[PATCH v4 1/2] powerpc/pseries/iommu: Share the per-cpu TCE page with the hypervisor.

Ram Pai linuxram at us.ibm.com
Thu Dec 5 07:27:51 AEDT 2019


On Wed, Dec 04, 2019 at 03:26:50PM -0300, Leonardo Bras wrote:
> On Sun, 2019-12-01 at 22:45 -0800, Ram Pai wrote:
> > @@ -206,8 +224,7 @@ static int tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum,
> >          * from iommu_alloc{,_sg}()
> >          */
> >         if (!tcep) {
> > -               tcep = (__be64 *)__get_free_page(GFP_ATOMIC);
> > -               /* If allocation fails, fall back to the loop implementation */
> > +               tcep = alloc_tce_page();
> >                 if (!tcep) {
> >                         local_irq_restore(flags);
> >                         return tce_build_pSeriesLP(tbl, tcenum, npages, uaddr,
> 
> The comment about failing allocation was removed, but I see no chage of
> behaviour here. 
> 
> Can you please explain what/where it changes? 

You observed it right. The comment should stay put.  Will have it fixed
in my next version.

Thanks,
RP



More information about the Linuxppc-dev mailing list