On Fri, Apr 7, 2017 at 10:41 AM, Kani, Toshimitsu <toshi.kani@xxxxxxx> wrote: > On Thu, 2017-04-06 at 13:59 -0700, Dan Williams wrote: >> Before we rework the "pmem api" to stop abusing __copy_user_nocache() >> for memcpy_to_pmem() we need to fix cases where we may strand dirty >> data in the cpu cache. The problem occurs when copy_from_iter_pmem() >> is used for arbitrary data transfers from userspace. There is no >> guarantee that these transfers, performed by dax_iomap_actor(), will >> have aligned destinations or aligned transfer lengths. Backstop the >> usage __copy_user_nocache() with explicit cache management in these >> unaligned cases. >> >> Yes, copy_from_iter_pmem() is now too big for an inline, but >> addressing that is saved for a later patch that moves the entirety of >> the "pmem api" into the pmem driver directly. > > The change looks good to me. Should we also avoid cache flushing in > the case of size=4B & dest aligned by 4B? Yes, since you fixed the 4B aligned case we should skip cache flushing in that case. I'll send a v2.