> -----Ursprüngliche Nachricht----- > Von: jan.kis...@web.de [mailto:jan.kis...@web.de] > Gesendet: Montag, 23. November 2009 09:50 > An: Debes, Thomas RAEK MRA > Cc: rtnet-users@lists.sourceforge.net > Betreff: Re: AW: [RTnet-users] unresolved symbol netdev_priv when > loading rtcap.o > > thomas.de...@manroland.com wrote: > >> -----Ursprüngliche Nachricht----- > >> Von: jan.kis...@web.de [mailto:jan.kis...@web.de] > >> Gesendet: Freitag, 20. November 2009 18:41 > >> An: Debes, Thomas RAEK MRA > >> Cc: rtnet-users@lists.sourceforge.net > >> Betreff: Re: [RTnet-users] unresolved symbol netdev_priv when > loading > >> rtcap.o > >> > >> thomas.de...@manroland.com wrote: > >>> Hello, > >>> > >>> recently I tried to build the latest RTnet release for our kernel > >>> 2.4.25 based MPC52000. Everything went fine until I loaded the > rtcap > >> module. > >>> This led to "unresolved symbol netdev_priv". Looking through the > git > >>> commit messages I found that this function was introduced with > >>> kernel 2.6.29. Is there a chance to make it work with kernel 2.4 > again? > >> If that was the only issue: please check latest git. > >> > > > > I did that but I still got an unresolved symbol. After including > rtnet_port.h in rtcap.c compiling was fine but insmoding led to a > segfault. Did I miss something? > > Inclusion fixed in git, thanks for pointing out. > > > > > *** RTnet 0.9.11 - built on Nov 20 2009 22:28:51 *** > > > > RTnet: initialising real-time networking > > RTnet: registered rteth0 > > RTproxy attached to rteth0 > > rtnetproxy installed as "rtproxy" > > RTcap: real-time capturing interface > > Oops: kernel access of bad area, sig: 11 > > NIP: D12C8DA4 XER: 20000000 LR: D12C8D90 SP: CD92DE40 REGS: cd92dd90 > TRAP: 0300 Not tainted > > MSR: 00009032 EE: 1 PR: 0 FP: 0 ME: 1 IR/DR: 11 > > DAR: 00000000, DSISR: 22000000 > > TASK = cd92c000[166] 'insmod' Last syscall: 128 last math cd92c000 > > last altivec 00000000 > > GPR00: D12C8D90 CD92DE40 CD92C000 CFBF6200 CDB0300A CFBF6206 CFBF6279 > > CFD42360 > > GPR08: CFBF6000 00000000 00000000 D12C88F8 84428428 10047088 00000000 > > 00000000 > > GPR16: 10040000 00000001 00000000 00000000 D12C994C D12D0000 D12D0000 > > 00000000 > > GPR24: CDB03004 0000006C 00000001 CFBF6200 D12C99B0 CDB03000 00000000 > > CDB030AC Call backtrace: > > D12C8D90 C0018340 C0005AFC 10088480 10003DB4 10004EC4 1000942C > > 10009654 0FEBCFC8 00000000 > > But this oops is not very helpful. No backtrace, not even a proper > resolution of the IP. Can you enable debug symbols for your kernel and > run this through ksymoops? >
Okay, I was not able to get debug enabled kernel, but I could track down the oops to line 467 in rtcap_init: *(struct rtnet_device **)netdev_priv(dev) = rtdev; By instrumenting it with some printk's it turned out that "dev->priv" is NULL. That seems to be a valid case because "rtdev" is assigned. Finally I manually replaced this line with the code before switching to netdev_priv (dev->priv = rtdev) and it worked until the next oops occurred at the following "register_netdev(dev)" call. Obviously there is something wrong with the dev structure. I will dig into this... By the way rtmac_vnic.c needs rtnet_port.h included too. Thomas -------------------------------------------------------- manroland AG Vorsitzender des Aufsichtsrates: Hanno C. Fiedler Vorstand: Gerd Finkbeiner (Vorsitzender), Dr. Ingo Koch, Dr. Markus Rall, Paul Steidle Sitz der Gesellschaft: Offenbach am Main, Registergericht: Amtsgericht Offenbach HRB-Nr. 42592 USt-Ident-Nr. DE 250200933 ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july _______________________________________________ RTnet-users mailing list RTnet-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/rtnet-users
