Re: [PATCH net-next v5 00/19] net: stmmac: preserve datapath state across MTU and resume failures

[Maxime Chevallier <maxime.chevallier@bootlin.com>]

Hi James,

On 9/27/26 23:59, James Hilliard wrote:
> Keep the stmmac datapath coherent after failed MTU changes or hardware
> resume without changing the interface's administrative state. Retain the
> working MTU configuration for rollback, and allow ordinary down/up
> recovery when hardware cannot be restored.
> 
> This revision incorporates the following previously posted fixes as
> prerequisites:
> 
> - Linkui Xiao's v3 MDIO reset GPIO fix:
>   https://lore.kernel.org/r/20260921015727.2643540-1-xiaolinkui@126.com
> - Ding Hui's v3 DMA descriptor allocation cleanup:
>   https://lore.kernel.org/r/20260919121436.1642724-1-dinghui1111@163.com
> - Lorenzo Bianconi's v3 EST series, patches 2-4:
>   https://lore.kernel.org/r/20260902-stmmac-est-reapply-after-open-v3-0-e72a6df5a7ef@oss.qualcomm.com

Then please wait for them to be merged.

Also why are you targetting net-next now ?

And most important, no more than 15 outstanding patches,
this series alone goes over the limit.

You need to split this up, tackle the issues one at a time, even
if the overall goal is "MTU and resume failures".

As you can see there's a lot of backlog in this driver, and if you try
to address sashiko's review by adding more patches and account for
all "pre-existing issues", this won't work.

This series already went from a 3-ish series that was
sent alongside the allwinner glue to 19 patches in about a week.

Please think about the people who have to review that :

>  39 files changed, 2390 insertions(+), 808 deletions(-)

Not only the quantity, but the quality of the commit logs, the changelogs
and the comments. This really feels like the raw output of an LLM and it's
very hard to read.

Maxime