From: Daniel P. <phi...@ar...> - 2002-07-11 22:04:18
|
Hi, I just resubscribed with my new mailing address, my old isp having taken a sudden exit out of the isp business. I'll be on the conference call tomorrow, ready to talk about config_nonlinear. I hope to post the current patch set before then. This include uml and mips32 architectures, though it's a practical certainty nobody, including me, will be able to test the mips32 version since the chip isn't in production yet. To properly support this thing, we need an x86 implementation. I can afford the time to rough that out to the point where it compiles, but I hope somebody else will be able to take it the rest of the way. I can imagine two flavors of the x86 version: one for numa, to divide up the zone_normal memory between nodes, and another that's basically a toy, but has the advantage of running on hardware that people actually have. -- Daniel |
From: Martin J. B. <Mar...@us...> - 2002-07-11 22:11:53
|
> To properly support > this thing, we need an x86 implementation. I can afford the time > to rough that out to the point where it compiles, but I hope > somebody else will be able to take it the rest of the way. I can > imagine two flavors of the x86 version: one for numa, to divide up > the zone_normal memory between nodes, and another that's basically > a toy, but has the advantage of running on hardware that people > actually have. I'll do this for x86 - if you have time to rough it out first, so much the better. I've promised to get the ZONE_NORMAL spread around in the next month anyway .... M. |