Re: [PATCH 00/25 v7] SBSA UART emulation support in Xen

From: Wei Liu <wei.liu2@citrix.com>
To: Julien Grall <julien.grall@arm.com>
Cc: Stefano Stabellini <sstabellini@kernel.org>,
	Wei Liu <wei.liu2@citrix.com>,
	George Dunlap <George.Dunlap@eu.citrix.com>,
	Andrew Cooper <andrew.cooper3@citrix.com>,
	Ian Jackson <ian.jackson@eu.citrix.com>, Tim Deegan <tim@xen.org>,
	Jan Beulich <jbeulich@suse.com>,
	Andre Przywara <andre.przywara@arm.com>,
	Bhupinder Thakur <bhupinder.thakur@linaro.org>,
	xen-devel@lists.xenproject.org
Subject: Re: [PATCH 00/25 v7] SBSA UART emulation support in Xen
Date: Thu, 10 Aug 2017 14:01:16 +0100	[thread overview]
Message-ID: <20170810130116.uzzidhus6hwfyovz@citrix.com> (raw)
In-Reply-To: <33b24693-7ea9-216f-dabf-182afa2e582b@arm.com>

On Thu, Aug 10, 2017 at 01:40:07PM +0100, Julien Grall wrote:
> Hi Wei,
> 
> On 10/08/17 12:40, Wei Liu wrote:
> > On Thu, Aug 10, 2017 at 01:29:04PM +0530, Bhupinder Thakur wrote:
> > > Hi Wei,
> > > 
> > > On 9 August 2017 at 16:33, Wei Liu <wei.liu2@citrix.com> wrote:
> > > > On Wed, Aug 09, 2017 at 04:28:14PM +0530, Bhupinder Thakur wrote:
> > > > > Hi Julien,
> > > > > 
> > > > > Thanks for the testing.
> > > > > 
> > > > > On 8 August 2017 at 21:29, Julien Grall <julien.grall@arm.com> wrote:
> > > > > > Hi Bhupinder,
> > > > > > 
> > > > > > I gave another and I have a couple of comments.
> > > > > > 
> > > > > > Booting Linux with earlycon enabled take quite a while. I can see the
> > > > > > characters coming slower than on the minitel. It seems to be a bit better
> > > > > > after switching off the bootconsole. Overall Linux is taking ~20 times to
> > > > > > boot with pl011 vs HVC console.
> > > > > > 
> > > > > > I do agree that pl011 is emulated and therefore you have to trap after each
> > > > > > character. But 20 times sounds far too much.
> > > > > > 
> > > > > I think this slowness could be due to ratelimiting of the pl011 events
> > > > > in xenconosle. Currently, the rate limit is
> > > > > set to 30 events per 200 msecs (see RATE_LIMIT_ALLOWANCE/RATE_LIMIT_PERIOD).
> > > > > 
> > > > > I increased the rate limit to 600 events (30 * 20) per 200 msecs. With
> > > > > this change,
> > > > > I see that the the find command is running faster and smoother.
> > > > > Earlier the find output would be jerky.
> > > > 
> > > > Please consider batching the events instead of bumping the limit.
> > > 
> > > If I try to batch the events then it may delay per character
> > > processing (if no further characters are coming) since we will wait
> > > for more events to come to batch them together. By keeping the rate
> > > limit high, it is ensured that it can handle large burst of events. If
> > > there is sporadic data then the rate limit is not hit.
> > 
> > But do you really need to send an event for every character (that's my
> > reading of your patch, please correct me if I'm wrong)? Suppose you have
> > a buffer you can send one event when the buffer is processed.
> 
> I am not sure what you mean by buffer here. The interface with the guest is
> a single data register. So characters are written one by one.

Yes, that's what the guest sees, but that is not necessarily what
xenconsoled sees.

> 
> > 
> > > 
> > > I am thinking of making the rate limit configuration per console.
> > > 
> > 
> > There is the question why vuart is special with regard to other types of
> > console.
> > 
> > If something is mandated by the spec it should be specifically called
> > out to justify the decision to bump the rate limit.
> 
> I might have other ideas how to improve the speed. I will answer on
> Bhupinder's e-mail.
> 

Sure.

> Cheers,
> 
> -- 
> Julien Grall

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel