[comp.sources.d] Recent PCmail posting

richard@calvin.spp.cornell.edu (Richard Brittain) (02/04/90)

Hi,
    I have persuaded the recent pcmail package to compile under Turbo-C,
and it more or less runs, but tc detects a null pointer assignment at
run time.  If I exercise any of the options, the prompts gradually
get destroyed as the character pointers get messed up.  There may be some
residual differences between MSC and TC, in particular it seems that
setjump/longjump take pointers to jmp_buf structs, rather than the
struct itself - can anyone with a MSC manual verify this.  
Anyway, it seems to go bad before any of the setjump code is reached.  Has
anyone else tried this with TC yet ?

Richard Brittain,                   School of Elect. Eng.,  Upson Hall   
                                    Cornell University, Ithaca, NY 14853
ARPA: richard@calvin.spp.cornell.edu	
UUCP: {uunet,uw-beaver,rochester,cmcl2}!cornell!calvin!richard

wswietse@lso.win.tue.nl (Wietse Venema) (02/06/90)

richard@calvin.spp.cornell.edu (Richard Brittain) writes:

>    I have persuaded the recent pcmail package to compile under Turbo-C,
>and it more or less runs, but tc detects a null pointer assignment at
>run time.  

Note from the author: pc-mail needs lots of stack space. I never tried less 
than 8 kbyte. The MicroSoft C default (2 kbyte?) is definitely too small.
Too small stacks will cause problems with other compilers as well.