diff --git a/doc/html/label.html b/doc/html/label.html index fd01e7d7..60af2f29 100644 --- a/doc/html/label.html +++ b/doc/html/label.html @@ -87,6 +87,50 @@ Place a label in the layout to another layer.
+
+ Attribute labels:++ A handful of labels are referred to as "attribute" labels. These + label types are placed on devices and affect how the device is + extracted.
+ + A label that is placed inside a device (e.g., a MOSFET gate) which + ends with the character "^" is a gate attribute. A + gate attribute in the form of "name=value^" + specifies an extra parameter to be passed to the device in addition + to the standard parameters calculated for that device. This is used + to capture parameters which cannot easily be inferred from the layout. + For example, an RF device model might be distinguished from a non-RF + device model by a parameter such as rfmode=1. Whether or not + a device is intended for RF use is not easily guessed from the layout, + and so "tagging" the gate with the parameter allows the correct model + parameters to be used for the device.
+ + A gate attribute that is not in the form of a parameter will be used + as the device's instance index in the netlist; e.g., a label of + "1^" on a MOSFET gate extracted as a MOSFET device would be an + entry "M1" in the netlist. This can be used to better track + device indexes between a schematic and layout.
+ + A label that is placed on the edge a device, specificlly a + MOSFET gate, and which ends with the character "$", is a + terminal attribute. The only terminal attributes recognized + by magic are S$ and D$, which specify which side of the + gate is to be considered the source and which is to be considered the + drain. Generally, MOSFET devices are symmetric, and their use in a + simulation does not depend on which side is in the position of the + "source" and which is in the position of the "drain". To the extent + that it matters, the terminal attributes can be used to ensure that + the source and drain connections appear in the netlist in their + intended orientation.
+ + Labels ending with "@" are node attributes. There is + currently no functional application for node attributes. When one + is applied, it will appear in the output netlist in a SPICE comment + line indicating the node and attribute. This could be used, say, + by a post-processing script, but as it is in a comment line, it has + no impact on simulation or LVS. +
label is implemented as a built-in command in magic.