Date: Tue, 24 Aug 1999 10:27:09 -0400 (EDT) From: Andrew Gallatin <gallatin@cs.duke.edu> To: Jeff Wheat <jeff@cetlink.net> Cc: Doug Rabson <dfr@nlsystems.com>, freebsd-alpha@FreeBSD.ORG Subject: Re: 1999-08-08 snapshot problem Message-ID: <14274.42890.841875.678443@grasshopper.cs.duke.edu> In-Reply-To: <XFMail.990824095327.jeff@cetlink.net> References: <Pine.BSF.4.10.9908240939280.72739-100000@salmon.nlsystems.com> <XFMail.990824095327.jeff@cetlink.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Jeff Wheat writes: > > tom-n-jerry# pstat -s > Device 1K-blocks Used Avail Capacity Type > /dev/od0b 130816 0 130816 0% Interleaved <...> > c++ -DMYSQL_SERVER -DDEFAULT_MYSQL_HOME="\"/usr/local\"" - > DDATADIR="\"/usr/local/var\"" -DSHAREDIR="\"/usr/local/share/m > ysql\"" -DHAVE_CONFIG_H -I./../include - > I./../regex -I. -I../include -I.. -I. -O -fno-implicit- > templates -DDBUG_OFF -c sql_yacc.cc > Aug 24 09:38:39 tom-n-jerry /kernel: swap_pager: out of swap space > Aug 24 09:38:39 tom-n-jerry /kernel: swap_pager: out of swap space > Aug 24 09:38:40 tom-n-jerry /kernel: swap_pager_getswapspace: failed > Aug 24 09:38:40 tom-n-jerry /kernel: swap_pager_getswapspace: failed > Aug 24 09:39:02 tom-n-jerry /kernel: pid 6183 (cc1plus), uid 0, was killed: out > of swap space > Aug 24 09:39:02 tom-n-jerry /kernel: pid 6183 (cc1plus), uid 0, was killed: out > of swap space <...> > It is running out of swap for some reason. This same code compiled without a > problem under 3.2-RELEASE. > > Regards, > Jeff Short answer: You need more swap space. A good rule of thumb is 3-4 times the size of physical memory. More helpful answer: There's some sort of funkyness in the egcs c++ optimizer which makes it take LOTS of space. I just built mysql for fun & watched the size of cc1plus get up over 280MB when building sql_yacc.cc. It took over a minute to compile on a 21264 with 640MB of ram. I've seen this sort of thing on x86's as well. Using -O0 cuts the size down to "only" 25MB or so, and cuts the compilation time down to 4 seconds. I think this problem is fixed in newer versions of gcc (gcc-2.95.1) which is soon to become the system compiler. Drew ------------------------------------------------------------------------------ Andrew Gallatin, Sr Systems Programmer http://www.cs.duke.edu/~gallatin Duke University Email: gallatin@cs.duke.edu Department of Computer Science Phone: (919) 660-6590 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?14274.42890.841875.678443>