From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <5de67ff6c489f32ca98e2a89d9895f36@quanstro.net> From: erik quanstrom Date: Fri, 6 Mar 2009 19:16:26 -0500 To: 9fans@9fans.net In-Reply-To: <13426df10903061605p3f96f3ej33d881810a0e8728@mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit Subject: Re: [9fans] Google Summer of code 2009 Topicbox-Message-UUID: b52054b2-ead4-11e9-9d60-3106f5b1d025 > > what's the pressure point here? > > > > The number of mainboards that have incorrect PIR, MP, tables that are > correct in ACPI. The fact that so many functions, for correct > operation, need ACPI. > > It's a mess. ACPI sucks. But it's what we have to work with. this is just my perspective on the matter, but i find the mp tables often get fixed, and when they don't the results are mostly annoying. my core i7 motherboard, for example, doesn't see it's ht processors. i can live with that. i still have 4. (the biggest sticking point with the mp tables is the interrupts. this can also be fixed by using msi interrupts. i think that would offer a bigger bang/buck factor. the performance of 8259 or even mp interrupts is pretty poor.) on the other hand, nvidia graphics drive me bats. it would make a bigger difference to me to have good amd or intel gma graphics support. just my own perspective. maybe i'm out of touch and have just been lucky picking motherboards. that's not to say acpi isn't a good thing to get done. it's just that personally i'd rank it lower than a few other things. ymmv. - erik