On Thu, 2019-07-25 at 19:46 -0700, Dan Williams wrote: > On Wed, Jul 24, 2019 at 2:57 PM Vishal Verma <[email protected]> wrote: > > Add a man page describing the new daxctl-reconfigure-device command. > > > > Cc: Pavel Tatashin <[email protected]> > > Cc: Dave Hansen <[email protected]> > > Cc: Dan Williams <[email protected]> > > Signed-off-by: Vishal Verma <[email protected]> > > --- > > Documentation/daxctl/Makefile.am | 3 +- > > .../daxctl/daxctl-reconfigure-device.txt | 139 ++++++++++++++++++ > > 2 files changed, 141 insertions(+), 1 deletion(-) > > create mode 100644 Documentation/daxctl/daxctl-reconfigure-device.txt > > [..] > > +* Run a process called 'some-service' using numactl to restrict its cpu > > +nodes to '0' and '1', and memory allocations to node 2 (determined using > > +daxctl_dev_get_target_node() or 'daxctl list') > > +---- > > +# daxctl reconfigure-device --mode=system-ram --no-online dax0.0 > > Any reason to use --no-online in this example? Presumably some-service > may not start if node2 has no online memory. > Yep just a copy/paste typo, removing it. > > > +[ > > + { > > + "chardev":"dax0.0", > > + "size":16777216000, > > + "target_node":2, > > + "mode":"system-ram" > > + } > > +] > > + > > +# numactl --cpunodebind=0-1 --membind=2 -- some-service --opt1 --opt2 > > +---- > > + > > +DESCRIPTION > > +----------- > > + > > +Reconfigure the operational mode of a dax device. This can be used to > > convert > > +a regular 'devdax' mode device to the 'system-ram' mode which allows for > > the dax > > s/allows/arranges/
ok. > > > +range to be hot-plugged into the system as regular memory. > > + > > +NOTE: This is a destructive operation. Any data on the dax device *will* be > > +lost. > > + > > +NOTE: Device reconfiguration depends on the dax-bus device model. If > > dax-class is > > +in use (via the dax_pmem_compat driver), the reconfiguration will fail. See > > +linkdaxctl:daxctl-migrate-device-model[1] for more information. > > Let's make sure that do_reconfig() bails with a common error message > for the compat case and quote that message here. You can check that by > comparing the device path 'subsystem' to /sys/class/dax. I.e. > > # ls -l /dev/dax0.0 > crw------- 1 root root 253, 4 Jul 25 12:22 /dev/dax0.0 > > # readlink -f /sys/dev/char/253\:4/subsystem > /sys/class/dax > > ...it will be /sys/bus/dax otherwise. Hm,. currently the failure is just: libdaxctl: daxctl_dev_enable: dax3.0: failed to enable Should we check the subsystem as above programatically, and print a better error in daxctl-reconfigure-device? > > > + > > +OPTIONS > > +------- > > +-r:: > > +--region=:: > > + Restrict the operation to devices belonging to the specified > > region(s). > > + A device-dax region is a contiguous range of memory that hosts one > > or > > + more /dev/daxX.Y devices, where X is the region id and Y is the > > device > > + instance id. > > + > > +-m:: > > +--mode=:: > > + Specify the mode to which the dax device(s) should be reconfigured. > > + - "system-ram": hotplug the device into system memory. > > + > > + - "devdax": switch to the normal "device dax" mode. This requires > > the > > + kernel to support hot-unplugging 'kmem' based memory. If this is > > not > > + available, a reboot is the only way to switch back to 'devdax' > > mode. > > + > > +-N:: > > +--no-online:: > > + By default, memory sections provided by system-ram devices will be > > + brought online automatically and immediately with the > > 'online_movable' > > + policy. Use this option to disable the automatic onlining behavior. > > Probably need to mention that the system might online the memory even > if this is specified, or for extra credit, coordinate with that > auto-online facility to arrange for it to be skipped. Good point, added a note regarding this. > > > + > > +-O:: > > +--attempt-offline:: > > + When converting from "system-ram" mode to "devdax", it is expected > > + that all the memory sections are first made offline. By default, > > + daxctl won't touch online memory. However with this option, attempt > > + to offline the memory on the NUMA node associated with the dax > > device > > + before converting it back to "devdax" mode. > > As mentioned in patch 7, this sounds like --force to me. Done. _______________________________________________ Linux-nvdimm mailing list [email protected] https://lists.01.org/mailman/listinfo/linux-nvdimm
