On Thu, Apr 23, 2020 at 09:46:02AM +0200, Alexander Leidinger wrote: > > Quoting Konstantin Belousov <kostikbel_at_gmail.com> (from Thu, 23 Apr 2020 > 10:04:12 +0300): > > > On Thu, Apr 23, 2020 at 08:30:08AM +0200, Alexander Leidinger wrote: > > > Quoting Konstantin Belousov <kostikbel_at_gmail.com> (from Thu, 23 Apr 2020 > > > 09:19:08 +0300): > > > > > > > On Thu, Apr 23, 2020 at 08:08:09AM +0200, Alexander Leidinger wrote: > > > > > Hi, > > > > > > > > > > link_elf_obj: symbol xdr_free undefined > > > > > linker_load_file: /boot/kernel/nfslockd.ko - unsupported file type > > > > > KLD nfsd.ko: depends on nfslockd - not available or version mismatch > > > > > linker_load_file: /boot/kernel/nfsd.ko - unsupported file type > > > > > > > > > > The xdr kernel module is loaded. > > > > > > > > The following should help. > > > > > > > /* So that loader and kldload(2) can find us, wherever we are.. */ > > > > +MODULE_DEPEND(nfslockd, xdr, 1, 1, 1); > > > > MODULE_DEPEND(nfslockd, krpc, 1, 1, 1); > > > > > > I thought MODULE_DEPEND is only a way to tell the loader / kldload that the > > > other module is to be loaded in case it is not in the kernel / loaded > > > already. Is this only a part of what it does and / or did this change and > > > now it is also needed so that a kernel module is able to resolve the > > > symbols? In my case the module is loaded already when I try to load the > > > nfslockd module. > > > > Did you tried the patch ? > > I was trying it after asking the question. I confirm it works. > > > You already said that the module was loaded. > > > > MODULE_DEPEND() handles symbol namespaces. > > Since when is this the case (rough figure would be enough if someone knows > it without looking it up)? It is definitely so from the 5.x times, but I highly suspect that this was the feature of 'new' modules from the beginning.Received on Thu Apr 23 2020 - 07:22:59 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:41:23 UTC