On Mon, 2007-11-05 at 23:22 -0500, jean-francois simon wrote: > Hi, > > thx a lot. i will check my clock.. > do you know what list i should use to post my request? > maybe linux@xxxxxxxxxxxxxxx > thx a lot > -jfs > > > > Sergio Monteiro Basto wrote: > > have you the clock running to fast ? > > your email have more than 3 hour in advance > > I don't know if it is the appropriated because this is the first post in > > months ! > > > > best regards, > > On Mon, 2007-11-05 at 19:33 -0500, jean-francois simon wrote: > >> hi, > >> we have designed an embedded motherboard with dual AMD opteron CPU and > >> Nvidia CK804 chipset. this board works fine. > >> lately we have replaced the two single core CPUs with two dual core > >> CPUs, giving us a total of 4 cores. > >> we are seeing that linux kernels from 2.6.21 to 2.6.23 are unable to > >> load the initial ramdisk (initrd) with that new configuration. Basically > >> the kernel complains that it can't load the ELFs with messages such as: > >> > >> "Inconsistency detected by ld.so: ../sysdeps/x86_64/dl-machine: 450: > >> elf-machine-rela-relative: Assertion '((reloc- r_info) & 0xffffffff) ==8' > >> failed! > >> > >> This is happening here: > >> http://lxr.linux.no/source/init/main.c#L762 > >> > >> When initrd support is disabled in the kernel (CONFIG_BLK_DEV_INITRD=N), > >> then the system is able to boot. > >> > >> When using a linux kernel up to 2.6.20.21, the problem doesn't happen. > >> > >> > >> i don't know where to start to try solving this problem. why would the > >> dual core CPUs cause a problem? i fear that just disabling the initrd > >> could just mask a more serious problem. > >> > >> thanks a lot for any help on this > >> > >> Regards, > >> jean-francois simon > >> - > >> To unsubscribe from this list: send the line "unsubscribe linux-x86_64" in > >> the body of a message to majordomo@xxxxxxxxxxxxxxx > >> More majordomo info at http://vger.kernel.org/majordomo-info.html > > -- Sérgio M. B.
Attachment:
smime.p7s
Description: S/MIME cryptographic signature
