I am a little puzzled about something I just noticed about the
executable file format of my Debian Sarge install....

It seems to be running gcc 3.3.5, and if I run file(1) on the output
of the compiler I get:

a.out: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), for GNU/Linux 
2.2.0, dynamically linked (uses shared libs), not stripped

On the old system I upgraded from (SuSE 7.3), the compiler was version 2.95.3
but running file(1) again (also on the new Debian) I get

a.out: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), for GNU/Linux 
2.2.5, dynamically linked (uses shared libs), not stripped

For some reason, the executable from the old comiler is for GNU/Linux '2.2.5',
but the output of the newer Debain compiler indicates '2.2.0'...

And finally, I have an older executable, purchased as part of a commercial
package so that I can't recompile it, which ran fine on my old SuSE kernel
(2.4.10), but my Debian 2.6.8 returns quite a misleading error message:

% file /tmp/foo
/tmp/foo: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically 
linked (uses shared libs), stripped

% /tmp/foo
bash: /tmp/foo: No such file or directory

Note that file(1) this time makes no mention of a GNU/Linux version. I was
rather expecting some potential for problems with dynamic libraries, but
I am not get any sensible messages:

% ldd /tmp/foo
/usr/bin/ldd: line 1: /tmp/foo: No such file or directory

If I run strace, it suggests that the  "No such file or directory",
status is being returned by execve(2):
% strace /tmp/foo 
strace: exec: No such file or directory
execve("/tmp/foo", ["/tmp/foo"], [/* 20 vars */]) = 0

But surely that would be inappropriate when the file does exist
% ls -l /tmp/foo
-r-xr-xr-x  1 root root 99096 2006-02-05 18:58 /tmp/foo

and is just (perhaps) an unsupported executable format....

Does anyone have any idea what is going on here?

I was rather hoping as a worst case to have gotten some form of
'unsupported executable format' error message if there was a
compatability problem...

Regards,
DigbyT
-- 
Digby R. S. Tarvin                                          digbyt(at)digbyt.com
http://www.digbyt.com


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Reply via email to