Date: Wed, 1 Sep 2004 21:54:03 +1000 From: Peter Jeremy <PeterJeremy@optushome.com.au> To: Ruslan Ermilov <ru@freebsd.org>, Wilko Bulte <wilko@freebsd.org>, ticso@freebsd.org Cc: freebsd-alpha@freebsd.org Subject: Re: FreeBSD 5.3-BETA2/alpha successes/failures Message-ID: <20040901115403.GA60090@cirb503493.alcatel.com.au> In-Reply-To: <20040901090019.GB24389@ip.net.ua> References: <20040829151021.GA43674@bobbi.cse.buffalo.edu> <20040830191325.GA53006@freebie.xs4all.nl> <20040831120529.GB58440@cirb503493.alcatel.com.au> <20040901082845.GA60151@cirb503493.alcatel.com.au> <20040901090019.GB24389@ip.net.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 2004-Sep-01 12:00:19 +0300, Ruslan Ermilov wrote: >> ===> gnu/usr.bin/cc/cc1plus >> gperf -o -C -E -k '1-6,$' -j1 -D -N 'libc_name_p' -L ANSI-C /home/src5/gnu/usr. >> bin/cc/cc1plus/../../../../contrib/gcc/cp/cfns.gperf > cfns.h >> No words in input file, did you forget to prepend %% or use -t accidentally? >> *** Error code 1 >> >> Stop in /home/src5/gnu/usr.bin/cc/cc1plus. ... >Where in buildworld does this happen, cross-tools stage or "everything" >stage? It's the second gnu/usr.bin/cc/cc1plus in "stage 3: cross tools". Now that I've reported it as repeatable, it doesn't fail. The only difference is "which gperf" and "env|sort" in cc1plus/Makefile (as per your suggestion. I wonder if this was some sort of NFS glitch - maybe timing related, though I can't explain why it failed identically twice in a row. Once this buildworld completes (in about a day), I'll try another and see if I can work out what is happening. It's painful when a test run takes ~24 hours. On Wed, 2004-Sep-01 11:51:18 +0200, Wilko Bulte wrote: >> On a AS4100 4xEV5/466 and 8GB RAM (logically reduced to 2GB), >> it hangs just after GEOM configures md0c. > >I think ticso has no such problems on his AS4100(?) ticso might like to compare what's in his AS4100 with mine (verbose boot log was in my previous posting). Mine has 5 SCSI busses - two of which have another AS4100 hanging off them. Given that the md0: report is well before the 15 second SCSI reset timeout (unless the serial output is buffered), I wonder if it's something about my SCSI config. I haven't tried FreeBSD on this particular machine before. I might try one of the others that isn't part of a cluster. >DS10 works, I have one. No need to try. ES45 will most likely not work >(but it would be interesting to see what happens :) I'll try "borrowing" an ES45 tomorrow morning if no-one's left testing running overnight. -- Peter Jeremy
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040901115403.GA60090>