Here is some email that IBM sent to me which (ostensibly) explains some of the trouble I and others have been encountering: "The problem is that Redhat seems to have a patch (linux-2.4.8-tasklets.patch) applied to their Kernel which changes the function tasklet_hi_schedule to __ tasklet_hi_schedule. Our Kernel modules are built against a vanilla 2.4.7 which uses tasklet_hi_schedule and therefore cannot be used with your Redhat Kernel. I have contacted Redhat and told them the problem." In Mark's last post to me, he said that he found no problems with the latest kernel image, so perhaps they have already resolved this problem. Of course, I might just wait for the next version of the release to be out, as it should be soon. From:"Post, Mark K" <mark.post@eds.com> on 09/18/2001 02:17 PM AST Sent by: "Post, Mark K" <mark.post@eds.com> To: "'Michael K Lambert'" <mlambe7@lsu.edu>, redhat-s390-list@redhat.com cc: Subject: RE: [Redhat-s390-list] (no subject) Michael, No bells, per se, but it does show that the symbol is being exported by the kernel. I don't have an OSA card on any of my systems, but I just downloaded the latest initrd and kernel images from ftp.redhat.de for another reason. If I copy the lcs-2.4.7-s390-1.o module to /lib/lcs.o, and do an 'insmod -f /lib/lcs.o' I don't get any messages about missing symbols. I do get the "no lcs capable cards found' error of course. You might want to try the latest kernel/initrd pair with that lcs module. Mark Post -----Original Message----- From: Michael K Lambert [mailto:mlambe7@lsu.edu] Sent: Tuesday, September 18, 2001 1:38 PM To: redhat-s390-list@redhat.com Subject: [Redhat-s390-list] (no subject) Mark, The results of cat/proc/ksyms|grep tasklet_hi_schedule was: 000257d4 __tasklet_hi_schedule_Rsmp_44a7ed12 Does that ring any bells for you? Michael Lambert
