cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Contributor
Contributor
857 Views
Registered: ‎02-12-2018

what is the microblaze init signal in MIG

I use the ultrascale FPGA to make the system testing memory.

I modify the code and then, microblaze(cal_riu) in MIG does not work properly.

cal_addr_decode.sv file and riu files take charge of the calibration from init to cal_done. which signal init the microblaze? (ub_initDone in cal_addr_decode.sv is always zero. so, next stage as MR init is not working and, finally calibration is not working. not cablibration fail)

 

Thank you in advance.

0 Kudos
Reply
6 Replies
Teacher
Teacher
844 Views
Registered: ‎07-09-2009

can I clarify,
your modifying the MicroBlaze used in the MIG to calibrate the DDR memory interface ?

If so, not a good idea.
<== If this was helpful, please feel free to give Kudos, and close if it answers your question ==>
0 Kudos
Reply
Contributor
Contributor
841 Views
Registered: ‎02-12-2018

I don't modify the microblaze code such as u_ddr_cal_riu, mcs.xci, and so on.
I modify the upper code include the microblaze like mem_intfc.sv. and cal_top, cal.sv code.

however, microblaze is not working.
I check that the bisc_complete is HIGH using ILA. but, microblaze is not working.

based on the code, I think that the bisc_compelte is High and then, microblaze is working by io_address_strobe, io_write_strobe and io_address.

but still init and done signals (ub_initdone, initdone(MR), ub_caldone, en_vtc(for signal compensation), and phy_ready) are Low.
0 Kudos
Reply
Teacher
Teacher
824 Views
Registered: ‎07-09-2009

Can I suggest that is not a good way to go ahead,

    That micorblaze in the MIG is at best 'special', VERY dedicated to its job of calibration and monitoring.

If you want to do more than look at it, I'd sugest that up have a long chat with your FAE.

 

 

 

<== If this was helpful, please feel free to give Kudos, and close if it answers your question ==>
0 Kudos
Reply
Contributor
Contributor
809 Views
Registered: ‎02-12-2018

Thank you. I'll contact the FAE to resolve the microblaze problem.

0 Kudos
Reply
Moderator
Moderator
785 Views
Registered: ‎08-21-2007

Was MIG working before you made modification?

-----------------------------------------------------Please don't forget to give kudos or accept as solution if information provided is helpful.---------------------------------------------------------------------
0 Kudos
Reply
Contributor
Contributor
779 Views
Registered: ‎02-12-2018

Yes. I worked three simulations.
First. example_design test-bench test with MIG.XCI(O) and PHY.XCI(O). I don't remove anything.
Second, example_design tb test with MIG.XCI(X) and PHY.XCI(O). i remove only MIG.XCI
Third, example_design tb test with MIG.XCI(X) and PHY.XCI(X). i remove two xci.

First and Second is working well (actually, i modify a little). but, Third is no.
i check the test-bench address and write data and read data in Third case using ILA. address ,add_cmd (RD:0, WR:1), and WR_Data is good. but, RD_data is unexpcted data.

I think that the reasons are two.
First is that the microblaze don't perform to calibrate the I/O.
Second is that XCI influence to PHY to align the data timing basis of observation. but, I just show the parameter and relative path version data in PHY.XCI and PHY.XML.
0 Kudos
Reply