summaryrefslogtreecommitdiff
path: root/README.txt
diff options
context:
space:
mode:
authoruvok2025-12-26 12:52:59 +0100
committeruvok2025-12-26 12:52:59 +0100
commit4eb00254cf157d041ee2ab1babbaa39c2a23443c (patch)
tree0a06373077765b69a83e13bd1fcc0336382ceaba /README.txt
parent2b2e6c244be4f7fc9e18ebdc5030bc34bb406990 (diff)
Update README
Diffstat (limited to 'README.txt')
-rw-r--r--README.txt6
1 files changed, 6 insertions, 0 deletions
diff --git a/README.txt b/README.txt
index 014cea5..fd579e2 100644
--- a/README.txt
+++ b/README.txt
@@ -3,9 +3,15 @@ Learnings:
- Anything that needs to "store" a state must be a reg?
=> wire's can't be assigned in always blocks, yosys complains
- regs must not lead to wires? (unsure where I read that)
+ - https://blog.waynejohnson.net/doku.php/verilog_wire_and_reg
+ -> wires of an outer module can be connecting to an inner module.
+ -> reg can be input to an inner module
+ -> regs can't be outputs
- Clock on the tang9k is 27 MHz
Questions:
- Why do so many examples use always @(posedge clk or nededge rst).
i.e., why is the clk always included?
+ - "async reset" - asynchronous events might be missed, and they
+ don't work well with clocked registers. \ No newline at end of file