Hello readers,
after some while I tried to use coreboot again on my well supported Gigabyte M57SLI and I was really happy, that the first built image booted the machine.
Last time I tried this, I stumbled across an error with the RAM, and had no time to investigate this further.
With the recent version of coreboot (r6013) I get some errors which I'd like to fix. I don't know if some changes to coreboot happened in the meanwhile or, if the kernel got changed so that this errors appear. My kernel version running is from debian sid with the version number 2.6.32-27 for AMD64.
The problems I get are the following: [ 0.000000] mtrr: your BIOS has set up an incorrect mask, fixing it up. [ 0.000000] Modules linked in: [ 0.000000] Pid: 0, comm: swapper Not tainted 2.6.32-5-amd64 #1 [ 0.000000] Call Trace: [ 0.000000] [<ffffffff8102189f>] ? generic_get_mtrr+0xbf/0xf9 [ 0.000000] [<ffffffff8102189f>] ? generic_get_mtrr+0xbf/0xf9 [ 0.000000] [<ffffffff8104dd8c>] ? warn_slowpath_common+0x77/0xa3 [ 0.000000] [<ffffffff8104de14>] ? warn_slowpath_fmt+0x51/0x59 [ 0.000000] [<ffffffff810219ab>] ? mtrr_wrmsr+0x1c/0x49 [ 0.000000] [<ffffffff81021bb0>] ? get_fixed_ranges+0x71/0x98 [ 0.000000] [<ffffffff810219ab>] ? mtrr_wrmsr+0x1c/0x49 [ 0.000000] [<ffffffff81021a6f>] ? prepare_set+0x97/0x9d [ 0.000000] [<ffffffff81021ac8>] ? post_set+0x53/0x60 [ 0.000000] [<ffffffff814fb0d6>] ? get_mtrr_state+0x2eb/0x2f6 [ 0.000000] [<ffffffff8102189f>] ? generic_get_mtrr+0xbf/0xf9 [ 0.000000] [<ffffffff812f96e3>] ? printk+0x4e/0x5b [ 0.000000] [<ffffffff814fbcad>] ? mtrr_trim_uncached_memory+0x91/0x311 [ 0.000000] [<ffffffff814fac46>] ? mtrr_bp_init+0x1b1/0x1d4 [ 0.000000] [<ffffffff8123a7b7>] ? dmi_name_in_serial+0x1d/0x28 [ 0.000000] [<ffffffff81504f2b>] ? early_gart_iommu_check+0x9b/0x287 [ 0.000000] [<ffffffff814f7141>] ? setup_arch+0x43f/0x9cb [ 0.000000] [<ffffffff811f6a9f>] ? extract_entropy+0x6a/0x125 [ 0.000000] [<ffffffff814f3140>] ? early_idt_handler+0x0/0x71 [ 0.000000] [<ffffffff814f39d0>] ? start_kernel+0xdb/0x3e8 [ 0.000000] [<ffffffff814f33b7>] ? x86_64_start_kernel+0xf9/0x106 [ 0.000000] ---[ end trace a7919e7f17c0a725 ]--- [ 0.000000] initial memory mapped : 0 - 20000000 [ 0.000000] init_memory_mapping: 0000000000000000-000000007ffef000 [ 0.000000] 0000000000 - 007fe00000 page 2M [ 0.000000] 007fe00000 - 007ffef000 page 4k [ 0.000000] kernel direct mapping tables up to 7ffef000 @ 8000-c000 [ 0.000000] RAMDISK: 37753000 - 37fefe82
According to /proc/mtrr it looks like something failed on the MTRR setup: reg00: base=0x000000000 ( 0MB), size= 2048MB, count=1: write-back
I don't really know how this table looked like the last time I had coreboot running without troubles (it was around my last commit about rev. 4362), but it definitely had no MTRR error and the table contained more lines with different values. Rudolf Marek told me yesterday on the IRC that there are some MTRR problems with different boards, but not the same type of problem.
The second one, which is delaying the boot process form the kernel is something regarding the USB controller on the board. Here are the messages from the dmesg, but right now I didn't look that up in detail until now: [ 1.038117] hub 2-0:1.0: USB hub found [ 1.396015] usb 1-10: new high speed USB device using ehci_hcd and address 4 [ 11.528167] usb 1-10: device descriptor read/all, error -110 [ 11.640020] usb 1-10: new high speed USB device using ehci_hcd and address 5 [ 21.772039] usb 1-10: device descriptor read/all, error -110 [ 21.884014] usb 1-10: new high speed USB device using ehci_hcd and address 6 [ 26.904039] usb 1-10: device descriptor read/8, error -110 Oh, I just saw that this could be related to a new USB device which doesn't give a proper name via USB identification. - Need to try booting without device or booting on the proprietary bios.
On the last point I'm not really sure if I should call it problem, or if this is just warnings/errors from kernel changes:[ 5.400416] k8temp 0000:00:18.3: Temperature readouts might be wrong - check erratum #141 [ 5.670067] EDAC MC: Ver: 2.1.0 Oct 30 2010 [ 5.671976] EDAC amd64_edac: Ver: 3.2.0 Oct 30 2010 [ 5.672079] EDAC amd64: This node reports that Memory ECC is currently disabled, set F3x44[22] (0000:00:18.3). [ 5.672087] EDAC amd64: ECC disabled in the BIOS or no ECC capability, module will not load. [ 5.672088] Either enable ECC checking or force module loading by setting 'ecc_enable_override'. [ 5.672089] (Note that use of the override may cause unknown side effects.) [ 5.672113] amd64_edac: probe of 0000:00:18.2 failed with error -22
About the temperature readouts I'm not sure if this is only a warning, because I know that there was some discussion about this erratum from AMD on the list. Does anyone know something about this one?
The ECC check might also only be a warning, as the mainboard (afaik) does not support ECC ram. Hints on that one?
For everyone who read until here, thank you for your time reading the mail.
Kind regards, Harald Gutmann