On Wed, Dec 9, 2015 at 6:37 PM, Dan Williams <dan.j.williams@xxxxxxxxx> wrote: > Changes since v1: [1] > > 1/ Rebase on latest -next (20151209). Fixed up PHYSICAL_PAGE_MASK, but > no other major collisions. > > 2/ Decreased the transfer size in "dax: increase granularity of > dax_clear_blocks() operations" to get the max latency to reschedule > under 1ms and average latency under 150us. (Andrew) > > 3/ Add cc's for x86 maintainers > > 4/ Add Tested-by: Logan Gunthorpe > > This update, as before, passes the ndctl [2] and nvml [3] test suites. > > [1]: https://lists.01.org/pipermail/linux-nvdimm/2015-December/003213.html > [2]: https://github.com/pmem/ndctl > [3]: https://github.com/pmem/nvml > > A git tree of this set is available here: > > git://git.kernel.org/pub/scm/linux/kernel/git/djbw/nvdimm libnvdimm-pending > Andrew, I went ahead and rebased the tree on next-20151211, and refreshed 3 patches with the following changes (sent as replies to avoid spamming folks with the other 22 patches that did not change): [-mm PATCH v3 04/25] dax: fix lifetime of in-kernel dax mappings with dax_map_atomic() Updated to the new definition of blk_queue_enter() from commit 6f3b0e8bcf3c "blk-mq: add a flags parameter to blk_mq_alloc_request" [-mm PATCH v3 09/25] mm, dax, pmem: introduce pfn_t Moved phys_to_pfn_t() out of line to enable some new unit tests. [-mm PATCH v3 10/25] mm: introduce find_dev_pagemap() Just a reflow to pick up the context differences from the changes in [-mm PATCH v3 09/25]. These updates (plus next-20151211) continue to pass all my tests and have been pushed out to the 'libnvdimm-pending' branch. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>