(Illustration by Gaich Muramatsu)
On Wed, 20 Jan 1999, Jason Duerstock wrote: > Perhaps some sort of watermark could be set for what is acceptable to > cache as a whole file and what is not? Um, easiest solution would be to cache whole files when they're being written to, since typically (i.e. non-dbm case) you write the whole file anyway. > stackptr is originally typed as (char *)...in Alpha-land, pointers are 8 > bytes and sizeof(void *) == sizeof(long), not sizeof(int)... So gcc's "long" on the alpha really is 8 bytes? Is there a "long long" too? > What kind of evil voodoo is this and what's the proper way to fix it? if (stackptr == (char*)-1) would be my guess. But that's still dangerous; bounds-checking compilers will flag it. I guess stackptr=NULL already means something else? -- <[email protected]> Kragen Sitaker <http://www.pobox.com/~kragen/> Computers are the tools of the devil. It is as simple as that. There is no monotheism strong enough that it cannot be shaken by Unix or any Microsoft product. The devil is real. He lives inside C programs. -- philg_at_mit.eduReceived on 1999-01-20 16:39:21