Re: [PATCH v3 01/11] PCI/P2PDMA: Support peer-to-peer memory

[Sinan Kaya]

On 3/13/2018 6:48 PM, Logan Gunthorpe wrote:
> 
> 
> On 13/03/18 04:29 PM, Sinan Kaya wrote:
>> If hardware doesn't support it, blacklisting should have been the right
>> path and I still think that you should remove all switch business from the code.
>> I did not hear enough justification for having a switch requirement
>> for P2P.
> 
> I disagree.
> 
>> You are also saying that root ports have issues not because of functionality but
>> because of performance. 
> 
> No... performance can also be an issue but the main justification for
> this is that many root ports do not support P2P at all and can fail in
> different ways (usually just dropping the TLPs).
> 
>> What if I come up with a very cheap/crappy switch (something like used in data
>> mining)?
> 
> Good luck. That's not how hardware is designed. PCIe switches that have
> any hope to compete with the existing market will need features like
> NTB, non-transparent ports, etc... and that implies a certain symmetry
> (ie there isn't a special host port because there may be more than one
> and it may move around) which implies that packets will always be able
> to forward between each ports which implies P2P will work.
> 

It is still a switch it can move packets but, maybe it can move data at
100kbps speed. 

What prevents that switch from trying P2P and having a bad user experience?

If everything is so broken, I was suggesting to at least list the switches
you have tested.

What's the problem with this?

Why do you want to assume that all switches are good and all root ports are
bad?

>> I have been doing my best to provide feedback. It feels like you are throwing
>> them over the wall to be honest.
>>
>> You keep implying "not my problem".
> 
> Well, the fact of the matter is that extending this in all the ways
> people like you want face huge problems on all sides. These are not
> trivial issues and holding back work that works for our problem because
> it doesn't solve your problem is IMO just going to grind development in
> this area to a halt. We have to have something we can agree on which is
> safe to start building on. The building can't just emerge fully formed
> in one go.
> 

What if the design is so canned that you can't change anything? 

I have been asking things like getting rid of switch search in ACS
enablement towards achieving generic P2P. You seem to be pushing back.
You said yourself P2P and isolation doesn't go together at this point
but you also care about isolation for other devices that are not doing
P2P.

Confusing...

> P2P proposal go back a long time and have never gotten anywhere because
> there are limitations and people want it to do things that are hard but
> don't want to contribute the work to solving those problems.
> 
>>> Well, if it's a problem for someone they'll have to solve it. We're
>>> targeting JBOFs that have no use for ACS / IOMMU groups at all.
>>
>> IMO, you (not somebody) should address this one way or the other before this
>> series land in upstream.
> 
> The real way to address this (as I've mentioned before) is with some way
> of doing ACS and iomem groups dynamically. But this is a huge project in
> itself and is never going to be part of the P2P patchset.

fair enough.

> 
>> Another assumption: There are other architectures like ARM64 where IOMMU
>> is enabled by default even if you don't use VMs for security reasons.
>> IOMMU blocks stray transactions.
> 
> True, but it doesn't change my point: ACS is not a requirement for Linux
> many many systems do not have it on at all or by default.

I don't think so.

It is not a requirement for you but it is a requirement for me (ARM64 guy).
Linux happens to run on multiple architectures. One exception invalidates your
point.

> 
>> Didn't the ACS behavior change suddenly for no good reason when we enabled
>> your code even though I might not be using the P2P but I happen to have
>> a kernel with P2P config option?
> 

If you are assuming that your kernel option should not be used by general
distributions like Ubuntu/redhat etc. and requires a kernel compilation,
creating a dependency to EXPERT is the right way to do. 

Distributions assume that no damage is done by enabling PCI bus options
under normal circumstances.

> Well no, presumably you made a conscious choice to turn the config
> option on and build a custom kernel for your box. That doesn't seem very
> sudden and the reason is that the two concepts are very much at odds
> with each other: you can't have isolation and still have transactions
> between devices.
> 
> Logan
> 
> 


-- 
Sinan Kaya
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.