[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20250725-amiable-strict-pudu-5cce71@kuoka>
Date: Fri, 25 Jul 2025 09:11:54 +0200
From: Krzysztof Kozlowski <krzk@...nel.org>
To: Lorenzo Bianconi <lorenzo@...nel.org>
Cc: Andrew Lunn <andrew+netdev@...n.ch>,
"David S. Miller" <davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>, Rob Herring <robh@...nel.org>,
Krzysztof Kozlowski <krzk+dt@...nel.org>, Conor Dooley <conor+dt@...nel.org>,
Simon Horman <horms@...nel.org>, Felix Fietkau <nbd@....name>,
linux-arm-kernel@...ts.infradead.org, linux-mediatek@...ts.infradead.org, netdev@...r.kernel.org,
devicetree@...r.kernel.org
Subject: Re: [PATCH net-next v5 1/7] dt-bindings: net: airoha: npu: Add
memory regions used for wlan offload
On Wed, Jul 23, 2025 at 07:19:50PM +0200, Lorenzo Bianconi wrote:
> Document memory regions used by Airoha EN7581 NPU for wlan traffic
> offloading. The brand new added memory regions do not introduce any
> backward compatibility issues since they will be used just to offload
> traffic to/from the MT76 wireless NIC and the MT76 probing will not fail
> if these memory regions are not provide, it will just disable offloading
> via the NPU module.
>
> Signed-off-by: Lorenzo Bianconi <lorenzo@...nel.org>
> ---
> .../devicetree/bindings/net/airoha,en7581-npu.yaml | 19 +++++++++++++++----
> 1 file changed, 15 insertions(+), 4 deletions(-)
>
> diff --git a/Documentation/devicetree/bindings/net/airoha,en7581-npu.yaml b/Documentation/devicetree/bindings/net/airoha,en7581-npu.yaml
> index 76dd97c3fb4004674dc30a54c039c1cc19afedb3..f99d60f75bb03931a1c4f35066c72c709e337fd2 100644
> --- a/Documentation/devicetree/bindings/net/airoha,en7581-npu.yaml
> +++ b/Documentation/devicetree/bindings/net/airoha,en7581-npu.yaml
> @@ -41,9 +41,18 @@ properties:
> - description: wlan irq line5
>
> memory-region:
> - maxItems: 1
> - description:
> - Memory used to store NPU firmware binary.
I still do not get why this cannot be kept backwards compatible. I
looked at your driver code and NPU offload support RFC, and they look
correct. Yet what stops any future developer from changing:
mt76_npu_init(mdev, pci_resource_start(pdev, 0),
pdev->bus && pci_domain_nr(pdev->bus) ? 3 : 2);
into:
ret = mt76_npu_init(...)
if (ret)
goto err...;
? Why would anyone NOT DO such change in the future?
I think I asked about this last time as well - why you cannot keep new
entries optional (minItems: 1)?
Best regards,
Krzysztof
Powered by blists - more mailing lists