Tracker: Bugs

5 configure, make, preprocessor - ID: 556119
Last Update: Settings changed ( hpa )

I have had some trouble with compiling 0.98.30 and 0.98.31 under
MinGW. Using bash instead of sh has solved some of my problems. I am
able to get both executables to compile, except 0.98.30 works and
0.98.31 does not.

I am not sure if this is because of the
configure/make have changed slightly or because of the actual
code.

The problem seems to be isolated to the preprocessor. In
my case, the processor has difficulty properly working out %ifidni
%elifidni %else %end blocks. Most of the errors seem to be snowball
from this.

NASM 0.98.30 has this problem when configured under
sh, but is okay when configured under bash. NASM 0.98.31 has this
problem under both sh and bash.

Without a better understanding
of hw NASM works I am pretty clueless as to how these problems could
relate, or if they do at all.

Dylan Bowker ( dbowker ) - 2002-05-14 14:30:41 PDT

5

Closed

Fixed

Nobody/Anonymous

None

None

Public


Comments ( 3 )

Date: 2002-05-22 16:02:42 PDT
Sender: hpaProject Admin

Logged In: YES
user_id=58697

Could you please try this with the current CVS release?

Date: 2002-05-22 15:38:32 PDT
Sender: hpaProject Admin

Logged In: YES
user_id=58697

This could be related to bug [ 558528 ] - %elif blocks broken.

Date: 2002-05-15 15:56:35 PDT
Sender: fbkotlerProject Admin

Logged In: YES
user_id=68913

I'm even more clueless, never having used MinGW. Is "sh" a
shell in itself, or a soft link to something else? (not that
I'd know the difference between bash and other shells...)

It seems odd that a configuration change would allow it to
compile at all, and not function properly, but that sounds
like what's happening.

Can you determine what *is* happening in this %ifidni block?

Attached File

No Files Currently Attached

Changes ( 3 )

Field Old Value Date By
status_id Open 2002-05-24 19:29:59 PDT hpa
close_date - 2002-05-24 19:29:59 PDT hpa
resolution_id None 2002-05-24 12:42:13 PDT dbowker