Makefile: Correct parameters for the Orange Crab 85F #336
Conversation
|
And... now with setting the OPENOCD_DEVICE_CONFIG variable correctly. |
|
I believe this might be motivated because the OrangeCrab that @antonblanchard got from @gregdavill, and which was used to write and test the initial makefiles in this repo, is different from OrangeCrab v2.1. See #294 (comment). Hence, maybe it's desirable to support two OrangeCrab boards/variants. |
This is correct. There should probably be a way to support for "non-standard" boards. But I'm not sure how that would look. |
The existing orange crab target is for an older board with a LFE5UM5G-85F device. Newer orange crab boards (v0.21) have a LFE5U-85F device in the -8 speed grade, so make a new target for them called ORANGE-CRAB-0.21. Also add flags to ecppack to indicate that the bitstream should be compressed and can be loaded at 38.8MHz. Signed-off-by: Paul Mackerras <[email protected]>
|
I updated the patch to leave the existing ORANGE-CRAB target as is and add a new one called ORANGE-CRAB-0.21. However, now the checks are failing because the ghdl container image has a bug that causes the error: ERROR: Assert `is_fully_const() && GetSize(chunks_) <= 1' failed in kernel/rtlil.cc:4532. I believe that has been fixed upstream, so hopefully the container images will get updated shortly. |
|
@paulusmack, please, try restarting the CI run. |
|
@umarcor Thanks, restarting did fix it. |
|
FTR, it was not just a restart. I did trigger the image build workflow in hdl/containers before I posted my previous comment. So, Paul was correct. |
The Orange Crab v2.1 with 85k LUT FPGA uses a LFE5U-85F device, not
an LFE5UM5G device. Specifying the wrong device causes the FPGA to
give a configuration error and not load the bitstream. Also add flags
to indicate that the device is the highest speed grade (the -8 grade)
and that the bitstream should be compressed, and can be loaded at
38.8MHz.
Signed-off-by: Paul Mackerras [email protected]