On Tue, 17 Sep 2019, Mike Snitzer wrote: > On Tue, Sep 17 2019 at 9:56am -0400, > John Dorminy <jdorminy@xxxxxxxxxx> wrote: > > > I'm confused here: > > > >...and then it fails on activation because DM table load detects old (or > > missing) dm-crypt feature. > > >(There was no way to get dm target version before table load if module is > > not loaded.) > > >And I tried to avoid modprobe calls from libcryptsetup. > > > > I'm not understanding how this could work. Let's say there's a module > > providing target 'splice' which is currently unloaded. Then `dmsetup > > target-version splice` is unaware of the splice target, and thus reports > > nothing. So in order to get the proper version number, we do have to do a > > modprobe first. > > But, if the module providing splice *is* already loaded, `dmsetup targets` > > will report splice's version number already, so `dmsetup target-version > > splice` is a convenience instead of parsing `dmsetup targets` output? > > dm_get_target_type() loads the requested module. No userspace modprobe > needed. The reason why Milan wanted it is - that it is bad practice to call external processes from libraries. He doesn't want to call modprobe from libcryptsetup. So I created this patch that calls modprobe from the kernel ioctl. Mikulas -- dm-devel mailing list dm-devel@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/dm-devel
