Hello,
I like to boot Grub from HDD in AHCI modus, but I can't get it working with the defaults for now. It only last until I power off the system.
Where I could obtain the option roms for AHCI and the IMC firmware. Do I need to contact AMD or is it possible to extract it from the UEFI? I only have had success with the video bios and PXE.
Please help!
Best regards, Reiner
--
Dear Reiner,
welcome to coreboot!
Am Sonntag, den 30.06.2013, 16:37 +0200 schrieb Reiner Luett:
I like to boot Grub from HDD in AHCI modus, but I can't get it working with the defaults for now. It only last until I power off the system.
I do not understand that paragraph. Could you please elaborate how you figure out that the SATA controller is not running in AHCI mode anymore after powering off.
In coreboot since some months AHCI mode is the default mode and therefore everything should work out of the box. Even if you selected NATIVE (IDE mode) only loading the payload like GRUB or SeaBIOS would take a little longer. Afterward for example the Linux kernel switches to AHCI mode anyway [1].
Where I could obtain the option roms for AHCI and the IMC firmware. Do I need to contact AMD or is it possible to extract it from the UEFI? I only have had success with the video bios and PXE.
I do not even know if the vendor BIOS ships these ROMs. But `bios_extract` would be the way to go to check.
Thanks,
Paul
Hello Paul,
thank a lot for reply.
Sorry for the delay. I have done some further testing.
On 02.07.2013 23:42, Paul Menzel wrote:
I do not understand that paragraph. Could you please elaborate how you figure out that the SATA controller is not running in AHCI mode anymore after powering off.
That was a bit strange, because I don't know why AHCI is not working with the involved SSD.
My setup is the following: I've compiled coreboot with the defaults and add the needed vBios. After flashing into the ROM and reboot the machine, I can boot from the connected Sandisk pSSD P4 (SDSA3BD-016G). After powering off, the AHCI mode is working, but I cannot boot from the device any more.
I've tried some other devices. USB flash drive with DOS don't work. Also some USB flash drive with grub doesn't work. But booting from a CD/DVD or traditional SATA drive was fine.
If I have changed to IDE Mode, I can boot every device. I guess, that there is a problem with this SSD. Maybe I need to debug the firmware, but I don't know how.
In coreboot since some months AHCI mode is the default mode and therefore everything should work out of the box. Even if you selected NATIVE (IDE mode) only loading the payload like GRUB or SeaBIOS would take a little longer. Afterward for example the Linux kernel switches to AHCI mode anyway [1].
Yes you are right. AHCI mode works. I only recognized a difference with the sensors (CPU internal temp) and the fan control in AHCI mode. With the IDE mode the values aren't correct shown. I would appreciate to use the AHCI mode with the SSD. Maybe you have some suggestions how I could get it working.
I do not even know if the vendor BIOS ships these ROMs. But `bios_extract` would be the way to go to check.
IMC firmware should be free. It comes with the sources. I cannot use bios_extract with the current Asrock UEFI.
Best regards, Reiner
--
Dear Reiner,
Am Samstag, den 20.07.2013, 13:47 +0200 schrieb Reiner Luett:
[…]
On 02.07.2013 23:42, Paul Menzel wrote:
I do not understand that paragraph. Could you please elaborate how you figure out that the SATA controller is not running in AHCI mode anymore after powering off.
That was a bit strange, because I don't know why AHCI is not working with the involved SSD.
My setup is the following: I've compiled coreboot with the defaults and add the needed vBios. After flashing into the ROM and reboot the machine, I can boot from the connected Sandisk pSSD P4 (SDSA3BD-016G).
with AHCI, right?
After powering off, the AHCI mode is working, but I cannot boot from the device any more.
I've tried some other devices. USB flash drive with DOS don't work. Also some USB flash drive with grub doesn't work. But booting from a CD/DVD or traditional SATA drive was fine.
If I have changed to IDE Mode, I can boot every device. I guess, that there is a problem with this SSD. Maybe I need to debug the firmware, but I don't know how.
What firmware? The one from the SSD device?
In coreboot since some months AHCI mode is the default mode and therefore everything should work out of the box. Even if you selected NATIVE (IDE mode) only loading the payload like GRUB or SeaBIOS would take a little longer. Afterward for example the Linux kernel switches to AHCI mode anyway [1].
Yes you are right. AHCI mode works. I only recognized a difference with the sensors (CPU internal temp) and the fan control in AHCI mode.
Please paste the differences.
With the IDE mode the values aren't correct shown. I would appreciate to use the AHCI mode with the SSD. Maybe you have some suggestions how I could get it working.
1. What payloads did you try? 2. Please paste the full logs with debugging output enabled in the payload(s), when it works and when it does not.
I do not even know if the vendor BIOS ships these ROMs. But `bios_extract` would be the way to go to check.
IMC firmware should be free. It comes with the sources. I cannot use bios_extract with the current Asrock UEFI.
Unfortunately, I have no idea how to fix `bios_extract` that.
Thanks,
Paul
Hello Paul,
On 20.07.2013 21:27, Paul Menzel wrote:
with AHCI, right?
Yes! It is starting in AHCI mode after flashing.
What firmware? The one from the SSD device?
Right, the firmware of the Sandisk SSD. I've tried an OCZ Vertex SSD and it fully functional in AHCI mode.
Please paste the differences.
I've forgotten to write, that I don't have suitable COM port adaptor. I will need to order one or ask someone, if he has this old style legacy comport header. With USB2serial cable from Siemens I have only garbled text on the console.
- What payloads did you try?
I've tried SeaBIOS and Grub with the same results. Next time I will try it with a kernel and initrd.
- Please paste the full logs with debugging output enabled in the
payload(s), when it works and when it does not.
Yes, for sure. I will paste the output as soon as I have the serial port header.
Unfortunately, I have no idea how to fix `bios_extract` that.
Me too. But thanks for help.
Regards, Reiner
--
-
Paul Menzel -
Reiner Luett