Showing posts with label LOW POWER VLSI. Show all posts
Showing posts with label LOW POWER VLSI. Show all posts

Thursday, June 14, 2018

LOW POWER VLSI




     UPF 1.0

 EDA tools have been more stable for UPF 1.0 support, which is very important for product design

Bottom-up hierarchy flow is the only feasible approach in the era of billion gate SoC design

Each block has its own UPF file 

 Easy to manage and reuse UPF file

 Unified Power Format plays a central role in mitigating dynamic and static power in the battle for low-power in advanced process technology.  UPF  is intended to ease the job of specifying, simulating and verifying IC designs that have a number of power states and power islands. UPF is designed to reflect the power intent of a design at a relatively high level.

  UPF scripts describe –   Which power rails should be routed to individual blocks, when blocks are expected to be powered up or shut down how voltage levels should be shifted as signals cross from one power domain to another and    if measures should be taken to retain register and memory-cell contents if the primary power supply to a domain is removed.

  In reality, this comes at the cost of exponentially increasing leakage power. This is because the minimum gate-to-source voltage differential that is needed in CMOS devices to create a conducting path between the source and the drain terminals (known as threshold voltage) has been pushed to its limit.   Leakage power is a function of the threshold voltage, and at smaller device geometries, its contribution to total energy dissipation becomes significant.

 Device supply voltage and leakage current directly contribute to:

    leakage power; 

   switching activity of the capacitive load on supply voltage ;

   its switching frequency contribute to dynamic power.

 It is obvious that reducing supply voltage can be used to control both leakage and dynamic power dissipation. Mainstream power management and reduction techniques are solely based on direct manipulation of voltage, in terms of supply power connectivity and voltage area or power network distributions on the chip.

 IEEE 1801-2015 is the latest version of the power-intent description standard for low-power VLSI design that started life as the Unified Power Format (UPF), and is sometimes referred to as UPF 2.1.

The previous version, IEEE 1801-2009 was also known as UPF 2.0.

The latest version of the standard incorporates elements of the Common Power Format (CPF), which is managed by the Silicon Integration Initiative (Si2).



UPF Components

       Power Domain: – Groups of elements which share a common set of power supply requirements

        Power Supply Network – Abstract description of power distribution (ports, nets, sets & switches)

        Power State Table – The legal combinations of states of each power domain

        Isolation Strategies – How the interface to a power domain should be isolated when its primary power supply is removed !

       Retention Strategies – What registered state in a power domain should be retained when its primary power supply is removed !

       Level Shifter Strategies – How signals connecting power domains operating at different voltages should be shifted !

Repeater Strategies – How domain ports should be bufffered

   Challenges in deploying UPF hierarchy flow in SoCs

  UPF file becomes too big to manage in billion gate SoC era.  The complexity of power state table (PST) increase almost exponential with the number of power domains.  Some RTL coding styles are difficult or not compatible with UPF flow.Ø  Custom design styles requires special attention in UPF flow.

                      UPF   AND LOW POWER

 

1) Why do we need UPF?

2) What is the latest version of vclp tool?

3) Why retention register are used ?

4) What is A0 (always on logic) or always on buffer ?

5) Why we use power switch?

6) What is clamp value ?

7) What is static power and dynamic power ?

8) For what reason we control leakage power ?

9) what is isolation strategy?

10) what is supply net ?

11) what is the flow of UPF?

12) how many and mention the types of retention register?

13) difference b/w vclp v1.0 and vclp v2.0 and which is better?

14) what is level shifter?

15) how many voltage levels are there to define on and off state ?

16) what is the default state in UPF?

17) what is the vclp tool flow?

18) vclp tool is dynamic power verification or static power verification?

19) what is clock gating (CG) and power gating (PG) ?

20) what is freqency scaling in upf ?

21)UPF vs CPF

22)UPF components

23)Challenges faced in deploying UPF in SoC

24)Latest Version of the tool used.Features

25)retention flop hardware

26)UPF command for the power domain

27)Different types of Always on logic

28)What is clock gating?How do you use the same?Hardware.

29)What is power gating?Hardware.

30)Power reduction techniques.

31)Advanced low power techniques.

32)What is crow bar current?causes

33)Different types of level shifter.

34)Different UPF commands.

35)What does set_scope does?

36)UPF for power system table.

37)steps for the environment setup for the upf

38)where are all you declare upf in the RTL To GDS2 flow

39)steps to reduce the compile in UPF 

40) What is PUT?

41)Why do we need PUT?

42)Who should deliver PUT?

43)How does  communication between library design and SoC designer is important to make UPF compliant library  at what conditions?

 44)Write upf for buffer insertion.

 45)UPF for the hierarchy flow.

 46)Write UPF for the SoC.

 47)Low power techniques for SoC

 48)What is the benefit of smaller power switch and bigger power switch?

 49)How do you place  the switch?

 50)DFT implications in low power?

 51)Power gating considerations

 52)Power gating  flow

 53)Parameters captured  in full power gating results.

 54) Parameters captured  in selective power gating results.

 55)Power aware design flow.

 56) Library Requirements for LPD.

 57)Memory retention methods

 58)Single control live slave hardware and working

 59)Dual control baloon and single control baloon

 60)Reduced VDD retention

 61) Diode Biasing

 62)Source Biasing

 63)Difference between the GUPF file and the UPF file.

 64)Different strategies for low power.

 65)Environment setting of the tool for UPF/CPF.

 66)Low power coverage methodology?

 67)What is incrementality elaboration?

68)Integration of low power UPF into the SoC?

 69)set_repeater and set_port_attributes.

 70)rtl optimization for the low power?

 71)Logic optimization for low power?

 72)Layout optimization for low power?

 73)What is zero idle power?

 74)Power management bugs.How to resolve?

 75)What is clamp value 0 and clamp value1?

 76)Tool didn’t understand the port connection defined in RTL through UPF.How to resolve?

 77)X value doesnot exist in emulation how to highlight OFF power domain?

 78)Power sequence failure  in hard macros?

 79)Hard macros interface corruption?

 80)How to corrupt an array which lise in multivoltage domain?

 81)What is meant by power down  and power up sequence in SoC?

 82)UPF 2.0 migration issues?How to resolve it?

 83)Different types of low power checks.explain