You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
However sometimes one may use signals defined from the slave point of view: LINKS_wb_s_o, LINKS_wb_s_i...
The problem appears if those two conventions meet in a single block. That happened to me, and resulted in a difficult to spot error. I had two signals - one connected to the master, and another to the slave. The synthesis tool didn't complain, but the design didn't work.
Therefore I suggest to either use only the signals named from the master perspective, or (maybe better) use suffixes describing the direction: LINKS_wb_m2s, LINKS_wb_s2m.
The text was updated successfully, but these errors were encountered:
I have just lost a significant amount of time in one of AGWB-using designs to fix my stupid mistake caused by inconsistent naming of signals.
We usually use signals like below to name the parts of the Wishbone bus:
agwb/tests/test/hdl/main.vhd
Line 48 in 73625ed
agwb/tests/test/hdl/main.vhd
Line 49 in 73625ed
However sometimes one may use signals defined from the slave point of view:
LINKS_wb_s_o
,LINKS_wb_s_i
...The problem appears if those two conventions meet in a single block. That happened to me, and resulted in a difficult to spot error. I had two signals - one connected to the master, and another to the slave. The synthesis tool didn't complain, but the design didn't work.
Therefore I suggest to either use only the signals named from the master perspective, or (maybe better) use suffixes describing the direction:
LINKS_wb_m2s
,LINKS_wb_s2m
.The text was updated successfully, but these errors were encountered: