beginner_EDA
Full Member level 4
To be honest I have only little experience working with Modelsim. I got stuck writing testbench especially the part which drive input. Therefore I used signal tap but as you say it's not a right approach.Why didn't you run a simulation on the design before implementing it? Debugging a design with signal tap is what you do when the simulation doesn't show any problem, but the errors only occur when the design is implemented.
As mentioned here:
https://www.altera.com/content/dam/altera-www/global/en_US/pdfs/literature/ug/ug_fifo.pdf
"The DCFIFO IP core rdempty output may momentarily glitch when the aclr input is asserted. To
prevent an external register from capturing this glitch incorrectly, ensure that one of the following is true:
• The external register must use the same reset which is connected to the aclr input of the DCFIFO IP
core, or
• The reset connected to the aclr input of the DCFIFO IP core must be asserted synchronous to the
clock which drives the external register.
"
and also in page 17 of same document. Does It have something to do with aclr? In my real design I am also using aclr for Dualclock fifo mixed width.
- - - Updated - - -
or
it is the case in dual clock fifo mixed width, when input data written to FIFO is not exactly multiple/sub multiple of output data.
I mean lets say when Input width is 4 bit and output width is 8 bit and when you write only 20 bits of data which is not multiple of 8. Then only 16 bit data(which is multiple of 8) is read and remaining 4 bit is not read because for last 4 bit rdempty might become high.