Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Sender: Tejun Heo <htejun@gmail.com>
Date:   Tue, 17 Aug 2021 08:28:21 -1000
From:   Tejun Heo <tj@kernel.org>
To:     Sean Anderson <sean.anderson@seco.com>
Cc:     Arend van Spriel <aspriel@gmail.com>,
        linux-wireless@vger.kernel.org, Kalle Valo <kvalo@codeaurora.org>,
        Chi-hsien Lin <chi-hsien.lin@cypress.com>,
        Franky Lin <franky.lin@broadcom.com>,
        SHA-cyfmac-dev-list@infineon.com,
        Hante Meuleman <hante.meuleman@broadcom.com>,
        Wright Feng <wright.feng@cypress.com>,
        Arend Van Spriel <arend.vanspriel@broadcom.com>,
        brcm80211-dev-list.pdl@broadcom.com
Subject: Re: [PATCH] brcmfmac: Set SDIO workqueue as WQ_HIGHPRI
Message-ID: <YRv/xUlC5sIMbCxI@slm.duckdns.org>
References: <20210802170904.3116223-1-sean.anderson@seco.com>
 <e6eb89af-b81e-b96e-70a6-c9498eb681fd@seco.com>
 <17b551dc6e0.279b.9696ff82abe5fb6502268bdc3b0467d4@gmail.com>
 <c5bba530-1002-a3fa-322a-1045bf7ee4c2@seco.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <c5bba530-1002-a3fa-322a-1045bf7ee4c2@seco.com>
Precedence: bulk

Hello,

On Tue, Aug 17, 2021 at 02:21:55PM -0400, Sean Anderson wrote:
> > While I understand the obvious gain it seems like a wrong move to me. What if all workqueues in the kernel would start using this flag? I bet the gain above would be negated and all are equal in the eyes of .. the kernel
> 
> Is there an official policy on what counts as high-priority? Using some
> very-scientific methodology [1], it seems like most high-priority
> workqueues are in drivers/net and fs. Making these queues high-priority
> seems to be commonplace. For example, in fe101716c7c9 ("rtw88: replace
> tx tasklet with work queue"), Po-Hao Huang remarks:

I think this is actually a good candidate for HIGHPRI. As you noted, stuff
which interacts with hardware in latency sensitive manner with impact on
observable performance is one of the common use cases. The alternatives
would be doing it from hard/softirqs which are higher priorities anyway.

Thanks.

-- 
tejun