In addition to Jack's suggestion, I think you can get an error like this if your VHDL top level module does not match the name of the VHDL file, so check in the HDL files as well.
On Wed, Nov 14, 2012 at 4:43 AM, Alex Zahn <[email protected]> wrote: > It looks like you're right about the mask script--I had a space in the mask > script name that was causing problems. > > Unfortunately, now EDK fails with an error I can't interpret: > > ERROR:EDK - IPNAME:adc6x1000_sysclk_interface INSTANCE: > simple_trigger_adc6x1000_sysclk - > > /home/betacage/Desktop/casper_startup/working_directory/onboard_clock/simple_ > trigger/XPS_ROACH_base/system.mhs line 279 - cannot find MPD for the > pcore > > Any ideas? > > -Alex > > > On Wed, Nov 14, 2012 at 3:54 AM, Jack Hickish <[email protected]> wrote: >> >> I believe an initialization failure amounts to an error whilst running the >> block mask script. If you run the mask script manually on your new yellow >> block from the matlab command line, if it fails, it should at least tell you >> something useful. >> >> >> On 14 November 2012 09:59, Alex Zahn <[email protected]> wrote: >>> >>> Hi all, >>> >>> I wanted to make a variant of an existing yellow block that we have (the >>> modification amounts to changing a single parameter in the verilog code). To >>> do this, I made copies of all the files in xps_library and xps_lib >>> pertaining to the yellow block and replaced every instance of the old name >>> with the new name in both the file contents and file names. Then I made a >>> Simulink model with the old yellow block, changed its tag parameter, and >>> dragged it into the xps_library model. >>> >>> When I try to compile with the modified yellow block, Simulink throws an >>> "Initialization commands cannot be evaluated" error on the new block, and >>> gives no further debug information. >>> >>> What is the correct procedure for doing this? Is there a way to get any >>> more information on the error out of Simulink? >>> >>> -Alex >> >> >
