jdevito@brutus.cs.uiuc.edu (John Devitofranceschi) (10/22/89)
Well, after playing around with the macro call to "ASM_OUTPUT_SKIP" in final.c (that is to say, eliminating it and replacing it with what was in 1.35 (for only final.c!)), I have what appears to be a 'working' gcc 1.36. I have put working in quotes because I noticed some weirdness going on when I compiled the thing with itself the second time around (stage2): 1) "cc1" got "fatal signal 9" seemingly randomly; the compiler would die once, but then work fine after starting it again on a given module. 2) When I "cmp"'d the stage 2 object files with the final object files, there were 2 files that had differences. Blowing those two files away and recompiling them made everything work okay. What I'd like to know is: has anyone else had these problems? If you're desperate enough to have the binaries for gcc 1.36, you are welcome to have some of mine. If you maintain an anonymous ftp unix-pc archive and would like to house these things for general retrieval, please let me know. No one should have to put up with the hassle of getting this beast up and running once it's already been done. Again, please send me e-mail with requests to test drive the program, give it a good home, or comment on the problems that I mentioned. I'll summarize in a couple of weeks if I get lots o' responses. cheers, jd ----------------------------------------------------------------------- John Devitofranceschi jdevito@a.cs.uiuc.edu (217) 332-2458
brant@manta.pha.pa.us (Brant Cheikes) (10/24/89)
John Devitofranceschi writes about his problems with gcc 1.36, and then says: >If you're desperate enough to have the binaries for gcc 1.36, you are >welcome to have some of mine. If you maintain an anonymous ftp >unix-pc archive and would like to house these things for general >retrieval, please let me know. I would simply like to remind everyone that a general archive of unix-pc software is maintained on osu-cis. This archive is accessible both by uucp and by anonymous Internet FTP. The archive presently contains binaries of gcc-1.36, compiled for the unix-pc. To the best of my knowledge, these binaries suffer none of the problems John describes (I've been using gcc 1.36 for a few weeks now with no problems). I maintain the osu-cis unix-pc archives, so if you have any questions about them, send me mail. Access info, briefly (note that complete access details are regularly posted by Karl Kleinpaste to newsgroup comp.sources.d). By UUCP, all files are in osu!~/att7300/*. Retrieve README.Z to see the latest listing of contents. In your L.sys file, you will need a line something like this: osu-cis Any ACU 1200 16142923112 "" \d\r\c Name? osu-cis GO \d\r\d\r in:--in: Uanon By FTP, just connect to cis.ohio-state.edu and poke around in the pub/archives/att7300 area. Happy snarfing. -- Brant Cheikes University of Pennsylvania, Department of Computer and Information Science brant@manta.pha.pa.us, brant@linc.cis.upenn.edu, bpa!manta!brant