From: ext Felipe Contreras <felipe.contreras@xxxxxxxxx> Subject: Re: [RFC/PATCH 3/3] omap3-iommu: remote registration Date: Mon, 18 May 2009 15:52:59 +0200 > On Mon, May 18, 2009 at 4:35 PM, Hiroshi DOYU <Hiroshi.DOYU@xxxxxxxxx> wrote: > > From: ext Russell King - ARM Linux <linux@xxxxxxxxxxxxxxxx> > > Subject: Re: [RFC/PATCH 3/3] omap3-iommu: remote registration > > Date: Mon, 18 May 2009 14:11:13 +0200 > > > >> On Sat, May 16, 2009 at 01:05:50PM +0300, Felipe Contreras wrote: > >> > This allows devices to be registered only when they are used. The > >> > current dsp-bridge driver for example is not using iommu so registering > >> > the iommu iva2 device would conflict. By allowing remote registration > >> > the dsp-bridge can decide when the iommu iva2 device is registered. > >> > >> I don't think that this is a good idea - what happens if two people > >> call omap_iommu_add() for the same IOMMU device independently? > >> > >> The real problem here seems to be the TI DSP bridge code, and if that's > >> the case why can't we just avoid registering IVA2 if the TI DSP bridge > >> code is enabled. That solves your stated problem without creating > >> additional management issues. > > > > How about the attached patch? I think this is enough. > > That wouldn't work when bridgedriver moves to iommu, how about: > > +#if !defined(CONFIG_MPU_BRIDGE) && !defined(CONFIG_OMAP_DSP_MODULE) > +#if defined(CONFIG_MPU_BRIDGE_IOMMU) Just now I read Russell's proposal. This would be better. Agreed. How about Hari? I think that he's been working on this for a while.... -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
