UVM Interview Questions

Q1: What is UVM? What is the advantage of UVM?

Ans: UVM (Universal Verification Methodology) is a standardized methodology for

verifying the both complex & simple digital design in simple way.

UVM Features:
 First methodology & second collection of class libraries for Automation
 Reusability through testbench
 Plug & Play of verification IPs
 Generic Testbench Development
 Vendor & Simulator Independent
 Smart Testbench i.e. generate legal stimulus as from pre-planned coverage plan
 Support CDV –Coverage Driven Verification
 Support CRV –Constraint Random Verification
 UVM standardized under the Accellera System Initiative
 Register modeling

Q2: UVM derived from which language?

Ans: Here is the detailed connection between SV, UVM, OVM and other methodologies.
Q3. What is the difference between uvm_component and uvm_object?
OR
We already have uvm_object, why do we need uvm_component which is actually
derived class of uvm_object?

Ans:
uvm_component:
 Quasi Static Entity (after build phase it is available throughout the simulation)
 Always tied to a given hardware(DUT Interface) Or a TLM port
 Having phasing mechanism for control the behavior of simulation
 Configuration Component Topology

uvm_object:
 Dynamic Entity (create when needed, transfer from one component to other & then
dereference)
 Not tied to a given hardware or any TLM port
 Not phasing mechanism

Q4: Why phasing is used? What are the different phases in uvm?

Ans: UVM Phases is used to control the behavior of simulation in a systematic way &
execute in a sequential ordered to avoid race condition. This could also be done in system
verilog but manually.
1. List of UVM Phases:
2. buid_phase
3. connect_phase
4. end_of_elaboration_phase
5. start_of_simulation_phase
6. run _phase (task)
Sub Phases of Reset Phase:
pre_reset_phase
reset_phase
post_reset_phase
pre_configure_phase
configure_phase
post_configure_phase
 pre_main_phase
main_phase
post_main_phase
pre_shutdown_phase
shutdown_phase
post_shutdown_phase
7. extract_phase
8. check_phase
9. report_phase
Below figure makes it more clear
Q5: Which uvm phase is top - down , bottom – up & parallel?

Ans: Only build phase is a top-down & other phases are bottom-up except run phase
which is parallel. The build phase works top-down since the testbench hierarchy may be
configure so we need to build the branches before leafs.

Q6: Why build phase is top – down & connect phase is bottom – up?

Ans: The connect phase is intended to be used for making TLM connections between
components, which is why it occur after build phase. It work bottom-up so that its got the
correct implementation all the way up the design hierarchy, if worked top-down this
would be not possible

Q7: Which phase is function & which phase is task?

Ans: Only run phase is a task (time consuming phase) & other phases are functions (non-
blocking)

Q8: Which phase takes more time and why?

Ans: As previously said the run phase is implemented as task and remaining all are
function. run phase will get executed from start of simulation to till the end of simulation.
run phase is time consuming, where the testcase is running.

Q9: How uvm phases initiate?

Ans: UVM phases initiate by calling run_test(“test1”) in top module. When run_test()
method call, it first create the object of test top & then call all phases.

Q10: How test cases run from simulation command line?

Ans: In top module write run_test(); i.e. Don't give anything in argument.
Then in command line : +UVM_TESTNAME=testname

Q11: Difference between module & class based TB?

Ans: A module is a static object present always during of the simulation.

A Class is a dynamic object because they can come and go during the life time of
simulation.
Q12: What is uvm_config_db ? What is difference between uvm_config_db &
uvm_resource_db?

Ans:
 uvm_config_db is a parameterized class used for configuration of different type of
parameter into the uvm database, So that it can be used by any component in the
lower level of hierarchy.

 uvm_config_db is a convenience layer built on top of uvm_resource_db, but that

convenience is very important. In particular, uvm_resource_db uses a "last write
wins" approach. The uvm_config_db, on the other hand, looks at where things are in
the hierarchy up through end_of_elaboration, so "parent wins." Once you start
start_of_simulation, the config_db becomes "last write wins."

 All of the functions in uvm_config_db#(T) are static, so they must be called using
the :: operator. It is extended from the uvm_resource_db#(T), so it is child class of
uvm_resource_db#(T)

Q13: What is the advantage and difference of `uvm_component_utils() and

`uvm_object_utils()?

Ans:
 The utils macros define the infrastructure needed to enable the object/component
for correct factory operation.

 The reason there are two macros is because the factory design pattern fixes the
number of arguments that a constructor can have. Classes derived from uvm_object
have constructors with one argument, a string name. Classes derived from
uvm_component have two arguments, a name and a uvm_component parent.

 The two `uvm_*utils macros inserts code that gives you a factory create() method
that delegates calls to the constructors of uvm_object or uvm_component. You need
to use the respective macro so that the correct constructor arguments get passed
through. This means that you cannot add extra constructor arguments when you
extend these classes in order to be able to use the UVM factory.
Q14: Difference between `uvm_do and `uvm_rand_send ?

Ans: `uvm_do perform the below steps:

1. create
2. start_item
3. randomize
4. finish_item
5. get_response (optional)

while `uvm_rand_send perform all the above steps except create. User needs to create
sequence / sequence_item.

Q15: Difference between uvm_transaction and uvm_seq_item?

Ans: class uvm_sequence_item extends uvm_transaction

uvm_sequence_item extended from uvm_transaction only, uvm_sequence_item class has

more functionality to support sequence & sequencer features. uvm_sequence_item
provides the hooks for sequencer and sequence , So you can generate transaction by using
sequence and sequencer , and uvm_transaction provide only basic methods like do_print
and do_record etc .

Q16: Is uvm is independent of systemverilog ?

Ans: UVM is a methodology based on SystemVerilog language and is not a language on its
own. It is a standardized methodology that defines several best practices in verification to
enable efficiency in terms of reuse and is also currently part of IEEE 1800.2 working
group.

Q17: What are the benefits of using UVM?

Ans: Some of the benefits of using UVM are :

 Modularity and Reusability – The methodology is designed as modular

components (Driver, Sequencer, Agents , env etc) which enables reusing components
across unit level to multi-unit or chip level verification as well as across projects.
 Separating Tests from Testbenches – Tests in terms of stimulus/sequencers are
kept separate from the actual testbench hierarchy and hence there can be reuse of
stimulus across different units or across projects
 Simulator independent – The base class library and the methodology is
supported by all simulators and hence there is no dependence on any specific simulator
 Better control on Stimulus generation – Sequence methodology gives good
control on stimulus generation. There are several ways in which sequences can be
developed which includes randomization, layered sequences, virtual sequences etc which
provides a good control and rich stimulus generation capability.
 Easy configuration – Config mechanisms simplify configuration of objects with
deep hierarchy. The configuration mechanism helps in easily configuring different
testbench components based on which verification environment uses it and without
worrying about how deep any component is in testbench hierarchy
 Factory mechanism – Factory mechanisms simplifies modification of components
easily. Creating each components using factory enables them to be overridden in different
tests or environments without changing underlying code base.

Q18: Can we have user defined phase in UVM?

Ans: In addition to the predefined phases available in uvm , the user has the option to add
his own phase to a component. This is typically done by extending the uvm_phase class
the constructor needs to call super.new which has three arguments
 Name of the phase task or function
 Top down or bottom up phase
 Task or function

The call_task or call_func and get_type_name need to be implemented to complete the

addition of new phase.
Below is a simple example

Example
class custom_phase extends uvm_phase;
function new();
super.new(“custom”,1,1);
endfunction

task call_task ( uvm_component parent);

 my_comp_type comp;
if ( $cast(comp,parent) )
comp.custom_phase();
endtask
virtual function string get_type_name();
return “custom”;
endfunction
endclass

Q19: What is uvm RAL model ? why it is required ?

Ans: In a verification context, a register model (or register abstraction layer) is a set of
classes that model the memory mapped behavior of registers and memories in the DUT in
order to facilitate stimulus generation and functional checking (and optionally some
aspects of functional coverage). The UVM provides a set of base classes that can be
extended to implement comprehensive register modeling capabilities.

Q20: What is the difference between new() and create?

Ans: We all know about new() method that is use to allocate memory to an object
instance. In UVM (and OVM), the create() method causes an object instance to be created
from the factory. This allows you to use factory overrides to replace the desired object
with an object of a different type without having to recode.

Q21: What is analysis port?

Analysis port (class uvm_tlm_analysis_port) — a specific type of transaction-level port

that can be connected to zero, one, or many analysis exports and through which a
component may call the method write implemented in another component, specifically a
subscriber.

port, export, and imp classes used for transaction analysis.

uvm_analysis_port
Broadcasts a value to all subscribers implementing a uvm_analysis_imp.
uvm_analysis_imp
Receives all transactions broadcasted by a uvm_analysis_port.
uvm_analysis_export
Exports a lower-level uvm_analysis_imp to its parent.
Q22: What is TLM FIFO?

In simpler words TLM FIFO is a FIFO between two UVM components, preferably between
Monitor and Scoreboard. Monitor keep on sending the DATA, which will be stored in TLM
FIFO, and Scoreboard can get data from TLM FIFO whenever needed.

// Create a FIFO with depth 4

tlm_fifo = new ("uvm_tlm_fifo", this, 4);

Q23: How sequence starts?

start_item starts the sequence

virtual task start_item ( uvm_sequence_item item,

int set_priority = -1,
uvm_sequencer_base sequencer = null )

start_item and finish_item together will initiate operation of a sequence item. If the item
has not already been initialized using create_item, then it will be initialized here to use the
default sequencer specified by m_sequencer.

Q24: What is the difference between UVM RAL model backdoor write/read and
front door write/read ?

 Front door access means using the standard access mechanism external to the DUT
to read or write to a register. This usually involves sequences of time-consuming
transactions on a bus interface.

 Backdoor access means accessing a register directly via hierarchical reference or

outside the language via the PLI. A backdoor reference usually in 0 simulation time.

Q25: What is objection?

 The objection mechanism in UVM is to allow hierarchical status communication

among components which is helpful in deciding the end of test.
  There is a built-in objection for each in-built phase, which provides a way for
components and objects to synchronize their testing activity and indicate when it is
safe to end the phase and, ultimately, the test end.

 The component or sequence will raise a phase objection at the beginning of an

activity that must be completed before the phase stops, so the objection will be
dropped at the end of that activity. Once all of the raised objections are dropped,
the phase terminates.

Raising an objection: phase.raise_objection(this);

Dropping an objection: phase.drop_objection(this);

Q26: What is p_sequencer ? OR Difference between m_sequencer and p_sequencer?

m_sequencer is the default handle for uvm_vitual_sequencer and p_sequencer is the hook
up for child sequencer.

m_sequencer is the generic uvm_sequencer pointer. It will always exist for the
uvm_sequence and is initialized when the sequence is started.

p_sequencer is a typed-specific sequencer pointer, created by registering the sequence to

the sequencer using macros (`uvm_declare_p_sequencer) . Being type specific, you will be
able to access anything added to the sequencer (i.e. pointers to other sequencers, etc.).
p_sequencer will not exist if we have not registered the sequence with the
`uvm_declare_p_sequencer macros.

The drawback of p_sequencer is that once the p_sequencer is defined, one cannot run the
sequence on any other sequencer type.

Q27: What is the difference between Active mode and Passive mode with respect to
agent?

An agent is a collection of a sequencer, a driver and a monitor.

In active mode, the sequencer and the driver are constructed and stimulus is generated
by sequences sending sequence items to the driver through the sequencer. At the same
time the monitor assembles pin level activity into analysis transactions.
In passive mode, only the monitor is constructed and it performs the same function as in
an active agent. Therefore, your passive agent has no need for a sequencer. You can set up
the monitor using a configuration object.

Q28: What is the difference between copy and clone?

The built-in copy() method executes the __m_uvm_field_automation() method with the
required copy code as defined by the field macros (if used) and then calls the built-
in do_copy() virtual function. The built-in do_copy() virtual function, as defined in the
uvm_object base class, is also an empty method, so if field macros are used to define the
fields of the transaction, the built-in copy() method will be populated with the proper
code to copy the transaction fields from the field macro definitions and then it will execute
the empty do_copy() method, which will perform no additional activity.

The copy() method can be used as needed in the UVM testbench. One common place
where the copy() method is used is to copy the sampled transaction and pass it into a
sb_calc_exp() (scoreboard calculate expected) external function that is frequently used by
the scoreboard predictor.

The clone() method calls the create() method (constructs an object of the same type) and
then calls the copy() method. It is a one-step command to create and copy an existing
object to a new object handle.

Q29: What is UVM factory?

UCM Factory is used to manufacture (create) UVM objects and components. Apart from
creating the UVM objects and components the factory concept essentially means that you
can modify or substitute the nature of the components created by the factory without
making changes to the testbench.

For example, if you have written two driver classes, and the environment uses only one of
them. By registering both the drivers with the factory, you can ask the factory to substitute
the existing driver in environment with the other type. The code needed to achieve this is
minimal, and can be written in the test.

Q30: What are the types of sequencer? Explain each?

There are two types of sequencers :

uvm_sequencer #(REQ, RSP) :
When the driver initiates new requests for sequences, the sequencer selects a sequence
from a list of available sequences to produce and deliver the next item to execute. In order
to do this, this type of sequencer is usually connected to a driver uvm_driver #(REQ, RSP).

uvm_push_sequencer #(REQ, RSP) :

The sequencer pushes new sequence items to the driver, but the driver has the ability to
block the item flow when its not ready to accept any new transactions. This type of
sequencer is connected to a driver of type uvm_push_driver #(REQ, RSP).

Q31. What is the difference between Active mode and Passive mode?
An agent is a collection of a sequencer, a driver and a monitor.

 In active mode, the sequencer and the driver are constructed and stimulus is generated by
sequences sending sequence items to the driver through the sequencer. At the same time the monitor
assembles pin level activity into analysis transactions.

 In passive mode, only the monitor is constructed and it performs the same function as in an active
agent. Therefore, your passive agent has no need for a sequencer. You can set up the monitor using a
configuration object.