mingw.org link in User's Guide - domain does not appear to be owned by mingw project

2022-07-10 Thread Lynn Winebarger
There's a link in the "Building DLLs" section of the user guide that
is supposed to go to the MingW home page.  It looks like that domain
name is no longer under the control of the MingW project.
https://www.cygwin.com/cygwin-ug-net/dll.html#dll-build

Lynn

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


Re: Bad EXE format (error 193)

2007-10-05 Thread Lynn Winebarger
I forgot to cc the list, but the origin of the issue might be of
interest to someone.

On 10/5/07, Dave Korn [EMAIL PROTECTED] wrote:
   Well, this was a weird one.  I think the underlying problem must be a bug in
 larceny's final link stage during the build.  First, this command made your
 binary executable for me:

 objcopy -R .comment larceny.bin larceny2.exe

 [Detailed analysis omitted]
   So, the file is regarded as malformed by the loader because it contains a
 section that isn't correctly aligned to the file alignment.  I don't know how
 it got that way, but it's clearly inconsistent.  It might be that the section
 was supposed to have EXCLUDE or some other flag that would have made the
 loader not care, I don't know, but since it's just a comment section,
 discarding it with objdump -R does the job nicely.

Thanks, Dave!  With that information, I have tracked down the
problem.  The original Makefile uses nasm -o foo.o foo.asm -f elf
-g.  I had subsequently changed this to -f gnuwin32 (no -g),  but
the make clean did not actually erase that particular object file.
 The comment only appears in files made with -f elf and -g (at
least, it doesn't appear with -f gnuwin32 -g, or -f win32 -g).
The section has alignment 2**0, compared to 2**2 of every other
section.
I'm surprised ld (or collect2, I don't know how different they
are) did not at least complain about this if it wasn't willing to pad
the section.

Thanks,

Lynn

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



Bad EXE format (error 193)

2007-10-04 Thread Lynn Winebarger
   I am trying to get the Larceny Scheme compiler
(http://www.ccs.neu.edu/home/will/Larceny/) to compile (and work)
under the most recent stable Cygwin release.  While I have managed to
get it to produce a file that very closely resembles a PE file, it
apparently is not close enough to satisfy Windows XP Home Edition SP
2.
   Unfortunately, objdump and other file analyzers have no problem
with the file, while none of the debuggers I tried had anything
helpful to say about the problem.  I have compiled other things under
Cygwin (PLT Scheme, for example) that not only produced an executable
image but produced one Windows would actually run.  I have tried doing
a diff against the headers between this and the Larceny image and
cannot see a glaring error (but I am not an expert in either Windows
executables or Larceny's source code).
I cannot find anything helpful on this error in the mailing list
archives.  Is there a tool that would identify the problem, or maybe a
kind expert with some guidance on resolving this issue?

Thanks,
Lynn

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



Re: Bad EXE format (error 193)

2007-10-04 Thread Lynn Winebarger
(Sorry for the formatting, I belatedly subscribed to the list and got
this as a thread digest)
 From: Dave Korn [EMAIL PROTECTED]

  On 04 October 2007 14:56, Dave Korn wrote:
 
We might be able to make guesses at what was wrong with the compiled
  program if you told us *in what way* it is not close enough to satisfy
  Windows XP.
 
Ah.  And I just took a closer look at the subject line.  So I'm guessing
  you see an error message along those lines, yes?  Hm.  Does it still happen
  if you compile the most basic sort of hello world program?


   It might also be informative to run cygcheck path-to-exe on your
 compiled executable; that'll display the dependent DLLs for you.  A bit of
 googling suggests that 193 can be caused by a corrupt/bogus executable, but
 equally by a good executable thatt depends on a DLL which is faulty.


Thanks, Dave.  I did compile a hello-world.c program, and it ran, but
I had also compiled PLT scheme v360 before (needed to bootstrap
Larceny) and had run it without problem as well (I had thought this
might be a permission issue, because what bash actually reports is
Access denied - the underlying error took further investigation).

$ cygcheck ./larcenybin.exe
.\larcenybin.exe
  C:\cygwin\bin\cygwin1.dll
C:\WINDOWS\system32\ADVAPI32.DLL
  C:\WINDOWS\system32\ntdll.dll
  C:\WINDOWS\system32\KERNEL32.dll
  C:\WINDOWS\system32\RPCRT4.dll


Thanks,
Lynn

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



Re: Bad EXE format (error 193)

2007-10-04 Thread Lynn Winebarger
On 10/4/07, Dave Korn [EMAIL PROTECTED] wrote:
 On 04 October 2007 15:26, Lynn Winebarger wrote:
  Thanks, Dave.  I did compile a hello-world.c program, and it ran, but
  I had also compiled PLT scheme v360 before (needed to bootstrap
  Larceny) and had run it without problem as well (I had thought this
  might be a permission issue, because what bash actually reports is
  Access denied - the underlying error took further investigation)

   Actually, I was trying to ask if you'd compiled a basic helloworld with
 your new larceny compiler, or indeed if this example of yours is already a
 very simple test program; i.e. is it only complex larceny executables that
 don't work, or even the very simplest ones.

   I did not understand.  No, the larceny binary (runtime
system/interpreter) won't load at all - it doesn't even make it to the
entry point.  I'm more than willing to send the executable.  It
might be something an expert could quickly spot.  Or the linker map
output, if that would be useful.

Thanks,
Lynn

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



Re: Bad EXE format (error 193)

2007-10-04 Thread Lynn Winebarger
On 10/4/07, Lynn Winebarger [EMAIL PROTECTED] wrote:
 On 10/4/07, Dave Korn [EMAIL PROTECTED] wrote:
  On 04 October 2007 15:26, Lynn Winebarger wrote:
   Thanks, Dave.  I did compile a hello-world.c program, and it ran, but
   I had also compiled PLT scheme v360 before (needed to bootstrap
   Larceny) and had run it without problem as well (I had thought this
   might be a permission issue, because what bash actually reports is
   Access denied - the underlying error took further investigation)
 
Actually, I was trying to ask if you'd compiled a basic helloworld with
  your new larceny compiler, or indeed if this example of yours is already a
  very simple test program; i.e. is it only complex larceny executables that
  don't work, or even the very simplest ones.
 
I did not understand.  No, the larceny binary (runtime
 system/interpreter) won't load at all - it doesn't even make it to the
 entry point.  I'm more than willing to send the executable.  It
 might be something an expert could quickly spot.  Or the linker map
 output, if that would be useful.

   Or, if someone knows of a program, or a pointer to what black magic
Windows does to determine that a PE file is valid.  I have briefly
reviewed the PE documentation MS provides, but nothing has leaped out
at me (e.g. the flags on the sections appear consistent with their
specification).

Lynn

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