Received-SPF: pass (google.com: domain of linux-nfs-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH rdma-next 00/10] Enable relaxed ordering for ULPs
To:     Haakon Bugge <haakon.bugge@oracle.com>
Cc:     Chuck Lever III <chuck.lever@oracle.com>,
        Jason Gunthorpe <jgg@nvidia.com>,
        Christoph Hellwig <hch@lst.de>,
        Leon Romanovsky <leon@kernel.org>,
        Doug Ledford <dledford@redhat.com>,
        Leon Romanovsky <leonro@nvidia.com>,
        Adit Ranadive <aditr@vmware.com>,
        Anna Schumaker <anna.schumaker@netapp.com>,
        Ariel Elior <aelior@marvell.com>,
        Avihai Horon <avihaih@nvidia.com>,
        Bart Van Assche <bvanassche@acm.org>,
        Bernard Metzler <bmt@zurich.ibm.com>,
        "David S. Miller" <davem@davemloft.net>,
        Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>,
        Devesh Sharma <devesh.sharma@broadcom.com>,
        Faisal Latif <faisal.latif@intel.com>,
        Jack Wang <jinpu.wang@ionos.com>,
        Jakub Kicinski <kuba@kernel.org>,
        Bruce Fields <bfields@fieldses.org>, Jens Axboe <axboe@fb.com>,
        Karsten Graul <kgraul@linux.ibm.com>,
        Keith Busch <kbusch@kernel.org>, Lijun Ou <oulijun@huawei.com>,
        CIFS <linux-cifs@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Linux NFS Mailing List <linux-nfs@vger.kernel.org>,
        "linux-nvme@lists.infradead.org" <linux-nvme@lists.infradead.org>,
        OFED mailing list <linux-rdma@vger.kernel.org>,
        "linux-s390@vger.kernel.org" <linux-s390@vger.kernel.org>,
        Max Gurtovoy <maxg@mellanox.com>,
        Max Gurtovoy <mgurtovoy@nvidia.com>,
        "Md. Haris Iqbal" <haris.iqbal@ionos.com>,
        Michael Guralnik <michaelgur@nvidia.com>,
        Michal Kalderon <mkalderon@marvell.com>,
        Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>,
        Naresh Kumar PBS <nareshkumar.pbs@broadcom.com>,
        Linux-Net <netdev@vger.kernel.org>,
        Potnuri Bharat Teja <bharat@chelsio.com>,
        "rds-devel@oss.oracle.com" <rds-devel@oss.oracle.com>,
        Sagi Grimberg <sagi@grimberg.me>,
        "samba-technical@lists.samba.org" <samba-technical@lists.samba.org>,
        Santosh Shilimkar <santosh.shilimkar@oracle.com>,
        Selvin Xavier <selvin.xavier@broadcom.com>,
        Shiraz Saleem <shiraz.saleem@intel.com>,
        Somnath Kotur <somnath.kotur@broadcom.com>,
        Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>,
        Steve French <sfrench@samba.org>,
        Trond Myklebust <trond.myklebust@hammerspace.com>,
        VMware PV-Drivers <pv-drivers@vmware.com>,
        Weihang Li <liweihang@huawei.com>,
        Yishai Hadas <yishaih@nvidia.com>,
        Zhu Yanjun <zyjzyj2000@gmail.com>
References: <20210405052404.213889-1-leon@kernel.org>
 <20210405134115.GA22346@lst.de> <20210405200739.GB7405@nvidia.com>
 <C2924F03-11C5-4839-A4F3-36872194EEA8@oracle.com>
 <20210406114952.GH7405@nvidia.com>
 <aeb7334b-edc0-78c2-4adb-92d4a994210d@talpey.com>
 <8A5E83DF-5C08-49CE-8EE3-08DC63135735@oracle.com>
 <4b02d1b2-be0e-0d1d-7ac3-38d32e44e77e@talpey.com>
 <1FA38618-E245-4C53-BF49-6688CA93C660@oracle.com>
From:   Tom Talpey <tom@talpey.com>
Message-ID: <7b9e7d9c-13d7-0d18-23b4-0d94409c7741@talpey.com>
Date:   Fri, 9 Apr 2021 13:49:15 -0400
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:68.0) Gecko/20100101
 Thunderbird/68.12.1
MIME-Version: 1.0
In-Reply-To: <1FA38618-E245-4C53-BF49-6688CA93C660@oracle.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 4/9/2021 12:27 PM, Haakon Bugge wrote:
> 
> 
>> On 9 Apr 2021, at 17:32, Tom Talpey <tom@talpey.com> wrote:
>>
>> On 4/9/2021 10:45 AM, Chuck Lever III wrote:
>>>> On Apr 9, 2021, at 10:26 AM, Tom Talpey <tom@talpey.com> wrote:
>>>>
>>>> On 4/6/2021 7:49 AM, Jason Gunthorpe wrote:
>>>>> On Mon, Apr 05, 2021 at 11:42:31PM +0000, Chuck Lever III wrote:
>>>>>   
>>>>>> We need to get a better idea what correctness testing has been done,
>>>>>> and whether positive correctness testing results can be replicated
>>>>>> on a variety of platforms.
>>>>> RO has been rolling out slowly on mlx5 over a few years and storage
>>>>> ULPs are the last to change. eg the mlx5 ethernet driver has had RO
>>>>> turned on for a long time, userspace HPC applications have been using
>>>>> it for a while now too.
>>>>
>>>> I'd love to see RO be used more, it was always something the RDMA
>>>> specs supported and carefully architected for. My only concern is
>>>> that it's difficult to get right, especially when the platforms
>>>> have been running strictly-ordered for so long. The ULPs need
>>>> testing, and a lot of it.
>>>>
>>>>> We know there are platforms with broken RO implementations (like
>>>>> Haswell) but the kernel is supposed to globally turn off RO on all
>>>>> those cases. I'd be a bit surprised if we discover any more from this
>>>>> series.
>>>>> On the other hand there are platforms that get huge speed ups from
>>>>> turning this on, AMD is one example, there are a bunch in the ARM
>>>>> world too.
>>>>
>>>> My belief is that the biggest risk is from situations where completions
>>>> are batched, and therefore polling is used to detect them without
>>>> interrupts (which explicitly). The RO pipeline will completely reorder
>>>> DMA writes, and consumers which infer ordering from memory contents may
>>>> break. This can even apply within the provider code, which may attempt
>>>> to poll WR and CQ structures, and be tripped up.
>>> You are referring specifically to RPC/RDMA depending on Receive
>>> completions to guarantee that previous RDMA Writes have been
>>> retired? Or is there a particular implementation practice in
>>> the Linux RPC/RDMA code that worries you?
>>
>> Nothing in the RPC/RDMA code, which is IMO correct. The worry, which
>> is hopefully unfounded, is that the RO pipeline might not have flushed
>> when a completion is posted *after* posting an interrupt.
>>
>> Something like this...
>>
>> RDMA Write arrives
>> 	PCIe RO Write for data
>> 	PCIe RO Write for data
>> 	...
>> RDMA Write arrives
>> 	PCIe RO Write for data
>> 	...
>> RDMA Send arrives
>> 	PCIe RO Write for receive data
>> 	PCIe RO Write for receive descriptor
> 
> Do you mean the Write of the CQE? It has to be Strongly Ordered for a correct implementation. Then it will shure prior written RO date has global visibility when the CQE can be observed.

I wasn't aware that a strongly-ordered PCIe Write will ensure that
prior relaxed-ordered writes went first. If that's the case, I'm
fine with it - as long as the providers are correctly coded!!

>> 	PCIe interrupt (flushes RO pipeline for all three ops above)
> 
> Before the interrupt, the HCA will write the EQE (Event Queue Entry). This has to be a Strongly Ordered write to "push" prior written CQEs so that when the EQE is observed, the prior writes of CQEs have global visibility.
> 
> And the MSI-X write likewise, to avoid spurious interrupts.

Ok, and yes agreed the same principle would apply.

Is there any implication if a PCIe switch were present on the
motherboard? The switch is allowed to do some creative routing
if the operation is relaxed, correct?

Tom.

> Thxs, Håkon
> 
>>
>> RPC/RDMA polls CQ
>> 	Reaps receive completion
>>
>> RDMA Send arrives
>> 	PCIe RO Write for receive data
>> 	PCIe RO write for receive descriptor
>> 	Does *not* interrupt, since CQ not armed
>>
>> RPC/RDMA continues to poll CQ
>> 	Reaps receive completion
>> 	PCIe RO writes not yet flushed
>> 	Processes incomplete in-memory data
>> 	Bzzzt
>>
>> Hopefully, the adapter performs a PCIe flushing read, or something
>> to avoid this when an interrupt is not generated. Alternatively, I'm
>> overly paranoid.
>>
>> Tom.
>>
>>>> The Mellanox adapter, itself, historically has strict in-order DMA
>>>> semantics, and while it's great to relax that, changing it by default
>>>> for all consumers is something to consider very cautiously.
>>>>
>>>>> Still, obviously people should test on the platforms they have.
>>>>
>>>> Yes, and "test" be taken seriously with focus on ULP data integrity.
>>>> Speedups will mean nothing if the data is ever damaged.
>>> I agree that data integrity comes first.
>>> Since I currently don't have facilities to test RO in my lab, the
>>> community will have to agree on a set of tests and expected results
>>> that specifically exercise the corner cases you are concerned about.
>>> --
>>> Chuck Lever
>