Dave Korn | 3 Jun 11:42 2008

FW: [Chicken-users] Minor problem with cmake-based chicken build on cygwin

felix winkelmann wrote on 02 June 2008 21:36:

> On Sat, May 31, 2008 at 8:00 PM, Nathan Thern <nthern <at> > wrote:
>> Hi all-
>> I just built chicken-3.2.0 on cygwin. The build went just fine, but the
>> install put the runtime dlls (cygchicken-0.dll and cyguchicken-0.dll) in
>> /usr/lib rather than /usr/bin. The installation did not work until I
>> moved the dlls to /usr/bin. 

  My problem is always that I can't compile cygchicken-0.dll without first
compiling cygegg-0.dll.  Unfortunately cygegg-0.dll has a dependency on
cyguchicken-0.dll so I can't compile cygegg-0.dll without first compiling
cygchicken-0.dll.

  Does anyone know a way to resolve this situation, or perhaps just a better
way to describe it?

>> A quick inspection of the makefiles revealed that DLLSINPATH is not set
>> in Makefile.cygwin (whereas it is in the mingw & msvc makefiles).
>> 
>> The obvious fix is to define DLLSINPATH in the make arguments, but since
>> the build is broken unless that is done, it looks like a bug to me.
>> Perhaps this is a cmake bug, or a bug in the chicken source?
> 
> Hi!
> 
> I guess this is a bug in our build - thanks for reporting it. BTW, we
> don't use CMake anymore, so this has nothing to do with CMake itself,
> cygwin or the now historical cmake-based chicken build.

(Continue reading)

Phil Betts | 3 Jun 14:03 2008

RE: [Chicken-users] Minor problem with cmake-based chicken build on cygwin

Dave Korn wrote on Tuesday, June 03, 2008 10:42 AM::

>   My problem is always that I can't compile cygchicken-0.dll without
> first compiling cygegg-0.dll.  Unfortunately cygegg-0.dll has a
> dependency on cyguchicken-0.dll so I can't compile cygegg-0.dll
> without first compiling cygchicken-0.dll.
> 
>   Does anyone know a way to resolve this situation, or perhaps just a
> better way to describe it?
> 

IIRC a hacked junglefowl package was used to create the initial 
cygegg-0.dll

Note: It has been hinted that large portions of the closed-source 
junglefowl can be found in the chicken sau^H^Hource, but so far the 
creator of junglefowl has provided no evidence to back up these claims.

Phil
--

-- 

One of the following statements is true:
This email has not been scanned by Ascribe PLC using Microsoft Antigen
for Exchange.

This email has been scanned by Ascribe PLC using Microsoft Antigen for Exchange.

Christopher Faylor | 3 Jun 17:30 2008

Re: FW: [Chicken-users] Minor problem with cmake-based chicken build on cygwin

On Tue, Jun 03, 2008 at 10:42:04AM +0100, Dave Korn wrote:
>felix winkelmann wrote on 02 June 2008 21:36:
>
>> On Sat, May 31, 2008 at 8:00 PM, Nathan Thern <nthern <at> > wrote:
>>> Hi all-
>>> I just built chicken-3.2.0 on cygwin. The build went just fine, but the
>>> install put the runtime dlls (cygchicken-0.dll and cyguchicken-0.dll) in
>>> /usr/lib rather than /usr/bin. The installation did not work until I
>>> moved the dlls to /usr/bin. 
>
>  My problem is always that I can't compile cygchicken-0.dll without first
>compiling cygegg-0.dll.  Unfortunately cygegg-0.dll has a dependency on
>cyguchicken-0.dll so I can't compile cygegg-0.dll without first compiling
>cygchicken-0.dll.

You need to use the road cross compiler, I think.  That should get you
to the other side of this problem.

cgf

Dave Korn | 3 Jun 17:55 2008

RE: FW: [Chicken-users] Minor problem with cmake-based chicken build on cygwin

Christopher Faylor wrote on 03 June 2008 16:30:

> On Tue, Jun 03, 2008 at 10:42:04AM +0100, Dave Korn wrote:
>> felix winkelmann wrote on 02 June 2008 21:36:
>> 
>>> On Sat, May 31, 2008 at 8:00 PM, Nathan Thern <nthern <at> > wrote:
>>>> Hi all-
>>>> I just built chicken-3.2.0 on cygwin. The build went just fine, but the
>>>> install put the runtime dlls (cygchicken-0.dll and cyguchicken-0.dll)
>>>> in /usr/lib rather than /usr/bin. The installation did not work until I
>>>> moved the dlls to /usr/bin.
>> 
>>  My problem is always that I can't compile cygchicken-0.dll without first
>> compiling cygegg-0.dll.  Unfortunately cygegg-0.dll has a dependency on
>> cyguchicken-0.dll so I can't compile cygegg-0.dll without first compiling
>> cygchicken-0.dll.
> 
> You need to use the road cross compiler, I think.  That should get you
> to the other side of this problem.
> 
> cgf

  ... <rimshot>

    cheers,
      DaveK
--

-- 
Can't think of a witty .sigline today....

(Continue reading)

Robert Pendell | 19 Jun 22:09 2008

Re: --add-stdcall-alias equivalent for Visual-C++ ?


Jim Kleckner wrote:
| Jim Kleckner wrote:
|> An off-topic request:
|>
|> The --add-stdcall-alias argument to ld exports symbols as both
|> "decorated" (with leading _ and trailing  <at> nn) and "undecorated".
|>
|> Anyone know of the easiest way to get VC++ to do this?
|> Yes, a manual exports .def file can be created (or semi-automated).
|>
|> But it is so nice that ld does this which is needed for Excel or other
|> DLL-consuming Windoze programs.
|>
|
| FYI, after searching and trying lots of hacks, I came up with
| including a "pragma comment" as the easiest way to manage
| this requirement.  Microsoft doesn't appear to want to make
| this easy the way gcc/ld does.
|
| Something like:
|
| #pragma comment(linker, "/EXPORT:func1=_func1 <at> 4")
|
|

Why am I thinking this is off-topic even for the talk list?  This is a
cygwin list.  Not VC++.  *whistles*

--
(Continue reading)

Dave Korn | 20 Jun 12:22 2008

RE: --add-stdcall-alias equivalent for Visual-C++ ?

Robert Pendell wrote on 19 June 2008 21:10:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Jim Kleckner wrote:
>> Jim Kleckner wrote:
>>> An off-topic request:
>>> 
>>> The --add-stdcall-alias argument to ld exports symbols as both
>>> "decorated" (with leading _ and trailing  <at> nn) and "undecorated".
>>> 
>>> Anyone know of the easiest way to get VC++ to do this?
>>> Yes, a manual exports .def file can be created (or semi-automated).
>>> 
>>> But it is so nice that ld does this which is needed for Excel or other
>>> DLL-consuming Windoze programs.
>>> 
>> 
>> FYI, after searching and trying lots of hacks, I came up with
>> including a "pragma comment" as the easiest way to manage
>> this requirement.  Microsoft doesn't appear to want to make
>> this easy the way gcc/ld does.
>> 
>> Something like:
>> 
>> #pragma comment(linker, "/EXPORT:func1=_func1 <at> 4")
>> 
>> 
> 
(Continue reading)

Dave Korn | 20 Jun 12:58 2008

RE: Problems compiling grep and friends

sandra200 wrote on 20 June 2008 02:02:

> Hello
>      Good Day
> My name is sandra i saw your profile today at (nabble.com) and became
> intrested
> in you,i will also like to know you the more,

  Good day Sandra, My name as you will know from nabble is Cygwin, and I am
an abstract concept embodied in a suite of software utilities.

>  and i want you
> to send an email to my email address, so i can give you my picture for you

  I am a suite of software.  I have no eyes.  I don't think there's even an
OCR package in the standard distro.  Still, it's nice of you to offer.

> to know whom i believe we can move from here to next level 

  But before we can do that we must conquer the end-of-level boss!  I hope
you brought your railgun, it's a hundred-foot tall half-zombie
half-shambling-mass that has piranhas on the ends of its fingers!

> I am waiting for 
> your mail to my maill address (sandra.johnson82 <at> yahoo.com )
> Remeber the distance or colour does not matter 

  Yes, but I would have thought being a living biological entity was at
least semi-relevant.

(Continue reading)

Igor Peshansky | 20 Jun 19:27 2008
Picon

RE: Problems compiling grep and friends

On Fri, 20 Jun 2008, Dave Korn wrote:

> sandra200 wrote on 20 June 2008 02:02:
> [snip]
>
> > to know whom i believe we can move from here to next level
>
>   But before we can do that we must conquer the end-of-level boss!  I
> hope you brought your railgun, it's a hundred-foot tall half-zombie
> half-shambling-mass that has piranhas on the ends of its fingers!

Ah, but the hero is a hundred-foot tall hippo, so the boss is no match for
him...  er, her...  er, it...

> [snip]
> > (sandra.johnson82 <at> yahoo.com) so i can give you my picture and will can
> > move
>
>   Will?  Will!?  Who the hell is Will?  Listen darling, two's company,
> three's a crowd.

Will is the next love.

> > from here to next love. my lovely one,

See, she's moving on to Will.  It's a rejection letter, don't you get it?
	Igor
--

-- 
				http://cs.nyu.edu/~pechtcha/
      |\      _,,,---,,_	    pechtcha <at> cs.nyu.edu | igor <at> watson.ibm.com
(Continue reading)

Dave Korn | 24 Jun 16:13 2008

FW: [rurban: Re: setup.exe command line parameters]

Dave Korn wrote on 24 June 2008 14:59:

> reasonable and proportionate to a Rea-Soon-Now release timeframe:

</walks off into sunset, releasing setup.exe to the strains of "Auberge">

    cheers,
      DaveK
--

-- 
Can't think of a witty .sigline today....

Christopher Faylor | 25 Jun 21:39 2008

Re: perl-5.10.0-4 going curr?

On Wed, Jun 25, 2008 at 02:49:55PM +0000, Eric Blake wrote:
>Unfortunately, this caught me at the start of a 2-week vacation;

What???  Did you clear this with anyone?

I don't think the Cygwin project can allow two weeks of vacation
from a key resource.

Maybe a couple of days, I could see but how is free software going to
survive if people keep taking vacations?  This tendency for open source
people to "go on vacation" shows a clear focus on fun and that's why
open source will *never* be taken seriously.

cgf


Gmane