netdev - Re: [PATCH net v3 3/3] bnxt_en: handle tpa

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] [day] [month] [year] [list]

Message-ID: <5336d624-8d8b-40a6-b732-b020e4a119a2@davidwei.uk>
Date: Mon, 16 Dec 2024 14:59:16 -0800
From: David Wei <dw@...idwei.uk>
To: Michael Chan <michael.chan@...adcom.com>
Cc: Somnath Kotur <somnath.kotur@...adcom.com>,
 Yunsheng Lin <linyunsheng@...wei.com>, netdev@...r.kernel.org,
 Andy Gospodarek <andrew.gospodarek@...adcom.com>,
 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>
Subject: Re: [PATCH net v3 3/3] bnxt_en: handle tpa_info in queue API
 implementation

On 2024-12-16 14:48, Michael Chan wrote:
> On Mon, Dec 16, 2024 at 1:41 PM David Wei <dw@...idwei.uk> wrote:
>>
>> On 2024-12-11 09:11, Michael Chan wrote:
>>> Yeah, I think it makes sense to add napi_disable().  Thanks.
>>
>> Michael, Som. I can't add napi_disable()/enable() because the NAPI
>> instance is shared between the Rx and Tx queues. If I disable a NAPI
>> instance, then it affects the corresponding Tx queue because it is not
>> quiesced. Only the Rx queue is quiesced indirectly by preventhing the HW
>> from receiving packets via the call to bnxt_hwrm_vnic_update().
>>
>> The other implementation of queue API (gve) will quiesce all queues for
>> an individual queue stop/start operation. To call
>> napi_disable()/enable() I believe we will need the same thing for bnxt.
> 
> Som is working on a set of driver patches to make
> queue_stop/queue_start work for the AMD PCIe TPH feature.  He will
> have to disable and re-enable the TX queue for TPH to work.  I think
> we can just include the napi_disable()/napi_enable() in Som's patches
> then.  Thanks.

Sounds good, thanks Michael.