sys-apps/net-tools-1.60-r12 cant compile against linux-headers-2.6.18
Created attachment 99344 [details, diff] net-tools fddi-fix patch this patch corrects the in_fddi error
The patch fixed the fddi error however I still get other errors on amd64: patching file pub/diag/Makefile patching file pub/diag/Makefile Hunk #1 succeeded at 15 with fuzz 1. patching file pub/diag/ether-wake.c touch .patch make -C pub/diag most make[1]: Entering directory `/var/tmp/portage/sys-apps/net-tools-1.60-r12/work/net-tools-1.60/ethercard-diag/pub/diag' x86_64-pc-linux-gnu-gcc -O -Wall -Wstrict-prototypes -DLIBFLASH -DLIBMII -c -o libflash.o libflash.c x86_64-pc-linux-gnu-gcc -O -Wall -Wstrict-prototypes -DLIBFLASH -DLIBMII -c -o libmii.o libmii.c x86_64-pc-linux-gnu-gcc -O -Wall -Wstrict-prototypes -DLIBFLASH -DLIBMII pci-config.c -o pci-config x86_64-pc-linux-gnu-gcc -O -Wall -Wstrict-prototypes -DLIBFLASH -DLIBMII ether-wake.c -o ether-wake x86_64-pc-linux-gnu-gcc -O -Wall -Wstrict-prototypes -DLIBFLASH -DLIBMII ne2k-pci-diag.c libflash.o -o ne2k-pci-diag ether-wake.c: In function 'main': ether-wake.c:184: warning: pointer targets in initialization differ in signedness x86_64-pc-linux-gnu-gcc -O -Wall -Wstrict-prototypes -DLIBFLASH -DLIBMII alta-diag.c libflash.o libmii.o -o alta-diag x86_64-pc-linux-gnu-gcc -O -Wall -Wstrict-prototypes -DLIBFLASH -DLIBMII eepro100-diag.c libflash.o libmii.o -o eepro100-diag alta-diag.c:89:27: error: asm/unaligned.h: No such file or directory make[1]: *** [alta-diag] Error 1 make[1]: *** Waiting for unfinished jobs.... make[1]: Leaving directory `/var/tmp/portage/sys-apps/net-tools-1.60-r12/work/net-tools-1.60/ethercard-diag/pub/diag' make: *** [most] Error 2 make: Leaving directory `/var/tmp/portage/sys-apps/net-tools-1.60-r12/work/net-tools-1.60/ethercard-diag' !!! ERROR: sys-apps/net-tools-1.60-r12 failed. Call stack: ebuild.sh, line 1568: Called dyn_compile ebuild.sh, line 937: Called src_compile net-tools-1.60-r12.ebuild, line 56: Called die !!! emake ethercard-diag failed !!! If you need support, post the topmost build error, and the call stack if relevant. !!! This ebuild is from an overlay: '/usr/local/portage'
This patch - net-tools_fddi-fix patch also helps my compile under gcc-3.4.6 with kernel 2.4.32-om-migshm. THX! PLS put in portage!
fixed with 2.6.19 headers
Alright, I am confused. I just installed linux-headers-2.6.19 and removed the overlay and I am back to the exact same fddi error. After applying the patch I am back to where I was before. I guess its time to go back to the stable kernel headers. Ok, I just did that emerged linux-headers-2.6.17-r2 and then emerging net-tools-1.60-r12 now works.
*** Bug 156983 has been marked as a duplicate of this bug. ***
for example problem occured on reiserfsprogs by exclude unaligned.h from 2.6.18. https://launchpad.net/distros/ubuntu/+source/reiserfsprogs/+bug/65842 http://cvs.fedora.redhat.com/viewcvs/rpms/reiserfs-utils/devel/header-fix.patch?view=markup resolved defining put_unaligned and get_unaligned by local header.(using fedora's better way)
Created attachment 103447 [details] local unaligned.h
Created attachment 103448 [details] ebuild using unaligned.h and fddi_fix.patch
*** Bug 157564 has been marked as a duplicate of this bug. ***
*** Bug 157994 has been marked as a duplicate of this bug. ***
Not fixed, reopen.
Same against 2.6.19. The problem is that __be16 is used in linux/if_fddi.h, however its definition is not included. The definition previously came from including linux/netdevices.h at the top of linux/if_fddi.h. Re-adding this include deals with the problem. Obviously upstream removed the include of linux/netdevices.h for a reason - so it would be useful to find out why.
Created attachment 104122 [details, diff] Replace the include of linux/netdevices.h in linux/if_fddi.h Patch for linux-headers-2.6.19 (and 2.6.18 I guess). Obviously when I mentioned upstream previously I was talking about the kernel, not nettools.
Found this: http://lists.netfilter.org/pipermail/netfilter-devel/2005-October/021648.html which fixes the same problem just for netfilter, by including <linux/types.h> in the source file that includes <linux/if_fddi.h> I don't know what the rules are for including linux headers; whether it's up to the includer to ensure all dependent headers are included, or whether it's the responsibility of the included file to pull in whatever it needs.
ive already fixed this
*** Bug 162885 has been marked as a duplicate of this bug. ***
*** Bug 182460 has been marked as a duplicate of this bug. ***
