Messages in this thread | | | Date | Sat, 22 Dec 2012 08:21:53 -0500 | From | Konrad Rzeszutek Wilk <> | Subject | Re: [PATCH v7 24/27] x86: Add swiotlb force off support |
| |
On Fri, Dec 21, 2012 at 07:23:03PM -0800, Eric W. Biederman wrote: > Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> writes: > > > On Fri, Dec 21, 2012 at 06:42:47PM -0800, Eric W. Biederman wrote: > >> Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> writes: > >> > >> > On Mon, Dec 17, 2012 at 11:15:56PM -0800, Yinghai Lu wrote: > >> >> So use could disable swiotlb from command line, even swiotlb support > >> >> is compiled in. Just like we have intel_iommu=on and intel_iommu=off. > >> > > >> > You really need to spell out why this is useful. > >> > >> YH why can't we safely autodetect that the swiotlb is unusable when > >> there is no memory below 4G free? > > > > I am not sure what 'YH' stands for (Yeah?). > > Yinghai Lu's nickname. > > > However we could turn SWIOTLB off altogether if it cannot allocate > > _some_ memory. It could try first 64MB, then 32MB, lastly 16MB. And > > if all that fails - print a nice warning and continue on. > > > > Later in the late initialization phase, when pci_swiotlb_late_init > > is called - it can then figure out whether 'iommu' has been set > > and it iself was never able to allocate. At that point it can try > > the dynamic allocation (swiotlb_late_init_with_default_size) > > ... and if that fails give up and panic. > > As far as I can tell panics should be avoided unless there is something > that actually needs an iommu, and the swiommu is the only option, and > the swiommu can not fulfill that request.
Right. > > In this case YH has been working on the case of loading a kernel > completely above 4G, and apparently he has also been testing the case of > running a kernel with no memory below 4G. > > Eric >
| |