See here for the proper LabVIEW bug reporting procedure.
Tags |
(This content has not been tagged yet)
|
![]() |
Jan 12 2007, 11:04 PM
Post
#1
|
|||
|
Changing the world, one VI at a time. JKI ![]() Posts: 1700 Joined: 22-October 02 From: San Francisco, CA Member No.: 17 Using LabVIEW Since:1995 LV:8.2.1 ,8.5 ,7.1.1
My Blog
My Gallery
|
[CAR 45BGR0J1]
The following VIs should obviously be reentrant (but they are not), in order for two parallel calls to work as expected: --------------------
|
||
|
|
|||
| Ad |
Jan 12 2007, 11:04 PM
Post
#
|
||
|
|
|
||
|
|
|||
Jan 12 2007, 11:51 PM
Post
#2
|
|||
![]() LV R&D Envoy NI ![]() Posts: 1226 Joined: 15-August 06 From: Austin, TX Member No.: 5877 Using LabVIEW Since:2000 LV:8.5.1 ,. ,.
My Gallery
|
This was reported to R&D (#45BGR0J1) for further investigation. It will be fixed in the next full release of LV.
-------------------- "A VI outside a class is a gun without a safety. Data outside a class is a target."
--- A message from LabVOOP R&D
|
||
|
|
|||
Jan 13 2007, 07:42 AM
Post
#3
|
|||
![]() (n0ob)² Member Posts: 756 Joined: 22-October 04 From: Duesseldorf / Germany Member No.: 885 Using LabVIEW Since:2001 LV:8.20 ,7.1.1 ,.
|
This was reported to R&D (#45BGR0J1) for further investigation. It will be fixed in the next full release of LV. It's nice to have a "non official but obviously fast and reliable mechanism with semi direct access to the developers for bug reporting" here Thank you Stephen --------------------
|
||
|
|
|||
Jan 15 2007, 01:57 AM
Post
#4
|
|||
![]() LV R&D Envoy NI ![]() Posts: 1226 Joined: 15-August 06 From: Austin, TX Member No.: 5877 Using LabVIEW Since:2000 LV:8.5.1 ,. ,.
My Gallery
|
Kring has posted a tidal wave of bugs against this VI.
http://forums.lavag.org/Wait-for-All-Notifications-should-be-reentrant-t5780.html http://forums.lavag.org/Wait-on-Notification-from-Multiple-output-notifier-values-are-wrong-on-error-t5807.html http://forums.lavag.org/Wait-for-All-Notifications-outputs-extra-null-notification-t5804.html http://forums.lavag.org/Wait-on-Notification-from-Multiple-output-values-can-be-wrong-on-timeout-t5802.html I'll ask the original author of the VI to take a long hard look at it and see if the example can be improved. I do own the Notifier primitives, but this particular example was written by someone else -- I asked that it be included in the shipping LabVIEW because the functionality gets asked for from time to time. This post has been edited by Aristos Queue: Jan 15 2007, 02:02 AM -------------------- "A VI outside a class is a gun without a safety. Data outside a class is a target."
--- A message from LabVOOP R&D
|
||
|
|
|||
Jan 15 2007, 03:19 AM
Post
#5
|
|||
|
Changing the world, one VI at a time. JKI ![]() Posts: 1700 Joined: 22-October 02 From: San Francisco, CA Member No.: 17 Using LabVIEW Since:1995 LV:8.2.1 ,8.5 ,7.1.1
My Blog
My Gallery
|
I'll ask the original author of the VI to take a long hard look at it and see if the example can be improved. I do own the Notifier primitives, but this particular example was written by someone else -- I asked that it be included in the shipping LabVIEW because the functionality gets asked for from time to time. Stephen, Thanks for pushing things forward on your side. One more thing... the description of the "notifiers out" indicator (which is shown in the context menu when you hover over the indicator) states: "notifiers out is the entire array of notifiers in, unless an error occurs. If an error occurs, notifiers out contains only those notifier refnums that caused the error." However, this is not the actual (or the desired, IMO) behavior. If some notifiers are received before the error occurs (for example, if one notifier becomes invalid after others are received), then the notifiers and notifications received should be output to "notifiers out" and "notifications" (respectively). I don't think that it is entirely possible to know which notifiers caused the error, since the "Wait on Notification from Multiple" primitive does not tell you which notifier generated the error. We could infer which notifiers caused the error by checking if any of them are Not a Refnum. But, in any case, the user of this VI could make this check, themselves, if an error did occur. If this VI does actually need to output an array of notifiers that caused the error, then I would prefer if this were an additional output ("notifiers with errors") and we used "notifiers out" to contain the notifiers that were actually fired. Thanks a million! --------------------
|
||
|
|
|||
Aug 1 2008, 09:00 PM
Post
#6
|
|||
![]() LV R&D Envoy NI ![]() Posts: 1226 Joined: 15-August 06 From: Austin, TX Member No.: 5877 Using LabVIEW Since:2000 LV:8.5.1 ,. ,.
My Gallery
|
All these bugs were fixed in LV8.5.
-------------------- "A VI outside a class is a gun without a safety. Data outside a class is a target."
--- A message from LabVOOP R&D
|
||
|
|
|||
![]() ![]() |
| Time is now: 3rd December 2008 - 08:38 AM |