Hi Chinmay,
You'd need to look through the vc707/system/vc707_system/vc707_system.runs/synth_1/runme.log and vc707/system/vc707_system/vc707_system.runs/impl_1/runme.log log files to see if there are any "CRITICAL WARNING" or "ERROR" messages. It's unlikely there are any errors at least if it produced a system.bit, but it's possible that Vivado 2013 doesn't support a command we use, or that one of the IP cores has some other bug in 2013 that was fixed by 2015.4. Is there still really no way you could get a licence for 2015.4 or newer? It may save you a lot of time and effort.
You may want to add some debug nets to the design, perhaps on the PC (ifu_exu_pc_d) and initial interrupt packet from ciop_iob. This would let you tell if the initial startup sequence is operating correctly.
Jon
--
You received this message because you are subscribed to the Google Groups "OpenPiton Discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpiton+...@googlegroups.com.
To post to this group, send email to open...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/openpiton/fd0f7c1d-c3d0-4ceb-abad-49a9c7a8a207%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
runme.log. There are 4 critical warnings:-
CRITICAL WARNING: [Netlist 29-160] Cannot set property 'VCCAUX_IO', because the property does not exist for objects of type 'pin'. [/home/esdm/project_piton/openpiton/piton/design/chipset/mc/xilinx/vc707/ip_cores/mig_7series_0/mig_7series_0/user_design/constraints/mig_7series_0.xdc:608]
I run protosyn -b vc707 then I program the FPGA with the generated bitstream. I insert SD card containing the OS image and press reset. Third led blinks the once and then nothing happens. I turned off switch 8 before programming fpga.
I already tested os with bitfile provided in openpiton package, it's working there.
Thanks,
Chinmay
Hi Chinmay,
What does the utilisation look like after synthesis? It sounds like the three cores might not fit since we moved to the newer SD controller, which is larger than the previous one. You could try downsizing a parameter of the core (are you still setting FPGA_SYN_1THREAD?), reducing the cache in the new SD controller, or switching back to the old SD controller. I’ll ask Ang to give you a pointer on the latter option as we had to do it for our Piton chipset too.
Jon
From: <open...@googlegroups.com> on behalf of chinmay shekhar <chinmays...@gmail.com>
Date: Wednesday, 8 November 2017 at 12:36
To: OpenPiton Discussion <open...@googlegroups.com>
--
You received this message because you are subscribed to the Google Groups "OpenPiton Discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpiton+...@googlegroups.com.
To post to this group, send email to open...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/openpiton/7793253c-fc29-4941-a6fb-5036f6161a64%40googlegroups.com.
Hi Chinmay,
This depends what you want to do. Possibilities include:
You may also be able to change PTON_NETWORK_CONFIG in pyhp_setup.tcl to xbar_config. I've never actually tried this on FPGA but in theory it should work since it works in simulation. You can only change PTON_X_TILES and PITON_NUM_TILES when you do this. Don't change PTON_Y_TILES.
To view this discussion on the web, visit https://groups.google.com/d/msgid/openpiton/c9b2666b-d1ca-4462-bdef-8bba3b1dce25%40googlegroups.com.
To unsubscribe from this group and stop receiving emails from it, send an email to openpiton+unsubscribe@googlegroups.com.