Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!rutgers!columbia!cs.columbia.edu!abrams From: abrams@cs.columbia.edu (Steven Abrams) Newsgroups: comp.os.minix Subject: Re: Have you ever had this problem...? Message-ID: Date: 13 Mar 90 10:06:40 GMT References: <4296@hub.UUCP> Sender: news@cs.columbia.edu (The Daily News) Organization: Columbia University Lines: 42 In-Reply-To: mcm@peach.ucsb.edu's message of 13 Mar 90 02:18:25 GMT In article <4296@hub.UUCP> mcm@peach.ucsb.edu (Marcelo C. Mourier) writes: > When make is processing kernel/system.c cpp complains with the following me- > ssage: > > "system.c", line 89: fatal error -- cannot find include file "protect.h" > "./proc.h", line 28: } missing > "./proc.h", line 28: ; missing > > However, file protect.h is indeed present and it has the correct crc. The > same thing happens while make is processing fs/misc.c, cpp complains with: > > "misc.c", line 26: fatal error -- cannot find include file "super.h" > "./file.h", line 6: type_identifier missing > "./file.h", line 6: identifier missing > "./file.h", line 6: (warning) no semicolon after declarator > "./file.h", line 6: } missing > "./file.h", line 6: ; missing > > Again, file super.h does exist and has the correct crc. There were no such > problems while making mm; I don't have any idea of what could possibly be > happening... Any suggestions..? > Yes, this plagued me during one of my earlier upgrade attempts. You need to chmem all passes of the compiler way up to handle much of the new code, especially cpp. I just hit them all with: chmem /bin/cxxx +=60000 and this solved the problem. The compiler really ought to know that it ran out of memory and not that there's something wrong with your files. Oh well... ~~~Steve -- /************************************************* * *Steven Abrams abrams@cs.columbia.edu * **************************************************/ #include #include