Hi kp, I agree that we don't really want a fourth kernel variant, but it does seem that the net4501 CPU needs "CONFIG_X86_ELAN=y" - at least with 2.6.35.11.
The changes for the ELAN kernel that is working OK for Graziano are in the diff output below. (Most of the removed lines were as a result of simply saying "y" at the CONFIG_X86_ELAN prompt.) I wonder if the "ELAN" kernel would be a candidate for running on a generic i486 machine? Perhaps Andrew or others could comment based on the diff output. There is some very recent chatter on the Linux Kernel Mailing List (LKML) about CONFIG_X86_ELAN with respect to 2.6.39-rc5 so maybe compatibility will be better with a later kernel. I suggest we park this until after the 4.0 release and revisit for 4.1. davidMbrooke $ diff linux-i486/.config linux-elan/.config 4c4 < # Fri Apr 29 10:45:38 2011 --- > # Fri Apr 29 11:27:17 2011 136a137 > CONFIG_HAVE_PERF_EVENTS=y 141a143 > CONFIG_PERF_COUNTERS=y 236c238 < # CONFIG_X86_ELAN is not set --- > CONFIG_X86_ELAN=y 242,268d243 < # CONFIG_M386 is not set < CONFIG_M486=y < # CONFIG_M586 is not set < # CONFIG_M586TSC is not set < # CONFIG_M586MMX is not set < # CONFIG_M686 is not set < # CONFIG_MPENTIUMII is not set < # CONFIG_MPENTIUMIII is not set < # CONFIG_MPENTIUMM is not set < # CONFIG_MPENTIUM4 is not set < # CONFIG_MK6 is not set < # CONFIG_MK7 is not set < # CONFIG_MK8 is not set < # CONFIG_MCRUSOE is not set < # CONFIG_MEFFICEON is not set < # CONFIG_MWINCHIPC6 is not set < # CONFIG_MWINCHIP3D is not set < # CONFIG_MGEODEGX1 is not set < # CONFIG_MGEODE_LX is not set < # CONFIG_MCYRIXIII is not set < # CONFIG_MVIAC3_2 is not set < # CONFIG_MVIAC7 is not set < # CONFIG_MPSC is not set < # CONFIG_MCORE2 is not set < # CONFIG_MATOM is not set < # CONFIG_GENERIC_CPU is not set < CONFIG_X86_GENERIC=y 270c245 < CONFIG_X86_INTERNODE_CACHE_SHIFT=6 --- > CONFIG_X86_INTERNODE_CACHE_SHIFT=4 272c247 < CONFIG_X86_L1_CACHE_SHIFT=6 --- > CONFIG_X86_L1_CACHE_SHIFT=4 274,276d248 < # CONFIG_X86_PPRO_FENCE is not set < CONFIG_X86_F00F_BUG=y < CONFIG_X86_INVD_BUG=y 282d253 < CONFIG_X86_INTEL_USERCOPY=y 283a255 > CONFIG_X86_DEBUGCTLMSR=y 429a402,403 > CONFIG_ELAN_CPUFREQ=m > CONFIG_SC520_CPUFREQ=m On Wed, 2011-05-11 at 08:19 +0200, Graziano Brioschi wrote: > Hi KP, > > to boot correctly on Soekris Engineering net4501 we **must** have an > ELAN kernel variant, that DavidMBrooke has built for my tests; my system > hangs during early boot process with no console messages using the i486 > BuC40 kernel. > > thanks > graziano > > Il 10/05/2011 21.52, KP Kirchdoerfer ha scritto: > > Am Montag, 9. Mai 2011, um 09:45:22 schrieb Graziano Brioschi: > >> Hi David > >> > >> i can boot automatically using two method: > >> > >> * First method: > >> o special ELAN kernel variant > >> o edit initrd.lrp /boot/etc/modules to append a "pata_legacy" > >> line (nothing else) > >> o devices are automaticaly created by "mdev" > >> * Second Method (preferred because there is not editing in > >> initrd.lrp package) > >> o special ELAN kernel variant > >> o edit syslinux.cfg, adding a "KMODULES=pata_legacy" on the > >> "DEFAULT" line (see the follow syslinux.cfg sample) > >> o device are automatically created by "mdev" during boot process > >> > >> A last note: net4501 has three NS DP83816 network card, so you must load > >> the "natsemi" kernel modules to activate these nic cards (making a "Find > >> & load modules for hardware" discover the natsemi driver correctly) > >> > >> I have booted correctly the system without "libata.dma=0" parameter, so > >> I think it is not necessary. > > Graziano; > > > > just for clarification - does it work *without* a "special ELAN kernel > > variant"? Keeping the no. of images as low as possible will easy > > maintenance. > > > > thx kp > > ------------------------------------------------------------------------------ Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay ------------------------------------------------------------------------ leaf-user mailing list: leaf-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/leaf-user Support Request -- http://leaf-project.org/
