Re: [PATCH net v5 0/4] bnxt_en: Make RING FREE more robust

[Joe Damato <joe@dama.to>]

On Mon, Nov 09, 2026 at 04:13:57PM -0800, Michael Chan wrote:
> On Mon, Nov 9, 2026 at 12:12 PM Joe Damato <joe@dama.to> wrote:
> 
> >   - Notifying bnxt_re when DMA is stopped.
> 
> I think this is probably the one item that should be added to the
> patchset.  When DMA is disabled, RoCE will not work at all so adding
> this will make the series more complete.  It may be as simple as
> calling bnxt_ulp_stop(), but I haven't looked into it in detail.

Hm. bnxt_stop_dma runs with rtnl and the netdev instance lock held and
bnxt_ulp_stop -> bnxt_re_suspend takes both. So, I don't think calling
bnxt_ulp_stop is going to work.

Maybe I need to make a new patch where bnxt_stop_dma queues a work item that
calls bnxt_ulp_stop without locks being held? Not sure.

Also: would you be OK with bnxt_re treating BNXT_STATE_DMA_STOPPED as
detached? I'm probably getting this wrong but I think with DMA off
bnxt_re_suspend only takes the detach path when FW_FATAL_COND is set.

LMK if I am missing something here. Just trying to limit the number of
revisions it takes to get something reasonable :)

> We have tested your patches by purposely unmapping the TX buffer early
> to simulate the IOMMU fault.  When the number of rings is small, the
> fall back logic to check the HWRM_RING_FREE response during message
> timeout works most of the time.  With a large number of rings, some of
> the HWRM_RING_FREE messages do not come back at all and DMA will then
> be disabled.  The FW team is still looking into this to understand it
> better.  Thanks.

Thanks for testing.