Rod,
I'm tending to agree after searching more for code associated with
tags. At present looking for code that may be active that does some kind
of polling or status requests at regular intervals and Includes any that
may become active if there is a change in the command serial number due
to status peek requests. So far this particular problem has been very
allusive which means that I'm looking in the wrong places and don't
understand what to look for.
But why would something transfer large blocks of data that appears to
contain no data (all zeros) and place it in the TCP buffer at the same
time a status requests data is being sent back to the remote at specific
intervals and do it say six or seven times in a sequence. I have
considered that this could possibly be a memcpy routines of some memory
space being sent in addition to the actual status data the remote requested.
Or perhaps just long strings that have been allocated for use but
contain no data for something that is not yet complete. That concept
kind of follows the increase in the NML status buffer size in the NML
configuration file. Divide that increase by six or seven and it could
approximate the total of the fathom data transfer bytes for each
iteration. This is just thinking out loud here and likely thinking to much.
My number 6 could be 7 if also considering the larger number bytes
transferred for number 7 in my lists in which it appears that data in
the TCP buffer is piling up, possibly due to being locked out for a bit
and not able to pull the data out of the TCP buffer.
On 10/28/2020 1:10 PM, Rod Webster wrote:
I don't believe the issue is related to state tags.
I've done a fair bit of work with extending state tags to include some more
useful data as motion pins at run time.
All State tags does is add an additional structure to emcStatus that
records the interpreter state of a given segment. So the interp primitives
update the tags.
The only annoyance is that interp is written in C++ and motion is in C.
Interp keeps its own C++ status structure which is copied across to
emcStatus ( using discrete read and write functions in interp). But that
well and truly predates state tags.
I recall reading the source notes you refer to and I think what it means is
that as long as you use the enums when referring to status bits, it does
not matter where the data is stored as the enum applies the correct index
to the data. The other issue is that some of the underlying tag data enums
are #defined private to interp which requires hard coded numerics on the
motion side which is not desirable and could introduce errors. This refers
https://github.com/LinuxCNC/linuxcnc/pull/900
Rod Webster
*1300 896 832*
+61 435 765 611
VMN®
www.vmn.com.au
On Thu, 29 Oct 2020 at 06:09, Johannes Fassotte <
johan...@automationassist.com> wrote:
Did the addition of state tags mess up status replies that are routed to
my TCP connected UI? Data is being sent even though it was not requested
at fixed intervals. This problem appeared after upgrading from a early
version of 2.9 pre0 master to current version of 2.9 pre0. master. No
changes have been made to it.
As far as I know the normal replies to status requests are handled by
the emcmaintask at whatever rate is requested by the remote as long as
it slower than the timing rate set in the .ini file. The below shows
that there are seven data transmissions sent that are causing serious
problems at regular intervals.
The normal transmission of status data to the remote is in response to a
status peek request of buffer 2 and such a request is always constant in
size with the location of data within it also being constant unless the
LinuxCNC configuration is changed. Inserting a bunch of added data
preset intervals is really bad when thinking about this from a remotes
point of view. In such a case the remote has to be expecting it and as a
minimum need to know the size or length of the data. Neither of those
two requirements can be met right now. I suspect that two separate
processes are fighting for control with the remote ending up being the
looser. A real time process and a none real time process.
I have looked at actual length of the data that is causing this issue
which may be 72980 bytes. Other other data is likely being merged into
it from the remote status peek requests. The long one on the end of each
bad data segment is likely the TCP buffer filling up due to the remotes
inability to empty it because of being locked out for a while.
I'm working around this issue temporarily by pulling the problem data
out of the TCP buffer and dumping it. This does cause the loss of some
status updates but allows me to continue to work on other UI work.
Is the bad data is likely being generated separately by a real time task
that sends it out at specific intervals?. The below pattern is repeating
over and over. Additional comment at the end.
Data error 74272 lenght is not 103716, Bad data Lenght: 74272 Data Not
part of emcStatus
Data error 74224 lenght is not 103716, Bad data Lenght: 74224 Data Not
part of emcStatus
Data error 74276 lenght is not 103716, Bad data Lenght: 74276 Data Not
part of emcStatus
Data error 74204 lenght is not 103716, Bad data Lenght: 74204 Data Not
part of emcStatus
Data error 74424 lenght is not 103716, Bad data Lenght: 74424 Data Not
part of emcStatus
Data error 74436 lenght is not 103716, Bad data Lenght: 74436 Data Not
part of emcStatus
Data error 280176 lenght is not 103716, Bad data Lenght: 280176 Data Not
part of emcStatus
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
Data error 74272 lenght is not 103716, Bad data Lenght: 74272 Data Not
part of emcStatus
Data error 74224 lenght is not 103716, Bad data Lenght: 74224 Data Not
part of emcStatus
Data error 74276 lenght is not 103716, Bad data Lenght: 74276 Data Not
part of emcStatus
Data error 74204 lenght is not 103716, Bad data Lenght: 74204 Data Not
part of emcStatus
Data error 74204 lenght is not 103716, Bad data Lenght: 74204 Data Not
part of emcStatus
Data error 74204 lenght is not 103716, Bad data Lenght: 74204 Data Not
part of emcStatus
Data error 280628 lenght is not 103716, Bad data Lenght: 280628 Data Not
part of emcStatus
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
Data error 74220 lenght is not 103716, Bad data Lenght: 74220 Data Not
part of emcStatus
Data error 74276 lenght is not 103716, Bad data Lenght: 74276 Data Not
part of emcStatus
Data error 74276 lenght is not 103716, Bad data Lenght: 74276 Data Not
part of emcStatus
Data error 74204 lenght is not 103716, Bad data Lenght: 74204 Data Not
part of emcStatus
Data error 74204 lenght is not 103716, Bad data Lenght: 74204 Data Not
part of emcStatus
Data error 74204 lenght is not 103716, Bad data Lenght: 74204 Data Not
part of emcStatus
Data error 280628 lenght is not 103716, Bad data Lenght: 280628 Data Not
part of emcStatus
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
103716 <<<<< Data bytes matches with what was sent (maintask)
Data error 74272 lenght is not 103716, Bad data Lenght: 74272 Data Not
part of emcStatus
Data error 74224 lenght is not 103716, Bad data Lenght: 74224 Data Not
part of emcStatus
Data error 74276 lenght is not 103716, Bad data Lenght: 74276 Data Not
part of emcStatus
Data error 74204 lenght is not 103716, Bad data Lenght: 74204 Data Not
part of emcStatus
Data error 74204 lenght is not 103716, Bad data Lenght: 74204 Data Not
part of emcStatus
Data error 74204 lenght is not 103716, Bad data Lenght: 74204 Data Not
part of emcStatus
Data error 280628 lenght is not 103716, Bad data Lenght: 280628 Data Not
part of emcStatus
I have read some notes within files that the index position of the tags
within a enum is not important. This struck me as strange since then the
only way to find the right one would by through a search of some kind
for name instead of direct addressing. This would seem to defeat part of
the purpose of tags due to decreasing speed. I mention this because a
remote would likely expect consistency in data locations.
_______________________________________________
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers
_______________________________________________
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers
_______________________________________________
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers
