aboutsummaryrefslogtreecommitdiffstats
path: root/fpga/docs/usrp3/sim
diff options
context:
space:
mode:
authorMartin Braun <martin.braun@ettus.com>2020-01-23 16:10:22 -0800
committerMartin Braun <martin.braun@ettus.com>2020-01-28 09:35:36 -0800
commitbafa9d95453387814ef25e6b6256ba8db2df612f (patch)
tree39ba24b5b67072d354775272e687796bb511848d /fpga/docs/usrp3/sim
parent3075b981503002df3115d5f1d0b97d2619ba30f2 (diff)
downloaduhd-bafa9d95453387814ef25e6b6256ba8db2df612f.tar.gz
uhd-bafa9d95453387814ef25e6b6256ba8db2df612f.tar.bz2
uhd-bafa9d95453387814ef25e6b6256ba8db2df612f.zip
Merge FPGA repository back into UHD repository
The FPGA codebase was removed from the UHD repository in 2014 to reduce the size of the repository. However, over the last half-decade, the split between the repositories has proven more burdensome than it has been helpful. By merging the FPGA code back, it will be possible to create atomic commits that touch both FPGA and UHD codebases. Continuous integration testing is also simplified by merging the repositories, because it was previously difficult to automatically derive the correct UHD branch when testing a feature branch on the FPGA repository. This commit also updates the license files and paths therein. We are therefore merging the repositories again. Future development for FPGA code will happen in the same repository as the UHD host code and MPM code. == Original Codebase and Rebasing == The original FPGA repository will be hosted for the foreseeable future at its original local location: https://github.com/EttusResearch/fpga/ It can be used for bisecting, reference, and a more detailed history. The final commit from said repository to be merged here is 05003794e2da61cabf64dd278c45685a7abad7ec. This commit is tagged as v4.0.0.0-pre-uhd-merge. If you have changes in the FPGA repository that you want to rebase onto the UHD repository, simply run the following commands: - Create a directory to store patches (this should be an empty directory): mkdir ~/patches - Now make sure that your FPGA codebase is based on the same state as the code that was merged: cd src/fpga # Or wherever your FPGA code is stored git rebase v4.0.0.0-pre-uhd-merge Note: The rebase command may look slightly different depending on what exactly you're trying to rebase. - Create a patch set for your changes versus v4.0.0.0-pre-uhd-merge: git format-patch v4.0.0.0-pre-uhd-merge -o ~/patches Note: Make sure that only patches are stored in your output directory. It should otherwise be empty. Make sure that you picked the correct range of commits, and only commits you wanted to rebase were exported as patch files. - Go to the UHD repository and apply the patches: cd src/uhd # Or wherever your UHD repository is stored git am --directory fpga ~/patches/* rm -rf ~/patches # This is for cleanup == Contributors == The following people have contributed mainly to these files (this list is not complete): Co-authored-by: Alex Williams <alex.williams@ni.com> Co-authored-by: Andrej Rode <andrej.rode@ettus.com> Co-authored-by: Ashish Chaudhari <ashish@ettus.com> Co-authored-by: Ben Hilburn <ben.hilburn@ettus.com> Co-authored-by: Ciro Nishiguchi <ciro.nishiguchi@ni.com> Co-authored-by: Daniel Jepson <daniel.jepson@ni.com> Co-authored-by: Derek Kozel <derek.kozel@ettus.com> Co-authored-by: EJ Kreinar <ej@he360.com> Co-authored-by: Humberto Jimenez <humberto.jimenez@ni.com> Co-authored-by: Ian Buckley <ian.buckley@gmail.com> Co-authored-by: Jörg Hofrichter <joerg.hofrichter@ni.com> Co-authored-by: Jon Kiser <jon.kiser@ni.com> Co-authored-by: Josh Blum <josh@joshknows.com> Co-authored-by: Jonathon Pendlum <jonathan.pendlum@ettus.com> Co-authored-by: Martin Braun <martin.braun@ettus.com> Co-authored-by: Matt Ettus <matt@ettus.com> Co-authored-by: Michael West <michael.west@ettus.com> Co-authored-by: Moritz Fischer <moritz.fischer@ettus.com> Co-authored-by: Nick Foster <nick@ettus.com> Co-authored-by: Nicolas Cuervo <nicolas.cuervo@ettus.com> Co-authored-by: Paul Butler <paul.butler@ni.com> Co-authored-by: Paul David <paul.david@ettus.com> Co-authored-by: Ryan Marlow <ryan.marlow@ettus.com> Co-authored-by: Sugandha Gupta <sugandha.gupta@ettus.com> Co-authored-by: Sylvain Munaut <tnt@246tNt.com> Co-authored-by: Trung Tran <trung.tran@ettus.com> Co-authored-by: Vidush Vishwanath <vidush.vishwanath@ettus.com> Co-authored-by: Wade Fife <wade.fife@ettus.com>
Diffstat (limited to 'fpga/docs/usrp3/sim')
-rw-r--r--fpga/docs/usrp3/sim/libs_axi.md266
-rw-r--r--fpga/docs/usrp3/sim/libs_general.md188
-rw-r--r--fpga/docs/usrp3/sim/running_testbenches.md133
-rw-r--r--fpga/docs/usrp3/sim/writing_testbenches.md471
4 files changed, 1058 insertions, 0 deletions
diff --git a/fpga/docs/usrp3/sim/libs_axi.md b/fpga/docs/usrp3/sim/libs_axi.md
new file mode 100644
index 000000000..a260aabdd
--- /dev/null
+++ b/fpga/docs/usrp3/sim/libs_axi.md
@@ -0,0 +1,266 @@
+# AXI Interface Libraries
+
+## AXI4 Stream (sim\_axis\_lib.vh)
+
+Defines ``axis_t``, an AXI Stream bus interface that implements several tasks to send and
+receive data on the bus.
+
+### Definition
+
+ interface axis_t #(parameter DWIDTH = 64)
+ (input clk);
+ logic [DWIDTH-1:0] tdata;
+ logic tvalid;
+ logic tlast;
+ logic tready;
+
+ modport master (output tdata, output tvalid, output tlast, input tready);
+ modport slave (input tdata, input tvalid, input tlast, output tready);
+
+
+### Operations
+
+#### push\_word
+
+ // Push a word onto the AXI-Stream bus and wait for it to transfer
+ // Args:
+ // - word: The data to push onto the bus
+ // - eop (optional): End of packet (asserts tlast)
+
+#### push\_bubble
+
+ // Push a bubble cycle onto the AXI-Stream bus
+
+#### pull\_word
+
+ // Wait for a sample to be transferred on the AXI Stream
+ // bus and return the data and last
+ // Args:
+ // - word: The data pulled from the bus
+ // - eop: End of packet (tlast)
+
+#### wait\_for\_bubble
+
+ // Wait for a bubble cycle on the AXI Stream bus
+
+#### wait\_for\_pkt
+
+ // Wait for a packet to finish on the bus
+
+#### push\_rand\_pkt
+
+ // Push a packet with random data onto to the AXI Stream bus
+ // Args:
+ // - num_samps: Packet size.
+
+#### push\_ramp\_pkt
+
+ // Push a packet with a ramp on to the AXI Stream bus
+ // Args:
+ // - num_samps: Packet size.
+ // - ramp_start: Start value for the ramp
+ // - ramp_inc: Increment per clock cycle
+
+## Compressed VITA [CHDR] (sim\_chdr\_lib.vh)
+
+Defines ``cvita_stream_t``, an AXI Stream bus interface that implements the CHDR protocol and
+several tasks to send and receive data on the bus.
+
+### CHDR
+
+ typedef enum logic [1:0] {
+ DATA=2'b00, FC=2'b01, CMD=2'b10, RESP=2'b11
+ } cvita_pkt_t;
+
+ typedef struct packed {
+ logic [31:0] sid;
+ logic [15:0] length;
+ logic [11:0] seqno;
+ logic eob;
+ logic has_time;
+ cvita_pkt_t pkt_type;
+ logic [63:0] timestamp;
+ } cvita_hdr_t;
+
+#### Operations
+
+ - ``flatten_chdr_no_ts``: Flatten header struct to a 64-bit bus. No timestamp.
+ - ``unflatten_chdr_no_ts``: Decode a 64-bit header and populate the ``cvita_hdr_t`` struct. No timestamp.
+ - ``unflatten_chdr``: Decode a 64-bit header and populate the ``cvita_hdr_t`` struct. Timestamp supported.
+
+### CVITA Stream Type
+
+#### Definition
+
+ interface cvita_stream_t (input clk);
+ axis_t #(.DWIDTH(64)) axis (.clk(clk));
+
+#### push\_hdr
+
+ // Push a CVITA header into the stream
+ // Args:
+ // - hdr: The header to push
+
+#### push\_data
+
+ // Push a word onto the AXI-Stream bus and wait for it to transfer
+ // Args:
+ // - word: The data to push onto the bus
+ // - eop: End of packet (asserts tlast)
+
+#### push\_bubble
+
+ // Push a bubble cycle on the AXI-Stream bus
+
+#### pull\_word
+
+ // Wait for a sample to be transferred on the AXI Stream
+ // bus and return the data and last
+ // Args:
+ // - word: The data pulled from the bus
+ // - eop: End of packet (tlast)
+
+#### wait\_for\_bubble
+
+ // Wait for a bubble cycle on the AXI Stream bus
+
+#### wait\_for\_pkt
+
+ // Wait for a packet to finish on the bus
+
+#### wait\_for\_pkt\_get\_info
+
+ // Wait for a packet to finish on the bus and extract the header and payload statistics.
+
+ typedef struct packed {
+ logic [31:0] count;
+ logic [63:0] sum;
+ logic [63:0] min;
+ logic [63:0] max;
+ logic [63:0] crc;
+ } cvita_stats_t;
+
+#### push\_rand\_pkt
+
+ // Push a packet with random data onto to the AXI Stream bus
+ // Args:
+ // - num_samps: Packet size.
+ // - hdr: Header to attach to packet (length will be ignored)
+ // - timestamp: Optional timestamp
+
+#### push\_ramp\_pkt
+
+ // Push a packet with a ramp on to the AXI Stream bus
+ // Args:
+ // - num_samps: Packet size.
+ // - ramp_start: Start value for the ramp
+ // - ramp_inc: Increment per clock cycle
+ // - hdr: Header to attach to packet (length will be ignored)
+ // - timestamp: Optional timestamp
+
+## Memory Mapped AXI4 (sim\_axi4\_lib.vh)
+
+Defines the following interfaces to group signals in the AXI4 bus.
+WIP: No functions or tasks implemented yet.
+
+#### Address
+
+ interface axi4_addr_t #(parameter AWIDTH=32, parameter IDWIDTH=4)
+ (input clk);
+
+ logic [IDWIDTH-1:0] id;
+ logic [AWIDTH-1:0] addr;
+ logic [7:0] len;
+ logic [2:0] size;
+ logic [1:0] burst;
+ logic lock;
+ logic [3:0] cache;
+ logic [2:0] prot;
+ logic [3:0] qos;
+ logic [3:0] region;
+ logic user;
+ logic valid;
+ logic ready;
+
+ modport master (output id,addr,len,size,burst,lock,cache,prot,qos,valid, input ready);
+ modport slave (input id,addr,len,size,burst,lock,cache,prot,qos,valid, output ready);
+
+ endinterface
+
+#### Write Data
+
+ interface axi4_wdata_t #(parameter DWIDTH=64)
+ (input clk);
+
+ logic [DWIDTH-1:0] data;
+ logic [(DWIDTH/8)-1:0] strb;
+ logic last;
+ logic user;
+ logic valid;
+ logic ready;
+
+ modport master(output data,strb,last,valid, input ready);
+ modport slave(input data,strb,last,valid, output ready);
+
+ endinterface
+
+#### Write Response
+
+ interface axi4_resp_t #(parameter IDWIDTH=4)
+ (input clk);
+
+ logic ready;
+ logic [IDWIDTH-1:0] id;
+ logic [1:0] resp;
+ logic user;
+ logic valid;
+
+ modport master(output ready, input id,resp,valid);
+ modport slave(input ready, output id,resp,valid);
+
+ endinterface
+
+#### Read Data
+
+ interface axi4_rdata_t #(parameter DWIDTH=64, parameter IDWIDTH=4)
+ (input clk);
+
+ logic ready;
+ logic [IDWIDTH-1:0] id;
+ logic [DWIDTH-1:0] data;
+ logic [1:0] resp;
+ logic user;
+ logic last;
+ logic valid;
+
+ modport master(output ready, input id,data,resp,last,valid);
+ modport slave(input ready, output id,data,resp,last,valid);
+
+ endinterface
+
+#### Meta: AXI4 Writer
+
+ interface axi4_wr_t #(parameter DWIDTH=64, parameter AWIDTH=32, parameter IDWIDTH=4)
+ (input clk);
+
+ axi4_addr_t #(.AWIDTH(AWIDTH), .IDWIDTH(IDWIDTH)) addr (.clk(clk));
+ axi4_wdata_t #(.DWIDTH(DWIDTH)) data (.clk(clk));
+ axi4_resp_t #(.IDWIDTH(IDWIDTH)) resp (.clk(clk));
+
+ modport master(output addr, output data, input resp);
+ modport slave(input addr, input data, output resp);
+
+ endinterface
+
+#### Meta: AXI4 Reader
+
+ interface axi4_rd_t #(parameter DWIDTH=64, parameter AWIDTH=32, parameter IDWIDTH=4)
+ (input clk);
+
+ axi4_addr_t #(.AWIDTH(AWIDTH), .IDWIDTH(IDWIDTH)) addr (.clk(clk));
+ axi4_rdata_t #(.DWIDTH(DWIDTH), .IDWIDTH(IDWIDTH)) data (.clk(clk));
+
+ modport master(output addr, output data);
+ modport slave(input addr, input data);
+
+ endinterface
diff --git a/fpga/docs/usrp3/sim/libs_general.md b/fpga/docs/usrp3/sim/libs_general.md
new file mode 100644
index 000000000..4bc154dc2
--- /dev/null
+++ b/fpga/docs/usrp3/sim/libs_general.md
@@ -0,0 +1,188 @@
+# General Purpose Libraries
+
+## Execution and Reporting (sim\_exec\_report.vh)
+
+Macros to do boilerplate testbench initialization and utilities to define test cases
+
+#### TEST\_BENCH\_INIT
+
+ // Initializes state for a test bench.
+ // This macro *must be* called within the testbench module but
+ // outside the primary initial block
+ // Its sets up boilerplate code for:
+ // - Logging to console
+ // - Test execution tracking
+ // - Gathering test results
+ // - Bounding execution time based on the SIM_RUNTIME_US vdef
+ //
+ // Usage: `TEST_BENCH_INIT(test_name,min_tc_run_count,ns_per_tick)
+ // where
+ // - tb_name: Name of the testbench. (Only used during reporting)
+ // - min_tc_run_count: Number of test cases in testbench. (Used to detect stalls and inf-loops)
+ // - ns_per_tick: The time_unit_base from the timescale declaration
+ //
+
+#### TEST\_CASE\_START
+
+ // Indicates the start of a test case
+ // This macro *must be* called inside the primary initial block
+ //
+ // Usage: `TEST_CASE_START(test_name)
+ // where
+ // - test_name: The name of the test.
+ //
+
+#### TEST\_CASE\_DONE
+ // Indicates the end of a test case
+ // This macro *must be* called inside the primary initial block
+ // The pass/fail status of test case is determined based on the
+ // the user specified outcome and the number of fatal or error
+ // ASSERTs triggered in the test case.
+ //
+ // Usage: `TEST_CASE_DONE(test_result)
+ // where
+ // - test_result: User specified outcome
+ //
+
+#### ASSERT\_FATAL
+
+ // Wrapper around a an assert.
+ // ASSERT_FATAL throws an error assertion and halts the simulator
+ // if cond is not satisfied
+ //
+ // Usage: `ASSERT_FATAL(cond,msg)
+ // where
+ // - cond: Condition for the assert
+ // - msg: Message for the assert
+ //
+
+#### ASSERT\_ERROR
+
+ // Wrapper around a an assert.
+ // ASSERT_ERROR throws an error assertion and fails the test case
+ // if cond is not satisfied. The simulator will *not* halt
+ //
+ // Usage: `ASSERT_ERROR(cond,msg)
+ // where
+ // - cond: Condition for the assert
+ // - msg: Message for the assert
+ //
+
+#### ASSERT\_WARNING
+
+ // Wrapper around a an assert.
+ // ASSERT_WARNING throws an warning assertion but does not fail the
+ // test case if cond is not satisfied. The simulator will *not* halt
+ //
+ // Usage: `ASSERT_WARNING(cond,msg)
+ // where
+ // - cond: Condition for the assert
+ // - msg: Message for the assert
+ //
+
+## Clocks and Resets (sim\_clks\_rsts.vh)
+
+Shortcut macros to create typical clock and reset signals.
+
+#### DEFINE\_CLK
+
+ // Generates a persistent clock that starts at t=0 and runs forever
+ //
+ // Usage: `DEFINE_CLK(clk_name,period,duty_cycle)
+ // where
+ // - clk_name: The clock net to be generated
+ // - period: Period of the clock in simulator ticks
+ // - duty_cycle: Percentage duty cycle
+ //
+
+#### DEFINE\_LATE\_START\_CLK
+
+ // Generates a clock that starts at the specified time and runs forever
+ //
+ // Usage: `DEFINE_LATE_START_CLK(clk_name,period,duty_cycle,start_time,start_time_res)
+ // where
+ // - clk_name: The clock net to be generated
+ // - period: Period of the clock in simulator ticks
+ // - duty_cycle: Percentage duty cycle
+ // - start_time: Start time for clock in simulator ticks
+ // - start_time_res: Start time resolution (must be > timescale increment and < start_time)
+ //
+
+#### DEFINE_RESET
+
+ // Generates an active high reset
+ //
+ // Usage: `DEFINE_RESET(reset_name,reset_time,reset_duration)
+ // where
+ // - reset_name: The reset net to be generated
+ // - reset_time: Time at which reset will be asserted (i.e. rst=1)
+ // - reset_duration: Duration of reset assertion
+ //
+
+#### DEFINE_RESET_N
+
+ // Generates an active low reset
+ //
+ // Usage: `DEFINE_RESET_N(reset_name,reset_time,reset_duration)
+ // where
+ // - reset_name: The reset net to be generated
+ // - reset_time: Time at which reset will be asserted (i.e. rst=0)
+ // - reset_duration: Duration of reset assertion
+ //
+
+## File I/O (sim\_file\_io.sv)
+
+### interface data\_file\_t
+
+Defines a ``data_file_t`` interface with the following functions:
+
+#### ctor
+
+ // Create a handle to a data_file with
+ // - FILENAME: Name of the file
+ // - FORMAT: Data format (HEX, DEC, OCT, BIN, FLOAT)
+ // - DWIDTH: Width of each element stored in the file (one line per word)
+ //
+
+#### open
+
+ // Open the data file for reading or writing.
+ //
+ // Usage: open(mode)
+ // where
+ // - mode: RW mode (Choose from: READ, WRITE, APPEND)
+ //
+
+#### close
+
+ // Close an open data file. No-op if file isn't already open
+ //
+ // Usage: close()
+ //
+
+#### is_eof
+
+ // Is end-of-file reached.
+ //
+ // Usage: is_eof() Returns eof
+ // where
+ // - eof: A boolean
+ //
+
+#### readline
+
+ // Read a line from the datafile
+ //
+ // Usage: readline() Returns data
+ // where
+ // - data: A logic array of width DWIDTH containing the read word
+ //
+
+#### writeline
+
+ // Write a line to the datafile
+ //
+ // Usage: writeline(data)
+ // where
+ // - data: A logic array of width DWIDTH to write to the file
+ // \ No newline at end of file
diff --git a/fpga/docs/usrp3/sim/running_testbenches.md b/fpga/docs/usrp3/sim/running_testbenches.md
new file mode 100644
index 000000000..2e2068e5e
--- /dev/null
+++ b/fpga/docs/usrp3/sim/running_testbenches.md
@@ -0,0 +1,133 @@
+# Running a Testbench
+
+Each executable testbench has its own Makefile that automatically pulls in support
+for all supported simulators. The build infrastructure supports the following simulators:
+
+ - Xilinx Vivado (XSim)
+ - Mentor Graphics ModelSim (may require an additional license)
+
+
+In general running ``make <sim_target>`` will run the
+simulation and report results in the console. Running ``make help`` will print out
+all supported simulator targets. Currently, the following targets will work:
+
+ Supported Targets:
+ ipclean: Cleanup all IP intermediate files
+ clean: Cleanup all simulator intermediate files
+ cleanall: Cleanup everything!
+ xsim: Run the simulation using the Xilinx Vivado Simulator
+ xclean: Cleanup Xilinx Vivado Simulator intermediate files
+ vsim: Run the simulation using Modelsim
+ vclean: Cleanup Modelsim intermediate files
+
+
+## Using Xilinx Vivado XSim
+
+XSim is the built-in simulator in the Xilinx Vivado toolchain. If you already met the
+prerequisites for building an FPGA image, then you don't need to install anything else.
+
+Follow these steps to run a testbench:
+
+ - Navigate to the directory that contains the top level testbench and Makefile
+ - Run the setenv.sh script for the USRP product that you are trying to simulate
+
+ ``$ source <repo>/usrp3/top/<product>/setupenv.sh``
+
+ This step is required even if the simulation is generic because the toolchain requires
+ an FPGA part number to load simulation models.
+ - Run the simulator specific target
+
+ ``$ make xsim``
+
+
+## Using Mentor Graphics ModelSim
+
+ModelSim is a third-party simulation tool that is compatible with Vivado and the USRP
+FPGA build infrastructure.
+
+Use the following one-time setup to install and configure Modelsim on your system
+
+ - Install Modelsim from the [Mentor Graphics](http://www.mentor.com/) website.
+ It is recommended that you install it to the default location (/opt/mentor/modelsim)
+ - Run the setenv.sh script for the USRP product that you are trying to simulate
+
+ ``$ source <repo>/usrp3/top/<product>/setupenv.sh``
+
+ This step is required even if the simulation is generic because the toolchain requires
+ an FPGA part number to load simulation models.
+ - Build the Xilinx simulation libraries
+ ``$ build_simlibs``
+
+
+To validate that everything was install properly run ``setupenv.sh`` again. You should see the following
+
+ Setting up X3x0 FPGA build environment (64-bit)...
+ - Vivado: Found (/opt/Xilinx/Vivado/2014.4/bin)
+ - Modelsim: Found (/opt/mentor/modelsim/modeltech/bin)
+ - Modelsim Compiled Libs: Found (/opt/Xilinx/Vivado/2014.4/modelsim)
+
+ Environment successfully initialized.
+
+Follow these steps to run a testbench:
+
+ - Navigate to the directory that contains the top level testbench and Makefile
+ - Run the setenv.sh script for the USRP product that you are trying to simulate
+
+ ``$ source <repo>/usrp3/top/<product>/setupenv.sh``
+
+ This step is required even if the simulation is generic because the toolchain requires
+ an FPGA part number to load simulation models.
+ - Run the simulator specific target
+
+ ``$ make vsim``
+
+
+## Troubleshooting
+
+#### Vivado Not Found
+
+If running the setupenv.sh script return an error like the following:
+
+ Vivado: Not found! (ERROR.. Builds and simulations will not work)
+
+then it is possible that Vivado was not installed or it was not installed in the default
+location. If Vivado is installed in a non-default location, just run the following:
+
+ ``$ source <repo>/usrp3/top/<product>/setupenv.sh --vivado-path=<PATH>``
+
+#### Modelsim Not Found
+
+If running the setupenv.sh script return an error like the following:
+
+ Setting up X3x0 FPGA build environment (64-bit)...
+ - Vivado: Found (/opt/Xilinx/Vivado/2014.4/bin)
+ - Modelsim: Not found! (WARNING.. Simulations with vsim will not work)
+
+ Environment successfully initialized.
+
+or something like this (even when Modelsim is installed)
+
+ Setting up X3x0 FPGA build environment (64-bit)...
+ - Vivado: Found (/opt/Xilinx/Vivado/2014.4/bin)
+
+ Environment successfully initialized.
+
+then it is possible that Modelsim was not installed or it was not installed in the default
+location. If Modelsim is installed in a non-default location, just run the following:
+
+ ``$ source <repo>/usrp3/top/<product>/setupenv.sh --modelsim-path=<PATH>``
+
+#### Modelsim Simulation Libraries Not Found
+
+If running the setupenv.sh script return an error like the following:
+
+ Setting up X3x0 FPGA build environment (64-bit)...
+ - Vivado: Found (/opt/Xilinx/Vivado/2014.4/bin)
+ - Modelsim: Found (/opt/mentor/modelsim/modeltech/bin)
+ - Modelsim Compiled Libs: Not found! (Run build_simlibs to generate them.)
+
+ Environment successfully initialized.
+
+just run the following
+
+ $ build_simlibs \ No newline at end of file
diff --git a/fpga/docs/usrp3/sim/writing_testbenches.md b/fpga/docs/usrp3/sim/writing_testbenches.md
new file mode 100644
index 000000000..cfbbcdbce
--- /dev/null
+++ b/fpga/docs/usrp3/sim/writing_testbenches.md
@@ -0,0 +1,471 @@
+# Writing a Testbench
+
+Writing a unit test or system level test is easy with the Vivado makefile infrastructure!
+Most of the overhead of building and running a testbench is handled by the build tools.
+Even recurring tasks like reporting and monitoring are implemented by framework libraries.
+
+Each executable FPGA unit test must have the following components:
+
+1. A Makefile
+2. A Testbench top-level module
+
+## Testbench Makefile
+
+The Testbench Makefile tell the build tools what to build, where to build it, dependency information and runtime information.
+The build infrastructure will handle the how-to part for each supported simulation tool.
+
+Here is a sample Makefile (you are encouraged to use this as a starting point)
+
+ #
+ # Copyright 2015 Ettus Research LLC
+ #
+
+ #-------------------------------------------------
+ # Top-of-Makefile
+ #-------------------------------------------------
+ # Define BASE_DIR to point to the "top" dir
+ BASE_DIR = $(abspath ../../..)
+ # Include viv_sim_preample after defining BASE_DIR
+ include $(BASE_DIR)/../tools/make/viv_sim_preamble.mak
+
+ #-------------------------------------------------
+ # Design Specific
+ #-------------------------------------------------
+ # Define part using PART_ID (<device>/<package>/<speedgrade>)
+ ARCH = kintex7
+ PART_ID = xc7k410t/ffg900/-2
+
+ # Include makefiles and sources for the DUT and its dependencies
+ include $(BASE_DIR)/../lib/fifo/Makefile.srcs
+ include $(BASE_DIR)/../lib/axi/Makefile.srcs
+ include $(BASE_DIR)/../lib/control/Makefile.srcs
+
+ DESIGN_SRCS = $(abspath \
+ $(FIFO_SRCS) \
+ $(AXI_SRCS) \
+ $(CONTROL_LIB_SRCS) \
+ )
+
+ #-------------------------------------------------
+ # IP Specific
+ #-------------------------------------------------
+ # If simulation contains IP, define the IP_DIR and point
+ # it to the base level IP directory
+ IP_DIR = ../../ip
+
+ # Include makefiles and sources for all IP components
+ # *after* defining the IP_DIR
+ include $(IP_DIR)/ddr3_32bit/Makefile.inc
+ include $(IP_DIR)/axi_intercon_2x64_128/Makefile.inc
+ include $(IP_DIR)/fifo_short_2clk/Makefile.inc
+ include $(IP_DIR)/fifo_4k_2clk/Makefile.inc
+ include $(IP_DIR)/axi4_bram_1kx64/Makefile.inc
+
+ DESIGN_SRCS += $(abspath \
+ $(IP_DDR3_32BIT_SRCS) \
+ $(IP_AXI_INTERCON_2X64_128_SRCS) \
+ $(IP_FIFO_4K_2CLK_SRCS) \
+ $(IP_FIFO_SHORT_2CLK_SRCS) \
+ $(IP_AXI4_BRAM_1KX64_SRCS) \
+ )
+
+ #-------------------------------------------------
+ # Testbench Specific
+ #-------------------------------------------------
+ include $(BASE_DIR)/../sim/general/Makefile.srcs
+ include $(BASE_DIR)/../sim/axi/Makefile.srcs
+
+ # Define only one toplevel module
+ SIM_TOP = dram_fifo_tb
+ # Simulation runtime in microseconds
+ SIM_RUNTIME_US = 80
+
+ SIM_SRCS = \
+ $(abspath dram_fifo_tb.sv) \
+ $(abspath axis_dram_fifo_single.sv) \
+ $(IP_DDR3_32BIT_SIM_OUTS) \
+ $(SIM_GENERAL_SRCS) \
+ $(SIM_AXI_SRCS)
+
+ #-------------------------------------------------
+ # Bottom-of-Makefile
+ #-------------------------------------------------
+ # Include all simulator specific makefiles here
+ # Each should define a unique target to simulate
+ # e.g. xsim, vsim, etc and a common "clean" target
+ include $(BASE_DIR)/../tools/make/viv_simulator.mak
+
+You will notice that the Makefile has 5 distinct sections.
+
+### Section 1: Boilerplate
+
+ #-------------------------------------------------
+ # Top-of-Makefile
+ #-------------------------------------------------
+ # Define BASE_DIR to point to the "top" dir
+ BASE_DIR = $(abspath ../../..)
+ # Include viv_sim_preample after defining BASE_DIR
+ include $(BASE_DIR)/../tools/make/viv_sim_preamble.mak
+
+Before declaring any variables or using any recipes, the following must be done (in order):
+
+- Define `BASE_DIR` to tell the build system where the `<repo>/usrp3/top` directory is relative to the
+ current testbench directory.
+- Include `viv_sim_preamble.mak` to initialize boilerplate variables and functions
+
+### Section 2: Design Specific
+
+ #-------------------------------------------------
+ # Design Specific
+ #-------------------------------------------------
+ # Define part using PART_ID (<device>/<package>/<speedgrade>)
+ ARCH = kintex7
+ PART_ID = xc7k410t/ffg900/-2
+
+ # Include makefiles and sources for the DUT and its dependencies
+ include $(BASE_DIR)/../lib/fifo/Makefile.srcs
+ include $(BASE_DIR)/../lib/axi/Makefile.srcs
+ include $(BASE_DIR)/../lib/control/Makefile.srcs
+
+ DESIGN_SRCS = $(abspath \
+ $(FIFO_SRCS) \
+ $(AXI_SRCS) \
+ $(CONTROL_LIB_SRCS) \
+ )
+
+This section contains pointers to sources and other variables for the DUT to function. In the
+example above, we are including all sources from the lib/fifo, lib/axi, lib/control directories.
+
+The following makefile variables are special and must be defined:
+
+- `ARCH`: The architecture targeted for the simulation.
+- `PART_ID`: The exact part targeted for the simulation. Format: `<device>/<package>/<speedgrade>`
+- `DESIGN_SRCS`: Space-separated paths to the DUT and all of its dependencies.
+
+### Section 3: IP Specific
+
+ #-------------------------------------------------
+ # IP Specific
+ #-------------------------------------------------
+ # If simulation contains IP, define the IP_DIR and point
+ # it to the base level IP directory
+ IP_DIR = ../../ip
+
+ # Include makefiles and sources for all IP components
+ # *after* defining the IP_DIR
+ include $(IP_DIR)/ddr3_32bit/Makefile.inc
+ include $(IP_DIR)/axi_intercon_2x64_128/Makefile.inc
+ include $(IP_DIR)/fifo_short_2clk/Makefile.inc
+ include $(IP_DIR)/fifo_4k_2clk/Makefile.inc
+ include $(IP_DIR)/axi4_bram_1kx64/Makefile.inc
+
+ DESIGN_SRCS += $(abspath \
+ $(IP_DDR3_32BIT_SRCS) \
+ $(IP_AXI_INTERCON_2X64_128_SRCS) \
+ $(IP_FIFO_4K_2CLK_SRCS) \
+ $(IP_FIFO_SHORT_2CLK_SRCS) \
+ $(IP_AXI4_BRAM_1KX64_SRCS) \
+ )
+
+If the DUT depends on any Xilinx IP then this section is required. It tell the tools
+which IP cores need to be built in order to run the simulation. The IP specific Makefile
+includes handle the "how" part of building IP. As long as the correct Mafefile is included
+and the IP XCI sources are added to `DESIGN_SRCS`, the IP intermediates will be built correctly.
+
+The `IP_DIR` variable must be defined to point to the base ip directory that contains XCI sources.
+
+### Section 4: Testbench Specific
+
+ #-------------------------------------------------
+ # Testbench Specific
+ #-------------------------------------------------
+ include $(BASE_DIR)/../sim/general/Makefile.srcs
+ include $(BASE_DIR)/../sim/axi/Makefile.srcs
+
+ # Define only one toplevel module
+ SIM_TOP = dram_fifo_tb
+ # Simulation runtime in microseconds
+ SIM_RUNTIME_US = 80
+
+ SIM_SRCS = \
+ $(abspath dram_fifo_tb.sv) \
+ $(abspath axis_dram_fifo_single.sv) \
+ $(IP_DDR3_32BIT_SIM_OUTS) \
+ $(SIM_GENERAL_SRCS) \
+ $(SIM_AXI_SRCS)
+
+This section contains all sources and parameters for the actual testbench. Any simulation
+dependency makefiles can be included here.
+
+The following variables must be defined:
+
+- `SIM_TOP`: The toplevel module name for the simulation project
+- `SIM_RUNTIME_US`: The maximum runtime of the simulation in microseconds. At this time $finish will be called to terminate the testbench.
+- `SIM_SRCS`: This is similar to DESIGN_SRCS except that that should contain a path to `SIM_TOP` and all of its dependencies.
+
+### Section 5: Tool Support
+
+ #-------------------------------------------------
+ # Bottom-of-Makefile
+ #-------------------------------------------------
+ # Include all simulator specific makefiles here
+ # Each should define a unique target to simulate
+ # e.g. xsim, vsim, etc and a common "clean" target
+ include $(BASE_DIR)/../tools/make/viv_simulator.mak
+
+Now that the Makefile knows all the basic information about the testbench, include tool-specific
+makefiles to implement simulation targets. Currently the following simulator makefiles exits:
+
+- ``<repo>/tools/make/viv_simulator.mak``
+
+Please refer to the next section for more information about targets
+
+
+## Testbench Top Level
+
+The top-level module will instantiate the DUT and implement self-checking behavior.
+Test benches could be written in any language (SystemVerilog, Verilog, VHDL) but
+to take advantage of our repository of simulation libraries, it is recommended that SystemVerilog be used.
+
+Here is a sample SystemVerilog top module (you are encouraged to use this as a starting point)
+
+ //
+ // Copyright 2015 Ettus Research LLC
+ //
+
+ `timescale 1ns/1ps
+ `define NS_PER_TICK 1
+ `define NUM_TEST_CASES 3
+
+ `include "sim_clks_rsts.vh"
+ `include "sim_exec_report.vh"
+ `include "sim_cvita_lib.sv"
+
+ module example_fifo_tb();
+ `TEST_BENCH_INIT("example_fifo_tb",`NUM_TEST_CASES,`NS_PER_TICK)
+
+ // Define all clocks and resets
+ `DEFINE_CLK(bus_clk, 1000/166.6667, 50) //166MHz bus_clk
+ `DEFINE_RESET(bus_rst, 0, 100) //100ns for GSR to deassert
+
+ cvita_stream_t chdr_i (.clk(bus_clk));
+ cvita_stream_t chdr_o (.clk(bus_clk));
+
+ // Initialize DUT
+ axi_fifo #(.WIDTH(65), .SIZE(24)) dut_single (
+ .clk(bus_clk),
+ .reset(bus_rst),
+ .clear(1'b0),
+
+ .i_tdata({chdr_i.axis.tlast, chdr_i.axis.tdata}),
+ .i_tvalid(chdr_i.axis.tvalid),
+ .i_tready(chdr_i.axis.tready),
+
+ .o_tdata({chdr_o.axis.tlast, chdr_o.axis.tdata}),
+ .o_tvalid(chdr_o.axis.tvalid),
+ .o_tready(chdr_o.axis.tready),
+
+ .space(),
+ .occupied()
+ );
+
+ //Testbench variables
+ cvita_hdr_t header, header_out;
+ cvita_stats_t stats;
+
+ //------------------------------------------
+ //Main thread for testbench execution
+ //------------------------------------------
+ initial begin : tb_main
+
+ `TEST_CASE_START("Wait for reset");
+ while (bus_rst) @(posedge bus_clk);
+ `TEST_CASE_DONE((~bus_rst));
+
+ repeat (200) @(posedge bus_clk);
+
+ header = '{
+ pkt_type:DATA, has_time:0, eob:0, seqno:12'h666,
+ length:0, sid:$random, timestamp:64'h0};
+
+ `TEST_CASE_START("Fill up empty FIFO then drain (short packet)");
+ chdr_o.axis.tready = 0;
+ chdr_i.push_ramp_pkt(16, 64'd0, 64'h100, header);
+ chdr_o.axis.tready = 1;
+ chdr_o.wait_for_pkt_get_info(header_out, stats);
+ `ASSERT_ERROR(stats.count==16, "Bad packet: Length mismatch");
+ `ASSERT_ERROR(header.sid==header_out.sid, "Bad packet: Wrong SID");
+ `ASSERT_ERROR(chdr_i.axis.tready, "Bus not ready");
+ `TEST_CASE_DONE(1);
+
+ header = '{
+ pkt_type:DATA, has_time:1, eob:0, seqno:12'h666,
+ length:0, sid:$random, timestamp:64'h0};
+
+ `TEST_CASE_START("Concurrent read and write (single packet)");
+ chdr_o.axis.tready = 1;
+ fork
+ begin
+ chdr_i.push_ramp_pkt(20, 64'd0, 64'h100, header);
+ end
+ begin
+ chdr_o.wait_for_pkt_get_info(header_out, stats);
+ end
+ join
+ `ASSERT_ERROR(stats.count==20, "Bad packet: Length mismatch");
+ `TEST_CASE_DONE(1);
+ end
+ endmodule
+
+
+Each testbench should have the following basic components:
+
+### Timescale Defines and Includes
+
+ `timescale 1ns/1ps
+ `define NS_PER_TICK 1
+ `define NUM_TEST_CASES 3
+
+ `include "sim_clks_rsts.vh"
+ `include "sim_exec_report.vh"
+ `include "sim_cvita_lib.sv"
+
+In addition to the timescale, the infrastructure needs to know the number of
+nanoseconds per simulator tick. This can be a floating point number.
+
+
+In addition to the timescale, you may include any Verilog/SystemVerilog headers here.
+
+### Main Module Definition
+
+ `include "sim_exec_report.vh"
+
+ module example_fifo_tb();
+ `TEST_BENCH_INIT("example_fifo_tb",`NUM_TEST_CASES,`NS_PER_TICK)
+
+ ...
+
+ //------------------------------------------
+ //Main thread for testbench execution
+ //------------------------------------------
+ initial begin : tb_main
+
+ ...
+
+ end
+ endmodule
+
+The name of the main module must match the ``SIM_TOP`` variable value in the Makefile.
+To register this module with the framework, the ``TEST_BENCH_INIT`` macro must be called.
+This macro is defined in ``<repo>/usrp3/sim/general/sim_exec_report.vh``.
+
+``TEST_BENCH_INIT``:
+
+ // Initializes state for a test bench.
+ // This macro *must be* called within the testbench module but
+ // outside the primary initial block
+ // Its sets up boilerplate code for:
+ // - Logging to console
+ // - Test execution tracking
+ // - Gathering test results
+ // - Bounding execution time based on the SIM_RUNTIME_US vdef
+ //
+ // Usage: `TEST_BENCH_INIT(test_name,min_tc_run_count,ns_per_tick)
+ // where
+ // - tb_name: Name of the testbench. (Only used during reporting)
+ // - min_tc_run_count: Number of test cases in testbench. (Used to detect stalls and inf-loops)
+ // - ns_per_tick: The time_unit_base from the timescale declaration
+
+The testbench must also have at least one initial block that consists tests cases (covered later).
+For the sake of convention it should be called ``tb_main``. *All test cases must live in ``tb_main``*. You may
+have other initial block but they must not call macros from ``sim_exec_report.vh`` because the code
+there is not thread-safe.
+
+### Test Cases
+
+A test case in this context is defined as an independent entity that validates an aspect of the DUT behavior
+and which is independent from other test cases i.e. the result of one test case should ideally not affect others.
+
+
+Test cases are wrapped in the ``TEST_CASE_START`` and ``TEST_CASE_DONE`` macros:
+
+ `TEST_CASE_START("Fill up empty FIFO then drain (short packet)");
+ chdr_o.axis.tready = 0;
+ chdr_i.push_ramp_pkt(16, 64'd0, 64'h100, header);
+ chdr_o.axis.tready = 1;
+ chdr_o.wait_for_pkt_get_info(header_out, stats);
+ `ASSERT_ERROR(stats.count==16, "Bad packet: Length mismatch");
+ `ASSERT_ERROR(header.sid==header_out.sid, "Bad packet: Wrong SID");
+ `ASSERT_ERROR(chdr_i.axis.tready, "Bus not ready");
+ `TEST_CASE_DONE(1);
+
+Here are the signatures of the two macros:
+
+``TEST_CASE_START``:
+
+ // Indicates the start of a test case
+ // This macro *must be* called inside the primary initial block
+ //
+ // Usage: `TEST_CASE_START(test_name)
+ // where
+ // - test_name: The name of the test.
+ //
+
+``TEST_CASE_DONE``:
+
+ // Indicates the end of a test case
+ // This macro *must be* called inside the primary initial block
+ // The pass/fail status of test case is determined based on the
+ // the user specified outcome and the number of fatal or error
+ // ASSERTs triggered in the test case.
+ //
+ // Usage: `TEST_CASE_DONE(test_result)
+ // where
+ // - test_result: User specified outcome
+ //
+
+In addition to the test case status, it is also possible to have asserts within
+a test case. We have wrappers for the different kinds of SystemVerilog asserts
+that additionally fail the test case in case the assert fails. An assert triggered
+in a test case will not affect the outcome of another (except for a fatal assert which
+halts the simulator). Supported assert macros:
+
+ // Wrapper around a an assert.
+ // ASSERT_FATAL throws an error assertion and halts the simulator
+ // if cond is not satisfied
+ //
+ // Usage: `ASSERT_FATAL(cond,msg)
+ // where
+ // - cond: Condition for the assert
+ // - msg: Message for the assert
+ //
+
+
+ // Wrapper around a an assert.
+ // ASSERT_ERROR throws an error assertion and fails the test case
+ // if cond is not satisfied. The simulator will *not* halt
+ //
+ // Usage: `ASSERT_ERROR(cond,msg)
+ // where
+ // - cond: Condition for the assert
+ // - msg: Message for the assert
+ //
+
+
+ // Wrapper around a an assert.
+ // ASSERT_WARNING throws an warning assertion but does not fail the
+ // test case if cond is not satisfied. The simulator will *not* halt
+ //
+ // Usage: `ASSERT_WARNING(cond,msg)
+ // where
+ // - cond: Condition for the assert
+ // - msg: Message for the assert
+ //
+
+### Optional Libraries
+
+It is encouraged to use (and create) reusable libraries in product specific
+test benches. Libraries can provide macros, modules, tasks and functions for
+ease-of-use with particular protocols and subsystems.
+
+The \ref md_usrp3_sim_writing_testbenches page has more information.