Messages in this thread | | | From | (Keith Schneider) | Subject | Re: 2.0.35 nls problem (make dep) SOLUTION | Date | 19 Jul 1998 05:50:58 GMT |
| |
keith@ugcs.caltech.edu (Keith Schneider) writes:
>I just got the released 2.0.35 kernel. I may be an idiot, but I have >compiled a number of kernels before without encountering the current >error. When I do 'make dep' it ends in the following error sequence: >(snipped from the end of 'make dep >& dep_log')
>----- >make[1]: Entering directory `/usr/src/linux/fs' >gcc -D__KERNEL__ -I/usr/src/linux/include -Wall -Wstrict-prototypes -O2 >-fomit-frame-pointer -fno-strength-reduce -pipe -m486 -malign-loops=2 >-malign-jumps=2 -malign-functions=2 -DCPU=586 -E -D__GENKSYMS__ >nls_cp437.c | /sbin/genksyms /usr/src/linux/include/linux/modules >make[1]: *** [/usr/src/linux/include/linux/modules/nls_cp437.ver] Error >139 >make[1]: Leaving directory `/usr/src/linux/fs' >make: *** [dep-files] Error 2 >-----
Hello, I figured this out after snooping around a bit. For some reason, I had CONFIG_MODVERSIONS set in my .config file. I've never had this set before so I didn't run into this problem. The help for this option mentions that the compile will fail if you don't have genksyms. Well, I had genksyms, but the compile still failed.
For some reason, genksyms was not producing an output file for the above (quoted) gcc pipe. Perhaps there are no symbols (or whatever--I don't know much about all of this as you can tell) in those codepage sources. When I simply fabricated the necessary files, e.g. 'touch /usr/src/linux/include/linux/modules/nls_cp437.ver', everything seemed to work fine. (I only needed to do this for the few codepage things I had selected plus one net file of some sort.)
The best option is to simply turn off the CONFIG_MODVERSIONS option, and then things also work fine. I don't know if there is a bug here or not (shouldn't the make process be able to handle the situation when genksyms doesn't produce a .ver file? or is the problem with genksyms?), but I thought I would post my finding here so that someone might review it and see.
keith
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
| |