LSI SAS3082E-R
#8
I don't have the same card as you, but the crash points are the same.
And there are many other similar reports. For example, the report shown below.
https://github.com/ayufan-rock64/linux-build/issues/254

Unfortunately, in my case, using the "wait" insertion method suggested in this thread did not improve the situation.

---

Next, I would like to express my view on "ll7.txt" that you presented.
The log shows that "Gen1 Training" succeeded, but "Gen2 Training" failed and fell back to "Gen1".
And it crashes right after "pci 0000: 00: 00.0: bridge configuration invalid ([bus 00-00]), reconfiguring".
If it is functioning properly, the vendor ID acquisition should continue after this, but it ended in a crash before that.

[    2.434952] rockchip-pcie f8000000.pcie: PCIe Gen1 link training completed
...
[    2.935625] rockchip-pcie f8000000.pcie: PCIe link training gen2 timeout, fall back to gen1!
[    2.936458] PCI host bridge /pcie@f8000000 ranges:
[    2.936903]   MEM 0xfa000000..0xfbdfffff -> 0xfa000000
[    2.938606]    IO 0xfbe00000..0xfbefffff -> 0xfbe00000
[    2.939665] rockchip-pcie f8000000.pcie: PCI host bridge to bus 0000:00
[    2.940277] pci_bus 0000:00: root bus resource [bus 00-1f]
[    2.940763] pci_bus 0000:00: root bus resource [mem 0xfa000000-0xfbdfffff]
[    2.941405] pci_bus 0000:00: root bus resource [io  0x0000-0xfffff] (bus address [0xfbe00000-0xfbefffff])
[    2.942889] pci 0000:00:00.0: bridge configuration invalid ([bus 00-00]), reconfiguring
--- Crash occurred ---
[    2.943789] Synchronous External Abort: synchronous external abort (0x96000210) at 0xffffff800c100000
[    2.944600] Internal error: : 96000210 [#1] SMP
[    2.945000] Modules linked in: midgard_kbase dw_hdmi_i2s_audio phy_rockchip_pcie rockchip_saradc
[    2.945796] CPU: 2 PID: 6 Comm: kworker/u12:0 Not tainted 4.4.180 #1
[    2.946352] Hardware name: Pine64 RockPro64 (DT)
[    2.946771] Workqueue: deferwq deferred_probe_work_func
[    2.947234] task: ffffffc0f3244380 task.stack: ffffffc0f3278000
[    2.947758] PC is at rockchip_pcie_rd_conf+0x164/0x1b4
[    2.948210] LR is at rockchip_pcie_rd_conf+0x154/0x1b4
[    2.948661] pc : [<ffffff80084a1220>] lr : [<ffffff80084a1210>] pstate: 600001c5
[    2.949305] sp : ffffffc0f327b7e0
[    2.949598] x29: ffffffc0f327b7e0 x28: 0000000000000000
[    2.950070] x27: 0000000000000001 x26: 0000000000000000
[    2.950542] x25: 0000000000000000 x24: ffffffc0f0d6c800

> but still no card detected
This is why the card is not detected by sudo lspci -vvv
If "rockchip-pcie f8000000.pcie: PCIe probe completed successfully" is executed correctly,
Even if there is no driver, the card itself is detected by "lspci -vvv".
Like the "PNY Nvidia Quadro lspci -vvv" log you presented.

---
This is a proposal.
You point out kernel build options.
Therefore, I think that you are at a level where you can easily change "dtb".

First try to limit the number of lanes ("num-lanes")
If there is still no improvement, try limiting the link speed ("max-link-speed")

I hope the situation will improve, even a little.
And the "driver presence" that you point out is after this hurdle has been overcome.

---
Finally, I will state my view.
My conclusion is that Rockpro64's "PCIE-Express" originally has little margin.
Many people point out that power supply is suspected, but I think this is not correct.

Because, I own three Rockpro64, but all three work differently.
(One unit is unstable, two units operate stably, but performance is different)
Except for Rockpro64, all conditions are the same, so it is not reasonable to doubt the power supply.


Messages In This Thread
LSI SAS3082E-R - by kram0000 - 11-01-2019, 03:05 PM
RE: LSI SAS3082E-R - by kram0000 - 11-02-2019, 11:54 AM
RE: LSI SAS3082E-R - by t4_4t - 11-04-2019, 01:52 AM
RE: LSI SAS3082E-R - by kram0000 - 11-04-2019, 05:57 PM
RE: LSI SAS3082E-R - by t4_4t - 11-04-2019, 08:36 PM
RE: LSI SAS3082E-R - by kram0000 - 11-05-2019, 11:26 AM
RE: LSI SAS3082E-R - by kram0000 - 11-05-2019, 12:27 PM
RE: LSI SAS3082E-R - by t4_4t - 11-05-2019, 08:34 PM
RE: LSI SAS3082E-R - by kram0000 - 11-14-2019, 12:21 AM
RE: LSI SAS3082E-R - by AndrewC78 - 05-10-2020, 05:40 AM
RE: LSI SAS3082E-R - by kram0000 - 09-13-2020, 08:59 PM

Forum Jump:


Users browsing this thread: 1 Guest(s)