Date: Mon, 26 Jul 2010 10:43:59 -0400 From: John Baldwin <jhb@freebsd.org> To: freebsd-current@freebsd.org, sbruno@freebsd.org Subject: Re: [patch and review please] 64 CPU Support Message-ID: <201007261044.00126.jhb@freebsd.org> In-Reply-To: <1280090933.14823.2.camel@localhost.localdomain> References: <20100722213836.GH15227@martini.nu> <7573B69C-3C37-449A-A27F-5B0B2ED84757@mac.com> <1280090933.14823.2.camel@localhost.localdomain>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sunday, July 25, 2010 4:48:53 pm Sean Bruno wrote: > Reposting from -stable. > > Kind of a large patch, but in order to make an omlette, you need to > break a few servers. > > This is a diff against -CURRENT, not stable-8 as I didn't get a chance > to test it. It is directly based off of changes that peter@ made to the > Yahoo FreeBSD 7 tree. > > I have compile and boot tested this on my local machines, but I don't > have 64 CPU machines to test upon. I think IPI_AST in the first hunk should be using ipi_cpu(). I would perhaps tackle ipi_cpu() as a first step: introduce ipi_cpu() on both i386 and amd64 (it should be ok to add a real version for i386 rather than the current macro, it should be a copy of the amd64 code). Other folks can help with other architectures. ipi_selected() should generally be a good clue as to how to implement ipi_cpu(). -- John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201007261044.00126.jhb>