Skip to content

Conversation

@adrian-nicolau
Copy link

@adrian-nicolau adrian-nicolau commented Sep 15, 2025

In kernel 6.12, only with loglevel=0 boot argument, sometimes we have modem FW asserts (ucode 0x4091 means empty RFs):

"vpp.service","dpdk             [notice]: 0003:01:00.0: Firmware error detected, assert codes FW 0x00004610, UCODE 0x00004091"

Armada 8k has 4 x PCI: cp0_pcie0, cp0_pcie1, cp1_pcie0, cp1_pcie1. The issue happens when they are not probed in this order, for example cp0_pcie1, cp1_pcie0, cp1_pcie1 and only then cp0_pcie0. This PR adds support for node dependencies, so the right order can be enforced:

&cp0_pcie1 {
    [...]
    depends-on = <&cp0_pcie0>;
};

At boot:

[   21.784820] armada8k-pcie f2620000.pcie: deferred probe: /cp0/pcie@f2600000 not ready
[   21.785034] armada8k-pcie f4600000.pcie: deferred probe: /cp0/pcie@f2620000 not ready
[   21.785223] armada8k-pcie f4620000.pcie: deferred probe: /cp1/pcie@f4600000 not ready

@shmuelhazan shmuelhazan changed the title DEV-6442: ptp_clockmatrix - wait to settle before registering the DPLL as HW clock DEV-6442: ptp_clockmatrix: wait to settle before registering the DPLL as HW clock Sep 16, 2025
@adrian-nicolau adrian-nicolau marked this pull request as draft September 16, 2025 11:14
@adrian-nicolau adrian-nicolau changed the title DEV-6442: ptp_clockmatrix: wait to settle before registering the DPLL as HW clock DEV-6442: pcie-armada8k: add support for node dependencies Sep 17, 2025
@adrian-nicolau adrian-nicolau changed the title DEV-6442: pcie-armada8k: add support for node dependencies DEV-7615: pcie-armada8k: add support for node dependencies Sep 18, 2025
@adrian-nicolau adrian-nicolau marked this pull request as ready for review September 18, 2025 07:34
@adrian-nicolau adrian-nicolau merged commit afeeddb into tg-v6.12 Sep 18, 2025
@adrian-nicolau adrian-nicolau deleted the adrianni-tg-v6.12-falcon-pll-pci-sync branch September 18, 2025 07:49
shmuelhazan pushed a commit that referenced this pull request Nov 6, 2025
In kernel 6.12, only with `loglevel=0` boot argument, sometimes we have
modem FW asserts (ucode 0x4091 means _empty RFs_):
```
"vpp.service","dpdk             [notice]: 0003:01:00.0: Firmware error detected, assert codes FW 0x00004610, UCODE 0x00004091"
```

Armada 8k has 4 x PCI: cp0_pcie0, cp0_pcie1, cp1_pcie0, cp1_pcie1. The
issue happens when they are not probed in this order, for example
cp0_pcie1, cp1_pcie0, cp1_pcie1 and only then cp0_pcie0. This PR adds
support for node dependencies, so the right order can be enforced:
```
&cp0_pcie1 {
    [...]
    depends-on = <&cp0_pcie0>;
};
```

At boot:
```
[   21.784820] armada8k-pcie f2620000.pcie: deferred probe: /cp0/pcie@f2600000 not ready
[   21.785034] armada8k-pcie f4600000.pcie: deferred probe: /cp0/pcie@f2620000 not ready
[   21.785223] armada8k-pcie f4620000.pcie: deferred probe: /cp1/pcie@f4600000 not ready
```
adrian-nicolau added a commit that referenced this pull request Nov 13, 2025
In kernel 6.12, only with `loglevel=0` boot argument, sometimes we have
modem FW asserts (ucode 0x4091 means _empty RFs_):
```
"vpp.service","dpdk             [notice]: 0003:01:00.0: Firmware error detected, assert codes FW 0x00004610, UCODE 0x00004091"
```

Armada 8k has 4 x PCI: cp0_pcie0, cp0_pcie1, cp1_pcie0, cp1_pcie1. The
issue happens when they are not probed in this order, for example
cp0_pcie1, cp1_pcie0, cp1_pcie1 and only then cp0_pcie0. This PR adds
support for node dependencies, so the right order can be enforced:
```
&cp0_pcie1 {
    [...]
    depends-on = <&cp0_pcie0>;
};
```

At boot:
```
[   21.784820] armada8k-pcie f2620000.pcie: deferred probe: /cp0/pcie@f2600000 not ready
[   21.785034] armada8k-pcie f4600000.pcie: deferred probe: /cp0/pcie@f2620000 not ready
[   21.785223] armada8k-pcie f4620000.pcie: deferred probe: /cp1/pcie@f4600000 not ready
```
adrian-nicolau added a commit that referenced this pull request Nov 28, 2025
In kernel 6.12, only with `loglevel=0` boot argument, sometimes we have
modem FW asserts (ucode 0x4091 means _empty RFs_):
```
"vpp.service","dpdk             [notice]: 0003:01:00.0: Firmware error detected, assert codes FW 0x00004610, UCODE 0x00004091"
```

Armada 8k has 4 x PCI: cp0_pcie0, cp0_pcie1, cp1_pcie0, cp1_pcie1. The
issue happens when they are not probed in this order, for example
cp0_pcie1, cp1_pcie0, cp1_pcie1 and only then cp0_pcie0. This PR adds
support for node dependencies, so the right order can be enforced:
```
&cp0_pcie1 {
    [...]
    depends-on = <&cp0_pcie0>;
};
```

At boot:
```
[   21.784820] armada8k-pcie f2620000.pcie: deferred probe: /cp0/pcie@f2600000 not ready
[   21.785034] armada8k-pcie f4600000.pcie: deferred probe: /cp0/pcie@f2620000 not ready
[   21.785223] armada8k-pcie f4620000.pcie: deferred probe: /cp1/pcie@f4600000 not ready
```
adrian-nicolau added a commit that referenced this pull request Dec 1, 2025
In kernel 6.12, only with `loglevel=0` boot argument, sometimes we have
modem FW asserts (ucode 0x4091 means _empty RFs_):
```
"vpp.service","dpdk             [notice]: 0003:01:00.0: Firmware error detected, assert codes FW 0x00004610, UCODE 0x00004091"
```

Armada 8k has 4 x PCI: cp0_pcie0, cp0_pcie1, cp1_pcie0, cp1_pcie1. The
issue happens when they are not probed in this order, for example
cp0_pcie1, cp1_pcie0, cp1_pcie1 and only then cp0_pcie0. This PR adds
support for node dependencies, so the right order can be enforced:
```
&cp0_pcie1 {
    [...]
    depends-on = <&cp0_pcie0>;
};
```

At boot:
```
[   21.784820] armada8k-pcie f2620000.pcie: deferred probe: /cp0/pcie@f2600000 not ready
[   21.785034] armada8k-pcie f4600000.pcie: deferred probe: /cp0/pcie@f2620000 not ready
[   21.785223] armada8k-pcie f4620000.pcie: deferred probe: /cp1/pcie@f4600000 not ready
```
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants