home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.mips
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!mips!sdd.hp.com!usc!orion.oac.uci.edu!gordius!gordius!dan
- From: dan@gordian.com (Dan Lambright)
- Subject: weird compiling on a RISC 5.0
- Message-ID: <1992Aug21.001007.24697@gordian.com>
- Keywords: mips 5.0 , compiling
- Sender: news@gordian.com
- Organization: Gordian; Costa Mesa, CA
- Date: Fri, 21 Aug 1992 00:10:07 GMT
- Lines: 28
-
-
- cc -g -N -T 0 -EL test.c -c
- ld -g -N -m -EL -T 00000000 test.o -o test.mon
-
- I am not able to use utilities such as "dbx" and "nm" when compiling
- a one line program using the above options on the new RISC 5.0 machines.
- I have not had that problem when I compile using a 4.52. I've traced
- the problem to the symbol tables.
-
- I compared the "test.mon" generated by the 5.0 compiler with
- the one compiled on 4.52. There was a difference in the
- file header. The f_nscns element from struct filehdr was 5 on our
- 5.0 machine, and 4 when compiled on the 4.52 machine.
-
- Dissasembling the symbol tables, I found that for both the 4.52
- and 5.0 compilers generated .text, .sdata, .sbss, and .bss sections,
- in that order. The 5.0 compiler, however, produced a 5th
- section: ".comment". Say what?!?
-
- My question is: why does the 5.0 compiler create a 5th section called
- .comment? .comment is undoccumented in our manuals. What is it?
-
- .comment is not produced when I compile in big endian.
-
-
- Dan
-
- dan@gordian.com
-