[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAALAos8+K4XL7GS7rzLrwO=LoFM0wfG1QdWvPX3ryED5spKVcw@mail.gmail.com>
Date: Fri, 18 Aug 2017 11:42:16 +0530
From: Anup Patel <anup.patel@...adcom.com>
To: Vinod Koul <vinod.koul@...el.com>
Cc: Rob Herring <robh+dt@...nel.org>,
Mark Rutland <mark.rutland@....com>,
Dan Williams <dan.j.williams@...el.com>,
Florian Fainelli <f.fainelli@...il.com>,
Scott Branden <sbranden@...adcom.com>,
Ray Jui <rjui@...adcom.com>,
Linux Kernel <linux-kernel@...r.kernel.org>,
Linux ARM Kernel <linux-arm-kernel@...ts.infradead.org>,
Device Tree <devicetree@...r.kernel.org>,
dmaengine@...r.kernel.org,
BCM Kernel Feedback <bcm-kernel-feedback-list@...adcom.com>
Subject: Re: [PATCH v2 09/16] dmaengine: bcm-sba-raid: Improve
sba_issue_pending() run duration
On Thu, Aug 17, 2017 at 12:06 PM, Vinod Koul <vinod.koul@...el.com> wrote:
> On Tue, Aug 01, 2017 at 04:07:53PM +0530, Anup Patel wrote:
>> The pending sba_request list can become very long in real-life usage
>> (e.g. setting up RAID array) which can cause sba_issue_pending() to
>> run for long duration.
>
> that raises the warning flags.. Even if we have a long pending list why
> would issue_pending run for long. The purpose of the issue_pending() is to
> submit a txn if idle and return. The interrupt and tasklet shall push the
> subsequent txn to hardware...
Yes, we are doing very similar thing in PATCH13 by further
simplifying sba_process_deferred_requests()
Regards,
Anup
Powered by blists - more mailing lists