Luke,
I didn't experience anything like you described. Here's as much detail as I can think of about what I did:
* Mac OS/X 10.8.4
* Xcode 4.6
* ghc-7.6.3 downloaded from http://haskell.org/ghc as the bootstrap compiler
* llvm - 3.0 built from source downloaded off the llvm site:
Low Level Virtual Machine (http://llvm.org/):
llvm version 3.0
Optimized build.
Built Jan 31 2012 (09:33:35).
Host: x86_64-apple-darwin12.4.0
Host CPU: penryn
Registered Targets:
alpha - Alpha [experimental]
arm - ARM
... and others ...
Before I had a problem with not enough memory but I bought some more. So I got a cross compiler built, but wasn't able to test it, because I can't authorize my iPod Touch for development at the moment (Apple systems are down). So I can't verify that it produces iOS executables that work.
The latest commit in the git log for GHC HEAD was this:
commit e8aa8ccba0c40884765281b21ff8f4411802dd41
Author: Richard Eisenberg <eir@cis.upenn.edu>
Date: Fri Aug 2 15:47:03 2013 +0100
Implement "roles" into GHC.
...
I patched it as attached here (same patch as in my last email).
The procedure I used was exactly the one described at http://ghc.haskell.org/trac/ghc/wiki/Building/CrossCompiling/iOS
Thanks for your work on this! That problem you're having is pretty weird!
Steve
On 11/08/13 20:14, Luke Iannini wrote:
And the truly final word for the moment : ) —
I built a tool to partially automate the indentation workaround for LLVM
3.0 and it yields the same "co-processor offset out of
range"/"unsupported relocation on symbol LCPI65_0" errors LLVM 3.3/3.4
did when it finally gets to integer-simple/GHC/Integer/Type.hs.
On Sun, Aug 11, 2013 at 3:06 AM, Luke Iannini <lukexipd@gmail.com<mailto:lukexipd@gmail.com>> wrote:
OK! So just to summarize:
Building GHC HEAD with LLVM 3.0 or 3.2 (using GHC 7.6.3 as the
bootstrap) on OS X 10.9 DP5/Xcode 5 DP5 exhibits very strange
behavior wherein layout-based code along with mixed-tabs-and-spaces
code fails to parse correctly, with issues in hundreds of files in
the GHC HEAD tree.
I don't have a 10.8 machine to check if this is a 10.9 exclusive
issue, so I'd love if someone can try using these binaries to build
GHC HEAD:
http://llvm.org/releases/3.0/clang+llvm-3.0-x86_64-apple-darwin11.tar.gz
Building GHC HEAD with LLVM 3.3 or 3.4 works great as a regular
compiler with the 10.9 workarounds I outlined in another thread, but
fails when compiling as a cross-compiler (./configure
--target=arm-apple-darwin10) with these errors:
https://gist.github.com/lukexi/2b129f34fa027172c5ee
So I'm between a rock and a hard place at the moment.
The only (very tedious and slow) workaround I've found for the
3.0/3.2 bug is to manually expand tabs to spaces, and to transform
do x
y
into
do
x
y
(similarly for where and let blocks)
Cheers
Luke
On Sun, Aug 11, 2013 at 1:53 AM, Luke Iannini <lukexipd@gmail.com<mailto:lukexipd@gmail.com>> wrote:
Argh, sorry for the confusion: 3.2 *does* exhibit the issue. 3.3
and 3.4 do not.
On Sun, Aug 11, 2013 at 1:39 AM, Luke Iannini<lukexipd@gmail.com <mailto:lukexipd@gmail.com>> wrote:<http://llvm.org/releases/3.0/clang+llvm-3.0-x86_64-apple-darwin11.tar.gz> and
Further investigation:
I grabbed 7.6.3 just to see if I somehow had a bad install
of GHC, but the problem still occurred.
The problem only occurs with LLVM 3.0.
It is not related to cross-compilation or Stephen's patches:
I tested this on multiple fresh clones with --with-gcc=clang.
LLVM 3.2, 3.3 and 3.4 do not exhibit the issue.
If anyone wants to try to reproduce, you can grab the LLVM
3.0 binaries here Clang Binaries for MacOS X/x86-64
just drop them in your path.
(Stephen, I'm now trying your patch with LLVM 3.2)
Cheers
Luke
On Sat, Aug 10, 2013 at 8:11 PM, Luke Iannini<lukexipd@gmail.com <mailto:lukexipd@gmail.com>> wrote:
The first error on a fresh checkout is
"/usr/local/bin/ghc" -hisuf hi -osuf o -hcsuf hc
-static -H32m -O -package-db
libraries/bootstrapping.conf -hide-all-packages -i
-iutils/hsc2hs/. -iutils/hsc2hs/dist/build
-iutils/hsc2hs/dist/build/autogen
-Iutils/hsc2hs/dist/build
-Iutils/hsc2hs/dist/build/autogen -optP-include
-optPutils/hsc2hs/dist/build/autogen/cabal_macros.h
-package base-4.6.0.1 -package containers-0.5.0.0
-package directory-1.2.0.1 -package filepath-1.3.0.1
-package process-1.1.0.2 -XHaskell98 -XCPP
-XForeignFunctionInterface -no-user-package-db
-rtsopts -odir utils/hsc2hs/dist/build -hidir
utils/hsc2hs/dist/build -stubdir utils/hsc2hs/dist/build
-c utils/hsc2hs/./C.hs -o utils/hsc2hs/dist/build/C.o
utils/hsc2hs/C.hs:155:3:
parse error (possibly incorrect indentation or
mismatched brackets)
There seem to be two classes of error: one is the layout
issue above, but other files can be fixed by simply
running 'expand' on them.
On Sat, Aug 10, 2013 at 6:42 PM, Luke Iannini<mailto:likeliest.complexions.stephen@blacksapphire.com>><lukexipd@gmail.com <mailto:lukexipd@gmail.com>> wrote:
Hi Stephen/all,
I got LLVM 3.0 installed and started building again
but hit a very strange problem now wherein tons of
layout-based code (as in
http://en.wikibooks.org/wiki/Haskell/Indentation) is
suddenly erroring out, e.g.
compiler/coreSyn/CoreUnfold.lhs:481:2:
parse error (possibly incorrect indentation or
mismatched brackets)
(some files also seem to be triggered by mixed tabs
and spaces)
I can fix the errors one by one by converting the
code to use more concrete indentation (like
do
thing1
thing2
)
but it's all over the tree.
Anyone have any idea what might cause this?
Cheers
Luke
On Fri, Aug 9, 2013 at 6:14 AM, Stephen Blackheath
[to GHC-iPhone]
<likeliest.complexions.stephen@blacksapphire.comhttps://gist.github.com/__lukexi/7ca55b36269703236f1f
wrote:
Luke,
Try llvm version 3.0 - that's what I'm using,
and it definitely worked before. llvm-3.1 is
broken for GHC+ARM. As for llvm >= 3.2, I'm not
sure if it's been fixed yet, but it wasn't
working last time I tried a couple of months
ago. I think this was because llvm is getting
fussier about its input and GHC hasn't been
"tightened up" yet.
It's really easy to build llvm from source.
Steve
On 09/08/13 20:35, Luke Iannini wrote:
v3 output:<mailto:lukexipd@gmail.com
<https://gist.github.com/lukexi/7ca55b36269703236f1f>
On Fri, Aug 9, 2013 at 4:34 AM, Luke Iannini
<lukexipd@gmail.com <mailto:lukexipd@gmail.com>https://gist.github.com/__lukexi/d9f8bfd8bca56d5d0ee9
<mailto:lukexipd@gmail.com>>> wrote:
OK, that got me past that one.
Now I'm stuck here during compilation
of integer-simple:<mailto:lukexipd@gmail.com
<https://gist.github.com/lukexi/d9f8bfd8bca56d5d0ee9>
("unsupported relocation on
symbol"/"co-processor offset out of range")
On Fri, Aug 9, 2013 at 4:00 AM, Luke
Iannini <lukexipd@gmail.com
<mailto:lukexipd@gmail.com>-Iincludes/dist-__derivedconstants/header
<mailto:lukexipd@gmail.com>>> wrote:
OK, I'm underway on this.
First roadbump was:
"inplace/bin/ghc-stage1" -static
-H32m -O -Iincludes
-Iincludes/dist
-Iincludes/dist-ghcconstants/__header -Irts<mailto:lukexipd@gmail.com
-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 -O2 -c
rts/Apply.cmm -o
rts/dist/build/Apply.o
You are using a new version of LLVM
that hasn't been tested yet!
We will try though...
/usr/local/bin/llc: : error: unable
to get target for
'arm-apple-darwin10', see --version
and --triple.
which I figured out were because
the homebrew LLVM 3.4 only
includes host platforms by default
(x86/x86-64)
Reinstalling it with all-targets
enables them all:
brew install llvm --with-clang
--all-targets --HEAD
Trying again now.
On Thu, Aug 8, 2013 at 5:29 PM,
Luke Iannini <lukexipd@gmail.com
<mailto:lukexipd@gmail.com><mailto:carter.schonwald@__gmail.com
<mailto:lukexipd@gmail.com>>> wrote:
Update: I've got GHC HEAD
building on 10.9 again, tonight
I'll dive into the iOS patch!
Cheers
Luke
On Wed, Aug 7, 2013 at 9:12 PM,
Carter Schonwald
<carter.schonwald@gmail.com
<mailto:carter.schonwald@gmail.com>
<mailto:lukexipd@gmail.com
<mailto:carter.schonwald@gmail.com>>> wrote:
if you're on 10.8, just DL
the 4.6 CLI tools to get
things working again.
If you're on 10.9, you'll
need to use recent GHC head +
CLANG HEAD, (ok, the most
recent XCODE 5 dev preview CLI
tools may have all the
needed clang patches, but thats
still needing its own
validation, orthogonal to the IOS
related issues )
On Tue, Aug 6, 2013 at 7:06
PM, Luke Iannini
<lukexipd@gmail.com
<mailto:lukexipd@gmail.com><likeliest.complexions.__stephen@blacksapphire.com
<mailto:lukexipd@gmail.com>>> wrote:
Hi Stephen,
I installed 10.9 for
another project and thus am in
the middle of fixing up
my system to be able to
build GHC again, but I
think I'm close — as soon as
I have it going I'll
try this.
In the meantime someone
still on 10.8 should
definitely try the patch!
Hope you're great!
Luke
On Tue, Aug 6, 2013 at
6:42 PM, Stephen Blackheath
[to GHC-iPhone]
<mailto:likeliest.complexions.stephen@blacksapphire.com>
<mailto:likeliest.complexions.__stephen@blacksapphire.comhttp://ghc.haskell.org/trac/____ghc/wiki/Building/____CrossCompiling/iOS
<mailto:likeliest.complexions.stephen@blacksapphire.com>>>
wrote:
Hi all,
A couple of months
ago I got some patches
applied to GHC
mainline for the ability to cross
compile to iOS.
I've been doing other things and
now I want to test
those changes. The trouble
is, I am having
Apple trouble and I can't get
Xcode to work.
If anyone is
interested in testing it for me,
I'd appreciate it.
The patch attached
here needs to be applied to
GHC head. Using
this, I was able to get the
cross compiler to
build. The instructions are here:
<http://ghc.haskell.org/trac/__ghc/wiki/Building/__CrossCompiling/iOS>_________________________________________________ <mailto:iPhone@haskell.org
<http://ghc.haskell.org/trac/__ghc/wiki/Building/__CrossCompiling/iOS
<http://ghc.haskell.org/trac/ghc/wiki/Building/CrossCompiling/iOS>>
Steve
<mailto:iPhone@haskell.org>>
http://www.haskell.org/__mailman/listinfo/iphone
<http://www.haskell.org/mailman/listinfo/iphone>
_________________________________________________ <mailto:iPhone@haskell.org
<mailto:iPhone@haskell.org>>
http://www.haskell.org/__mailman/listinfo/iphone
<http://www.haskell.org/mailman/listinfo/iphone>