[ 
https://issues.apache.org/jira/browse/IMAGING-379?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Arnout Engelen updated IMAGING-379:
-----------------------------------
    Description: 
Processing images can take considerable amounts memory, and running out of 
memory can impact the entire application. It would be great to have a mechanism 
to make sure image processing remains within 'reasonable' bounds, and fail 
gracefully when it exceeds them. Of course what is 'reasonable' is 
application-dependent.

This issue exists to track the ideas and experiments around implementing such a 
mechanism, which will likely have impact on various APIs. The 'Allocator' API 
on the master branch is the current work towards this goal.

Potentially interesting testcases for this infrastructure include:

* [https://oss-fuzz.com/testcase?key=4885302936469504]

  was:
Processing images can take considerable amounts memory, and running out of 
memory can impact the entire application. It would be great to have a mechanism 
to make sure image processing remains within 'reasonable' bounds, and fail 
gracefully when it exceeds them. Of course what is 'reasonable' is 
application-dependent.

This issue exists to track the ideas and experiments around implementing such a 
mechanism, which will likely have impact on various APIs. The 'Allocator' API 
on the master branch is the current work towards this goal.


> Generic solution for bounding memory use
> ----------------------------------------
>
>                 Key: IMAGING-379
>                 URL: https://issues.apache.org/jira/browse/IMAGING-379
>             Project: Commons Imaging
>          Issue Type: Improvement
>          Components: imaging.*
>            Reporter: Arnout Engelen
>            Priority: Minor
>
> Processing images can take considerable amounts memory, and running out of 
> memory can impact the entire application. It would be great to have a 
> mechanism to make sure image processing remains within 'reasonable' bounds, 
> and fail gracefully when it exceeds them. Of course what is 'reasonable' is 
> application-dependent.
> This issue exists to track the ideas and experiments around implementing such 
> a mechanism, which will likely have impact on various APIs. The 'Allocator' 
> API on the master branch is the current work towards this goal.
> Potentially interesting testcases for this infrastructure include:
> * [https://oss-fuzz.com/testcase?key=4885302936469504]



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to