home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!gatech!news.ans.net!cmcl2!adm!news
- From: philip@concave.cs.wits.ac.za (Philip Machanick)
- Newsgroups: comp.sys.sgi
- Subject: Cfront 3.0.1 bug?
- Message-ID: <34594@adm.brl.mil>
- Date: 12 Dec 92 13:58:21 GMT
- Sender: news@adm.brl.mil
- Lines: 31
-
- I just tried to compile a program that cfront 2 compiled.
-
- I have a virtual function that is called via an element of an array
- of objects. It is quite a complex situation because the array is
- 3-dimensional and implemented via classes (overloading "[]") rather
- than a "C"-style 3-D array.
-
- Cfront 3.0.1 generates a direct call to the function without going
- via the virtual function pointer and also neglects to insert the
- "this" pointer.
-
- Only when the resulting mangled code gets through to the C compiler
- is an error reported:
-
- accom: Error: M_particle.u, line 51: Actual parameters count 0 while
- formal parameter count is 1
-
- This looks like a bug to me.
-
- Anyone encountered something similar?
-
- (I can supply the source and cfront 2 and 3 output by mail on request -
- probably not of general interest.)
-
- Is there a bug report procedure if not on maintenance? Is there such a
- thing as a bug on warantee repair? (Luckily the -v2 flag allows me to
- revert to cfront 2.)
- --
- Philip Machanick
- Computer Science Dept, Univ of the Witwatersrand, 2050 Wits, South Africa
- philip@concave.cs.wits.ac.za phone: 27 (11) 716-3759 fax: 339-7965
-