search

Giako68 / TEXT_DEMO

Arduino MKR VIDOR 4000 -- HDMI Text Screen demo
10 stars 1 forks source link

Build failure under Quartus 18.1 #1

Open g3zbu opened 5 years ago

g3zbu commented 5 years ago

I seem to get error messages when compiling TEXT_Demo-master ;-(

Analysis & Synthesis is green ;-) Fitter (Place and Route) is green ;-) Design Assistant (Post-Fitting) is green ;-)

But Compile Design is red... Quartus Prime EDA Netlist writer was unsuccessful. 1 error 293001 Quartus Prime Full Compilation was unsuccessful. 3 errors, 85 warnings.

quartuserrors

I wonder if you can possibly assist?

g3zbu commented 5 years ago

The EDA report file states Error (20268): Functional simulation is off but it is the only supported netlist type for this device. Info (204019): Generated file TEXT_Demo_6_1200mv_85c_slow.vo in folder "D:/APRW/Vidor4000/TEXT_DEMO-master/TEXT_DEMO-master/TEXT_DEMO/FPGA/simulation/modelsim/" for EDA simulation tool Error: Quartus Prime EDA Netlist Writer was unsuccessful. 1 error, 1 warning Error: Peak virtual memory: 4674 megabytes Error: Processing ended: Sun Dec 30 15:50:33 2018 Error: Elapsed time: 00:00:05 Error: Total CPU time (on all processors): 00:00:05

Sadly I cannot discover how to switch on 'Generate functional simulation netlist'. Could that be my problem?

Giako68 commented 5 years ago

Go to: Assignments -> Settings ... -> EDA Tool Settings -> Simulation -> More EDA Netlist Writer Settings and change the Generate functional simulation netlist entry to ON.

I use the 18.0 version of Quartus Prime, maybe that's what makes the difference.

g3zbu commented 5 years ago

Perfect! Zero errors with 18.1.Mille grazie!Happy new year.AlOn 31 December 2018 at 14:44 Giako68 notifications@github.com wrote: Go to: Assignments -> Settings ... -> EDA Tool Settings -> Simulation -> More EDA Netlist Writer Settings and change the Generate functional simulation netlist entry to ON.I use the 18.0 version of Quartus Prime, maybe that's what makes the difference.—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or mute the thread.

g3zbu commented 5 years ago

I now have an SOF file which is converted to an SVF file (1,432KB in size). But the Makeimage.py generates an output file FPGA_Image_RLE.h that is 0KB in size. Is the header part of SVF critical as it says SDR 5748760 TDI but the python code is looking for 👍 if (s[0:17] == "SDR 4087056 TDI ("): Thanks for your rapid replies.

Giako68 commented 5 years ago

You probably modified the project target. Make sure it is the 10CL016YU256C8G chip and rebuild it.

g3zbu commented 5 years ago

I have set the chip and now have an image file fpga_image_RLE.h of 707kb.Another step forwards. Compile of Arduino has stalled so will look at that tomorrow. HNY.On 31 December 2018 at 21:21 Giako68 notifications@github.com wrote: You probably modified the project target. Make sure it is the 10CL016YU256C8G chip and rebuild it.—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or mute the thread.! html>

Giako68 commented 5 years ago

Ok. Happy New Year! :)

g3zbu commented 5 years ago

Hi Giako, a couple of issues I am sorry to say.

I notice that your 18.0 files have Altera_mf set as sim library files. Is that critical? Sadly I cannot find the box/menu to select that, or rather I may have seen it over the last few days but cannot remember where...

Secondly my header file does not match yours so I have copied the first 1 or 2 lines of each in case that gives a clue. Line lengths are way different. My FPGA_Image_RLE.h 0xB3, 0xFF, 0x01, 0x6A, 0x83, 0xF7, 0x17, 0xF3, 0xFB, 0xFB, 0xF9, 0xFB, 0xFA, 0xF0, 0xF8, 0xF8, 0xF9, 0xF8, 0xF9, 0xF8, 0xF8, 0xFA, 0xF8, 0xF8, 0xFB, 0xFB, 0xFA, 0xFB, 0xF9, 0xF9, 0x83, 0xF8, Your file, just line 1: 0xB3, 0xFF, 0x01, 0x6A, 0x83, 0xF7, 0x07, 0xF3, 0xFB, 0xFA, 0xF8, 0xFB, 0xFB, 0xF0, 0x80, 0xF8, 0x0D, 0xF9, 0xF8, 0xF9, 0xF9, 0xFB, 0xF8, 0xF8, 0xFB, 0xFB, 0xFA, 0xFB, 0xF9, 0xF9, 0x83, 0xF8, 0x82, 0xFA, 0x02, 0x40, 0x13, 0xFF, 0xFF, 0xCF, 0xFF, 0xFF, 0x00, 0xFF, 0x00, 0xCD, 0x00, 0x8D, 0x02, 0x8D, 0x00, 0x91, 0x02, 0x8D, 0x00, 0x91, 0x02, 0x8D, 0x00, 0x91, 0x02, 0x8D, 0x00, 0x91, 0x02, 0x8D, 0x00, 0x91, 0x02, 0x8D, 0x00, 0x81, 0x02, 0x81, 0x00, 0x81, 0x02, 0x9D, 0x00, 0x81, 0x02, 0x9D, 0x00, 0x81, 0x02, 0x81, 0x00, 0x81, 0x02, 0x95, 0x00, 0x81, 0x02, 0x81, 0x00, 0x81, 0x02, 0x95, 0x00, 0x81, 0x02, 0x81, 0x00, 0x81, 0x02, 0x8D, 0x00, 0x89, 0x02, 0x81, 0x00, 0x81, 0x02, 0x91, 0x00, 0x81, 0x02, 0x9D, 0x00, 0x81, 0x02, 0x9D, 0x00, 0x81, 0x02, 0x81, 0x00, 0x81, 0x20, 0xA1, 0x00, 0x85, 0x02, 0x81, 0x20, 0x89, 0x00, 0x81, 0x02, 0x81, 0x00, 0x81, 0x20, 0x85, 0x22, 0x8D, 0x00, 0x81, 0x60, 0x81, 0x40, 0x81, 0x20, 0x81, 0x00, 0x81, 0x22, 0x81, 0x40, 0x81, 0x62, 0x81, 0x42, 0x81, 0x60, 0x89, 0x40, 0x81, 0x50,

BTW I am running Win10 not Linux. Hence I was suspicious of the 0x0D at first!

Cheers, Al

Giako68 commented 5 years ago

Most likely, the .svf file generated under Windows is slightly different from the one generated under Linux due to the different handling of text files. To solve the problem you should modify the script in Python, but on this I can not help you since I do not use Windows for many years.

g3zbu commented 5 years ago

Hi Giako,

Thanks. I can easily deal with the Python stuff. I used to use Ubuntu in the past but the wi-fi driver code got a bit forked-up and I didn't want to play kernels with it. Do I need the Altera-mf sim library setting?

Al

On 01 January 2019 at 09:33 Giako68 notifications@github.com wrote:

Most likely, the .svf file generated under Windows is slightly different from the one generated under Linux due to the different handling of text files.
To solve the problem you should modify the script in Python, but on this I can not help you since I do not use Windows for many years.

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub https://github.com/Giako68/TEXT_DEMO/issues/1#issuecomment-450718747 , or mute the thread https://github.com/notifications/unsubscribe-auth/AHdqD5kj46vuSfc62WgUmKd3IcnUaypcks5u-yuAgaJpZM4ZlVad .
Giako68 commented 5 years ago

I think it's only used if you try to simulate the circuit using Modelsim.

g3zbu commented 5 years ago

Thanks.

Giako68 commented 5 years ago

I'm thinking that the configuration file might be different due to the different version of Quartus, but maybe it works correctly. Exactly what error do you have when compiling in Arduino IDE?

g3zbu commented 5 years ago

Hi Giako,

I don't get any compile errors in Arduino now. The Arduino IDE downloads to the Vidor4000. After about 30 seconds the red LED comes on. With your header file linked in the HDMI display comes up and over a few seconds the text is displayed on the screen, but with my header there is no HDMI output at all. Red LED comes on after about 30 seconds, same as with your code.

Al

On 01 January 2019 at 15:24 Giako68 notifications@github.com wrote:

I'm thinking that the configuration file might be different due to the different version of Quartus, but maybe it works correctly.
Exactly what error do you have when compiling in Arduino IDE?

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub https://github.com/Giako68/TEXT_DEMO/issues/1#issuecomment-450737663 , or mute the thread https://github.com/notifications/unsubscribe-auth/AHdqDxfCBqhjZ7mjLN3ysIDbVCFGP7Htks5u-32SgaJpZM4ZlVad .
Giako68 commented 5 years ago

If in the next few days I will have time to do it, I will upgrade to version 18.1 to see if it also has the same problem on Linux.

g3zbu commented 5 years ago

Hi Giako,

I am just about to install 18.0 on this Win10 PC.

This morning's attempt to put Ubuntu 16.04-04 on my i386 laptop failed. It now refuses to boot up which I suspect is because of the defective WiFi driver. And I'm not entirely sure if 18.1 will run in 32-bit mode anyway...conflicting info on Intel's site about that.

Will report back my findings with 18.0 soon. In fact all my attempts to use 18.1 on the GitHub Vidor FPGA projects have ended in failure so trying 18.0 could be interesting.

Al

On 01 January 2019 at 15:35 Giako68 notifications@github.com wrote:

If in the next few days I will have time to do it, I will upgrade to version 18.1 to see if it also has the same problem on Linux.

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub https://github.com/Giako68/TEXT_DEMO/issues/1#issuecomment-450738284 , or mute the thread https://github.com/notifications/unsubscribe-auth/AHdqD-wuUWqke5da2q8NbHYZwB0CTiteks5u-4AqgaJpZM4ZlVad .
g3zbu commented 5 years ago

Hi Giako,

You will be very pleased to hear that 18.0 seems to generate a working header file ;-)

BTW I used its programmer menu to generate the .svf file (1012kb) at 25MHz. Not sure about the frequency but I left it as-is.

The python file skipped the quartus_cpf.exe bit ( files not found, etc, typical Windows issue!) but the RLE section happily converted .svf to .h (686kb).

Conclusion: something not quite right with 18.1...

Thanks for your assistance.

Al

On 01 January 2019 at 15:35 Giako68 notifications@github.com wrote:

If in the next few days I will have time to do it, I will upgrade to version 18.1 to see if it also has the same problem on Linux.

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub https://github.com/Giako68/TEXT_DEMO/issues/1#issuecomment-450738284 , or mute the thread https://github.com/notifications/unsubscribe-auth/AHdqD-wuUWqke5da2q8NbHYZwB0CTiteks5u-4AqgaJpZM4ZlVad .
Giako68 commented 5 years ago

In the meantime I installed version 18.1 on Linux and the resulting .h file is identical between the two versions. It therefore seems that on Linux there are no problems with 18.1.

g3zbu commented 5 years ago

Thanks. I can alter the .v code and confirm that the changes I make do have an effect. I halved the blink rate to prove it.

I will stick to 18.0 for the time being.

Al

On 01 January 2019 at 18:12 Giako68 notifications@github.com wrote:

In the meantime I installed version 18.1 on Linux and the resulting .h file is identical between the two versions. It therefore seems that on Linux there are no problems with 18.1.

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub https://github.com/Giako68/TEXT_DEMO/issues/1#issuecomment-450747472 , or mute the thread https://github.com/notifications/unsubscribe-auth/AHdqD16PNlLV8wwgaFtTzDZUm0LYBluVks5u-6UNgaJpZM4ZlVad .
g3zbu commented 5 years ago

Hi Giako,

I am writing up my project for the Element14 community blog. Can I mention your name?

Ciao,

Al

Giako68 commented 5 years ago

Yes, no problem.

Ciao!

Il giorno dom 27 gen 2019, 14:19 g3zbu notifications@github.com ha scritto:

Hi Giako,

I am writing up my project for the Element14 community blog. Can I mention your name?

Ciao,

Al

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Giako68/TEXT_DEMO/issues/1#issuecomment-457917172, or mute the thread https://github.com/notifications/unsubscribe-auth/AKkQ0PtorO24V4p1N4qifvMHLn63qOwwks5vHadbgaJpZM4ZlVad .