Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Questions about standard cells & standard cells library

Status
Not open for further replies.

R1kky

Junior Member level 3
Joined
Nov 3, 2011
Messages
27
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,283
Activity points
1,454
Hi, Friends!

could you, please, help me with several questions regarding std cells & libraries.

1. what it (std cells libraries) can be (i.e. classification on vt,coarse-grain, fine-grain etc)

my suppose - vt could be high or low, also as mentioned coarse- or fine- grain, what else?


2. what track is?

my suppose - tracks mean routing resources, track is the path in which wires can pass through. One track is approximately the minimum spacing between metal1 and metal2 via in a technology node . Track is generally used as a unit to define the height of the std cell. For example, a 12 track cell will be taller than a 9 track cell, a 12 track std cell will be taller, that means more metal1 routing space is available within the cell, hence cells will be faster. where as in a 9 track cell, the cell will be compact, but speed is less compared to 12 track.


3. how many metal level using in std cells?

my suppose - 2 or 3 levels


4. what groups of std cells libraries consist? (for example, filler

my suppose - even don't know, as told there
A standard cell library is a collection of low-level logic functions such as AND, OR, INVERT, flip-flops, latches, and buffers. These cells are realized as fixed-height, variable-width full-custom cells. The key aspect with these libraries is that they are of a fixed height, which enables them to be placed in rows, easing the process of automated digital layout. The cells are typically optimized full-custom layouts, which minimize delays and area.
A typical standard-cell library contains two main components:
Library Database - Consists of a number of views often including layout, schematic, symbol, abstract, and other logical or simulation views. From this, various information may be captured in a number of formats including the Cadence LEF format, and the Synopsys Milkyway format, which contain reduced information about the cell layouts, sufficient for automated "Place and Route" tools.
Timing Abstract - Generally in Liberty format, to provide functional definitions, timing, power, and noise information for each cell.
A standard-cell library may also contain the following additional components:
A full layout of the cells
Spice models of the cells
Verilog models or VHDL Vital models
Parasitic Extraction models
DRC rule decks
An example is a simple XOR logic gate, which can be formed from OR, INVERT and AND gates.
so from that i can point only "logical elements group", what else?

thanks in advance.
 

soloktanjung

Full Member level 6
Joined
Nov 20, 2006
Messages
364
Helped
51
Reputation
100
Reaction score
43
Trophy points
1,308
Location
nowhere
Activity points
3,194
3. how many metal level using in std cells?

my suppose - 2 or 3 levels

The number of metal layers depend on the foundry and the technology (45 nm, 65 nm). I have used 130 nm technology with 6 metal layers.

Thank you.
 
  • Like
Reactions: R1kky

    R1kky

    Points: 2
    Helpful Answer Positive Rating

R1kky

Junior Member level 3
Joined
Nov 3, 2011
Messages
27
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,283
Activity points
1,454
The number of metal layers depend on the foundry and the technology (45 nm, 65 nm). I have used 130 nm technology with 6 metal layers.

Thank you.

thanks for answer, is it correct, that used number of levels of metal for whole SoC and for stnd cell libs is the same number or it differs?
 

soloktanjung

Full Member level 6
Joined
Nov 20, 2006
Messages
364
Helped
51
Reputation
100
Reaction score
43
Trophy points
1,308
Location
nowhere
Activity points
3,194
that used number of levels of metal for whole SoC and for stnd cell libs is the same number or it differs?

I dont quite understand your question. In my design, I have a logic + memory block both using 130 nm and 6 metal layers in one die.

Thank you.
 

oratie

Full Member level 6
Joined
Jan 10, 2007
Messages
348
Helped
176
Reputation
350
Reaction score
170
Trophy points
1,323
Activity points
3,743
Usually, the std cell library contains only Metal1, sometimes it may contain M2. I never saw more than 2 metal layers in the std. cell library.

In addition to multi-Vt libs, exist multi-channel_length std cell libraries (also for leakage optimization).

Groups of cells (I do not know are there standards in naming of such groups):
Simple logic (nand, not, inv, buf, ...)
Complex logic (aoi, oai, 1-bit adder...)
Sequential (flip-flops, latches, scanFF)
Clock (clock buffers, clock inverters, clock-gating cells)
Misc (fillers, decap, tap, antenna diode, bus holder ...)
ECO (spare cells)
Power managetment (power switch, level-shifter, isolation, retention, always-on)
 

jeet_asic

Full Member level 3
Joined
Nov 18, 2011
Messages
156
Helped
20
Reputation
40
Reaction score
19
Trophy points
1,298
Location
Bangalore
Activity points
2,196
Usually, the std cell library contains only Metal1, sometimes it may contain M2. I never saw more than 2 metal layers in the std. cell library.
Thanks sir. While doing a project on Std. cell we were allowed to use only M1 . But when we started layouts for higher drive strength gates , its became almost impossible to route , as M1 was overlapping. So how to come out of this problem and why don't we use M2 layer? Track size was 9.
 

oratie

Full Member level 6
Joined
Jan 10, 2007
Messages
348
Helped
176
Reputation
350
Reaction score
170
Trophy points
1,323
Activity points
3,743
But when we started layouts for higher drive strength gates , its became almost impossible to route , as M1 was overlapping.
I've used std. cell library with x40 drive strength - and M1 was enough to draw such cells.

So how to come out of this problem and why don't we use M2 layer?
Less metal layers in the std cells -> more metal layers for routing -> less chip area. The library, that contains M1+M2, used M2 only for power rails (reduce IR-drop and prevent EM issues).

The answer - if M1 is enough to draw all needed cells without area increasing of these cells - no reason to use additional metal layers inside cells.
 

jeet_asic

Full Member level 3
Joined
Nov 18, 2011
Messages
156
Helped
20
Reputation
40
Reaction score
19
Trophy points
1,298
Location
Bangalore
Activity points
2,196
Less metal layers in the std cells -> more metal layers for routing
Sorry , my bad. But i didn't get this line. Sir can you elaborate this little more.
 

Somashekhar

Member level 2
Joined
Jan 4, 2010
Messages
47
Helped
9
Reputation
18
Reaction score
8
Trophy points
1,288
Location
India
Activity points
1,538
Metal2 usage does not depend on the drive strength of the cell. Complex cells like flop, scan flops and mux require metal2 (7 or 9 track require).

Metal2 usage blocks lot of routing resource when the particular cell is used in the design.
For example : U have used metal2 horizontally to connect two nets in "X" standard cell. And if routing direction of metal2 in the design is vertical.
If the "X" standard cell is used in the design, it blocks lot met2 routing resource.
 

manu.srg

Newbie level 3
Joined
Feb 20, 2010
Messages
3
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
bangalore
Activity points
1,293
How to choose the power rail width while designing a standard cell? I am designing a standard cell library in 180nm (cadence gpdk180). please help me in this regard..,
 

Prashanthanilm

Full Member level 5
Joined
Aug 24, 2012
Messages
303
Helped
36
Reputation
72
Reaction score
36
Trophy points
1,308
Activity points
2,950
Re: Questions about standard cells & standard cells library

As discussed above, the MET2 is used only when MET1 routing is not possible.
Eg: IN High drive cells, MET1 is used as there is plenty of routing area.
Whereas in Flops,Latches and Scan Flops the MET2 usage might be present and it should be very minimal .

- - - Updated - - -

How to choose the power rail width while designing a standard cell? I am designing a standard cell library in 180nm (cadence gpdk180). please help me in this regard..,

Hi Manu,

I have not worked on this.
But, based on the Understanding , Power Rail Width is measured by designing Inverters and running it in desired frequency , temperature and checking the voltage drop in VDD.

Based on the multiple iterations the power rail is chosen .
 

aks0203

Newbie level 1
Joined
Nov 27, 2012
Messages
1
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,290
Thanks sir. While doing a project on Std. cell we were allowed to use only M1 . But when we started layouts for higher drive strength gates , its became almost impossible to route , as M1 was overlapping. So how to come out of this problem and why don't we use M2 layer? Track size was 9.

The reason behind using the lowest possible metal in designing the std.cell is that we do not create congestion for any other proceeding layer. Lets take an example : if i have a freedom of using all the metals available to design my std. cell, all the cells will have all the metals and once they are imported in the design i can't edit it any further. i will encounter problem while making the connections across the cells (one cell to another) also, another major reason is if i can utilize lower metals as much as possible then i would have more than sufficient area to do my routing in higher metals.

Now that the technology is developing things have come so far that we have most of the std.cell library across different technologies that have all their cells made using only metal 1.

in technology like 14nm/16nm we now have a Metal 0 that comes into existence.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top