Hi
Its good to see the work on this issue progressing, however there was a 
solution ready in https://github.com/qos-ch/logback/pull/477 for review, which 
has been working in a very large closed source systems for over a year now
The fix that is not in master seems not to be threadsafe though. LinkedHashMap 
on which it is based isn't and that why I used CHM
Maintaining a threadsafe no blocking LRU is complex, so I would suggest that to 
implementation in master is at least synchronized

I an happy to work on a better lock free solution though, but I am unconvinced 
that we need to have a limit on the size, and there are simpler approaches to 
limiting the growth if memory constraints are the concern

Regards
Mike

-----Original Message-----
From: logback-dev <[email protected]> On Behalf Of Ceki Gulcu
Sent: 07 August 2021 22:08
To: [email protected]
Subject: [logback-dev] [qos-ch/logback] fd9006: ongoing work on LOGBACK-1509

  Branch: refs/heads/master
  Home:   https://github.com/qos-ch/logback
  Commit: fd9006981ac7ff52fbc79187254b1f3c1bd70bf8
      
https://github.com/qos-ch/logback/commit/fd9006981ac7ff52fbc79187254b1f3c1bd70bf8
  Author: Ceki Gulcu <[email protected]>
  Date:   2021-08-07 (Sat, 07 Aug 2021)

  Changed paths:
    M 
logback-classic/src/main/java/ch/qos/logback/classic/pattern/NamedConverter.java
    M 
logback-classic/src/test/java/ch/qos/logback/classic/pattern/LoggerNameConverterPerfTest.java

  Log Message:
  -----------
  ongoing work on LOGBACK-1509

Signed-off-by: Ceki Gulcu <[email protected]>


_______________________________________________
logback-dev mailing list
[email protected]
http://mailman.qos.ch/mailman/listinfo/logback-dev

_______________________________________________
logback-dev mailing list
[email protected]
http://mailman.qos.ch/mailman/listinfo/logback-dev

Reply via email to