UltraScale Architecture

Clocking Resources

User Guide

UG572 (v1.10.2) February 1, 2023

Xilinx is creating an environment where employees, customers,

and partners feel welcome and included. To that end, we’re
removing non-inclusive language from our products and related
collateral. We’ve launched an internal initiative to remove
language that could exclude people or reinforce historical biases,
including terms embedded in our software and IPs. You may still
find examples of non-inclusive language in our older products as
we work to make these changes and align with evolving industry
standards. Follow this link for more information.
Table of Contents
Chapter 1: Overview
Introduction to UltraScale Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Clocking Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Clocking Architecture Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Clocking Differences from Previous FPGA Generations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Chapter 2: Clocking Resources

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Global Clock Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Byte Clock Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Clock Buffers and Clock Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Chapter 3: Clock Management Tile

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
MMCMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
PLLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Dynamic Reconfiguration Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
VHDL and Verilog Templates and the Clocking Wizard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Clocking Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

Appendix A: Additional Resources and Legal Notices

Xilinx Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Solution Centers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Revision History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
Please Read: Important Legal Notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

UltraScale Architecture Clocking Resources Send Feedback

2
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 1

Overview

Introduction to UltraScale Architecture

The Xilinx® UltraScale™ architecture is the first ASIC-class architecture to enable
multi-hundred gigabit-per-second levels of system performance with smart processing,
while efficiently routing and processing data on-chip. UltraScale architecture-based devices
address a vast spectrum of high-bandwidth, high-utilization system requirements by using
industry-leading technical innovations, including next-generation routing, ASIC-like
clocking, 3D-on-3D ICs, multiprocessor SoC (MPSoC) technologies, and new power
reduction features. The devices share many building blocks, providing scalability across
process nodes and product families to leverage system-level investment across platforms.

Virtex® UltraScale+™ devices provide the highest performance and integration capabilities
in a FinFET node, including both the highest serial I/O and signal processing bandwidth, as
well as the highest on-chip memory density. As the industry's most capable FPGA family,
the Virtex UltraScale+ devices are ideal for applications including 1+Tb/s networking and
data center and fully integrated radar/early-warning systems.

Virtex UltraScale devices provide the greatest performance and integration at 20 nm,
including serial I/O bandwidth and logic capacity. As the industry's only high-end FPGA at
the 20 nm process node, this family is ideal for applications including 400G networking,
large scale ASIC prototyping, and emulation.

Kintex® UltraScale+ devices provide the best price/performance/watt balance in a FinFET

node, delivering the most cost-effective solution for high-end capabilities, including
transceiver and memory interface line rates as well as 100G connectivity cores. Our newest
mid-range family is ideal for both packet processing and DSP-intensive functions and is well
suited for applications including wireless MIMO technology, Nx100G networking, and data
center.

Artix® UltraScale+ devices provide high serial bandwidth and signal compute density in a
cost-optimized device for critical networking applications, vision and video processing, and
secured connectivity. Coupled with the innovative InFO packaging, which provides excellent
thermal and power distribution, Artix UltraScale+ devices are perfectly suited to
applications requiring high compute density in a small footprint.

Kintex UltraScale devices provide the best price/performance/watt at 20 nm and include

the highest signal processing bandwidth in a mid-range device, next-generation

UltraScale Architecture Clocking Resources Send Feedback

3
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 1: Overview

transceivers, and low-cost packaging for an optimum blend of capability and

cost-effectiveness. The family is ideal for packet processing in 100G networking and data
centers applications as well as DSP-intensive processing needed in next-generation medical
imaging, 8k4k video, and heterogeneous wireless infrastructure.

Zynq® UltraScale+ devices provide 64-bit processor scalability while combining real-time
control with soft and hard engines for graphics, video, waveform, and packet processing.
Integrating an Arm®-based system for advanced analytics and on-chip programmable
logic for task acceleration creates unlimited possibilities for applications including 5G
Wireless, next generation ADAS, and industrial Internet-of-Things.

This user guide describes the UltraScale architecture clocking resources and is part of the
UltraScale architecture documentation suite available at: www.xilinx.com/ultrascale.

Clocking Overview
This chapter provides an overview of clocking and a comparison between clocking in the
UltraScale architecture and previous FPGA generations. For detailed information on usage
of clocking resources, see Chapter 2, Clocking Resources and Chapter 3, Clock Management
Tile. For more information refer to the Clocking Guidelines section in the UltraFast Design
Methodology Guide for the Vivado Design Suite (UG949) [Ref 1].

Clocking Architecture Overview

The UltraScale architecture clocking resources manage complex and simple clocking
requirements with dedicated global clocks distributed on clock routing and clock
distribution resources. The clock management tiles (CMTs) provide clock frequency
synthesis, deskew, and jitter filtering functionality. Non-clock resources such as local
routing are not recommended when designing for clock functions.

• The device is subdivided into columns and rows of segmented clock regions (CRs). CRs
differ from previous families because they are arranged in tiles and do not span half the
width of a device. A CR contains configurable logic blocks (CLBs), DSP slices,
block RAMs, interconnect, and associated clocking. The height of a CR is 60 CLBs,
24 DSP slices, and 12 block RAMs with a horizontal clock spine (HCS) at its center. The
HCS contains the horizontal routing and distribution resources, leaf clock buffers, clock
network interconnections, and the root of the clock network. Clock buffers drive
directly into the HCS. There are 52 I/Os per bank and four gigabit transceivers (GTs)
that are pitch matched to the CRs. A core column contains configuration, System
Monitor (SYSMON), and PCIe® blocks to complete a basic device.
• Adjacent to the input/output block columns are the physical layer (PHY) blocks with
CMTs, global clock buffers, global clock multiplexing structures, and I/O logic

UltraScale Architecture Clocking Resources Send Feedback

4
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 1: Overview

management functions. The clocking drives vertical and horizontal connectivity

through separate clock routing and clock distribution resources via HCS into the CRs
and I/Os.
• Horizontal clock routing and distribution tracks drive horizontally into the CRs. Vertical
routing and distribution tracks drive vertically adjacent CRs. The tracks are
segmentable at the CR boundaries in both the horizontal and vertical directions. This
allows for the creation of device-wide global clocks or local clocks of variable size.
• The distribution tracks drive the clocking of synchronous elements across the device.
Distribution tracks are driven by routing tracks or directly by the clocking structures in
the PHY.
• I/Os are directly driven from the PHY clocking and/or an adjacent PHY via routing
tracks.
• A CMT contains one mixed-mode clock manager (MMCM) and two phase-locked loops
(PLLs).

Clock Routing Resources Overview

Each I/O bank contains global clock input pins to bring user clocks onto the device clock
management and routing resources. The global clock inputs bring user clocks onto:

• Clock buffers in the PHY adjacent to the same bank

• CMTs in the PHY adjacent to the same bank

Each device has three global clock buffers: BUFGCTRL, BUFGCE, and BUFGCE_DIV. In
addition, there is a local BUFCE_LEAF clock buffer for driving leaf clocks from horizontal
distribution to various blocks in the device. BUFGCTRL has derivative software
representations of types BUFGMUX, BUFGMUX1, BUFGMUX_CTRL, and BUFGCE_1. BUFGCE
is for glitchless clock gating and has software derivative BUFG (BUFGCE with clock enable
tied High). The global clock buffers drive routing and distribution tracks into the device
logic via HCS rows. There are 24 routing and 24 distribution tracks in each HCS row. There
is also a BUFG_GT that generates divided clocks for GT clocking. The clock buffers:

• Can be used as a clock enable circuit to enable or disable clocks either globally, locally,
or within a CR for fine-grained power control.
• Can be used as a glitch-free multiplexer to:

° select between two clock sources.

° switch away from a failed clock source.

• Are often driven by a CMT to:

° eliminate the clock distribution delay.

° adjust clock delay relative to another clock.

UltraScale Architecture Clocking Resources Send Feedback

5
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 1: Overview

Chapter 2, Clocking Resources, has further details on global clocks, I/O, and GT clocking. It
also describes which clock routing resources to utilize for various applications.

CMT Overview
Each device has a CMT as part of the PHY next to each of the I/O banks. A CMT consists of
one MMCM and two PLLs. The MMCM is the primary block for frequency synthesis for a
wide range of frequencies, and serves as a jitter filter for either external or internal clocks,
and deskew clocks among a wide range of other functions. The PLL’s primary purpose is to
provide clocking to the PHY I/Os, but can also be used for clocking other resources in the
device in a limited fashion. The device clock input connectivity allows multiple resources to
provide the reference clock(s) to the MMCM and PLL.

MMCMs have infinite fine phase shift capability in either direction and can be used in
dynamic phase shift mode. MMCMs also have a fractional counter in either the feedback
path or in one output path, enabling further granularity of frequency synthesis capabilities.

The LogiCORE™ IP clocking wizard is available to assist in utilizing MMCMs and PLLs to
create clock networks in UltraScale architecture designs. The GUI interface is used to collect
clock network parameters. The clocking wizard chooses the appropriate CMT resource and
optimally configures the CMT resource and associated clock routing resources.

Chapter 3, Clock Management Tile, has further details on the CMT block features and
connectivity.

Clocking Differences from Previous FPGA

Generations
UltraScale architecture-based devices have significant innovations in the clocking
architecture. In general, there is a minimal difference between global and local clock
buffers. Thus, the 7 series regional clock buffers have been replaced by new clock buffers
with more global reach while automatically utilizing local clock buffers for local distribution
of the clocks. The CMT block consists of one MMCM and two PLLs. The MMCM is very
similar to the 7 series family while the PLL has new features for the I/O PHY clocking, but a
reduced set of functionality and connectivity with respect to clocking the rest of the device.

UltraScale Architecture Clocking Resources Send Feedback

6
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 1: Overview

Key Differences from 7 Series FPGAs

• BUFMRs, BUFRs, and BUFIOs, and the associated routing resources have been removed
from this architecture and are replaced by new clock buffers, clock routing, and a
completely new I/O clocking architecture.
• The BUFGCTRL and its derivatives are still available. Two new global clock buffer
resources BUFGCE and BUFGCE_DIV have been introduced in the new architecture. At
the local clocking level, a new BUFCE_LEAF clock buffer provides local, vertical clocking
with additional features.
• A BUFG_GT buffer for clock division of GT clocks has been added.
• A new and improved clock routing architecture is available. There are now two types of
global routing tracks called routing and distribution. Both types of routing provide a
segmentable clock network at the CR level. Both types can be driven by the global
clock buffers. The distribution tracks can be driven by routing tracks or directly by clock
buffer resources. The distribution tracks provide connectivity to all clocking points in
UltraScale devices.
• The CMTs now have two PLLs instead of one.
• MMCMs are similar to the MMCM in the 7 series devices. PLLs have new features
related to I/O PHY clocking. However, other clocking related functionality and
connectivity has been reduced as compared to the 7 series FPGAs. For example, the
PLLs do not support phase compensation or external feedback, have fewer outputs,
share a voltage-controlled oscillator (VCO) with the PHY clocking, and have other
features removed as compared to the 7 series devices. For this reason, most customers
should use the MMCM for general clocking. However, leftover PLLs are also available
for use.
• The MMCM output clock frequencies can be dynamically changed without resetting the
MMCM.
• The definition of clock region has changed. A clock region no longer spans half a
device width in the horizontal direction. UltraScale architecture clock regions have a
rectangular shape with a fixed width and height and are organized in tiles. Horizontal
and vertical clock tracks are segmented at the clock region boundaries.
• The clock capable pins (CC) have been replaced by global clock pins (GC). In addition,
the UltraScale+ architecture has high-density (HD) I/O banks. These banks contain four
global clock pins called HDGC which can connect to the BUFGCEs.

UltraScale Architecture Clocking Resources Send Feedback

7
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2

Clocking Resources

Overview
UltraScale™ architecture-based devices have several clock routing resources to support
various clocking schemes and requirements, including high fanout, short propagation
delay, and extremely low skew. To best utilize the clock routing resources, the designer must
understand how to get user clocks from the PCB to the UltraScale devices, decide which
clock routing resources are optimal, and then access those clock routing resources by
utilizing the appropriate I/O and clock buffers.

Global Clock Inputs

External global user clocks must be brought into the UltraScale device on differential clock
pin pairs called global clock (GC) inputs. There are four GC pin pairs in each bank that have
direct access to the global clock buffers, MMCMs, and PLLs that are in the CMT adjacent to
the same I/O bank. The UltraScale+ architecture has four HDGC pins per HD I/O bank. HD
I/O banks are only part of the UltraScale+ family. Since HD I/O banks do not have a XIPHY
and CMT next to them, the HDGC pins can only directly drive BUFGCEs (BUFGs) and not
MMCMs/PLLs. Therefore, clocks that are connected to an HDGC pin can only connect to
MMCMs/PLLs through the BUFGCEs. To avoid a design rule check (DRC) error, set the
property CLOCK_DEDICATED_ROUTE = FALSE. GC inputs provide dedicated, high-speed
access to the internal global and regional clock resources. GC inputs use dedicated routing
and must be used for clock inputs where the timing of various clocking features is
imperative. General-purpose I/O with local interconnects should not be used for clock
signals.

Each I/O bank is located in a single clock region and includes 52 I/O pins. Of the 52 I/O pins
in each I/O bank in every I/O column, there are four global clock input pin pairs (a total of
eight pins). Each global clock input:

• Can be connected to a differential or single-ended clock on the PCB.

• Can be configured for any I/O standard, including differential I/O standards.
• Has a P-side (master), and an N-side (slave).

UltraScale Architecture Clocking Resources Send Feedback

8
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

Single-ended clock inputs must be assigned to the P (master) side of the GC input pin pair.
If a single-ended clock is connected to the P-side of a differential clock pin pair, the N-side
cannot be used as another single-ended clock pin—it can only be used as a user I/O. For pin
naming conventions, refer to the UltraScale Architecture Packaging and Pinout User Guide
(UG575) [Ref 2].

GC inputs can be used as regular I/O if not used as clocks. When used as regular I/O, global
clock input pins can be configured as any single-ended or differential I/O standard. GC
inputs can connect to the PHY adjacent to the banks they reside in.

Byte Clock Inputs

Byte-lane clock (DBC and QBC) input pin pairs are dedicated clock inputs directly driving
source synchronous clocks to the bit slices in the I/O banks. In memory applications, these
are also known as DQS. When not used for I/O byte clocking these pin have other functions
such as general purpose I/Os. For more information, consult the UltraScale Architecture
SelectIO Resources User Guide (UG571) [Ref 3].

Clock Buffers and Clock Routing

Global clocks are a dedicated network of interconnects specifically designed to reach all
clock inputs to the various resources in a device. These networks are designed to have low
skew and low duty cycle distortion, low power, and improved jitter tolerance. They are also
designed to support very high-frequency signals.

Understanding the signal path for a global clock expands the understanding of the various
global clocking resources. The global clocking resources and network consist of these paths
and components:

• Clock Structure, page 10

• Clock Buffers, page 14
• BUFGCTRL Clock Buffer Primitives, page 15
• Additional BUFGCTRL Use Models, page 23
• BUFGCE Clock Buffers, page 27
• BUFG Clock Buffer, page 28
• BUFCE_LEAF Clock Buffer, page 28

UltraScale Architecture Clocking Resources Send Feedback

9
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

Clock Structure
The basic device architecture is composed of blocks of CRs. CRs are organized into tiles and
thus build columns and rows. Each CR contains slices (CLBs), DSPs, and 36K block RAM
blocks. The mix of slice, DSP, and block RAM columns in each CR can be different, but are
always identical when stacked in the vertical direction, thus building columns of those
resources for the entire device. I/O and GT columns are then inserted with columns of CRs.
In addition, there is a single column that contains the configuration logic, SYSMON, and
PCIe blocks. An HCS runs horizontally through the device in the center of each row of CRs,
I/Os, and GTs. The HCS contains the horizontal routing and distribution tracks as well as leaf
clock buffers and clock network interconnects between horizontal/vertical routing and
distribution. Vertical tracks of routing and distribution connect all CRs in a column, while
vertical routing spans an entire I/O column. There are 24 horizontal routing and 24
distribution tracks (Figure 2-1), and 24 vertical routing and 24 distribution tracks
(Figure 2-2). The purpose of the clock routing resources is to route a clock from the global
clock buffers to a central point from where it is connected to the loads via the distribution
resources. This central point of the clock network is called a clock root in the UltraScale
architecture. The root can be in any CR in a device from where it is routed to the loads via
the clock distribution resources. This architecture optimized clock skew. Routing and
distribution resources can either connect to adjacent CRs or disconnect (isolated) at the
border of the CR as needed. This concept extends to SSI devices as well.

UltraScale Architecture Clocking Resources Send Feedback

10
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

X-Ref Target - Figure 2-1

PHY and
CR Column with PCIe,
Clocking
Configuration, and
SYSMON Column
I/O Column CR Column I/O Column CR Column GT Column

HCS

24 Distribution Tracks

24 Routing Tracks
X16662-111516

Figure 2-1: Horizontal Clocking

UltraScale Architecture Clocking Resources Send Feedback

11
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

X-Ref Target - Figure 2-2

PHY and
CR Column with PCIe,
Clocking
Configuration, and
SYSMON Column
I/O Column CR Column I/O Column CR Column GT Column

24 Routing Tracks 24 Distribution Tracks 12 Distribution and

12 Routing Tracks
on each side
X16663-111516

Figure 2-2: Vertical Clocking

The clocks can be distributed from their sources in one of two ways (Figure 2-3):

• The clocks can go onto routing tracks that take the clocks to a central point in a CR
without going to any loads. The clocks can then drive the distribution tracks
unidirectionally from which the clock networks fan out. In this way, the clock buffers
can drive to a specific point in the CRs from which the clock buffers travel vertically and
then horizontally on the distribution tracks to drive the clocking points. The clocking

UltraScale Architecture Clocking Resources Send Feedback

12
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

points are driven via leaf clocks with clock enable (CE) in that CR and adjacent CRs, if
needed. Distribution tracks cannot drive routing tracks.

This distribution scheme is used to move the root for all the loads to be at a specific
location for improved, localized skew. Furthermore, both routing and distribution tracks
can drive into horizontally or vertically adjacent CRs in a segmented fashion. Routing
tracks can drive both routing and distribution tracks in the adjacent CRs while the
distribution tracks can drive other horizontal distribution tracks in adjacent CRs. The CR
boundary segmentation allows construction of either truly global, device-wide clock
networks or more local clock networks of variable sizes by reusing clocking tracks.

• Alternatively, clock buffers can drive straight onto the distribution tracks and distribute
the clock in that manner. This reduces the clock insertion delay.
X-Ref Target - Figure 2-3

Vertical From/To Vertical

Routing CR Above Distribution

CR Boundary

Columns of Columns of
CLBs, Block RAMs, 24 CLBs, Block RAMs,
24
DSPs DSPs

Horizontal BUFCE_LEAFs
Distribution CE CE CE CE

24
From/To Next TS From/To Next
CR or From Root CE
HCS CR or From
Clock Buffers Clock Buffers
TS
24

Horizontal CE CE CE CE
BUFCE_LEAFs
Routing

Columns of Columns of
CLBs, Block RAMs, CLBs, Block RAMs,
DSPs DSPs

From/To
CR Below

X16681-111516

Figure 2-3: Clock Region Clocking

• Each of the four bytes in the XIPHY BITSLICE have six connections from the HCS to their
global clocking pins. Therefore, only six BUFGs can drive the BITSLICE clocking pins in
either half of an I/O bank (a maximum of 6 clocks can drive any half of an I/O bank).

UltraScale Architecture Clocking Resources Send Feedback

13
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

Clock Buffers
The PHY global clocking contains several sets of BUFGCTRLs, BUFGCEs, and BUFGCE_DIVs.
Each set can be driven by four GC pins from the adjacent bank, MMCMs, PLLs in the same
PHY, and interconnect. The clock buffers then drive the routing and distribution resources
across the entire device. Each PHY contains 24 BUFGCEs, 8 BUFGCTRLs, and 4 BUFGCE_DIVs
but only 24 of them can be used at the same time.

IMPORTANT: It is recommended to only allow the Vivado® Placer to assign all global clock buffers to
specific locations. Each CR contains 24 BUFGCEs, 8 BUFGCTRLs and 4 BUFGCE_DIVs. These clock
buffers share the 24 routing tracks and therefore collisions may occur resulting in unroutable designs.
If the design requires a number of global clock buffers to be in a certain CR then it is recommended to
attach the CLOCK_REGION property to these buffers instead of a specific LOCATION property.

In the clocking architecture, BUFGCTRL multiplexers and all derivatives can be cascaded to
adjacent clock buffers, effectively creating a ring of eight BUFGMUXes (BUFGCTRL
multiplexers). Figure 2-4 shows a simplified diagram of cascading BUFGCTRLs.
X-Ref Target - Figure 2-4

X16664-111516

Figure 2-4: Cascading BUFGCTRLs

The following subsections detail the various configurations, primitives, and use models of
the clock buffers.

UltraScale Architecture Clocking Resources Send Feedback

14
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

BUFGCTRL Clock Buffer Primitives

The primitives in Table 2-1 are different configurations of the clock BUFGCTRL buffers. The
Vivado tools manage the configuration of all these primitives, and the Vivado Design Suite
User Guide: Using Constraints (UG903) [Ref 4] describes the LOC constraint.

Table 2-1: BUFGCTRL Clock Buffer Primitives

Primitive Input Output Control
BUFGCTRL I0, I1 O CE0, CE1, IGNORE0, IGNORE1, S0, S1
BUFGCE_1 I O CE
BUFGMUX I0, I1 O S
BUFGMUX_1 I0, I1 O S
BUFGMUX_CTRL I0, I1 O S

BUFGCTRL
The BUFGCTRL primitive shown in Figure 2-5 can switch between two asynchronous clocks.
All other global clock buffer primitives are derived from certain configurations of
BUFGCTRL.

BUFGCTRL has four select lines, S0, S1, CE0, and CE1. It also has two additional control lines,
IGNORE0 and IGNORE1. These six control lines are used to control the inputs I0 and I1.
X-Ref Target - Figure 2-5

BUFGCTRL

IGNORE1

CE1

S1

I1

I0

S0

CE0

IGNORE0
X16665-111516

Figure 2-5: BUFGCTRL Primitive

UltraScale Architecture Clocking Resources Send Feedback

15
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

BUFGCTRL is designed to switch between two clock inputs without the possibility of a glitch.
When the presently selected clock transitions from High to Low after S0 and S1 change, the
output is kept Low until the other (to-be-selected) clock transitions from High to Low. Then,
the new clock starts driving the output.The default configuration for BUFGCTRL is
falling-edge sensitive and held at Low prior to the input switching. BUFGCTRL can also be
rising-edge sensitive and held at High prior to the input switching by using the INIT_OUT
attribute.

In some applications, the conditions previously described are not desirable. Asserting the
IGNORE pins bypasses the BUFGCTRL from detecting the conditions for switching between
two clock inputs. In other words, asserting IGNORE causes the MUX to switch the inputs at
the instant the select pin changes. IGNORE0 causes the output to switch away from the I0
input immediately when the select pin changes, while IGNORE1 causes the output to switch
away from the I1 input immediately when the select pin changes.

Selection of an input clock requires a “select” pair (S0 and CE0, or S1 and CE1) to be
asserted High. If either S or clock enable (CE) is not asserted High, the desired input is not
selected. In normal operation, both S and CE pairs (all four select lines) are not expected to
be asserted High simultaneously. Typically, only one pin of a “select” pair is used as a select
line, while the other pin is tied High. The truth table is shown in Table 2-2.

Table 2-2: Truth Table for Clocking Resources

CE0 S0 CE1 S1 O
1 1 0 X I0
1 1 X 0 I0
0 X 1 1 I1
X 0 1 1 I1
1 1 1 1 Old Input (1)

Notes:
1. Old input refers to the valid input clock before this state is achieved.
2. For all other states, the output becomes the value of INIT_OUT and does not toggle.

Although both S and CE are used to select a desired output, only S is suggested for
glitch-free switching. This is because when using CE to switch clocks, the change in clock
selection can be faster than when using S. A violation in the setup/hold time of the CE pins
causes a glitch at the clock output. On the other hand, using the S pins allows the user to
switch between the two clock inputs without regard to setup/hold times. As a result, using
S to switch clocks does not result in a glitch. See BUFGMUX_CTRL, page 22.

The timing diagram in Figure 2-6 illustrates various clock switching conditions using the
BUFGCTRL primitives. Exact timing numbers are best found using the speed specification.

UltraScale Architecture Clocking Resources Send Feedback

16
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

X-Ref Target - Figure 2-6

1 2 3 4 5 6
I0

I1
TBCCKK_CE
CE0

CE1

S0

S1

IGNORE0

IGNORE1
TBCCKO_O TBCCKO_O TBCCKO_O

at I0 Begin I1 Begin I0
X16666-111516

Figure 2-6: BUFGCTRL Timing Diagram

• Before time event 1, output O uses input I0.

• At time TBCCCK_CE, before the rising edge at time event 1, both CE0 and S0 are
deasserted Low. At about the same time, both CE1 and S1 are asserted High.
• At time TBCCKO_O, after time event 3, output O uses input I1. This occurs after a
High-to-Low transition of I0 (event 2) followed by a High-to-Low transition of I1.
• At time event 4, IGNORE1 is asserted.
• At time event 5, CE0 and S0 are asserted High while CE1 and S1 are deasserted Low. At
TBCCKO_O, after time event 6, output O has switched from I1 to I0 without requiring a
High-to-Low transition of I1.

Other capabilities of BUFGCTRL are:

• Pre-selection of the I0 and I1 inputs are made after configuration but before device
operation.
• The initial output after configuration can be selected as either High or Low.
• Clock selection using CE0 and CE1 only (S0 and S1 tied High) can change the clock
selection without waiting for a High-to-Low transition on the previously selected clock.

UltraScale Architecture Clocking Resources Send Feedback

17
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

Table 2-3 summarizes the attributes for the BUFGCTRL primitive.

Table 2-3: BUFGCTRL Attributes

Attribute Name Description Possible Values
INIT_OUT Initializes the BUFGCTRL output to the specified value after 0 (default), 1
configuration. Sets the positive or negative edge behavior. Sets
the output level when changing clock selection.
PRESELECT_I0 If TRUE, BUFGCTRL output uses the I0 input after FALSE (default), TRUE
configuration.(1)
PRESELECT_I1 If TRUE, BUFGCTRL output will use the I1 input after FALSE (default), TRUE
configuration.(1)

Notes:
1. Both PRESELECT attributes cannot be TRUE at the same time.

BUFGCE_1
BUFGCE_1 is a clock buffer with one clock input, one clock output, and a clock enable line.
This primitive is based on BUFGCTRL with some pins connected to logic High or Low.
Figure 2-7 illustrates the relationship of BUFGCE_1 and BUFGCTRL. The LOC constraint is
available for manually placing the BUFGCE_1 location. See the Vivado Design Suite User
Guide: Using Constraints (UG903) [Ref 4] for more information.
X-Ref Target - Figure 2-7

BUFGCE_1 as BUFGCTRL
IGNORE1
VDD
CE1
GND
BUFGCE_1 GND S1
CE

I1
VDD
O O
I
I0
I

VDD S0
CE CE0
GND IGNORE0
X16668-111516

Figure 2-7: BUFGCE_1 as BUFGCTRL

The switching condition for BUFGCE_1 is similar to BUFGCTRL with INIT_OUT set to 1. If the
CE input is Low prior to the incoming falling clock edge, the following clock pulse does not
pass through the clock buffer, and the output stays High. Any level change of CE during the
incoming clock Low pulse has no effect until the clock transitions High. The output stays
High when the clock is disabled. However, when the clock is being disabled, it completes
the clock Low pulse.

UltraScale Architecture Clocking Resources Send Feedback

18
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

IMPORTANT: Because the clock enable line uses the CE pin of the BUFGCTRL, the select signal must
meet the setup time requirement. Violating this setup time can result in a glitch.

Figure 2-8 illustrates the timing diagram for BUFGCE_1.

X-Ref Target - Figure 2-8

BUFGCE_1(I)
TBCCCK_CE
BUFGCE_1(CE)

BUFGCE_1(O)

TBCCKO_O
X16669-111516

Figure 2-8: BUFGCE_1 Timing Diagram

BUFGMUX and BUFGMUX_1

BUFGMUX is a clock buffer with two clock inputs, one clock output, and a select line. This
primitive is based on BUFGCTRL with some pins connected to logic High or Low.

Figure 2-9 illustrates the relationship of BUFGMUX and BUFGCTRL. The LOC constraint is
available for manually placing the BUFGMUX and BUFGCTRL locations. See the Vivado
Design Suite User Guide: Using Constraints (UG903) [Ref 4] for more information.
X-Ref Target - Figure 2-9

IGNORE1
GND
S CE1

S1
VDD

BUFGMUX

I1 I1

O O
I0
I0

S
S0
VDD
CE0

IGNORE0
GND
X16670-111516

Figure 2-9: BUFGMUX as BUFGCTRL

UltraScale Architecture Clocking Resources Send Feedback

19
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

IMPORTANT: Because BUFGMUX uses the CE pins as select pins, when using the select, the setup time
requirement must be met. Violating this setup time can result in a glitch.

Switching conditions for BUFGMUX are the same as the CE pins on BUFGCTRL. Figure 2-10
illustrates the timing diagram for BUFGMUX.
X-Ref Target - Figure 2-10

TBCCCK_CE

I0

I1

O
TBCCKO_O
begin TBCCKO_O
switching using I1
X16671-111516

Figure 2-10: BUFGMUX Timing Diagram

In Figure 2-10:

• The current clock is I0.

• S is activated High.
• If I0 is currently High, the multiplexer waits for I0 to deassert Low.
• After I0 is Low, the multiplexer output stays Low until I1 transitions from High to Low.
• When I1 transitions from High to Low, the output switches to I1.
• If setup/hold times are met, no glitches or short pulses can appear on the output.

BUFGMUX_1 is rising-edge sensitive and held at High prior to input switch. Figure 2-11
illustrates the timing diagram for BUFGMUX_1. The LOC constraint is available for manually
placing the BUFGMUX and BUFGMUX_1 locations. See the Vivado Design Suite User Guide:
Using Constraints (UG903) [Ref 4] for more information.

UltraScale Architecture Clocking Resources Send Feedback

20
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

X-Ref Target - Figure 2-11

TBCCCK_CE

I0

I1

TBCCKO_O

X16672-111516

Figure 2-11: BUFGMUX_1 Timing Diagram

In Figure 2-11:

• The current clock is I0.

• S is activated High.
• If I0 is currently Low, the multiplexer waits for I0 to be asserted High.
• After I0 is High, the multiplexer output stays High until I1 transitions from Low to High.
• When I1 transitions from Low to High, the output switches to I1.
• If setup/hold times are met, no glitches or short pulses can appear on the output.

Table 2-4 summarizes the attributes for the BUFGMUX primitive.

Table 2-4: BUFGMUX Attributes

Attribute Name Description Possible Values
CLK_SEL_TYPE Specifies synchronous or asynchronous clock switching. SYNC (default), ASYNC

UltraScale Architecture Clocking Resources Send Feedback

21
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

BUFGMUX_CTRL
BUFGMUX_CTRL is a clock buffer with two clock inputs, one clock output, and a select line.
This primitive is based on BUFGCTRL with some pins connected to logic High or Low.
Figure 2-12 illustrates the relationship of BUFGMUX_CTRL and BUFGCTRL.
X-Ref Target - Figure 2-12

IGNORE1
GND
CE1
VDD
S S1

BUFGMUX_CTRL

I1 I1
O O
I0
I0

S
S0

CE0
VDD
IGNORE0
GND
X16673-111516

Figure 2-12: BUFGMUX_CTRL as BUFGCTRL

BUFGMUX_CTRL uses the S pins as select pins. S can switch anytime without causing a
glitch. The setup/hold time on S is for determining whether the output passes an extra
pulse of the previously selected clock before switching to the new clock. If S changes as
shown in Figure 2-13 prior to the setup time TBCCCK_S and before I0 transitions from High
to Low, the output does not pass an extra pulse of I0. If S changes following the hold time
for S, the output passes an extra pulse. If S violates the setup/hold requirements, the output
might pass the extra pulse but it will not glitch. In any case, the output changes to the new
clock within three clock cycles of the slower clock.

The setup/hold requirements for S0 and S1 are with respect to the falling clock edge, not
the rising edge as for CE0 and CE1.

Switching conditions for BUFGMUX_CTRL are the same as the S pin of BUFGCTRL.
Figure 2-13 illustrates the timing diagram for BUFGMUX_CTRL.

UltraScale Architecture Clocking Resources Send Feedback

22
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

X-Ref Target - Figure 2-13

I0

I1

O
TBCCKO_O
TBCCKO_O
X16674-111516

Figure 2-13: BUFGMUX_CTRL Timing Diagram

Other capabilities of the BUFGMUX_CTRL primitive are:

• I0 and I1 inputs can be preselected after configuration.

• Initial output can be selected as High or Low after configuration.

Additional BUFGCTRL Use Models

Asynchronous MUX Using BUFGCTRL

In some cases an application requires immediate switching between clock inputs or

bypassing the edge sensitivity of BUFGCTRL. An example is when one of the clock inputs is
no longer switching. If this happens, the clock output would not have the proper switching
conditions because the BUFGCTRL never detected a clock edge. This case uses the
asynchronous MUX. Figure 2-14 illustrates an asynchronous MUX with BUFGCTRL design
example.

UltraScale Architecture Clocking Resources Send Feedback

23
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

X-Ref Target - Figure 2-14

IGNORE1
VDD

VDD CE1

S S1

Asynchronous MUX
Design Example

I1 I1
O
O
I0
I0

S0

VDD CE0

VDD IGNORE0
X16675-111516

Figure 2-14: Asynchronous MUX with BUFGCTRL Design Example

Figure 2-15 shows the asynchronous MUX timing diagram.
X-Ref Target - Figure 2-15

I1

I0

TBCCKO_O TBCCKO_O

at I0 Begin I1
X16676-111516

Figure 2-15: Asynchronous MUX Timing Diagram

UltraScale Architecture Clocking Resources Send Feedback

24
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

In Figure 2-15:

• The current clock is from I0.

• S is activated High.
• The clock output immediately switches to I1.
• When ignore signals are asserted High, glitch protection is disabled.

BUFGMUX_CTRL with a Clock Enable

A BUFGMUX_CTRL with a clock enable BUFGCTRL configuration allows you to choose

between the incoming clock inputs. If needed, the clock enable is used to disable the
output. Figure 2-16 illustrates the BUFGCTRL usage design example and Figure 2-17 shows
the timing diagram.
X-Ref Target - Figure 2-16

IGNORE1
GND
CE CE1

S S1
BUFGMUX_CTRL+CE
Design Example

I1 I1

O O
I0 I0

S
S0
CE
CE0
IGNORE0
GND
X16677-111516

Figure 2-16: BUFGMUX_CTRL with a CE and BUFGCTRL

UltraScale Architecture Clocking Resources Send Feedback

25
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

X-Ref Target - Figure 2-17

1 2 3

I0

I1

S
TBCCCK_CE
CE
TBCCKO_O TBCCKO_O

Begin I1
at I0 Clock Off
X16678-111516

Figure 2-17: BUFGMUX_CTRL with a CE Timing Diagram

In Figure 2-17:

• At time event 1, output O uses input I0.

• Before time event 2, S is asserted High.
• At time TBCCKO_O, after time event 2, output O uses input I1. This occurs after a
High-to-Low transition of I0 followed by a High-to-Low transition of I1 is completed.
• At time TBCCCK_CE, before time event 3, CE is asserted Low. To avoid any output clock
glitches, the clock output is switched Low and kept at Low until after a High-to-Low
transition of I1 is completed.

UltraScale Architecture Clocking Resources Send Feedback

26
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

BUFGCE Clock Buffers

BUFGCE is a clock buffer with one clock input, one clock output, and a clock enable line
(Figure 2-18). This buffer provides glitchless clock gating. BUFGCE can directly drive the
routing resources and is a clock buffer with a single gated input. Its O output is 0 when CE
is Low (inactive). When CE is High, the I input is transferred to the O output.
X-Ref Target - Figure 2-18

BUFGCE
CE

X16679-111516

Figure 2-18: BUFGCE Buffer

Table 2-5 lists the BUFGCE pins.

Table 2-5: BUFGCE Pins

Pin Name Type Invertible Description
CE Input TRUE Clock enable
I Input FALSE Clock buffer
O Output FALSE Clock buffer

Table 2-6 shows the BUFGCE attributes.

Table 2-6: BUFGCE Attributes

Attribute Name Values Default Type Description
CE_TYPE SYNC, ASYNC SYNC STRING Sets the clock enable behavior where
SYNC allows for glitchless transition
while ASYNC allows immediate
transition.

UltraScale Architecture Clocking Resources Send Feedback

27
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

Figure 2-19 shows the BUFGCE timing diagram.

X-Ref Target - Figure 2-19

I
TBCCCK_CE

CE

TBCCKO_O
X16680-111516

Figure 2-19: BUFGCE Timing Diagram

BUFG Clock Buffer

BUFG is a clock buffer with one clock input and one clock output. This primitive is based on
BUFGCE with the CE pin connected to High, as shown in Figure 2-20.
X-Ref Target - Figure 2-20

BUFG

O
I

X16667-111516

Figure 2-20: BUFG Buffer

BUFCE_LEAF Clock Buffer

BUFCE_LEAF is a clock buffer with CE for leaf driving off horizontal HCS row. This buffer is an
interconnect leaf clock buffer driving the clocking point of the various blocks with a single
gated input. Its O output is 0 when CE is Low (inactive). When CE is High, the I input is
transferred to the O output. Table 2-7 shows the BUFCE_LEAF attributes.

Table 2-7: BUFCE_LEAF Attributes

Attribute Values Default Type Description
Name
CE_TYPE SYNC, ASYNC SYNC STRING Sets the clock enable behavior where SYNC
allows for glitchless transition while ASYNC
allows immediate transition.

The BUFGCE_LEAF is documented for information purpose only and is not user accessible in
Vivado design suite (e.g., for instantiation, placement, etc.).

UltraScale Architecture Clocking Resources Send Feedback

28
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

BUFGCE_DIV
BUFGCE_DIV is a clock buffer with one clock input (I), one clock output (O), one clear input
(CLR) and a clock enable (CE) input. BUFGCE_DIV can directly drive the routing and
distribution resources and is a clock buffer with a single gated input and a reset. Its O
output is 0 when CLR is High (active). When CE is High, the I input is transferred to the O
output. CE is synchronous to the clock for glitch-free operation. CLR is an asynchronous
reset assertion and synchronous reset deassertion to this buffer. BUFGCE_DIV can also
divide the input clock by 1 to 8.

When CLR (reset) is deasserted, the output clock transitions from Low to High on the first
edge after the CLR is deasserted, regardless of the divide value. Therefore, BUFGCE_DIV
output clocks are always aligned, regardless of the divide value. The output clock then
toggles at the divided frequency. When CLR is asserted, the clock stops toggling after some
clock-to-out time. For an odd divide, the duty cycle is not 50% because the clock is High
one cycle less than it is Low. For example, for a divide value of 7, the clock is High for 3
cycles and Low for 4 cycles.

When CE is deasserted, the output stops at its current state, High or Low. When CE is
reasserted, the internal counter restarts from where it stopped. For example, if the divide
value is 8 and CE is deasserted two input clock cycles after the last output High transition,
the output stays High. Then when CE is reasserted, the output transitions Low after two
input clock cycles. If the reset input is used, upon assertion the output transitions Low
immediately if the current output is High, otherwise it stays Low.

Since reset is synchronously deasserted, when reset is deasserted in the previous example,
the output transitions High at the next input clock edge and transitions Low four input
clock cycles later.

Table 2-8 shows the BUFGCE_DIV pins.

Table 2-8: BUFGCE_DIV Pins

Pin Name Type Invertible Description
I Input FALSE Clock input
CLR Input TRUE Reset
CE Input TRUE Clock enable
O Output FALSE Clock output

Table 2-9 shows the BUFGCE_DIV attributes.

Table 2-9: BUFGCE_DIV Attributes

Attribute Name Values Default Type Description
BUFGCE_DIVIDE 1, 2, 3, 4, 5, 6, 7, 8 1 STRING Defines whether the output
clock is a divided version of the
input clock.

UltraScale Architecture Clocking Resources Send Feedback

29
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

BUFG_GT and BUFG_GT_SYNC

The BUFG_GTs are driven by the gigabit transceivers (GTs) and the ADC/DAC blocks in the
RFSoC devices. BUFG_GTs provide the only means for those blocks to drive the clock routing
resources. Only GTs, ADCs, and DACs can drive BUFG_GTs. BUFG_GT (Figure 2-21) is a clock
buffer with one clock input (I), one clock output (O), one clear input (CLR) with CLR mask
input (CLRMASK), a clock enable (CE) input with a CE mask input (CEMASK) and a 3-bit
divide (DIV[2:0]) input. BUFG_GT_SYNC is the synchronizer circuit for the BUFG_GTs and is
shown here explicitly. The BUFG_GT_SYNC primitive is automatically inserted by the Vivado
tools, if not present in the design. This buffer can directly drive the routing and distribution
resources and is a clock buffer with a single gated input and a reset. When CE is deasserted
(Low) the output stops at its current state, High or Low. When CE is High, the I input is
transferred to the O output. Both edges of CE and the deassertion of CLR are automatically
synchronized to the clock for glitch-free operation. The Vivado tools do not support timing
for the CE pin, therefore, a deterministic latency cannot be achieved. CLR is an
asynchronous reset assertion and synchronous reset deassertion to the BUFG_GTs. The
synchronizers have two stages, but the CLR pin does not have a setup/hold timing arc
assigned. Therefore, the latency is not deterministic. BUFG_GTs can also divide the input
clock by 1 to 8. The DIV[2:0] value is the actual divide minus 1 (i.e., 3'b000 corresponds to
1 while 3'b111 corresponds to 8). The divide value (DIV inputs), CEMASK, and CLRMASK
must be changed while the buffer is held in reset. The input clock is allowed to change while
CE is deasserted or reset is asserted. However, there is a minimum deassertion/assertion
time for those control signals.

IMPORTANT: In RFSoC devices, the ADC and DAC tiles replace the GTH transceivers that are present in
the MPSoC devices. Therefore, ADC and DAC utilize the existing BUFG_GT clock buffers to drive the
global clock trees in the device and then back into the ADC/DAC tiles from the fabric. However, the DIV
function cannot be used when connecting to the ADC/DAC clocks. Hence the BUFG_GT functions more
like a simple global clock buffer with CE and CLR.

IMPORTANT: For devices in Zynq Ultrascale+ and selected devices in Kintex Ultrascale+ families
(XCKU9P and above), assigning the clock root in the same region as the BUFG_GT driver (X0 column)
can cause an unroutable situation and prevent the output clocks from reaching loads that are placed
in the clock regions to the right of the Zynq UltraScale+ device PS or Kintex UltraScale+ empty PL
regions in the Y0, Y1, and Y2 rows. To avoid the issue, users need to assign clock root one clock region
to the right, in this case the X1 column.

UltraScale Architecture Clocking Resources Send Feedback

30
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

X-Ref Target - Figure 2-21

3
DIV[2:0]

Clear Mask

Enable Mask

CEMASK
CLRMASK
MGT and DIV
ADC/DAC I O Clock
Clock
CLR
BUFG_GT
CE

BUFG_GT_SYNC
CLK
CESYNC
Clock Enable CE
CLRSYNC
Reset CLR

X19390-060717

Figure 2-21: BUFG_GT Primitive

UltraScale devices have 24 BUFG_GTs and 10 BUFG_GT_SYNCs per GT Quad. UltraScale+

devices also have 24 BUFG_GTs but they have 14 BUFG_GT_SYNCs per GT Quad. Any of the
GT output clocks in a Quad can be multiplexed to any of the BUFG_GTs. In UltraScale
devices, there are 10 CE and CLR pins which correspond to the 10 BUFG_GT_SYNCs and that
can drive the 24 BUFG_GTs. In UltraScale+ devices, there are 14 CE and CLR pins which
correspond to the 14 BUFG_GT_SYNCs and that can drive the 24 BUFG_GTs. Each of the
BUFG_GT buffers have an individual mask for both CE and CLR (24). All BUFG_GTs driven by
the same clock source must also have a common CE and CLR signal. Tying off CE and CLR to
a constant signal in this case is not allowed, but a mask can be set to provide the same
functionality. The output clocks of the BUFG_GTs connected to the same input clock are
synchronized (phase aligned) to each other when coming out of reset (CLR) or on CE
assertion. Individual mask pins can be used to control which BUFG_GT(s) out of the group of
24 respond to CE and CLR and therefore are synchronized to each other or retain their
previous phase and divide value. These clock buffers are located in the HCS and are directly
driven by the GT output clocks. Their purpose is to directly drive hard blocks and logic in the
CRs via routing and distribution resources. GTs have no other direct, dedicated connections
to other clock resources. However, they can connect to the CMT via the BUFG_GT and the
clock routing resources.

When CLR (reset) is deasserted, the output transitions High at the next input clock edge and
transitions Low divide_value/2 input clock cycles later. Because reset is synchronously
deasserted, two clock cycles of synchronization latency need to be added to the output to
transition it to High. The next transition to Low then occurs four input clock cycles after that
(divide by 8). The output transitions to High a number of clock cycles later, determined by
the divide value specified, after which the output clock toggles at the divided frequency.

UltraScale Architecture Clocking Resources Send Feedback

31
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 2: Clocking Resources

When CLR is asserted, the clock stops toggling at Low after some clock-to-out time. For an
odd divide, the duty cycle is not 50% because the clock is High one cycle less than it is Low.
For example, for a divide value of 7, the clock is High for 3 cycles and Low for 4 cycles.

When CE is deasserted, the output stops at its current state, High or Low. When CE is
reasserted, the internal counter restarts from where it stopped. For example, if the divide
value is 8 and CE is deasserted two input clock cycles after the last output High transition,
the output stays High. Then, when CE is reasserted, the output transitions Low four input
clock cycles later (two for synchronization and two to complete the High time period of the
output clock because of being a divide by 8). If the reset input is used, upon assertion the
output transitions Low immediately if the current output is High, otherwise it stays Low.
Because reset is synchronously deasserted, when reset is deasserted in the previous
example, the output transitions High two input clock cycles later due to synchronization
and transitions Low four input clock cycles after that (divide by 8).

The mask pins (CEMASK and CLRMASK) control how a specific, single BUFG_GT responds to
the CE/CLR control inputs. When a mask pin is deasserted, its respective control pin has
their normal function. When a mask pin is asserted, the respective control pin is ignored, in
effect allowing the clock to propagate through (i.e., CE is effectively High and reset is
effectively Low). The internal synchronizers phase align the clock outputs of the BUFG_GTs
that are not masked. Both edges of CE are synchronized while only the deassertion of reset
is synchronized. Assertion of reset immediately causes the output of the BUFG_GT to go
Low if it was previously High. This can cause a potential glitch or runt pulse. If this is not
acceptable, CE should be used to stop the output. A reset should then be asserted after two
input clock cycles plus half the “divide value.” This ensures that the output clock High time
(if the output clock happened to be disabled High) is no less than normal.

IMPORTANT: While the synchronizers ensure that all BUFG_GTs driven by the same clock come out of
reset in phase, they might not be in phase with BUFG_GTs that have not been reset (i.e., that have their
reset mask asserted).

BUFG_PS
The BUFG_PS is a simple clock buffer with one clock input (I), one clock output (O). This
clock buffer is a resource for the Zynq UltraScale+ MPSoC processor system (PS) and
provides access to the programmable logic (PL) clock routing resources for clocks from the
processor into the PL. Up to 18 PS clocks can drive the BUFG_PS. This clock buffer resides
next to the PS.

UltraScale Architecture Clocking Resources Send Feedback

32
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3

Clock Management Tile

Overview
In UltraScale™ architecture-based devices, the clock management tile (CMT) includes a
mixed-mode clock manager (MMCM) and two phase-locked loops (PLLs). The main purpose
of the PLL is to generate clocking for the I/Os. But it also contains a limited subset of the
MMCM functions that can be used for general clocking purposes.

The clock input connectivity allows multiple resources to provide the reference clock(s) to
the MMCM. The number of output counters (dividers) is eight, with some of them capable
of driving out an inverted clock signal (180° phase shift). MMCMs have infinite fine phase
shift capability in either direction and can be used in dynamic phase shift mode. The
resolution of the fine phase shift depends on the voltage-controlled oscillator (VCO)
frequency. Fractional divide functionality in increments of 1/8th (0.125) for CLKFBOUT and
CLKOUT0 are available to support greater clock frequency synthesis capability. UltraScale
architecture-based devices have a spread spectrum (SS) capability. If the MMCM
spread-spectrum feature is not used, a spread spectrum on an external input clock will not
be filtered and thus passed on to the output clock.

MMCMs
UltraScale architecture-based devices contain one CMT per I/O bank. The MMCMs serve as
frequency synthesizers for a wide range of frequencies, and as jitter filters for either
external or internal clocks, and deskew clocks.

Input multiplexers select the reference and feedback clocks from either the global clock
I/Os or the clock routing or distribution resources. Each clock input has a programmable
counter divider (D). The phase-frequency detector (PFD) compares both phase and
frequency of the rising edges of both the input (reference) clock and the feedback clock. If
a minimum High/Low pulse is maintained, the duty cycle is ancillary. The PFD is used to
generate a signal proportional to the phase and frequency between the two clocks. This
signal drives the charge pump (CP) and loop filter (LF) to generate a reference voltage to
the VCO. The PFD produces an up or down signal to the charge pump and loop filter to
determine whether the VCO should operate at a higher or lower frequency. When VCO
operates at a frequency that is too high, the PFD activates a down signal causing the control

UltraScale Architecture Clocking Resources Send Feedback

33
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

voltage to be reduced, thus decreasing the VCO operating frequency. When the VCO
operates at a frequency that is too low, an up signal increases voltage. The VCO produces
eight output phases and one variable phase for fine-phase shifting. Each output phase can
be selected as the reference clock to the output counters (Figure 3-1). Each counter can be
independently programmed for a given customer design. A special counter M is also
provided. This counter controls the feedback clock of the MMCM, allowing a wide range of
frequency synthesis.

In addition to integer divide output counters, MMCMs add a fractional counter for
CLKOUT0 and CLKFBOUT.
X-Ref Target - Figure 3-1

General Clock
Lock Detect
Routing Switch Lock
Circuit Lock Monitor

8-phase taps + 1 variable phase tap

CLKIN1 9
D
CLKIN2 PFD CP LF VCO CLKOUT0
Fractional Divide
CLKOUT0B

CLKFB CLKOUT1
O1 CLKOUT1B

CLKOUT2
O2 CLKOUT2B

CLKOUT3
O3 CLKOUT3B

CLKOUT4
O4

CLKOUT5
O5

CLKOUT6
O6

M CLKFBOUT
(Fractional Divide) CLKFBOUTB

X16683-111516

Figure 3-1: Detailed MMCM Block Diagram

UltraScale Architecture Clocking Resources Send Feedback

34
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

General Usage Description

MMCM Primitives
The UltraScale device MMCM primitives, MMCME3_BASE and MMCME3_ADV, are shown in
Figure 3-2. The UltraScale+ devices have the same primitives with an E4 instead of an E3. In
this user guide, MMCME4_ADV is the same as the MMCME3_ADV, and MMCME4_BASE is
the same as MMCME3_BASE.
X-Ref Target - Figure 3-2

CLKIN1 CLKOUT0 CLKIN1 CLKOUT0

CLKOUT0B CLKIN2 CLKOUT0B

CLKFBIN CLKOUT1 CLKFBIN CLKOUT1
RST CLKOUT1B RST
CLKOUT1B
PWRDWN CLKOUT2 PWRDWN
CLKOUT2
CLKOUT2B CLKINSEL
CLKOUT2B
CLKOUT3 DADDR[6:0]
CLKOUT3
CLKOUT3B DI[15:0]
CLKOUT3B
CLKOUT4 DWE

CLKOUT5 CLKOUT4
DEN

CLKOUT6 DCLK CLKOUT5

CLKFBOUT PSINCDEC CLKOUT6

CLKFBOUTB PSEN CLKFBOUT

LOCKED PSCLK CLKFBOUTB

LOCKED

CDDCREQ DO[15:0]

PSDONE

CLKINSTOPPED
MMCME3_BASE or
MMCME4_BASE CLKFBSTOPPED

CDDCDONE

MMCME3_ADV or
MMCME4_ADV
X16684-111516

Figure 3-2: MMCM Primitives

MMCME3_BASE or MMCME4_BASE Primitives

The MMCME#_BASE primitives provide access to the most frequently used features of a
stand-alone MMCM. Clock deskew, frequency synthesis, coarse phase shifting, and duty

UltraScale Architecture Clocking Resources Send Feedback

35
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

cycle programming are available to use with the MMCME#_BASE. The ports are listed in
Table 3-1.

Table 3-1: MMCME#_BASE Ports

Description Ports
Clock input CLKIN1, CLKFBIN
Control inputs RST
Clock output CLKOUT0 to CLKOUT6, CLKOUT0B to CLKOUT3B, CLKFBOUT, and CLKFBOUTB
Status and data outputs LOCKED
Power control PWRDWN

MMCME3_ADV and MMCME4_ADV Primitives

The MMCME#_ADV primitive provides access to all MMCME#_BASE features plus additional
ports for clock switching, access to the dynamic reconfiguration port (DRP), and dynamic
fine-phase shifting. The MMCME#_ADV ports are listed in Table 3-2.

Table 3-2: MMCME#_ADV Ports

Description Ports
Clock input CLKIN1, CLKIN2, CLKFBIN, DCLK, PSCLK
Control and data input RST, CLKINSEL, DWE, DEN, DADDR, DI, PSINCDEC, PSEN, CDDCREQ
Clock output CLKOUT0 to CLKOUT6, CLKOUT0B to CLKOUT3B, CLKFBOUT, and CLKFBOUTB
Status and data output LOCKED, DO, DRDY, PSDONE, CLKINSTOPPED, CLKFBSTOPPED, CDDCDONE
Power control PWRDWN

The MMCM is a mixed-signal block designed to support clock network deskew, frequency
synthesis, and jitter reduction. These three modes of operation are discussed in more detail
in this section. The VCO operating frequency can be determined by using the following
relationship:
M
F VCO = F CLKIN  --- Equation 3-1
D

M
F OUT = F CLKIN  ------------ Equation 3-2
DO
where the M, D, and O counters are shown in Figure 3-2, page 35. The value of M
corresponds to the CLKFBOUT_MULT_F setting, the value of D to the DIVCLK_DIVIDE, and O
to the CLKOUT_DIVIDE.

The seven “O” counters can be independently programmed. For example, O0 can be
programmed to do a divide-by-two while O1 is programmed for a divide-by-three. The only
constraint is that the VCO operating frequency must be the same for all the output counters
because a single VCO drives all the counters.

UltraScale Architecture Clocking Resources Send Feedback

36
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Clock Network Deskew

In many cases, designers do not want to incur the delay on a clock network in their I/O
timing budget. Therefore, an MMCM is used to compensate for the clock network delay.
This feature is supported in UltraScale architecture-based devices. A clock output matching
the reference clock CLKIN frequency (always CLKFBOUT) is connected to a clock buffer of
the same type driving the logic and fed back to the CLKFBIN feedback pin of the MMCM.
The remaining outputs can still be used to divide the clock down for additionally
synthesized frequencies. In this case, all output clocks have a defined phase relationship to
the input reference clock.

Frequency Synthesis Only Using Integer Divide

The MMCMs can also be used for stand-alone frequency synthesis. In this application, the
MMCM is not used to deskew a clock network. Rather, it generates an output clock
frequency for other blocks. In this mode, the MMCM feedback paths are internal, which
keeps all the routing local, minimizing the jitter. Figure 3-3 shows the MMCM configured as
a frequency synthesizer. In this example, an external 33 MHz reference clock is available.
The reference clock can be a crystal oscillator or the output of another MMCM. Setting the
M counter to 32 makes the VCO oscillate at 1056 MHz (33 MHz x 32). The MMCM outputs
are programmed to provide (for example) a 528 MHz processor clock, a 264 MHz gasket
clock, a 176 MHz clock, a 132 MHz memory interface clock, a 66 MHz interface, and a
33 MHz interface. In this example, there are no required phase relationships between the
reference clock and the output clocks, but there are required relationships between the
output clocks.
X-Ref Target - Figure 3-3

33 MHz
Reference D=1 PFD, CP, O0 = 2 Processor
Clock LF, VCO

O1 = 4 Gasket
M = 32
O2 = 6 CLBs

O3 = 8 Memory Interface

O4 = 16 66 MHz Interface

O5 = 32 33 MHz Interface

O6 = 1 not used
X16685-111516

Figure 3-3: MMCM as a Frequency Synthesizer

UltraScale Architecture Clocking Resources Send Feedback

37
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Frequency Synthesis Using Fractional Divide in the MMCM

Devices support fractional (non-integer) divides in the CLKOUT0 output path. The
resolution of the fractional divide is 1/8 or 0.125, effectively increasing the number of
synthesizeable frequencies by a factor of eight. For example, if the CLKIN frequency is
100 MHz and the M divide value is set to 8, the VCO frequency is 800 MHz. CLKOUT0 can be
used to further fractionally divide the 800 MHz VCO frequency (e.g., CLKOUT0_DIVIDE = 2.5,
resulting in a 320 MHz output frequency).

When using the fractional divider, the duty cycle is not programmable for outputs used in
the fractional mode.

Jitter Filter
MMCMs reduce the jitter inherent on a reference clock. The MMCM can be instantiated as
a stand-alone function to only support filtering jitter from an external clock before it is
driven into another block. As a jitter filter, it is usually assumed that the MMCM act as a
buffer and regenerate the input frequency on the output (e.g., FIN = 100 MHz,
FOUT = 100 MHz). In general, greater jitter filtering is possible by using the MMCM attribute
BANDWIDTH set to Low. Setting the BANDWIDTH to Low can incur an increase in the static
offset of the MMCM.

Limitations
The MMCM has some restrictions that must be adhered to. These are summarized in the
MMCM electrical specifications in the UltraScale device data sheets [Ref 5]. In general, the
major limitations are VCO operation range, input frequency, duty cycle programmability,
and phase shift. In addition, there are connectivity limitations to other clocking elements
(pins, GTs, and clock buffers). Cascading MMCMs can only occur through the clock routing
network.

VCO Operating Range

The minimum and maximum VCO operating frequencies are defined in the electrical
specification of the UltraScale device data sheets [Ref 5]. These values can also be extracted
from the speed specification.

Minimum and Maximum Input Frequencies

The minimum and maximum CLKIN input frequencies are defined in the electrical
specification of the UltraScale device data sheets [Ref 5].

UltraScale Architecture Clocking Resources Send Feedback

38
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Duty Cycle Programmability

Only discrete duty cycles are possible given a VCO operating frequency. Depending on the
CLKOUT_DIVIDE value, a minimum and maximum range is possible with a step size that is
also dependent on the CLKOUT_DIVIDE value. The Clocking Wizard tool gives the possible
values for a given CLKOUT_DIVIDE.

Phase Shift
In many cases, there needs to be a phase shift between clocks. The MMCM has multiple
options to implement phase shifting. Static phase shifting can be achieved by selecting one
of the eight VCO output phases with additional fine phase shifting available in the CLKOUT
output counters depending on the CLKOUT divide value. There is also an interpolated phase
shifting capability in either fixed or dynamic mode. The MMCM phase shifting capabilities
are very powerful, which can lead to complex scenarios. By using the Clocking Wizard, the
allowable phase shift values are determined based on the MMCM configuration settings.

Static Phase Shift Mode (MMCM and PLL)

The static phase shift (SPS) resolution in time units is defined as:
1 D
SPS = -------------- period or ------------- period Equation 3-3
8F VCO 8MF IN

Because the VCO can provide eight phase-shifted clocks at 45° each; always providing
possible settings for 0°, 45°, 90°, 135°, 180°, 225°, 270°, and 315° of phase shift. The higher
the VCO frequency is, the smaller the phase shift resolution. Because the VCO has a distinct
operating range, it is possible to bound the phase shift resolution using from
1 1
---------------------- to ----------------------- period.
8F VCOMIN 8FVCOMAX

Each CLKOUT output counter is individually programmable allowing each to have an

additional phase shift resolution in degrees based on the phase of the VCO selected and the
CLKOUT counter divide value. The granularity of the CLKOUT phase shift value can be
calculated as 45°/CLKOUT_DIVIDE value. The maximum phase shift range is also determined
by the CLKOUT_DIVIDE value. The maximum phase shift is 360° when CLKOUT_DIVIDE  64.
When CLKOUT_DIVIDE is > 64, the maximum phase shift is:

63
Maximum Phase Shift =  ----------------------------------------  360 +  7  Phase Shift Value  Equation 3-4
 CLKOUT_ DIVIDE 

It is possible to phase shift the CLKFBOUT feedback clock. In that case, all CLKOUT output
clocks are negatively phase shifted with respect to CLKIN.

UltraScale Architecture Clocking Resources Send Feedback

39
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

The two fractional counters (CLKFBOUT and CLKOUT0) also have static phase shift
capability. A phase shift step is defined as:

360 45
SPS  frac  = --------------------------------------------------------- or ------------------------------------------------- Equation 3-5
8  fractional_divide_value fractional_divide_value

For example, if the fractional divide value is 2.125, a static phase shift step is
360/(2.125 x 8) = 21.176 degrees.

Interpolated Fine Phase Shift in Fixed or Dynamic Mode in the MMCM

Interpolated fine phase shift (IFPS) mode in the MMCM has linear shift behavior
independent of the CLKOUT_DIVIDE value, and the phase shift resolution only depends on
the VCO frequency. In this mode, the output clocks can be rotated 360° round robin
1
in linear increments of ---------
------- .
56F VCO

If the VCO runs at 600 MHz, the phase resolution is approximately (rounded) 30 ps, and at
1.6 GHz is approximately (rounded) 11 ps.

No initial phase shift value can be programmed during configuration. When using fine
phase shift, no initial phase shift amount can be set. The phase always starts at zero and can
then be dynamically incremented or decremented. The dynamic phase shift is controlled by
the PS interface of the MMCME#_ADV. This phase shift mode equally affects all CLKOUT
output clocks that are selected for this mode by setting the USE_FINE_PS attribute to TRUE.
In interpolated fine phase shift mode, a clock must always be connected to the PSCLK pin
of the MMCM. Regardless of the interpolated fine phase shift mode (fixed or dynamic) a
clock is in, the clock must always be connected to the PSCLK pin of the MMCM. Each
individual CLKOUT counter can independently either select the interpolated phase shift, the
previously described static phase shift mode, or none. Fractional divide is not allowed in
either fixed or dynamic interpolated fine phase shift mode. Fixed or dynamic phase shifting
of the feedback path results in a negative phase shift of all output clocks with respect to
CLKIN. The dynamic phase shift interface cannot be used when the phase shift mode is set
to fixed.

UltraScale Architecture Clocking Resources Send Feedback

40
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Dynamic Phase Shift Interface in the MMCM

The MMCME#_ADV primitive provides three inputs and one output for dynamic fine phase
shifting. Each CLKOUT and the CLKFBOUT divider can be individually selected for phase
shifting. The attributes CLKOUT[0:6]_USE_FINE_PS and CLKFBOUT_USE_FINE_PS select the
output clocks to be dynamically phase shifted. The dynamic phase shift amount is common
to all the output clocks selected.

The variable phase shift is controlled by the PSEN, PSINCDEC, PSCLK, and PSDONE ports
(Figure 3-4). The phase of the MMCM output clock(s) increments/decrements according to
the interaction of PSEN, PSINCDEC, PSCLK, and PSDONE from the initial or previously
performed dynamic phase shift. PSEN, PSINCDEC, and PSDONE are synchronous to PSCLK.
When PSEN is asserted for one PSCLK clock period, a phase shift increment/decrement is
initiated. When PSINCDEC is High, an increment is initiated and when PSINCDEC is Low, a
decrement is initiated. Each increment adds to the phase shift of the MMCM clock outputs
by 1/56th of the VCO period. Similarly, each decrement decreases the phase shift by 1/56th
of the VCO period. PSEN must be active for one PSCLK period. PSDONE is High for exactly
one clock period when the phase shift is complete. The number of PSCLK cycles is
deterministic (12 PSCLK cycles). After initiating the phase shift by asserting PSEN and the
completion of the phase shift signaled by PSDONE, the MMCM output clocks gradually drift
from their original phase shift to an increment/decrement phase shift in a linear fashion.
The completion of the increment or decrement is signaled when PSDONE asserts High. After
PSDONE has pulsed High, another increment/decrement can be initiated. There is no
maximum phase shift or phase shift overflow. An entire clock period (360°) can always be
phase shifted regardless of frequency. When the end of the period is reached, the phase
shift wraps around round-robin style.
X-Ref Target - Figure 3-4

PSCLK

PSEN

PSDONE

PSINCDEC
X16686-111516

Figure 3-4: Phase Shift Timing Diagram

UltraScale Architecture Clocking Resources Send Feedback

41
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

MMCM Clock Divide Dynamic Change

The Clock Divide Dynamic Change (CDDC) feature supports the dynamic change of the
clock output dividers (CLKOUT[6:0]_DIVIDE) in conjunction with the DRP interface without
the need for resetting the MMCM. Effectively, one or more of the MMCM output clock
frequencies can be changed while leaving other output clocks untouched and running
continuously. Two pins (CDDCREQ and CDDCDONE) control the handshaking. The
application requests a change of output counter values (the CLKOUT_DIVIDE value) by
asserting the CDDCREQ signal. New values are written through the standard DRP interface
one port at a time and governed by standard DRP protocol (DEN, DWE, and DRDY). The DRP
address of the CLKOUT[6:0] counter written to determines which output clocks are affected.
After a CLKOUTx counter is written to, the associated clock output stops toggling. This can
be followed by more changes (DRP writes) to other CLKOUT counters in an identical fashion.
When all DRP writes have been completed (after the last DRDY), the CDDCREQ input must
be deasserted, after which the affected output counter(s) are synchronously restarted. The
MMCM acknowledges that the changes have taken place and the new output clocks
(frequencies) are available for use by asserting CDDCDONE.

CLKOUT ports not affected by the CDDC change continue to function uninterrupted during
this operation and maintain their phase relationship to each other (Figure 3-5). However,
the output clocks (ports) that were changed via the CDDC procedure are not phase aligned
(synchronized) to the other output clocks not affected by CDDCREQ. Clocks affected by
CDDCREQ should not be used after the signal has been asserted because the output might
glitch and the clocks stop toggling. This feature is not available in fractional mode.
X-Ref Target - Figure 3-5

CLKOUT

No CDDC Change

CLKOUT

With CDDC Change

CDDREQ
Last DRP
Access Done

DRDY

CDDCDONE

X16698-111516

Figure 3-5: CDDC Timing Diagram

UltraScale Architecture Clocking Resources Send Feedback

42
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

MMCM Counter Cascading

The CLKOUT6 divider (counter) can be cascaded with the CLKOUT4 divider. This provides the
capability of an output divider that is larger than 128. CLKOUT6 feeds the input of the
CLKOUT4 divider. There is a static phase offset between the output of the cascaded divider
and all other output dividers.

MMCM Programming
Programming of the MMCM must follow a set flow to ensure configuration that guarantees
stability and performance. This section describes how to program the MMCM based on
certain design requirements. A design can be implemented in two ways, directly through
the GUI interface (the Clocking Wizard) or implementing the MMCM through instantiation.
Regardless of the method selected, the following information is necessary to program the
MMCM:

• Reference clock period

• Output clock frequencies (up to seven maximum)
• Output clock duty cycle (default is 50%)
• Output clock phase shift in number of degrees relative to the original 0 phase of the
clock
• Desired bandwidth of the MMCM (default is OPTIMIZED and the bandwidth is chosen in
software)
• Compensation mode (automatically determined by the software)
• Reference clock jitter in UI (i.e., a percentage of the reference clock period)

Determine the Input Frequency

The first step is to determine the input frequency. This allows all possible output
frequencies to be determined by using the minimum and maximum input frequencies to
define the D counter range, the VCO operating range to determine the M counter range,
and the output counter range. There can be a very large number of frequencies. When using
integer divides, in the worst case there are 106 x 64 x 136 = 868,363 possible combinations.
In reality, the total number of different frequencies is less because the entire range of the M
and D counters cannot be realized, and there is overlap between the various settings.

As an example, consider FIN = 100 MHz, F VCO = between 600 MHz and 1600 MHz, and
FPFD = between 10 MHz and 550 MHz.

For a FPFDMIN of 10 MHz, the value of D can only be between 1 and 10.

D MIN (see Equation 3-6) = 1

D MAX (see Equation 3-7) = 10

UltraScale Architecture Clocking Resources Send Feedback

43
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

• For D = 1, M can only have a value between 6 and 16.

• For D = 2, M can only have a value between 12 and 32.
• For D = 3, M can only have a value between 24 and 64.

In addition, D = 1 M = 4 is a subset of D = 2 M = 8, D = 4 M = 16, and D = 8 M = 32

allowing these cases to be dropped. For this case, only D = 1, 3, 5, 6, 7, and 9 are considered
because all other D values are subsets of these cases. This drastically reduces the number of
possible output frequencies. The output frequencies are sequentially selected. The desired
output frequency should be checked against the possible output frequencies generated.
After the first output frequency is determined, an additional constraint can be imposed on
the values of M and D. This can further limit the possible output frequencies for the second
output frequency. This process is continued until all the output frequencies are selected.

The constraints used to determine the allowed M and D values are shown in these
equations:
f IN
D MIN = roundup ------------------- Equation 3-6
f PFD MAX

f IN
D MAX = rounddown -------------
---------- Equation 3-7
f PFD MIN( )

f
M MIN = roundup  --VCOMIN
----------  D MIN
 ------
Equation 3-8
f IN

f
M MAX = rounddown  ---VCOMAX
----------  D MAX
 ------
Equation 3-9
f IN

Determine the M and D Values

Determining the input frequency can result in several possible M and D values. The next
step is to determine the optimum M and D values. The starting M value is first determined.
This is based off the VCO target frequency, the ideal operating frequency of the VCO.

D  f VCOMAX
M IDEAL = ----MIN
-----------------
--------------- Equation 3-10
f IN

The goal is to find the M value closest to the ideal operating point of the VCO. The
minimum D value is used to start the process. The goal is to make D and M values as small
as possible while keeping  VCO as high as possible.

UltraScale Architecture Clocking Resources Send Feedback

44
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

MMCM Ports
Table 3-3 summarizes the MMCM ports.

Table 3-3: MMCM Ports(1)

Pin Name I/O Pin Description
CLKIN1 Input General clock input. See CLKIN1 – Primary Reference Clock Input.
CLKIN2 Input Secondary clock input for the MMCM reference clock. See CLKIN2 – Secondary
Clock Input.
CLKFBIN Input Feedback clock input. See CLKFBIN – Feedback Clock Input.
CLKINSEL Input This signal controls the state of the clock input MUX, High = CLKIN1,
Low = CLKIN2. CLKINSEL dynamically switches the MMCM reference clock. See
CLKINSEL – Clock Input Select.
RST Input Asynchronous reset signal. The RST signal is an asynchronous reset for the
MMCM. The MMCM synchronously re-enables itself when this signal is released
(i.e., MMCM re-enabled). A reset is required when the input clock conditions
change (e.g., frequency). See RST – Asynchronous Reset Signal.
PWRDWN Input Powers down instantiated but unused MMCMs. See PWRDWN – Power Down.
DADDR[6:0] Input The dynamic reconfiguration address (DADDR) input bus provides a
reconfiguration address for the dynamic reconfiguration. When not used, all bits
must be assigned zeros. See DADDR[6:0] – Dynamic Reconfiguration Address.
DI[15:0] Input The dynamic reconfiguration data input (DI) bus provides reconfiguration data.
When not used, all bits must be set to zero. See DI[15:0] – Dynamic
Reconfiguration Data Input.
DWE Input The dynamic reconfiguration write enable (DWE) input pin provides the write
enable control signal to write the DI data into the DADDR address. When not used,
it must be tied Low. See DWE – Dynamic Reconfiguration Write Enable.
DEN Input The dynamic reconfiguration enable (DEN) provides the enable control signal to
access the dynamic reconfiguration feature. When the dynamic reconfiguration
feature is not used, DEN must be tied Low. See DEN – Dynamic Reconfiguration
Enable Strobe.
DCLK Input The DCLK signal is the reference clock for the dynamic reconfiguration port. See
DCLK – Dynamic Reconfiguration Reference Clock.
PSCLK Input Phase shift clock. See PSCLK – Phase Shift Clock.
PSEN Input Phase shift enable. See PSEN – Phase Shift Enable.
PSINCDEC Input Phase shift increment/decrement control. See PSINCDEC – Phase Shift
Increment/Decrement Control.
CLKOUT[0:6] Output User configurable clock outputs (0 through 6) that can be divided versions of the
VCO phase outputs (user controllable) from 1 (bypassed) to 128. The output
clocks are phase aligned to each other (unless phase shifted) and aligned to the
input clock with a proper feedback configuration.
CLKOUT[0:3]B Output Inverted CLKOUT[0:3]. See CLKOUT[0:3]B – Inverted Output Clocks.
CLKFBOUT Output Dedicated MMCM feedback output. See CLKFBOUT – Dedicated MMCM Feedback
Output.
CLKFBOUTB Output Inverted CLKFBOUT. See CLKFBOUTB – Inverted CLKFBOUT.

UltraScale Architecture Clocking Resources Send Feedback

45
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Table 3-3: MMCM Ports(1) (Cont’d)

Pin Name I/O Pin Description
CLKINSTOPPED Output Status pin indicating that the input clock has stopped. See CLKINSTOPPED – Input
Clock Status.
CLKFBSTOPPED Output Status pin indicating that the feedback clock has stopped. See CLKFBSTOPPED –
Feedback Clock Status.
LOCKED Output An output from the MMCM that indicates when the MMCM has achieved phase
alignment within a predefined window and frequency matching within a
predefined PPM range. The MMCM automatically locks after power on. No extra
reset is required. LOCKED is deasserted if the input clock stops or the phase
alignment is violated (e.g., input clock phase shift). The MMCM must be reset after
LOCKED is deasserted.
DO[15:0] Output The dynamic reconfiguration output bus provides MMCM data output when using
dynamic reconfiguration. See DO[15:0] – Dynamic Reconfiguration Output Bus.
DRDY Output The dynamic reconfiguration ready output (DRDY) provides the response to the
DEN signal for the MMCM’s dynamic reconfiguration feature. See DRDY –
Dynamic Reconfiguration Ready.
PSDONE Output Phase shift done. See PSDONE – Phase Shift Done.
CDDCREQ Input Requests a dynamic frequency change for selected clock outputs. See MMCM
Clock Divide Dynamic Change, page 42.
CDDCDONE Output Signals that the dynamic frequency change is completed. See MMCM Clock Divide
Dynamic Change, page 42.

Notes:
1. All control and status signals except PSINCDEC are active High.

TIP: The port names generated by the clocking wizard can differ from the port names used on the
primitive.

MMCM Port Descriptions

CLKIN1 – Primary Reference Clock Input

CLKIN1 can be driven by a global clock I/O directly when in the same bank adjacent to the
PHY tile.

CLKIN2 – Secondary Clock Input

CLKIN2 can be driven by a global clock I/O directly when in the same bank adjacent to the
PHY tile.

CLKFBIN – Feedback Clock Input

CLKFBIN must be connected either directly to the CLKFBOUT for internal feedback, or to the
CLKFBOUT via a BUFG for clock buffer feedback matching, or IBUFG (through a global clock
pin for external deskew) or interconnect (not recommended). For clock alignment, the
feedback path clock buffer type should match the forward clock buffer type.

UltraScale Architecture Clocking Resources Send Feedback

46
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

IMPORTANT: The internal compensation mode setting is determined by a direct connection (wire) from
the CLKFBOUT to the CLKFBIN port in the source. However, synthesis optimizes this connection away
such that the CLKFBOUT to CLKFBIN connection is removed from all subsequent representations in
Vivado design suite. However, the INTERNAL compensation attribute attached to the MMCM/PLL
indicates that the compensation is still internal to the MMCM/PLL.

CLKFBOUT – Dedicated MMCM Feedback Output

For possible configuration of CLKFBOUT, see MMCM Use Models, page 56. CLKFBOUT can
also drive logic if the feedback path contains a clock buffer.

CLKFBOUTB – Inverted CLKFBOUT

This signal should not be used for feedback. It provides an additional, inverted CLKFBOUT
output clock. CLKFBOUTB can drive logic if the feedback path contains a clock buffer.

CLKINSEL – Clock Input Select

The CLKINSEL signal controls the state of the clock input multiplexers. High = CLKIN1, Low
= CLKIN2 (see Reference Clock Switching, page 56). The MMCM must be held in RESET
during clock switchover.

RST – Asynchronous Reset Signal

The RST signal is an asynchronous reset for the MMCM. The MMCM is synchronously
re-enabled when this signal is deasserted.

PWRDWN – Power Down

This signal powers down instantiated but currently unused MMCMs. This mode can be used
to save power for temporarily inactive portions of the design and/or MMCMs that are not
active in certain system configurations. No MMCM power is consumed in this mode.

DADDR[6:0] – Dynamic Reconfiguration Address

The dynamic reconfiguration address (DADDR) input bus provides a reconfiguration

address for dynamic reconfiguration. The address value on this bus specifies the 16
configuration bits that are written or read with the next DCLK cycle. When not used, all bits
must be assigned zeros.

DI[15:0] – Dynamic Reconfiguration Data Input

The dynamic reconfiguration data input (DI) bus provides reconfiguration data. The value of
this bus is written to the configuration cells. The data is presented in the cycle that DEN and
DWE are active. The data is captured in a shadow register and written at a later time. DRDY
indicates when the DRP port is ready to accept another write. When not used, all bits must
be set to zero.

UltraScale Architecture Clocking Resources Send Feedback

47
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

DWE – Dynamic Reconfiguration Write Enable

The dynamic reconfiguration write enable (DWE) input pin provides the write/read enable
control signal to write the DI data into or read the DO data from the DADDR address. When
not used, it must be tied Low.

DEN – Dynamic Reconfiguration Enable Strobe

The dynamic reconfiguration enable strobe (DEN) provides the enable control signal to
access the dynamic reconfiguration feature and enables all DRP port operations. When the
dynamic reconfiguration feature is not used, DEN must be tied Low.

DCLK – Dynamic Reconfiguration Reference Clock

The DCLK signal is the reference clock for the dynamic reconfiguration port. The rising edge
of this signal is the timing reference for all other port signals. The setup time is specified in
the UltraScale device data sheets [Ref 5]. There is no hold time requirement for the other
input signals relative to the rising edge of the DCLK. The pin can be driven by an IBUF,
IBUFG, BUFGCE, or BUFGCTRL. There are no dedicated connections to this clock input.

PSCLK – Phase Shift Clock

This input pin provides the source clock for the dynamic phase shift interface. All other
inputs are synchronous to the positive edge of this clock. The pin can be driven by an IBUF,
IBUFG, BUFG, or BUFGCE. There are no dedicated connections to this clock input.

PSEN – Phase Shift Enable

A dynamic (variable) phase shift operation is initiated by synchronously asserting this

signal. PSEN must be activated for one cycle of PSCLK. After initiating a phase shift, the
phase is gradually shifted until a High pulse on PSDONE indicates that the operation is
complete. There are no glitches or sporadic changes during the operation. From the start to
the end of the operation, the phase is shifted in a continuous analog manner.

PSINCDEC – Phase Shift Increment/Decrement Control

This input signal synchronously indicates if the dynamic phase shift is an increment or
decrement operation (positive or negative phase shift). PSENCDEC is asserted High for
increment and Low for decrement. There is no phase shift overflow associated with the
dynamic phase shift operation. If 360° or more are shifted, the phase wraps around, starting
at the original phase.

UltraScale Architecture Clocking Resources Send Feedback

48
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

CLKOUT[0:6] – Output Clocks

These user-configurable clock outputs (CLKOUT0 through CLKOUT6) can be divided

versions of the VCO phase outputs (user controllable) from 1 (bypassed) to 128. The input
clock and output clocks can be phase aligned.

RECOMMENDED: CLKOUT0 is first used to place the root clock point. In ZHOLD mode, it is used to set
the compensation. Therefore, Xilinx recommends using CLKOUT0 as the main clock.

For possible configurations, see MMCM Use Models, page 56. In the MMCM, CLKOUT0 and
CLKFBOUT can be used in fractional divide mode. All CLKOUT outputs can be used in
non-fractional mode to provide a static or dynamic phase shift. In fractional mode, only
fixed phase shift is allowed. See Static Phase Shift Mode (MMCM and PLL), page 39 for more
information.

CLKOUT[0:3]B – Inverted Output Clocks

This is the inverted (180° phase shift) CLKOUT[0:3].

CLKINSTOPPED – Input Clock Status

This is a status pin indicating that the input clock has stopped. This signal is asserted within
two CLKFBOUT clock cycles of clock stoppage. The signal is deasserted after the clock has
restarted and LOCKED is achieved, or the clock is switched to the alternate clock input and
the MMCM has re-locked.

CLKFBSTOPPED – Feedback Clock Status

This is a status pin indicating that the feedback clock has stopped. CLKFBSTOPPED is
asserted within one clock cycle of clock stoppage. The signal is deasserted after the
feedback clock has restarted and the MMCM has re-locked.

LOCKED

This is an output from the MMCM used to indicate when the MMCM has achieved phase
and frequency alignment of the reference clock and the feedback clock at the input pins.
Phase alignment is within a predefined window and frequency matching within a
predefined PPM range. The MMCM automatically locks after power on; no extra reset is
required. LOCKED is deasserted within one PFD clock cycle if the input clock stops, the
phase alignment is violated (e.g., input clock phase shift), or the frequency has changed.
The MMCM must be reset when LOCKED is deasserted. The clock outputs should not be
used prior to the assertion of LOCKED.

UltraScale Architecture Clocking Resources Send Feedback

49
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

DO[15:0] – Dynamic Reconfiguration Output Bus

The dynamic reconfiguration output bus provides MMCM data output when using dynamic
reconfiguration. If DWE is inactive while DEN is active at the rising edge of DCLK, this bus
holds the content of the configuration cells addressed by DADDR. The DO bus must be
captured on the rising edge of DCLK when DRDY is active. The DO bus value is held until the
next DRP operation.

DRDY – Dynamic Reconfiguration Ready

The dynamic reconfiguration ready output (DRDY) provides the response to the DEN signal
for the MMCM’s dynamic reconfiguration feature. This signal indicates that a DEN/ DCLK
operation has completed.

PSDONE – Phase Shift Done

The phase shift done output signal is synchronous to the PSCLK. When the current phase
shift operation is completed, the PSDONE signal is asserted for one clock cycle indicating
that a new phase shift cycle can be initiated.

CDDCREQ – Request a Clock Output Divide Change

This is a request signal for dynamically changing the output clock divide value and
therefore the frequency. When asserted High, a request is sent to all affected counters and
must stay asserted until the last change via the DRP has been completed.

CDDCDONE – Clock Output Change Done

This is an acknowledge signal from the MMCM that the output clock divide change is
complete and the output is valid.

UltraScale Architecture Clocking Resources Send Feedback

50
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

MMCM Attributes
Table 3-4 lists the attributes for the MMCME#_BASE and MMCME#_ADV primitives.

Table 3-4: MMCM Attributes

Attribute Type Allowed Values Default Description
BANDWIDTH String HIGH OPTIMIZED Specifies the MMCM programming
LOW algorithm affecting the jitter, phase
OPTIMIZED(6) margin, and other characteristics of
the MMCM.
CLKOUT[1:6]_DIVIDE Integer 1 to 128 1 Specifies the amount to divide the
associated CLKOUT clock output if a
CLKOUT[0]_DIVIDE_F (1) Integer 1 to 128 or 1
different frequency is desired. This
or 2.000 to 128.000 in
number, in combination with the
Real increments of 0.125
CLKFBOUT_MULT_F and
DIVCLK_DIVIDE values, determines
the output frequency.
CLKOUT[0:6]_PHASE Real –360.000 to 360.000 0.0 Specifies the output phase
See equations in the relationship of the associated
Static Phase Shift CLKOUT clock output in number of
Mode (MMCM and degrees offset (i.e., 90 indicates a
PLL) section. 90° or ¼ cycle offset phase offset
while 180 indicates a 180° offset or
½ cycle phase offset).
CLKOUT[0:6]_DUTY_CYCLE Real 0.01 to 0.99 0.50 Specifies the duty cycle of the
associated CLKOUT clock output as
a percentage (i.e., 0.50 generates a
50% duty cycle).
CLKFBOUT_MULT_F(1) Integer 2 to 64 or 5 This CLKFBOUT_MULT_F range
or 2.000 to 64.000 in applies to UltraScale devices.
Real increments of 0.125 Specifies the amount to multiply all
CLKOUT clock outputs if a different
frequency is desired. This number,
in combination with the associated
CLKOUT#_DIVIDE value and
DIVCLK_DIVIDE value, determines
the output frequency.
CLKFBOUT_MULT_F(1) Integer 2 to 128 or 5 This CLKFBOUT_MULT_F range
or 2.000 to 128.000 in applies to UltraScale+ devices.
Real increments of 0.125 Specifies the amount to multiply all
CLKOUT clock outputs if a different
frequency is desired. This number,
in combination with the associated
CLKOUT#_DIVIDE value and
DIVCLK_DIVIDE value, determines
the output frequency.
DIVCLK_DIVIDE Integer 1 to 106 1 Specifies the division ratio for all
output clocks with respect to the
input clock. Effectively divides the
CLKIN going into the PFD.

UltraScale Architecture Clocking Resources Send Feedback

51
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Table 3-4: MMCM Attributes (Cont’d)

Attribute Type Allowed Values Default Description
CLKFBOUT_PHASE Real –360.000 to 360.000 0.0 Specifies the phase offset in
degrees of the clock feedback
output. Shifting the feedback clock
results in a negative phase shift of
all output clocks to the MMCM.
REF_JITTER1 Real 0.000 to 0.999 0.010 Allows specification of the expected
REF_JITTER2 jitter on the reference clock to
better optimize MMCM
performance. A bandwidth setting
of OPTIMIZED attempts to choose
the best parameter for input
clocking when unknown. If known,
the value provided should be
specified in terms of the unit
interval (UI) (the maximum
peak-to-peak value) of the expected
jitter on the input clock.
CLKIN1_PERIOD Real 0.968 to 100.000 0.000 Specifies the input period in ns to
the MMCM CLKIN1 input.
Resolution is down to the ps. This
information is mandatory and must
be supplied.
CLKIN2_PERIOD Real 0.968 to 100.000 0.000 Specifies the input period in ns to
the MMCM CLKIN2 input.
Resolution is down to the ps. This
information is mandatory and must
be supplied.
CLKFBOUT_USE_FINE_PS (2) String FALSE, TRUE FALSE CLKFBOUT counter variable fine
phase shift enable.
CLKOUT0_USE_FINE_PS (2) String FALSE, TRUE FALSE CLKOUT0 counter variable fine
phase shift enable.
CLKOUT0_DIVIDE must be an
integer. Therefore, fractional divide
is not allowed.
CLKOUT[1:6]_USE_FINE_PS (2) String FALSE, TRUE FALSE CLKOUT[1:6] variable fine phase
shift enable.
STARTUP_WAIT String FALSE, TRUE FALSE Wait during the configuration
start-up cycle for the MMCM to
lock.
CLKOUT4_CASCADE String FALSE, TRUE FALSE Cascades the output divider
(counter) CLKOUT6 into the input of
the CLKOUT4 divider for an output
clock divider that is greater than
128, effectively providing a total
divide value of 16,384.

UltraScale Architecture Clocking Resources Send Feedback

52
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Table 3-4: MMCM Attributes (Cont’d)

Attribute Type Allowed Values Default Description
COMPENSATION String AUTO (3), AUTO Clock input compensation. Must be
ZHOLD, set to AUTO. Defines how the
EXTERNAL, MMCM feedback is configured.
INTERNAL(5), ZHOLD: Indicates the MMCM is
BUF_IN configured to provide a negative
hold time at the I/O registers.
EXTERNAL: Indicates a network
external to the device is being
compensated.
INTERNAL: Indicates the MMCM is
using its own internal feedback path
so no delay is being compensated.
BUF_IN: Indicates that the
configuration does not match with
the other compensation modes. The
CLKIN and CLKFBIN pins are aligned
in a way that a delay in the feedback
path is compensated with respect to
CLKIN.
SS_EN String FALSE, TRUE FALSE Enables spread spectrum
generation.
SS_MODE String DOWN_LOW, CENTER_ Controls the spread spectrum
DOWN_HIGH, HIGH frequency deviation and the spread
CENTER_LOW, type.
CENTER_HIGH
SS_MOD_PERIOD Integer 4000–40000 10000 Specifies the spread spectrum
modulation period (ns).

Notes:
1. The Vivado tools round up or down to the nearest multiple of 0.125 if the value is not specified as an exact 1/8th fraction.
2. When using the variable fine phase shift, the initial phase shift value is always zero and cannot be preset to a static, initial
phase.
3. The COMPENSATION attribute values are documented for informational purpose only. The Vivado tools automatically select
the appropriate compensation based on circuit topology. Do not manually select a compensation value, leave the attribute
at the default value.
4. The specifications for the VCO frequencies MMCM_FVCOMIN/MMCM_FVCOMAX and minimum out frequency
MMCM_FOUTMIN are different for the UltraScale and UltraScale+ families. Consult the appropriate data sheets.
5. The direct source code connection (wire) from CLKFBOUT to CLKFBIN is optimized away during synthesis.
6. When using an SEM-IP in Ultrascale devices only, additional noise is coupled into VCO of MMCM and PLL. This results in
higher TIE jitter value as described in AR:71314. Refer to the AR for guidance and mitigation techniques. To resolve the issue
of TIE jitter for SEM-IP, there are two new configurable properties: BITSTREAM.MMCM.BANDWIDTH and
BITSTREAM.PLL.BANDWIDTH. If the properties are set to POSTCRC, each MMCM instance that has the BANDWIDTH attribute
set to OPTIMIZED, or to a PLL instance with an implicit BANDWIDTH=OPTIMIZED attribute, will get configured to POSTCTRC
bandwidth settings. Refer to the Vivado Design Suite User Guide [Ref 13] for BITSTREAM property information.

UltraScale Architecture Clocking Resources Send Feedback

53
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

MMCM Clock Input Signals

The possible clock sources for the MMCM include:

• IBUF – Global clock input buffer. The MMCM compensates the delay of this path. IBUF
represents a global clock pin in the same region. The IBUF must be located at a global
clock pin location.
• BUFGCTRL – Internal global clock buffer. The MMCM does not compensate the delay of
this path.
• BUFGCE – Global clock buffer. The MMCM does not compensate the delay of this path.

Counter Control
The MMCM output counters provide a wide variety of synthesized clocks using a
combination of DIVIDE, DUTY_CYCLE, and PHASE. Figure 3-6 illustrates how the counter
settings impact the counter output.

The top waveform represents the output from the VCO.

X-Ref Target - Figure 3-6

Counter Clock
Input (VCO)

DIVIDE = 2
DUTY_CYCLE = 0.5
PHASE = 0

DIVIDE = 2
DUTY_CYCLE = 0.5
PHASE = 180

DIVIDE = 2
DUTY_CYCLE = 0.75
PHASE = 180

DIVIDE = 1
DUTY_CYCLE = 0.5
PHASE = 0

DIVIDE = 1
DUTY_CYCLE = 0.5
PHASE = 360

DIVIDE = 3
DUTY_CYCLE = 0.33
PHASE = 0

DIVIDE = 3
DUTY_CYCLE = 0.5
PHASE = 0
X16687-111516

Figure 3-6: Output Counter Clock Synthesis Examples

UltraScale Architecture Clocking Resources Send Feedback

54
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Detailed VCO and Output Counter Waveforms

Figure 3-7 shows the eight VCO phase outputs and four different counter outputs. Each
VCO phase is shown with the appropriate start-up sequence. The phase relationship and
start-up sequence are guaranteed to ensure the correct phase is maintained. This means the
rising edge of the 0° phase happens before the rising edge of the 45° phase. The O0 counter
is programmed to do a simple divide-by-two with the 0° phase tap as the reference clock.
The O1 counter is programmed to do a simple divide-by-two but uses the 180° phase tap
from the VCO. This counter setting can be used to generate a clock for a DDR interface
where the reference clock is edge aligned to the data transition. The O2 counter is
programmed to do a divide-by-three. The O3 output has the same programming as the O2
output except the phase is set for a one cycle delay. Phase shifts greater than one VCO
period are possible.

X-Ref Target - Figure 3-7

0°

45°

90°

135°
VCO
8 Phases 180°

225°

270°

315°

O0

O1
Counter
Outputs O2

O3

One Cycle Delay

X16688-111516

Figure 3-7: Selecting VCO Phases

If the MMCM is configured to provide a certain phase relationship and the input frequency
is changed, this phase relationship is also changed because the VCO frequency changes and
therefore the absolute shift in picoseconds changes. This aspect must be considered when
designing with the MMCM. When an important aspect of the design is to maintain a certain
phase relationship among various clock outputs, (e.g., CLK and CLK90), this relationship is
maintained regardless of the input frequency.

All O counters can be equivalent; anything O0 can do, O1 can do. The O0 counter has the
additional capability to be used in fractional divide mode. The MMCM outputs are flexible
when connecting to the global clock network because they are identical. In most cases, this

UltraScale Architecture Clocking Resources Send Feedback

55
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

level of detail is imperceptible because the software and Clocking Wizard determine the
proper settings through the MMCM attributes and Wizard inputs.

Reference Clock Switching

The MMCM reference clock can be dynamically switched by using the CLKINSEL pin. The
switching is done asynchronously. After the clock switches, the MMCM is likely to lose
LOCKED and automatically lock onto the new clock. Therefore, after the clock switches, the
MMCM must be reset. The MMCM clock MUX switching is shown in Figure 3-8. The
CLKINSEL signal directly controls the MUX. No synchronization logic is present.
X-Ref Target - Figure 3-8

CLKINSEL

CLKIN1

IBUFG (CC)
BUFGCTRL
BUFGCE
Local Routing
(Not Recommended)

MMCM
CLKIN

CLKIN2

IBUFG (CC)
BUFGCTRL
BUFGCE
Local Routing
(Not Recommended)
X16689-111516

Figure 3-8: Input Clock Switching

Missing Input Clock or Feedback Clock

When the input clock or feedback clock is lost, the CLKINSTOPPED or CLKFBSTOPPED status
signal is asserted. The MMCM deasserts the LOCKED signal. After the clock returns, the
CLKINSTOPPED signal is deasserted and a RESET must be applied.

MMCM Use Models

The examples in this section show the MMCM. There are several methods to design with the
MMCM. The Clocking Wizard in the Vivado tools can assist with generating the various
MMCM parameters. Additionally, the MMCM can be manually instantiated as a component.
It is also possible for the MMCM to be merged with an IP core. The IP core would contain
and manage the MMCM.

UltraScale Architecture Clocking Resources Send Feedback

56
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Clock Network Deskew

One of the predominant uses of the MMCM is for clock network deskew. Figure 3-9 shows
the MMCM in this mode. The clock output from one of the CLKOUT counters is used to drive
logic within the device and/or the I/Os. The feedback counter is used to control the exact
phase relationship between the input clock and the output clock (if, for example a 90° phase
shift is required). The associated clock waveforms are shown to the right for the case where
the input clock and output clock need to be phase aligned. The configuration in Figure 3-9
is the most flexible, but it does require two global clock networks.
X-Ref Target - Figure 3-9

IBUFG BUFG

1 2 4 5
CLKIN1 CLKOUT0 To Logic
3
CLKFBIN CLKOUT0B
RST CLKOUT1

CLKOUT1B

CLKOUT2

CLKOUT2B

CLKOUT3

CLKOUT3B 1

CLKOUT4
2
CLKOUT5
BUFG
CLKOUT6 3
6
CLKFBOUT
4
CLKFBOUTB

LOCKED 5
MMCM
6

X16690-111516

Figure 3-9: Global Clock Network Deskew Using Two BUFGs

There are certain restrictions on implementing the feedback. The CLKFBOUT output can be
used to provide the feedback clock signal. When an MMCM is driving both BUFGs and
BUFGCTRL, only one of the clock buffers that is also used in the feedback path is deskewed.
The fundamental restriction is that both input frequencies to the PFD must be identical.
Therefore, this relationship must be met:
f IN f VCO
- = f FB = ----
---- ------ Equation 3-11
D M
As an example, if ƒ IN is 166 MHz, D = 1, M = 6, and O = 2, then VCO is 996 MHz and the
clock output frequency is 498 MHz. Because the M value in the feedback path is 6, both
input frequencies at the PFD are 166 MHz.

UltraScale Architecture Clocking Resources Send Feedback

57
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Another more complex scenario has an input frequency of 66.66 MHz and D = 2, M = 30,
and O = 4. The VCO frequency in this case is 1000 MHz and the CLKOUT output frequency
is 250 MHz. Therefore, the feedback frequency at the PFD is 1000/30 or 33.33 MHz,
matching the 66.66 MHz/2 input clock frequency at the PFD.

MMCM with Internal Feedback

The MMCM feedback can be internal to the MMCM when the MMCM is used as a
synthesizer or jitter filter, and there is no required phase relationship between the MMCM
input clock and the MMCM output clock. The MMCM performance increases because the
feedback clock is not subjected to noise on the core supply since it never passes through a
block powered by this supply. However, noise introduced on the CLKIN signal and the BUFG
are still present (Figure 3-10).
X-Ref Target - Figure 3-10

IBUFG BUFG

CLKIN1 CLKOUT0 To Logic

CLKFBIN CLKOUT0B
RST CLKOUT1

CLKOUT1B

CLKOUT2

CLKOUT2B

CLKOUT3

CLKOUT3B

CLKOUT4

CLKOUT5

CLKOUT6

CLKFBOUT

CLKFBOUTB

LOCKED
MMCM

X16691-111516

Figure 3-10: MMCM with Internal Feedback

UltraScale Architecture Clocking Resources Send Feedback

58
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Zero Delay Buffer

The MMCM can also be used to generate a zero delay buffer clock. A zero delay buffer can
be useful for applications where there is a single clock signal fanout to multiple
destinations with a low skew between them. This configuration is shown in Figure 3-11.
Here, the feedback signal drives off chip and the board trace feedback is designed to match
the trace to the external components. In this configuration, it is assumed that the clock
edges are aligned at the input of the UltraScale device and the input of the external
component. The input clock buffers for CLKIN and CLKFBIN must be in the same bank.
X-Ref Target - Figure 3-11

IBUFG BUFG OBUF

Inside UltraScale Device
CLKIN1 CLKOUT0
CLKOUT0B To
External
CLKOUT1 Components
CLKFBIN
CLKOUT1B
RST CLKOUT2
CLKOUT2B
CLKOUT3
CLKOUT3B
CLKOUT4
CLKOUT5
BUFG
CLKOUT6
CLKFBOUT
CLKFBOUTB
MMCM LOCKED

X16692-111516

Figure 3-11: Zero Delay Buffer

In some cases, precise alignment cannot occur because of the difference in loading
between the input capacitance of the external component and the feedback path
capacitance of the UltraScale device. For example, the external components can have an
input capacitance of 1 pF to 4 pF while the part has an input capacitance as specified in the
UltraScale device data sheets [Ref 5]. There is a difference in the signal slope, which is
basically skew. Designers should be aware of this effect to ensure timing.

CMT to CMT Connection

The MMCM can be cascaded through the routing resources only. There is no compensation
for routing delays.

UltraScale Architecture Clocking Resources Send Feedback

59
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Spread-Spectrum Clock Generation

Spread-spectrum clock generation (SSCG) is widely used by manufacturers of electronic
devices to reduce the spectral density of the electromagnetic interference (EMI) generated
by these devices. Manufacturers must ensure that levels of electromagnetic energy emitted
do not interfere with the operation of other nearby electronic devices. For example, the
clarity of a phone call should not degrade when the phone is next to a video display. In the
same way, the display should not be affected when the phone is used.

Electromagnetic compatibility (EMC) regulations are used to control the noise or EMI that
causes these disturbances. Typical solutions for meeting EMC requirements involve adding
expensive shielding, ferrite beads, or chokes. These solutions can adversely impact the cost
of the final product by complicating PCB routing and forcing longer product development
cycles.

SSCG spreads the electromagnetic energy over a large frequency band to effectively reduce
the electrical and magnetic field strengths measured within a narrow window of
frequencies. The peak electromagnetic energy at any one frequency is reduced by
modulating the SSCG output.

The MMCME# can generate a spread-spectrum clock from a standard fixed frequency
oscillator when SS_EN is set to TRUE (see Figure 3-12). Within the MMCME#, the VCO
frequency is modulated along with CLKFBOUT and CLKOUT[6:4,1,0]. Clock outputs
CLKOUT[3:2] are used to control the modulation period and are not available for general
use. As long as the clock frequency is adjusted slowly, the spread-spectrum does not affect
the period jitter of the MMCME#.
X-Ref Target - Figure 3-12

Modulation Period
Frequency

FIN
Frequency
Deviation

Time
X16693-111516

Figure 3-12: Center-Spread Modulation

Adjusting the modulation period SS_MOD_PERIOD allows you to direct the tools to select
the closest modulation period based on the MMCME# settings. The spread-spectrum
modulation reduces EMI as long as the modulation frequencies are higher than the audible
frequency range of 30 KHz. Typically, lower modulation frequencies are preferred to
minimize the impact of the introduction of spread-spectrum.

UltraScale Architecture Clocking Resources Send Feedback

60
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Increasing the frequency deviation with SS_MODE (CENTER_HIGH or DOWN_HIGH)

increases the overall EMI reduction, but care must be taken to ensure that the increased
range of frequencies does not affect the overall system operation (see Figure 3-13).
Because the spread-spectrum clock and the input clock are operating at different
frequencies, any data being transferred between the clock domains should use an
asynchronous FIFO to ensure that data is not lost. Increasing the frequency deviation
requires a larger FIFO.
X-Ref Target - Figure 3-13

FIN CENTER_HIGH
Frequency

CENTER_LOW

Time
X16694-111516

Figure 3-13: Center-Spread Modulation

(CENTER_LOW vs. CENTER_HIGH)
Another design trade-off is the decision to use a center spread or down spread. Selecting
SS_MODE (DOWN_HIGH, DOWN_LOW) spreads the frequencies to lower frequencies as
shown in Figure 3-14. DOWN_HIGH has similar frequency deviation to CENTER_LOW.

UltraScale Architecture Clocking Resources Send Feedback

61
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

X-Ref Target - Figure 3-14

FIN
Frequency

DOWN_LOW
DOWN_HIGH

Time
X16695-111516

Figure 3-14: Down-Spread Modulation

The decision to use down spread is often the result of considering the timing analysis
impact of spread-spectrum. When using a spread-spectrum clock, the design must meet
timing at the highest frequency in the frequency deviation. Therefore, if a 100 MHz clock
with SS_MODE (CENTER_LOW) produces a 3% (±1.5%) center spread, the 100 MHz clock
with 3% center spread must pass timing analysis as a 101.5 MHz clock. However, if
SS_MODE (DOWN_HIGH) produces a 3% down spread, the input frequency is the highest
frequency within the frequency deviation. Consequently, for a 100 MHz clock with 3% down
spread, the down-spread clock would continue to be analyzed by timing analysis as a
100 MHz clock.

Table 3-5: Manual SS Timing Adjustment Using Input Frequency for UltraScale Devices
Input Frequency Input Frequency Adjustment
Parameter M
(MHz) (FIN_SS)
25 < F IN < 35 M = 28 FIN_SS = F IN x 56/55
M = 21 FIN_SS = F IN x 42/41
35 < F IN < 50
M = 22 FIN_SS = FIN x 44/43
SS_MODE(CENTER_HIGH)
50 < F IN < 75 M = 28 FIN_SS = F IN x 56/55
M = 21 FIN_SS = F IN x 42/41
75 < F IN < 150
M = 22 FIN_SS = F IN x 44/43
25 < F IN < 35 M = 56 FIN_SS = F IN x 112/111
M = 42 FIN_SS = F IN x 84/83
35 < F IN < 50
M= 44 FIN_SS = FIN x 88/87
SS_MODE (CENTER_LOW)
50 < F IN < 75 M = 56 FIN_SS = F IN x 112/111
M = 42 FIN_SS = F IN x 84/83
75 < F IN < 150
M = 44 FIN_SS = F IN x 88/87

UltraScale Architecture Clocking Resources Send Feedback

62
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Table 3-5: Manual SS Timing Adjustment Using Input Frequency for UltraScale Devices (Cont’d)
Input Frequency Input Frequency Adjustment
Parameter M
(MHz) (FIN_SS)
25 < F IN < 35 M=28 FIN_SS = F IN
35 < F IN < 50 M = 21, 22 FIN_SS = F IN
SS_MODE (DOWN_HIGH) 50 < F IN < 75 M = 28 FIN_SS = F IN
75 < F IN < 100 M = 21, 22 FIN_SS = F IN
100 < F IN < 150 M = 21, 22 FIN_SS = F IN
25 < F IN < 35 M = 56 FIN_SS = F IN
35 < F IN < 50 M = 42, 44 FIN_SS = F IN
SS_MODE (DOWN_LOW) 50 < F IN < 75 M = 56 FIN_SS = F IN
75 < F IN < 100 M = 42, 44 FIN_SS = F IN
100 < F IN < 150 M = 42, 44 FIN_SS = F IN

Table 3-6: Manual SS Timing Adjustment Using Input Frequency for UltraScale+ Devices
Input Frequency Input Frequency Adjustment
Parameter M D
(MHz) (FIN_SS)
30 < F IN < 40 M = 28 D=1 FIN_SS = F IN x 56/55
M = 21 D=1 FIN_SS = F IN x 42/41
40 < F IN < 60
M = 22 D=1 FIN_SS = FIN x 44/43
SS_MODE 60 < F IN < 80 M = 28 D=2 FIN_SS = F IN x 56/55
(CENTER_HIGH)
M = 21 D=2 FIN_SS = F IN x 42/41
80 < F IN < 120
M = 22 D=2 FIN_SS = F IN x 44/43
M = 21 D=3 FIN_SS = F IN x 42/41
120 < F IN < 150
M = 22 D=3 FIN_SS = F IN x 44/43
30 < F IN < 40 M = 56 D=2 FIN_SS = F IN x 112/111
M = 42 D=2 FIN_SS = F IN x 84/83
40 < F IN < 60
M = 44 D=2 FIN_SS = F IN x 88/87

SS_MODE 60 < F IN < 80 M = 56 D=4 FIN_SS = F IN x 112/111

(CENTER_LOW) M = 42 D=4 FIN_SS = F IN x 84/83
80 < F IN < 120
M = 44 D=4 FIN_SS = F IN x 88/87
M = 42 D=6 FIN_SS = F IN x 84/83
120 < F IN < 150
M = 44 D=6 FIN_SS = F IN x 88/87
35 < F IN < 40 M=28 D=1 FIN_SS = F IN
40 < F IN < 60 M = 21, 22 D=1 FIN_SS = F IN
SS_MODE
60 < F IN < 80 M = 28 D=2 FIN_SS = F IN
(DOWN_HIGH)
80 < F IN < 120 M = 21, 22 D=2 FIN_SS = F IN
120 < F IN < 150 M = 21, 22 D=3 FIN_SS = F IN

UltraScale Architecture Clocking Resources Send Feedback

63
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Table 3-6: Manual SS Timing Adjustment Using Input Frequency for UltraScale+ Devices (Cont’d)
Input Frequency Input Frequency Adjustment
Parameter M D
(MHz) (FIN_SS)
35 < F IN < 40 M = 56 D=2 FIN_SS = F IN
40 < F IN < 60 M = 42, 44 D=2 FIN_SS = F IN
SS_MODE
60 < F IN < 80 M = 56 D=4 FIN_SS = F IN
(DOWN_LOW)
80 < F IN < 120 M = 42, 44 D=4 FIN_SS = F IN
120 < F IN < 150 M = 42, 44 D=6 FIN_SS = F IN

For a 25 MHz input clock, the new timing constraints would be:

• SS_MODE(CENTER_HIGH) = 25 x 56/55 = 25.45 MHz

• SS_MODE (CENTER_LOW) = 25 x 112/111 = 25.23 MHz
• SS_MODE (DOWN_HIGH) = 25 MHz
• SS_MODE (DOWN_LOW) = 25 MHz

For an 80 MHz input clock, the new timing constraints would be:

• SS_MODE(CENTER_HIGH) = 80 x 44/43 = 81.86 MHz

• SS_MODE (CENTER_LOW) = 80 x 88/87 = 80.92 MHz
• SS_MODE (DOWN_HIGH) = 80 MHz
• SS_MODE (DOWN_LOW) = 80 MHz

Table 3-5 and Table 3-6 provide information which allows the manual adjustment of timing
constraints to the frequency range of the spread-spectrum enabled clock. This is for the
generation of timing constraints in an XDC file used by Vivado tools.

Table 3-5 and Table 3-6 show that timing constraints should be modified when
spread-spectrum clocking parameter SS_MODE is set to CENTER_LOW or CENTER_HIGH.
When SS_MODE attribute is set to DOWN_LOW or DOWN_HIGH timing constraint
adjustment is not necessary.

Also note that manually adjusting timing constraints is not needed because the Vivado
tools detect when spread-spectrum clocking in a design. Vivado tools (static timing
analysis) automatically account for any timing spread caused by the spread-spectrum
enabled clocks. When spread-spectrum clocks are used, Vivado static timing analysis adds
a spread-spectrum (SS) uncertainty value of the total uncertainty calculation formula. The
formula used by the static analysis tools is as follows:

Equation 3-12
1
  TS J 2 – D J 2 -2- 
-------------------------------- + PE + SS
2

UltraScale Architecture Clocking Resources Send Feedback

64
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

where:

• TS J2 = Total system jitter

• D J2 = Discrete jitter
• PE = Phase error
• SS = Spread-spectrum

CAUTION! When using spread-spectrum clocking in a design, it is necessary to use appropriated clock
domain crossing (CDC) circuitry for all signals, data and non-data, crossing clock and spread-spectrum
clock domains, and vice versa.

Asynchronous FIFOs should be used to transfer data between two clock domains. The depth
of the FIFO depends on the modulation frequency in the clock. The slower the modulation,
the deeper the FIFO needs to be:
frequencyDeviation
FIFO depth is proportional to = --------------------------------------------------- Equation 3-13
ModulationFrequency

When spread-spectrum clocking is used with SS_MODE set as DOWN_LOW or DOWN_HIGH

the calculated F~IN_SS~ frequency (using data from Table 3-5 and/or Table 3-6) is lower
than the original clock frequency (Refer to the examples after Table 3-6). If no precautions
are taken, the used FIFO can fill up and over-run. Prevent this by using a FIFO with throttle
control.

Table 3-7: Spread-Spectrum Generation Restrictions for UltraScale Devices

Parameter Value
Minimum 25 [KHz]
F MODULATION
Maximum 250 [KHz]
Minimum 25 [MHz]
Input Clock Frequency
Maximum 150 [MHz]
M = 28
25 MHz < F IN < 35 MHz
D=1
M = 21, 22
35 MHz < F IN < 50 MHz
D=1
M = 28
SS_MODE (CENTER_HIGH) 50 MHz < F IN < 75 MHz
D=2
M = 21, 22
75 MHz < F IN < 100 MHz
D=2
M = 21, 22
100 MHz < F IN < 150 MHz
D=3

UltraScale Architecture Clocking Resources Send Feedback

65
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Table 3-7: Spread-Spectrum Generation Restrictions for UltraScale Devices (Cont’d)

Parameter Value
M = 56
25 MHz < F IN < 35 MHz
D=2
M = 42, 44
35 MHz < F IN < 50 MHz
D=2
M = 56
SS_MODE (CENTER_LOW) 50 MHz < F IN < 75 MHz
D=4
M = 42, 44
75 MHz < F IN < 100 MHz
D=4
M = 42, 44
100 MHz < F IN < 150 MHz
D=6
M = 28
25 MHz < F IN < 35 MHz
D=1
M = 21, 22
35 < F IN < 50 MHz
D=1
M = 28
SS_MODE (DOWN_HIGH) 50 MHz < F IN < 75 MHz
D=2
M = 21, 22
75 MHz < F IN < 100 MHz
D=2
M = 21, 22
100 MHz < F IN < 150 MHz
D=3
M = 56
25 MHz < F IN < 35 MHz
D=2
M = 42, 44
35 MHz < F IN < 50 MHz
D=2
M = 56
SS_MODE (DOWN_LOW) 50 MHz < F IN < 75 MHz
D=4
M = 42, 44
75 MHz < F IN < 100 MHz
D=4
M = 42, 44
100 MHz < F IN < 150 MHz
D=6
CLKOUT[3:2]_DIVIDE N/A
CLKOUT[6:4,1,0]_DIVIDE 1 to 128
Bandwidth Low

Table 3-8: Spread-Spectrum Generation Restrictions for UltraScale+ Devices

Parameter Value
Minimum 30 [KHz]
F MODULATION
Maximum 250 [KHz]
Minimum 25 [MHz]
Input Clock Frequency
Maximum 150 [MHz]

UltraScale Architecture Clocking Resources Send Feedback

66
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Table 3-8: Spread-Spectrum Generation Restrictions for UltraScale+ Devices (Cont’d)

Parameter Value
M = 28
30 MHz < FIN < 40 MHz
D=1
M = 21, 22
40 MHz < FIN < 60 MHz
D=1
M = 28
SS_MODE (CENTER_HIGH) 60 MHz < FIN < 80 MHz
D=2
M = 21, 22
80 MHz < F IN < 120 MHz
D=2
M = 21, 22
120 MHz < FIN < 150 MHz
D=3
M = 56
30 MHz < FIN < 40 MHz
D=2
M = 42, 44
40 MHz < FIN < 60 MHz
D=2
M = 56
SS_MODE (CENTER_LOW) 60 MHz < FIN < 80 MHz
D=4
M = 42, 44
80 MHz < F IN < 120 MHz
D=4
M = 42, 44
120 MHz < FIN < 150 MHz
D=6
M = 28
30 MHz < FIN < 40 MHz
D=1
M = 21, 22
40 MHz < F IN < 60 MHz
D=1
M = 28
SS_MODE (DOWN_HIGH) 60 MHz < FIN < 80 MHz
D=2
M = 21, 22
80 MHz < F IN < 120 MHz
D=2
M = 21, 22
120 MHz < FIN < 150 MHz
D=3
M = 56
30 MHz < FIN < 40 MHz
D=2
M = 42, 44
40 MHz < FIN < 60 MHz
D=2
M = 56
SS_MODE (DOWN_LOW) 60 MHz < FIN < 80 MHz
D=4
M = 42, 44
80 MHz < F IN < 120 MHz
D=4
M = 42, 44
120 MHz < FIN < 150 MHz
D=6
CLKOUT[3:2]_DIVIDE N/A

UltraScale Architecture Clocking Resources Send Feedback

67
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Table 3-8: Spread-Spectrum Generation Restrictions for UltraScale+ Devices (Cont’d)

Parameter Value
CLKOUT[6:4,1,0]_DIVIDE 1 to 128
Bandwidth Low

When using spread-spectrum generation, the VCO frequency is set by the clocking wizard
based on the input frequency and SS_MODE. As a result, the clocking wizard is
recommended to set the output frequencies for CLKOUT[6:4,1,0].

Based on the VCO frequency and SS_MOD_PERIOD, the clocking wizard also determines the
correct modulation settings to set the modulation frequency within 10% of
SS_MOD_PERIOD. Because the modulation frequency is dependent on the VCO frequency,
the modulation frequency scales as the input frequency changes for a given compilation.

UltraScale Architecture Clocking Resources Send Feedback

68
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

MMCM Application Example

These MMCM attribute settings result in a wide variety of synthesized clocks:

CLKOUT0_PHASE = 0;
CLKOUT0_DUTY_CYCLE = 0.5;
CLKOUT0_DIVIDE = 2;
CLKOUT1_PHASE = 90;
CLKOUT1_DUTY_CYCLE = 0.5;
CLKOUT1_DIVIDE = 2;
CLKOUT2_PHASE = 0;
CLKOUT2_DUTY_CYCLE = 0.25;
CLKOUT2_DIVIDE = 4;
CLKOUT3_PHASE = 90;
CLKOUT3_DUTY_CYCLE = 0.5;
CLKOUT3_DIVIDE = 8;
CLKOUT4_PHASE = 0;
CLKOUT4_DUTY_CYCLE = 0.5;
CLKOUT4_DIVIDE = 8;
CLKOUT5_PHASE = 135;
CLKOUT5_DUTY_CYCLE = 0.5;
CLKOUT5_DIVIDE = 8;
CLKFBOUT_PHASE = 0;
CLKFBOUT_MULT_F = 8;
DIVCLK_DIVIDE = 1;
CLKIN1_PERIOD = 10.0;

Figure 3-15 displays the resulting waveforms.

X-Ref Target - Figure 3-15

REFCLK

VCOCLK

CLKOUT0

CLKOUT1

CLKOUT2

CLKOUT3

CLKOUT4

CLKOUT5
X16696-111516

Figure 3-15: Example Waveform

UltraScale Architecture Clocking Resources Send Feedback

69
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

PLLs
There are two PLLs per CMT that provide clocking to the PHY logic and I/Os. In addition,
they can be used as frequency synthesizers for a wide range of frequencies, serve as jitter
filters, and provide basic phase shift capabilities and duty cycle programming. The PLLs
differ from the MMCM in number of outputs, cannot deskew clock nets, and do not have
advanced phase shift capabilities, Multipliers and input dividers have a smaller value range
and do not have many of the other advanced features of the MMCM.

General Usage Description

PLL Primitives
The UltraScale device PLL primitives, PLLE3_BASE and PLLE3_ADV, are shown in Figure 3-16.
For UltraScale+ devices have the same primitives with an E4 instead of an E3. In this user
guide, PLLE4_ADV is the same as the PLLE3_ADV, and PLLE4_BASE is the same as
PLLE3_BASE.
X-Ref Target - Figure 3-16

CLKIN CLKOUT0 CLKIN CLKOUT0

CLKFBIN CLKOUT0B CLKFBIN CLKOUT0B

CLKOUTPHYEN CLKOUT1 CLKOUTPHYEN CLKOUT1

RST CLKOUT1B RST CLKOUT1B

CLKFBOUT PWRDWN CLKFBOUT

PWRDWN
CLKOUTPHY DADDR[6:0] CLKOUTPHY
DCLK LOCKED
LOCKED
DEN DO[15:0]

PLLE3_BASE or PLLE4_BASE DI[15:0] DRDY

DWE
PLLE3_ADV or PLLE4_ADV
X16697-111516

Figure 3-16: PLL Primitives

UltraScale Architecture Clocking Resources Send Feedback

70
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

PLLE3_BASE and PLLE4_BASE Primitive

The PLLE#_BASE primitive provides access to the most frequently used features of a
stand-alone PLL. Clock deskew, frequency synthesis, and duty cycle programming are
available to use with the PLLE#_BASE. The ports are listed in Table 3-9.

Table 3-9: PLLE#_BASE Ports

Description Ports
Clock input CLKIN, CLKFBIN
Control inputs RST, CLKOUTPHYEN
Clock output CLKOUT0, CLKOUT1, CLKOUT0B, CLKOUT1B, CLKOUTPHY, CLKFBOUT
Status and data outputs LOCKED
Power control PWRDWN

PLLE3_ADV and PLLE4_ADV Primitive

The PLLE#_ADV primitive provides access to all PLLE#_BASE features plus additional ports
for access to the DRP. The ports are listed in Table 3-10.

Table 3-10: PLLE#_ADV Ports

Description Ports
Clock input CLKIN, DCLK, CLKFBIN
Control and data input RST, CLKOUTPHYEN, DWE, DEN, DADDR, DI
Clock output CLKOUT0, CLKOUT1, CLKOUT0B, CLKOUT1B, CLKOUTPHY, CLKFBOUT
Status and data output LOCKED, DO, DRDY
Power control PWRDWN

PLL Ports
Table 3-11 summarizes the PLL ports.

Table 3-11: PLL Ports

Pin Name I/O Pin Description
CLKIN Input General clock input.
RST Input Asynchronous reset signal. The RST signal is an asynchronous reset for the PLL.
The PLL synchronously re-enables itself when this signal is released (i.e., PLL
re-enabled). A reset is required when the input clock conditions change (e.g.,
frequency).
PWRDWN Input Powers down instantiated but unused PLLs. See PWRDWN – Power Down,
page 72.
CLKOUT[0:1] Output User configurable clock outputs 0 and 1 and their inverted versions. The
CLKOUT[0:1]B CLKOUT can be divided versions of the VCO phase outputs (user controllable)
from 1 (bypassed) to 128.

UltraScale Architecture Clocking Resources Send Feedback

71
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Table 3-11: PLL Ports (Cont’d)

Pin Name I/O Pin Description
CLKFBOUT Output Dedicated PLLE3 or PLLE4 feedback output.
CLKFBIN Input Feedback clock input.
CLKOUTPHYEN Input Enable PHY clocking.
CLKOUTPHY Output Dedicated PHY clock.
LOCKED Output An output from the PLL that indicates when the PLL has achieved phase
alignment within a predefined window and frequency matching within a
predefined PPM range. The PLL automatically locks after power on; no extra
reset is required. LOCKED is deasserted if the input clock stops or the phase
alignment is violated (e.g., input clock phase shift). The PLL must be reset after
LOCKED is deasserted.
DADDR[6:0] Input The dynamic reconfiguration address (DADDR) input bus provides a
reconfiguration address for the dynamic reconfiguration. When not used, all
bits must be assigned zeros.
DI[15:0] Input The dynamic reconfiguration data input (DI) bus provides reconfiguration
data. When not used, all bits must be set to zero.
DWE Input The dynamic reconfiguration write enable (DWE) input pin provides the write
enable control signal to write the DI data into the DADDR address. When not
used, DWE must be tied Low.
DEN Input The dynamic reconfiguration enable (DEN) provides the enable control signal
to access the dynamic reconfiguration feature. When the dynamic
reconfiguration feature is not used, DEN must be tied Low.
DCLK Input The DCLK signal is the reference clock for the dynamic reconfiguration port.
DO[15:0] Output The dynamic reconfiguration output bus provides PLL data output when using
dynamic reconfiguration.
DRDY Output The dynamic reconfiguration ready output (DRDY) provides the response to
the DEN signal for the PLL’s dynamic reconfiguration feature.

PLL Port Descriptions

CLKIN – Reference Clock Input

This is a general clock input to the PLL.

RST – Asynchronous Reset Signal

The RST signal is an asynchronous reset for the PLL. The PLL is synchronously re-enabled
when this signal is deasserted.

PWRDWN – Power Down

This signal powers down instantiated but currently unused PLLs. This mode can be used to
save power for temporarily inactive portions of the design and/or PLLs that are not active in
certain system configurations. No PLL power is consumed in this mode.

UltraScale Architecture Clocking Resources Send Feedback

72
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

CLKOUT[0:1] – Output Clocks

These are user-configurable clock outputs and can be divided versions of the VCO phase
outputs (user controllable) from 1 (bypassed) to 128. The input clock and output clocks can
be phase aligned.

CLKFBOUT – Dedicated PLL Feedback Output

For the possible configurations of CLKFBOUT, see Figure 3-17 and Figure 3-18. Unlike the
MMCM, the CLKFBOUT cannot drive logic.
X-Ref Target - Figure 3-17

1 Local Clock 2 1
CLKIN CLKOUT0 5
Network
CLKOUTPHYEN CLKOUT0B
2
RST CLKOUT1

BUF_IN PWRDWN CLKOUT1B 3

Local Clock 3
CLKFBIN CLKOUTPHY_P
Network
4
Compensation CLKOUTPHY_N

LOCKED 5

CLKFBOUT 4
PLLE3

X18045-111516

Figure 3-17: Clock Deskew Using BUF_IN Compensation Mode

X-Ref Target - Figure 3-18

CLKIN CLKOUT0

CLKOUTPHYEN CLKOUT0B
IBUFG BUFG
RST CLKOUT1

PWRDWN CLKOUT1B

CLKFBIN CLKOUTPHY_P

CLKOUTPHY_N

LOCKED

CLKFBOUT
PLLE3

X18046-111516

Figure 3-18: PLL Internal Feedback

UltraScale Architecture Clocking Resources Send Feedback

73
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

CLKFBIN – Feedback Clock Input

CLKFBIN must be connected either directly to the CLKFBOUT for internal feedback, or to the
CLKFBOUT through a BUF_IN. Using BUF_IN in the feedback path compensates for the clock
network delay in the same XIPHY bank as shown in Figure 3-17 where the nodes 1 and 5 are
phase aligned.

CLKOUTPHY – PHY Clock Output

This is a dedicated clock output for use by the PHY byte logic and I/O. It can be 2X, 1X, or
0.5X of the VCO frequency.

CLKOUTPHYEN – PHY Clock Enable

CLKOUTPHYEN enables the CLKOUTPHY clock outputs. The PLL employs enable logic to
synchronize the asynchronous CLKOUTPHYEN signal from your design and controls when
the CLKOUTPHY clocks are released. After the CLKOUTPHY clock is released, the rising edge
is aligned to the rising edge of the input clock CLKIN. Glitch-free enabling and disabling of
the CLKOUTPHY output clock is assured for all configurations.

However, phase alignment between multiple PLL CLKOUTPHY clocks is only assured when
both the CLKFBOUT_MULT and CLKOUT[0:1]_DIVIDE values are set to 1, 2, 4, or 8. Rising
edges do not align for CLKFBOUT = 3, 5, 6, 7, 9,...

CLKOUT[0:1]B – Inverted Output Clocks

This is the inverted (180° phase shift) CLKOUT[0:1].

LOCKED

This output from the PLL is used to indicate when the PLLs have achieved frequency
alignment of the reference clock and the internal feedback. Frequency alignment is within a
predefined window of frequency matching within a predefined PPM range. The PLL
automatically locks after power on; no extra reset is required. LOCKED is deasserted within
one PFD clock cycle if the input clock stops or the frequency has changed. The PLL must be
reset when LOCKED is deasserted. The clock outputs should not be used prior to the
assertion of LOCKED.

DADDR[6:0] – Dynamic Reconfiguration Address

The dynamic reconfiguration address (DADDR) input bus provides a reconfiguration

address for the dynamic reconfiguration. The address value on this bus specifies the 16
configuration bits that are written or read with the next DCLK cycle. When not used, all bits
must be assigned zeros.

DI[15:0] – Dynamic Reconfiguration Data Input

The dynamic reconfiguration data input (DI) bus provides reconfiguration data. The value of
this bus is written to the configuration cells. The data is presented in the cycle that DEN and

UltraScale Architecture Clocking Resources Send Feedback

74
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

DWE are active. The data is captured in a shadow register and written at a later time. DRDY
indicates when the DRP port is ready to accept another write. When not used, all bits must
be set to zero.

DWE – Dynamic Reconfiguration Write Enable

The dynamic reconfiguration write enable (DWE) input pin provides the write/read enable
control signal to write the DI data into or read the DO data from the DADDR address. When
not used, DWE must be tied Low.

DEN – Dynamic Reconfiguration Enable Strobe

The dynamic reconfiguration enable strobe (DEN) provides the enable control signal to
access the dynamic reconfiguration feature and enable all DRP port operations. When the
dynamic reconfiguration feature is not used, DEN must be tied Low.

DCLK – Dynamic Reconfiguration Reference Clock

DCLK is the reference clock for the dynamic reconfiguration port. The rising edge of this
signal is the timing reference for all other port signals. The setup time is specified in the
UltraScale device data sheets [Ref 5]. There is no hold time requirement for the other input
signals relative to the rising edge of DCLK. This signal can be driven by an IBUF, IBUFG,
BUFGCE, or BUFGCTRL. There are no dedicated connections to this clock input.

DO[15:0] – Dynamic Reconfiguration Output Bus

The dynamic reconfiguration output bus provides PLL data output when using dynamic
reconfiguration. If DWE is inactive while DEN is active at the rising edge of DCLK, this bus
holds the content of the configuration cells addressed by DADDR. The DO bus must be
captured on the rising edge of DCLK when DRDY is active. The DO bus value is held until the
next DRP operation.

DRDY – Dynamic Reconfiguration Ready

The dynamic reconfiguration ready output (DRDY) provides the response to the DEN signal
for the PLL’s dynamic reconfiguration feature. This signal indicates that a DEN/ DCLK
operation has completed.

UltraScale Architecture Clocking Resources Send Feedback

75
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

PLL Attributes
Table 3-12 lists the attributes for the PLLE#_BASE and PLLE#_ADV primitives.

Table 3-12: PLL Attributes

Attribute Type Allowed Values Default Description
CLKOUT[0:1]_DIVIDE Integer 1 to 128 1 Specifies the amount to divide the
associated CLKOUT clock output if a
different frequency is desired. This number,
in combination with the CLKFBOUT_MULT
values, determines the output frequency.
CLKOUT[0:1]_ Real 0.01 to 0.99 0.50 Specifies the duty cycle of the associated
DUTY_CYCLE CLKOUT clock output in percentages (i.e.,
0.50 generates a 50% duty cycle).
CLKFBOUT_MULT Boolean 1 to 19 5 This CLKFBOUT_MULT range applies to
UltraScale devices. Specifies the amount to
multiply all CLKOUT clock outputs if a
different frequency is desired. This number,
in combination with the associated
CLKOUT#_DIVIDE value and DIVCLK_DIVIDE
value, determines the output frequency.
CLKFBOUT_MULT Integer 2 to 21 5 This CLKFBOUT_MULT range applies to
UltraScale+ devices. Specifies the amount to
multiply all CLKOUT clock outputs if a
different frequency is desired. This number,
in combination with the associated
CLKOUT#_DIVIDE value and DIVCLK_DIVIDE
value, determines the output frequency.
DIVCLK_DIVIDE Integer 1 to 15 1 Specifies the division ratio for all output
clocks with respect to the input clock.
REF_JITTER Real 0.000 to 0.999 0.010 Allows specification of the expected jitter on
the reference clock to better optimize PLL
performance. A bandwidth setting of
OPTIMIZED attempts to choose the best
parameter for input clocking when unknown.
If known, the value provided should be
specified in terms of the unit interval (UI)
(the maximum peak-to-peak value) of the
expected jitter on the input clock.
CLKIN_PERIOD Real 0.938 to 14.286 0.000 Specifies the input period in ns to the PLL
CLKIN input. Resolution is down to the ps.
This information is mandatory and must be
supplied.
STARTUP_WAIT String FALSE, TRUE FALSE Wait during the configuration start-up cycle
for the PLL to lock.

UltraScale Architecture Clocking Resources Send Feedback

76
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Table 3-12: PLL Attributes (Cont’d)

Attribute Type Allowed Values Default Description
CLKOUT[0:1]_PHASE Real –360.000 to 0.000 Allows specification of the output phase
360.000 relationship of the associated CLKOUT clock
output in number of degrees offset (i.e., 90
indicates a 90° offset or ¼ cycle phase offset
while 180 indicates a 180° offset or ½ cycle
phase offset). Valid phase shifts are in
360 CLKOUT[0:1]_DIVIDE degree
increments.
CLKFBOUT_PHASE Real –360.000 to Specifies the phase offset in degrees of the
360.000 clock feedback output. Shifting the feedback
clock results in a negative phase shift of all
output clocks to the PLL. Valid phase shifts
are in 360 CLKFBOUT_MULT degree
increments.
COMPENSATION String AUTO(1), AUTO Clock input compensation. Must be set to
PHY_ALIGN, AUTO. Defines how the PLL feedback is
BUF_IN, configured.
INTERNAL INTERNAL: Indicates that the PLL is using its
own internal feedback path so no delay is
being compensated.

BUF_IN: Indicates that the clock network

delay within the same XIPHY bank is
compensated. Both CLKIN and CLKFB must
be the same frequency at the PFD (F IN/D =
F VCO /M).
The feedback must be limited to within the
same XIPHY and cannot be routed to
adjacent banks or outside the device.

PHY_ALIGN (for UltraScale+ devices only):

allows fine- grained adjustments of the PLL
output phase for the alignment of the XIPHY
and internal logic flip-flops using the same
source clock. Primarily for memory
controller use.
CLKOUTPHY_MODE String VCO_2X, VCO_2X Determines the clock output frequency
VCO, based on the VCO frequency for the
VCO_HALF BITSLICE_CONTROL block.

Notes:
1. The specifications for the VCO frequencies PLL_FVCOMIN/PLL_FVCOMAX and minimum out frequency PLL_FOUTMIN are
different for the UltraScale and UltraScale+ families. Consult the appropriate data sheets.

UltraScale Architecture Clocking Resources Send Feedback

77
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Dynamic Reconfiguration Port

In most circumstances, the MMCM and/or PLL used in a design are configured using static
calculated values to setup the used outputs. A wizard can be used to calculate all the values
and generate an instantiable wrapper containing a configured MMCM or PLL. The MMCM
and/or PLL primitive can also be instantiated as primitives and the values to make the
primitive function correctly can be calculated using the equations provided in the MMCM
Programming section.

The DRP port provides the ability to use a MMCM and/or PLL as a dynamic element in a
design. The DRP port setup is that of a common microcontroller peripheral and gives the
user access to a set of registers in the MMCM or PLL. These registers allow the user to fully
control the MMCM or PLL. Inputs pins and the values to define output clocks are turned
into register bits making it possible to use the primitives as active elements in a design.

Using the DRP port means reading and writing of registers of a peripheral. When using the
Clocking Wizard the DRP port can be enabled through an AXI-Lite controller to a hard or
soft microcontroller in the FPGA. Nevertheless, it might be necessary by design and other
requirements to use the DRP port in a bare metal configuration (also selectable in the
Clocking Wizard). The DRP port can then be used as such through a state machine based
design. To help with this the provided description of the functioning of the DRP port can be
used.

For additional DRP usage information, see MMCM and PLL Dynamic Reconfiguration
(XAPP888) [Ref 6] and the associated reference.

The DRP port connections are shown in the following figure.

X-Ref Target - Figure 3-19

DI[15:0] DOUT[15:0]
DADDR[n:0] DRDY
DWE
DEN
DCLK

X21932-111618

Figure 3-19: DRP Port Connections

UltraScale Architecture Clocking Resources Send Feedback

78
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Table 3-13: DRP Port Signals

Port Size I/O Description
The DCLK signal is the reference clock for the
dynamic reconfiguration port. This clock is
DCLK 1 Input normally about 100 MHz to 200 MHz. The newer
the technology of the FPGA family used, the faster
this clock can be.
The dynamic reconfiguration enable (DEN)
provides the enable control signal to access the
DEN 1 Input dynamic reconfiguration feature. When the
dynamic reconfiguration feature is not used, DEN
must be tied Low.
The dynamic reconfiguration write enable (DWE)
input pin provides the write enable control signal
DWE 1 Input to write the data on the DI port into the register
selected by the DADDR address. When not used,
DWE must be tied Low.
The dynamic reconfiguration address (DADDR)
input bus provides a reconfiguration address to
DADDR n(1) Input access a specific register in the primitive for
dynamic reconfiguration. When not used, all bits
must be assigned zeros.
The dynamic reconfiguration data input (DI) bus
provides reconfiguration data that is written into a
DI 16 Input
specified address (DADDR) of the register set.
When not used, all bits must be set to zero.
The dynamic reconfiguration output bus provides
data output of the register selected by the DADDR
DO 16 Output
bus. This port can be used to control DRP register
contents.
The dynamic reconfiguration ready output (DRDY)
provides the response to the DEN signal for the
DRDY 1 Output PLLs dynamic reconfiguration feature. This signal is
pulsed high when a write or read operation is
successful.

Notes:
1. The width of the DADDR bus depends on the primitive that the DRP port is a part of. For a MMCM, the address bus
is 7-bit wide and for a PLL the DRP address bus is 7-bit wide (DADDR(6:0)). The DADDR port of an ADC/DAC in a
RFSoC device is 12-bit wide while the DADDR port of a GTP is 10-bit wide.

UltraScale Architecture Clocking Resources Send Feedback

79
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Writing to the DRP Port

1. Put the address to write to and the data that needs to be written on the buses,
DADDR[n:0] and DI[15:0] respectively.
2. Make the DWE (Data Write Enable) signal High.
3. Pulse the DEN signal High for one clock cycle. The DEN signal is the trigger that makes
the DRP port function. When this signal is captured on the rising edge of the clock the
internals of the DRP port capture address and data and fill the correct register in the
DRP map.
4. When the DEN signal goes Low for one pulse, make the DWE signal Low.
5. The DRP ports pulse the DRDY High for a clock cycle to confirm that the provided data
is written into the provided address space. This also signals that a new write or read
operation can start.
X-Ref Target - Figure 3-20

DCLK 1 2 3 n n+1

DEN

DWE

DADDR[n:0]

DI[15:0]

DO[15:0]

DRDY

Address and Data valid

WRITE

X21935-111618

Figure 3-20: Writing to the DRP Port

Reading from the DRP Port

1. Put the address of the register to read from on the DADDR[n:0] bus.
2. Leave the DWE signal Low at all times when reading.
3. The value ON/OFF of the DI[15:0] bus does not matter.
4. Pulse the DEN signal High for one clock cycle. The DEN signal is the trigger that makes
the DRP port function. When this signal is captured on the rising edge of the clock the
internals of the DRP port capture the address to make sure that the contents of the
correct register in the DRP map are reflected on the DO[15:0] output.
5. The DRP ports pulse the DRDY High for a clock cycle to confirm that the provided data
is written into the provided address space. This also signals that a new write or read
operation can start.

UltraScale Architecture Clocking Resources Send Feedback

80
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

X-Ref Target - Figure 3-21

DCLK 1 2 3 n n+1

DEN

DWE

DADDR[n:0]

DI[15:0]

DO[15:0]

DRDY

Address valid
READ Data valid

X21937-111618

Figure 3-21: Reading from the DRP Port

Read—Write Operation
A read-write operation must always be executed with respect to the DRDY signal. Only
when the DRDY signal pulses High, a new read or write operation can be initiated. If the
DRDY signal is not controlled after a read or write operation, from or to the DRP port, it is
not certain that the written bits are set or the obtained bits are representing the value of the
register.
X-Ref Target - Figure 3-22

DCLK 1 2 3 n n+1 1 2 3 n n+1 1 2 3 n n+1 1 2 3 n n+1

DEN

DWE

DADDR[11:0] Addr_N Addr_M Addr_N Addr_M

DI[15:0] Data_DD Data_EE

DO[15:0] Data_DD Data_EE

DRDY

WRITE WRITE READ READ

X21936-111618

Figure 3-22: Read—Write Operation

UltraScale Architecture Clocking Resources Send Feedback

81
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

DRP Register Set

The user accessible DRP register set is described in this section. The DRP register map spans
from address 0x00 to address 0x7F (7-bit address bus). The figure below shows the layout
of the register map of user accessible registers. Be aware that values of different counters
overlap register boundaries.
X-Ref Target - Figure 3-23

MMCM PLL
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

4F loop filter R 4F loop filter R

4E loop filter C 4E loop filter C

27 interpolator

1A lock ref dly 1A lock ref dly

19 lock fb dly 19 lock fb dly
18 lock cnt 18 lock cnt

16 divck 16 divck
15 15
ckfbout
14 ckfbout 14
13
13
ckout6
12
11
ckout4
10
0F
ckout3
0E
0D
ckout2
0C
0B 0B ckoutphy
ckout1
0A 0A ckout1
9 9
ckout0
8 ckout0 8
7
7
ckout5
6
5 compensation 5 compensation
4 spread spectrum

0 inversion control 0 inversion control

These register positions are not user accessible.

X21942-073020

Figure 3-23: MMCM DRP Register Set

IMPORTANT: When operating a DRP port, it is recommended that the existing contents of the register
that is going to be changed are first read. Write back the register contents where only the required bits
are modified. Modify only the colored bits and always maintain the state of the gray bits.

UltraScale Architecture Clocking Resources Send Feedback

82
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

MMCM Registers
Reg 4F ADDR: 0x4F
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W
15 mc_res(3)
12 mc_res(2)
Loop filter resistor setting.
11 mc_res(1)
8 mc_res(0)
7 mc_lfhf(1)
Loop filter high frequency capacitor setting.
4 mc_lfhf(0)

Reg 4E ADDR: 0x4E

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 0
Access R/W R/W R/W R/W
15 mc_cp(3)
12 mc_cp(2)
Charge pump settings.
11 mc_cp(1)
8 mc_cp(0)

Registers 0x4F and 0x4E define the values for the loop filters. Pick the appropriate values
for these filters from the MMCM and PLL Dynamic Reconfiguration (XAPP888) [Ref 6].

UltraScale Architecture Clocking Resources Send Feedback

83
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Reg 27 ADDR: 0x27

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W
15 mc_interp_en(7)
12 mc_interp_en(6)
11 mc_interp_en(5)
8 mc_interp_en(4)
Interpolator selection. Default value is 00010000.
7 mc_interp_en(3)
4 mc_interp_en(2)
3 mc_interp_en(1)
0 mc_interp_en(0)

Notes:
1. If any of the output counters is using fine phase shift then mc_interp_en[3:0] must be set to 1111 otherwise
mc_interp_en[3:0] must be set to 0000.
2. mc_interp_en(4) is always set to 1.
3. If any of the output counters is using a phase of VCO other than 0 or 180, uses fractional division for a counter, or uses
spread-spectrum mode then mc_interp_en[7:5] must be set to 111 otherwise mc_interp_en[7:5] must be set to 000.

Reg 1A ADDR: 0x1A

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 1 1 1 1 1 1 0 1 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
14:10 mc_lock_ref_dly[4:0] Window setting for the lock circuit of the reference clock.
9:0 mc_lock_sat_high[9:0] Maximum value of the lock counter. Default value is d1001.

Reg 19 ADDR: 0x19

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
14:10 mc_lock_fb_dly[4:0] Window setting for the lock circuit of the feedback clock.
9:0 mc_lock_sat_high[9:0] Counter setting the number of clock cycles the MMCM needs to have
CLKREF and CLKFB misaligned within a certain window before
deasserting the LOCKED output. Default value is 1.

UltraScale Architecture Clocking Resources Send Feedback

84
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Reg 18 ADDR: 0x18

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 1 1 1 1 1 0 1 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
9:0 mc_lock_cnt[9:0] Counter setting the number of clock cycles the MMCM needs to have
CLKREF and CLKFB aligned within a certain window before the LOCKED
output is asserted. Default value is 1000.

Refer to MMCM and PLL Dynamic Reconfiguration (XAPP888) [Ref 6] to determine the values
for registers 0x1A, 0x19, and 0x18.

Reg 16 ADDR: 0x16

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 1 0 0 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
13 mc_divck_edge High to low clock edge transition control.
12 mc_divck_nocount Bypass counter.
11:6 mc_divck_ht[5:0] Counter high time.
5:0 mc_divck_lt[5:0] Counter low time.

Register 0x16 controls the divider (D counter) shown in Figure 3-1.

Reg 15 ADDR: 0x15

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 1 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
14:12 mc_ckfbout_frac[2:0] Fractional phase.
11 mc_ckfbout_frac_en Enable fractional counter.
10 mc_ckfbout_frac_wf_r Fractional mode. Rising edge wait.
7 mc_ckfbout_edge High to low clock edge transition control.
6 mc_ckfbout_nocount Bypass counter.
5:0 mc_ckfbout_frac[5:0] Counter delay.

UltraScale Architecture Clocking Resources Send Feedback

85
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Reg 14 ADDR: 0x14

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
15:13 mc_ckfbout_pm_r[2:0] VCO phase selection mux and rising edge control.
12 mc_ckfbout_en Counter enable
11:6 mc_ckfbout_ht[5:0] Counter high time
5:0 mc_ckfbout_lt[5:0] Counter low time

Reg 13 ADDR: 0x13

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 0 0 0 1 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
15:13 mc_ckfbout_pm_f[2:0] VCO phase selection mux and falling edge control.
12 mc_ckfbout_frac_wf_f Fractional mode. Falling edge wait.
10 mc_ckout6_cddc_en Clock divide dynamic change enable (DRP only)
7 mc_ckout6_edge High to low clock edge control
6 mc_ckout6_nocount Counter bypass
5:0 mc_ckout6_dt[5:0] Counter delay

Registers 0x15, 0x14, and bits [15:12] of 0x13 control the fractional feedback (M counter)
shown in Figure 3-1.

Reg 12 ADDR: 0x12

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
15:13 mc_ckout6_pm[2:0] VCO phase selection multiplexer
12 mc_ckout6_en Counter enable
11:6 mc_ckout6_ht[5:0] Counter high time
5:0 mc_ckout6_lt[5:0] Counter low time

Bits [10:0] of register 0x13 and register 0x12 control the CLKOUT6 counter.

UltraScale Architecture Clocking Resources Send Feedback

86
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Reg 11 ADDR: 0x11

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 1 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W
10 mc_ckout4_cddc_en Clock divide dynamic change control enable (DRP only).
7 mc_ckout4_edge High to low clock edge control.
6 mc_ckout4_nocount Counter bypass.
5:0 mc_ckout4_dt[5:0] Counter delay.

Reg 10 ADDR: 0x10

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
15:13 mc_ckout4_pm[2:0] VCO phase selection multiplexer.
12 mc_ckout4_en Counter enable.
11:6 mc_ckout4_ht[5:0] Counter high time.
5:0 mc_ckout4_lt[5:0] Counter low time.

Registers 0x11 and 0x10 control the output counter for CLKOUT4.

Reg 0F ADDR: 0x0F

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 1 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W
10 mc_ckout3_cddc_en Clock divide dynamic change control enable (DRP only).
7 mc_ckout3_edge High to low clock edge control.
6 mc_ckout3_nocount Counter bypass.
5:0 mc_ckout3_dt[5:0] Counter delay.

Reg 0E ADDR: 0x0E

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
15:13 mc_ckout3_pm[2:0] VCO phase selection multiplexer.
12 mc_ckout3_en Counter enable.
11:6 mc_ckout3_ht[5:0] Counter high time.
5:0 mc_ckout3_lt[5:0] Counter low time.

UltraScale Architecture Clocking Resources Send Feedback

87
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Registers 0x0F and 0x0E control the output counter for CLKOUT3.

Reg 0D ADDR: 0x0D

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 1 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W
10 mc_ckout2_cddc_en Clock divide dynamic change control enable (DRP only).
7 mc_ckout2_edge High to low clock edge control.
6 mc_ckout2_nocount Counter bypass.
5:0 mc_ckout2_dt[5:0] Counter delay.

Reg 0C ADDR: 0x0C

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
15:13 mc_ckout2_pm[2:0] VCO phase selection multiplexer.
12 mc_ckout2_en Counter enable.
11:6 mc_ckout2_ht[5:0] Counter high time.
5:0 mc_ckout2_lt[5:0] Counter low time.

Registers 0x0D and 0x0C control the output counter for CLKOUT2.

Reg 0B ADDR: 0x0B

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 1 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W
10 mc_ckout1_cddc_en Clock divide dynamic change control enable (DRP only).
7 mc_ckout1_edge High to low clock edge control.
6 mc_ckout1_nocount Counter bypass.
5:0 mc_ckout1_dt[5:0] Counter delay.

UltraScale Architecture Clocking Resources Send Feedback

88
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Reg 0A ADDR: 0x0A

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
15:13 mc_ckout1_pm[2:0] VCO phase selection multiplexer.
12 mc_ckout1_en Counter enable.
11:6 mc_ckout1_ht[5:0] Counter high time.
5:0 mc_ckout1_lt[5:0] Counter low time.

Registers 0x0B and 0x0A control the output counter for CLKOUT1.

Reg 09 ADDR: 0x09

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 0 0 0 1 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
15 mc_ckout0_cddc_en Clock divide dynamic change control enable (DRP only).
14:12 mc_ckout0_frac[2:0] Fractional phase.
11 mc_ckout0_frac_en Enable fractional phase counter.
10 mc_ckout0_frac_wf_r Falling edge wait.
7 mc_ckout0_edge High to low clock edge control.
6 mc_ckout0_nocount Counter bypass.
5:0 mc_ckout0_dt[5:0] Counter delay.

Reg 08 ADDR: 0x8

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
15:13 mc_ckout0_pm_r[2:0] Rising edge control.
12 mc_ckout0_en Counter enable.
11:6 mc_ckout0_ht[5:0] Counter high time.
5:0 mc_ckout0_lt[5:0] Counter low time.

UltraScale Architecture Clocking Resources Send Feedback

89
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Reg 07 ADDR: 0x07

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 0 0 0 1 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
15:13 mc_ckout0_pm_f[2:0] Falling edge control.
12 mc_ckout0_frac_wf_f Rising edge wait.
10 mc_ckout0_cddc_en Clock divide dynamic change control enable (DRP only).
7 mc_ckout5_edge High to low clock edge control.
6 mc_ckout5_nocount Counter bypass.
5:0 mc_ckout5_dt[5:0] Counter delay.

The fractional output counter for CLKOUT0 is controlled by registers 0x09, 0x08, and bits
[15:12] of register 0x07.

Reg 06 ADDR: 0x06

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
15:13 mc_ckout5_pm[2:0] VCO phase selection multiplexer.
12 mc_ckout5_en Counter enable.
11:6 mc_ckout5_ht[5:0] Counter high time.
5:0 mc_ckout5_lt[5:0] Counter low time.

Bits [10:0] of register 0x07 and register 0x06 control the CLKOUT5 counter.

The MMCM clock outputs are all defined by a configurable counter. The parameters
defining the clock outputs CLKOUT6 to CLKOUT1 are explained in the following table. Refer
to MMCM and PLL Dynamic Reconfiguration (XAPP888) [Ref 6] for calculation instructions
and methods.

Table 3-14: Clock Output Parameters

Type Description
CDDC Clock divide dynamic change. The possibility to change the DRP registers without
the need for a MMCM reset. When this option is enabled it functions with the
CDDCREQ and CDDCDONE handshake pins. For more information read MMCM
Clock Divide Dynamic Change.
MX Clock input multiplexer control.
EDGE Clock edge identification. Identify the clock edge used for a high to low transition
of the counter.
NOCOUNT Counter bypass.

UltraScale Architecture Clocking Resources Send Feedback

90
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Table 3-14: Clock Output Parameters (Cont’d)

Type Description
DT Delay time. Counter delay or coarse phase shift setting.
PM VCO phase selection. Used to select one of the eight possible VCO outputs.
EN Counter enable.
HT Counter high time. Set the delay the counter needs to output a high value.
LT Counter low time. Set the delay the counter needs to output a low value.

Two of the counters, CLKFBOUT and CLKOUT0, are fractional counters. A fractional counter
uses two non-fractional counters, an extra state, and adder logic. This is the reason a
fractional counter has two enables (one for each counter to allow non-fractional use) and
two VCO phase selection settings. For the adder and state logic, the VCO phase selection is
extra split in rising and falling settings. Additional register configuration options defining
the fractional counters are listed in the following table.

Table 3-15: Register Configuration Options

Type Description
FRAC Select a VCO phase to operate the fractional counter.
FRAC_EN Enable the fractional counter.
FRAC_WF_R Fractional counter wait for rising edge.
FRAC_WF_L Fractional counter wait for falling edge.
PM_R Select one of the eight VCO phased outputs as rising edge counter clock.
PM_L Select one of the eight VCO phased outputs as falling edge counter clock.

Fractional counter mode is enabled when both mc_ckout_en and mc_ckout_frac_en are set.
Both counters take different phases from the VCO outputs.

Example 1

Assume that a division of 2.5 is required, then:

• Counter A would take phase 0 (0 degrees) of the VCO

• Counter B would take phase 4 (180 degree offset)

Output starts High with rising edge on counter A and goes Low with second rising edge
counter B. It goes High again with second rising edge of counter A after that and so on.

UltraScale Architecture Clocking Resources Send Feedback

91
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

X-Ref Target - Figure 3-24

VCO phase 0
(0 degrees)
VCO phase 4
(180 degrees)

Divide by 2

CLKOUT
Divide by 2.5 State & adder logic assemble a
fractional clock from the two
selected VCO phases using the
real divide value
X21938-111418

Figure 3-24: Fractional Counter Mode Example 1

Example 2

Assume a division by 2.125 is required, then:

• Counter A would take phase 0 (0 degrees) of the VCO

• Counter B would take as start phase 2 (45 degree offset)

Output starts High with rising edge on counter A and goes Low with second rising edge
counter B. It goes High again with next rising edge of counter B, counter B switches to VCO
phase 90 and the output goes Low again by the second rising edge of that phase and so on.
X-Ref Target - Figure 3-25

VCO phase 0
(0 degrees)
VCO phase 1
(45 degrees)
VCO phase 2
(90 degrees)
VCO phase 3
(135 degrees)

CLKOUT
Divide by 2.125 State & adder logic assemble a fractional
Clock from the two selected VCO phases
using the real divide value
X21939-111418

Figure 3-25: Fractional Counter Mode Example 2

UltraScale Architecture Clocking Resources Send Feedback

92
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Reg 04 ADDR: 0x04

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 1 0 0 1 1 1
Access R/W R/W R/W R/W R/W R/W
5:3 mc_ss_steps_init[2:0] Start the correct spread based on the SS_MODE attribute. Default value
is 100.
2:0 mc_ss_steps[2:0] Control the spread of the spread-spectrum clocking based on the
SS_MODE attribute. Default value is 111.

The settings for register 4 are controlled by the SS_MODE attribute of the MMCM. Refer to
the Spread-Spectrum Clock Generation section for detailed information on
spread-spectrum clocking set up and behavior.

ss_steps_init ss_steps
DOWN_LOW 100 011
DOWN_HIGH 100 011
CENTER_LOW 100 111
CENTER_HIGH 100 111

Reg 00 ADDR: 0x00

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0
Access R/W R/W R/W
3 mc_inv_clkfbin Enable inversion on the CLKFBIN input. This is the same as setting the
attribute IS_CLKFBIN_INVERTED to 1.
2 mc_inv_clkin2 Enable inversion on the CLKIN2 input. This is the same as setting the
attribute IS_CLKIN2_INVERTED to 1.
1 mc_inv_clkin1 Enable inversion on the CLKIN1 input. This is the same as setting the
attribute IS_CLKIN1_INVERTED to 1.

Register 0 represents the bits that are also available as attributes of the MMCM primitive.
For the functional explanation of these bits, refer to the MMCM Attributes section.

UltraScale Architecture Clocking Resources Send Feedback

93
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

PLL Registers
The PLL DRP register set is similar and runs parallel with that of the MMCM. The number of
possible changeable registers in the PLL DRP resister set is smaller than that of the MMCM
because the PLL has only two clock outputs and doe not use a selectable VCO output
multiplexer and interpolator.

Reg 73 ADDR: 0x73

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 1
Access R/W R/W R/W
15 mc_gts_wait Wait for the GTS_CFG_B signal before starting the LOCKED process.
14 mc_startup_wait Wait during the configuration start-up cycle for the MMCM to lock.
12 mc_mmcm_en Enable the PLL.

Reg 4F ADDR: 0x4F

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W
15 mc_res(3)
12 mc_res(2) Loop filter resistor setting.
11 mc_res(1)
8 mc_res(0)
7 mc_lfhf(1) Loop filter high frequency capacitor setting.
4 mc_lfhf(0)

Reg 4E ADDR: 0x4E

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W
15 mc_cp(3)
12 mc_cp(2)
Charge pump setttings.
11 mc_cp(1)
8 mc_cp(0)
7 mc_cp_bias_trip_set Control of the low trip point sense circuit.
4 mc_cp_res(1)
Charge pump reference current control.
3 mc_cp_res(0)

UltraScale Architecture Clocking Resources Send Feedback

94
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Reg 19 ADDR: 0x19

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
14:10 mc_lock_fb_dly[4:0] Window setting for the lock circuit on the feedback clock.
9:0 mc_unlock_cnt[9:0] Counter setting the number of clock cycles the PLL needs to have
CLKREF and CLKFB misaligned within a certain window before
deasserting the LOCKED output. Default value is 1.

Reg 18 ADDR: 0x18

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 0 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
9:0 mc_lock_cnt[9:0] Counter setting the number of clock cycles the PLL needs to have
CLKREF and CLKFB aligned within a certain window before the LOCKED
output is asserted. Default value is 1000.

Reg 16 ADDR: 0x16

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 1 0 0 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
13 mc_divck_edge High to low clock edge transition control.
12 mc_divck_nocount Bypass counter.
11:6 mc_divck_ht[5:0] Counter high time.
5:0 mc_divck_lt[5:0] Counter low time.

Reg 15 ADDR: 0x15

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 1 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W
7 mc_ckfbout_edge High to low clock edge transition control.
6 mc_ckfbout_nocount Bypass counter.
5:0 mc_ckfbout_dt[5:0] Counter delay.

UltraScale Architecture Clocking Resources Send Feedback

95
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Reg 14 ADDR: 0x14

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 1 0 0 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
12 mc_ckfbout_en Counter enable.
11:6 mc_ckfbout_ht[5:0] Counter high time.
5:0 mc_ckfbout_lt[5:0] Counter low time.

Reg 0B ADDR: 0x0B

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 1 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
14:13 mc_ckoutphy_mode[1:0] CLKOUTPHY mode (VCO_2X, VCO, VCO_HALF).
7 mc_ckout1_edge High to low clock edge control.
6 mc_ckout1_nocount Counter bypass.
5:0 mc_ckout1_dt[5:0] Counter delay.

Reg 0A ADDR: 0x0A

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 1 0 0 0 0 0 1 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
12 mc_ckout1_en Counter enable.
11:6 mc_ckout1_ht[5:0] Counter high time.
5:0 mc_ckout1_lt[5:0] Counter low time.

Reg 09 ADDR: 0x09

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 1 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W R/W
7 mc_ckout0_edge High to low clock edge control.
6 mc_ckout0_nocount Counter bypass.
5:0 mc_ckout0_dt[5:0] Counter delay.

UltraScale Architecture Clocking Resources Send Feedback

96
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Reg 08 ADDR: 0x08

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 1 0 0 0 0 0 1 0 0 0 0 0 1
Access R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W
12 mc_ckout0_en Counter enable.
11:6 mc_ckout0_ht[5:0] Counter high time.
5:0 mc_ckout0_lt[5:0] Counter low time.

Reg 05 ADDR: 0x05

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0 0 0 0 0 0
Access R/W R/W R/W R/W R/W R/W R/W
15:10 mc_in_dly_set[5:0] Counter delay setting. Control how much delay is inserted in the
9:4 mc_in_dly_mx_dvdd[5:0] path.

3 mc_direct_path_cntrl Reserved.
2 mc_in_dly_en Compensation delay enable.

Reg 00 ADDR: 0x00

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Default 0 0
Access R/W R/W
3 mc_inv_clkfbin Enable inversion on the CLKFBIN input. This is the same as setting the
attribute IS_CLKFBIN_INVERTED to 1.
1 mc_inv_clkin Enable inversion on the CLKIN1 input. This is the same as setting the
attribute IS_CLKIN1_INVERTED to 1.

VHDL and Verilog Templates and the Clocking

Wizard
The VHDL and Verilog code for all clocking resource primitives and Vivado tools language
templates are available in the UltraScale Architecture Libraries Guide (UG974) [Ref 7].

The Clocking Wizard helps to correctly set up the MMCM and PLL resources. Additionally,
the Clocking Wizard reports the jitter and supports phase and frequency synthesis. See
LogiCORE IP Clocking Wizard User Guide (PG065) [Ref 8] for more information.

UltraScale Architecture Clocking Resources Send Feedback

97
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Chapter 3: Clock Management Tile

Clocking Guidelines
Clocking in a design is not just applying clock buffers, instantiating MMCM and/or PLL, and
applying one or a couple of constraints in a XDC file. Clocking and the setup of a clocking
network needs attention. To create a design, that is implemented (synthesize, place, and
route) using all of Vivado Design Suite features, and when downloaded makes the FPGA
function at optimal conditions, follow the guidelines provided in the chapters Clocking
Guidelines and Clock Domain Crossing of the UltraFast Design Methodology Guide for the
Vivado Design Suite (UG949) [Ref 1].

The UltraFast Design Methodology Guide for the Vivado Design Suite (UG949) [Ref 1] offers
a set of best practices intended to help streamline the design process for new devices. The
size and complexity of these designs require specific steps and design tasks to ensure
success at each stage of the design. Following these steps and adhering to the best
practices will help you achieve your desired design goals as quickly and efficiently as
possible. Two other documents that can be useful for designing are:

• UltraFast Design Methodology Quick Reference Guide (UG1231) [Ref 9]

• UltraFast Design Methodology Checklist (XTP301) [Ref 10]

UltraScale Architecture Clocking Resources Send Feedback

98
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Appendix A

Additional Resources and Legal Notices

Xilinx Resources
For support resources such as Answers, Documentation, Downloads, and Forums, see Xilinx
Support.

Solution Centers
See the Xilinx Solution Centers for support on devices, software tools, and intellectual
property at all stages of the design cycle. Topics include design assistance, advisories, and
troubleshooting tips.

References
1. UltraFast Design Methodology Guide for the Vivado Design Suite (UG949)
2. UltraScale Architecture Packaging and Pinout User Guide (UG575)
3. UltraScale Architecture SelectIO Resources User Guide (UG571)
4. Vivado Design Suite User Guide: Using Constraints (UG903)
5. UltraScale and UltraScale+ device data sheets:

° UltraScale Architecture and Products Overview (DS890)

° Zynq UltraScale+ MPSoC Overview (DS891)

° Kintex UltraScale Architecture Data Sheet: DC and AC Switching Characteristics

(DS892)

° Virtex UltraScale Architecture Data Sheet: DC and AC Switching Characteristics

(DS893)

° Kintex UltraScale+ FPGAs Data Sheet: DC and AC Switching Characteristics (DS922)

° Zynq UltraScale+ MPSoC Data Sheet: DC and AC Switching Characteristics (DS925)

UltraScale Architecture Clocking Resources Send Feedback

99
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Appendix A: Additional Resources and Legal Notices

° Virtex UltraScale+ FPGAs Data Sheet: DC and AC Switching Characteristics (DS923)

° Artix UltraScale+ FPGA Data Sheet: DC and AC Switching Characteristics (DS931)

6. MMCM and PLL Dynamic Reconfiguration (XAPP888)
7. UltraScale Architecture Libraries Guide (UG974)
8. LogiCORE IP Clocking Wizard User Guide (PG065)
9. UltraFast Design Methodology Quick Reference Guide (UG1231)
10. UltraFast Design Methodology Checklist (XTP301)
11. Zynq UltraScale+ MPSoC Packaging and Pinout User Guide (UG1075)
12. Zynq UltraScale+ MPSoC Technical Reference Manual (UG1085)
13. Vivado Design Suite User Guide (UG908)

Revision History
The following table shows the revision history for this document.

Date Version Revision

02/01/2023 1.10.2 Editorial updates only. No technical content updates.
08/25/2021 1.10.1 Editorial updates only. No technical content updates.
08/28/2020 1.10 Chapter 3: Updated Figure 3-23.
Updated the table for register 15 in MMCM Registers.
10/31/2019 1.9 Chapter 3: Updated Table 3-4 footnote. Updated Spread-Spectrum Clock Generation
section with new content and equations. Updated Table 3-12. Updated UltraScale+
note on page 30.
12/19/2018 1.8 Chapter 3: Updated the example in Determine the Input Frequency section.
Added the new sections Dynamic Reconfiguration Port and Clocking Guidelines.
04/09/2018 1.7 Chapter 2: Updated the BUFG_GT and BUFG_GT_SYNC section.
Chapter 3: In Table 3-4, updated note 3.
06/06/2017 1.6 Chapter 3: In Table 3-4, updated the description of BUF_IN for the COMPENSATION
attribute on page 53.
03/15/2017 1.5 Chapter 2: Updated the discussion on page 13. Added clarification to the BUFG_GT
and BUFG_GT_SYNC section.
Chapter 3: Updated the Dynamic Phase Shift Interface in the MMCM section. Added
Table 3-6 and Table 3-8. In Table 3-12, updated the descriptions for
CLKOUT[0:1]_PHASE and CLKFBOUT_PHASE.

UltraScale Architecture Clocking Resources Send Feedback

100
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Appendix A: Additional Resources and Legal Notices

Date Version Revision

12/12/2016 1.4 Chapter 1: Updated the discussion on page 7 about the differences between clock
capable and global clock pins.
Chapter 2: Added clarification to the Global Clock Inputs section. Added further
information following Figure 2-3. Updated the BUFGCE_DIV section. Revised the
BUFG_GT_SYNC description on page 31 to include the UltraScale+ devices.
Chapter 3: Added the UltraScale+ device MMCME4 and PLLE4 primitives to the
MMCM Primitives and PLL Primitives sections. Updated the description of PSCLK
cycles in the Dynamic Phase Shift Interface in the MMCM section. Added a
Recommended note on page page 49. Updated the CLKINSTOPPED – Input Clock
Status section. Added CLKFBOUT and CLKFBIN to Table 3-11 and their descriptions
below the table. Updated the CLKOUTPHYEN – PHY Clock Enable description. Added
Figure 3-17 and Figure 3-18. In Table 3-12, updated the DIVCLK_DIVIDE allowed
values and added PHY_ALIGN to the COMPENSATION attribute.
Updated the Please Read: Important Legal Notices section.
11/24/2015 1.3 Under Introduction to UltraScale Architecture, page 3, added new introductory text
for UltraScale+ devices. Added ninth bullet under Key Differences from 7 Series
FPGAs, page 7. Updated first paragraph under Global Clock Inputs, page 8 to include
information about HDGC pins. Updated first paragraph under Clock Structure,
page 10. Added Important note under Clock Buffers, page 14. Added second
paragraph under BUFCE_LEAF Clock Buffer, page 28. Added first two sentences
under BUFG_GT and BUFG_GT_SYNC, page 30. Added BUFG_PS, page 32 section.
Updated Frequency Synthesis Using Fractional Divide in the MMCM, page 38, by
changing 0.125 degrees to 0.125. Revised the heading Static Phase Shift Mode
(MMCM and PLL), page 39 by adding (MMCM and PLL). Revised the heading MMCM
Clock Divide Dynamic Change, page 42 by adding MMCM. Added Important note
under CLKFBIN – Feedback Clock Input, page 46. In Table 3-4, added a row of
UltraScale+ device MMCM attributes for CLKFBOUT_MULT_F(1), page 51, changed
default value for COMPENSATION, page 53 from ZHOLD to AUTO, revised the
COMPENSATION Description, added note 4, and note 5. In Table 3-12, added a row
of UltraScale+ device PLL attributes for CLKFBOUT_MULT, page 76, revised the
COMPENSATION Description, and added note 1. Added Dynamic Reconfiguration
Port, page 78 section. Updated References, page 99.
02/23/2015 1.2 In Table 3-4, changed the Allowed Values attribute for CLKIN1_PERIOD and
CLKIN2_PERIOD, page 52. In Table 3-12, changed the Allowed Values attribute for
CLKIN_PERIOD, page 76.
08/21/2014 1.1 Replaced clock-capable with global clock in Global Clock Inputs. Updated Byte Clock
Inputs. Added BUFG_GT_SYNC to BUFG_GT and BUFG_GT_SYNC. Updated Figure 3-3
and added tip for Table 3-3. Updated Figure 3-11.
12/10/2013 1.0 Initial Xilinx release.

UltraScale Architecture Clocking Resources Send Feedback

101
UG572 (v1.10.2) February 1, 2023 www.xilinx.com
 Appendix A: Additional Resources and Legal Notices

Please Read: Important Legal Notices

The information disclosed to you hereunder (the "Materials") is provided solely for the selection and use of Xilinx products. To the
maximum extent permitted by applicable law: (1) Materials are made available "AS IS" and with all faults, Xilinx hereby DISCLAIMS
ALL WARRANTIES AND CONDITIONS, EXPRESS, IMPLIED, OR STATUTORY, INCLUDING BUT NOT LIMITED TO WARRANTIES OF
MERCHANTABILITY, NON-INFRINGEMENT, OR FITNESS FOR ANY PARTICULAR PURPOSE; and (2) Xilinx shall not be liable (whether
in contract or tort, including negligence, or under any other theory of liability) for any loss or damage of any kind or nature related
to, arising under, or in connection with, the Materials (including your use of the Materials), including for any direct, indirect, special,
incidental, or consequential loss or damage (including loss of data, profits, goodwill, or any type of loss or damage suffered as a
result of any action brought by a third party) even if such damage or loss was reasonably foreseeable or Xilinx had been advised
of the possibility of the same. Xilinx assumes no obligation to correct any errors contained in the Materials or to notify you of
updates to the Materials or to product specifications. You may not reproduce, modify, distribute, or publicly display the Materials
without prior written consent. Certain products are subject to the terms and conditions of Xilinx's limited warranty, please refer to
Xilinx's Terms of Sale which can be viewed at www.xilinx.com/legal.htm#tos; IP cores may be subject to warranty and support terms
contained in a license issued to you by Xilinx. Xilinx products are not designed or intended to be fail-safe or for use in any
application requiring fail-safe performance; you assume sole risk and liability for use of Xilinx products in such critical applications,
please refer to Xilinx's Terms of Sale which can be viewed at www.xilinx.com/legal.htm#tos.
AUTOMOTIVE APPLICATIONS DISCLAIMER
AUTOMOTIVE PRODUCTS (IDENTIFIED AS “XA” IN THE PART NUMBER) ARE NOT WARRANTED FOR USE IN THE DEPLOYMENT OF
AIRBAGS OR FOR USE IN APPLICATIONS THAT AFFECT CONTROL OF A VEHICLE (“SAFETY APPLICATION”) UNLESS THERE IS A
SAFETY CONCEPT OR REDUNDANCY FEATURE CONSISTENT WITH THE ISO 26262 AUTOMOTIVE SAFETY STANDARD (“SAFETY
DESIGN”). CUSTOMER SHALL, PRIOR TO USING OR DISTRIBUTING ANY SYSTEMS THAT INCORPORATE PRODUCTS, THOROUGHLY
TEST SUCH SYSTEMS FOR SAFETY PURPOSES. USE OF PRODUCTS IN A SAFETY APPLICATION WITHOUT A SAFETY DESIGN IS FULLY
AT THE RISK OF CUSTOMER, SUBJECT ONLY TO APPLICABLE LAWS AND REGULATIONS GOVERNING LIMITATIONS ON PRODUCT
LIABILITY.
© Copyright 2013–2023 Advanced Micro Devices, Inc. Xilinx, the Xilinx logo, Artix, ISE, Kintex, Spartan, Virtex, Vivado, Zynq, and
other designated brands included herein are trademarks of Xilinx in the United States and other countries. AMBA, AMBA Designer,
ARM, ARM1176JZ-S, CoreSight, Cortex, PrimeCell, and MPCore are trademarks of ARM in the EU and other countries. PCI, PCIe,
and PCI Express are trademarks of PCI-SIG and used under license. All other trademarks are the property of their respective owners.

UltraScale Architecture Clocking Resources Send Feedback

102
UG572 (v1.10.2) February 1, 2023 www.xilinx.com