Hi,
My HP Proliant dl145 g3 with Coreboot Bios and SeaBIOS payload cannot read my hard disk. Here is the serial console output.
http://coreboot.pastebin.com/HYee3u0t
For easy reference, here is the coreboot page on the HP Proliant dl145 g3.
http://www.coreboot.org/HP_DL145_G3
To ensure the problem was not caused by an HP BIOS or the LSI SAS controller raid initialization of hard drives, I used a hard drive with OS created on a non HP(Asus) without either hardware or software raid. No hard drive with boot sector was readable by coreboot/seabios.
Thanks, j
On Sat, Feb 19, 2011 at 06:26:30PM -0500, jarray52 jarray52 wrote:
Hi,
My HP Proliant dl145 g3 with Coreboot Bios and SeaBIOS payload cannot read my hard disk. Here is the serial console output.
http://coreboot.pastebin.com/HYee3u0t
For easy reference, here is the coreboot page on the HP Proliant dl145 g3.
http://www.coreboot.org/HP_DL145_G3
To ensure the problem was not caused by an HP BIOS or the LSI SAS controller raid initialization of hard drives, I used a hard drive with OS created on a non HP(Asus) without either hardware or software raid. No hard drive with boot sector was readable by coreboot/seabios.
The log shows SeaBIOS found an ATA controller, but did not find any drives attached to the controller. It did not find an option rom on any cards - so no scsi option rom was run.
Can you post the output of lspci and identify which device you expected to boot from?
-Kevin
On 02/21/2011 03:53 PM, Kevin O'Connor wrote:
On Sat, Feb 19, 2011 at 06:26:30PM -0500, jarray52 jarray52 wrote:
Hi,
My HP Proliant dl145 g3 with Coreboot Bios and SeaBIOS payload cannot read my hard disk. Here is the serial console output.
http://coreboot.pastebin.com/HYee3u0t
For easy reference, here is the coreboot page on the HP Proliant dl145 g3.
http://www.coreboot.org/HP_DL145_G3
To ensure the problem was not caused by an HP BIOS or the LSI SAS controller raid initialization of hard drives, I used a hard drive with OS created on a non HP(Asus) without either hardware or software raid. No hard drive with boot sector was readable by coreboot/seabios.
The log shows SeaBIOS found an ATA controller, but did not find any drives attached to the controller. It did not find an option rom on any cards - so no scsi option rom was run.
He removed the SAS card from the system, as running its option ROM would freeze the boot process. He has two SATA drives connected to the onboard controller. Initially, I thought the error had to do with the fact that the drives were in RAID, and the boot sector was offset, but he tried with a non-RAID drive connected (with linux already installed), and the result was the same.
The coreboot log does show the SATA controller to be initialized (if I am reading it correctly).
PCI: 01:0e.0 [1166/024a] ops PCI: 01:0e.0 [1166/024a] enabled malloc Enter, size 68, free_mem_ptr 00150224 malloc 00150224 PCI: 01:0e.1 [1166/024a] ops PCI: 01:0e.1 [1166/024a] enabled POST: 0x25
The 1166:024a is the SATA controller.
Can you post the output of lspci and identify which device you expected to boot from?
Better lspci -nn ?
Alex
On Mon, Feb 21, 2011 at 06:16:55PM +0200, Alex G. wrote:
On 02/21/2011 03:53 PM, Kevin O'Connor wrote:
The log shows SeaBIOS found an ATA controller, but did not find any drives attached to the controller. It did not find an option rom on any cards - so no scsi option rom was run.
He removed the SAS card from the system, as running its option ROM would freeze the boot process.
A boot log with debug level set to 8 from that event would be helpful.
[...]
The coreboot log does show the SATA controller to be initialized (if I am reading it correctly).
PCI: 01:0e.0 [1166/024a] ops PCI: 01:0e.0 [1166/024a] enabled malloc Enter, size 68, free_mem_ptr 00150224 malloc 00150224 PCI: 01:0e.1 [1166/024a] ops PCI: 01:0e.1 [1166/024a] enabled POST: 0x25
The 1166:024a is the SATA controller.
It doesn't look like the class field of that pci device identified itself as ATA or AHCI. (SeaBIOS just skipped over the device.)
Can you post the output of lspci and identify which device you expected to boot from?
Better lspci -nn ?
Yeah - actually "lspci -vnn" would be better.
-Kevin
On 02/21/2011 09:01 PM, Kevin O'Connor wrote:
On Mon, Feb 21, 2011 at 06:16:55PM +0200, Alex G. wrote:
On 02/21/2011 03:53 PM, Kevin O'Connor wrote:
The log shows SeaBIOS found an ATA controller, but did not find any drives attached to the controller. It did not find an option rom on any cards - so no scsi option rom was run.
He removed the SAS card from the system, as running its option ROM would freeze the boot process.
A boot log with debug level set to 8 from that event would be helpful.
jarray52, please build SeaBIOS with debug level 8, test, and post the boot log. Leave the SAS card out. We'll deal with that once we get the SATA booting.
It doesn't look like the class field of that pci device identified itself as ATA or AHCI. (SeaBIOS just skipped over the device.)
I don't know how SeaBIOS does things. The init code for the SATA is in src/southbridge/broadcom/bcm5785/sata.c but it's not very well (read "not at all") commented.
Can you post the output of lspci and identify which device you expected to boot from?
Better lspci -nn ?
Yeah - actually "lspci -vnn" would be better.
jarray52, please also post the output of "lspci -vnn" when you get the chance (pastebin, or as a text attachment to the list).
Alex
On Mon, Feb 21, 2011 at 09:19:08PM +0200, Alex G. wrote:
On 02/21/2011 09:01 PM, Kevin O'Connor wrote:
On Mon, Feb 21, 2011 at 06:16:55PM +0200, Alex G. wrote:
On 02/21/2011 03:53 PM, Kevin O'Connor wrote:
The log shows SeaBIOS found an ATA controller, but did not find any drives attached to the controller. It did not find an option rom on any cards - so no scsi option rom was run.
He removed the SAS card from the system, as running its option ROM would freeze the boot process.
A boot log with debug level set to 8 from that event would be helpful.
jarray52, please build SeaBIOS with debug level 8, test, and post the boot log. Leave the SAS card out. We'll deal with that once we get the SATA booting.
The earlier message had the log for debug level 8 for the non-SAS case. A log with debug level 8 for the SAS case would also be useful.
It doesn't look like the class field of that pci device identified itself as ATA or AHCI. (SeaBIOS just skipped over the device.)
I don't know how SeaBIOS does things. The init code for the SATA is in src/southbridge/broadcom/bcm5785/sata.c but it's not very well (read "not at all") commented.
SeaBIOS looks for devices with class=0x0101 for ATA, and class=0x0106 and prog-if=0x01 for AHCI.
-Kevin
-
Alex G. -
jarray52 jarray52 -
Kevin O'Connor