Date: Sun, 12 Mar 2000 11:21:33 +0100 From: Bernd Walter <ticso@cicely.de> To: Doug Rabson <dfr@nlsystems.com> Cc: Bernd Walter <ticso@cicely.de>, freebsd-alpha@freebsd.org Subject: Re: 64bit address space on alpha Message-ID: <20000312112132.A75380@cicely8.cicely.de> In-Reply-To: <Pine.BSF.4.10.9907181540560.58023-100000@salmon.nlsystems.com> References: <19990718123721.A88813@cicely8.cicely.de> <Pine.BSF.4.10.9907181540560.58023-100000@salmon.nlsystems.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Jul 18, 1999 at 03:41:12PM +0100, Doug Rabson wrote: > On Sun, 18 Jul 1999, Bernd Walter wrote: > > > On Sun, Jul 18, 1999 at 09:27:14AM +0100, Doug Rabson wrote: > > > On Sat, 17 Jul 1999, Bernd Walter wrote: > > > > > > > I would like to mmap a large file into a single process address space. > > > > Is it save to increase MAXDSIZ to more than 4G? > > > > If not is there already work in progress? > > > > > > It should be safe to increase MAXDSIZ. The maximum user address space is > > > about 4T on FreeBSD/alpha. I have programs which mmap about 12G of device > > > addresses. > > I tried to compile a kernel with MAXDSIZ set to 128G and got several overflow > > compiler warnings: > > > > Are these realy save to ignore? > > I'll check. Just to get you informed about this point: It was completely my fault and it took me realy a half year to recheck and find this stupid thing :( I put (128*1024*1024*1024) in config following the example in LINT. The point is that gcc uses an int to calculate the number behind. Now that I'm using a constant everything is fine... -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de 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?20000312112132.A75380>