You can subscribe to this list here.
2009 |
Jan
|
Feb
(2) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
---|
From: Joshua A. <jo...@us...> - 2009-02-14 17:01:27
|
I am in the process of adding support to FlexotaskThreadFactory to support having scheduler threads at different priority (we needed to hack this kind of support for the Ptides scheduler). To aid in my own organization, I have updated sourceforge already and am now in the process of updating the alphaWorks release, since this facility requires cooperation from the VM. See https://sourceforge.net/tracker2/?func=detail&aid=2595618&group_id=242572&atid=1119824#. For some period, then, it may be that anyone using the latest dev code will see UnsatisfiedLinkErrors when running on a realtime VM. This can be fixed by commenting out the call to vmBridge.adoptPriority in FlexotaskSystemSupport. Obviously, this code won't be released to the public download until alphaWorks has been updated, so it only effects those running code checked directly out of sourceforge, which is hopefully just us developers. Josh |
From: Joshua A. <jo...@us...> - 2009-02-11 19:21:58
|
I just committed a change to the validator and some of the runtime data structures that could cause a "regression" in some multi-threaded schedulers (the streaming scheduler might be affected). I believe that any such "regression" would not be a real regression but rather a revealing of a latent bug. However, let me know of any problems. This checking was added as part of a successful attempt to track down a nasty and very intermittent bug in the Ptides scheduler. Josh |