search

therahuldahiya / iphone-dev

Automatically exported from code.google.com/p/iphone-dev
0 stars 0 forks source link

errors making odcctools...errors.c on Windows XP #40

Open GoogleCodeExporter opened 8 years ago

GoogleCodeExporter commented 8 years ago 
What steps will reproduce the problem?
1. cd ~/iphone-dev/odcctools
2. make
3. allocate.c builds, but errors.c fails

What is the expected output? What do you see instead?
errors.o expected, but get this instead:
gcc -Wall -Wno-import  -DHAVE_CONFIG_H -D__LITTLE_ENDIAN__=1   -I../include
-I..
/include -include ../include/extern.h -I../include/foreign -g -O2
-fno-builtin-r
ound -fno-builtin-trunc   -c -o errors.o errors.c
In file included from /usr/include/stdio.h:46,
                 from errors.c:26:
/usr/include/sys/types.h:104: error: parse error before "clock_t"
/usr/include/sys/types.h:109: error: parse error before "time_t"
/usr/include/sys/types.h:115: error: parse error before "time_t"
/usr/include/sys/types.h:117: error: parse error before '}' token
/usr/include/sys/types.h:120: error: field `it_interval' has incomplete type
/usr/include/sys/types.h:121: error: field `it_value' has incomplete type
/usr/include/sys/types.h:256: error: parse error before "clockid_t"
/usr/include/sys/types.h:261: error: parse error before "timer_t"
In file included from /usr/include/cygwin/types.h:21,
                 from /usr/include/sys/types.h:373,
                 from /usr/include/stdio.h:46,
                 from errors.c:26:
/usr/include/stdint.h:20: error: conflicting types for 'int32_t'
../include/i386/types.h:82: error: previous declaration of 'int32_t' was here
In file included from /usr/include/sys/types.h:373,
                 from /usr/include/stdio.h:46,
                 from errors.c:26:
/usr/include/cygwin/types.h:43: error: parse error before "loff_t"
In file included from ../include/mach/machine/vm_types.h:29,
                 from ../include/mach/port.h:79,
                 from ../include/mach/std_types.h:63,
                 from ../include/mach/mach.h:60,
                 from errors.c:29:
../include/mach/i386/vm_types.h:96: error: conflicting types for 'vm_offset_t'
/usr/include/cygwin/types.h:163: error: previous declaration of
'vm_offset_t' wa
s here
../include/mach/i386/vm_types.h:107: error: conflicting types for 'vm_size_t'
/usr/include/cygwin/types.h:168: error: previous declaration of 'vm_size_t' was
here
In file included from ../include/foreign/machine/endian.h:32,
                 from ../include/mach/mach_traps.h:72,
                 from ../include/mach/mach_init.h:75,
                 from ../include/mach/mach.h:64,
                 from errors.c:29:
../include/i386/endian.h:86:1: warning: "LITTLE_ENDIAN" redefined
In file included from /usr/include/cygwin/types.h:22,
                 from /usr/include/sys/types.h:373,
                 from /usr/include/stdio.h:46,
                 from errors.c:26:
/usr/include/endian.h:28:1: warning: this is the location of the previous
defini
tion
In file included from ../include/foreign/machine/endian.h:32,
                 from ../include/mach/mach_traps.h:72,
$
                 from ../include/mach/mach.h:64,
                 from errors.c:29:
../include/i386/endian.h:87:1: warning: "BIG_ENDIAN" redefined
In file included from /usr/include/cygwin/types.h:22,
                 from /usr/include/sys/types.h:373,
                 from /usr/include/stdio.h:46,
                 from errors.c:26:
/usr/include/endian.h:29:1: warning: this is the location of the previous
defini
tion
In file included from ../include/foreign/machine/endian.h:32,
                 from ../include/mach/mach_traps.h:72,
                 from ../include/mach/mach_init.h:75,
                 from ../include/mach/mach.h:64,
                 from errors.c:29:
../include/i386/endian.h:88:1: warning: "PDP_ENDIAN" redefined
In file included from /usr/include/cygwin/types.h:22,
                 from /usr/include/sys/types.h:373,
                 from /usr/include/stdio.h:46,
                 from errors.c:26:
/usr/include/endian.h:30:1: warning: this is the location of the previous
defini
tion
In file included from ../include/foreign/machine/endian.h:32,
                 from ../include/mach/mach_traps.h:72,
                 from ../include/mach/mach_init.h:75,
                 from ../include/mach/mach.h:64,
                 from errors.c:29:
../include/i386/endian.h:90:1: warning: "BYTE_ORDER" redefined
In file included from /usr/include/cygwin/types.h:22,
                 from /usr/include/sys/types.h:373,
                 from /usr/include/stdio.h:46,
                 from errors.c:26:
/usr/include/endian.h:31:1: warning: this is the location of the previous
defini
tion
errors.c:135: warning: visibility attribute not supported in this
configuration;
 ignored
make[1]: *** [errors.o] Error 1
make[1]: Leaving directory `/home/ahmiller/iphone-dev/odcctools/libstuff'
make: *** [libstuff] Error 2

What version of the product are you using? On what operating system?
Checked out revision 170 of iphone-dev, using Cygwin 1.5.24 on Windows XP SP2
$ uname -a
CYGWIN_NT-5.1 ahmiller-t60 1.5.24(0.156/4/2) 2007-01-31 10:57 i686 Cygwin

Please provide any additional information below.
Thanks for the allocate fix...I'll keep hammering away with XP as long as
your willing to provide fixes or advice!

Original issue reported on code.google.com by aaronmiller@gmail.com on 15 Sep 2007 at 6:09

GoogleCodeExporter commented 8 years ago

Original comment by nightwat...@gmail.com on 15 Sep 2007 at 1:44

GoogleCodeExporter commented 8 years ago 
I get this too with cygwin, I'm trying MinGW now.

Original comment by geo...@gmail.com on 17 Sep 2007 at 8:32

GoogleCodeExporter commented 8 years ago 
[deleted comment]
GoogleCodeExporter commented 8 years ago 
Looks like the -I../include/foreign pulls in odcctools/include/i386/types.h and
defines _MACHTYPES_H_, so cygwin never parses /usr/include/machine/types.h which
contains clock_t and time_t.  This patch to gets through this file, but the 
next file
has the reverse problem: type.h loads from the cygwin-side first...

[Add to odcctools/include/i386/types.h after #include]:
#ifdef __CYGWIN__
// Define non-BSD types that Cygwin is expecting.
#define _CLOCK_T_   unsigned long       /* clock() */
#define _TIME_T_    long            /* time() */
#define _CLOCKID_T_     unsigned long
#define _TIMER_T_       unsigned long
typedef long int __loff_t;
// Inform Cygwin that basic types have been defined BSD-style.
#define __int8_t_defined
#define __vm_offset_t_defined
#define __vm_size_t_defined
#endif

Original comment by martinturon on 23 Sep 2007 at 11:42

GoogleCodeExporter commented 8 years ago 
This revision will compile farther, but it is neither clean nor complete:

[Add to odcctools/include/i386/types.h after #include]:

#ifdef __CYGWIN__
// Define non-BSD types that Cygwin is expecting.
#define _CLOCK_T_   unsigned long       /* clock() */
#define _TIME_T_    long            /* time() */
#define _CLOCKID_T_     unsigned long
#define _TIMER_T_       unsigned long
// Inform Cygwin that basic types have been defined BSD-style.
#define __vm_offset_t_defined
#define __vm_size_t_defined
typedef long int __loff_t;

#ifdef __int8t_defined
// Force Cygwin to foreign INT definitions
#define __int8_t_defined
#else
// Use Cygwin INT definitions
#define _INT8_T
#define _INT16_T
#define _INT32_T
#endif

#endif  //__CYGWIN__

Original comment by martinturon on 23 Sep 2007 at 12:11

GoogleCodeExporter commented 8 years ago 
Thanks Martin...after your suggestion, it does get a little farther, but alas, 
still
fails compilation:

$ make
cd libstuff && make
make[1]: Entering directory `/home/ahmiller/iphone-dev/odcctools/libstuff'
gcc -Wall -Wno-import  -DHAVE_CONFIG_H -D__LITTLE_ENDIAN__=1   -I../include -I..
/include -include ../include/extern.h -I../include/foreign -g -O2 -fno-builtin-r
ound -fno-builtin-trunc   -c -o errors.o errors.c
In file included from /usr/include/cygwin/types.h:21,
                 from /usr/include/sys/types.h:373,
                 from /usr/include/stdio.h:46,
                 from errors.c:26:
/usr/include/stdint.h:20: error: conflicting types for 'int32_t'
../include/i386/types.h:82: error: previous declaration of 'int32_t' was here
In file included from ../include/foreign/machine/endian.h:32,
                 from ../include/mach/mach_traps.h:72,
                 from ../include/mach/mach_init.h:75,
                 from ../include/mach/mach.h:64,
                 from errors.c:29:
../include/i386/endian.h:86:1: warning: "LITTLE_ENDIAN" redefined
In file included from /usr/include/cygwin/types.h:22,
                 from /usr/include/sys/types.h:373,
                 from /usr/include/stdio.h:46,
                 from errors.c:26:
/usr/include/endian.h:28:1: warning: this is the location of the previous defini
tion
In file included from ../include/foreign/machine/endian.h:32,
                 from ../include/mach/mach_traps.h:72,
                 from ../include/mach/mach_init.h:75,
                 from ../include/mach/mach.h:64,
                 from errors.c:29:
../include/i386/endian.h:87:1: warning: "BIG_ENDIAN" redefined
In file included from /usr/include/cygwin/types.h:22,
                 from /usr/include/sys/types.h:373,
                 from /usr/include/stdio.h:46,
                 from errors.c:26:
/usr/include/endian.h:29:1: warning: this is the location of the previous defini
tion
In file included from ../include/foreign/machine/endian.h:32,
                 from ../include/mach/mach_traps.h:72,
                 from ../include/mach/mach_init.h:75,
                 from ../include/mach/mach.h:64,
                 from errors.c:29:
../include/i386/endian.h:88:1: warning: "PDP_ENDIAN" redefined
In file included from /usr/include/cygwin/types.h:22,
                 from /usr/include/sys/types.h:373,
                 from /usr/include/stdio.h:46,
                 from errors.c:26:
/usr/include/endian.h:30:1: warning: this is the location of the previous defini
tion
In file included from ../include/foreign/machine/endian.h:32,
                 from ../include/mach/mach_traps.h:72,
                 from ../include/mach/mach_init.h:75,
                 from ../include/mach/mach.h:64,
                 from errors.c:29:
../include/i386/endian.h:90:1: warning: "BYTE_ORDER" redefined
In file included from /usr/include/cygwin/types.h:22,
                 from /usr/include/sys/types.h:373,
                 from /usr/include/stdio.h:46,
                 from errors.c:26:
/usr/include/endian.h:31:1: warning: this is the location of the previous defini
tion
errors.c:135: warning: visibility attribute not supported in this configuration;
 ignored
make[1]: *** [errors.o] Error 1
make[1]: Leaving directory `/home/ahmiller/iphone-dev/odcctools/libstuff'
make: *** [libstuff] Error 2

Original comment by aaronmiller@gmail.com on 24 Sep 2007 at 8:39

GoogleCodeExporter commented 8 years ago 
Martin I'm trying to understand what you meant by this code here :

#ifdef __int8t_defined
// Force Cygwin to foreign INT definitions
#define __int8_t_defined
#else
// Use Cygwin INT definitions
#define _INT8_T
#define _INT16_T
#define _INT32_T
#endif

what is the '__int8t_defined' versus '__int8_t_defined'? I tried changing the 
code to
this :
#ifndef __int8_t_defined
// Force Cygwin to foreign INT definitions
#define __int8_t_defined
#else
// Use Cygwin INT definitions
#define _INT8_T
#define _INT16_T
#define _INT32_T
#endif

That seems like what you may have intended. It fixes some of the problems but 
there
were several warnings and another error.

odcctools/libstuff/bytesex.c
In file included from /usr/include/sys/types.h:373,
from ../../../odcctools/include/mach-o/ranlib.h:28,
from ../../../odcctools/libstuff/bytesex.c:207
/usr/include/cygwin/types.h:195: error: conflicting types for 'register_t'
../../../odcctools/include/i386/types.h:120: error: previous declaration of
'register_t' was here

But it doesn't say where it was :P

Original comment by leyl...@gmail.com on 2 Oct 2007 at 1:31

GoogleCodeExporter commented 8 years ago 
I was looking up what odcctools is and if any one has built it under cygwin, I 
cam
across this page...
http://labs.docudesk.com/latest-technologies/2007/5/8/macos-target-gcj-create-ma
ctelppc-binaries-from-os-x-linux-and-cygwin.html

...don't have time to look into it further though.

Original comment by leyl...@gmail.com on 2 Oct 2007 at 1:48

GoogleCodeExporter commented 8 years ago 
This docudesk.com instructions basically work (except that OpenDarwin project 
closed
and you need to replace odcctools URL in Makefile for, say,
http://biolpc22.york.ac.uk/pub/linux-mac-cross/odcctools-20060413.tar.bz2), but 
they
use older, simpler version of odcctools, which do not include conflicting 
types.h file.

And they do not build the freshest GNU C/C++, but it is OK - they build i386/PPC
odcctools at least.

I am trying to analyze differences between older odcctools and the ones from
iphone-dev project.

Original comment by vic...@gmail.com on 2 Oct 2007 at 5:28

GoogleCodeExporter commented 8 years ago 
How can I join project?

Original comment by vic...@gmail.com on 3 Oct 2007 at 6:36

GoogleCodeExporter commented 8 years ago 
odcctools built successfully.

There are several problems, most harmful is clash of
odcctools/include/foreign/machine with native Cygwin machine include 
directories.
Patch can not remove/rename directories, so you need to do it manually:

cd odcctools
patch -p0 <odcctools_cygwin.patch
mv include/foreign/machine include/foreign/_machine

then cd ../build/odcctools

configure, make, make install

You're here!

Original comment by vic...@gmail.com on 5 Oct 2007 at 5:05

Attachments:

GoogleCodeExporter commented 8 years ago 
Am I the first having built the toolkit under Cygwin? ;-)
It actually built without a hitch after odcctools patch.

Testing how it compiles iPhone apps.

Original comment by vic...@gmail.com on 5 Oct 2007 at 6:27

GoogleCodeExporter commented 8 years ago 
nice work, guys. 
lucky me, i first tried after you published the patch   ;) 
odcctools builds in cygwin, just do not leave odcctools directory as supposed 
in the
main build instructions. 
let's head on to the next traps.

Original comment by naughty....@gmail.com on 7 Oct 2007 at 9:25

GoogleCodeExporter commented 8 years ago 
I managed to build Motion app from http://lucasnewman.com/phonedemos.zip, but I
needed to copy GraphicsServices.h from Berlios repository at:

http://svn.berlios.de/svnroot/repos/iphone-binutils/trunk

Somehow it contains fresher version with more functions defined.

Also I found that there is probably some deficiency in Objective-C compiler.

It does not like construct:

LKTransform sublayerTransform = LKTransformIdentity;

where LKTransfromIdentity is defined in LayerKit/LKTransform.h as

extern LKTransform LKTransformIdentity;

It complains
FlipACoinApp.m:19: error: initializer element is not constant

Is it known bug?

Original comment by vic...@gmail.com on 8 Oct 2007 at 6:23

GoogleCodeExporter commented 8 years ago 
It builds FlipACoin, include GraphicsServices was missed in the source.

Original comment by vic...@gmail.com on 9 Oct 2007 at 5:14

GoogleCodeExporter commented 8 years ago 
In attach you can find logs for my install session toolchain on CygWin.
I worry about 2 things:
1. I have many warnings(see attach for logs) when build toolchain. is they are 
critical?
2. Size of executables of test apps(iphonedemos from Lucas Newman with modified 
makefile) differ from size of apps builded on CygWin. Also i'm warry about this.
On early builds toolchain i compile FlipACoin, and the executable runs on 
iPhone 
extremely fast and don't play animation with sunbeams.

My working toolchain based on llvm rev.42269. Test apps in attach build on it.

Original comment by yo.l...@gmail.com on 23 Oct 2007 at 9:15

Attachments:

GoogleCodeExporter commented 8 years ago 
Hard to tell - you enclosed so many files, many of them are not even logs, that
signal/noise ratio is not very high. From what I've seen, there are many benign
warnings. Some warnings are for real deficiencies of the programs, but it is 
not my
job to fix them.

Executable sizes - many exes are still built with toolchain 0.20 on MacOS or 
Linux.
They use different backend version and so must be different, the point is they 
are
running.

FlipACoin - yea, you;re right, I have the sunbeams, but in very fast manner. I 
will
look at the code, how the animation is implemented. I just built the toolchain 
and
verified that it builds some stuff. I don't have many iPhone programs, 
especially
really complicated ones to test toolchain.

For some time nobody was even interested in Cygwin port at all, so I stopped 
actively
working on it.

Original comment by vic...@gmail.com on 25 Oct 2007 at 9:25

GoogleCodeExporter commented 8 years ago 

victzh, Thank you for the odcctools patch it worked like a charm.

I had already compiled odcctools and I'm now stopped on the last step. OMG, I'm 
almost there.

When I ran the command "make LLVM_VERSION_INFO=2.0-svn-iphone-dev-0.3-svn" 
after 35 
minutes it give me an error. "... -lc".

I've checked LLVMOBJDIR and HEAVENLY and the symlinks into iPhone paths. 

My question is simple, are you finished to compile the toolchain and build any 
HelloWorld to make sure it works in cygwin?

If you have any tip in my problem it will be appreciated.

Original comment by igorcha...@gmail.com on 6 Nov 2007 at 2:19

GoogleCodeExporter commented 8 years ago 
I followed the instruction from here: 
http://iphone.fiveforty.net/wiki/index.php/Toolchain_installation_-_Windows

However, when i tried to run ./configure. I have got the following errors: 
/home/llvm-svn
$'r':command not found
syntax error near the unexpected token 'new line'

Any ideas what could be the problem?

Original comment by eln...@gmail.com on 7 Nov 2007 at 4:15

GoogleCodeExporter commented 8 years ago 
[deleted comment]
GoogleCodeExporter commented 8 years ago 
Hello, I followed the same instructions on iphone.fiveforty.net, but I'm having
trouble with the Mac OSX Headers.

On the Apple Tools download page, I have a choice between XCode 3.0 and XCode 
2.5.
Nowhere in the instructions does it say which one I need for IPhone dev ??

I've tried downloading the XCode 3.0 and I can extract the PKG files using
HFSExplorer or TransMac. Strangely everyone in here talks of these files as
directories?? I am unable to extract headers from the PKG files on windows or in
Cygwin. None of the mentioned tools will do it.

Please tell me how to extract files from PKG so I can get at the .pax.gz header
files! This step is missing from the instructions.

Original comment by anders.e...@gmail.com on 7 Nov 2007 at 1:23

GoogleCodeExporter commented 8 years ago 
Anders, I took XCode 2.5, since 3.0 seemed specifically for Leopard. Then dug 
into
the .dmg with TransMac and copied out the Archive.pax.gz file from the
MacOSX10.4.Universal.pkg/Contents dir and followed the advice from the main
instructions to pull out the headers. You'll have to replace some of the links 
to
certain libraries with the actual libraries...things must work differently in 
Cygwin
than other platforms. Good luck!

Original comment by aaronmiller@gmail.com on 9 Nov 2007 at 5:44

GoogleCodeExporter commented 8 years ago 
Thanks a bunch, Aaron, this information would have saved me some hours of 
frustration
since the XCode 3.0 dmg file contains PKG files instead of folders and there is
apparently no way of extracting these archives on a windows PC even with 
cygwin...
(correct me if I'm wrong).

Thanks to victzh's patch file I've now been able to compile and install 
everything
for iphone dev and hopefully I can compile and run hello world within moments :)

Original comment by anders.e...@gmail.com on 10 Nov 2007 at 10:42

GoogleCodeExporter commented 8 years ago 
Tried to follow the Wiki page and instructions but fail after the make for llvm:

arm-apple-darwin-lipo -output libgcc_s.10.4.dylib -create libgcc_s.10.4.dylib_T*

arm-apple-darwin-lipo: can't figure out the architecture type of: libgcc_s.10.4.
dylib_T
make[1]: *** [libgcc_s.10.4.dylib] Error 1

Any ideas?

Original comment by NetMage....@gmail.com on 10 Nov 2007 at 9:46

GoogleCodeExporter commented 8 years ago 
@NetMage.SCW: Did you use the installation guidelines on iphone.fiveforty.net ? 
I've
been able to install and compile the toolset twice on different machines now 
using
those guidelines, even managed to compile and run HelloWorld. If you did use 
them,
try manually deleting the entire checkout of llvm and checking it out again, 
then
follow the guidelines again, closely, step by step.

Original comment by anders.e...@gmail.com on 11 Nov 2007 at 4:21

GoogleCodeExporter commented 8 years ago 
@NetMage.SCW: Just realized you may rather be missing "usr/lib/libc.dylib" from 
your
iphone filesystem. You need this to be able to build the toolchain. Get a hold 
of
that and try again.

Original comment by anders.e...@gmail.com on 11 Nov 2007 at 4:27

GoogleCodeExporter commented 8 years ago 
Everybory, just to inform that I am building iPhone binaries from my Windows / 
Cygwin since Nov, 6.

Thanks victzh for the patch without it I would never get it.

Original comment by igorcha...@gmail.com on 12 Nov 2007 at 1:11

GoogleCodeExporter commented 8 years ago 
Yup, many many thanks to victch for getting us this far!

NetMage, I've run across that error myself and usually a "make clean" then a:
#  $ make LLVM_VERSION_INFO=2.0-svn-iphone-dev-0.3-svn
# $ make install 
solves it.

Also, to everyone else, while I have a "working" toolchain, I've noticed that 
not all
apps build correctly. For example, when I compile the "Motion" sample, I only 
get 1/3
indicators showing up. The FlipACoin app refuses to compile, throwing a syntax 
error
on code that works for others. Anyway, I suspect not all is well with toolchains
built on Windows/Cygwin, so proceed with caution...or tell us what other magic 
you're
using! :)

Original comment by aaronmiller@gmail.com on 12 Nov 2007 at 8:31

GoogleCodeExporter commented 8 years ago 
OK, glad to report that after replacing the GraphicsServices.h from the 
iphone-dev
repository with the one from
http://svn.berlios.de/svnroot/repos/iphone-binutils/trunk (as victzh noted), 
graphics
dependent apps like FlipACoin and Motion now compile & work. So, if you get a
non-functional Motion app or a non-compiling FlipACoin, be sure to check your
/usr/local/arm-apple-darwin/include/GraphicsServices/GraphicsServices.h file.

Original comment by aaronmiller@gmail.com on 12 Nov 2007 at 10:16

GoogleCodeExporter commented 8 years ago 
am I the only one getting an "undefined symbol: ___eprintf" whenever I try to 
compile
code using asserts? What is the correct way to fix this?

Original comment by anders.e...@gmail.com on 13 Nov 2007 at 1:58

GoogleCodeExporter commented 8 years ago 
Saw this note in Issue #64: recompiling with -D__DARWIN_UNIX03, helps with 
"___eprintf"

Original comment by aaronmiller@gmail.com on 13 Nov 2007 at 7:40

GoogleCodeExporter commented 8 years ago 
the topic in #iphone on irc.osx86.hu has included a link to
http://wiki.iphonegameover.com/Windows_Cygwin_Binary_Toolchain_Installation for 
some
time.

Original comment by beej666@gmail.com on 4 Jan 2008 at 11:49

GoogleCodeExporter commented 8 years ago 
When compiling this: http://wiki.iphonegameover.com/Hello_World_Tutorial
The HelloWorld program compiles fine ... but when I upload the HelloWorld.app 
in my
Iphone, the blue screen with Hello World doesn't show and the screen freezes 
... 
Anyone has any idea why ?
Please ... and Thankyou

Original comment by matin...@gmail.com on 23 Feb 2008 at 8:18

GoogleCodeExporter commented 8 years ago 
Never mind ...  I had to change the permissions ... to executable.

Original comment by matin...@gmail.com on 25 Feb 2008 at 12:54

GoogleCodeExporter commented 8 years ago 
It seems newest cygwin adopted mingw-gcc to generate native win32 exe, which 
causes
the libstdc++-v3 build failed. I tried to modify the libstdc++-v3/configure 
line 2390
from yes to no for bypassing the check and it seems works.

PS, as I used prefix to assign different root directory for installing, there 
are
some troubles while building. Finally, I find to create symbolic link under
/opt/iphone-toolchain/arm-apple-darwin/{include,lib,libexec,sys-include} is 
helpful,
they are all in upper level directory after a series of make install.

Original comment by pin...@gmail.com on 9 Jul 2008 at 4:15

GoogleCodeExporter commented 8 years ago 
I've applied the patch above but i still get similar errors on execute.c.

gcc -Wall -Wno-import  -DHAVE_CONFIG_H -D__LITTLE_ENDIAN__=1   -I../include -I..
/../../odcctools/include -include ../../../odcctools/include/extern.h -I../../..
/odcctools/include/foreign -g -O2 -fno-builtin-round -fno-builtin-trunc   -c -o
execute.o ../../../odcctools/libstuff/execute.c
In file included from /usr/include/sys/types.h:25,
                 from /usr/include/stdio.h:46,
                 from ../../../odcctools/libstuff/execute.c:26:
/usr/include/machine/_types.h:59: error: conflicting types for 'uint32_t'
/usr/include/stdint.h:28: error: previous declaration of 'uint32_t' was here
make[1]: *** [execute.o] Error 1
make[1]: Leaving directory `/home/iphone-dev/build/odcctools/libstuff'
make: *** [libstuff] Error 2

Does anybody know what can cause this? I've followed everything exactly.

Original comment by gearfo...@gmail.com on 18 Jul 2008 at 12:31

GoogleCodeExporter commented 8 years ago 
I think i know the cause partially, it's because i'm compiling for 2.0 using 
the 
tutorial on the wikee.iphwn.org website.
They've updated their odccctools there so probably the patch doesn't apply 
correctly...
Do you know if someone can make a patch for this version please 
(odcctools-ld9.2 
branch from svn).
Thanks in advance.

Original comment by gearfo...@gmail.com on 18 Jul 2008 at 1:33

GoogleCodeExporter commented 8 years ago 
@gearforce I had that very same error.  My quick fix for this is just erase 
your 
odcctools build directory and configure again, then make, make install, etc.  
I'm sorta 
stuck at including the OS X headers, but oh well.  I've been trying to do this 
for a 
good three days on my eeePC with cygwin under XP and this is by far the best 
guide.  
Thanks victzh, that patch really pushed my installation ahead!

Original comment by sudo.adam.carruthers@gmail.com on 18 Dec 2009 at 6:51