Date: Thu, 16 Mar 2006 09:35:28 +1100 From: Andrew Reilly <andrew-freebsd@areilly.bpc-users.org> To: JoaoBR <joao@matik.com.br> Cc: freebsd-amd64@freebsd.org, Kris Kennaway <kris@obsecurity.org> Subject: Re: amd64 slower than i386 on identical AMD 64 system? Message-ID: <20060315223528.GA40682@gurney.reilly.home> In-Reply-To: <200603150601.26135.joao@matik.com.br> References: <200603140740.38388.joao@matik.com.br> <200603141914.54442.joao@matik.com.br> <20060315022800.GA47353@xor.obsecurity.org> <200603150601.26135.joao@matik.com.br>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Mar 15, 2006 at 06:01:25AM -0300, JoaoBR wrote: > SMP with single dual-core processors on standard MBs are sensitive (crashing > easily or time-outs) with non polling NICs Just to add another data point: My 1G RAM, AMD64-X2 box is running a mostly-SMP kernel, but with kernel.smp.disabled="1" in boot/loader.conf (because that's the only way to get the 4front-tech sound card driver to load; otherwise it is unable to allocate a necessary contiguous chunk of memory). The on-board nForce MCP9 network adaptor has never worked properly (time-outs and failure to detect carrier, rather than crashes), but I was having good success with a dc (Intel 21143) up until last weekend's cvsup/rebuild, when that stopped working too: unable to detect carrier, which I guess is an interrupt-related activity. The dc0 driver is working happily, now, with DEVICE_POLLING enabled, and configured on the interface. -- Andrew
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060315223528.GA40682>