Hi On Fri, Oct 22, 2010 at 9:49 PM, Christian Stroetmann <stroetmann@xxxxxxxxxxxxx> wrote: > Thanks for further informations. > > On the 22.10.2010 17:46, Viji V Nair wrote: >> >> Hi >> >> On Fri, Oct 22, 2010 at 6:23 PM, Christian Stroetmann >> <stroetmann@xxxxxxxxxxxxx> wrote: >>> >>> Hola; >>> >>> On the 22.10.2010 12:02, Viji V Nair wrote: >>>> >>>> Hi, >>>> >>>> I have been trying to build reiser4 as a module, but the module which >>>> is getting compiled having the "symbol" issue. >>>> >>>> <snip> >>>> reiser4: Unknown symbol lzo1x_1_compress >>>> reiser4: Unknown symbol might_fault >>>> reiser4: Unknown symbol mutex_lock_nested >>>> reiser4: Unknown symbol zlib_deflate >>>> reiser4: Unknown symbol writeback_inodes_wbc >>>> </snip> >>> >>> You mentioned this problem in the thread [1] as well. >>> Check your pathes. >> >> Nobody replied to that thread :). I am planning to push reiser4 in the >> fedora official repository. >> I have checked the paths, looks fine. Even I tried with some other >> modules (cifs) and those are getting compiled and loads with out any >> issue. > > This is surprising. Maybe it is in the scripts of the compilation toolchain > (configure, make, and so on). >>>> >>>> I could see from the reiser4 2.6.34 patch that it is adding >>>> "EXPORT_SYMBOL" to few files in fs& mm directories of the kernel >>>> source tree. Can we have all these symbols defined in the core source >>>> tree of resier4 also? Do we have a patch for this already? >>> >>> For what? See above. > > I still believe that they belong to the files in the fs and mm directories > of the kernel source tree. > > Also somehow fuzzy, the fault messages you have sent are referencing > existing items that are contained in the unpatched/original libraries. Are > the missing symbols related with the ones that are included in the R4 2.6.34 > patch? > > So, maybe you should take a second and wait for the comments by the > maintainer about the "EXPORT_SYMBOL" point. >>> >>> Btw.: To be honest, I find it a little bit off-topic, because it's about >>> compiling a kernel module and not a R4 issue. >> >> This was working fine untill 2.6.10 patch. You can find the same issue >> in Open Suse/Ubuntu/Mandriva mailing lists. > > Ah, okay. Good to know this. >> >> The missing symbols need to exported in the module level, I think. If >> I compile the kernel completely everything works fine, but that is not >> I want. I dont want to disturb the stock kernel. > > Okay. > I still guess the problem is in the scripts that compile the module, if you > say that compiling kernel modules with other file systems and the complete > kernel with R4 support works correctly. Compare the scipts that are used to > compile the complete kernel with R4 support with the scripts of the R4 > kernel module compilation. There must be the searched difference. You are right, I am digging into the fedora kernel source tree, some scripts are exporting all these symbols globally on a complete compilation of the kernel. Need to figure out which one, anyway, for resier4 we should have this symbol exported separately to avoid the dependency with the above scripts. >>>> >>>> Thanks >>>> Viji >>> >>> [1] Compilation Issue - Reiser4 >>> http://www.spinics.net/lists/reiserfs-devel/msg02508.html >>> >>> Have fun >>> Christian O>-< *<:o) _(D)>-< >>> >> Thank you so much for your support >> >> Viji > > Good luck > Christian O>-< *<:o) _(D)>-< > Thanks Viji -- To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
