![Page 1: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/1.jpg)
Clock Gating for Power Optimization in ASIC Design Cycle:
Theory & Practice
Jairam S, Madhusudan Rao, Jithendra Srinivas, Parimala Vishwanath,
Udayakumar H, Jagdish RaoSoC Center of Excellence, Texas Instruments, India
(sjairam, bgm-rao, jithendra, pari, uday, j-rao) @ti.com
1
![Page 2: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/2.jpg)
JS/BGM – ISLPED08
AGENDA
• Introduction• Combinational Clock Gating
– State of the art– Open problems
• Sequential Clock Gating– State of the art– Open problems
• Clock Power Analysis and Estimation• Clock Gating In Design Flows
2
![Page 3: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/3.jpg)
JS/BGM – ISLPED08
AGENDA
• Introduction• Combinational Clock Gating
– State of the art– Open problems
• Sequential Clock Gating– State of the art– Open problems
• Clock Power Analysis and Estimation• Clock Gating In Design Flows
3
![Page 4: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/4.jpg)
JS/BGM – ISLPED08
Clock Gating Overview
4
![Page 5: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/5.jpg)
JS/BGM – ISLPED08
Clock Gating Overview• System level gating: Turn off entire block disabling all functionality.• Conditions for disabling identified by the designer
4
![Page 6: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/6.jpg)
JS/BGM – ISLPED08
Clock Gating Overview• System level gating: Turn off entire block disabling all functionality.• Conditions for disabling identified by the designer
• Suspend clocks selectively• No change to functionality• Specific to circuit structure • Possible to automate gating atRTL or gate-level
4
![Page 7: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/7.jpg)
JS/BGM – ISLPED08
Clock Network Power
5
![Page 8: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/8.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
5
![Page 9: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/9.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power
5
![Page 10: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/10.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power– Clock Tree dynamic power due to wires
5
![Page 11: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/11.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power– Clock Tree dynamic power due to wires– CLK->Q sequential internal power
5
![Page 12: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/12.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power– Clock Tree dynamic power due to wires– CLK->Q sequential internal power
• Leaf-levels drive the highest capacitance in the tree
5
![Page 13: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/13.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power– Clock Tree dynamic power due to wires– CLK->Q sequential internal power
• Leaf-levels drive the highest capacitance in the tree
5
![Page 14: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/14.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power– Clock Tree dynamic power due to wires– CLK->Q sequential internal power
• Leaf-levels drive the highest capacitance in the tree
• ~80% of the clock network dynamic power is consumed by the leaf driver stage
5
![Page 15: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/15.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power– Clock Tree dynamic power due to wires– CLK->Q sequential internal power
• Leaf-levels drive the highest capacitance in the tree
• ~80% of the clock network dynamic power is consumed by the leaf driver stage− The clock pins of registers are considered as
loads
5
![Page 16: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/16.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power– Clock Tree dynamic power due to wires– CLK->Q sequential internal power
• Leaf-levels drive the highest capacitance in the tree
• ~80% of the clock network dynamic power is consumed by the leaf driver stage− The clock pins of registers are considered as
loads– Leaf cap = wire cap + (constant) pin cap
5
![Page 17: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/17.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power– Clock Tree dynamic power due to wires– CLK->Q sequential internal power
• Leaf-levels drive the highest capacitance in the tree
• ~80% of the clock network dynamic power is consumed by the leaf driver stage− The clock pins of registers are considered as
loads– Leaf cap = wire cap + (constant) pin cap
− Good clustering during synthesis reduces wire-cap
5
![Page 18: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/18.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power– Clock Tree dynamic power due to wires– CLK->Q sequential internal power
• Leaf-levels drive the highest capacitance in the tree
• ~80% of the clock network dynamic power is consumed by the leaf driver stage− The clock pins of registers are considered as
loads– Leaf cap = wire cap + (constant) pin cap
− Good clustering during synthesis reduces wire-cap
Clock network consumes 30-50% ofthe total dynamic power of the chip
5
![Page 19: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/19.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power– Clock Tree dynamic power due to wires– CLK->Q sequential internal power
• Leaf-levels drive the highest capacitance in the tree
• ~80% of the clock network dynamic power is consumed by the leaf driver stage− The clock pins of registers are considered as
loads– Leaf cap = wire cap + (constant) pin cap
− Good clustering during synthesis reduces wire-cap
• Effective clock gating isolates this leaf level buffers and cap, providing large dynamic power savings
Clock network consumes 30-50% ofthe total dynamic power of the chip
5
![Page 20: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/20.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power– Clock Tree dynamic power due to wires– CLK->Q sequential internal power
• Leaf-levels drive the highest capacitance in the tree
• ~80% of the clock network dynamic power is consumed by the leaf driver stage− The clock pins of registers are considered as
loads– Leaf cap = wire cap + (constant) pin cap
− Good clustering during synthesis reduces wire-cap
• Effective clock gating isolates this leaf level buffers and cap, providing large dynamic power savings
• Larger savings with CGs higher up in the tree
Clock network consumes 30-50% ofthe total dynamic power of the chip
5
![Page 21: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/21.jpg)
JS/BGM – ISLPED08
Clock Network Power• Clock network power consists of
– Clock Tree Buffer Power– Clock Tree dynamic power due to wires– CLK->Q sequential internal power
• Leaf-levels drive the highest capacitance in the tree
• ~80% of the clock network dynamic power is consumed by the leaf driver stage− The clock pins of registers are considered as
loads– Leaf cap = wire cap + (constant) pin cap
− Good clustering during synthesis reduces wire-cap
• Effective clock gating isolates this leaf level buffers and cap, providing large dynamic power savings
• Larger savings with CGs higher up in the tree– A trade-off with timing
Clock network consumes 30-50% ofthe total dynamic power of the chip
5
![Page 22: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/22.jpg)
JS/BGM – ISLPED08
Clock Gating and Power consumption
• Power dissipation of a flop due to clock toggles lies in it’s CLK-Q transition power arc
• Disable the clock to a flop when the D pin does not toggle– Disable the CLK-Q arc– Identify all the D Pin non-toggle scenarios
• Can non-toggling of a D-pin be used to find gating scenarios across the clock boundary– Multi Cycle Scenarios
6
![Page 23: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/23.jpg)
D
Clk
Q
Q
Control Logic
AND
LATCHOR
CLK
Scan Enable
EN
Control point for
testability
Main Gate
Could be NAND/AND/NOR/OR
depending on the register style
Gated Clock
Integrated Clock Gate
JS/BGM – ISLPED08
Construction of a Clock Gate
7
![Page 24: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/24.jpg)
JS/BGM – ISLPED08
AGENDA
• Introduction• Combinational Clock Gating
– State of the art– Open problems
• Sequential Clock Gating– State of the art– Open problems
• Clock Power Analysis and Estimation• Clock Gating In Design Flows
8
![Page 25: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/25.jpg)
JS/BGM – ISLPED08
Combinational CG: State-of-the-art - 1• Compile the logic (RTL
or netlist) and detect a structural scenario leading to data gating– Identify Load-enable
registers • Most common is the
mux-feedback loop (MFL) from an output to an input of a flop
• Reduces datapath delay and area
9
![Page 26: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/26.jpg)
JS/BGM – ISLPED08
Combinational CG: State-of-the-art - 2• Identify registers with low data
activity • Additional CGs would cost
area– Grouping registers and
building an XOR tree, introduces a single CG for the group
• To guarantee power reduction, method should be based on placement information– Timing and congestion are
affectedCG
10
![Page 27: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/27.jpg)
JS/BGM – ISLPED08
Combinational CG: Open Problems• Activity driven clock gating
– Clock gating should be done if it helps improve overall power, based on switching activity
– There can exist more than one scenarios that need to be optimized
– Clock gating should not be done for high switching activity registers
• Placement-driven optimisation– Cloning/Merging of clock gates
• Observability Don’t Care– Registers whose outputs are not observable, during a clock
cycle, should be isolated• Leakage/Static Power Impact
– All clock gating techniques should comprehend total power
11
![Page 28: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/28.jpg)
JS/BGM – ISLPED08
An ODC Illustration
12
![Page 29: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/29.jpg)
JS/BGM – ISLPED08
AGENDA
• Introduction• Combinational Clock Gating
– State of the art– Open problems
• Sequential Clock Gating– State of the art– Open problems
• Clock Power Analysis and Estimation• Clock Gating In Design Flows
13
![Page 30: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/30.jpg)
JS/BGM – ISLPED08
Sequential Gating : State-of-the-art - 1
• Ability that can ‘observe’ a logic path beyond a clock-to-clock boundary
• Scenarios– De-Assert a data path if its forward stage is gated – De-Assert forward stage, if the current stage is gated
• Advantages– Apart from sequential power savings, combinational
logic cones can also be gated
14
![Page 31: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/31.jpg)
JS/BGM – ISLPED08
Sequential Gating : State-of-the-art - 2
din
vld
d_1 d_2
dout
vld_1 vld_2Original RTL
Power Optimized RTL
din
vld
d_1 d_2
dout
vld_1 vld_2
CGCGCG
Combinational Analysis
Sequential Analysis
Observability based CG (Backward Traversal)
Source : Mitch Dale, http://www.chipdesignmag.com/display.php?articleId=915
15
![Page 32: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/32.jpg)
JS/BGM – ISLPED08
Sequential Gating : State-of-the-art - 3
f_1 f_2
g_1 g_2
dout
din_1
vld_1
din_2
vld_2
Original RTL
din_1
vld_1
din_2
vld_2
f_2
g_1 g_2
f_1
dout
CG
Power Optimized RTL
Combinational Analysis
Sequential Analysis
CG CG
CG CG
Input-Stability based CG (Forward Traversal)
Source : Mitch Dale, http://www.chipdesignmag.com/display.php?articleId=915
16
![Page 33: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/33.jpg)
JS/BGM – ISLPED08
Sequential Gating: The Next Leap
• Pushing up the abstraction levels– The ESL Platform
• Compilation paradigms for ESL to identify sequential opportunities at RTL
• Power Aware ESL coding styles to ease RTL clock gating
• Verification Requirements : A critical enabler– Alteration to pipelines means a change in functionality– Hence the need to verify the optimized RTL– Formal Approaches gaining precedence over
simulation based methods
17
![Page 34: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/34.jpg)
JS/BGM – ISLPED08
AGENDA
• Introduction• Combinational Clock Gating
– State of the art– Open problems
• Sequential Clock Gating– State of the art– Open problems
• Clock Power Analysis and Estimation • Clock Gating In Design Flows
18
![Page 35: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/35.jpg)
JS/BGM – ISLPED08
Power Estimation Methodology
19
![Page 36: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/36.jpg)
JS/BGM – ISLPED08
Power Estimation Methodology
• Estimation needs to be performed at RTL, netlist and physical design stages
• One constant input at every stage of estimation is the switching profile of the circuit– Ideally, a peak power “testcase” switching profile is desired
for both optimisation and estimation– However, there could be multiple application scenarios which
consume similar power, but with different switching profiles
19
![Page 37: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/37.jpg)
JS/BGM – ISLPED08
Power Estimation Methodology
• Estimation needs to be performed at RTL, netlist and physical design stages
• One constant input at every stage of estimation is the switching profile of the circuit– Ideally, a peak power “testcase” switching profile is desired
for both optimisation and estimation– However, there could be multiple application scenarios which
consume similar power, but with different switching profiles• Switching profiles are derived from simulation of
circuits with appropriate testbenches - costly to do multiple times in the implementation cycle
19
![Page 38: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/38.jpg)
JS/BGM – ISLPED08
Power Estimation Methodology
• Estimation needs to be performed at RTL, netlist and physical design stages
• One constant input at every stage of estimation is the switching profile of the circuit– Ideally, a peak power “testcase” switching profile is desired
for both optimisation and estimation– However, there could be multiple application scenarios which
consume similar power, but with different switching profiles• Switching profiles are derived from simulation of
circuits with appropriate testbenches - costly to do multiple times in the implementation cycle
• Can the source RTL simulation activity for each scenario be used consistently at all stages?
19
![Page 39: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/39.jpg)
JS/BGM – ISLPED08
Capturing Simulation Data
• Will not effect functional profile • CG addition not to effect the Q profile
20
![Page 40: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/40.jpg)
JS/BGM – ISLPED08
Clock Gate Analysis Metrics Formulation
21
![Page 41: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/41.jpg)
JS/BGM – ISLPED08
Clock Gate Analysis Metrics Formulation
• Metric should address the following concerns:
• How good is a current implementation?– Effectiveness of a clock gate
• How much is left on the table?– Granularity of sequential sinks
• How much can be obtained out of the available?– Quality of a gating signal
21
![Page 42: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/42.jpg)
JS/BGM – ISLPED08
Metric Definitions - 1
22
![Page 43: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/43.jpg)
JS/BGM – ISLPED08
Metrics Definitions - 2
• Clock Gating Efficiency (CGE)– Length of time CG is asserted to disable the clock– Average % of time each register is gated
• Data Non-Toggling Ratio (DNT)– Active clock is defined as the percentage time clock
reaches a sequential sink– DNT defined as % time data is non-active for an active
clock• Clustering Efficiency
– Quality of ‘enable’ in proportion to correlation of enable logic to the sequential cluster
23
![Page 44: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/44.jpg)
JS/BGM – ISLPED08
AGENDA
• Introduction• Combinational Clock Gating
– State of the art– Open problems
• Sequential Clock Gating– State of the art– Open problems
• Clock Power Analysis and Estimation• Clock Gating In Design Flows
24
![Page 45: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/45.jpg)
JS/BGM – ISLPED08
Additive CG Gain in RTL2GDSII
• Given the list of available methods, we need a design flow which: – Is additive in power savings– Provides a seamless interface for design tools– Has ability to integrate (and also generate) switching
scenarios at all design stages to enable activity base optimization
– Provides a power estimation framework at all design stages to aid optimization
25
![Page 46: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/46.jpg)
JS/BGM – ISLPED08
CG Flow Sequencing
• RTL Design– Apply sequential gating at RTL design stage– Verify RTL post sequential clock gating– Verify power savings
• Synthesis– Apply combinational clock gating– Apply cluster constraints based of fan-out/bitwidths– Apply CG optimization based on activity
• Physical Design– Validate cluster efficiency based on layout– Add/Refine enable logic, based on cluster refinement
26
![Page 47: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/47.jpg)
JS/BGM – ISLPED08
Results
• Proposed methods were applied to a 65nm data flow centric IP ( ~400K)– A very power sensitive application needing
optimization for different use modes– Optimization was needed to be performed across
multiple use case scenarios• Analysis showed ~40% of total dynamic
consumption in the clock network– Hence scope for power reduction through clock gating
27
![Page 48: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/48.jpg)
JS/BGM – ISLPED08
Incremental Power Savings
# Stage Method Savings1 Synthesis Combinational
(MFL)50%*
2 RTL Sequential 15%
3 Placement IO Exclusivity 6%
4 CTS Cluster Refinement & CTS Implementation
4%
• * Savings reported over a non clock gated design. This can vary across designs
28
![Page 49: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/49.jpg)
JS/BGM – ISLPED08
References
• Automatic synthesis of low-power gated-clock finite-state machines, Benini, L.; De Micheli, G.; IEEE Trans. CAD Volume 15, Issue 6, June 1996 Page(s):630 – 643
• New clock-gating techniques for low-power flip-flops, Strollo, A.G.M.; Napoli, E.; De Caro, D; Proc. ISLPED 2000 Page(s):114 – 119
• DCG: Deterministic clock-gating for low-power microprocessor design, Hai Li; Bhunia, S; Yiran Chen; Roy, K.; Vijaykumar, T.N.;IEEE Trans. VLSI Systems, Volume 12, Issue 3, March 2004 Page(s):245 - 254
• Guarded evaluation: pushing power management to logic synthesis/design; Tiwari, V.; Malik, S.; Ashar, P.; IEEE Transactions on CAD, Volume 17, Issue 10, Oct. 1998 Page(s):1051 - 1060
• Power Compiler Manual, Synopsys Inc.
29
![Page 50: Clock Gating for Power Optimization in ASIC Design · PDF fileClock Gating for Power Optimization in ASIC Design Cycle: Theory & Practice ... JS/BGM – ISLPED08 AGENDA ... –Clock](https://reader031.vdocuments.us/reader031/viewer/2022021418/5a8ef6be7f8b9a7f398d511e/html5/thumbnails/50.jpg)
JS/BGM – ISLPED08
THANK YOU
30