Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size

From	Auger Eric <eric.auger@redhat.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size
Date	2017-09-20 16:30 +0200
Message-ID	<urOaJ-4ey-5@gated-at.bofh.it> (permalink)
References	<uru2l-7SS-7@gated-at.bofh.it> <uruYq-8tS-5@gated-at.bofh.it> <urxjA-1CN-15@gated-at.bofh.it> <urI5j-8tt-3@gated-at.bofh.it> <urMVk-3rX-17@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

Hi Sinan,

On 20/09/2017 15:01, Sinan Kaya wrote:
> On 9/20/2017 3:59 AM, Auger Eric wrote:
>>> My impression is that MRRS is predominantly device and driver
>>> dependent, not topology dependent.  A device can send a read request
>>> with a size larger than MPS, which implies that the device supplying
>>> the read data would split it into multiple TLPs based on MPS.
>> I read that too on the net. However in in 6.3.4.1. (3.0. Nov 10), Rules
>> for SW Configuration it is written:
>> "Software must set Max_Read_Request_Size of an isochronous-configured
>> device with a value that does not exceed the Max_Payload_Size set for
>> the device."
>>
>> But on the the other hand some drivers are setting the MMRS directly
>> without further checking the MPS?
> 
> We discussed this on LPC. MRRS and MPS are two independent concepts and
> are not related to each other under normal circumstances. 
> 
> The only valid criteria is that MRRS needs to be a multiple of MPS.
> 
> https://linuxplumbersconf.org/2017/ocw//system/presentations/4732/original/crs.pdf
> 
> Because completions are required to be a minimum of MPS size. If MRRS > MPS,
> read response is sent as multiple completions.

With that patch, you can end up with MRRS < MPS. Do I understand
correctly this is an issue?

Thanks

Eric
> 
> The only reason you want to match MRRS==MPS is that you don't want a single
> device to hog system resources. You can have MRRS 4k and MPS 128 bytes. Completions
> come in as 128 x N packets on the PCI bus. 
> 
> If you are sharing the same PCI bus with some other device, switch; you are effectively
> stalling other devices. That's why, isochronous devices are requesting small MRRS.
> 
> An Isochronous device is an exception not norm.
>

Thread

[PATCH] vfio/pci: Virtualize Maximum Payload Size Alex Williamson <alex.williamson@redhat.com> - 2017-09-19 19:00 +0200
  Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size Auger Eric <eric.auger@redhat.com> - 2017-09-19 20:00 +0200
    Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size Alex Williamson <alex.williamson@redhat.com> - 2017-09-19 22:30 +0200
      Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size Auger Eric <eric.auger@redhat.com> - 2017-09-20 10:00 +0200
        Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size Sinan Kaya <okaya@codeaurora.org> - 2017-09-20 15:10 +0200
        Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size Sinan Kaya <okaya@codeaurora.org> - 2017-09-20 15:10 +0200
          Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size Auger Eric <eric.auger@redhat.com> - 2017-09-20 16:30 +0200
            Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size Sinan Kaya <okaya@codeaurora.org> - 2017-09-20 17:10 +0200
            Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size Alex Williamson <alex.williamson@redhat.com> - 2017-09-20 18:20 +0200
              Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size Sinan Kaya <okaya@codeaurora.org> - 2017-09-20 18:30 +0200
                Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size Auger Eric <eric.auger@redhat.com> - 2017-09-21 08:40 +0200
  Re: [PATCH] vfio/pci: Virtualize Maximum Payload Size Auger Eric <eric.auger@redhat.com> - 2017-09-21 08:40 +0200

csiph-web