Re: [PATCH net v7 0/7] can: fix bus-off restart and device close races

[Heiko Stübner]

Hi,

Am Freitag, 6. November 2026, 10:37:19 Mitteleuropäische Normalzeit schrieb Cunhao Lu via B4 Relay:
> This work started from a real failure on a production robot using a
> LubanCat-5IO with an RK3588. RX continued, but the TX queue remained
> stopped and subsequent sends returned ENOBUFS. The interface still
> reported ERROR-ACTIVE, and the hardware indicated neither a bus-off
> condition nor a pending TX request.

[...]

> This series depends on the three patches in "can: skb: make CAN skb
> freeing safe in any IRQ context" [1] and the three patches in "can:
> rockchip_canfd: fix TX stalls and races" [2].
> 
> [1] https://lore.kernel.org/linux-can/tencent_CD05F020BAFB2537F5D1F4CBB425588EA80A@xxxxxx/
> [2] https://lore.kernel.org/linux-can/tencent_E25C9BE6689264E4138A464803B810FDFD09@xxxxxx/

Both these series are from july and seemingly from before you started
using the b4-relay. So they may or may not have survived the kernel
releases and their threading is broken both on lore and in any maibox
they arrived in.

And if there haven't been any responses to them since then, I'd think
they have been lost or so.

So I'd suggest rolling both your dependencies forward on top of the can
tree and sending them as a new version, so that reviewers can find them?


Heiko