World of-umm
Upcoming SlideShare
Loading in...5
×
 

World of-umm

on

  • 231 views

Resource: LCU13

Resource: LCU13
Name: World of Unified Memory Management
Date: 28-10-2013
Speaker: Tom Gall, John Stultz

Statistics

Views

Total Views
231
Slideshare-icon Views on SlideShare
231
Embed Views
0

Actions

Likes
0
Downloads
1
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    World of-umm World of-umm Presentation Transcript

    • World of Unified Memory Management Tom Gall, John Stultz, Oct 28th 2013
    • www.linaro.org Android Solutions ● Android Sync ● Atomic Display Framework ● ION Overview Upstream Solutions ● Dmabuf-fences ● KMS Atomic mode-setting & Nuclear page-flipping ● Dmabuf constraint solving & allocation
    • www.linaro.org ● Implicit vs explicit sync contracts ● Seemingly agreement on allowing for optional explicit sync arguments in the APIs – Likely trying to implement Android's Sync api on top of kernel-internal dmabuf-fences ● Next steps: – Maarten started looking at this, but may need assistance – Keep the discussion going ● Issues: – Lack of resources Android Sync vs dmabuf-fences
    • www.linaro.org ● ADF: Simplified KMS-like framework. – Sort of an experiment to see what an ideal framework for Android would look like – Similar to upstream atomic mode-setting / nuclear pageflip work – Integrates Android sync points in the API – Presented as fodder for how KMS could be extended ● Next steps: – Continue pushing community and Android developers to collaborate on this. – Hopefully avoid forking display frameworks! ● Issues: – Lack of resources ADF vs Atomic mode-setting/Nuclear pagefilp
    • www.linaro.org ● Discussed ION interface issues ● Upstream preferred solution is post-attach delayed allocation – Android developers have issues with this approach ● Likely to continue having separate interfaces, but hopefully will share underlying logic. ● Next steps: – Push ION into staging – Try to integrate ION's “heap” allocators in the dmabuf post-attach delayed allocation approach. ● Issues: – No open graphics drivers for proof of concept ION vs Dmabuf constraint solving
    • More about Linaro: http://www.linaro.org/about/ More about Linaro engineering: http://www.linaro.org/engineering/ How to join: http://www.linaro.org/about/how-to-join Linaro members: www.linaro.org/members connect.linaro.org