RE: [PATCH v1 2/3] scsi: ufs: Optimize host lock on transfer requests send/compl paths

From: Avri Altman <Avri.Altman@wdc.com>
To: "Asutosh Das (asd)" <asutoshd@codeaurora.org>,
	Bart Van Assche <bvanassche@acm.org>,
	Can Guo <cang@codeaurora.org>,
	"nguyenb@codeaurora.org" <nguyenb@codeaurora.org>,
	"hongwus@codeaurora.org" <hongwus@codeaurora.org>,
	"linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
	"kernel-team@android.com" <kernel-team@android.com>
Cc: Stanley Chu <stanley.chu@mediatek.com>,
	Alim Akhtar <alim.akhtar@samsung.com>,
	"James E.J. Bottomley" <jejb@linux.ibm.com>,
	"Martin K. Petersen" <martin.petersen@oracle.com>,
	Matthias Brugger <matthias.bgg@gmail.com>,
	Bean Huo <beanhuo@micron.com>, Jaegeuk Kim <jaegeuk@kernel.org>,
	Adrian Hunter <adrian.hunter@intel.com>,
	Kiwoong Kim <kwmad.kim@samsung.com>,
	Satya Tangirala <satyat@google.com>,
	open list <linux-kernel@vger.kernel.org>,
	"moderated list:ARM/Mediatek SoC support" 
	<linux-arm-kernel@lists.infradead.org>,
	"moderated list:ARM/Mediatek SoC support" 
	<linux-mediatek@lists.infradead.org>
Subject: RE: [PATCH v1 2/3] scsi: ufs: Optimize host lock on transfer requests send/compl paths
Date: Tue, 25 May 2021 08:24:30 +0000	[thread overview]
Message-ID: <DM6PR04MB65752DD2F442C178B2D0233AFC259@DM6PR04MB6575.namprd04.prod.outlook.com> (raw)
In-Reply-To: <d4ff8e1a-f368-6720-798a-a2a31a4d41fb@codeaurora.org>

> On 5/24/2021 1:10 PM, Bart Van Assche wrote:
> > On 5/24/21 1:36 AM, Can Guo wrote:
> >> Current UFS IRQ handler is completely wrapped by host lock, and because
> >> ufshcd_send_command() is also protected by host lock, when IRQ handler
> >> fires, not only the CPU running the IRQ handler cannot send new
> requests,
> >> the rest CPUs can neither. Move the host lock wrapping the IRQ handler
> into
> >> specific branches, i.e., ufshcd_uic_cmd_compl(), ufshcd_check_errors(),
> >> ufshcd_tmc_handler() and ufshcd_transfer_req_compl(). Meanwhile, to
> further
> >> reduce occpuation of host lock in ufshcd_transfer_req_compl(), host lock
> is
> >> no longer required to call __ufshcd_transfer_req_compl(). As per test, the
> >> optimization can bring considerable gain to random read/write
> performance.
> >
> 
> > An additional question is whether it is necessary for v3.0 UFS devices
> > to serialize the submission path against the completion path? Multiple
> > high-performance SCSI LLDs support hardware with separate submission
> and
> > completion queues and hence do not need any serialization between the
> > submission and the completion path. I'm asking this because it is likely
> > that sooner or later multiqueue support will be added in the UFS
> > specification. Benefiting from multiqueue support will require to rework
> > locking in the UFS driver anyway.
> >
> Hi Bart,
> No it's not necessary to serialize both the paths. I think this series
> attempts to remove this serialization to a certain degree, which is
> what's giving the performance improvement.
> 
> Even if multiqueue support would be available in the future, I think
> this change is apt now for the current available specification.
I agree - this looks like the harbinger of a major change,
And going further with respect of hw queues,
will need the spec support - e.g. doorbell per lane, etc.

Thanks,
Avri
 
> > Thanks,
> >
> > Bart.
> >
> 
> 
> Thanks,
> -asd
> 
> --
> The Qualcomm Innovation Center, Inc. is a member of the Code Aurora
> Forum,
> Linux Foundation Collaborative Project