On 2019/06/27 17:26, Christoph Hellwig wrote: > On Thu, Jun 27, 2019 at 08:14:56AM +0000, Damien Le Moal wrote: >> which I guessed is for the architectures that do not need the flush/invalidate >> vmap functions. I copied. Is there a better way ? The point was to avoid doing >> the loop on the bvec for the range length on architectures that have an empty >> definition of invalidate_kernel_vmap_range(). > > No, looks like what you did is right. I blame my lack of attention > on the heat wave here and the resulting lack of sleep.. No problem ! At least we checked twice now ! -- Damien Le Moal Western Digital Research
