Legacy PCIE HBA in LGA1700 MB
- Thread starter linuxperson
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
I have been looking at some. The scanner I have is a Canoscan. It's a decent scanner. I don't use it all that often.
You bring up an interesting point making me ponder about the bootstrap process. When the hardware initializes , I figure the PCIE slots are searched for adapters, videocards, etc. Would there a whitelist (devices IDs perhaps), in the BIOS? PCIE is advertised as backwards compatible so any PCIE card that is detected should at least show up. For instance in Linux there is a utility called lspci that shows all the devices on the PCIE bus even if they are not claimed by a driver yet. Here is an example:
27:04.0 Multimedia audio controller: C-Media Electronics Inc CMI8788 [Oxygen HD Audio]
Subsystem: ASUSTeK Computer Inc. CMI8786 (Xonar DGX)
your older pcie card only has a bios option rom and no efi driver / BSD. To use the option rom, you'll need CSM enabled and to set all relevant option roms (including video) to use legacy bios. However, it's quite possible (likely) that your scanner doesn't need the option rom in order to be used by the OS; you just won't see it during preboot. Does the
/dev/sg device show up? SANE hardware support is a bit complex.
Last edited:
No the scanner doesn't show up. There are errors in the dmesg output.
Here's the fun stuff:
05:08.0 SCSI storage controller: Broadcom / LSI 53c1030 PCI-X Fusion-MPT Dual Ultra320 SCSI (rev c1)
Subsystem: Broadcom / LSI 53c1030 PCI-X Fusion-MPT Dual Ultra320 SCSI
[ 2.157972] pci 0000:05:08.0: [1000:0030] type 00 class 0x010000
[ 2.158961] pci 0000:05:08.0: reg 0x10: [io 0x3000-0x30ff]
[ 2.159955] pci 0000:05:08.0: reg 0x14: [mem 0x4fc20000-0x4fc3ffff 64bit]
[ 2.160955] pci 0000:05:08.0: reg 0x1c: [mem 0x4fc00000-0x4fc1ffff 64bit]
[ 2.161957] pci 0000:05:08.0: reg 0x30: [mem 0x4fb00000-0x4fbfffff pref]
[ 2.163002] pci 0000:05:08.0: supports D1 D2
[ 4.894735] mptspi 0000:05:08.0: enabling device (0000 -> 0002)
[ 65.039364] systemd-udevd[475]: 0000:05:08.0: Worker [550] processing SEQNUM=1923 is taking a long time
[ 185.039456] systemd-udevd[475]: 0000:05:08.0: Worker [550] processing SEQNUM=1923 killed
[ 292.039452] systemd-udevd[888]: 0000:05:08.0: Worker [893] processing SEQNUM=3502 is taking a long time
[ 412.039441] systemd-udevd[888]: 0000:05:08.0: Worker [893] processing SEQNUM=3502 killed
[ 412.054500] systemd-udevd[888]: 0000:05:08.0: Worker [893] failed
[ 248.998115] mptscsih: ioc0: attempting task abort! (sc=0000000004411113)
[ 259.227074] mptscsih: ioc0: WARNING - Issuing Reset from mptscsih_IssueTaskMgmt!! doorbell=0x24000000
[ 260.046094] mptscsih: ioc0: task abort: SUCCESS (rv=2002) (sc=0000000004411113)
Maybe a timing issue? Anyway it's way down in the weeds.
if the kernel is right that's a pci-x card not a pcie? if it's pci-x as claimed it won't even fit into a pcie slot
you previously said it was a LSI 20320 - that too is pci-x not pci-e - shouldn't be possible to fit that into a pcie slot
you sure you got the model number correct?
Last edited:
This is the core of the issue. No CSM = No legacy BIOS Option ROMs. And MSI seems to be enforcing to use a discrete video card to enable CSM for some weird reason. Check here: MSI Global English Forum
Remember that we're at the point where CSM is supposed to be already phased out, I'm even surprised it is actually still available.
same here with ASRock Z590 PG Velocita. if you wanna enable CSM, then you need discrete graphic card - as i understand it - there is no CPU video support in CSM mode..
It is in fact a LSI20320IE.
This is an update. After hours of fiddling like installing the HBA in three different computers, A Z97 Intel based, an AMD EPYC SuperMicro, and the MSI B660M A PRO, which was in the target box, I managed to get the scanner detected and was able to scan a magazine cover. Honestly, I'm not sure what I did to get it to work because I tried everything I could think of like cleaning the PCIE connector, different SCSI cables and so on. Nothing seemed to help. I installed an older PCI Ultra320 SCSI card I had in the Z97 based computer and verified that the cables were good and the scanner worked. I was "this close" to giving up on the MSI based Alder Lake computer with the LSI20320IE when I thought I'd give it one more try before contacting the seller of the LSI20320IE for a possible return of the HBA.
Funny, I saw a thread on another forum where the poster was having similar issues and errors from the HBA and driver. He took the system apart and re-fitted everything in his system started working too.
Probably everyone here has made that experience at some point
