Mail Archives: djgpp-workers/2000/04/07/08:02:20
At 20:32 06/04/00 +0200, you wrote:
>On 6 Apr 2000, at 18:48, Pierre Muller wrote:
>
>>
>> Using CVS repository
>> I tried to compile the current tree,
>>
>> First, I wolud like to state that indeed configure works without
problems !
>>
>> Second, I wanted to use -gstabs option instead of -g
>> (the is a line count limitation with -g that is overflooded)
>>
>> I am using gcc 2.952 with DJGPP 2.03
>>
>> but changing "CFLAGS= -g -O" into "CFLAGS=-gstabs -O"
>> leads to an unexpected and unnice result :
>>
>> The labels used for the stabn generation of source line encoding are
labeled
>> by
>> LM???
>> where ??? represents a number,
>>
>> the problem is that gcc only considers .L???? labels as local labels
>> thus it keep all these LM??? labels
>> and when debugging gdb with itself a backtrace gives
>> a list of LM??? without args :(
>>
>> Does anyone have an idea what is wrong here ???
>>
>> I am quite sure it worked better with older gcc versions !!
>> (I tested gcc v 2.951 but this one also generates LM???)
>
>I suspect You are using binutils-2.8.1. I suggest to install
>bnu251b.zip.
Thanks this worked OK !
I still got several warnings in gnu-regex.c compilation
configure has set
ENABLE_NLS and HAVE_GETTEXT to 1
I have gtext032b installed and this seems to be the latest version !
Warnings that I get
gnu-regex.c: In function `re_compile_pattern':
gnu-regex.c:5470: warning: return makes pointer from integer without a cast
gnu-regex.c: In function `re_comp':
gnu-regex.c:5500: warning: return makes pointer from integer without a cast
gnu-regex.c: In function `regerror':
gnu-regex.c:5824: warning: assignment makes pointer from integer without a
cast
all lines have expressions of style :
return gettext (re_error_msgid[(int) ret]);
Note that there are some other line with re_error_msgid that
have an explicit (char *) typecast for the result of gettext !
I could not find the definition of gettext in libintl.h !!
Is this problem DJGPP specific ?
Shouldn't gettext allways be (char *) ??
Or it is a bug in the CVS tree ?
Pierre Muller
Institut Charles Sadron
6,rue Boussingault
F 67083 STRASBOURG CEDEX (France)
mailto:muller AT ics DOT u-strasbg DOT fr
Phone : (33)-3-88-41-40-07 Fax : (33)-3-88-41-40-99
- Raw text -