvm.org/bugs/show_bug.cgi?id=14080 May be interesting to discuss it here first. Cheers, Gabor On 10/16/12, David Terei davidte...@gmail.com wrote: What is the problem you face? That you want to allow passing a dynamic number of float or double arguments? I assume you are keeping it fixed at 6 still but want to allow those 6 to be either 6 floats, 6 doubles or any other combination. (Long term is anything that can be put in a xmm* register...) Why do you want Cmm liveness on entry? Is it simply to solve the above so you can generate specialized function types instead of the universal one the LLVM backend uses? If so, have you tried just using the most general universal type? e.g., how badly is the generated code affected if you have all arguments be Doubles and then do type casts on jumps and function entries? This is the original commit I wrote for tracking STG reg liveness: 4384e146640230399b38cd62e8e5df391f72c3a7 This is the one Simon Marlow wrote to do the same but in the new code gen: c6a61235aa0baf6f9e8a41c5a771ccc7e32c23a5 I'd suggest you look at these commits and they should show you where to get the information from. (I can't remember anymore sorry). Cheers, David On 16 October 2012 03:45, Geoffrey Mainland mainl...@apeiron.net wrote: Hi all, I'm working on changing GHC's calling convention to support overlapping register classes. On x86-64 this will mean that a function can receive up to six Float/Double arguments (any mixture) in registers. Right now only two Double arguments can be passed in registers on x86-64, even if a function takes zero Float arguments---the register classes for Floats and Doubles cannot overlap even though both are passed in xmm* registers. This is working fine with the native back-end, but I'm not sure how to get it working with the LLVM back-end. Right now LLVM translates a Cmm procedure to an LLVM function of a single universal type( that is, all LLVM functions that the LLVM back-end produces for Cmm procedures have the same type) that takes all STG registers as arguments. A Cmm procedure call passes all the STG registers, although the back-end takes advantage of the liveness information attached to Cmm jumps to pass undefined values for non-live registers. Now that register classes can overlap, I need to change this. Since jumps have liveness information attached, I can simply pass the live registers. But when translating Cmm procedures, I need to know which registers are live on entry to the procedure. The slightly longer-term goal is to add support for passing SSE vectors in registers. What I'd like to have ideally is the set of live registers *and their Cmm types* at each procedure entry and jump. I envision having a single STG register class, say X16, for all 128-bit-wide SSE registers, but I'd like the Cmm type information when generating LLVM type signatures so I can generate the appropriate LLVM types---a 128-bit-wide vector of 32-bit integers does not have the same type as a 128-bit-wide vector of double precision floats. I might be able to do without the Cmm type information by inserting appropriate casts in the generated LLVM code, but I really need the set of live registers on entry to a Cmm procedure. I don't see how to get this unfortunately. Any hints? Thanks, Geoff ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: Determining live registers on Cmm procedure entry?
On 10/17/12, David Terei davidte...@gmail.com wrote: On 16 October 2012 14:48, Gabor Greif ggr...@gmail.com wrote: David, this reminds me, I have added a feature bugzilla to LLVM: http://llvm.org/bugs/show_bug.cgi?id=14080 May be interesting to discuss it here first. I think opening a ticket with LLVM is probably a bad move sorry. It's not our community and so we want to be careful not to harass them so we maintain good-will with them (and they've shown they are willing to As I've been a contributor to LLVM since 2005, I do not think they'll understand this as harassment :-) Currently I focus on selling Haskell *as such* at my workplace, hopefully that will be over in December, and then I can tackle some of my LLVM enhancement PRs (such as this one). I just wanted to record some thoughts. engage with us and help with us but that is a limited use resource, hence conserve!). Discussion is probably best done on the mailing list. But more importantly, we should only pull them into a discussion when we have someone committed to working on the problem. That isn't me at this point, I don't have time. I assume you mean this Haskell list (cvs.ghc@). So if you want to discuss TNTC / LLVM solutions, great! But unless you are also prepared to drive the discussion and implement proposed solutions, lets keep it internal to GHC. Yeah, I won't implement anything before discussing this with the relevant persons (in both communities). Sometimes people have very good arguments why something is not a good idea. I think there are ideas right now of how to move forward on TNTC, its more a matter of implementing one or two of them to experiment and demonstrate results. Okay, as soon as I am ready to kick, I'll rekindle these threads. Thanks for your honest opinions, Gabor Cheers, David Sure. Cheers, Gabor On 10/16/12, David Terei davidte...@gmail.com wrote: What is the problem you face? That you want to allow passing a dynamic number of float or double arguments? I assume you are keeping it fixed at 6 still but want to allow those 6 to be either 6 floats, 6 doubles or any other combination. (Long term is anything that can be put in a xmm* register...) Why do you want Cmm liveness on entry? Is it simply to solve the above so you can generate specialized function types instead of the universal one the LLVM backend uses? If so, have you tried just using the most general universal type? e.g., how badly is the generated code affected if you have all arguments be Doubles and then do type casts on jumps and function entries? This is the original commit I wrote for tracking STG reg liveness: 4384e146640230399b38cd62e8e5df391f72c3a7 This is the one Simon Marlow wrote to do the same but in the new code gen: c6a61235aa0baf6f9e8a41c5a771ccc7e32c23a5 I'd suggest you look at these commits and they should show you where to get the information from. (I can't remember anymore sorry). Cheers, David On 16 October 2012 03:45, Geoffrey Mainland mainl...@apeiron.net wrote: Hi all, I'm working on changing GHC's calling convention to support overlapping register classes. On x86-64 this will mean that a function can receive up to six Float/Double arguments (any mixture) in registers. Right now only two Double arguments can be passed in registers on x86-64, even if a function takes zero Float arguments---the register classes for Floats and Doubles cannot overlap even though both are passed in xmm* registers. This is working fine with the native back-end, but I'm not sure how to get it working with the LLVM back-end. Right now LLVM translates a Cmm procedure to an LLVM function of a single universal type( that is, all LLVM functions that the LLVM back-end produces for Cmm procedures have the same type) that takes all STG registers as arguments. A Cmm procedure call passes all the STG registers, although the back-end takes advantage of the liveness information attached to Cmm jumps to pass undefined values for non-live registers. Now that register classes can overlap, I need to change this. Since jumps have liveness information attached, I can simply pass the live registers. But when translating Cmm procedures, I need to know which registers are live on entry to the procedure. The slightly longer-term goal is to add support for passing SSE vectors in registers. What I'd like to have ideally is the set of live registers *and their Cmm types* at each procedure entry and jump. I envision having a single STG register class, say X16, for all 128-bit-wide SSE registers, but I'd like the Cmm type information when generating LLVM type signatures so I can generate the appropriate LLVM types---a 128-bit-wide vector of 32-bit integers does not have the same type as a 128-bit-wide vector of double precision floats. I might be able to do without the Cmm type information by inserting appropriate casts in the generated LLVM code
[commit: ghc] master: fix obvious typos (b7fbf3a)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/b7fbf3a11ed939f6758fb6c5ef55d577351523e2 --- commit b7fbf3a11ed939f6758fb6c5ef55d577351523e2 Author: Gabor Greif ggr...@gmail.com Date: Fri Oct 12 09:40:09 2012 +0200 fix obvious typos these can go into the release 7.6.2 too, I guess --- docs/users_guide/glasgow_exts.xml |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/docs/users_guide/glasgow_exts.xml b/docs/users_guide/glasgow_exts.xml index f574199..541b309 100644 --- a/docs/users_guide/glasgow_exts.xml +++ b/docs/users_guide/glasgow_exts.xml @@ -632,7 +632,7 @@ type Typ data TypView = Unit | Arrow Typ Typ -view :: Type - TypeView +view :: Typ - TypView -- additional operations for constructing Typ's ... /programlisting ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: Fix typo in the documentation of CApiFFI (fc8f9c1)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/fc8f9c1be4f05ddf58c128ecca7213351765168b --- commit fc8f9c1be4f05ddf58c128ecca7213351765168b Author: shelarcy shela...@gmail.com Date: Tue Oct 2 17:10:13 2012 +0900 Fix typo in the documentation of CApiFFI Signed-off-by: Gabor Greif ggr...@gmail.com --- docs/users_guide/ffi-chap.xml |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/docs/users_guide/ffi-chap.xml b/docs/users_guide/ffi-chap.xml index 2425d82..e778c03 100644 --- a/docs/users_guide/ffi-chap.xml +++ b/docs/users_guide/ffi-chap.xml @@ -164,7 +164,7 @@ foreign import ccall interruptible sect2 id=ffi-capi titleThe CAPI calling convention/title para - The literalCAPI/literal extension allows a calling + The literalCApiFFI/literal extension allows a calling convention of literalcapi/literal to be used in foreign declarations, e.g. ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [PATCH] Fix typo in the documentation of CApiFFI
done in commit fc8f9c1be4f05ddf58c128ecca7213351765168b. Thanks! Gabor On 10/2/12, shelarcy shela...@gmail.com wrote: Hello, GHC HEAD and 7.6.1 User's Guide 8.1.5. The CAPI calling convention section describes wrong extention name. So, I made patch to fix that. Best Regards, -- shelarcy shelarcyhotmail.co.jp http://page.freett.com/shelarcy/ ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[Q] Potential confusion in docs
Which is correct: to0 or to ? $ git grep map to genum docs/users_guide/glasgow_exts.xml:enum = map to genum $ git grep map to0 genum docs/users_guide/glasgow_exts.xml:the default implementation literalmap to0 genum/literal is filled-in, It appears that to0 is the correct one grepping the GHC sources, but who knows... Should this be corrected? Cheers, Gabor ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [commit: ghc] master: whitespace and panic message fixup (254f1a6)
Hi Simon, -getCondCode other = pprPanic getCondCode(2)(x86,sparc) (ppr other) +getCondCode other = pprPanic getCondCode(2)(x86,x96_64) (ppr other) 'x96_64' sounds wrong... Cheers, Gabor ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typo (0678289)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/067828956acdcb046c0640568638fbc496eda9f3 --- commit 067828956acdcb046c0640568638fbc496eda9f3 Author: Gabor Greif ggr...@gmail.com Date: Mon Sep 17 16:10:47 2012 +0200 typo --- compiler/ghci/Linker.lhs |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/compiler/ghci/Linker.lhs b/compiler/ghci/Linker.lhs index 565cf0b..6b47db3 100644 --- a/compiler/ghci/Linker.lhs +++ b/compiler/ghci/Linker.lhs @@ -772,7 +772,7 @@ dynLinkObjs dflags pls objs = do mapM_ loadObj (map nameOfObject unlinkeds) --- Link the all together +-- Link them all together ok - resolveObjs -- If resolving failed, unload all our ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] cross-compiler-alienless: Checkpointing some makefile simplifications that I attempted. (a18f009)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : cross-compiler-alienless http://hackage.haskell.org/trac/ghc/changeset/a18f0096c6c9b1d649067fc73c04f43fef7bef82 --- commit a18f0096c6c9b1d649067fc73c04f43fef7bef82 Author: Gabor Greif ggr...@gmail.com Date: Thu Sep 13 14:02:41 2012 +0200 Checkpointing some makefile simplifications that I attempted. This is not in a working state yet, as we are trying to build something that goes into the build/ dir but it is not yet created. Not sure how to add dependency tracking properly. Btw., Ian is working in this area on the master branch so there will be a bunch of conflicts anyway :-( --- includes/ghc.mk | 44 +++- 1 files changed, 11 insertions(+), 33 deletions(-) diff --git a/includes/ghc.mk b/includes/ghc.mk index b41f92e..5b7bd51 100644 --- a/includes/ghc.mk +++ b/includes/ghc.mk @@ -141,8 +141,13 @@ DerivedConstants.h : else includes_dist-derivedconstants_CC_OPTS = -Iincludes/dist-derivedconstants/build +includes_dist-derivedconstants_C_SRCS = mkDerivedConstants.c +includes_dist-derivedconstants_PROG = mkDerivedConstants$(exeext) ifeq $(BuildingCrossCompiler)-$(AlienScript) YES- +includes/dist-derivedconstants/build/Capability.cross.h \ +includes/dist-derivedconstants/build/Rts.cross.h: $(includes_H_CONFIG) $(includes_H_PLATFORM) $(includes_H_FILES) $$(rts_H_FILES) + includes/dist-derivedconstants/build/Capability.cross.h: rts/Capability.h | $$(dir $$@)/. $(CC_STAGE1) -E -DPROFILING -DTHREADED_RTS $(CONF_CPP_OPTS_STAGE1) $(rts_CC_OPTS) $ $@ includes/dist-derivedconstants/build/Rts.cross.h: includes/Rts.h | $$(dir $$@)/. @@ -155,17 +160,7 @@ includes/dist-derivedconstants/build/mkDerivedConstants.cross.o: includes/dist-d includes/dist-derivedconstants/build/SizeMacros.h: includes/mkSizeMacros.cross.awk includes/dist-derivedconstants/build/SizeMacros.h: includes/dist-derivedconstants/build/mkDerivedConstants.cross.o | $$(dir $$@)/. $(NM) $ | $(SORT) | awk -f includes/mkSizeMacros.cross.awk $@ - -includes_dist-derivedconstants_C_SRCS = mkDerivedConstants.c # XXX NM_STAGE1 AWK -includes_dist-derivedconstants_PROG = mkDerivedConstants$(exeext) - -includes/dist-derivedconstants/build/mkDerivedConstants$(exeext) : includes/dist-derivedconstants/build/SizeMacros.h -includes/dist-derivedconstants/build/mkDerivedConstants$(exeext) : includes/mkDerivedConstants.c - $(CC_STAGE0) $(includes_dist-derivedconstants_CC_OPTS) $(CONF_CPP_OPTS_STAGE0) $(rts_CC_OPTS) $(includes_CC_OPTS) $ -o $@ - -$(INPLACE_BIN)/mkDerivedConstants$(exeext) : includes/dist-derivedconstants/build/mkDerivedConstants$(exeext) - $(CP) $ $@ else includes/dist-derivedconstants/build/SizeMacros.h : | $$(dir $$@)/. @echo #define OFFSET(s_type, field) ((size_t)(((s_type*)0)-field)) $@ @@ -184,14 +179,6 @@ includes/dist-derivedconstants/build/SizeMacros.h : | $$(dir $$@)/. @echo #define b_bitmap b.bitmap $@ @echo $@ -includes_dist-derivedconstants_C_SRCS = mkDerivedConstants.c -includes_dist-derivedconstants_PROG = mkDerivedConstants$(exeext) - -$(eval $(call build-prog,includes,dist-derivedconstants,0)) - -$(includes_dist-derivedconstants_depfile_c_asm) : $(includes_H_CONFIG) $(includes_H_PLATFORM) $(includes_H_FILES) $$(rts_H_FILES) -includes/dist-derivedconstants/build/mkDerivedConstants.o : includes/dist-derivedconstants/build/SizeMacros.h $(includes_H_CONFIG) $(includes_H_PLATFORM) - ifneq $(AlienScript) $(INPLACE_BIN)/mkDerivedConstants$(exeext): includes/$(includes_dist-derivedconstants_C_SRCS) | $$(dir $$@)/. $(WhatGccIsCalled) -o $@ $ $(CFLAGS) $(includes_CC_OPTS) @@ -199,6 +186,9 @@ endif endif ifneq $(BINDIST) YES +$(includes_dist-derivedconstants_depfile_c_asm) : | includes/dist-derivedconstants/build/SizeMacros.h +$(includes_dist-derivedconstants_depfile_c_asm) : $(includes_H_CONFIG) $(includes_H_PLATFORM) $(includes_H_FILES) $$(rts_H_FILES) +$(eval $(call build-prog,includes,dist-derivedconstants,0)) $(includes_DERIVEDCONSTANTS) : $(INPLACE_BIN)/mkDerivedConstants$(exeext) | $$(dir $$@)/. ifeq $(AlienScript) ./$ $@ @@ -222,25 +212,14 @@ $(includes_GHCCONSTANTS) : else +ifneq $(BINDIST) YES includes_dist-ghcconstants_CC_OPTS = -DGEN_HASKELL -Iincludes/dist-derivedconstants/build - -ifeq $(BuildingCrossCompiler)-$(AlienScript) YES- -includes/dist-ghcconstants/build/mkDerivedConstants$(exeext) : includes/dist-derivedconstants/build/SizeMacros.h -includes/dist-ghcconstants/build/mkDerivedConstants$(exeext) : includes/mkDerivedConstants.c | $$(dir $$@)/. - $(CC_STAGE0) $(includes_dist-ghcconstants_CC_OPTS) $(CONF_CPP_OPTS_STAGE0) $(rts_CC_OPTS) $(includes_CC_OPTS) $ -o $@ -$(INPLACE_BIN)/mkGHCConstants$(exeext) : includes/dist-ghcconstants
Re: [commit: ghc] cross-compiler-alienless: Checkpointing some makefile simplifications that I attempted. (a18f009)
On 9/13/12, Ian Lynagh i...@well-typed.com wrote: On Thu, Sep 13, 2012 at 05:12:56AM -0700, Gabor Greif wrote: Btw., Ian is working in this area on the master branch so there will be a bunch of conflicts anyway :-( Sorry, I hadn't realised you were working on this in a branch. If merging your changes to master is possible (i.e. if they don't break the build, even if what they add isn't actually used yet) then I'd recommend doing that, to avoid conflicts. No worries, I like the direction you going :-) I just cannot work full time on the branch (mere few hours a week), so I am always lagging behind. You guys are a productive bunch... It's starting to feel like we're close to a full cross-compilation compiler, incidentally. I'm currently working on getting all the platform-specific values into a platformConstants file, which is read by the compiler at run-time. Once that's done, modulo a handful of bits of That is fine, as my branch's main contribution is to interrogate a cross-gcc about what those constants are, and to make them available to mkDerivedConstants.c. Did you look into the mechanism how this is done? In short, I create an .o file that has objects whose *size* equals the platform constants. I extract the sizes via 'nm' and create the SizeMacros.h. This in turn gets included by mkDerivedConstants.c. CPP etc that remain, it ought to be possible to make a single-platform cross-compiler simply by substituting the 'settings' and 'platformConstants' files with those from another platform. In reality, we'll probably find more places where target = host assumptions are getting baked in, but we'll have to deal with those when we come to them. Yes. The other half of the puzzle (perhaps what you're looking at) is altering the build system to be able to (a) generate 'settings' and 'platformConstants' files for platforms other than host, and (b) build Yes. See above. for multiple platforms in a single tree. I suspect that (a) means Actually, I am happy with one HOST != TARGET platform per GHC already. getting rid of mkDerivedConstants, and using a configure script instead, but I haven't really thought about this half yet. Yes, this is part of the plan, but can probably wait. Passing around the autoconf cache file to sub-configures is probably more lucrative short-term. Cheers, Gabor Thanks Ian ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [Haskell] ANNOUNCE: GHC version 7.6.1
And there was this c-vs-c comment from (IIRC) axman on the #ghc IRC channel: class Category (c :: k - k - *) where type Ob c :: k - Constraint id :: Ob c a = c a a (.) :: (Ob c a, Ob c b, Ob c c) = c b c - c a b - c a c This is pretty confusing for anybody whose brain does not yet instantly 'autocorrect' these kinds of mistakes. Cheers, Gabor On 9/7/12, Simon Peyton-Jones simo...@microsoft.com wrote: Yes, it should be 'z', as you say. Paolo, could you fix this typo please? Thanks Simon | -Original Message- | From: haskell-boun...@haskell.org [mailto:haskell-boun...@haskell.org] On | Behalf Of Bas van Dijk | Sent: 07 September 2012 10:01 | To: hask...@haskell.org; glasgow-haskell-us...@haskell.org | Subject: Re: [Haskell] ANNOUNCE: GHC version 7.6.1 | | On 6 September 2012 18:05, Ian Lynagh i...@well-typed.com wrote: | The GHC Team is pleased to announce a new major release of GHC, 7.6.1. | | Great! | | * It is now possible to defer type errors until runtime using the | -fdefer-type-errors flag. | | In section 7.13.1 it says: | | ...given the following code: | | x :: Int | x = 0 | | y :: Char | y = x | | z :: Int | z = y | | evaluating x will result in a runtime type error. | | Shouldn't this be: | | evaluating z will result in a runtime type error. | | Cheers, | | Bas | | ___ | Haskell mailing list | hask...@haskell.org | http://www.haskell.org/mailman/listinfo/haskell ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] cross-compiler-alienless: employ make variable BuildingCrossCompiler in all relevant places (78bea85)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : cross-compiler-alienless http://hackage.haskell.org/trac/ghc/changeset/78bea858089392ee404674bad97d11b8c2a25c20 --- commit 78bea858089392ee404674bad97d11b8c2a25c20 Author: Gabor Greif ggr...@gmail.com Date: Fri Aug 31 11:14:20 2012 +0200 employ make variable BuildingCrossCompiler in all relevant places --- ghc/ghc.mk |6 +++--- libffi/ghc.mk |4 ++-- rules/build-dependencies.mk |4 ++-- rules/build-package-data.mk |4 ++-- rules/cross-compiling.mk| 24 rules/distdir-way-opts.mk |4 ++-- rules/hs-suffix-rules-srcdir.mk |2 +- rules/shell-wrapper.mk |4 ++-- 8 files changed, 14 insertions(+), 38 deletions(-) diff --git a/ghc/ghc.mk b/ghc/ghc.mk index c78fa3b..12b837c 100644 --- a/ghc/ghc.mk +++ b/ghc/ghc.mk @@ -45,7 +45,7 @@ ghc_stage1_C_FILES_NODEPS = ghc/hschooks.c ghc_stage2_MKDEPENDC_OPTS = -DMAKING_GHC_BUILD_SYSTEM_DEPENDENCIES ghc_stage3_MKDEPENDC_OPTS = -DMAKING_GHC_BUILD_SYSTEM_DEPENDENCIES -ifneq $(TARGETPLATFORM) $(HOSTPLATFORM) +ifeq $(BuildingCrossCompiler) YES ghc_stage2_MKDEPENDC_OPTS += -DCOMPILING_GHC endif @@ -77,7 +77,7 @@ ghc_stage3_PROG = ghc-stage3$(exeext) ghc_stage1_SHELL_WRAPPER = YES ghc_stage2_SHELL_WRAPPER = YES ghc_stage3_SHELL_WRAPPER = YES -ifneq $(TARGETPLATFORM) $(HOSTPLATFORM) +ifeq $(BuildingCrossCompiler) YES ghc_stage1_SHELL_WRAPPER_NAME = ghc/ghc-cross.wrapper ghc_stage2_SHELL_WRAPPER_NAME = ghc/ghc-cross.wrapper else @@ -100,7 +100,7 @@ ifneq $(filter-out 1,$(stage)) ghc_stage1_NOT_NEEDED = YES endif -ifneq $(TARGETPLATFORM) $(HOSTPLATFORM) +ifeq $(BuildingCrossCompiler) YES ghc_stage2_NOT_NEEDED = YES ghc_stage3_NOT_NEEDED = YES else diff --git a/libffi/ghc.mk b/libffi/ghc.mk index c5c1ff7..dad675d 100644 --- a/libffi/ghc.mk +++ b/libffi/ghc.mk @@ -1,6 +1,6 @@ # - # -# (c) 2009 The University of Glasgow +# (c) 2009-2012 The University of Glasgow # # This file is part of the GHC build system. # @@ -44,7 +44,7 @@ endif # libffi has a different nomenclature from ours # regarding 'host' (our target) and 'build' (also our build) -ifneq $(TARGETPLATFORM) $(HOSTPLATFORM) +ifeq $(BuildingCrossCompiler) YES LIBFFI_PLATFORMS = --host=$(TARGETPLATFORM) --build=$(HOSTPLATFORM) else # XXX: --host=$(TARGETPLATFORM) also? diff --git a/rules/build-dependencies.mk b/rules/build-dependencies.mk index 867ea14..db999a3 100644 --- a/rules/build-dependencies.mk +++ b/rules/build-dependencies.mk @@ -1,6 +1,6 @@ # - # -# (c) 2009 The University of Glasgow +# (c) 2009-2012 The University of Glasgow # # This file is part of the GHC build system. # @@ -24,7 +24,7 @@ $1_$2_C_FILES_DEPS = $$(filter-out $$($1_$2_C_FILES_NODEPS),$$($1_$2_C_FILES)) $1_$2_MKDEPENDHS_FLAGS = -dep-makefile $$($1_$2_depfile_haskell).tmp $$(foreach way,$$(filter-out v,$$($1_$2_WAYS)),-dep-suffix $$(way)) $1_$2_MKDEPENDHS_FLAGS += -include-pkg-deps -ifneq $(TARGETPLATFORM) $(HOSTPLATFORM) +ifeq $(BuildingCrossCompiler) YES $1_$2_MKDEPENDHS_FLAGS += -DCOMPILING_GHC endif diff --git a/rules/build-package-data.mk b/rules/build-package-data.mk index 34f09a0..92e0413 100644 --- a/rules/build-package-data.mk +++ b/rules/build-package-data.mk @@ -1,6 +1,6 @@ # - # -# (c) 2009 The University of Glasgow +# (c) 2009-2012 The University of Glasgow # # This file is part of the GHC build system. # @@ -64,7 +64,7 @@ $1_$2_CONFIGURE_OPTS += $$(BOOT_PKG_CONSTRAINTS) endif ifeq $3 1 -ifneq $(TARGETPLATFORM) $(HOSTPLATFORM) +ifeq $(BuildingCrossCompiler) YES $1_$2_CONFIGURE_OPTS += --configure-option=--host=$(HOSTPLATFORM) $1_$2_CONFIGURE_OPTS += --configure-option=--target=$(TARGETPLATFORM) endif diff --git a/rules/cross-compiling.mk b/rules/cross-compiling.mk deleted file mode 100644 index 9f9ec6f..000 --- a/rules/cross-compiling.mk +++ /dev/null @@ -1,24 +0,0 @@ -# - -# -# (c) 2012 The University of Glasgow -# -# This file is part of the GHC build system. -# -# To understand how the build system works and how to modify it, see -# http://hackage.haskell.org/trac/ghc/wiki/Building/Architecture -# http://hackage.haskell.org/trac/ghc/wiki/Building/Modifying -# -# - - -define cross-compiling # $1 = then, $2 = else, $3 = then, ... -ifneq $(TARGETPLATFORM) $(HOSTPLATFORM) -ifneq $(BUILDPLATFORM) $(HOSTPLATFORM) -$(warning When cross-compiling, the build and host platforms must
Warnings in RTS
I am seeing these warnings: rts/Interpreter.c: In function `interpretBCO': rts/Interpreter.c:1410:0: warning: value computed is not used rts/Interpreter.c:1411:0: warning: value computed is not used rts/sm/Sanity.c: In function `memInventory': rts/sm/Sanity.c:857:0: warning: value computed is not used Should apply the obvious fixes and push? Cheers, Gabor ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] cross-compiler-alienless: cross compiler is placed in stage1, no need for this wrapper name (df53a4d)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : cross-compiler-alienless http://hackage.haskell.org/trac/ghc/changeset/df53a4d5eaac320c0876a083e40c5d14c1d87238 --- commit df53a4d5eaac320c0876a083e40c5d14c1d87238 Author: Gabor Greif ggr...@gmail.com Date: Fri Aug 31 14:32:43 2012 +0200 cross compiler is placed in stage1, no need for this wrapper name --- ghc/ghc.mk |1 - 1 files changed, 0 insertions(+), 1 deletions(-) diff --git a/ghc/ghc.mk b/ghc/ghc.mk index 12b837c..43484e6 100644 --- a/ghc/ghc.mk +++ b/ghc/ghc.mk @@ -79,7 +79,6 @@ ghc_stage2_SHELL_WRAPPER = YES ghc_stage3_SHELL_WRAPPER = YES ifeq $(BuildingCrossCompiler) YES ghc_stage1_SHELL_WRAPPER_NAME = ghc/ghc-cross.wrapper -ghc_stage2_SHELL_WRAPPER_NAME = ghc/ghc-cross.wrapper else ghc_stage1_SHELL_WRAPPER_NAME = ghc/ghc.wrapper ghc_stage2_SHELL_WRAPPER_NAME = ghc/ghc.wrapper ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] cross-compiler-alienless: make sure that the C compiler is noted in the cross-ghc's wrapper (420aff3)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : cross-compiler-alienless http://hackage.haskell.org/trac/ghc/changeset/420aff3ab85d9b0102de636de3244d9d1fa68abc --- commit 420aff3ab85d9b0102de636de3244d9d1fa68abc Author: Gabor Greif ggr...@gmail.com Date: Fri Aug 31 14:34:44 2012 +0200 make sure that the C compiler is noted in the cross-ghc's wrapper --- rules/shell-wrapper.mk |3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/rules/shell-wrapper.mk b/rules/shell-wrapper.mk index 2f27357..55eabfa 100644 --- a/rules/shell-wrapper.mk +++ b/rules/shell-wrapper.mk @@ -36,8 +36,10 @@ all_$1_$2 : $$(INPLACE_BIN)/$$($1_$2_PROG) ifeq $(BuildingCrossCompiler) YES $1_$2_PGMGCC = $$(CC_STAGE1) +$1_$2_INSTALL_CROSS_SHELL_WRAPPER_EXTRA = (echo 'pgmgcc=$$($1_$2_PGMGCC)'; echo) else $1_$2_PGMGCC = $$(WhatGccIsCalled) +$1_$2_INSTALL_CROSS_SHELL_WRAPPER_EXTRA = echo endif @@ -83,6 +85,7 @@ install_$1_$2_wrapper: echo 'topdir=$$(topdir)' $$(WRAPPER) $$($1_$2_SHELL_WRAPPER_EXTRA) $$($1_$2_INSTALL_SHELL_WRAPPER_EXTRA) + $$($1_$2_INSTALL_CROSS_SHELL_WRAPPER_EXTRA) $$(WRAPPER) cat $$($1_$2_SHELL_WRAPPER_NAME) $$(WRAPPER) $$(EXECUTABLE_FILE) $$(WRAPPER) ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: make sure to remove the right link before calling 'ln -s' (could we use 'ln -sf'?) (b660cc0)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/b660cc0b3f6ea09ecc7f8fdef9ac79704c3ccaf0 --- commit b660cc0b3f6ea09ecc7f8fdef9ac79704c3ccaf0 Author: Gabor Greif ggr...@gmail.com Date: Thu Aug 30 15:02:56 2012 +0200 make sure to remove the right link before calling 'ln -s' (could we use 'ln -sf'?) --- ghc/ghc.mk |4 ++-- utils/ghc-pkg/ghc.mk |4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/ghc/ghc.mk b/ghc/ghc.mk index a13f03b..c45e288 100644 --- a/ghc/ghc.mk +++ b/ghc/ghc.mk @@ -1,6 +1,6 @@ # - # -# (c) 2009 The University of Glasgow +# (c) 2009-2012 The University of Glasgow # # This file is part of the GHC build system. # @@ -159,7 +159,7 @@ ifeq $(Windows) NO install: install_ghc_link .PHONY: install_ghc_link install_ghc_link: - $(call removeFiles,$(DESTDIR)$(bindir)/ghc) + $(call removeFiles,$(DESTDIR)$(bindir)/$(CrossCompilePrefix)ghc) $(LN_S) $(CrossCompilePrefix)ghc-$(ProjectVersion) $(DESTDIR)$(bindir)/$(CrossCompilePrefix)ghc else # On Windows we install the main binary as $(bindir)/ghc.exe diff --git a/utils/ghc-pkg/ghc.mk b/utils/ghc-pkg/ghc.mk index 8ec3fd0..ba553d2 100644 --- a/utils/ghc-pkg/ghc.mk +++ b/utils/ghc-pkg/ghc.mk @@ -1,6 +1,6 @@ # - # -# (c) 2009 The University of Glasgow +# (c) 2009-2012 The University of Glasgow # # This file is part of the GHC build system. # @@ -116,7 +116,7 @@ install: install_utils/ghc-pkg_link .PHONY: install_utils/ghc-pkg_link install_utils/ghc-pkg_link: $(call INSTALL_DIR,$(DESTDIR)$(bindir)) - $(call removeFiles,$(DESTDIR)$(bindir)/ghc-pkg) + $(call removeFiles,$(DESTDIR)$(bindir)/$(CrossCompilePrefix)ghc-pkg) $(LN_S) $(CrossCompilePrefix)ghc-pkg-$(ProjectVersion) $(DESTDIR)$(bindir)/$(CrossCompilePrefix)ghc-pkg endif ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] cross-compiler-alienless: Merge branch 'master' of https://github.com/ghc/ghc into cross-compiler-alienless (b8f6e55)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : cross-compiler-alienless http://hackage.haskell.org/trac/ghc/changeset/b8f6e55ec9fcdf44cff96b57e523e94c6c4cf9ba --- commit b8f6e55ec9fcdf44cff96b57e523e94c6c4cf9ba Merge: 90f7395... 4eb02c1... Author: Gabor Greif ggr...@gmail.com Date: Mon Aug 27 14:46:40 2012 +0200 Merge branch 'master' of https://github.com/ghc/ghc into cross-compiler-alienless aclocal.m4 | 16 +- compiler/basicTypes/BasicTypes.lhs |2 +- compiler/basicTypes/RdrName.lhs| 21 +- compiler/cmm/Cmm.hs| 15 +- compiler/cmm/CmmBuildInfoTables.hs | 99 ++-- compiler/cmm/CmmCallConv.hs| 52 +- compiler/cmm/CmmContFlowOpt.hs | 95 +++- compiler/cmm/CmmCvt.hs |2 +- compiler/cmm/CmmExpr.hs|2 + compiler/cmm/CmmInfo.hs| 110 +++-- compiler/cmm/CmmLayoutStack.hs | 153 -- compiler/cmm/CmmLint.hs|6 +- compiler/cmm/CmmMachOp.hs |8 +- compiler/cmm/CmmNode.hs| 36 +- compiler/cmm/CmmOpt.hs | 22 +- compiler/cmm/CmmParse.y|9 +- compiler/cmm/CmmPipeline.hs| 145 -- compiler/cmm/CmmProcPoint.hs | 26 +- compiler/cmm/CmmRewriteAssignments.hs | 29 +- compiler/cmm/CmmSink.hs| 580 ++-- compiler/cmm/MkGraph.hs| 232 + compiler/cmm/OldCmm.hs | 20 +- compiler/cmm/PprC.hs |4 +- compiler/cmm/PprCmm.hs |2 +- compiler/cmm/PprCmmDecl.hs |2 - compiler/codeGen/CgCallConv.hs | 76 ++-- compiler/codeGen/CgClosure.lhs | 19 +- compiler/codeGen/CgCon.lhs | 13 +- compiler/codeGen/CgForeignCall.hs |6 +- compiler/codeGen/CgHeapery.lhs | 12 +- compiler/codeGen/CgInfoTbls.hs | 26 +- compiler/codeGen/CgMonad.lhs | 13 +- compiler/codeGen/CgTailCall.lhs| 14 +- compiler/codeGen/CgUtils.hs| 255 ++--- compiler/codeGen/ClosureInfo.lhs | 37 +- compiler/codeGen/CodeGen/Platform.hs | 67 +++ compiler/codeGen/CodeGen/Platform/ARM.hs |9 + compiler/codeGen/CodeGen/Platform/NoRegs.hs|8 + compiler/codeGen/CodeGen/Platform/PPC.hs |9 + compiler/codeGen/CodeGen/Platform/PPC_Darwin.hs| 10 + compiler/codeGen/CodeGen/Platform/SPARC.hs |9 + compiler/codeGen/CodeGen/Platform/X86.hs |9 + compiler/codeGen/CodeGen/Platform/X86_64.hs|9 + compiler/codeGen/StgCmm.hs | 129 ++--- compiler/codeGen/StgCmmBind.hs | 525 ++ compiler/codeGen/StgCmmClosure.hs | 55 +- compiler/codeGen/StgCmmCon.hs | 71 ++- compiler/codeGen/StgCmmEnv.hs | 29 +- compiler/codeGen/StgCmmExpr.hs | 415 +++ compiler/codeGen/StgCmmForeign.hs | 38 +- compiler/codeGen/StgCmmHeap.hs | 185 --- compiler/codeGen/StgCmmLayout.hs | 59 ++- compiler/codeGen/StgCmmMonad.hs| 206 +--- compiler/codeGen/StgCmmPrim.hs |2 +- compiler/codeGen/StgCmmProf.hs | 46 +- compiler/codeGen/StgCmmUtils.hs| 498 +++-- compiler/coreSyn/CoreSyn.lhs | 13 + compiler/coreSyn/CoreUtils.lhs |2 +- compiler/coreSyn/MkCore.lhs|6 +- compiler/coreSyn/TrieMap.lhs | 111 - compiler/deSugar/Coverage.lhs | 98 +++-- compiler/deSugar/Desugar.lhs |3 +- compiler/deSugar/DsMeta.hs | 253 ++ compiler/ghc.cabal.in |8 + compiler/ghc.mk| 14 - compiler/hsSyn/Convert.lhs | 92 ++- compiler/hsSyn/HsExpr.lhs |2 +- compiler/hsSyn/HsSyn.lhs | 77 ++-- compiler/iface/BinIface.hs |5 +- compiler/iface/MkIface.lhs | 15 +- compiler/llvmGen/LlvmCodeGen.hs
[commit: ghc] cross-compiler-alienless: Merge branch 'master' of https://github.com/ghc/ghc into cross-compiler-alienless (c7ec369)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : cross-compiler-alienless http://hackage.haskell.org/trac/ghc/changeset/c7ec369af4ce00476090f02c3f374a081eb5a71b --- commit c7ec369af4ce00476090f02c3f374a081eb5a71b Merge: 6741921... b660cc0... Author: Gabor Greif ggr...@gmail.com Date: Thu Aug 30 15:06:24 2012 +0200 Merge branch 'master' of https://github.com/ghc/ghc into cross-compiler-alienless ghc/ghc.mk |4 ++-- utils/ghc-pkg/ghc.mk |4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] cross-compiler-alienless's head updated: Merge branch 'master' of https://github.com/ghc/ghc into cross-compiler-alienless (c7ec369)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc Branch 'cross-compiler-alienless' now includes: 1d094f9... Remove nativeGen/PprInstruction.hs cc3d982... Merge branch 'master' of darcs.haskell.org:/srv/darcs//ghc 6c2c07c... bugfix: cope when a source reg is dead at a join point f68b427... Fixes for the stack layout algorithm to handle join points fe3753e... Merge sinking and inlining to get better results. f1ed6a1... New codegen: do not split proc-points when using the NCG 1cea9a5... optimise away some unnecessary stack checks a4b249c... Small optimisation to the code generated for CAFs e26161f... no need to removeUnreachableBlocks 9388337... bug fixes for the sinker a2e0fbe... bug fix for control-flow optimisation a0020c1... fix merge bugs 8b31090... fix warnings a25c974... fix haddock parse error 6f346d4... Merge branch 'master' of darcs.haskell.org:/srv/darcs//ghc 9e7acbe... Fix compile on linux-powerpc64. (#7084). 598ee1a... Fix #7087 (integer overflow in getDelayTarget()) d236142... Improve code generated when real registers are clobbered 6ede006... Explicitly share some return continuations 7930221... Don't shortcut call-returns when not splitting proc points dae976c... add a sinking pass before stack layout (currently disabled) 08c16ba... Code reformatting a915d9b... Inline into the last node 40b6598... Improve accuracy of memory conflict tracking 325f2f3... Disable the mini-inliner when using the new codegen 9568636... No need to do removeDeadAssignments, just do cmmLiveness instead 3ae875c... Eliminate r = r assignments. 4aaa3c1... fix a bug in the inliner 46adcee... comment wibble 15e4f93... applying simonpj's fix from #7022 (with 80-col reformatting) 4200c4a... FloutOut.wrapTick: don't forget to tick the args of a constructor app b11a5ef... comments and refactoring 7974afb... Fix references to repositories in the user guide 89900ff... Merge branch 'master' of darcs.haskell.org:/srv/darcs//ghc 0e7aad2... De-orphan the Outputable Char instance 14b05c0... Whitespace only in hsSyn/HsSyn.lhs 5ae0cea... Whitespace only in utils/Outputable.lhs be691e1... Remove another unnecessary Platform argument e9a587a... De-orphan the Outputable Fingerprint instance 541cc50... Remove an unsafe use of head (#7118) ecc43c1... Convert prefix uses of () to infix 2d2650b... Move linker flags into the settings file; fixes #4862 c17f301... Remove LD_X; it's no longer used ddf9d40... Build fix 34b2906... Whitespace only in compiler/simplCore/Simplify.lhs 52307c1... Rename trace_dump to dump 13f3a31... Remove pprDefiniteTrace c9820b2... Merge branch 'master' of darcs.haskell.org:/srv/darcs//ghc 1235c27... Improve some code layout in SysTools 668151c... More more ld-related settings into the settings file c7c2d6c... Whitespace only in UniqFM 4cdfe67... Define LdCmd in the bindist configure.ac 5620662... finish a comment 290ec75... Add a comment to explain why the FCode monad is lazy 6228e31... Use ReturnedTo when generating safe foreign calls 22d5822... Foreign calls may clobber caller-saves registers f67a8b8... Continue by jumping to the top-of-stack after a safe foreign call 82fa790... Fix two bugs in the sinker. cf3fb95... MO_F_Mul and MO_F_Add are commutative 7db26e0... comment 190c555... fix warning 6997bb5... Don't define STOLEN_X86_REGS in Cmm.h 8e7fb28... Start separating out the RTS and Haskell imports of MachRegs.h e6ef5ab... Make tablesNextToCode dynamic f917eeb... Add Unregisterised as a field in the settings file 46b5c19... Define callerSaves for all platforms 8d3e9fd... small cleanup 0b75e45... Generate one fewer temps per heap allocation b534f42... Make lint check for undefined variables in Cmm 149e04b... A closure with void args only should be a function, not a thunk ddd6af0... Cleanup and fixes to profiling 5c3f13f... Fix update frames for profiling 3cf6050... fix maybeSaveCostCentre: cases were reversed bccd9e8... Add missing cases in hand-coded instance Eq GlobalReg 1469a12... Node calling convs should use R1 even if it isn't a register df7a20c... Eliminate r = r in mkAssign 313740e... fix a warning ef58afe... Small optimisation babe3c6... entryHeapCheck: fix calls to stg_gc_fun and stg_gc_enter_1 69cda9e... maybeInvertComparison: remove floating-point comparisons 396f090... Fix a bug in the handling of recent_activity 62ab993... fix warning 1edad87... Add missing flag for respecting EXTRA_LIBDIR c2a532a... Set the value of Unregisterised in the bindist configure.ac 415598b... Merge branch 'master' of darcs.haskell.org:/srv/darcs//ghc 50f5016... Add cast to placate gcc 4.6
[commit: ghc] master: v7.4.1 is required for building, so we can drop the v7.2.1 workaround (bcef1e28)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/bcef1e2847ce009f0b139abce1fb664733768c50 --- commit bcef1e2847ce009f0b139abce1fb664733768c50 Author: Gabor Greif ggr...@gmail.com Date: Wed Aug 29 14:47:02 2012 +0200 v7.4.1 is required for building, so we can drop the v7.2.1 workaround --- mk/config.mk.in | 10 -- 1 files changed, 0 insertions(+), 10 deletions(-) diff --git a/mk/config.mk.in b/mk/config.mk.in index a3f75cd..0005ab0 100644 --- a/mk/config.mk.in +++ b/mk/config.mk.in @@ -526,16 +526,6 @@ else GhcCanonVersion = $(GhcMajVersion)$(GhcMinVersion) endif -# Work around #5453, which causes the stage1 compiler to segfault when -# built with 7.2.1. -ifeq $(GhcVersion) 7.2.1 -CONF_HC_OPTS_STAGE0 += -fno-full-laziness -compiler/cmm/Bitmap_HC_OPTS += -ffull-laziness -# urgh. Compiling cmm/Bitmap.hs with -fno-full-laziness results in -# gmp: overflow in mpz type -# for some unknown reason, so turn full-laziness back on for this module. -endif - GHC_PACKAGE_DB_FLAG = @GHC_PACKAGE_DB_FLAG@ #- ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typo (c7a7c49)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/c7a7c49a81fa92bc46d04a9200be2afc5e4d1338 --- commit c7a7c49a81fa92bc46d04a9200be2afc5e4d1338 Author: Gabor Greif ggr...@gmail.com Date: Tue Jul 24 13:01:08 2012 +0200 typo --- includes/ghc.mk |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/includes/ghc.mk b/includes/ghc.mk index 73704b4..2cf835d 100644 --- a/includes/ghc.mk +++ b/includes/ghc.mk @@ -113,7 +113,7 @@ endif @echo #define BUILD_VENDOR \$(HostVendor_CPP)\ $@ @echo #define HOST_VENDOR \$(TargetVendor_CPP)\ $@ @echo $@ - @echo /* These TARGET macros are for backwards compatibily... DO NOT USE! */ $@ + @echo /* These TARGET macros are for backwards compatibility... DO NOT USE! */ $@ @echo #define TargetPlatform_TYPE $(TargetPlatform_CPP) $@ @echo #define $(TargetPlatform_CPP)_TARGET 1 $@ @echo #define $(TargetArch_CPP)_TARGET_ARCH 1 $@ ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: GHCConstants.h should not contain preprocessor definitions (ab14d99)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/ab14d99b524cef7fe412e30f7607fc76da72563a --- commit ab14d99b524cef7fe412e30f7607fc76da72563a Author: Gabor Greif ggr...@gmail.com Date: Fri Jul 27 23:12:49 2012 +0200 GHCConstants.h should not contain preprocessor definitions --- includes/mkDerivedConstants.c | 25 + 1 files changed, 25 insertions(+), 0 deletions(-) diff --git a/includes/mkDerivedConstants.c b/includes/mkDerivedConstants.c index 4042498..3fcf128 100644 --- a/includes/mkDerivedConstants.c +++ b/includes/mkDerivedConstants.c @@ -79,8 +79,12 @@ field_offset_(str(s_type,field),s_type,field); /* An access macro for use in C-- sources. */ +#if defined(GEN_HASKELL) +#define struct_field_macro(str) /* nothing */ +#else #define struct_field_macro(str) \ printf(#define str (__ptr__) REP_ str [__ptr__+OFFSET_ str ]\n); +#endif /* Outputs the byte offset and MachRep for a field */ #define struct_field(s_type, field)\ @@ -125,8 +129,12 @@ closure_size(s_type) /* An access macro for use in C-- sources. */ +#if defined(GEN_HASKELL) +#define closure_field_macro(str) /* nothing */ +#else #define closure_field_macro(str) \ printf(#define str (__ptr__) REP_ str [__ptr__+SIZEOF_StgHeader+OFFSET_ str ]\n); +#endif #define closure_field_offset_(str, s_type,field) \ def_offset(str, OFFSET(s_type,field) - TYPE_SIZE(StgHeader)); @@ -134,8 +142,12 @@ #define closure_field_offset(s_type,field) \ closure_field_offset_(str(s_type,field),s_type,field) +#if defined(GEN_HASKELL) +#define closure_payload_macro(str) /* nothing */ +#else #define closure_payload_macro(str) \ printf(#define str (__ptr__,__ix__) W_[__ptr__+SIZEOF_StgHeader+OFFSET_ str + WDS(__ix__)]\n); +#endif #define closure_payload(s_type,field) \ closure_field_offset_(str(s_type,field),s_type,field); \ @@ -164,26 +176,39 @@ def_offset(str(s_type,field), OFFSET(s_type,field) - TYPE_SIZE(StgHeader) - TYPE_SIZE(StgTSOProfInfo)); /* Full byte offset for a TSO field, for use from Cmm */ +#if defined(GEN_HASKELL) +#define tso_field_offset_macro(str) /* nothing */ +#else #define tso_field_offset_macro(str) \ printf(#define TSO_OFFSET_ str (SIZEOF_StgHeader+SIZEOF_OPT_StgTSOProfInfo+OFFSET_ str )\n); +#endif #define tso_field_offset(s_type, field) \ tso_payload_offset(s_type, field); \ tso_field_offset_macro(str(s_type,field)); +#if defined(GEN_HASKELL) +#define tso_field_macro(str) /* nothing */ +#else #define tso_field_macro(str) \ printf(#define str (__ptr__) REP_ str [__ptr__+TSO_OFFSET_ str ]\n) +#endif + #define tso_field(s_type, field) \ field_type(s_type, field); \ tso_field_offset(s_type,field);\ tso_field_macro(str(s_type,field)) +#if defined(GEN_HASKELL) +#define opt_struct_size(s_type, option) /* nothing */ +#else #define opt_struct_size(s_type, option) \ printf(#ifdef #option \n);\ printf(#define SIZEOF_OPT_ #s_type SIZEOF_ #s_type \n); \ printf(#else\n); \ printf(#define SIZEOF_OPT_ #s_type 0\n); \ printf(#endif\n\n); +#endif #define FUN_OFFSET(sym) (OFFSET(Capability,f.sym) - OFFSET(Capability,r)) ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typos (93c03f8)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/93c03f84c4bfebea5faeb60d75d708a578413e4a
---
commit 93c03f84c4bfebea5faeb60d75d708a578413e4a
Author: Gabor Greif ggr...@gmail.com
Date: Fri Jul 27 23:29:21 2012 +0200
typos
---
rts/parallel/Pack.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/rts/parallel/Pack.c b/rts/parallel/Pack.c
index fe64a75..58fe7ed 100644
--- a/rts/parallel/Pack.c
+++ b/rts/parallel/Pack.c
@@ -2801,7 +2801,7 @@ UnpackFetchMe (StgWord ***bufptrP, StgClosure **graphP) {
ToDo: implement packing of MUT_ARRAYs
*/
-//@cindex UnackArray
+//@cindex UnpackArray
static void
UnpackArray(StgWord ***bufptrP, StgClosure *graph)
{
@@ -2995,7 +2995,7 @@ UnpackPAP(StgWord ***bufptrP, StgClosure *graph)
//ASSERT(type==BLACKHOLE || type==CAF_BLACKHOLE || type==BLACKHOLE_BQ);
IF_PAR_DEBUG(pack,
-belch(*** UnackPAP @ %p: UPDATE_FRAME,
+belch(*** UnpackPAP @ %p: UPDATE_FRAME,
p));
((StgUpdateFrame *)p)-header.info = (StgInfoTable*)*bufptr++;
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] cross-compiler-alienless: initial batch of cross compiler changes (90f7395)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : cross-compiler-alienless http://hackage.haskell.org/trac/ghc/changeset/90f73952dd713c641ef9313e384a2f0b24fc69cb --- commit 90f73952dd713c641ef9313e384a2f0b24fc69cb Author: Gabor Greif ggr...@gmail.com Date: Mon Jul 30 00:09:39 2012 +0200 initial batch of cross compiler changes notable missing functionality is: - the inclusion of the cross-gcc into the wrapper script - disabling of some extra packages - specifying linker per stage - some patching of libffi/.../powerpc/ffi.c These features will be included in subsequent commits. aclocal.m4 |8 - compiler/nativeGen/PPC/Ppr.hs |7 +++- configure.ac| 75 --- ghc.mk |3 +- ghc/ghc.mk | 15 includes/ghc.mk | 71 includes/mkDerivedConstants.c | 41 +++-- includes/mkSizeMacros.cross.awk |1 + libffi/ghc.mk | 25 + mk/config.mk.in |4 +- rules/build-dependencies.mk |3 ++ rules/build-package-data.mk |7 rules/build-package-way.mk |2 +- rules/distdir-way-opts.mk |4 ++ rules/hs-suffix-rules-srcdir.mk | 11 -- rules/shell-wrapper.mk |9 - 16 files changed, 236 insertions(+), 50 deletions(-) Diff suppressed because of size. To see it, use: git show 90f73952dd713c641ef9313e384a2f0b24fc69cb ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] branch 'cross-compiler-alienless' created
Repository : ssh://darcs.haskell.org//srv/darcs/ghc New branch : cross-compiler-alienless Referencing: 90f73952dd713c641ef9313e384a2f0b24fc69cb ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
'sizeof' infestation in rts/
Hi all,
this week I managed to build a cross-compiler, but on trying hello
Cross-World! I got a SEGV. After debugging it, it turned out that the
usage of sizeofW(...) is the culprit. I'll fix this on my (still
local) branch.
But...
Upon closer inspection of the rts/... files I found many uses of
sizeof(W_) which are totally cross-unfriendly.
I'd like to submit sizeof(W_) --- SIZEOF_HSWORD to master directly.
Okay to commit patches of this style (see below)?
As an extra bonus compilation will be faster, since SIZEOF_HSWORD is
resolved in the preprocessor.
Cheers,
Gabor
-
diff --git a/rts/sm/BlockAlloc.c b/rts/sm/BlockAlloc.c
index 0b8955c..0a8b387 100644
--- a/rts/sm/BlockAlloc.c
+++ b/rts/sm/BlockAlloc.c
@@ -1,6 +1,6 @@
/*
-
*
- * (c) The GHC Team 1998-2008
+ * (c) The GHC Team 1998-2012
*
* The block allocator and free list manager.
*
@@ -347,7 +347,7 @@ allocGroup (nat n)
if (ln == MAX_FREE_LIST) {
#if 0 /* useful for debugging fragmentation */
if ((W_)mblocks_allocated * BLOCKS_PER_MBLOCK * BLOCK_SIZE_W
- - (W_)((n_alloc_blocks - n) * BLOCK_SIZE_W)
(2*1024*1024)/sizeof(W_)) {
+ - (W_)((n_alloc_blocks - n) * BLOCK_SIZE_W)
(2*1024*1024)/SIZEOF_HSWORD) {
debugBelch(Fragmentation, wanted %d blocks:, n);
RtsFlags.DebugFlags.block_alloc = 1;
checkFreeListSanity();
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: 'sizeof' infestation in rts/
Hi Simon,
thanks for the quick answer!
Sure, it can wait. Next week is fine.
The problem with 'sizeof' is that while building the cross-compiler
(on a 64-bit intel system, to be run on the same system) bakes in the
sizeof(void*) into the binary (=8), but the target system is 32-bit
only, so sizeof(void*) should be 4. This messes up all the
offset/array-size calculations in GC etc.
The value of SIZEOF_HSWORD comes from configure and correctly expands
to 4 in my case. So rts will be correctly built with the size
assumptions of the targeted platform.
Hope this explains it.
Cheers,
Gabor
On 7/27/12, Simon Marlow marlo...@gmail.com wrote:
Please hold off on this until next week as I'm on holiday right now.
What's the problem with sizeof?
Cheers,
Simon
From: Gabor Greif
Sent: 27/07/2012 10:05
To: cvs-ghc; Simon Marlow
Subject: 'sizeof' infestation in rts/
Hi all,
this week I managed to build a cross-compiler, but on trying hello
Cross-World! I got a SEGV. After debugging it, it turned out that the
usage of sizeofW(...) is the culprit. I'll fix this on my (still
local) branch.
But...
Upon closer inspection of the rts/... files I found many uses of
sizeof(W_) which are totally cross-unfriendly.
I'd like to submit sizeof(W_) --- SIZEOF_HSWORD to master directly.
Okay to commit patches of this style (see below)?
As an extra bonus compilation will be faster, since SIZEOF_HSWORD is
resolved in the preprocessor.
Cheers,
Gabor
-
diff --git a/rts/sm/BlockAlloc.c b/rts/sm/BlockAlloc.c
index 0b8955c..0a8b387 100644
--- a/rts/sm/BlockAlloc.c
+++ b/rts/sm/BlockAlloc.c
@@ -1,6 +1,6 @@
/*
-
*
- * (c) The GHC Team 1998-2008
+ * (c) The GHC Team 1998-2012
*
* The block allocator and free list manager.
*
@@ -347,7 +347,7 @@ allocGroup (nat n)
if (ln == MAX_FREE_LIST) {
#if 0 /* useful for debugging fragmentation */
if ((W_)mblocks_allocated * BLOCKS_PER_MBLOCK * BLOCK_SIZE_W
- - (W_)((n_alloc_blocks - n) * BLOCK_SIZE_W)
(2*1024*1024)/sizeof(W_)) {
+ - (W_)((n_alloc_blocks - n) * BLOCK_SIZE_W)
(2*1024*1024)/SIZEOF_HSWORD) {
debugBelch(Fragmentation, wanted %d blocks:, n);
RtsFlags.DebugFlags.block_alloc = 1;
checkFreeListSanity();
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: 'sizeof' infestation in rts/
Okay, yes... I am the one who is confused. Sorry for the confusion, I got carried away in my grepping and code-reading and forgot that all of rts/ is built for target. Please disregard this thread, sorry for disturbing your vacation :-( Regretfully, Gabor PS: Anyway my internal commit, just in case: 1f9927c1b840a3bb1202f3e4b0e8444d1fc1d971 On 7/27/12, Simon Marlow marlo...@gmail.com wrote: I'm confused. Shouldn't the RTS be built for the target platform? ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: pgj2 (amd64 FreeBSD HEAD), build 757, Failure
I see, it should have been configured (by build.mk) to build the simple-integer library, right? Thanks and cheers, Gabor On 7/19/12, Ian Lynagh ig...@earth.li wrote: On Thu, Jul 19, 2012 at 04:31:31PM +0200, Gabor Greif wrote: Both FreeBSD bots failed in the same way: libraries/hpc/ghc.mk:3: libraries/hpc/dist-boot/package-data.mk: No such file or directory libraries/Cabal/Cabal/ghc.mk:3: libraries/Cabal/Cabal/dist-boot/package-data.mk: No such file or directory libraries/binary/ghc.mk:3: libraries/binary/dist-boot/package-data.mk: No such file or directory libraries/bin-package-db/ghc.mk:3: libraries/bin-package-db/dist-boot/package-data.mk: No such file or directory libraries/hoopl/ghc.mk:3: libraries/hoopl/dist-boot/package-data.mk: No such file or directory compiler/ghc.mk:450: compiler/stage1/package-data.mk: No such file or directory utils/hsc2hs/ghc.mk:14: utils/hsc2hs/dist/package-data.mk: No such file or directory ghc/ghc.mk:106: ghc/stage1/package-data.mk: No such file or directory Those messages are normal. The problem is: /usr/bin/ld: cannot find -lgmp gmake[1]: *** [libraries/integer-gmp/dist-install/build/libHSinteger-gmp-0.3.0.0-ghc7.5.20120719.so] Error 1 Thanks Ian ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: pgj2 (amd64 FreeBSD HEAD), build 757, Failure
Both FreeBSD bots failed in the same way: libraries/hpc/ghc.mk:3: libraries/hpc/dist-boot/package-data.mk: No such file or directory libraries/Cabal/Cabal/ghc.mk:3: libraries/Cabal/Cabal/dist-boot/package-data.mk: No such file or directory libraries/binary/ghc.mk:3: libraries/binary/dist-boot/package-data.mk: No such file or directory libraries/bin-package-db/ghc.mk:3: libraries/bin-package-db/dist-boot/package-data.mk: No such file or directory libraries/hoopl/ghc.mk:3: libraries/hoopl/dist-boot/package-data.mk: No such file or directory compiler/ghc.mk:450: compiler/stage1/package-data.mk: No such file or directory utils/hsc2hs/ghc.mk:14: utils/hsc2hs/dist/package-data.mk: No such file or directory ghc/ghc.mk:106: ghc/stage1/package-data.mk: No such file or directory What commit could have caused this? N.B. many other nightlies do not show this, perhaps FreeBSD-only (or exotic)? Thanks, Gabor On 7/19/12, Builder bit.buc...@galois.com wrote: pgj2 (amd64 FreeBSD HEAD), build 757 Build failed Details: http://darcs.haskell.org/ghcBuilder/builders/pgj2/757.html git clone| Success create mk/build.mk | Success get subrepos | Success repo versions| Success touching clean-check files | Success setting version date | Success booting | Success configuring | Success creating check-remove-before | Success compiling| Failure: Just (ExitFailure 2) Build failed Details: http://darcs.haskell.org/ghcBuilder/builders/pgj2/757.html ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typos in comments (9ef3b76)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/9ef3b765e85adb158dfe87f0c95fad290e12af93
---
commit 9ef3b765e85adb158dfe87f0c95fad290e12af93
Author: Gabor Greif ggr...@gmail.com
Date: Wed Jul 18 10:33:46 2012 +0200
typos in comments
this is also my test commit and
reduces the lines in the cross-
compiler branch
---
utils/genapply/GenApply.hs |6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/utils/genapply/GenApply.hs b/utils/genapply/GenApply.hs
index d00324f..e39f42e 100644
--- a/utils/genapply/GenApply.hs
+++ b/utils/genapply/GenApply.hs
@@ -1,6 +1,6 @@
{-# OPTIONS -cpp -fglasgow-exts #-}
{-# OPTIONS -w #-}
--- The above warning supression flag is a temporary kludge.
+-- The above warning suppression flag is a temporary kludge.
-- While working on this module you are encouraged to remove it and fix
-- any warnings in the module. See
-- http://hackage.haskell.org/trac/ghc/wiki/Commentary/CodingStyle#Warnings
@@ -179,7 +179,7 @@ mb_tag_node arity | Just tag - tagForArity arity =
mkTagStmt tag semi
mkTagStmt tag = text (R1 = R1 + ++ show tag)
genMkPAP regstatus macro jump ticker disamb
-no_load_regs-- don't load argumnet regs before jumping
+no_load_regs-- don't load argument regs before jumping
args_in_regs-- arguments are already in regs
is_pap args all_args_size fun_info_label
is_fun_case
@@ -223,7 +223,7 @@ genMkPAP regstatus macro jump ticker disamb
else shuffle_extra_args,
-- for a PAP, we have to arrange that the stack contains a
--- return address in the even that stg_PAP_entry fails its
+-- return address in the event that stg_PAP_entry fails its
-- heap check. See stg_PAP_entry in Apply.hc for details.
if is_pap
then text R2 = mkApplyInfoName this_call_args semi
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: typos in trace strings (which I observed in compile output) (3ae9f0f)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/3ae9f0f97583a3084a76cf7e31931bd9ef5f4173 --- commit 3ae9f0f97583a3084a76cf7e31931bd9ef5f4173 Author: Gabor Greif ggr...@gmail.com Date: Wed Jul 18 12:25:58 2012 +0200 typos in trace strings (which I observed in compile output) --- compiler/simplCore/SimplCore.lhs |2 +- compiler/specialise/Rules.lhs|2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/compiler/simplCore/SimplCore.lhs b/compiler/simplCore/SimplCore.lhs index d8c6732..731f551 100644 --- a/compiler/simplCore/SimplCore.lhs +++ b/compiler/simplCore/SimplCore.lhs @@ -586,7 +586,7 @@ simplifyPgmIO pass@(CoreDoSimplify max_iterations mode) -- about to begin, with '1' for the first | iteration_no max_iterations -- Stop if we've run out of iterations = WARN( debugIsOn (max_iterations 2) -, hang (ptext (sLit Simplifier baling out after) + int max_iterations +, hang (ptext (sLit Simplifier bailing out after) + int max_iterations + ptext (sLit iterations) + (brackets $ hsep $ punctuate comma $ map (int . simplCountN) (reverse counts_so_far))) diff --git a/compiler/specialise/Rules.lhs b/compiler/specialise/Rules.lhs index 498302a..0cf858e 100644 --- a/compiler/specialise/Rules.lhs +++ b/compiler/specialise/Rules.lhs @@ -726,7 +726,7 @@ match_co :: RuleEnv match_co renv subst (CoVarCo cv) co = match_var renv subst cv (Coercion co) match_co _ _ co1 _ - = pprTrace match_co baling out (ppr co1) Nothing + = pprTrace match_co bailing out (ppr co1) Nothing - rnMatchBndr2 :: RuleEnv - RuleSubst - Var - Var - RuleEnv ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: made comments C-style, so that no gcc-extension is needed (f25718b)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/f25718b039f8a31dc158ab8dad3c9523ee6f87e5
---
commit f25718b039f8a31dc158ab8dad3c9523ee6f87e5
Author: Gabor Greif ggr...@gmail.com
Date: Wed Jul 18 12:14:16 2012 +0200
made comments C-style, so that no gcc-extension is needed
also removed an unnecessary 'struct' tag (since the struct is
not recursive); this is in line with the other struct definitions
fixed a typo, updated copyright
it remains to remove the tabs and align the structure members
accordingly
---
includes/stg/Regs.h | 10 +-
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/includes/stg/Regs.h b/includes/stg/Regs.h
index bfc3d4b..bf17b7e 100644
--- a/includes/stg/Regs.h
+++ b/includes/stg/Regs.h
@@ -1,6 +1,6 @@
/*
-
*
- * (c) The GHC Team, 1998-2009
+ * (c) The GHC Team, 1998-2012
*
* Registers in the STG machine.
*
@@ -21,7 +21,7 @@
*
* The register set is backed by a table in memory (struct
* StgRegTable). If a particular STG register is not mapped to a
- * machine register, then the apprpriate slot in this table is used
+ * machine register, then the appropriate slot in this table is used
* instead.
*
* This table is itself pointed to by another register, BaseReg. If
@@ -58,7 +58,7 @@ typedef union {
*register, probably because there's a shortage of real registers.
* 2) caller-saves registers are saved across a CCall
*/
-typedef struct StgRegTable_ {
+typedef struct {
StgUnion rR1;
StgUnion rR2;
StgUnion rR3;
@@ -80,13 +80,13 @@ typedef struct StgRegTable_ {
StgPtr rSpLim;
StgPtr rHp;
StgPtr rHpLim;
- struct CostCentreStack_ * rCCCS; // current cost-centre-stack
+ struct CostCentreStack_ * rCCCS; /* current cost-centre-stack */
struct StgTSO_ * rCurrentTSO;
struct nursery_ *rNursery;
struct bdescr_ * rCurrentNursery; /* Hp/HpLim point into this block */
struct bdescr_ * rCurrentAlloc; /* for allocation using allocate() */
StgWord rHpAlloc;/* number of *bytes* being allocated in heap */
- StgWord rRet; // holds the return code of the thread
+ StgWord rRet; /* holds the return code of the thread */
} StgRegTable;
#if IN_STG_CODE
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: use idiomatic type (6af4e5d)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/6af4e5dc8d224dff3c065843e065c3861320ec33
---
commit 6af4e5dc8d224dff3c065843e065c3861320ec33
Author: Gabor Greif ggr...@gmail.com
Date: Sun Jan 8 15:51:39 2012 +0100
use idiomatic type
---
includes/rts/prof/CCS.h |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/includes/rts/prof/CCS.h b/includes/rts/prof/CCS.h
index 2492bb3..e6c746b 100644
--- a/includes/rts/prof/CCS.h
+++ b/includes/rts/prof/CCS.h
@@ -1,6 +1,6 @@
/*
-
*
- * (c) The GHC Team, 2009
+ * (c) The GHC Team, 2009-2012
*
* Macros for profiling operations in STG code
*
@@ -107,7 +107,7 @@ typedef struct IndexTable_ {
CostCentre *cc;
CostCentreStack *ccs;
struct IndexTable_ *next;
-unsigned int back_edge;
+nat back_edge;
} IndexTable;
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: comment-only typo (9417612)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc On branch : master http://hackage.haskell.org/trac/ghc/changeset/9417612ec9d4b27cacd8100cf4831e3e703c976b --- commit 9417612ec9d4b27cacd8100cf4831e3e703c976b Author: Gabor Greif ggr...@gmail.com Date: Wed Jul 18 14:15:05 2012 +0200 comment-only typo --- includes/Cmm.h |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/includes/Cmm.h b/includes/Cmm.h index bfac1ee..1788122 100644 --- a/includes/Cmm.h +++ b/includes/Cmm.h @@ -1,6 +1,6 @@ /* - * - * (c) The University of Glasgow 2004 + * (c) The University of Glasgow 2004-2012 * * This file is included at the top of all .cmm source files (and * *only* .cmm files). It defines a collection of useful macros for @@ -50,7 +50,7 @@ * StgTSO_what_next(CurrentTSO) = x * * where the StgTSO_what_next() macro is automatically generated by - * mkDerivedConstnants.c. If you need to access a field that doesn't + * mkDerivedConstants.c. If you need to access a field that doesn't * already have a macro, edit that file (it's pretty self-explanatory). * * -- */ ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[commit: ghc] master: First batch of file additions for cross-compiler: (6ae696a)
Repository : ssh://darcs.haskell.org//srv/darcs/ghc
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/6ae696a1d1f25bf52923a3dd1c3b4a08e2033bfd
---
commit 6ae696a1d1f25bf52923a3dd1c3b4a08e2033bfd
Author: Gabor Greif ggr...@gmail.com
Date: Wed Jul 18 15:03:31 2012 +0200
First batch of file additions for cross-compiler:
new file: ghc/ghc-cross.wrapper
new file: includes/mkDerivedConstants.cross.awk
new file: includes/mkSizeMacros.cross.awk
new file: rules/cross-compiling.mk
These are expected to sit quietly in the tree until
the rest of the machinery matures on an (upcoming)
branch. Reviews will begin to make sense as soon as
that has happened. Anyway, comments are welcome. See
http://www.haskell.org/pipermail/cvs-ghc/2012-July/074456.html
for background.
Disclaimer: these source files are not (yet) up to the
quality standards set by the rest of the tree.
Cleanups, move-arounds and rewrites (i.e. .awk - .hs), as
well as additional comments and documentation will happen
as soon as the basic functionality of a cross-compiler is
working reliably.
ghc/ghc-cross.wrapper |1 +
includes/mkDerivedConstants.cross.awk | 350 +++
includes/mkSizeMacros.cross.awk | 82 +++
rules/{all-target.mk = cross-compiling.mk} | 18 +-
4 files changed, 445 insertions(+), 6 deletions(-)
Diff suppressed because of size. To see it, use:
git show 6ae696a1d1f25bf52923a3dd1c3b4a08e2033bfd
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[PING] Cross-compiler next steps
ping :-) On 7/11/12, Gabor Greif ggr...@gmail.com wrote: Hi all, coming back on the cross compilers issue. As you know, I had a working PPC cross compiler in a github fork that did not require any alien tools, just a cross-gcc to distill all characteristics of the platform that are needed by GHC. Unfortunately this fork has bitrotten somewhat (mostly because the alien-cross merge). Now I have some energy to port my solution to HEAD. The question is how you feel about my suggested approach: 1) get all non-conflicting changes on HEAD (i.e. put them into the official repo). These are: * the 'awk' scripts (*) and makefile changes to create the target constants (sizes, offsets). 2) make sure there are no regressions for some nights, so we can be confident that I did not break anything. 3) The currently conflicting (though not invasively) modifications. These include: * the 2-stage build changes (**) still needed in makefiles and configury (if still necessary). These include --with-cross-gcc detection etc. * the modifications to the GHC wrapper script. I have something small working, but an experienced GHC implementor surely can point me to a more general approach. * modifications to the building of the libraries, so that the correct cross-compiling options are passed to the package builds. 4) again, observe the bot for some time whether stuff borked, get some beta testers into the boat, implement feedback. 5) Add some tests that exercise cross compilation (e.g. x86_64 host, i386 target or x64 target with 32-bit pointer API on Linux). For setting up these I'd ask for some help from the more experienced of you. It remains to clarify whether I can obtain commit privileges for the above steps w.r.t. the HEAD branch of the official repo. I believe that these privileges would speed up the porting a lot. For 1) I would ask for post-commit review and for 3) I'd suggest pre-integration on a branch followed by a pre-merge review, then merge to HEAD. Alternatively, I'd just try to pull this off in a new github fork and ask you for pulls to the official (if you feel hesitant about granting me the privileges). A bit of background... I got official permission to evaluate Haskell in an embedded programming context for research projects from my employer (Alcatel-Lucent). I have to present concrete evidence that this direction of attack results in usable artifacts for our research this summer, as the proposal for continuation of said activity will be concluded in September. Thanks for your attention and in hope for cooperation, cheers, Gabor (*) I promise to convert all 'awk' scripts to 'runhaskell' scripts as soon as the cross-compilation works from vanilla HEAD for PowerPC and ARM. (**) ultimately we'll need a 3-stage build for cross compilation, because I consider Template Haskell as essential. ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Cross-compiler next steps
Hi all, coming back on the cross compilers issue. As you know, I had a working PPC cross compiler in a github fork that did not require any alien tools, just a cross-gcc to distill all characteristics of the platform that are needed by GHC. Unfortunately this fork has bitrotten somewhat (mostly because the alien-cross merge). Now I have some energy to port my solution to HEAD. The question is how you feel about my suggested approach: 1) get all non-conflicting changes on HEAD (i.e. put them into the official repo). These are: * the 'awk' scripts (*) and makefile changes to create the target constants (sizes, offsets). 2) make sure there are no regressions for some nights, so we can be confident that I did not break anything. 3) The currently conflicting (though not invasively) modifications. These include: * the 2-stage build changes (**) still needed in makefiles and configury (if still necessary). These include --with-cross-gcc detection etc. * the modifications to the GHC wrapper script. I have something small working, but an experienced GHC implementor surely can point me to a more general approach. * modifications to the building of the libraries, so that the correct cross-compiling options are passed to the package builds. 4) again, observe the bot for some time whether stuff borked, get some beta testers into the boat, implement feedback. 5) Add some tests that exercise cross compilation (e.g. x86_64 host, i386 target or x64 target with 32-bit pointer API on Linux). For setting up these I'd ask for some help from the more experienced of you. It remains to clarify whether I can obtain commit privileges for the above steps w.r.t. the HEAD branch of the official repo. I believe that these privileges would speed up the porting a lot. For 1) I would ask for post-commit review and for 3) I'd suggest pre-integration on a branch followed by a pre-merge review, then merge to HEAD. Alternatively, I'd just try to pull this off in a new github fork and ask you for pulls to the official (if you feel hesitant about granting me the privileges). A bit of background... I got official permission to evaluate Haskell in an embedded programming context for research projects from my employer (Alcatel-Lucent). I have to present concrete evidence that this direction of attack results in usable artifacts for our research this summer, as the proposal for continuation of said activity will be concluded in September. Thanks for your attention and in hope for cooperation, cheers, Gabor (*) I promise to convert all 'awk' scripts to 'runhaskell' scripts as soon as the cross-compilation works from vanilla HEAD for PowerPC and ARM. (**) ultimately we'll need a 3-stage build for cross compilation, because I consider Template Haskell as essential. ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: Cross-compiler next steps
Thanks for the heads-up, ARM is not that essential, and PowerPC would
be sufficient for my purposes a.t.m.
Cheers,
Gabor
On 7/11/12, Karel Gardas karel.gar...@centrum.cz wrote:
Hi Gabor,
thanks a lot for your effort on GHC cross-compilation. You are talking
about cross-compiling to ARM too. I would like to warn you here as GHC
HEAD build for ARM is broken since February due to LLVM backend
miscompiling stage2 -- at least that's my analysis of this so far. See
http://hackage.haskell.org/trac/ghc/ticket/7049
Hopefully once GHC/LLVM unbreaks I will be more than happy to test your
stuff for {solaris|linux}/x86-linux/arm cross-compilation.
Thanks!
Karel
On 07/11/12 12:19 PM, Gabor Greif wrote:
Hi all,
coming back on the cross compilers issue. As you know, I had a working
PPC cross compiler in a github fork that did not require any alien
tools, just a cross-gcc to distill all characteristics of the platform
that are needed by GHC.
Unfortunately this fork has bitrotten somewhat (mostly because the
alien-cross merge). Now I have some energy to port my solution to
HEAD. The question is how you feel about my suggested approach:
1) get all non-conflicting changes on HEAD (i.e. put them into the
official repo). These are:
* the 'awk' scripts (*) and makefile changes to create the target
constants (sizes, offsets).
2) make sure there are no regressions for some nights, so we can be
confident that I did not break anything.
3) The currently conflicting (though not invasively) modifications.
These include:
* the 2-stage build changes (**) still needed in makefiles and
configury (if still necessary). These include --with-cross-gcc
detection etc.
* the modifications to the GHC wrapper script. I have something
small working, but an experienced GHC implementor surely can point me
to a more general approach.
* modifications to the building of the libraries, so that the
correct cross-compiling options are passed to the package builds.
4) again, observe the bot for some time whether stuff borked, get some
beta testers into the boat, implement feedback.
5) Add some tests that exercise cross compilation (e.g. x86_64 host,
i386 target or x64 target with 32-bit pointer API on Linux). For
setting up these I'd ask for some help from the more experienced of
you.
It remains to clarify whether I can obtain commit privileges for the
above steps w.r.t. the HEAD branch of the official repo. I believe
that these privileges would speed up the porting a lot. For 1) I would
ask for post-commit review and for 3) I'd suggest pre-integration on a
branch followed by a pre-merge review, then merge to HEAD.
Alternatively, I'd just try to pull this off in a new github fork and
ask you for pulls to the official (if you feel hesitant about granting
me the privileges).
A bit of background... I got official permission to evaluate Haskell
in an embedded programming context for research projects from my
employer (Alcatel-Lucent). I have to present concrete evidence that
this direction of attack results in usable artifacts for our research
this summer, as the proposal for continuation of said activity will be
concluded in September.
Thanks for your attention and in hope for cooperation,
cheers,
Gabor
(*) I promise to convert all 'awk' scripts to 'runhaskell' scripts as
soon as the cross-compilation works from vanilla HEAD for PowerPC and
ARM.
(**) ultimately we'll need a 3-stage build for cross compilation,
because I consider Template Haskell as essential.
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
Typo in docs
forkIOWIthUnmask - forkIOWithUnmask in http://www.haskell.org/ghc/docs/latest/html/libraries/base/Control-Concurrent.html Cheers, Gabor ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [commit: testsuite] master: Test Trac #6035, #6036 (2f9e8f3)
Hi Simon,
I might be blind, but are these tests being run at all?
Cheers,
Gabor
On 4/25/12, Simon Peyton Jones simo...@microsoft.com wrote:
Repository : ssh://darcs.haskell.org//srv/darcs/testsuite
On branch : master
http://hackage.haskell.org/trac/ghc/changeset/2f9e8f3072921e1cfe7ed83445848e91573dda56
---
commit 2f9e8f3072921e1cfe7ed83445848e91573dda56
Author: Simon Peyton Jones simo...@microsoft.com
Date: Tue Apr 24 17:04:05 2012 +0100
Test Trac #6035, #6036
---
tests/polykinds/T6035.hs | 28
tests/polykinds/T6036.hs | 17 +
2 files changed, 45 insertions(+), 0 deletions(-)
diff --git a/tests/polykinds/T6035.hs b/tests/polykinds/T6035.hs
new file mode 100644
index 000..c9e4342
--- /dev/null
+++ b/tests/polykinds/T6035.hs
@@ -0,0 +1,28 @@
+{-# LANGUAGE PolyKinds, DataKinds, TypeFamilies, KindSignatures, GADTs,
TypeOperators
+ #-}
+
+module T6035 where
+
+data Nat = Zero | Succ Nat
+
+type family Sing (a :: k) :: k - *
+
+data SNat n where
+ SZero :: SNat Zero
+ SSucc :: SNat n - SNat (Succ n)
+
+data SList (a :: [k]) where
+ SNil :: SList '[]
+ SCons :: Sing h h - SList t - SList (h ': t)
+
+type instance Sing (a :: Nat) = SNat
+type instance Sing (a :: [k]) = SList
+
+nil :: SList '[]
+nil = SNil
+
+zero :: SList '[ '[] ]
+zero = SCons SNil SNil
+
+term :: SList '[ '[Zero], '[]]
+term = SCons (SCons SZero SNil) (SCons SNil SNil)
diff --git a/tests/polykinds/T6036.hs b/tests/polykinds/T6036.hs
new file mode 100644
index 000..4425c85
--- /dev/null
+++ b/tests/polykinds/T6036.hs
@@ -0,0 +1,17 @@
+{-# LANGUAGE DataKinds, TypeFamilies, PolyKinds, GADTs #-}
+
+module T6036 where
+
+data family Sing (a :: k)
+
+data instance Sing (a :: Maybe k) where
+ SNothing :: Sing 'Nothing
+ SJust :: Sing b - Sing ('Just b)
+
+data Nat = Zero | Succ Nat
+
+data instance Sing (a :: Nat) where
+ SZero :: Sing Zero
+ SSucc :: Sing n - Sing (Succ n)
+
+term = SJust SZero
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[PATCH] fix ticket #6001 by reporting the expected signature instead of the actual one
--- compiler/typecheck/TcInstDcls.lhs |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/compiler/typecheck/TcInstDcls.lhs b/compiler/typecheck/TcInstDcls.lhs index 64b839c..97b55f4 100644 --- a/compiler/typecheck/TcInstDcls.lhs +++ b/compiler/typecheck/TcInstDcls.lhs @@ -836,7 +836,7 @@ mkMethIds sig_fn clas tyvars dfun_ev_vars inst_tys sel_id ; inst_sigs - xoptM Opt_InstanceSigs ; if inst_sigs then checkTc (sig_ty `eqType` local_meth_ty) -(badInstSigErr sel_name sig_ty) +(badInstSigErr sel_name local_meth_ty) else addErrTc (misplacedInstSig sel_name hs_ty) ; return sig_ty } -- 1.7.9.3 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [LLVMdev] LLVM GHC Backend: Tables Next To Code
Chris said: I really like the idea of adding this as an inline asm blob at the start of a function, and biasing the actual address of the closure based on the size of the table. I'm not 100% confident that it will work (not being very familiar with TNTC) but it seems quite plausible and the impact on LLVM would be quite reasonable (some new calling convention work?) While reading this I had the idea that the LLVM code generator could watch out for the specific combination of inline asm and calling convention and strip off the fat. This way the code increase could be dealt with. OTOH the inline asm would still need to be target specific, which is very ugly. What about a new intrinsic which holds a reference to the global and creates the right assembly in the backend? Since the reference to the global (which is the table) would not be used otherwise, the linker could drop it, thus no code increase and no redundant global. Does this make sense? Cheers, Gabor ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[PING] Re: [PATCH] Many small typos and formatting
Ping. Anything wrong with this or just overlooked? Cheers, Gabor On 2/16/12, Gabor Greif ggr...@gmail.com wrote: See my previous mail... I believe I did precisely what you called for: https://github.com/ggreif/ghc/commits/misc-cleanups/ repo: https://github.com/ggreif/ghc.git branch: misc-cleanups Cheers, Gabor On 2/16/12, Simon Marlow marlo...@gmail.com wrote: On 15/02/2012 17:33, Gabor Greif wrote: Hi all, on my fork https://github.com/ggreif/ghc I have accumulated a number of small patches that are results of my browsing of the GHC sources and correcting minor issues. These are without functional impact and mostly apply to comments and formatting. By transferring these to mainline GHC I hope to lessen the distance of my fork to mainline. When pulling patches from mainline via github I sometimes use the fork queue (https://github.com/ggreif/ghc/forkqueue). By using the corresponding device (https://github.com/ghc/ghc/forkqueue) of the official mirror it might be quite easy to locate and apply the below patches. There might be an even better method by resorting to the command-line git tool. Alternatively I could set up a github clone of the original mirror (a topic branch maybe?) and push these changes there, so you would only need to pull them into the official repo. Please indicate how this can be handled best. The easiest for us would be if you could give us a repo URL that we can pull from. I've done this before when people have submitted patches via github - they make a branch on their github repo containing the changes, and give us the URL and branch name, we pull and validate the changes and push to GHC's repo. Cheers, Simon Cheers, Gabor https://github.com/ggreif/ghc/commit/77320765550bfa603052f0334d2d681134b63da0 https://github.com/ggreif/ghc/commit/598306aca074875bbe2cc025a5dd267a53a0d3bd done# https://github.com/ggreif/ghc/commit/fe443ed82538d04191212a08e44bd3f2e27613af https://github.com/ggreif/ghc/commit/95c2d50da10e49bb936e20a3b58fb29c7c5b4d13 https://github.com/ggreif/ghc/commit/3c79e184304ca38b8ed4484ccc3e8bac9724963a https://github.com/ggreif/ghc/commit/e47aa109b35ec02b544b4dec5640714c4ce66d74 https://github.com/ggreif/ghc/commit/e4c4cc978fbc1cc445e076adfd08765b4ef3a39c https://github.com/ggreif/ghc/commit/22e1ba53b7c0d1d94cc4ace70f3d8bd868e634fe https://github.com/ggreif/ghc/commit/4e048f0d2862989265e30d72c8025d087c8e69cd https://github.com/ggreif/ghc/commit/81026cc203cfc4dd0b118061d0cd9719888f3cf3 https://github.com/ggreif/ghc/commit/e5eb2378295ddce58fc5d2b1979f291b8ecd7875 https://github.com/ggreif/ghc/commit/fbbc9dcf224aae7cbcba0f8045159e345f0dc16b https://github.com/ggreif/ghc/commit/700a3e41159fa1ae3b9a836cc82fe566d7f61c34 https://github.com/ggreif/ghc/commit/3498437648f75507d3a059c5eba1acffdb7c0b16 https://github.com/ggreif/ghc/commit/84410c60b7e65ba0be31ab9f97b61102819882f2 https://github.com/ggreif/ghc/commit/0269459cd528db1486b51bfb98ec3ee674a0edff https://github.com/ggreif/ghc/commit/61ac72160329ed2d6200cac26238bc2053a4b4c9 https://github.com/ggreif/ghc/commit/ec67dbacd14f160d1c11470d6606985ea2363852 https://github.com/ggreif/ghc/commit/116b6b6338fe7c3b813f4c15db99627421ea0b9f ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [PATCH] Many small typos and formatting
See my previous mail... I believe I did precisely what you called for: https://github.com/ggreif/ghc/commits/misc-cleanups/ repo: https://github.com/ggreif/ghc.git branch: misc-cleanups Cheers, Gabor On 2/16/12, Simon Marlow marlo...@gmail.com wrote: On 15/02/2012 17:33, Gabor Greif wrote: Hi all, on my fork https://github.com/ggreif/ghc I have accumulated a number of small patches that are results of my browsing of the GHC sources and correcting minor issues. These are without functional impact and mostly apply to comments and formatting. By transferring these to mainline GHC I hope to lessen the distance of my fork to mainline. When pulling patches from mainline via github I sometimes use the fork queue (https://github.com/ggreif/ghc/forkqueue). By using the corresponding device (https://github.com/ghc/ghc/forkqueue) of the official mirror it might be quite easy to locate and apply the below patches. There might be an even better method by resorting to the command-line git tool. Alternatively I could set up a github clone of the original mirror (a topic branch maybe?) and push these changes there, so you would only need to pull them into the official repo. Please indicate how this can be handled best. The easiest for us would be if you could give us a repo URL that we can pull from. I've done this before when people have submitted patches via github - they make a branch on their github repo containing the changes, and give us the URL and branch name, we pull and validate the changes and push to GHC's repo. Cheers, Simon Cheers, Gabor https://github.com/ggreif/ghc/commit/77320765550bfa603052f0334d2d681134b63da0 https://github.com/ggreif/ghc/commit/598306aca074875bbe2cc025a5dd267a53a0d3bd done# https://github.com/ggreif/ghc/commit/fe443ed82538d04191212a08e44bd3f2e27613af https://github.com/ggreif/ghc/commit/95c2d50da10e49bb936e20a3b58fb29c7c5b4d13 https://github.com/ggreif/ghc/commit/3c79e184304ca38b8ed4484ccc3e8bac9724963a https://github.com/ggreif/ghc/commit/e47aa109b35ec02b544b4dec5640714c4ce66d74 https://github.com/ggreif/ghc/commit/e4c4cc978fbc1cc445e076adfd08765b4ef3a39c https://github.com/ggreif/ghc/commit/22e1ba53b7c0d1d94cc4ace70f3d8bd868e634fe https://github.com/ggreif/ghc/commit/4e048f0d2862989265e30d72c8025d087c8e69cd https://github.com/ggreif/ghc/commit/81026cc203cfc4dd0b118061d0cd9719888f3cf3 https://github.com/ggreif/ghc/commit/e5eb2378295ddce58fc5d2b1979f291b8ecd7875 https://github.com/ggreif/ghc/commit/fbbc9dcf224aae7cbcba0f8045159e345f0dc16b https://github.com/ggreif/ghc/commit/700a3e41159fa1ae3b9a836cc82fe566d7f61c34 https://github.com/ggreif/ghc/commit/3498437648f75507d3a059c5eba1acffdb7c0b16 https://github.com/ggreif/ghc/commit/84410c60b7e65ba0be31ab9f97b61102819882f2 https://github.com/ggreif/ghc/commit/0269459cd528db1486b51bfb98ec3ee674a0edff https://github.com/ggreif/ghc/commit/61ac72160329ed2d6200cac26238bc2053a4b4c9 https://github.com/ggreif/ghc/commit/ec67dbacd14f160d1c11470d6606985ea2363852 https://github.com/ggreif/ghc/commit/116b6b6338fe7c3b813f4c15db99627421ea0b9f ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[PATCH] Many small typos and formatting
Hi all, on my fork https://github.com/ggreif/ghc I have accumulated a number of small patches that are results of my browsing of the GHC sources and correcting minor issues. These are without functional impact and mostly apply to comments and formatting. By transferring these to mainline GHC I hope to lessen the distance of my fork to mainline. When pulling patches from mainline via github I sometimes use the fork queue (https://github.com/ggreif/ghc/forkqueue). By using the corresponding device (https://github.com/ghc/ghc/forkqueue) of the official mirror it might be quite easy to locate and apply the below patches. There might be an even better method by resorting to the command-line git tool. Alternatively I could set up a github clone of the original mirror (a topic branch maybe?) and push these changes there, so you would only need to pull them into the official repo. Please indicate how this can be handled best. Cheers, Gabor https://github.com/ggreif/ghc/commit/77320765550bfa603052f0334d2d681134b63da0 https://github.com/ggreif/ghc/commit/598306aca074875bbe2cc025a5dd267a53a0d3bd done# https://github.com/ggreif/ghc/commit/fe443ed82538d04191212a08e44bd3f2e27613af https://github.com/ggreif/ghc/commit/95c2d50da10e49bb936e20a3b58fb29c7c5b4d13 https://github.com/ggreif/ghc/commit/3c79e184304ca38b8ed4484ccc3e8bac9724963a https://github.com/ggreif/ghc/commit/e47aa109b35ec02b544b4dec5640714c4ce66d74 https://github.com/ggreif/ghc/commit/e4c4cc978fbc1cc445e076adfd08765b4ef3a39c https://github.com/ggreif/ghc/commit/22e1ba53b7c0d1d94cc4ace70f3d8bd868e634fe https://github.com/ggreif/ghc/commit/4e048f0d2862989265e30d72c8025d087c8e69cd https://github.com/ggreif/ghc/commit/81026cc203cfc4dd0b118061d0cd9719888f3cf3 https://github.com/ggreif/ghc/commit/e5eb2378295ddce58fc5d2b1979f291b8ecd7875 https://github.com/ggreif/ghc/commit/fbbc9dcf224aae7cbcba0f8045159e345f0dc16b https://github.com/ggreif/ghc/commit/700a3e41159fa1ae3b9a836cc82fe566d7f61c34 https://github.com/ggreif/ghc/commit/3498437648f75507d3a059c5eba1acffdb7c0b16 https://github.com/ggreif/ghc/commit/84410c60b7e65ba0be31ab9f97b61102819882f2 https://github.com/ggreif/ghc/commit/0269459cd528db1486b51bfb98ec3ee674a0edff https://github.com/ggreif/ghc/commit/61ac72160329ed2d6200cac26238bc2053a4b4c9 https://github.com/ggreif/ghc/commit/ec67dbacd14f160d1c11470d6606985ea2363852 https://github.com/ggreif/ghc/commit/116b6b6338fe7c3b813f4c15db99627421ea0b9f ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [PATCH] Many small typos and formatting
Okay, managed to put all these (after resolving a conflict) on a branch: https://github.com/ggreif/ghc/commits/misc-cleanups/ merging that on head should be it. The branch 'misc-cleanups' is based on today's official master branch, so even a fast-forward merge should be possible. Cheers, Gabor On 2/15/12, Gabor Greif ggr...@gmail.com wrote: Hi all, on my fork https://github.com/ggreif/ghc I have accumulated a number of small patches that are results of my browsing of the GHC sources and correcting minor issues. These are without functional impact and mostly apply to comments and formatting. By transferring these to mainline GHC I hope to lessen the distance of my fork to mainline. When pulling patches from mainline via github I sometimes use the fork queue (https://github.com/ggreif/ghc/forkqueue). By using the corresponding device (https://github.com/ghc/ghc/forkqueue) of the official mirror it might be quite easy to locate and apply the below patches. There might be an even better method by resorting to the command-line git tool. Alternatively I could set up a github clone of the original mirror (a topic branch maybe?) and push these changes there, so you would only need to pull them into the official repo. Please indicate how this can be handled best. Cheers, Gabor https://github.com/ggreif/ghc/commit/77320765550bfa603052f0334d2d681134b63da0 https://github.com/ggreif/ghc/commit/598306aca074875bbe2cc025a5dd267a53a0d3bd done# https://github.com/ggreif/ghc/commit/fe443ed82538d04191212a08e44bd3f2e27613af https://github.com/ggreif/ghc/commit/95c2d50da10e49bb936e20a3b58fb29c7c5b4d13 https://github.com/ggreif/ghc/commit/3c79e184304ca38b8ed4484ccc3e8bac9724963a https://github.com/ggreif/ghc/commit/e47aa109b35ec02b544b4dec5640714c4ce66d74 https://github.com/ggreif/ghc/commit/e4c4cc978fbc1cc445e076adfd08765b4ef3a39c https://github.com/ggreif/ghc/commit/22e1ba53b7c0d1d94cc4ace70f3d8bd868e634fe https://github.com/ggreif/ghc/commit/4e048f0d2862989265e30d72c8025d087c8e69cd https://github.com/ggreif/ghc/commit/81026cc203cfc4dd0b118061d0cd9719888f3cf3 https://github.com/ggreif/ghc/commit/e5eb2378295ddce58fc5d2b1979f291b8ecd7875 https://github.com/ggreif/ghc/commit/fbbc9dcf224aae7cbcba0f8045159e345f0dc16b https://github.com/ggreif/ghc/commit/700a3e41159fa1ae3b9a836cc82fe566d7f61c34 https://github.com/ggreif/ghc/commit/3498437648f75507d3a059c5eba1acffdb7c0b16 https://github.com/ggreif/ghc/commit/84410c60b7e65ba0be31ab9f97b61102819882f2 https://github.com/ggreif/ghc/commit/0269459cd528db1486b51bfb98ec3ee674a0edff https://github.com/ggreif/ghc/commit/61ac72160329ed2d6200cac26238bc2053a4b4c9 https://github.com/ggreif/ghc/commit/ec67dbacd14f160d1c11470d6606985ea2363852 https://github.com/ggreif/ghc/commit/116b6b6338fe7c3b813f4c15db99627421ea0b9f ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: Backtrace from crashing target program
We have a byte swapping(-like problem).
Looking at the dump of the info table on PPC
(gdb) p *info
$7 = {layout = {payload = {ptrs = 0, nptrs = 4}, bitmap = 4,
large_bitmap_offset = 4, selector_offset = 4}, type = 0,
srt_bitmap = 40, code = 0x102cab2c `\02028c\212K%\215}
(gdb) disassemble 0x102cab2c
Dump of assembler code for function stg_BLOCKING_QUEUE_CLEAN_info:
0x102cab2c stg_BLOCKING_QUEUE_CLEAN_info+0: lis r3,4146
0x102cab30 stg_BLOCKING_QUEUE_CLEAN_info+4: addir3,r3,-29988
0x102cab34 stg_BLOCKING_QUEUE_CLEAN_info+8: bl 0x1029d0c0 barf
0x102cab38 stg_BLOCKING_QUEUE_CLEAN_info+12: .long 0x4
0x102cab3c stg_BLOCKING_QUEUE_CLEAN_info+16: .long 0x28
While on x86_64 we have
(gdb) p *info
$2 = {layout = {payload = {ptrs = 4, nptrs = 0}, bitmap = 4,
large_bitmap_offset = 4, __pad_large_bitmap_offset = 4,
selector_offset = 4},
type = 40, srt_bitmap = 0,
code = 0x47f950 I\211u\020I\211}\030M\211E
M\211M(\363A\017\021MP\363A\017\021UT\363A\017\021]X\363A\017\021e\\\362A\017\021m`\362A\017\021uh\277\300%L}
(gdb) disassemble 0x47f950
Dump of assembler code for function stg_BLOCKING_QUEUE_CLEAN_info:
0x0047f950 stg_BLOCKING_QUEUE_CLEAN_info+0: mov%rsi,0x10(%r13)
0x0047f954 stg_BLOCKING_QUEUE_CLEAN_info+4: mov%rdi,0x18(%r13)
0x0047f958 stg_BLOCKING_QUEUE_CLEAN_info+8: mov%r8,0x20(%r13)
'type' and 'srt_bitmap' are wrong on PowerPC, and swapped relative to
x86_64. The latter is correct. I believe this is CMM stuff where can I
dig further?
How did PowerPC linux ever work?
Cheers,
Gabor
On 1/16/12, Gabor Greif ggr...@gmail.com wrote:
It appears that I ended up with the threaded runtime. While not
optimal for debugging, I see this:
Breakpoint 1, schedule (initialCapability=0x1032b680, task=0x10336ec8)
at rts/Schedule.c:486
486 r = StgRun((StgFunPtr) stg_returnToStackTop, cap-r);
(gdb) p cap-r
$6 = {rR1 = {w = 0, a = 0x0, c = 0, f = 0, i = 0, p = 0x0}, rR2 = {w =
0, a = 0x0,
c = 0, f = 0, i = 0, p = 0x0}, rR3 = {w = 0, a = 0x0, c = 0, f = 0, i =
0,
p = 0x0}, rR4 = {w = 0, a = 0x0, c = 0, f = 0, i = 0, p = 0x0},
rR5 = {w = 0,
a = 0x0, c = 0, f = 0, i = 0, p = 0x0}, rR6 = {w = 0, a = 0x0, c =
0, f = 0,
i = 0, p = 0x0}, rR7 = {w = 0, a = 0x0, c = 0, f = 0, i = 0, p =
0x0}, rR8 = {
w = 0, a = 0x0, c = 0, f = 0, i = 0, p = 0x0}, rR9 = {w = 0, a =
0x0, c = 0,
f = 0, i = 0, p = 0x0}, rR10 = {w = 0, a = 0x0, c = 0, f = 0, i =
0, p = 0x0},
rF1 = 0, rF2 = 0, rF3 = 0, rF4 = 0, rD1 = 0, rD2 = 0, rL1 = 0, rSp = 0x0,
rSpLim = 0x0, rHp = 0x0, rHpLim = 0x0, rCCCS = 0x0, rCurrentTSO =
0x301033c0,
rNursery = 0x1032f178, rCurrentNursery = 0x30100040, rCurrentAlloc =
0x30100060,
rHpAlloc = 0, rRet = 0}
After loading the thread state,
stg_returnToStackTop
{
LOAD_THREAD_STATE();
CHECK_SENSIBLE_REGS();
jump %ENTRY_CODE(Sp(0));
}
rightly barfs that registers are zero. Which function sets up those?
From C or Haskell?
How can I select the single-threaded runtime?
Sorry for the abundance of questions...
Cheers,
Gabor
On 1/16/12, Simon Marlow marlo...@gmail.com wrote:
On 15/01/2012 00:07, Gabor Greif wrote:
Hello hackers,
my hello world program compiles and links, but when running on the
embedded PowerPC board it crashes like this:
(gdb) run
Starting program: /root/test
test: internal error: stg_ap_p_ret
(GHC version 7.5.20120111 for powerpc_unknown_linux)
Please report this as a GHC bug:
http://www.haskell.org/ghc/reportabug
Program received signal SIGABRT, Aborted.
0x0fdc0a4c in raise () from /lib/tls/libc.so.6
(gdb) bt
#0 0x0fdc0a4c in raise () from /lib/tls/libc.so.6
#1 0x0fdc2488 in abort () from /lib/tls/libc.so.6
#2 0x1029b300 in rtsFatalInternalErrorFn ()
#3 0x1029b024 in barf ()
#4 0x102b3f9c in stg_ap_p_info ()
#5 0x1029d0ac in scheduleWaitThread ()
#6 0x102c69a8 in rts_evalLazyIO ()
#7 0x1029ae80 in real_main ()
#8 0x1029afc8 in hs_main ()
#9 0x10002168 in main ()
(gdb) c
Continuing.
Program terminated with signal SIGABRT, Aborted.
The program no longer exists.
This back trace tells us almost nothing, I'm afraid - many different
kinds of corruption end up as crashes in stg_ap_p_info or some other bit
of the RTS.
I suggest following the instructions in
http://hackage.haskell.org/trac/ghc/wiki/Debugging/CompiledCode
and see how far you get. Come back if you get stuck.
Cheers,
Simon
'git grep' does not turn up any hits for 'stg_ap_p_info', and
'scheduleWaitThread' looks pretty much innocent. Where is the best
place to get into debugging this?
I gather stg_ap_p_info is a (global data) StgInfoTable, but then again
it appears in the backtrace, (tables beside code?). Why would it barf
at all?
I am thinking that some host configury may still be leaking into the
target, so I might be fighting a deceived RTS ('revenge tastes best
when served cold'). But the host
Re: Backtrace from crashing target program
Just for the record, I believe this is resolved now, in my cross-compilation setup (for PowerPC) WORDS_BIGENDIAN was not set. Not sure yet about FLOAT_WORDS_BIGENDIAN yet. Cheers, Gabor On 1/18/12, Gabor Greif ggr...@gmail.com wrote: We have a byte swapping(-like problem). ... ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
MO_WriteBarrier and PowerPC (NCG)
Hi all,
looks like I have an (almost) running binary on a PPC embedded target.
'Almost' because the native backend emits a 'lwsync' (light-weight
sync) instruction for the MO_WriteBarrier C-- instruction.
Unfortunately this is an e500 core that traps with illegal
instruction on 'lwsync'. Below I added some references that show that
many projects feel (felt) the pain.
From what I gather, write-barriers in GHC are only necessary in
multi-processor settings; when I am sure to have a uniprocessor, then
a simple 'nop' instruction will do. Can somebody confirm this?
So how to get out?
1) We could probe the target C compiler (when specified
'--with-target-gcc') whether it provides a __NO_LWSYNC__ macro (won't
work for us, i.e. my employer, as our GCC is probably older than
that).
2) some configure-time switch like '--enable-write-barrier={sync|nop|lwsync}'
sounds like a good solution, but clutters './configure --help'
3) Using the target triple to specify e500 or uniprocessor
how? are there conventions already?
Then we have to communicate the above to NCG; we could use
a) a #define and conditional compilation
quick, dirty, effective
b) Sub-architecture (ArchPPC e500)
elegant, but more work
I am very interested in input, do somebody have preferences or ideas
about the above?
Cheers,
Gabor
PS: added Wolfgang to 'CC:', as he wrote the 'lwsync' patch in the first place
http://gcc.gnu.org/ml/gcc-patches/2006-11/msg01238.html
http://code.google.com/p/google-perftools/issues/detail?id=140
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
Backtrace from crashing target program
Hello hackers,
my hello world program compiles and links, but when running on the
embedded PowerPC board it crashes like this:
(gdb) run
Starting program: /root/test
test: internal error: stg_ap_p_ret
(GHC version 7.5.20120111 for powerpc_unknown_linux)
Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug
Program received signal SIGABRT, Aborted.
0x0fdc0a4c in raise () from /lib/tls/libc.so.6
(gdb) bt
#0 0x0fdc0a4c in raise () from /lib/tls/libc.so.6
#1 0x0fdc2488 in abort () from /lib/tls/libc.so.6
#2 0x1029b300 in rtsFatalInternalErrorFn ()
#3 0x1029b024 in barf ()
#4 0x102b3f9c in stg_ap_p_info ()
#5 0x1029d0ac in scheduleWaitThread ()
#6 0x102c69a8 in rts_evalLazyIO ()
#7 0x1029ae80 in real_main ()
#8 0x1029afc8 in hs_main ()
#9 0x10002168 in main ()
(gdb) c
Continuing.
Program terminated with signal SIGABRT, Aborted.
The program no longer exists.
'git grep' does not turn up any hits for 'stg_ap_p_info', and
'scheduleWaitThread' looks pretty much innocent. Where is the best
place to get into debugging this?
I gather stg_ap_p_info is a (global data) StgInfoTable, but then again
it appears in the backtrace, (tables beside code?). Why would it barf
at all?
I am thinking that some host configury may still be leaking into the
target, so I might be fighting a deceived RTS ('revenge tastes best
when served cold'). But the host and target should be pretty
comparable linux OSs.
I am thankful for any hints.
Cheers,
Gabor
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: Backtrace from crashing target program
Hi Gaby, here you go: https://github.com/ggreif/ghc But in 2 places you will still need some manual intervention: $TOP/settings (check word size) the other settings file 'find $TOP -name settings' $ find . -name settings ./inplace/lib/settings ./settings And be sure to check ./includes/ghcautoconf.h because it is currently filled with host configury, so you'll have to edit it. Making this automatic is on my list after the binary starts to run on target. I had to adjust the UL and L sizes (and alignments) from 8 to 4 for my PPC target. Then you have to $ perl boot $ ./configure --host=triple --target=triple --with-target-gcc=path-to-cross-cc Cheers, Gabor PS: What are you trying to target? PPS: Crossing fingers that you run into the same problem and that you can help me debugging it ;-) On 1/15/12, Gabriel Dos Reis g...@integrable-solutions.net wrote: On Sat, Jan 14, 2012 at 6:07 PM, Gabor Greif ggr...@gmail.com wrote: Hello hackers, my hello world program compiles and links, but when running on the embedded PowerPC board it crashes like this: Hi Gabor, where can I find links to your patches enabling cross-compilation/cross-building? ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: Backtrace from crashing target program
Not sure you can trick GHC into generting code for those. I am fortunate, because ArchPPC is supported. Anyway, good luck, Gabor On 1/15/12, Gabriel Dos Reis g...@integrable-solutions.net wrote: On Sat, Jan 14, 2012 at 6:35 PM, Gabor Greif ggr...@gmail.com wrote: Hi Gaby, here you go: https://github.com/ggreif/ghc But in 2 places you will still need some manual intervention: $TOP/settings (check word size) the other settings file 'find $TOP -name settings' $ find . -name settings ./inplace/lib/settings ./settings And be sure to check ./includes/ghcautoconf.h because it is currently filled with host configury, so you'll have to edit it. Making this automatic is on my list after the binary starts to run on target. I had to adjust the UL and L sizes (and alignments) from 8 to 4 for my PPC target. Then you have to $ perl boot $ ./configure --host=triple --target=triple --with-target-gcc=path-to-cross-cc Thank you! Cheers, Gabor PS: What are you trying to target? I've got some AVR micro controllers to play with. It should be fun -- at least, I hope not too painful :-) PPS: Crossing fingers that you run into the same problem and that you can help me debugging it ;-) :-) -- Gaby ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: ghc ==7.* and 7.5.20120111, what the heck?
Right, and the reason for why dph-lifted-vseq is configured before the ghc package seems to be that I told compiler_stage2_NOT_NEEDED = YES. Logically this would be correct, as there is no stage2 compiler in a cross-compilation setting, but by saying so, the build system lumps all packages for being configured together and performs it using the stage1 compiler (obviously before the ghc package is built). I do not understand this fully yet, anyway this https://github.com/ggreif/ghc/commit/56522be8b2b7c09b492d11ba911a2b8f2406999d solved my problems (I hope). Any insight why this is happening is welcome. Cheers, Gabor On 1/12/12, Simon Marlow marlo...@gmail.com wrote: On 11/01/2012 21:40, Roman Leshchinskiy wrote: On 11/01/2012, at 19:37, Gabor Greif wrote: libraries/dph/dph-lifted-vseg/Data/Array/Parallel/PArray/PData/Base.hs:20:8: Could not find module `SpecConstr' Use -v to see a list of the files searched for. make[1]: *** [libraries/dph/dph-lifted-vseg/dist-install/build/.depend-v.haskell] Error 1 make: *** [all] Error 2 Turns out there is no such module in Base any more. FWIW, SpecConstr is a ghc module which (I assume) isn't found because you removed the dependency on ghc. I have no idea why the ghc==7.* dependency is failing, it looks like a bug to me. The usual reason for this kind of failure is if there is a missing dependency such that dph-lifted-vseq is being configured before the ghc package. Cheers, Simon ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: ghc-cabal in HEAD and libraries/primitive (from ./sync-all)
Hi Ian, Roman,
this may be a problem with my cross-compiler, as the problem persists
in my setup:
$ inplace/bin/ghc-cabal configure
--with-ghc=/home/ggreif/ghc-head/inplace/bin/ghc-stage1
--with-ghc-pkg=/home/ggreif/ghc-head/inplace/bin/ghc-pkg
--enable-library-for-ghci --configure-option=CFLAGS=
--configure-option=LDFLAGS=--configure-option=CPPFLAGS=
--with-gcc=false --configure-option=--with-cc=false
--with-ar=false --with-ranlib=true -- dist-install
libraries/primitive
Configuring primitive-0.4.0.1...
ghc-cabal: Missing dependency on a foreign library:
* Missing (or bad) header file: primitive-memops.h
This problem can usually be solved by installing the system package that
provides this library (you may need the -dev version). If the library is
already installed but in a non-standard location then you can use the flags
--extra-include-dirs= and --extra-lib-dirs= to specify where it is.
If the header file does exist, it may contain errors that are caught by the C
compiler at the preprocessing stage. In this case you can re-run configure
with the verbosity flag -v3 to see the error messages.
$ cat /home/ggreif/ghc-head/inplace/bin/ghc-stage1
#!/bin/sh
executablename=/home/ggreif/ghc-head/inplace/lib/ghc-stage1
datadir=/home/ggreif/ghc-head/inplace/lib
bindir=/home/ggreif/ghc-head/inplace/bin
topdir=/home/ggreif/ghc-head/inplace/lib
pgmgcc=/home/ggreif/devkit/ppc/85xx/bin/ppc_85xx-gcc
exec $executablename -B$topdir ${1+$@}
Is it possible that ghc-cabal somehow uses a C compiler (host/target
gcc) to discover dependencies?
Cheers,
Gabor
On 1/11/12, Roman Leshchinskiy r...@cse.unsw.edu.au wrote:
On 10/01/2012, at 20:28, Gabor Greif wrote:
My suspicion is that ghc-cabal is not prepared to see includes: or
some other weirdness with Include-Dirs: or Install-Includes:.
From bytestring.cabal in the GHC tree:
include-dirs: include
includes: fpstring.h
install-includes: fpstring.h
Does bytestring have the same problem for you? This sounds very weird, I've
never seen this error before.
Roman
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
ghc ==7.* and 7.5.20120111, what the heck?
So, I have a cross-ghc configured like this: ./configure --host=x86_64-apple-darwin --target=i386-apple-darwin After successfully building stage1 I get this error when configuring dph/dph-lifted-base: $ inplace/bin/ghc-cabal configure --with-ghc=/Users/ggreif/ghc/inplace/bin/ghc-stage1 --with-ghc-pkg=/Users/ggreif/ghc/inplace/bin/ghc-pkg --enable-library-for-ghci --configure-option=CFLAGS= -m32 -fno-stack-protector --configure-option=LDFLAGS= -m32 --configure-option=CPPFLAGS= -m32 --with-gcc=/usr/bin/gcc --configure-option=--with-cc=/usr/bin/gcc --with-ar=/usr/bin/ar --with-ranlib=ranlib -- dist-install libraries/dph/dph-lifted-base Configuring dph-lifted-base-0.5.2.0... ghc-cabal: At least the following dependencies are missing: ghc ==7.* make[1]: *** [libraries/dph/dph-lifted-base/dist-install/package-data.mk] Error 1 make: *** [all] Error 2 $ inplace/bin/ghc-stage1 --version The Glorious Glasgow Haskell Compilation System, version 7.5.20120111 Another weirdness I am experiencing. I can remove the depencency from the cabal file (which helps) but I get the same issue in the next configure, etc. Another issue (after correcting the above 4 times) libraries/dph/dph-lifted-vseg/Data/Array/Parallel/PArray/PData/Base.hs:20:8: Could not find module `SpecConstr' Use -v to see a list of the files searched for. make[1]: *** [libraries/dph/dph-lifted-vseg/dist-install/build/.depend-v.haskell] Error 1 make: *** [all] Error 2 Turns out there is no such module in Base any more. Removing the imports (in some places) allows the configuring to continue. Down the road, I even see package ghc unknown when configuring utils/mkUserGuidePart. What is going on? Has anybody expreienced such things? Am I getting the wring dependencies? Or just the perils of living near the bleeding edge? I am preparing for a long debug-night. Cheers, Gabor ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
ghc-cabal in HEAD and libraries/primitive (from ./sync-all)
Hi all, for some reason either the libraries/primitive version in tarballs is broken (should be includes: cbits/primitive-memops.h ?) or the ghc-cabal is not able to resolve the included header... Anybody seen this? I have a fresh checkout. Following change helped, but I assume is not acceptable for a patch :-) diff --git a/primitive.cabal b/primitive.cabal index 3e19e84..985e2e7 100644 --- a/primitive.cabal +++ b/primitive.cabal @@ -57,6 +57,5 @@ Library Include-Dirs: cbits Install-Includes: primitive-memops.h - includes: primitive-memops.h c-sources: cbits/primitive-memops.c Cheers, Gabor ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: ghc-cabal in HEAD and libraries/primitive (from ./sync-all)
Ouch, I was in a hurry leaving work and forgot copying the error message. It was along the lines cannot resolve header path: primitive-memops.h I cannot remember the exact wording. Anyway, the package does not contain such a header at toplevel, only at cbits/primitive-memops.h. I tried to symlink it, no way. Also copying it over did not resolve the error and changing the .cabal file to read: includes: cbits/primitive-memops.h did not help. I can try to reproduce the error at home later. My suspicion is that ghc-cabal is not prepared to see includes: or some other weirdness with Include-Dirs: or Install-Includes:. By 'tarballs' I mean the pre-canned libraries that '.sync-all get' pulls into the ghc source tree (by 'git'). On 1/10/12, Ian Lynagh ig...@earth.li wrote: Hi Gabor, On Tue, Jan 10, 2012 at 08:06:44PM +0100, Gabor Greif wrote: for some reason either the libraries/primitive version in tarballs is broken What do you mean by in tarballs? Anybody seen this? What are you seeing exactly? Thanks Ian ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [Q] CMM error (cross-powerpc-linux backend)
On 1/9/12, Simon Marlow marlo...@gmail.com wrote: On 04/01/2012 15:48, Gabor Greif wrote: Hi hackers, I got my cross-ghc so far that it compiles its own libraries, now onto the rts! BTW, if you're working on cross compilation, are you aware of Stephen Blackheath's work? http://www.haskell.org/pipermail/iphone/2011-November/94.html Hi Simon, no, I did not see this before. Unfortunately (for me) it looks like it is llvm-based so, it is wildly out of our possibilities. We only have a vanilla gcc-3.4.3 for our PPC target. So I'll have pull it off myself. Anyway, I'll have a look into those patches in a calm hour. Thanks and cheers, Gabor Stephen, I haven't looked at your patches, can you summarise what you've done? Do you think we'll want to pull it in as is, or will it need more work? Does it follow the general plan in http://hackage.haskell.org/trac/ghc/wiki/CrossCompilation ? Cheers, Simon ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [PATCH] use (GHC) idiomatic types
Hi Simon, as I am now done with OFFSET and TYPE_SIZE computations for cross-ghc, these also popped up https://github.com/ggreif/ghc/commit/77320765550bfa603052f0334d2d681134b63da0 https://github.com/ggreif/ghc/commit/4edcfb16c3b45dc595630269d4a566b87183884d this one is cleanup: https://github.com/ggreif/ghc/commit/9bc6f79d4a764a6a19bbc4fb3cc04c24feee2923 Cheers, Gabor On 1/9/12, Simon Marlow marlo...@gmail.com wrote: Thanks, I'll push. Cheers, Simon ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[PATCH] use (GHC) idiomatic types
---
includes/rts/storage/GC.h |8
1 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/includes/rts/storage/GC.h b/includes/rts/storage/GC.h
index fef8e00..26c5593 100644
--- a/includes/rts/storage/GC.h
+++ b/includes/rts/storage/GC.h
@@ -67,7 +67,7 @@ typedef struct nursery_ {
} nursery;
typedef struct generation_ {
-unsigned int no; // generation number
+natno; // generation number
bdescr * blocks; // blocks in this gen
memcount n_blocks;// number of blocks
@@ -85,9 +85,9 @@ typedef struct generation_ {
struct generation_ *to;// destination gen for live objects
// stats information
-unsigned int collections;
-unsigned int par_collections;
-unsigned int failed_promotions;
+nat collections;
+nat par_collections;
+nat failed_promotions;
//
// Fields below are used during GC only
--
1.7.7.4
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [PATCH] update copyright, cite realistic version numbers, fix a smattering of typos
No problem! as you can see I am involved in some janitorial activity to get up to pace with git, so here are some more patches (from my github fork for cross-compilation) https://github.com/ggreif/ghc/commit/da8c8345512958fc55650fc265f2cdc2549a98b2 https://github.com/ggreif/ghc/commit/2e6c2551ff7b414a185779075d47a96679acd716 the latter one I have also sent to the list: http://www.haskell.org/pipermail/cvs-ghc/2012-January/069529.html No idea which format is easier to you... A bit of motivation for the second patch. I have to figure out the target system byte offsets from all relevant struct members and feed mkDerivedConstants from that. I have a small toolchain that kicks in when cross-compiling, and it helps me reduce the toolchain complexity by not resorting to types like unsigned int for example. 'nat' is being used in many places already and seems only natural (sic) to use it here too. Cheers, Gabor On 1/7/12, Ian Lynagh ig...@earth.li wrote: On Thu, Jan 05, 2012 at 03:33:32AM +0100, Gabor Greif wrote: --- configure.ac | 20 ++-- 1 files changed, 10 insertions(+), 10 deletions(-) Thanks for the patch. This was applied as 9186d92f381c49720da96b8fc59339af7be951bd. Thanks Ian ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [PATCH] use C-style comments
Thanks for the explanation, I'll back out this one from my local set. Cheers, Gabor On 1/6/12, Simon Marlow marlo...@gmail.com wrote: I tend to use C++ style rather than C style for single-line comments, except for code that will be #included into Haskell source. ... ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [PATCH] abstract away from the 'build-toolchain'-dependent sizeof(...) operator
On 1/6/12, Simon Marlow marlo...@gmail.com wrote: On 05/01/2012 18:41, Gabor Greif wrote: The sizes obtained this way do not work on a target system in general. So in a future cross-compilable setup we need another way of obtaining expansions for the macros OFFSET, FIELD_SIZE and TYPE_SIZE. Guarded against accidental use of 'sizeof' by poisoning. Verified that the generated *Constants.h/hs files are unchanged. Ok, I'll push this one. Awesome, what about this one? http://www.haskell.org/pipermail/cvs-ghc/2012-January/069412.html Thanks and cheers, Gabor Cheers, Simon --- ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [Q] CMM error (cross-powerpc-linux backend)
Okay, mystery solved. The STG registers obtain the size of a host
pointer (which in my case is 64 bits) even for the target
architectures that are 32 bits.
include/Types.h takes care to check target pointer size here:
#if SIZEOF_VOID_P == 4
but goes on to define
typedef StgWord* StgPtr;
which ends up as the datatype of registers:
typedef union {
StgWordw;
...
StgPtr p;
} StgUnion;
typedef struct StgRegTable_ {
StgUnion rR1;
StgUnion rR2;
...
So StgUnion has size of 8 (in my --host=x86_64-redhat-linux), which
messes up the offset calculations into StgRegTable_. These are done by
includes/mkDerivedConstants.c which is compiled and executed on the
host platform. So the pointer sizes of host leak into the target
offsets.
I have an elegant idea how to get this right (without any impact for
HOST == TARGET).
Now I just have to implement my idea, finish the cross toolset and
submit the patches.
No more creepy interactions between HOST and TARGET!
Cheers,
Gabor
On 1/4/12, Gabor Greif ggr...@gmail.com wrote:
Hi hackers,
I got my cross-ghc so far that it compiles its own libraries, now onto the
rts!
All goes right, until I get:
inplace/bin/ghc-stage1 -H64m -O0 -fasm -Iincludes -Irts
-Irts/dist/build -DCOMPILING_RTS -package-name rts -dcmm-lint -i
-irts -irts/dist/build -irts/dist/build/autogen -Irts/dist/build
-Irts/dist/build/autogen-optc-DCOMPILING_GHC-optc-O2 -c
rts/Apply.cmm -o rts/dist/build/Apply.o
Cmm lint error:
in proc stg_AP_STACK_entry
in basic block c2g
in assignment:
_c21::I32 = I64[_c20::I32 + 4 + 8];
Reg ty: I32
Rhs ty: I64
For some reason I64 appears but that should be avoided in the
powerpc-linux backend :-(
My --host is x86_64-linux, so it may creep in from there.
...
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
[PATCH] abstract away from the 'build-toolchain'-dependent sizeof(...) operator
The sizes obtained this way do not work on a target system in general. So in a future cross-compilable setup we need another way of obtaining expansions for the macros OFFSET, FIELD_SIZE and TYPE_SIZE. Guarded against accidental use of 'sizeof' by poisoning. Verified that the generated *Constants.h/hs files are unchanged. --- includes/mkDerivedConstants.c | 24 ++-- 1 files changed, 14 insertions(+), 10 deletions(-) diff --git a/includes/mkDerivedConstants.c b/includes/mkDerivedConstants.c index 6688330..8d2e789 100644 --- a/includes/mkDerivedConstants.c +++ b/includes/mkDerivedConstants.c @@ -1,6 +1,6 @@ /* -- * - * (c) The GHC Team, 1992-2004 + * (c) The GHC Team, 1992-2012 * * mkDerivedConstants.c * @@ -17,11 +17,11 @@ * We need offsets of profiled things... better be careful that this * doesn't affect the offsets of anything else. */ + #define PROFILING #define THREADED_RTS #include Rts.h - #include Stable.h #include Capability.h @@ -30,6 +30,10 @@ #define str(a,b) #a _ #b #define OFFSET(s_type, field) ((size_t)(((s_type*)0)-field)) +#define FIELD_SIZE(s_type, field) ((unsigned long)sizeof(((s_type*)0)-field)) +#define TYPE_SIZE(type) (sizeof(type)) + +#pragma GCC poison sizeof #if defined(GEN_HASKELL) #define def_offset(str, offset) \ @@ -44,7 +48,7 @@ #define ctype(type) /* nothing */ #else #define ctype(type) \ -printf(#define SIZEOF_ #type %lu\n, (unsigned long)sizeof(type)); +printf(#define SIZEOF_ #type %lu\n, (unsigned long)TYPE_SIZE(type)); #endif #if defined(GEN_HASKELL) @@ -59,7 +63,7 @@ */ #define field_type_(str, s_type, field) \ printf(#define REP_ str b); \ -printf(%lu\n, (unsigned long)sizeof (__typeof__(s_type*)0)-field * 8); +printf(%lu\n, FIELD_SIZE(s_type, field) * 8); #define field_type_gcptr_(str, s_type, field) \ printf(#define REP_ str gcptr\n); #endif @@ -105,18 +109,18 @@ #endif #define struct_size(s_type) \ -def_size(#s_type, sizeof(s_type)); +def_size(#s_type, TYPE_SIZE(s_type)); /* * Size of a closure type, minus the header, named SIZEOF_type_NoHdr * Also, we #define SIZEOF_type to be the size of the whole closure for .cmm. */ #define closure_size(s_type) \ -def_size(#s_type _NoHdr, sizeof(s_type) - sizeof(StgHeader)); \ -def_closure_size(#s_type, sizeof(s_type) - sizeof(StgHeader)); +def_size(#s_type _NoHdr, TYPE_SIZE(s_type) - TYPE_SIZE(StgHeader)); \ +def_closure_size(#s_type, TYPE_SIZE(s_type) - TYPE_SIZE(StgHeader)); #define thunk_size(s_type) \ -def_size(#s_type _NoThunkHdr, sizeof(s_type) - sizeof(StgThunkHeader)); \ +def_size(#s_type _NoThunkHdr, TYPE_SIZE(s_type) - TYPE_SIZE(StgThunkHeader)); \ closure_size(s_type) /* An access macro for use in C-- sources. */ @@ -124,7 +128,7 @@ printf(#define str (__ptr__) REP_ str [__ptr__+SIZEOF_StgHeader+OFFSET_ str ]\n); #define closure_field_offset_(str, s_type,field) \ -def_offset(str, OFFSET(s_type,field) - sizeof(StgHeader)); +def_offset(str, OFFSET(s_type,field) - TYPE_SIZE(StgHeader)); #define closure_field_offset(s_type,field) \ closure_field_offset_(str(s_type,field),s_type,field) @@ -156,7 +160,7 @@ /* Byte offset for a TSO field, minus the header and variable prof bit. */ #define tso_payload_offset(s_type, field) \ -def_offset(str(s_type,field), OFFSET(s_type,field) - sizeof(StgHeader) - sizeof(StgTSOProfInfo)); +def_offset(str(s_type,field), OFFSET(s_type,field) - TYPE_SIZE(StgHeader) - TYPE_SIZE(StgTSOProfInfo)); /* Full byte offset for a TSO field, for use from Cmm */ #define tso_field_offset_macro(str) \ -- 1.7.7.4 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[PATCH] use C-style comments
Also remove underscore identifier as
it is not needed, StgFunTable is now
consistent with StgRegTable.
---
includes/stg/Regs.h | 10 +-
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/includes/stg/Regs.h b/includes/stg/Regs.h
index bfc3d4b..bf17b7e 100644
--- a/includes/stg/Regs.h
+++ b/includes/stg/Regs.h
@@ -1,6 +1,6 @@
/*
-
*
- * (c) The GHC Team, 1998-2009
+ * (c) The GHC Team, 1998-2012
*
* Registers in the STG machine.
*
@@ -21,7 +21,7 @@
*
* The register set is backed by a table in memory (struct
* StgRegTable). If a particular STG register is not mapped to a
- * machine register, then the apprpriate slot in this table is used
+ * machine register, then the appropriate slot in this table is used
* instead.
*
* This table is itself pointed to by another register, BaseReg. If
@@ -58,7 +58,7 @@ typedef union {
*register, probably because there's a shortage of real registers.
* 2) caller-saves registers are saved across a CCall
*/
-typedef struct StgRegTable_ {
+typedef struct {
StgUnion rR1;
StgUnion rR2;
StgUnion rR3;
@@ -80,13 +80,13 @@ typedef struct StgRegTable_ {
StgPtr rSpLim;
StgPtr rHp;
StgPtr rHpLim;
- struct CostCentreStack_ * rCCCS; // current cost-centre-stack
+ struct CostCentreStack_ * rCCCS; /* current cost-centre-stack */
struct StgTSO_ * rCurrentTSO;
struct nursery_ *rNursery;
struct bdescr_ * rCurrentNursery; /* Hp/HpLim point into this block */
struct bdescr_ * rCurrentAlloc; /* for allocation using allocate() */
StgWord rHpAlloc;/* number of *bytes* being allocated in heap */
- StgWord rRet; // holds the return code of the thread
+ StgWord rRet; /* holds the return code of the thread */
} StgRegTable;
#if IN_STG_CODE
--
1.7.7.4
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
Re: [Q] CMM error (cross-powerpc-linux backend)
On 1/5/12, Simon Marlow marlo...@gmail.com wrote: On 05/01/2012 12:33, Gabor Greif wrote: Okay, mystery solved. The STG registers obtain the size of a host pointer (which in my case is 64 bits) even for the target architectures that are 32 bits. ... So StgUnion has size of 8 (in my --host=x86_64-redhat-linux), which messes up the offset calculations into StgRegTable_. These are done by includes/mkDerivedConstants.c which is compiled and executed on the host platform. So the pointer sizes of host leak into the target offsets. Yes, the contents of DerivedConstants.h and HaskellConstants.hs should be correct for the target platform, not the host platform. So you either have to compile and run mkDerivedConstants on the target, or do it using the cross toolchain on the host platform (which ought to be possible, since autoconf does this kind of thing). Yeah, this is my plan. I'll start submitting my cross-compilation branch on github next week but in the meantime I am sending minor cleanup patches to this list to get comfortable with the git workflow. I would be thrilled if someone could give them a look (or even commit :-) Cheers, Gabor Cheers, Simon I have an elegant idea how to get this right (without any impact for HOST == TARGET). Now I just have to implement my idea, finish the cross toolset and submit the patches. No more creepy interactions between HOST and TARGET! Cheers, Gabor On 1/4/12, Gabor Greifggr...@gmail.com wrote: Hi hackers, I got my cross-ghc so far that it compiles its own libraries, now onto the rts! All goes right, until I get: inplace/bin/ghc-stage1 -H64m -O0 -fasm -Iincludes -Irts -Irts/dist/build -DCOMPILING_RTS -package-name rts -dcmm-lint -i -irts -irts/dist/build -irts/dist/build/autogen -Irts/dist/build -Irts/dist/build/autogen-optc-DCOMPILING_GHC-optc-O2 -c rts/Apply.cmm -o rts/dist/build/Apply.o Cmm lint error: in proc stg_AP_STACK_entry in basic block c2g in assignment: _c21::I32 = I64[_c20::I32 + 4 + 8]; Reg ty: I32 Rhs ty: I64 For some reason I64 appears but that should be avoided in the powerpc-linux backend :-( My --host is x86_64-linux, so it may creep in from there. ... ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[Q] CMM error (cross-powerpc-linux backend)
Hi hackers, I got my cross-ghc so far that it compiles its own libraries, now onto the rts! All goes right, until I get: inplace/bin/ghc-stage1 -H64m -O0 -fasm -Iincludes -Irts -Irts/dist/build -DCOMPILING_RTS -package-name rts -dcmm-lint -i -irts -irts/dist/build -irts/dist/build/autogen -Irts/dist/build -Irts/dist/build/autogen-optc-DCOMPILING_GHC-optc-O2 -c rts/Apply.cmm -o rts/dist/build/Apply.o Cmm lint error: in proc stg_AP_STACK_entry in basic block c2g in assignment: _c21::I32 = I64[_c20::I32 + 4 + 8]; Reg ty: I32 Rhs ty: I64 For some reason I64 appears but that should be avoided in the powerpc-linux backend :-( My --host is x86_64-linux, so it may creep in from there. So my question, which include I should look at? I am pretty newbie with CMM. Cheers and thanks, Gabor PS: include/ghcautoconf.h (almost) looks good: /* The size of a `char', as computed by sizeof. */ #define SIZEOF_CHAR 1 /* The size of a `double', as computed by sizeof. */ #define SIZEOF_DOUBLE 8 /* The size of a `float', as computed by sizeof. */ #define SIZEOF_FLOAT 4 /* The size of a `int', as computed by sizeof. */ #define SIZEOF_INT 4 /* The size of a `long', as computed by sizeof. */ #define SIZEOF_LONG 4 /* The size of a `long long', as computed by sizeof. */ #define SIZEOF_LONG_LONG 8 /* The size of a `short', as computed by sizeof. */ #define SIZEOF_SHORT 2 /* The size of a `unsigned char', as computed by sizeof. */ #define SIZEOF_UNSIGNED_CHAR 1 /* The size of a `unsigned int', as computed by sizeof. */ #define SIZEOF_UNSIGNED_INT 4 /* The size of a `unsigned long', as computed by sizeof. */ #define SIZEOF_UNSIGNED_LONG 4 /* The size of a `unsigned long long', as computed by sizeof. */ #define SIZEOF_UNSIGNED_LONG_LONG 8 /* The size of a `unsigned short', as computed by sizeof. */ #define SIZEOF_UNSIGNED_SHORT 2 /* The size of a `void *', as computed by sizeof. */ #define SIZEOF_VOID_P 4 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
[PATCH] update copyright, cite realistic version numbers, fix a smattering of typos
---
configure.ac | 20 ++--
1 files changed, 10 insertions(+), 10 deletions(-)
diff --git a/configure.ac b/configure.ac
index eda926b..b9f000d 100644
--- a/configure.ac
+++ b/configure.ac
@@ -3,7 +3,7 @@ dnl (run grep '^dnl \*' configure.ac | sed -e 's/dnl / /g;
s/\*\*/ +/g;'
dnl (or some such) to see the outline of this file)
dnl
#
-# (c) The University of Glasgow 1994-2004
+# (c) The University of Glasgow 1994-2012
#
# Configure script template for GHC
#
@@ -18,11 +18,11 @@ AC_INIT([The Glorious Glasgow Haskell Compilation System],
[7.5], [glasgow-haske
# Set this to YES for a released version, otherwise NO
: ${RELEASE=NO}
-# The primary version (e.g. 6.7, 6.6.1) is set in the AC_INIT line
+# The primary version (e.g. 7.5, 7.4.1) is set in the AC_INIT line
# above. If this is not a released version, then we will append the
-# date to the version number (e.g. 6.7.20070204). The date is
+# date to the version number (e.g. 7.4.20111220). The date is
# constructed by finding the date of the most recent patch in the
-# darcs repository. If this is a source distribution (not a darcs
+# git repository. If this is a source distribution (not a git
# checkout), then we ship a file 'VERSION' containing the full version
# when the source distribution was created.
@@ -257,7 +257,7 @@ then
NM=$hardtop/inplace/mingw/bin/nm.exe
fp_prog_ar_raw=$hardtop/inplace/mingw/bin/ar.exe
-# NB. If you update the tarbballs to a new version of gcc, don't
+# NB. If you update the tarballs to a new version of gcc, don't
# forget to tweak the paths in driver/gcc/gcc.c.
if ! test -d inplace/mingw ||
test inplace/mingw -ot ghc-tarballs/mingw/binutils*.tar.lzma ||
@@ -669,8 +669,8 @@ then
AC_DEFINE([HAVE_LIBM], [1], [Define to 1 if you need to link with libm])
fi
-dnl ** check whether this machine has BFD and liberty installed (used for
debugging)
-dnlthe order of these tests matters: bfd needs liberty
+dnl ** check whether this machine has BFD and libiberty installed (used for
debugging)
+dnlthe order of these tests matters: bfd needs libiberty
AC_CHECK_LIB(iberty, xmalloc)
AC_CHECK_LIB(bfd,bfd_uncompress_section_contents)
@@ -694,7 +694,7 @@ dnl --
dnl ** can we get alloca?
AC_FUNC_ALLOCA
-dnl ** Working vfork?
+dnl ** working vfork?
AC_FUNC_FORK
dnl ** determine whether or not const works
@@ -750,7 +750,7 @@ if test -n $GTK_CONFIG; then
fi
AC_SUBST([GTK_CONFIG])
-#Checking for PAPI
+# checking for PAPI
AC_CHECK_LIB(papi, PAPI_library_init, HavePapiLib=YES, HavePapiLib=NO)
AC_CHECK_HEADER([papi.h], [HavePapiHeader=YES], [HavePapiHeader=NO])
AC_SUBST(HavePapiLib)
@@ -799,7 +799,7 @@ AC_CONFIG_FILES([mk/config.mk mk/install.mk mk/project.mk
compiler/ghc.cabal ghc
AC_CONFIG_COMMANDS([mk/stamp-h],[echo timestamp mk/stamp-h])
AC_OUTPUT
-# We get caught by
+# We got caught by
# http://savannah.gnu.org/bugs/index.php?1516
# $(eval ...) inside conditionals causes errors
# with make 3.80, so warn the user if it looks like they're about to
--
1.7.7.4
___
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc
HTYPE_USECONDS_T missing on RHEL 5
Hi all, I am trying to bootstrap the latest source snapshot on a RHEL system: ggreif@mine [!138] uname -a Linux mine 2.6.18-274.7.1.el5 #1 SMP Mon Oct 17 11:57:14 EDT 2011 x86_64 x86_64 x86_64 GNU/Linux causes compilation error when building base lib with a stage1 compiler: inplace/bin/ghc-stage1 -prof -H32m -O-package-name base-4.5.0.0 -hide-all-packages -i -ilibraries/base/. -ilibraries/base/dist-install/build -ilibraries/base/dist-install/build/autogen -Ilibraries/base/dist-install/build -Ilibraries/base/dist-install/build/autogen -Ilibraries/base/include -optP-DOPTIMISE_INTEGER_GCD_LCM -optP-include -optPlibraries/base/dist-install/build/autogen/cabal_macros.h -package ghc-prim-0.2.0.0 -package integer-gmp-0.4.0.0 -package rts-1.0 -split-objs -package-name base -XHaskell98 -XCPP -O2 -no-user-package-conf -rtsopts -odir libraries/base/dist-install/build -hidir libraries/base/dist-install/build -stubdir libraries/base/dist-install/build -hisuf p_hi -osuf p_o -hcsuf p_hc -c libraries/base/./Foreign/C/Types.hs -o libraries/base/dist-install/build/Foreign/C/Types.p_o libraries/base/Foreign/C/Types.hs:210:31: Not in scope: type constructor or class `HTYPE_USECONDS_T' libraries/base/Foreign/C/Types.hs:210:189: Not in scope: type constructor or class `HTYPE_USECONDS_T' libraries/base/Foreign/C/Types.hs:210:269: Not in scope: type constructor or class `HTYPE_USECONDS_T' libraries/base/Foreign/C/Types.hs:210:385: Not in scope: type constructor or class `HTYPE_USECONDS_T' libraries/base/Foreign/C/Types.hs:210:462: Not in scope: type constructor or class `HTYPE_USECONDS_T' libraries/base/Foreign/C/Types.hs:210:545: Not in scope: type constructor or class `HTYPE_USECONDS_T' looking at configury: ./libraries/base/config.log:configure:12452: checking Haskell type for suseconds_t ./libraries/base/config.log:fptools_cv_htype_sup_suseconds_t=yes ./libraries/base/config.log:fptools_cv_htype_sup_useconds_t=no ./libraries/base/config.log:fptools_cv_htype_suseconds_t=Int64 ./libraries/base/config.log:fptools_cv_htype_useconds_t=NotReallyAType these checks end up in ggreif@mine [!129] find . -name *.h | xargs grep useconds_t ./libraries/base/include/HsBaseConfig.h:/* Define to Haskell type for suseconds_t */ ./libraries/base/include/HsBaseConfig.h:/* Define to Haskell type for useconds_t */ but without a #define for HTYPE_USECONDS_T! So I edited the Foreign/C/Types.hs file to read ggreif@mine [!127] xemacs +210 libraries/base/Foreign/C/Types.hs -- | Haskell type representing the C @clock_t@ type. ARITHMETIC_TYPE(CClock,tyConCClock,CClock,HTYPE_CLOCK_T) -- | Haskell type representing the C @time_t@ type. ARITHMETIC_TYPE(CTime,tyConCTime,CTime,HTYPE_TIME_T) #ifdef HTYPE_USECONDS_T -- | Haskell type representing the C @useconds_t@ type. ARITHMETIC_TYPE(CUSeconds,tyConCUSeconds,CUSeconds,HTYPE_USECONDS_T) #elif defined(HTYPE_SUSECONDS_T) ARITHMETIC_TYPE(CUSeconds,tyConCUSeconds,CUSeconds,HTYPE_SUSECONDS_T) #endif #ifdef HTYPE_SUSECONDS_T -- | Haskell type representing the C @suseconds_t@ type. ARITHMETIC_TYPE(CSUSeconds,tyConCSUSeconds,CSUSeconds,HTYPE_SUSECONDS_T) #elif defined(HTYPE_USECONDS_T) ARITHMETIC_TYPE(CSUSeconds,tyConCSUSeconds,CSUSeconds,HTYPE_USECONDS_T) #endif The strategy is to offer the opposite signed type when either one is missing. Does this make sense and should it be included into the next v7.4 snapshot (as well as mainline)? The 7.2 branch could also get this change... Cheers, Gabor ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc
