Re: [BUG] rockpro64: PCI BAR reassignment broken by commit 9d57e61bf723 ("of/pci: Add IORESOURCE_MEM_64 to resource flags for 64-bit memory addresses")

From: Qu Wenruo
Date: Wed May 19 2021 - 07:16:39 EST



On 2021/5/19 下午5:20, Alexandru Elisei wrote:
Hi,

On 5/19/21 8:05 AM, Qu Wenruo wrote:


[...]

Is there anything special needed to boot the kernel with that offending commit
reverted?

I believe this is a separate bug which is being investigated [1].

Things that I tried:

1. Reverted the offending commit from v5.13-rc2, kernel booted.

2. Checkout out tag v5.13-rc1, reverted the offending commit from v5.13-rc1,
kernel hangs during boot. Didn't clean before building.

3. Did a clean and rebuild, kernel boots correctly. This seems to point at the
regression [1], since cleaning was one of the workarounds reported.

Another bug that I noticed is a kernel panic on shutdown, but that's unrelated to
the offending commit (I'll try to bisect it when I have the time).

[1] https://lkml.org/lkml/2021/5/17/2284

Awesome! Thanks for your quick mention about the bug.

Indeed after a clean and offending commit reverted, now I can boot a v5.13-rc2 kernel from NVME without problem.

Thank you very much!
Qu


Thanks,

Alex


Thanks,
Qu


Thanks,
Qu


[..]
[    0.307381] rockchip-pcie f8000000.pcie: host bridge /pcie@f8000000 ranges:
[    0.307445] rockchip-pcie f8000000.pcie:      MEM
0x00fa000000..0x00fbdfffff ->
0x00fa000000
[    0.307481] rockchip-pcie f8000000.pcie:       IO
0x00fbe00000..0x00fbefffff ->
0x00fbe00000
[    0.308406] rockchip-pcie f8000000.pcie: supply vpcie1v8 not found, using
dummy
regulator
[    0.308534] rockchip-pcie f8000000.pcie: supply vpcie0v9 not found, using
dummy
regulator
[    0.374676] rockchip-pcie f8000000.pcie: PCI host bridge to bus 0000:00
[    0.374701] pci_bus 0000:00: root bus resource [bus 00-1f]
[    0.374723] pci_bus 0000:00: root bus resource [mem 0xfa000000-0xfbdfffff]
[    0.374746] pci_bus 0000:00: root bus resource [io  0x0000-0xfffff] (bus
address [0xfbe00000-0xfbefffff])
[    0.374808] pci 0000:00:00.0: [1d87:0100] type 01 class 0x060400
[    0.374943] pci 0000:00:00.0: supports D1
[    0.374961] pci 0000:00:00.0: PME# supported from D0 D1 D3hot
[    0.379473] pci 0000:00:00.0: bridge configuration invalid ([bus 00-00]),
reconfiguring
[    0.379712] pci 0000:01:00.0: [144d:a808] type 00 class 0x010802
[    0.379815] pci 0000:01:00.0: reg 0x10: [mem 0x00000000-0x00003fff 64bit]
[    0.379997] pci 0000:01:00.0: Max Payload Size set to 256 (was 128, max 256)
[    0.380607] pci 0000:01:00.0: 8.000 Gb/s available PCIe bandwidth, limited by
2.5 GT/s PCIe x4 link at 0000:00:00.0 (capable of 31.504 Gb/s with 8.0 GT/s PCIe
x4 link)
[    0.394239] pci_bus 0000:01: busn_res: [bus 01-1f] end is updated to 01
[    0.394285] pci 0000:00:00.0: BAR 14: assigned [mem 0xfa000000-0xfa0fffff]
[    0.394312] pci 0000:01:00.0: BAR 0: assigned [mem 0xfa000000-0xfa003fff
64bit]
[    0.394374] pci 0000:00:00.0: PCI bridge to [bus 01]
[    0.394395] pci 0000:00:00.0:   bridge window [mem 0xfa000000-0xfa0fffff]
[    0.394569] pcieport 0000:00:00.0: enabling device (0000 -> 0002)
[    0.394845] pcieport 0000:00:00.0: PME: Signaling with IRQ 78
[    0.395153] pcieport 0000:00:00.0: AER: enabled with IRQ 78
[..]

And here is the output of lspci when BAR reassignment works:

# lspci -v
00:00.0 PCI bridge: Fuzhou Rockchip Electronics Co., Ltd RK3399 PCI Express Root
Port (prog-if 00 [Normal decode])
     Flags: bus master, fast devsel, latency 0, IRQ 78
     Bus: primary=00, secondary=01, subordinate=01, sec-latency=0
     I/O behind bridge: 00000000-00000fff [size=4K]
     Memory behind bridge: fa000000-fa0fffff [size=1M]
     Prefetchable memory behind bridge: 00000000-000fffff [size=1M]
     Capabilities: [80] Power Management version 3
     Capabilities: [90] MSI: Enable+ Count=1/1 Maskable+ 64bit+
     Capabilities: [b0] MSI-X: Enable- Count=1 Masked-
     Capabilities: [c0] Express Root Port (Slot+), MSI 00
     Capabilities: [100] Advanced Error Reporting
     Capabilities: [274] Transaction Processing Hints
     Kernel driver in use: pcieport
lspci: Unable to load libkmod resources: error -2

01:00.0 Non-Volatile memory controller: Samsung Electronics Co Ltd NVMe SSD
Controller SM981/PM981/PM983 (prog-if 02 [NVM Express])
     Subsystem: Samsung Electronics Co Ltd NVMe SSD Controller SM981/PM981/PM983
     Flags: bus master, fast devsel, latency 0, IRQ 77, NUMA node 0
     Memory at fa000000 (64-bit, non-prefetchable) [size=16K]
     Capabilities: [40] Power Management version 3
     Capabilities: [50] MSI: Enable- Count=1/1 Maskable- 64bit+
     Capabilities: [70] Express Endpoint, MSI 00
     Capabilities: [b0] MSI-X: Enable+ Count=33 Masked-
     Capabilities: [100] Advanced Error Reporting
     Capabilities: [148] Device Serial Number 00-00-00-00-00-00-00-00
     Capabilities: [158] Power Budgeting <?>
     Capabilities: [168] Secondary PCI Express
     Capabilities: [188] Latency Tolerance Reporting
     Capabilities: [190] L1 PM Substates
     Kernel driver in use: nvme

I can provide more information if needed (the board is sitting on my desk) and I
can help with testing the fix.

Thanks,

Alex


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel



_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel