prjxray/docs/db_dev_process/newpart.rst

Guide to adding a new device to an existing family
==================================================

This documents how to add support for a new device. The running example
is the addition of the xc7a100t device to the Artix-7 family.

Adding a new device to an existing family is much simpler than adding a
new family, since the building blocks (tiles) are already known. There
are just more or fewer of them, arranged differently. No new fuzzers are
needed. You just need to rerun some fuzzers for the new device to
understand how the tiles are connected to each other and to IOs.

*If you are*\ \  **just**\ \  *adding a new package for a device that is
already supported, you can skip Steps 2 through 5.*

Note: Since this guide was written, the xc7a100t has become the primary
device in the database, not a secondary device as it was when it was
originally added. Therefore the files currently in the repo don’t match
what is described here. But if you look at the original PRs, they match
what is described in the examples here.

The main PR from the example is
`#1313 <https://github.com/SymbiFlow/prjxray/pull/1313>`__. Followup
fixes for problems revealed during testing are
`#1334 <https://github.com/SymbiFlow/prjxray/pull/1334>`__ and
`#1336 <https://github.com/SymbiFlow/prjxray/pull/1336>`__.

Step 0
~~~~~~

Fork a copy of https://github.com/SymbiFlow/prjxray on GitHub (go to the
page, click “Fork” button, select your own workspace).

Clone your fork, and make a new branch, with a name related to the new
device/package:

::

   git clone git@github.com:<yourUserID>/prjxray.git
   cd prjxray
   git checkout -b <new_branch_name>

Step 1
~~~~~~

Follow the Project X-Ray developer setup instructions in the
`documentation <https://symbiflow.readthedocs.io/en/latest/prjxray/docs/db_dev_process/readme.html>`__,
up through Step 7 and choose Option 1 (invoke the
``./download-latest-db.sh`` script). This script will clone the official
prjxray-db database under ``database/``. The following steps will make
changes under this directory. You may want to put these changes on your
own fork of ``prjxray-db`` for testing. This is explained at the end,
under “Database Updates”.

Step 2
~~~~~~

Add a new settings file. Usually you will start with an existing
settings file and modify it. Assuming you’re in prjxray/,

::

   cp settings/<baseline_device>.sh settings/<new_device>.sh
   git add settings/<new_device>.sh

Example:

::

   cp settings/artix7_200t.sh settings/artix7_100t.sh
   git add settings/artix7_100t.sh

Update the following values in the new settings file:

-  ``XRAY_PART`` –
   Important: choose a package that is fully bonded (typically the one with
   the largest number of pins). If the part that you’re actually interested
   in is different (with fewer bonded pins), it will be handled later. In
   the running example, the actual part of interest was the xc7a100tcsg324,
   since that is on the Arty A7-100T board. But here, the xc7a100tfgg676
   part is used; the xc7a100tcsg324 is handled later.

-  ``XRAY_ROI_TILEGRID`` – modify the bounding boxes to be a tight fit on
   your new part.

-  ``XRAY_IOI3_TILES`` – These tiles need special
   handling for an irregularity in Xilinx 7-series FPGAs. See the
   `comments <https://github.com/SymbiFlow/prjxray/blob/master/fuzzers/005-tilegrid/generate_full.py#L401>`__
   in the 005 fuzzer for more information.

`This <https://github.com/SymbiFlow/prjxray/blob/master/settings/artix7_100t.sh>`__
is what the new settings file looked like in the example.

Source this new settings file:

::

   source settings/<new_device>.sh

Step 3
~~~~~~

Add all informations about the part. YAML files for each family are located
at database/<family>/mapping/, which contain the part information (parts.yaml),
device to fabric mapping (devices.yaml) or hints about resources
(resources.yaml).

The first file contains a mapping between a part number and informations
about the device, package and speed grade used by the fuzzers. The
complete part number is used as key. Device, package and speedgrade are parts
of the part numbers.

::
  "xc7a100tcsg324-1":
    device: "xc7a100t"
    package: "csg324"
    speedgrade: "1"
    pins:
      0: "N15"
      1: "U17"
      2: "V17"
      3: "V16"
      4: "V14"
      5: "U14"
      6: "U16"

The second file maps devices to fabrics. Because some fabrics are added to
multiple devices, they are only generated for one and parts with the same link
to the result.

::
  "xc7a50t":
    fabric: "xc7a50t"
  "xc7a35t":
    fabric: "xc7a50t"

The last file contains information about the information about a resource
to support the fuzzers generating the informations. The dictionary pins
defines package pins with the following purpose:

-  ``00`` – this must be a clock pin. You can look at the device in the Vivado
   GUI interactively (click on IOs and check their properties until you find
   one with IS_CLOCK=true), or run a small clocked design in Vivado and see
   which pin is assigned to ‘clk’.

-  ``01`` and on – these should be normal data pins on the device.

Step 4
~~~~~~

Edit the top Makefile

-  Update the Makefile by adding the new device to the `correct
   list <https://github.com/tcal-x/prjxray/blob/fbf4dd897d5a1025ebfeb7c51c5077a6b6c9bc47/Makefile#L171>`__,
   so that the Makefile generates targets for the new device (used in
   Step 4). ``<new_device>`` is the basename of the new settings file
   that you just created.

::

   <FAMILY>_PARTS=<existing_devices> <new_device>

-  In our running example, we add ``artix7_100t`` to ``ARTIX_PARTS``:

::

   ARTIX_PARTS=artix7_200t artix7_100t

Step 5
~~~~~~

Make sure you’ve sourced your new device settings file (see the end of
step 2). Now it is time to run some fuzzers to figure out how the tiles
on your new device are connected.

Make the following target, with ``<new_device>`` as above, and setting
the parallelism factor ``-j<n>`` appropriate for the number of cores
your host has. The make job can benefit from large numbers of cores.

::

   make -j<n> MAX_VIVADO_PROCESS=<n> db-part-only-<new_device>

Again, ``<new_device>`` must match the base name of the new settings
file that was added. For example,

::

   make -j32 MAX_VIVADO_PROCESS=32 db-part-only-artix7_100t

-  It should run fuzzers 000, 001, 005, 072, 073, 074, and 075.

-  005 will take a long time. Using multiple cores will help.

-  074 *will fail* the first time, since it hasn’t been told to ignore
   certain wires.

   -  After it fails, go to the build directory
      ``cd fuzzers/074-dump_all/build_<XRAY_PART>`` (this is the
      ``XRAY_PART`` from the new settings script; in our example, the
      build directory is
      ``fuzzers/074-dump_all/build_xc7a100tfgg676-1/``).
   -  Run
      ``python3 ../analyze_errors.py --output_ignore_list  > new-ignored``
   -  Inspect and compare ``new-ignored`` against existing ignored wire
      files in ``../ignored_wires/``.
   -  If it looks good, copy it to an appropriately-named file:
      ``cp new-ignored ../ignored_wires/artix7/<XRAY_PART>_ignored_wires.txt``
      (in our example, it is
      ``../ignored_wires/artix7/xc7a100tfgg676-1_ignored_wires.txt``).
   -  Add it:
      ``git add ../ignored_wires/artix7/<XRAY_PART>_ignored_wires.txt``

-  Return to prjxray/ directory, and clean up 074 to prepare for the
   rerun: ``make -C fuzzers/074-dump-all clean``

-  Rerun the top make command,
   e.g. ``make -j32 MAX_VIVADO_PROCESS=32 db-part-only-artix7_100t``

Step 6
~~~~~~

The next task is handling the extra parts – those not fully bonded out.
These are usually the parts you actually have on the boards you buy.

-  Add a new entry in the appropriate ‘harness’ section for any
   alternative packages (typically with fewer pins, in this example, 324
   versus 676). If any ``XRAY_PIN_<XX>`` values you listed in the
   settings file are not bonded out on the new part, you must specify
   alternatives. In this example, we need to specify a new clock pin,
   ``XRAY_PIN_00=N15``. Here, ``XRAY_PART`` is the extra part, and
   ``XRAY_EQUIV_PART`` is the original, fully-bonded version:

::

   db-extras-artix7-harness:
       +source settings/artix7.sh && \
             XRAY_PART=xc7a35tftg256-1 $(MAKE) -C fuzzers roi_only
   +   +source settings/artix7_100t.sh && \
   +         XRAY_PART=xc7a100tcsg324-1 $(MAKE) -C fuzzers roi_only
       +source settings/artix7_200t.sh && \
             XRAY_PIN_00=V10 XRAY_PIN_01=W10 XRAY_PIN_02=Y11 XRAY_PIN_03=Y12 \
             XRAY_PART=xc7a200tsbg484-1 XRAY_EQUIV_PART=xc7a200tffg1156-1 \
             $(MAKE) -C fuzzers roi_only

Make the appropriate harness target (adjusting for your family):

::

   make -j32 db-extras-artix7-harness

This target will make updates for the extra parts of all of the family
devices, not just your new device.

Step 7
~~~~~~

Do a spot check.

-  Check that there are new part directories in the database under the family subdirectory, for example:

::

   $ ll database/artix7/xc7a*
   xc7a35tftg256-1:
   total 48
   -rw-rw-r-- 1 daniel daniel  8234 Jan  9 13:01 package_pins.csv
   -rw-rw-r-- 1 daniel daniel 18816 Jan  9 13:01 part.json
   -rw-rw-r-- 1 daniel daniel 13099 Jan  9 13:01 part.yaml

   xc7a50t:
   total 15480
   -rw-rw-r-- 1 daniel daniel  695523 Jan  9 12:53 node_wires.json
   -rw-rw-r-- 1 daniel daniel 8587682 Jan  9 12:53 tileconn.json
   -rw-rw-r-- 1 daniel daniel 6562851 Jan  9 10:31 tilegrid.json

   xc7a50tfgg484-1:
   total 52
   -rw-rw-r-- 1 daniel daniel 13056 Jan  9 09:54 package_pins.csv
   -rw-rw-r-- 1 daniel daniel 18840 Jan  9 09:58 part.json
   -rw-rw-r-- 1 daniel daniel 13099 Jan  9 09:58 part.yaml

In this case, the tile grid is the same size since it’s the same chip,
but the size of the package pins files differs, since there are
different numbers of bonded pins.

Note: These changes/additions under ``database/`` do *not* get checked
in. They are in the ``prjxray-db`` repo. This spot check is to make sure
that your changes in ``prjxray`` will do the right thing when the
official database is fully rebuilt. See “Database Updates” below for
more information.

Step 8
~~~~~~

Assuming everything looks good, commit to your ``prjxray`` fork/branch.
You should have a new file under settings/, a new ignored_wires file,
and a modified Makefile (see the `initial
PR <https://github.com/SymbiFlow/prjxray/pull/1313/files?file-filters%5B%5D=>`__
of the example for reference).

::

   git add Makefile settings/artix7_100t.sh
   git status
   git commit --signoff

Step 9
~~~~~~

Push to GitHub:

::

   git push origin <new_branch_name>

Then make a pull request. Navigate to the GitHub page for your
``prjxray`` fork/branch, and click the “New pull request” button.
Making the pull request will kick off continuous integration tests.
Watch the results and fix any issues.

Database Updates
~~~~~~~~~~~~~~~~

The process above (steps 4 and 5) will create some new files and modify
some existing files under database/, which is a different repo,
``prjxray-db``.

To test these changes before the next official prjxray-db gets built
(and even before your PR on prjxray is merged), you can put these
changes on your own fork of prjxray-db, and then test them in the
context of
`symbiflow-arch-defs <https://github.com/SymbiFlow/symbiflow-arch-defs>`__.

To put the db updates on your own fork, create your fork of
https://github.com/SymbiFlow/prjxray-db if you haven’t already. Then
follow one of the approaches suggested in the checked solution of this
StackOverflow
`post <https://stackoverflow.com/questions/25545613/how-can-i-push-to-my-fork-from-a-clone-of-the-original-repo>`__.

You are NEVER going to send a pull request on `prjxray-db`. The database is always rebuilt
from scratch. After your changes on prjxray are merged, they will
reflected in the next prjxray-db rebuild. The changes submitted to your
prjxray-db fork are only for your own testing.

To use your new repo/branch under
symbiflow-arch-defs/third_party/prjxray-db/, you will need to change the
submodule reference to point to your fork/branch of ``prjxray-db``.