netdev - Re: [PATCH net-next v2] net: airoha: Implement BQL support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20241015090952.6bcb5856@kernel.org>
Date: Tue, 15 Oct 2024 09:09:52 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Lorenzo Bianconi <lorenzo@...nel.org>
Cc: Felix Fietkau <nbd@....name>, Sean Wang <sean.wang@...iatek.com>, Mark
 Lee <Mark-MC.Lee@...iatek.com>, "David S. Miller" <davem@...emloft.net>,
 Eric Dumazet <edumazet@...gle.com>, Paolo Abeni <pabeni@...hat.com>,
 Matthias Brugger <matthias.bgg@...il.com>, AngeloGioacchino Del Regno
 <angelogioacchino.delregno@...labora.com>,
 linux-arm-kernel@...ts.infradead.org, linux-mediatek@...ts.infradead.org,
 netdev@...r.kernel.org, upstream@...oha.com
Subject: Re: [PATCH net-next v2] net: airoha: Implement BQL support

On Tue, 15 Oct 2024 17:54:59 +0200 Lorenzo Bianconi wrote:
> > Oh, thought its called on stop. In that case we're probably good
> > from BQL perspective.
> > 
> > But does it mean potentially very stale packets can sit on the Tx
> > ring when the device is stopped, until it's started again?  
> 
> Do you mean the packets that the stack is transmitting when the .ndo_stop()
> is run?

Whatever is in the queue at the time ndo_stop() gets called.
Could be the full descriptor ring I presume?

> In airoha_dev_stop() we call netif_tx_disable() to disable the transmission
> on new packets and inflight packets will be consumed by the completion napi,
> is it not enough?

They will only get consumed if the DMA gets to them right?
Stop seems to stop the DMA.

> I guess we can even add netdev_tx_reset_subqueue() for all netdev
> queues in airoha_dev_stop(), I do not have a strong opinion about it. What
> do you prefer?

So to be clear I think this patch is correct as of the current driver
code. I'm just wondering if we should call airoha_qdma_cleanup_tx_queue()
on stop as well, and then that should come with the reset.
I think having a packet stuck in a queue may lead to all sort of oddness
so my recommendation would be to flush the queues.