Just compiled the new 2.6.18 kernel. It works great, minus some DMA problems I found a workaround for...
Just compiled the new 2.6.18 kernel. It works great, minus some DMA problems I found a workaround for...
My Blog: http://rannb.blogspot.com
Sure, but it pretty complicated. Here goes:
When I installed my kernel, and rebooted, the bootup wouldn't go past 'Loading Hardware Drivers' (I think that's what it's called, the 6th item anyway). After having messed around a bit, I found that pressing ctrl-alt-F1, then ctrl-alt-F8 got me to the text version of the boot up screen. I noticed that my hda was timing out on DMA.
So, I rebooted into my old kernel, dropped to a root shell (ctrl-alt F1 at the graphical login screen, login as root. Or you could do all this with sudo), and started messing around with hdparm and /etc/hdparm.conf.
Eventually, I found that running, then editing /etc/hdparm.conf to includeCode:hdparm -d1 -X udma2and rebooting allowed me access to my new kernel. However, udma2 was still not enabled. So, now I need to typeCode:/dev/hda { mult_sect_io = 16 write_cache = off dma = on transfer_mode=udma2 }every time I boot up. Actually, I need to type it twice. The first time it tells me that the device isn't ready for the command.Code:sudo hdparm -d1 -X udma2
Weird, eh? If anyone has a solution, I'd appreciate it.
One more thing: if anyone has MadWifi 0.9.1 compiled from source (which you have to if you're using it with a custom kernel), note that it doesn't work with 2.6.18. You need to update to 0.9.2.
My Blog: http://rannb.blogspot.com
Worked great except a linux-restricted-modules folder wasn't generated for the new kernel. Some of my hardware needs restricted modules. How do I regenerate the linux-restricted-modles folder when upgrading the kernel?
Much thanks,
GaryH
This tutorial worked perfectly for me with the 2.6.18.1 kernel. After compiling, restarted. X failed to start, went in, ran script to recompile nvidia drivers and ndiswrapper load them into kernel, and restarted gdm and came up perfectly.
Thanks for the tutorial.
Help: Can anybody explain me how to get rid of those messages?
Kernel 2.6.17 from Kernel.org and Beyond4 = 2.6.17.13 or so.
System Edgy EFT-
I need to take 2.6.17 because of some undervolt patches and beyond sources which are not yet available for 2.6.18
Code:AR arch/i386/lib/lib.a GEN .version CHK include/linux/compile.h UPD include/linux/compile.h CC init/version.o LD init/built-in.o LD .tmp_vmlinux1 init/built-in.o: In function `try_name': do_mounts.c:(.text+0x604): undefined reference to `__stack_chk_fail' init/built-in.o: In function `name_to_dev_t': (.text+0x909): undefined reference to `__stack_chk_fail' init/built-in.o: In function `mount_block_root': (.init.text+0xbd3): undefined reference to `__stack_chk_fail' init/built-in.o: In function `do_header': initramfs.c:(.init.text+0x1924): undefined reference to `__stack_chk_fail' arch/i386/kernel/built-in.o: In function `arch_ptrace': (.text+0x5281): undefined reference to `__stack_chk_fail' arch/i386/kernel/built-in.o:intel_cacheinfo.c:(.text+0x9e43): more undefined references to `__stack_chk_fail' follow make[1]: *** [.tmp_vmlinux1] Error 1 make[1]: Leaving directory `/usr/src/linux-2.6.17' make: *** [debian/stamp-build-kernel] Error 2 root@felix-laptop:/usr/src/linux#
I'm getting the same with 2.6.17.14 and 2.6.17.
Must be an Edgy thing, I swear this compiled fine in Dapper.Code:CC init/version.o LD init/built-in.o LD .tmp_vmlinux1 init/built-in.o: In function `try_name': do_mounts.c:(.text+0x4b8): undefined reference to `__stack_chk_fail' init/built-in.o: In function `name_to_dev_t': (.text+0x77f): undefined reference to `__stack_chk_fail' init/built-in.o: In function `change_floppy': (.init.text+0x8ea): undefined reference to `__stack_chk_fail' init/built-in.o: In function `mount_block_root': (.init.text+0xb75): undefined reference to `__stack_chk_fail' init/built-in.o: In function `do_header': initramfs.c:(.init.text+0x4113): undefined reference to `__stack_chk_fail' arch/i386/kernel/built-in.o:(.text+0x46f6): more undefined references to `__stack_chk_fail' follow make[1]: *** [.tmp_vmlinux1] Error 1 make[1]: Leaving directory `/usr/src/linux-2.6.17.14' make: *** [debian/stamp-build-kernel] Error 2
- EDIT -
Solution, http://ubuntuforums.org/showthread.p...stack_chk_fail
Last edited by OPaul; October 30th, 2006 at 12:40 AM.
I have always wished for a computer that would be as easy to use as my telephone. My wish came true. I no longer know how to use my telephone.
--Bjarne Stroustrup
I got same problem after I upgraded to Edgy but that solution doesn't work with this Or am I doing something wrong? I edited Makefile and changed CFLAG to this: HOSTCFLAGS = -fno-stack-protector -Wall -Wundef Wstrict-prototypes -Wno-trigraphs \ -fno-strict-aliasing -fno-common
Edit Hehe, it look's like my problem was that I edited HOSTCFLAGS -line ... Now I edited correct line and it seems to work.
Last edited by Xirkka; October 30th, 2006 at 11:53 AM.
I'm tring to compile the new kernel downloaded from kernel.org 2.6.18.1 but i get this error
any idea on how can i get it out??Code:for module in /usr/src/modules/fglrx ; do \ if test -d $module; then \ (cd $module; \ if ./debian/rules KVERS="2.6.18.1-ale" KSRC="/usr/src/linux" \ KMAINT="Unknown Kernel Package Maintainer" KEMAIL="unknown@unconfigured.in.etc.kernel-pkg.conf" \ KPKG_DEST_DIR="/usr/src/linux/.." \ KPKG_MAINTAINER="Unknown Kernel Package Maintainer" \ KPKG_EXTRAV_ARG="EXTRAVERSION=.1-ale" \ ARCH="x86_64" \ KDREV="10.00.Custom" kdist_image; then \ echo "Module $module processed fine"; \ else \ echo "Module $module failed."; \ if [ "X" != "X" ]; then \ echo "Perhaps $module does not understand --rootcmd?"; \ echo "If you see messages that indicate that it is not"; \ echo "in fact being built as root, please file a bug "; \ echo "against $module."; \ fi; \ echo "Hit return to Continue"; \ read ans; \ fi; \ ); \ else \ echo "Module $module does not exist"; \ echo "Hit return to Continue?"; \ fi; \ done make[1]: Entering directory `/usr/src/modules/fglrx' dh_testdir /usr/bin/make -C /usr/src/linux SUBDIRS=/usr/src/modules/fglrx modules make[2]: Entering directory `/usr/src/linux-2.6.18.1' CC [M] /usr/src/modules/fglrx/firegl_public.o /usr/src/modules/fglrx/firegl_public.c:233: error: ‘UTS_RELEASE’ undeclared here (not in a function) /usr/src/modules/fglrx/firegl_public.c:447: warning: initialisation from incompatible pointer type /usr/src/modules/fglrx/firegl_public.c: In function ‘firegl_stub_open’: /usr/src/modules/fglrx/firegl_public.c:570: warning: assignment discards qualifiers from pointer target type /usr/src/modules/fglrx/firegl_public.c: In function ‘firegl_put_user_ptr’: /usr/src/modules/fglrx/firegl_public.c:1330: warning: cast from pointer to integer of different size /usr/src/modules/fglrx/firegl_public.c:1330: warning: cast from pointer to integer of different size /usr/src/modules/fglrx/firegl_public.c:1330: warning: cast from pointer to integer of different size /usr/src/modules/fglrx/firegl_public.c:1330: warning: cast from pointer to integer of different size /usr/src/modules/fglrx/firegl_public.c: In function ‘__ke_unregister_ioctl32_conversion’: /usr/src/modules/fglrx/firegl_public.c:2515: warning: ‘return’ with a value, in function returning void /usr/src/modules/fglrx/firegl_public.c: In function ‘__ke_vm_map’: /usr/src/modules/fglrx/firegl_public.c:3175: error: ‘VM_SHM’ undeclared (first use in this function) /usr/src/modules/fglrx/firegl_public.c:3175: error: (Each undeclared identifier is reported only once /usr/src/modules/fglrx/firegl_public.c:3175: error: for each function it appears in.) make[3]: *** [/usr/src/modules/fglrx/firegl_public.o] Error 1 make[2]: *** [_module_/usr/src/modules/fglrx] Error 2 make[2]: Leaving directory `/usr/src/linux-2.6.18.1' make[1]: *** [build] Error 2 make[1]: Leaving directory `/usr/src/modules/fglrx' Module /usr/src/modules/fglrx failed. Hit return to Continue
Regards
Bookmarks