[ANNOUNCEMENT] Updated: llvm/clang-3.4.2-1

[ANNOUNCEMENT] Updated: llvm/clang-3.4.2-1

[Yaakov Selkowitz]

The following packages have been updated in the Cygwin distribution:

* llvm-3.4.2-1
* llvm-doc-3.4.2-1
* libllvm3.4-3.4.2-1
* libllvm-devel-3.4.2-1
* libllvm-devel-static-3.4.2-1
* clang-3.4.2-1
* clang-analyzer-3.4.2-1
* libclang-3.4.2-1
* libclang-devel-3.4.2-1
* libclang-devel-static-3.4.2-1
* python-clang-3.4.2-1
* python3-clang-3.4.2-1

The LLVM suite provides libraries and tools for code generation and
optimization.  Also included is Clang, an LLVM native C/C++/ObjC
compiler, and the Clang Static Analyzer, a tool that automatically finds
bugs in code.

This is an update to the latest upstream release, and the first release
for x86_64.  PLEASE NOTE that clang will NOT work as a native code
compiler for x86_64 at this time, but should still work as a code
analyzer and LLVM bytecode compiler; PTC.

Packaging changes in this version:

* clang is configured to use gcc-4.8.3.

* Static libraries for the individual LLVM/Clang APIs are available in
separate -devel-static packages.

* libclang bindings are also available for Python 3.

* The OCaml bindings were dropped because they failed to build.

-- 
Yaakov


CYGWIN-ANNOUNCE UNSUBSCRIBE INFO
================================

If you want to unsubscribe from the cygwin-announce mailing list, please
use the automated form at:

http://cygwin.com/lists.html#subscribe-unsubscribe

If this does not work, then look at the "List-Unsubscribe: " tag in the
email header of this message.  Send email to the address specified
there.  It will be in the format:

cygwin-announce-unsubscribe-you=yourdomain.com@cygwin.com

If you need more information on unsubscribing, start reading here:

http://sourceware.org/lists.html#unsubscribe-simple

Please read *all* of the information on unsubscribing that is available
starting at this URL.

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple

Re: [ANNOUNCEMENT] Updated: llvm/clang-3.4.2-1

[Corinna Vinschen]

[-- Attachment #1: Type: text/plain, Size: 1081 bytes --]

On Jul 23 21:23, Angelo Graziosi wrote:
> Corinna Vinschen wrote:
> >The default for native Windows applications is the small code model
> >[...]
> >Therefore my
> >collegue Kai Tietz provided GCC with implementations of a medium and
> >large code model
> 
> Gulp! Seems to re-read the Borland C++/TPascal manuals at the beginning of
> the 1990s... only the huge model is missed (and tiny, obviously..)

It's not that bad.  It's just about using a trampolin for symbols not
marked as dllimport/dllexport.  The medium and large code models allow
to drop the requirement to mark symbols as DLL symbols and still link
the executable correctly, even in the big 44 bit address room of x86_64.
The problem stems from the fact that the x86_64 64 bit instruction set
doesn't know relative jumps with more than 31 bit displacement.

The only really bad problem is, sombody has to create a patch for clang...


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Maintainer                 cygwin AT cygwin DOT com
Red Hat

[-- Attachment #2: Type: application/pgp-signature, Size: 819 bytes --]

Re: [ANNOUNCEMENT] Updated: llvm/clang-3.4.2-1

[Angelo Graziosi]

Corinna Vinschen wrote:
> The default for native Windows applications is the small code model
> [...]
> Therefore my
> collegue Kai Tietz provided GCC with implementations of a medium and
> large code model

Gulp! Seems to re-read the Borland C++/TPascal manuals at the beginning 
of the 1990s... only the huge model is missed (and tiny, obviously..)


Ciao,
Angelo.

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple

Re: [ANNOUNCEMENT] Updated: llvm/clang-3.4.2-1

[Corinna Vinschen]

[-- Attachment #1: Type: text/plain, Size: 1766 bytes --]

On Jul 23 18:37, Angelo Graziosi wrote:
> Sorry,
> 
> Yaakov Selkowitz wrote:
> >This is an update to the latest upstream release, and the first release
> >for x86_64.  PLEASE NOTE that clang will NOT work as a native code
> >compiler for x86_64 at this time, but should still work as a code
> >analyzer and LLVM bytecode compiler; PTC.
> 
> That explains why I get
> 
> $ cat t.c
> #include <stdio.h>
> 
> int main()
> {
>   printf("Hello, World!/n");
>   return 0;
> }
> 
> $ clang t.c
> /tmp/t-12d75f.o:fake:(.text+0x15): rilocazione adattata per troncamento:
> R_X86_64_32 contro ".rdata"
> collect2: error: ld returned 1 exit status
> clang: error: linker (via gcc) command failed with exit code 1 (use -v to
> see invocation)

This looks pretty much like a problem with the code model.  Cygwin
The default for native Windows applications is the small code model,
accessing all code and data via 32 bit code-relative instructions
and jump trampolines where necessary.

When we worked on x86_64 Cygwin, we realized that we need to change GCC
to support another model, otherwise we would see the above kind of build
error a lot for otherwise perfectly correct POSIX code.  Therefore my
collegue Kai Tietz provided GCC with implementations of a medium and
large code model with different types of relative and absolute
addressing.  64 bit Cygwin applications use the medium code model by
default.

Having said that, I assume the problem here is that clang doesn't
support anything other than the small code model.  This needs some
serious patching, I guess.


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Maintainer                 cygwin AT cygwin DOT com
Red Hat

[-- Attachment #2: Type: application/pgp-signature, Size: 819 bytes --]

Re: [ANNOUNCEMENT] Updated: llvm/clang-3.4.2-1

[Yaakov Selkowitz]

On Wed, 2014-07-23 at 18:37 +0200, Angelo Graziosi wrote:
> Yaakov Selkowitz wrote:
> > This is an update to the latest upstream release, and the first release
> > for x86_64.  PLEASE NOTE that clang will NOT work as a native code
> > compiler for x86_64 at this time, but should still work as a code
> > analyzer and LLVM bytecode compiler; PTC.
> 
> That explains why I get
> 
> $ clang t.c
> /tmp/t-12d75f.o:fake:(.text+0x15): rilocazione adattata per troncamento: 
> R_X86_64_32 contro ".rdata"
> collect2: error: ld returned 1 exit status
> clang: error: linker (via gcc) command failed with exit code 1 (use -v 
> to see invocation)
> 
> ?

Yes.


Yaakov



--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple

Re: [ANNOUNCEMENT] Updated: llvm/clang-3.4.2-1

[Angelo Graziosi]

Sorry,

Yaakov Selkowitz wrote:
> This is an update to the latest upstream release, and the first release
> for x86_64.  PLEASE NOTE that clang will NOT work as a native code
> compiler for x86_64 at this time, but should still work as a code
> analyzer and LLVM bytecode compiler; PTC.

That explains why I get

$ cat t.c
#include <stdio.h>

int main()
{
   printf("Hello, World!/n");
   return 0;
}

$ clang t.c
/tmp/t-12d75f.o:fake:(.text+0x15): rilocazione adattata per troncamento: 
R_X86_64_32 contro ".rdata"
collect2: error: ld returned 1 exit status
clang: error: linker (via gcc) command failed with exit code 1 (use -v 
to see invocation)

?

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple