<div dir="ltr">This set is now merged.<div>Thanks.</div></div><div class="gmail_extra"> <div class="gmail_quote">On Tue, Jul 18, 2017 at 9:29 AM, Martin Wilck <<a href="mailto:mwilck@suse.com" target="_blank">mwilck@suse.com</a>> wrote: <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Current code fails to set up multipath maps for NVME devices in a Linux target/Linux host combination. This series enables at least basic operation. Patch 1/4 fixes a crash that happens if over-long WWIDs are encountered, and is not specific to NVME as such. Patch 2/4 drops the broken test uevent_can_discard_by_devpath(). Patch 3/4 compensates for the additional event processing required by 2/4. Patch 4/4 mangles overlong "nvme.*" WWIDs to make them usable for multipath (related discussion in [1]; WWID_SIZE can't be simply increased because it has to match device mapper's DM_NAME_LEN). Changes wrt v1: 1/4: assure 0-termination by using strlcpy() (Bart van Assche) 4/4: drop extra length test (Ben Marzinski) I kept the v1 Reviewed-by: and Acked-by tags because the changes are minor, I hope that's ok for the reviewers. [1] <a href="http://lists.infradead.org/pipermail/linux-nvme/2017-July/011960.html" rel="noreferrer" target="_blank">http://lists.infradead.org/pipermail/linux-nvme/2017-July/011960.html</a> Martin Wilck (4): libmultipath: get_udev_uid: make sure pp->wwid is 0-terminated libmultipath: drop uevent_can_discard_by_devpath libmultipath: only listen for uevents with DEVTYPE=disk libmultipath: fix over-long NVME WWIDs libmultipath/discovery.c | 85 +++++++++++++++++++++++++++++++++++++++++++++--- libmultipath/uevent.c | 27 +-------------- 2 files changed, 82 insertions(+), 30 deletions(-) -- 2.13.2 </blockquote></div> </div>