GIGABYTE Latest Beta BIOS

[sdavids6]

Possible to get x570S gaming X? Some similar boards here but i dont think this one specifically

[TweakinOut!]

F35d, Secure Boot State does not enable
Xtreme Rev. 1.0

EDIT: windows ms32info does not recognize Secure Boot State as being 'On'

[jwsg]

F35d, Secure Boot State does not enable
Xtreme Rev. 1.0

EDIT: windows ms32info does not recognize Secure Boot State as being 'On'

Firstly Secure Boot is a useful extra to have but its not actually needed to upgrade to W11
But care is needed when enabling it as it is possible to get into a boot loop by not following certain steps, so I would suggest these steps -

First make sure your boot drive is GPT and then disable CSM in BIOS

Then in Secure Boot settings if it says Mode:User you can simply Enable Secure Boot

Otherwise if Mode:Setup
- Disable Secure Boot if it says Enabled
- If it says Standard change to Custom
- Change Custom to Standard accepting Factory Defaults
- Enable Secure Boot

[rossi94]

Any news on B550 / B570? It's been weeks since the older boards got agesa 1.2.0.5. I would really like to update to a bios that adressed the safety issues and is not buggy.

[Lefuneste]

Hi there

I post here having found this forum as I am struggling with my GB B550M Aorus Pro P (Ryzen 9 5900X / 32GB ECC Unbuffered) to have a HBA card recognized by the BIOS. I have tried F13, F13i and F14e versions and I have the same issue.

The HBA is a LSI SAS 3008 crossed flash from a Fujitsu PSAS CP 400i

I have checked the card on every angle and it seems to work perfectly on another system based on an i7 3770k. Below are the PCI status I get when running on this system.

01:00.0 Serial Attached SCSI controller: Broadcom / LSI SAS3008 PCI-Express Fusion-MPT SAS-3 (rev 02)
Subsystem: Broadcom / LSI SAS9300-8i
Flags: bus master, fast devsel, latency 0, IRQ 16
I/O ports at e000 [size=256]
Memory at f7940000 (64-bit, non-prefetchable) [size=64K]
Memory at f7900000 (64-bit, non-prefetchable) [size=256K]
Expansion ROM at f7800000 [disabled] [size=1M]
Capabilities: [50] Power Management version 3
Capabilities: [68] Express Endpoint, MSI 00
Capabilities: [a8] MSI: Enable- Count=1/1 Maskable+ 64bit+
Capabilities: [c0] MSI-X: Enable+ Count=96 Masked-
Capabilities: [100] Advanced Error Reporting
Capabilities: [1e0] Secondary PCI Express
Capabilities: [1c0] Power Budgeting <?>
Capabilities: [190] Dynamic Power Allocation <?>
Capabilities: [148] Alternative Routing-ID Interpretation (ARI)
Kernel driver in use: mpt3sas
Kernel modules: mpt3sas


The SSID of the Device is :
01:00.0 0107: 1000:0097 (rev 02)

For those familiar, the config export of the card are :

Adapter Selected is a Avago SAS: SAS3008(C0)

Controller Number : 0
Controller : SAS3008(C0)
PCI Address : 00:01:00:00
SAS Address : 5003005-7-0218-1ff0
NVDATA Version (Default) : 0e.01.00.07
NVDATA Version (Persistent) : 0e.01.00.07
Firmware Product ID : 0x2221 (IT)
Firmware Version : 16.00.12.00
NVDATA Vendor : LSI
NVDATA Product ID : SAS9300-8i
BIOS Version : 08.37.00.00
UEFI BSD Version : 18.00.00.00
FCODE Version : N/A
Board Name : SAS9300-8i
Board Assembly : D3307A13
Board Tracer Number : 00000056478213


As you can see the card is totally operational on the other system based on an Intel i7 3770k with a PCIe gen 3.0 x16 primary slot.

FYI this HBA board requires PCIe Gen 3.0 x8 to run by design. I don't think it can work with less lanes.

Now on the B550M Aorus Pro P, the symptoms are :

-The card is not detected by the BIOS at all. Its heartbeat LED does not even light up, whereas on the working system it light up very fast after I switch it on (like half a second).
-The pre-bios prompt for the Avago utility does not show up at all, even when I put the BIOS to Legacy mode for all devices. It does not show up after POST neither like the PXE boot does.
-Obviously the card is not seen inside the OS.
-The card is slightly warm to the touch whereas on a working system it gets quite hot once it is fully powered. I believe something is blocking the card to fully power up. I have 750W brand new Corsair PSU with an almost empty system. Not a PSU issue.
-The slot I use is the primary PCIe gen 4.0 x16, I use the third slot for my Graphic adapter which is a GT1030 requiring a PCIe Gen 3.0 x4 (same as the third slot). GPU is working with and without the HBA connected.
-I have tried to disable all embedded devices (LAN, Audio, SATA...), remove NVME (I have 2), the result is the same.
-I have tried all possible settings in the BIOS like forcing PCIe mode, bifurcation and so on, the result is the same.
-I have tried the first slot with a Gen 2.0 x16 Graphic adapter and it works. Also works with the GT1030. So the slot itself seems OK in terms of hardware connectivity.

At this point the only possible remaining explanation is that it is a BIOS related issue.

More specifically, it seems that very early in the boot process, the BIOS refuses to power the adapter up. Again on a working system this occurs very fast in less than a second, the card is powered (heartbeat LED turning ON).

I'd be interested to know if this is a known issue with this particular MB, chipset, CPU... And hopefully get a corrected BIOS if this is a confirmed issue.

Thank you

[Lefuneste]

Ok I have confirmed the BIOS issue.

All voltages are correct on the PCIe slot when the card is inserted and the system is running (all pins from the first part of the slot).

--> Except pin A11 PWRGD / PCI RST which probes at 0.69V where it should be at 3.3V. My running GPU card has a 3.3V value for pin A11.

So it is indeed the BIOS that refuses to power the HBA Card.

My diagnostic is thus :

-Either the BIOS refuses to switch the card on because of its SSID for some obscure reason (blacklist or predefined know vendor id list)

-Or it refuses due to its initial power drain which is likely above 35W. As the slot 1 is a "GPU only" slot in terms of marketing, there may be a specific test to protect the MB itself from a GPU inserted without a proper secondary PSU line connected to it. I cannot test this card on the third slot as it requires a Gen 3.0 x8 and the third slot is only x4. So ideally there should be a setting in the BIOS to disable this routine from blocking a working device to start.

[Lefuneste]

Upon further inspection, I realize that the problem comes from the chipset not releasing the PERST# (A11 pin after boot.
This explains why I get 0.69V instead of 3.3V as it it pulled down to GND. Basically it means that the chipset doesn't allow the card to start.

Very interesting when looking at the PCIe boot sequence on this page.

https://www.eetimes.com/the-many-facets-of-power-management-for-computer-peripherals/#

Also following similar issues with Dell HBA controllers you can find everywhere on Google, I have tried a few days ago to block B5 and B6 pins which are SMBus communication pins as they prohibit boot from some similar controllers, It is a very common trick and it is reversible. I had exactly the same behavior. So I believe the problem is not linked with SMBus communication.

Because all this is quite technical and can be confusing for many, keep in mind that the card fully works when put in another system! I have tried again and again...

From all this I can conclude that the issue is not due to software reasons (ie bad SSID), but for some hardware reasons. The chipset is conducting a test and refuses the PCIe card to be started. I have thus tried to inject 12V from the PSU directly into the board to alleviate the PCIe rail supply but this did not do anything. Maybe it is related to the 3.3V rail. Overall it is definitely a FW/Chipset//BIOS issue.

[Selexin]

I am having a lot of problems troubleshooting the USB issue with my Gigabyte B550I Aorus Pro AX. I got some Logitech G560 USB speakers which unfortunately stutter/distort when the PC is 'under load' of say playing a game or having multiple applications with audio playing. 3.5mm headphones appear unaffected, and it is only the USB audio (multiple USB ports tried) that seem to have the issue.

Can I request the latest beta bios release for Gigabyte B550I Aorus Pro AX?

Cheers,
Selexin

[stasio]

Can I request the latest beta bios release for Gigabyte B550I Aorus Pro AX?

Cheers,
Selexin

https://download.gigabyte.com/FileList/BIOS/mb_bios_b550i-aorus-pro-ax_f14c.zip

https://download.gigabyte.com/FileList/BIOS/mb_bios_b550i-aorus-pro-ax_f14d.zip

https://download.gigabyte.com/FileList/BIOS/mb_bios_b550i-aorus-pro-ax_f14e.zip

[TweakinOut!]

Firstly Secure Boot is a useful extra to have but its not actually needed to upgrade to W11
But care is needed when enabling it as it is possible to get into a boot loop by not following certain steps, so I would suggest these steps -

First make sure your boot drive is GPT and then disable CSM in BIOS

Then in Secure Boot settings if it says Mode:User you can simply Enable Secure Boot

Otherwise if Mode:Setup
- Disable Secure Boot if it says Enabled
- If it says Standard change to Custom
- Change Custom to Standard accepting Factory Defaults
- Enable Secure Boot

Thanks for the reply, I did have it enabled in F35c, yes it's in GPT. I currently have CSM disabled which opens the Secure Boot option in UEFI, that's enabled but it says it's Not Active: set to Standard. I do have AMD CPU fTPM Enabled and Trusted Computing 2.0 is active...

EDIT: ran the procedure and it worked, Thanks!

[Raider]

any newer bioses for Z690 UD (1.0) ? F6a is not good enough i have many BSODs using xmp on ADATA AX5U5200C3816G-DCLABK

[oschmadi]

Ok I have confirmed the BIOS issue.

All voltages are correct on the PCIe slot when the card is inserted and the system is running (all pins from the first part of the slot).

--> Except pin A11 PWRGD / PCI RST which probes at 0.69V where it should be at 3.3V. My running GPU card has a 3.3V value for pin A11.

So it is indeed the BIOS that refuses to power the HBA Card.

My diagnostic is thus :

-Either the BIOS refuses to switch the card on because of its SSID for some obscure reason (blacklist or predefined know vendor id list)

-Or it refuses due to its initial power drain which is likely above 35W. As the slot 1 is a "GPU only" slot in terms of marketing, there may be a specific test to protect the MB itself from a GPU inserted without a proper secondary PSU line connected to it. I cannot test this card on the third slot as it requires a Gen 3.0 x8 and the third slot is only x4. So ideally there should be a setting in the BIOS to disable this routine from blocking a working device to start.

Have you tried disabling Secure BOOT in the BIOS of the B550M Aorus Pro P? The LSI BIOS/UEFI from older cards (or older Firmware versions) tend to use unsigned UEFI Code, which is then not allowed to start, if Secure BOOT is enabled. Try disabling that and see, if this works.

[Lefuneste]

I have just received a reply from GB support to my issue with PCIe detection for HBA card. And they indeed went the same route as I did.

They ask to test the card by masking the B5 and B6 pins in order to avoid SMBus conflict. As stated above this has already be tried and resulted in exactly the same behavior (card not being enabled or detected in BIOS). Unfortunately I cannot even reply to them as this functionality is not offered on the GB support website.

In the meantime I have tried to put a second PCIe card (an old PICe Gen 2.0 x16 Radeon 7550) in place of the first slot, to test the behavior of the motherboard. And it is detected. Not directly though as I had to force the PCIe x16 slot mode from Auto to Gen2. In Auto mode it would not detect it. This motherboard seems to be very strict on PCIe enumeration. I have never seen such a thing in 30+ years building systems.

Out of desperation I purchased a B550M Asus MB today just in order to check the behavior on another brand/BIOS. Unfortunately this HBA controller is required on this build as it is a virtualization platform with 8 HDD and 4 SSD.

[g111]

Upon further inspection, I realize that the problem comes from the chipset not releasing the PERST# (A11 pin after boot.
This explains why I get 0.69V instead of 3.3V as it it pulled down to GND. Basically it means that the chipset doesn't allow the card to start.

Very interesting when looking at the PCIe boot sequence on this page.

https://www.eetimes.com/the-many-facets-of-power-management-for-computer-peripherals/#

Also following similar issues with Dell HBA controllers you can find everywhere on Google, I have tried a few days ago to block B5 and B6 pins which are SMBus communication pins as they prohibit boot from some similar controllers, It is a very common trick and it is reversible. I had exactly the same behavior. So I believe the problem is not linked with SMBus communication.

Because all this is quite technical and can be confusing for many, keep in mind that the card fully works when put in another system! I have tried again and again...

From all this I can conclude that the issue is not due to software reasons (ie bad SSID), but for some hardware reasons. The chipset is conducting a test and refuses the PCIe card to be started. I have thus tried to inject 12V from the PSU directly into the board to alleviate the PCIe rail supply but this did not do anything. Maybe it is related to the 3.3V rail. Overall it is definitely a FW/Chipset//BIOS issue.

Hi there, thanks for the analysis, it is very interesting. Unfortunately I think your analysis here will fall on deaf ears. There's not many knowledgable people here, just BIOS beggars and other people with issues - there's rarely any real advice or help here other than the most basic of things.

I'd suggest reaching out directly to Gigabyte and ask their support to forward your findings to their BIOS engineer team (if they even have one because it is a joke in the first place) and hope they will implement a fix for future BIOS revisions

[Olnium]

Hi there

I post here having found this forum as I am struggling with my GB B550M Aorus Pro P (Ryzen 9 5900X / 32GB ECC Unbuffered) to have a HBA card recognized by the BIOS. I have tried F13, F13i and F14e versions and I have the same issue.

The HBA is a LSI SAS 3008 crossed flash from a Fujitsu PSAS CP 400i

I have checked the card on every angle and it seems to work perfectly on another system based on an i7 3770k. Below are the PCI status I get when running on this system.

01:00.0 Serial Attached SCSI controller: Broadcom / LSI SAS3008 PCI-Express Fusion-MPT SAS-3 (rev 02)
Subsystem: Broadcom / LSI SAS9300-8i
Flags: bus master, fast devsel, latency 0, IRQ 16
I/O ports at e000 [size=256]
Memory at f7940000 (64-bit, non-prefetchable) [size=64K]
Memory at f7900000 (64-bit, non-prefetchable) [size=256K]
Expansion ROM at f7800000 [disabled] [size=1M]
Capabilities: [50] Power Management version 3
Capabilities: [68] Express Endpoint, MSI 00
Capabilities: [a8] MSI: Enable- Count=1/1 Maskable+ 64bit+
Capabilities: [c0] MSI-X: Enable+ Count=96 Masked-
Capabilities: [100] Advanced Error Reporting
Capabilities: [1e0] Secondary PCI Express
Capabilities: [1c0] Power Budgeting <?>
Capabilities: [190] Dynamic Power Allocation <?>
Capabilities: [148] Alternative Routing-ID Interpretation (ARI)
Kernel driver in use: mpt3sas
Kernel modules: mpt3sas


The SSID of the Device is :
01:00.0 0107: 1000:0097 (rev 02)

For those familiar, the config export of the card are :

Adapter Selected is a Avago SAS: SAS3008(C0)

Controller Number : 0
Controller : SAS3008(C0)
PCI Address : 00:01:00:00
SAS Address : 5003005-7-0218-1ff0
NVDATA Version (Default) : 0e.01.00.07
NVDATA Version (Persistent) : 0e.01.00.07
Firmware Product ID : 0x2221 (IT)
Firmware Version : 16.00.12.00
NVDATA Vendor : LSI
NVDATA Product ID : SAS9300-8i
BIOS Version : 08.37.00.00
UEFI BSD Version : 18.00.00.00
FCODE Version : N/A
Board Name : SAS9300-8i
Board Assembly : D3307A13
Board Tracer Number : 00000056478213


As you can see the card is totally operational on the other system based on an Intel i7 3770k with a PCIe gen 3.0 x16 primary slot.

FYI this HBA board requires PCIe Gen 3.0 x8 to run by design. I don't think it can work with less lanes.

Now on the B550M Aorus Pro P, the symptoms are :

-The card is not detected by the BIOS at all. Its heartbeat LED does not even light up, whereas on the working system it light up very fast after I switch it on (like half a second).
-The pre-bios prompt for the Avago utility does not show up at all, even when I put the BIOS to Legacy mode for all devices. It does not show up after POST neither like the PXE boot does.
-Obviously the card is not seen inside the OS.
-The card is slightly warm to the touch whereas on a working system it gets quite hot once it is fully powered. I believe something is blocking the card to fully power up. I have 750W brand new Corsair PSU with an almost empty system. Not a PSU issue.
-The slot I use is the primary PCIe gen 4.0 x16, I use the third slot for my Graphic adapter which is a GT1030 requiring a PCIe Gen 3.0 x4 (same as the third slot). GPU is working with and without the HBA connected.
-I have tried to disable all embedded devices (LAN, Audio, SATA...), remove NVME (I have 2), the result is the same.
-I have tried all possible settings in the BIOS like forcing PCIe mode, bifurcation and so on, the result is the same.
-I have tried the first slot with a Gen 2.0 x16 Graphic adapter and it works. Also works with the GT1030. So the slot itself seems OK in terms of hardware connectivity.

At this point the only possible remaining explanation is that it is a BIOS related issue.

More specifically, it seems that very early in the boot process, the BIOS refuses to power the adapter up. Again on a working system this occurs very fast in less than a second, the card is powered (heartbeat LED turning ON).

I'd be interested to know if this is a known issue with this particular MB, chipset, CPU... And hopefully get a corrected BIOS if this is a confirmed issue.

Thank you

I have a similar issue with my B550m Aorus Elite and an Intel AX200 WiFi card. The card works fine in an old Intel system I have and the slot on the B550m is fine because it works if I plug in an old GPU. In my case, the card power light does come on at boot but it is not seen by the BIOS nor device manager. I did see a blog post about covering some pins on the card to prevent the MB sending a 3.3v PWR OFF signal to the card but I've yet to trace out the pins on the M.2 to PCIe adapter that the WiFi card resides in so I can try it.

stasio Do you have any comment on these issues?