Extremely long programming time

[langfordpang]

I'm having trouble programming MCS file to flash. Basically, after I run prog-fpga.tcl, it takes more than 8 hrs elapsed time to program the flash, while the CPU time is just a few minutes. I am not even sure if the flash is correctly programmed after 8 hrs because all I have right now is the MCS file from the master branch, which I believe is the same one as default.

I am using CentOS 7.6 and Vivado 2018.3, which I don't believe makes a difference in this case. I thought about other possibilities are USB device busy(?) or JTAG speed too low(?) but none of them make sense. I cannot explain. Please advise.

[langfordpang]

Follow up on this issue -

1. tried another image using different chip ID and confirmed the image can be successfully programmed. But it takes the same amount of time (8+hrs)
2. while building XVCD, it failed to find function ftdi_write_data_async. I searched in the header file and library and no such a symbol. I also failed to find a reliable source online to build myself. So I removed the ASYNC macro. I guess this could be the reason. If it’s true, can you advise how I should get over this? Thank you.

[RHSResearchLLC]

Normal programming time is 20 minutes, which isn't great but better than 8hr. You can try an executable I pre-built for Linux: https://github.com/RHSResearchLLC/xvcd/blob/master/linux/bin/xvcd

It depends on a few libraries:

linux-vdso.so.1 (0x00007fffe7969000)
	libftdi.so.1 => /usr/lib/x86_64-linux-gnu/libftdi.so.1 (0x00007f32c195f000)
	libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007f32c156e000)
	libusb-0.1.so.4 => /lib/x86_64-linux-gnu/libusb-0.1.so.4 (0x00007f32c1365000)
	/lib64/ld-linux-x86-64.so.2 (0x00007f32c1d6b000)

[resinbeard]

As a data point, my programming time takes 28min

[RHSResearchLLC]

Just added https://github.com/RHSResearchLLC/PicoEVB/tree/master/spi-flash-program-openocd. This can program the entire flash in under 5 minutes. The sample project programs in 2 minutes.

[RHSResearchLLC]

You'd replace image.bin with an output from vivado- flash image in binary format

…

On Sat, Feb 5, 2022 at 2:57 PM Maksim Levental ***@***.***> wrote: @RHSResearchLLC <https://github.com/RHSResearchLLC> sorry for the n00b question, but for our own projects, what do we replace in flash.cfg? i.e. here jtagspi_init 0 bscan_spi_xc7a50t.bit jtagspi_program image.bin 0x0 do we replace bscan_spi_xc7a50t.bit or image.bin with the corresponding files produced by vivado? I actually tried both permutations and replacing bscan_spi_xc7a50t.bit got me the same behavior as the example, and the inverse didn't successfully flash. — Reply to this email directly, view it on GitHub <#14 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AG3OOAAEMCEZ2UU4WET6MUDUZV6MLANCNFSM4I2HMJYQ> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you were mentioned.Message ID: ***@***.***>