F8M is unstable

[Eung Ho Kim]

Re: F8M is unstable

Thanks, Lsd.

I'm looking forward to new bios. Do you know when it will be released?

And push them to release formal bios, not beta.
GIGABYTE tend to release beta bios after formal bios released.
It's hard to see formal bios since last August.

[Pablo]

Re: F8M is unstable

Currently running ud9 on F4 bios with mode set to ide now, working ok.

I think with mine, when i updated to F5p bios, then went back to F4, its not erased something from the F5 bios.

But as im booting up ok in this state, i will wait for a fixed bios also.

Odd that Gigabyte asked for a bios screenshot, its a bit like a car manufacturer, asking the customer how to make them.

Anyway, its good they are looking into it.

If they dont know around their own bioses, it doesent give me much confidence in them fixing it?

[Lsdmeasap]

Re: F8M is unstable

I am not sure when they will get this fixed, they thought they had it today with F8n and force updating the Marvell Firmware, but I had already tested and failed with that so I had to let them know if they had success with it that it was random success or something else is causing the issue. Once they do correct it the BIOS will be a beta, until a final is rolled out after that.

Ya, the GGTS helpers only go by what's shown in the manual usually, so they wouldn't know about this setting unless they actually looked in a recent BIOS because it wasn't part of the original BIOS (So not in manual)

It was funny to hear though!

[Pablo]

Re: F8M is unstable

Yeah, prob a random succcess.

As with all the combinations i tried, you think youve got it, then next boot, it fails again.

But once theyve got it with the ud7 bios/ should fix it for other boards and bioses with similar or same issue.

Happy Easter.

[Lsdmeasap]

Re: F8M is unstable

Ya, I let them know you guys had it working for a few times, then on reboot it would fail again.

I'm sure they will get it figured out.

For now you should just set the BIOS to Firmware mode for that setting shown above, then it will be fine.

Happy Easter to you as well!

[Lsdmeasap]

Re: F8M is unstable

Hi guys! Got word back from Taiwan tonight, with a fix on this issue!

I have tested it on both ports, in firmware and bypass modes through several reboots from windows, and it is working fine in either mode from either port.

You will need to update the Marvell firmware for the fix to apply, this my suggested method to update the Marvell Firmware.


Update the BIOS with Qflash, Disable "Keep DMI Data" before the flash.

Once the flash is done reboot to the BIOS and load optimized defaults, then save/apply/F10 and reboot back to the BIOS once more. This time in the BIOS go to Integrated Peripherals BIOS page and set the Marvell SATA3 Firmware Selection mode to "Force" as shown in the image below, and at this time also set Intel RAID to enabled if you have a RAID array.



Then hit Save/apply/F10 and reboot back to the BIOS once more time. This time in the BIOS go into Advanced BIOS page and then into Hard Drive Boot Priority and move your OS drive to the top of this list, verify the above Firmware setting is still set to "Force" then F10 to save and apply.

Once you reboot this time to windows you will see the Marvell Firmware being updated at the top of the screen on the last POST Screen (The IRQ Screen), then the system will reboot. On reboot you may want to again enter the BIOS to verify that your hard disk boot priority is still set to the disk you want to boot from.


Here is the BIOS update
X58A-UD7-F8o.zip

[Pablo]

Re: F8M is unstable

Sounds long and complex, but if that fixes it , then nice one.

However, i need the UD9 version please, or do i just follow the procedure as above, but using the ud9 F5p one?

I guess i need a UD9 fixed version though?

Or no doubt, just a version that flashes and is modded to use the new marvel firmware.?

[Eung Ho Kim]

Re: F8M is unstable

Thanks for your hard work, Lsd.

I update it successfully.
What is the Marvell firmware version in F8o?

[Lsdmeasap]

Re: F8M is unstable

Thanks for your hard work, Lsd.

I update it successfully.
What is the Marvell firmware version in F8o?

Anytime, glad they were able to fix this for you!

The Marvell ROM is the same as the older F8 versions, they just changed some other setting to fix this, or a setting within the Marvell ROM but the version is the same.

Sounds long and complex, but if that fixes it , then nice one.

However, i need the UD9 version please, or do i just follow the procedure as above, but using the ud9 F5p one?

I guess i need a UD9 fixed version though?

Or no doubt, just a version that flashes and is modded to use the new marvel firmware.?

Sorry, when I first asked about this I linked them to this thread and forgot to mention UD9 also needed fixed.

Please follow the above method to update, here is the fixed BIOS
X58A-UD9-F5q.zip

[Eung Ho Kim]

Re: F8M is unstable

Anytime, glad they were able to fix this for you!

The Marvell ROM is the same as the older F8 versions, they just changed some other setting to fix this, or a setting within the Marvell ROM but the version is the same.



Sorry, when I first asked about this I linked them to this thread and forgot to mention UD9 also needed fixed.

Please follow the above method to update, here is the fixed BIOS
X58A-UD9-F5q.zip

Lsd, F8o is working fine! Good.

By the way, Intel will release Core i7-995X Extreme Edition(3.60GHz) soon. Is it compatible with current F8o bios? Or need to additional bios update?

[Lsdmeasap]

Re: F8M is unstable

Great to hear, thanks for the update!

I'm sure it probably has already been added into the BIOS because I'm sure Gigabyte has been testing those CPU's for a while now.

[Eung Ho Kim]

Re: F8M is unstable

Lsd, X58A's new formal BIOSes newly posted. Waited too long!

Compared with previous F8o version, what changed?

[dbguy]

Re: F8M is unstable

Hi guys! Got word back from Taiwan tonight, with a fix on this issue!

I have tested it on both ports, in firmware and bypass modes through several reboots from windows, and it is working fine in either mode from either port.

You will need to update the Marvell firmware for the fix to apply, this my suggested method to update the Marvell Firmware.


Update the BIOS with Qflash, Disable "Keep DMI Data" before the flash.

Once the flash is done reboot to the BIOS and load optimized defaults, then save/apply/F10 and reboot back to the BIOS once more. This time in the BIOS go to Integrated Peripherals BIOS page and set the Marvell SATA3 Firmware Selection mode to "Force" as shown in the image below, and at this time also set Intel RAID to enabled if you have a RAID array.



Then hit Save/apply/F10 and reboot back to the BIOS once more time. This time in the BIOS go into Advanced BIOS page and then into Hard Drive Boot Priority and move your OS drive to the top of this list, verify the above Firmware setting is still set to "Force" then F10 to save and apply.

Once you reboot this time to windows you will see the Marvell Firmware being updated at the top of the screen on the last POST Screen (The IRQ Screen), then the system will reboot. On reboot you may want to again enter the BIOS to verify that your hard disk boot priority is still set to the disk you want to boot from.


Here is the BIOS update
X58A-UD7-F8o.zip

Pardon the intrusion, but is the "fix" you mention here, which is for the X58A-UD7, also available for the X58A-UD3R? Thank you.

[dbguy]

Re: F8M is unstable

OK, guys, I am able to reproduce this error now! I was doing some TRIM testing for some users at Crucial, and immediately ran into this issue mid-way through my testing process.

The error affects Marvell SATA 6Gb/s port 6 (against PCB), when the BIOS is set to Firmware mode Bypass and a C300 is connected to port 6 the POST will hang at IDE detection. This may also affect any other disk or SSD, I did not test.

Port 7 is fine, and either port is fine if firmware mode is set to firmware instead of bypass.

I will let my contacts know as soon as I hear back from them, hopefully this week!

Please pardon the interruption, but does this mean that there is hope for TRIM being passed via the 9128 to a 6GB/s SSD? What did you find? Thanks very much.

[Lsdmeasap]

Re: F8M is unstable

No TRIM in any of my testing for 9128, this fix is in order to solve another issue.

As for those official BIOSes guys, I am not sure, have you tested them yet? If not you can test and see what Intel RAID ROM is loaded during bootup, if it's the newer one then the fix may not be carried over, that is if the use of the old ROM is what actually fixed the problem. And if that is what fixed the problem, and the new BIOS does not have the old ROM, I can put the old ROM in there for you guys if you need it. I'll check the BIOS later tonight and possibly go ahead and mod in the old ROM for you so you guys can test if you like.

As for the UD3R, are you having detection issues on port 6, but no issues on port 7? If not then you wouldn't need the fix, if so, and the above is true that the old ROM is what fixes the issue I can fix it for you. As for an official fix, I did not report that board as having a problem, so no they didn't send me a fixed BIOS for that board. If the guys posting here can test the new BIOS and it has a newer ROM and the issue is still fixed, then it's not something I can fix for you by putting in an old ROM.

To anyone using Marvell ports, the Intel ports will give you better performance anyway, so this is something you should definitely keep in mind especially if you are having issues. Don't believe me? See my C300 review linked in my signature.