[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <0507e2d9-6535-277c-bd9a-a009c11bf795@solarflare.com>
Date: Tue, 27 Oct 2020 13:24:24 +0000
From: Edward Cree <ecree@...arflare.com>
To: Sasha Levin <sashal@...nel.org>, <linux-kernel@...r.kernel.org>,
<stable@...r.kernel.org>
CC: Jakub Kicinski <kuba@...nel.org>, <netdev@...r.kernel.org>
Subject: Re: [PATCH AUTOSEL 5.9 054/147] sfc: add and use efx_tx_send_pending
in tx.c
On 26/10/2020 23:47, Sasha Levin wrote:
> From: Edward Cree <ecree@...arflare.com>
>
> [ Upstream commit 1c0544d24927e4fad04f858216b8ea767a3bd123 ]
>
> Instead of using efx_tx_queue_partner(), which relies on the assumption
> that tx_queues_per_channel is 2, efx_tx_send_pending() iterates over
> txqs with efx_for_each_channel_tx_queue().
That assumption was valid for the code as of v5.9; this change was only
needed to support the extra queues that were added for encap offloads.
Thus, this patch shouldn't be backported, unless -stable is also planning
to backport that feature (e.g. 0ce8df661456, 24b2c3751aa3), which I
doubt (it's nearly 20 patches, and can't be considered a bugfix).
-ed
Powered by blists - more mailing lists