search

epics-modules / mrfioc2

EPICS driver for Micro Research Finland event timing system devices
http://epics-modules.github.io/mrfioc2/
Other
8 stars 29 forks source link

CML/GTX supports for EVR-VME300 #157

Closed hongran closed 3 weeks ago

hongran commented 1 month ago

Because the EVR-VME230 uses 20bit CML word rather than 40bit, I need to handle this exception for the VME64 form factor. In the setPattern function for waveform mode, the bits and word arrangement does not work for the EVR-VME300. Since I don't have a EVR-VME230RF card at hand, it is hard for me to decide if it is an generic difference between the firmware versions. Therefore, I kept the old code only for the firmware version of the 230-series hardware.

Will people at ESS test if this patch works for the mtca EVR? (This change only affects the waveform mode, not the 4x pattern or the frequency mode.)

gabrielfedel commented 1 month ago

@hongran I don't know when I will be able to test this change (not sure if I have the setup ready for that yet). But I will do it as soon I can.

@zioven or @agaget do you have interest to check this change also ?

agaget commented 1 month ago

@hongran I don't know when I will be able to test this change (not sure if I have the setup ready for that yet). But I will do it as soon I can.

@zioven or @agaget do you have interest to check this change also ?

Sorry, I'm not sure to understand what we are talking about here... Anyway we have only MTCA EVR300U, and no CML input on it. So I don't think we can test anything with it.

hongran commented 1 month ago

@hongran I don't know when I will be able to test this change (not sure if I have the setup ready for that yet). But I will do it as soon I can. @zioven or @agaget do you have interest to check this change also ?

Sorry, I'm not sure to understand what we are talking about here... Anyway we have only MTCA EVR300U, and no CML input on it. So I don't think we can test anything with it.

CML is an output type, not input. For the mtca-evr-300u, the TCLKA and TCLKB outputs on the backplane uses the CML/GTX logic. Timo said that the TCLKA/TCLKB delivers the sampling clock to some of our ADC applications. Is there a way to route these signals to a scope and check if the output signal is consistent with the configuration?

hongran commented 1 month ago

I just remade a commit to remove two temporary diagnostics output lines.

zioven commented 1 month ago

I have briefly checked the pull request but didn't have the time to test on HW with oscilloscope, and I have couple of comments:

Will try to test this on one of our MTCA-EVR-300RF cards that has GTX output handling available on the UNIV-IO modules outputs

agaget commented 1 month ago

Is there a way to route these signals to a scope and check if the output signal is consistent with the configuration?

With those type of card you can have access to the backplane of your crate https://nateurope.com/product/namc-ext-rtm/

I know for Sure that Jukka from MRF is using that type of card to monitor the backplane

(picture provided by him few years ago) Probe

hongran commented 1 month ago

I have briefly checked the pull request but didn't have the time to test on HW with oscilloscope, and I have couple of comments:

* `shaddowPattern` affects both `waveform` mode and `4x pattern` mode according to `syncPattern`

* there is an `MRFVersion` variable type available with comparsion operators defined as well in `mrfCommon`  as defined in commit [493e32e](https://github.com/epics-modules/mrfioc2/commit/493e32e834765f61f5a373a30120aa68eeceaa8c)

Will try to test this on one of our MTCA-EVR-300RF cards that has GTX output handling available on the UNIV-IO modules outputs

Thanks for your comments.

hongran commented 3 weeks ago

Another PR updated the master branch so this PR is closed automatically. I will rebase these changes and make a new PR on a different branch other than the master.