![]() | |||||||||||||
Messages in this thread |
On Tue, Jun 07, 2005 at 10:55:27PM -0700, Andrew Grover wrote: > On 6/7/05, Jeff Garzik <jgarzik@pobox.com> wrote: > > If the driver has to _undo_ something that it did not do, that's pretty > > lame. Non-orthogonal. > > I would think the number of MSI and MSI-X capable devices is going to > explode over the next five years. I'm not sure it's right to make all > these device's drivers pay a complexity cost because some of the first > attempted MSI implementations were buggy. Exactly. Couldnt agree more. > > > Also, it looks like all the PCI MSI drivers need touching for this > > scheme -- which defeats the original intention. At this rate, the best > > API is the one we've already got. > > For now...but I'm bringing this up again in five years!! *sets egg timer* I think we should have the right (default MSI) API by default. If we wait 5 years we end up with lots of mess in the drivers. -Andi - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ | ||||||||||||
| Last update: 2005-06-08 13:41 [from the cache] ©2003-2008 | |||||||||||||