3 This section describes the workflow and some best practices for developing
4 the Libre-SOC hardware. We use nmigen, yosys and symbiyosys, and this
5 page is intended not just to help you get set up, it is intended to
6 help advise you of some tricks and practices that will help you become
7 effective team contributors.
9 It is particularly important to bear in mind that we are not just
10 "developing code", here: we are creating a "lasting legacy educational
11 resource" for other people to learn from, and for businesses and students
12 alike to be able to use, learn from and augment for their own purposes.
14 It is also important to appreciate and respect that we are funded under
15 NLNet's Privacy and Enhanced Trust Programme <http://nlnet.nl/PET>. Full
16 transparency, readability, documentation, effective team communication
17 and formal mathematical proofs for all code at all levels is therefore
20 Therefore, we need not only to be "self-sufficient" (absolutely under no circumstances critically reliant on somebody else's servers **or protocols**) we also need to ensure that everything (including all communication such as the mailing list archives) are recorded, replicable, and accessible in perpetuity. Use of slack or a "forum" either actively prevents or makes that much harder.
22 # Collaboration resources
24 The main message here: **use the right tool for the right job**.
26 * mailing list: general communication and discussion.
27 * irc channel #libre-soc: real(ish)-time communication.
28 * bugtracker: task-orientated, goal-orientated *focussed* discussion.
29 * ikiwiki: document store, information store, and (editable) main website
30 * git repositories: code stores (**not binary or auto-generated output store**)
31 * ftp server (<https://ftp.libre-soc.org/>): large (temporary,
32 auto-generated) file store.
34 We will add an IRC channel at some point when there are enough people
35 to warrant having one (and it will be publicly archived)
37 Note also the lack of a "forum" in the above list. this is very deliberate. forums are a serious distraction when it comes to technical heavily goal-orientated development. recent internet users may enjoy looking up the "AOL metoo postings" meme.
39 Note also the complete lack of "social platforms". if we wanted to tell everybody how much better each of us are than anyone else in the team, how many times we made a commit (look at me, look at me, i'm so clever), and how many times we went to the bathroom, we would have installed a social media based project "management" system.
41 ## Main contact method: mailing list
43 To respect the transparency requirements, conversations need to be
44 public and archived (i.e not skype, not telegram, not discord,
45 and anyone seriously suggesting slack will be thrown to the
46 lions). Therefore we have a mailing list. Everything goes through
47 there. <https://lists.libre-soc.org/mailman/listinfo/libre-soc-dev>
48 therefore please do google "mailing list etiquette" and at the very
49 minimum look up and understand the following:
51 * This is a technical mailing list with complex topics. Top posting
52 is completely inappropriate. Don't do it unless you have mitigating
53 circumstances, and even then please apologise and explain ("hello sorry
54 using phone at airport flight soon, v. quick reply: ....")
55 * Always trim context but do not cut excessively to the point where people
56 cannot follow the discussion. Especially do not cut the attribution
57 ("On monday xxx wrote") of something that you are actually replying
59 * Use inline replies i.e. reply at the point in the relevant part of
60 the conversation, as if you were actually having a conversation.
61 * Follow standard IETF reply formatting, using ">" for cascaded
62 indentation of other people's replies. If using gmail, please: SWITCH
63 OFF RICH TEXT EDITING.
64 * Please for god's sake do not use "my replies are in a different
65 colour". Only old and highly regarded people still using AOL are allowed
66 to get away with that (such as Mitch).
67 * Start a new topic with a relevant subject line. If an existing
68 discussion changes direction, change the subject line to reflect the
69 new topic (or start a new conversation entirely, without using the
71 * DMARC is a pain on the neck. Try to avoid GPG signed messages. sigh.
72 * Don't send massive attachments. Put them online (no, not on facebook or
73 google drive or anywhere else that demands privacy violations) and provide
74 the link. Which should not require any kind of login to access. ask the
75 listadmin if you don't have anywhere suitable: FTP access can be arranged.
77 ### Actionable items from mailing list
79 If discussions result in any actionable items, it is important not to
80 lose track of them. Create a bugreport, find the discussion in the
81 archives <https://lists.libre-soc.org/pipermail/libre-soc-dev/>,
82 and put the link actually in the bugtracker as one of the comments.
84 At some point in any discussion, the sudden realisation may dawn on one
85 or more people that this is an "actionable" discussion. at that point
86 it may become better to use <https://bugs.libre-soc.org/>
87 itself to continue the discussion rather than to keep on dropping copies
88 of links into the bugtracker. The bugtracker sends copies of comments
89 *to* the list however this is 'one-way' (note from lkcl: because this
90 involves running an automated perl script from email, on every email,
91 on the server, that is a high security risk, and i'm not doing it. sorry.)
93 ### Mailing list != editable document store
95 Also, please do not use the mailing list as an "information or document
96 store or poor-man's editor". We have the wiki for that. Edit a page and
97 tell people what you did (summarise rather than drop the entire contents
98 at the list) and include the link to the page.
100 Or, if it is more appropriate, commit a document (or source code)
101 into the relevant git repository then look up the link in the gitweb
102 source tree browser and post that (in the bugtracker or mailing list)
103 See <https://git.libre-soc.org/>
105 ### gmail "spam"ifying the list
107 See <https://blog.kittycooper.com/2014/05/keeping-my-mailing-list-emails-out-of-gmails-spam-folder/>
109 Basically it is possible to select any message from the list, create a "filter" (under "More"),
110 and, on the 2nd dialog box, click the "never send this to Spam" option.
114 bugzilla. old and highly effective. sign up in the usual way. any
115 problems, ask on the list.
117 Please do not ask for the project to be transferred to github or other
118 proprietary nonfree service "because it's soooo convenient", as the
119 lions are getting wind and gout from overfeeding on that one.
123 Runs the main libre-soc.org site (including this page). effective,
124 stunningly light on resources, and uses a git repository not a database.
125 That means it can be edited offline.
127 Usual deal: register an account and you can start editing and contributing
130 Hint: to create a new page, find a suitable page that would link to it,
131 first, then put the link in of the page you want to create, as if the
132 page already exists. Save that page, and you will find a question mark
133 next to the new link you created. click that link, and it will fire up a
134 "create new page" editor.
136 Wiki pages are formatted in [[markdown|ikiwiki/markdown]] syntax.
138 Hint again: the wiki is backed by a git repository. Don't go overboard
139 but at the same time do not be afraid that you might "damage" or "lose"
140 pages. Although it would be a minor pain, the pages can always be
141 reverted or edited by the sysadmins to restore things if you get in a tiz.
143 Assistance in creating a much better theme greatly appreciated. e.g.
144 <http://www.math.cmu.edu/~gautam/sj/blog/20140720-ikiwiki-navbar.html>
148 We use git. More on this below. We also use [gitolite3](https://gitolite.com/gitolite/) running on a dedicated server. again, it is extremely effective and low resource utilisation. Reminder: lions are involved if github is mentioned.
150 [gitweb](https://git.wiki.kernel.org/index.php/Gitweb) is provided which does a decent job. <https://git.libre-soc.org/>
152 [Git](https://en.wikipedia.org/wiki/Git) does version control, ie it tracks changes to files so that previous versions can be got back or compared.
156 <https://ftp.libre-soc.org/> is available for storing large files
157 that do not belong in a git repository, if we have (or ever need)
158 any. Images (etc.) if small and appropriate should go into the
159 wiki, however .tgz archives (etc.) and, at some point, binaries,
160 should be on the ftp server.
162 Ask on the list if you have a file that belongs on the ftp server.
166 As an aside: all this is "old school" and run on a single core 512MB
167 VM with only a 20GB HDD allocation. it costs only 8 GBP per month from
168 mythic-beasts and means that the project is in no way dependent on anyone
169 else - not microsoft, not google, not facebook, not amazon.
171 We tried [gitlab](https://about.gitlab.com/). it didn't go well. please don't ask to replace the above extremely resource-efficient services with it.
175 RAM is the biggest requirement. Minimum 16GB, the more the better (32
176 or 64GB starts to reach "acceptable" levels. Disk space is not hugely
177 critical: 256GB SSD should be more than adequate. Simulations and
178 FPGA compilations however are where raw processing power is a must.
179 High end Graphics Cards are nonessential.
181 What is particularly useful is to have hi-res screens (curved is
182 *strongly* recommended if the LCD is over 24in wide, to avoid eyeballs
183 going "prism" through long term use), and to have several of them: the
184 more the better. Either a DisplayLink UD160A (or more modern variant)
185 or simply using a second machine (lower spec hardware because it will
186 run editors) is really effective.
188 Also it is really recommended to have a UHD monitor (4k - 3840x2160),
189 or at least 2560x1200. If given a choice, 4:3 aspect ratio is better
190 than 16:9 particularly when using several of them. However, caveat
191 (details below): please when editing do not assume that everyone will
192 have access to such high resolution screens.
196 First install and become familiar with [Debian](https://www.debian.org/) ([Ubuntu](https://ubuntu.com/) if you absolutely
197 must) for standardisation cross-team and so that toolchain installation
198 is greatly simplified. yosys in particular warns that trying to use
199 Windows, BSD or MacOS will get you into a world of pain.
201 Only a basic GUI desktop is necessary: fvwm2, xfce4, lxde are perfectly
202 sufficient (alongside wicd-gtk for network management). Other more
203 complex desktops can be used however may consume greater resources.
205 # editors and editing
207 Whilst this is often a personal choice, the fact that many editors are
208 GUI based and run full-screen with the entire right hand side *and* middle
209 *and* the majority of the left side of the hi-res screen entirely unused
210 and bereft of text leaves experienced developers both amused and puzzled.
212 At the point where such full-screen users commit code with line lengths
213 well over 160 characters, that amusement quickly evaporates.
215 Where the problems occur with full-screen editor usage is when a project
216 is split into dozens if not hundreds of small files (as this one is). At
217 that point it becomes pretty much essential to have as many as six to
218 eight files open *and on-screen* at once, without overlaps i.e. not in
219 hidden tabs, next to at least two if not three additional free and clear
220 terminals into which commands are regularly and routinely typed (make,
221 git commit, nosetests3 etc). Illustrated with the following 3840x2160
222 screenshot (click to view full image), where *every one* of those 80x70
223 xterm windows is *relevant to the task at hand*.
225 [[!img 2020-01-24_11-56.png size=640x ]]
227 (hint/tip: fvwm2 set up with "mouse-over to raise focus, rather than
228 additionally requiring a mouse click, can save a huge amount of cumulative
229 development time here, switching between editor terminal(s) and the
232 Once this becomes necessary, it it turn implies that having greater
233 than 80 chars per line - and running editors full-screen - is a severe
234 hinderance to an essential *and highly effective* workflow technique.
236 Additionally, care should be taken to respect that not everyone will have
237 200+ column editor windows and the eyesight of a hawk. They may only have
238 a 1280 x 800 laptop which barely fits two 80x53 xterms side by side.
239 Consequently, having excessively long functions is also a hindrance to
240 others, as such developers with limited screen resources would need to
241 continuously page-up and page-down to read the code even of a single
244 This helps explain in part, below, why compliance with [pep8](https://pep8.org/) is enforced, including its 80 character limit. In short: not everyone has the same "modern" GUI workflow or has access to the same computing resources as
245 you, so please do respect that.
247 More on this concept is
248 [here](https://www.linuxjournal.com/content/line-length-limits).
249 Note *very pointedly* that Linus Torvalds *specifically* states that
250 he does not want Linux kernel development to become the exclusive
251 domain of the "wealthy". That means **no** to assumptions about
252 access to ultra-high resolution screens.
254 # Software prerequisites
256 Whilst many resources online advocate "`sudo`" in front of all root-level
257 commands below, this quickly becomes tiresome. run "`sudo bash`", get a
258 root prompt, and save yourself some typing.
261 * apt-get install vim exuberant-ctags
262 * apt-get install build-essential
263 * apt-get install git python3.7 python3.7-dev python-nosetest3
264 * apt-get install graphviz xdot gtkwave
265 * apt-get install python3-venv
266 * apt-get install python-virtualenv # this is an alternative to python3-venv
267 * return to user prompt (ctrl-d)
269 This will get you python3 and other tools that are needed. [graphviz](https://graphviz.org/) is essential for showing the interconnections between cells, and [gtkwave](http://gtkwave.sourceforge.net/) is essential for debugging.
271 If you would like to save yourself a lot more typing, check out the
272 [dev-env-setup](https://git.libre-soc.org/?p=dev-env-setup.git;a=summary)
273 repository, examine the scripts there and use them to automate much of
278 Look up good tutorials on how to use git effectively. There are so many
279 it is hard to recommend one. This is however essential. If you are not
280 comfortable with git, and you let things stay that way, it will seriously
281 impede development progress.
283 If working all day you should expect to be making at least two commits per
284 hour, so should become familiar with it very quickly. If you are *not*
285 doing around 2 commits per hour, something is wrong and you should read
286 the workflow instructions below more carefully, and also ask for advice
289 Worth noting: *this project does not use branches*. All code is committed
290 to master and we *require* that it be either zero-impact additions or that
291 relevant unit tests pass 100%. This ensures that people's work does not
292 get "lost" or isolated and out of touch due to major branch diversion,
293 and that people communicate and coordinate with each other.
297 Follow the source code (git clone) instructions here:
298 <http://www.clifford.at/yosys/download.html>
300 Or, alternatively, use the
301 [yosys-et-al](https://git.libre-soc.org/?p=dev-env-setup.git;a=blob;f=yosys-et-al;hb=HEAD)
302 script (which also installs symbiyosys and its dependencies)
304 Do not try to use a fixed revision of yosys (currently 0.9), nmigen is evolving
305 and frequently interacts with yosys.
307 [Yosys](http://www.clifford.at/yosys/) is a framework for Verilog RTL.
308 [Verilog](https://en.wikipedia.org/wiki/Verilog) is a hardware description language.
309 RTL [Register Transfer Level](https://en.wikipedia.org/wiki/Register-transfer_level) models how data moves between [registers](https://en.wikipedia.org/wiki/Hardware_register).
313 To install follow the [instructions here](https://symbiyosys.readthedocs.io/en/latest/install.html) Once done look at [A simple BMC example](https://symbiyosys.readthedocs.io/en/latest/quickstart.html)
315 You do not have to install all of those (avy, boolector can be left
316 out if desired) however the more that are installed the more effective
317 the formal proof scripts will be (less resource utilisation in certain
320 [SymbiYosys](https://symbiyosys.readthedocs.io/en/latest/) (sby) is a front-end driver program for Yosys-based formal hardware verification flows.
324 [nmigen](https://m-labs.hk/gateware/nmigen/) may be installed as follows:
328 * git clone https://github.com/nmigen/nmigen.git
331 * python3 setup.py develop
334 Testing can then be carried out with "python3 setup.py test"
336 nmigen is a Python toolbox for building complex digital hardware.
338 ## Softfloat and sfpy
340 These are a test suite dependency for the [ieee754fpu](https://www.gaisler.com/index.php/products/ipcores/ieee754fpu) library, and
341 will be changed in the future to use Jacob's [simple-soft-float](https://crates.io/crates/simple-soft-float)
342 library. In the meantime, sfpy can be built as follows:
344 git clone --recursive https://github.com/billzorn/sfpy.git
347 git apply ../softposit_sfpy_build.patch
348 git apply /path/to/ieee754fpu/SoftPosit.patch
349 cd ../berkely-softfloat-3
350 # Note: Do not apply the patch included in sfpy for berkely-softfloat,
351 # it contains the same changes as this one
352 git apply /path/to/ieee754fpu/berkeley-softfloat.patch
355 # prepare a virtual environment for building
358 # or, if you prefer the old way:
359 # virtualenv -p python3 .env
361 # install dependencies
362 source .env/bin/activate
363 pip3 install --upgrade -r requirements.txt
368 make inplace -j$(nproc)
372 deactivate # deactivates venv, optional
373 pip3 install dist/sfpy*.whl
375 You can test your installation by doing the following:
378 >>> from sfpy import *
381 It should print out `Posit8(1.3125)`
383 ## qemu, cross-compilers, gdb
385 As we are doing POWER ISA, POWER ISA compilers, toolchains and
386 emulators are required.
388 Install powerpc64 gcc:
390 apt-get install gcc-9-powerpc64-linux-gnu
394 apt-get install qemu-system-ppc
396 Install gdb from source. Obtain the latest tarball, unpack it, then:
398 cd gdb-9.1 (or other location)
401 ../configure --srcdir=.. --host=x86_64-linux --target=powerpc64-linux-gnu
405 [gdb](https://en.wikipedia.org/wiki/GNU_Debugger) lets you debug running programs.
406 [qemu](https://www.qemu.org/) emulates processors, you can run programs under qemu.
408 ## power_instruction_analyzer (pia)
410 We have a custom tool built in rust by programmerjake to help analyze
411 the power instructions execution on *actual* hardware.
413 Note: a very recent version of pip3 is required for this to work.
417 curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
419 Make sure we have the correct and up-to-date rust compiler (rustc):
421 rustup default stable
424 Use rust's package manager *cargo* to install the rust-python build tool maturin:
426 cargo install maturin
428 Install from git source by doing the following:
430 git clone https://salsa.debian.org/Kazan-team/power-instruction-analyzer.git pia
432 maturin build --cargo-extra-args=--features=python-extension
433 python3 -m pip install --user target/wheels/*.whl
435 Note: an ongoing bug in maturin interferes with successful installation. This can be worked around by explicitly installing only the `.whl` files needed rather than installing everything (`\*.whl`).
439 As this is an actual ASIC, we do not rely on an FPGA's JTAG TAP interface, instead require a full complete independent implementation of JTAG. Staf Verhaegen has one, with a full test suite, and it is superb and well-written. The Libre-SOC version includes DMI (Debug Memory Interface):
441 git clone https://git.libre-soc.org/git/c4m-jtag.git/
443 Included is an IDCODE tap point, Wishbone Master (for direct memory read and write, fully independent of the core), IOPad redirection and testing, and general purpose shift register capability for any custom use.
445 We added a DMI to JTAG bridge in LibreSOC which is directly connected to the core, to access registers and to be able to start and stop the core and change the PC. In combination with the JTAG Wishbone interface the test [ASIC](https://en.wikipedia.org/wiki/Application-specific_integrated_circuit) can have a bootloader uploaded directly into onboard [SRAM](https://en.wikipedia.org/wiki/Static_random-access_memory) and execution begun.
447 [Chips4Makers](https://chips4makers.io/) make it possible for makers and hobbyists to make their own open source chips.
449 [JTAG](https://en.wikipedia.org/wiki/JTAG) (Joint Test Action Group) is an industry standard for verifying designs and testing printed circuit boards after manufacture.
451 The [Wishbone bus](https://en.wikipedia.org/wiki/Wishbone_%28computer_bus%29) is an open source hardware computer bus intended to let the parts of an integrated circuit communicate with each other.
455 See [[HDL_workflow/coriolis2]] page, for those people doing layout work.
457 # Registering for git repository access
459 After going through the onboarding process and having agreed to take
460 responsibility for certain tasks, ask on the mailing list for git
461 repository access, sending in a public key (`id_rsa.pub`). If you do
462 not have one then generate it with `ssh-keygen -t rsa`. You will find it
465 NEVER SEND ANYONE THE PRIVATE KEY. By contrast the public key, on
466 account of being public, is perfectly fine to make... err... public.
468 Create a file `~/.ssh/config` with the following lines:
470 Host git.libre-soc.org
473 Wait for the Project Admin to confirm that the ssh key has been added
474 to the required repositories. Once confirmed, you can clone any of the
475 repos at https://git.libre-soc.org/:
477 git clone gitolite3@git.libre-soc.org:REPONAME.git
479 Alternatively, the .ssh/config can be skipped and this used:
481 git clone ssh://gitolite3@git.libre-soc.org:922/REPONAME.git
485 Although there are methods online which describe how (and why) these
486 settings are normally done, honestly it is simpler and easier to open
487 ~/.gitconfig and add them by hand.
489 core.autocrlf is a good idea to ensure that anyone adding DOS-formatted
490 files they don't become a pain. pull.rebase is something that is greatly
491 preferred for this project because it avoids the mess of "multiple
492 extra merge git tree entries", and branch.autosetuprebase=always will,
493 if you want it, always ensure that a new git checkout is set up with rebase.
502 autosetuprebase = always
504 # Checking out the HDL repositories
506 Before running the following, install the dependencies. This is easiest
507 done with this script <https://git.libre-soc.org/?p=dev-env-setup.git;a=blob;f=install-hdl-apt-reqs;hb=HEAD>
511 * git clone gitolite3@git.libre-soc.org:nmutil.git
512 * git clone gitolite3@git.libre-soc.org:ieee754fpu.git
513 * git clone gitolite3@git.libre-soc.org:nmigen-soc.git
514 * git clone gitolite3@git.libre-soc.org:soc.git
516 In each of these directories, in the order listed, track down the
517 `setup.py` file, then, as root (`sudo bash`), run the following:
519 * python3 setup.py develop
521 The reason for using "develop" mode is that the code may be edited
522 in-place yet still imported "globally". There are variants on this theme
523 for multi-user machine use however it is often just easier to get your
524 own machine these days.
526 The reason for the order is because soc depends on ieee754fpu, and
527 ieee754fpu depends on nmutil
529 If "`python3 setup.py install`" is used it is a pain: edit, then
530 install. edit, then install. It gets extremely tedious, hence why
531 "develop" was created.
537 * new members, add yourself to the [[about_us]] page and create yourself a home page using someone else's page as a template.
538 * communicate on the mailing list or the bugtracker an intent to take
539 responsibility for a particular task.
540 * assign yourself as the bug's owner
541 * *keep in touch* about what you are doing, and why you are doing it.
542 * edit your home page regularly, particularly to track tasks so that they can be paid by NLNet.
543 * if you cannot do something that you have taken responsibility for,
544 then unless it is a dire personal emergency please say so, on-list. we
545 won't mind. we'll help sort it out.
547 Regarding the above it is important that you read, understand, and agree
548 to the [[charter]] because the charter is about ensuring that we operate
549 as an effective organisation. It's *not* about "setting rules and meting
554 for actual code development
558 * plan in advance to write not just code but a full test suite for
559 that code. **this is not optional**. large python projects that do not
560 have unit tests **FAIL** (see separate section below).
561 * Prioritise writing formal proofs and a single clear unit test that is more like a "worked example".
562 We receive NLNet funds for writing formal proofs, plus they
563 cover corner cases and take far less time to write
565 ### Commit tested or zero-dependent code
567 * only commit code that has been tested (or is presently unused). other
568 people will be depending on you, so do take care not to screw up.
569 not least because, as it says in the [[charter]] it will be your
570 responsibility to fix. that said, do not feel intimidated: ask for help
571 and advice, and you'll get it straight away.
575 * commit often. several times a day, and "git push" it. this is
576 collaboration. if something is left even overnight uncommitted and not
577 pushed so that other people can see it, it is a red flag. if you find
578 yourself thinking "i'll commit it when it's finished" or "i don't want to
579 commit something that people might criticise" *this is not collaboration*,
580 it is making yourself a bottleneck. pair-programming is supposed to help
581 avoid this kind of thing however pair-programming is difficult to organise
582 for remote collaborative libre projects (suggestions welcomed here)
584 ### Enable editor auto-detection of file changes by external programs
586 This is important. "`git pull`" will merge in changes. If you then
587 arbitrarily save a file without re-loading it, you risk destroying
590 ### Absolutely no auto-generated output
592 * **do not commit autogenerated output**. write a shell script and commit
593 that, or add a `Makefile` to run the command that generates the output, but
594 **do not** add the actual output of **any** command to the repository.
595 ever. this is really important. even if it is a human-readable file
596 rather than a binary object file.
597 it is very common to add pdfs (the result of running `latex2pdf`) or configure.in (the result of running `automake`), they are an absolute nuisance and interfere hugely with git diffs, as well as waste hard disk space *and* network bandwidth. don't do it.
599 ### Write commands that do tasks and commit those
601 * if the command needed to create any given autogenerated output is not
602 currently in the list of known project dependencies, first consult on
603 the list if it is okay to make that command become a hard dependency of
604 the project (hint: java, node.js php and .NET commands may cause delays
605 in response time due to other list participants laughing hysterically),
606 and after a decision is made, document the dependency and how its source
607 code is obtained and built (hence why it has to be discussed carefully)
608 * if you find yourself repeating commands regularly, chances are high
609 that someone else will need to run them, too. clearly this includes
610 yourself, therefore, to make everyone's lives easier including your own,
611 put them into a `.sh` shell script (and/or a `Makefile`), commit them to
612 the repository and document them at the very minimum in the README,
613 INSTALL.txt or somewhere in a docs folder as appropriate. if unsure,
614 ask on the mailing list for advice.
616 ### Keep commits single-purpose
618 * edit files making minimal *single purpose* modifications (even if
619 it involves multiple files. Good extreme example: globally changing
620 a function name across an entire codebase is one purpose, one commit,
621 yet hundreds of files. miss out one of those files, requiring multiple
622 commits, and it actually becomes a nuisance).
624 ### Run unit tests prior to commits
626 * prior to committing make sure that relevant unit tests pass, or that
627 the change is a zero-impact addition (no unit tests fail at the minimum)
629 ### Do not break existing code
631 * keep working code working **at all times**. find ways to ensure that this is the case. examples include writing alternative classes that replace existing functionality and adding runtime options to select between old and new code.
633 ### Small commits with relevant commit message
635 * commit no more than around 5 to 10 lines at a time, with a CLEAR message
636 (no "added this" or "changed that").
637 * if as you write you find that the commit message involves a *list* of
638 changes or the word "and", then STOP. do not proceed: it is a "red flag"
639 that the commit has not been properly broken down into separate-purpose
640 commits. ask for advice on-list on how to proceed.
642 ### Exceptions to small commit: atomic single purpose commit
644 * if it is essential to commit large amounts of code, ensure that it
645 is **not** in use **anywhere** by any other code. then make a *small*
646 (single purpose) followup commit which actually puts that code into use.
648 This last rule is kinda flexible, because if you add the code *and* add
649 the unit test *and* added it into the main code *and* ran all relevant
650 unit tests on all cascade-impacted areas by that change, that's perfectly
651 fine too. however if it is the end of a day, and you need to stop and
652 do not have time to run the necessary unit tests, do *not* commit the
653 change which integrates untested code: just commit the new code (only)
654 and follow up the next day *after* running the full relevant unit tests.
656 ### Why such strict rules?
658 The reason for all the above is because python is a dynamically typed language.
659 make one tiny change at the base level of the class hierarchy and the
660 effect may be disastrous.
662 It is therefore worth reiterating: make absolutely certain that you *only*
663 commit working code or zero-impact code.
665 Therefore, if you are absolutely certain that a new addition (new file,
666 new class, new function) is not going to have any side-effects, committing
667 it (a large amount of code) is perfectly fine.
669 As a general rule, however, do not use this an an excuse to write code
670 first then write unit tests as an afterthought. write *less* code *in
671 conjunction* with its (more basic) unit tests, instead. then, folliw up with
672 additions and improvements.
674 The reason for separating out commits to single purpose only becomes
675 obvious (and regretted if not followed) when, months later, a mistake
676 has to be tracked down and reverted. if the commit does not have an
677 easy-to-find message, it cannot even be located, and once found, if the
678 commit confuses several unrelated changes, not only the diff is larger
679 than it should be, the reversion process becomes extremely painful.
683 * all code needs to conform to pep8. use either pep8checker or better
684 run autopep8. however whenever committing whitespace changes, *make a
685 separate commit* with a commit message "whitespace" or "autopep8 cleanup".
686 * pep8 REQUIRES no more than 80 chars per line. this is non-negotiable. if
687 you think you need greater than 80 chars, it *fundamentally* indicates
688 poor code design. split the code down further into smaller classes
691 ### Docstring checker
693 * TBD there is a docstring checker. at the minimum make sure to have
694 an SPD license header, module header docstring, class docstring and
695 function docstrings on at least non-obvious functions.
697 ### Clear code commenting and docstrings
699 * make liberal but not excessive use of comments. describe a group of
700 lines of code, with terse but useful comments describing the purpose,
701 documenting any side-effects, and anything that could trip you or other
702 developers up. unusual coding techniques should *definitely* contain
705 ### Only one class per module (ish)
707 * unless they are very closely related, only have one module (one class)
708 per file. a file only 25 lines long including imports and docstrings
709 is perfectly fine however don't force yourself. again, if unsure,
712 ### File and Directory hierarchy
714 * *keep files short and simple*. see below as to why
715 * create a decent directory hierarchy but do not go mad. ask for advice
720 * please do not use "from module import \*". it is extremely bad practice,
721 causes unnecessary resource utilisation, makes code readability and
722 tracking extremely difficult, and results in unintended side-effects.
724 Example: often you want to find the code from which a class was imported.
725 nirmally you go to the top of the file, check the imports, and you know
726 exactly which file has the class because of the import path. by using
727 wildcards, you have absolutely *no clue* which wildcard imported which
730 Example: sometimes you may accidentally have duplicate code maintained
731 in two or more places. editing one of them you find, puzzlingly, that
732 the code behaves in some files with the old behaviour, but in others it
733 works. after a massive amount of investigation, you find that the working
734 files happen to have a wildcard import of the newer accidental duplicate
735 class **after** the wildcard import of the older class with exactly the
736 same name. if you had used explicit imports, you would have spotted
737 the double import of the class from two separate locations, immediately.
739 Really. don't. use. wildcards.
741 ### Keep file and variables short but clear
743 * try to keep both filenames and variable names short but not ridiculously
744 obtuse. an interesting compromise on imports is "from ridiculousfilename
745 import longsillyname as lsn", and to assign variables as well: "comb =
746 m.d.comb" followed by multiple "comb += nmigen_stmt" lines is a good trick
747 that can reduce code indentation by 6 characters without reducing clarity.
749 Additionally, use comments just above an obtuse variable in order to
750 help explain what it is for. In combination with keeping the the module
751 itself short, other readers will not need to scroll back several pages
752 in order to understand the code.
754 Yes it is tempting to actually use the variables as
755 self-explanatory-comments and generally this can be extremely good
756 practice. the problem comes when the variable is so long that a function
757 with several parameters csn no longer fit on a single line, and takes
758 up five to ten lines rather than one or two. at that point, the length
759 of the code is adversely affected and thus so is readability by forcing
760 readers to scroll through reams of pages.
762 It is a tricky balance: basically use your common sense, or just ask
763 someone else, "can you understand this code?"
765 ### Reasons for code structure
767 Regarding code structure: we decided to go with small modules that are
768 both easy to analyse, as well as fit onto a single page and be readable
769 when displayed as a visual graph on a full UHD monitor. this is done
772 * using the capability of nmigen (TODO crossref to example) output the
773 module to a yosys ilang (.il) file
774 * in a separate terminal window, run yosys
775 * at the yosys prompt type "read_ilang modulename.il"
776 * type "show top" and a graphviz window should appear. note that typing
777 show, then space, then pressing the tab key twice will give a full list
778 of submodules (one of which will be "top")
780 You can now fullsize the graphviz window and scroll around. if it looks
781 reasonably obvious at 100% zoom, i.e the connections can be clearly
782 related in your mind back to the actual code (by matching the graph names
783 against signals and modules in the original nmigen code) and the words are
784 not tiny when zoomed out, and connections are not total incomprehensible
785 spaghetti, then congratulations, you have well-designed code. If not,
786 then this indicates a need to split the code further into submodules
787 and do a bit more work.
789 The reasons for doing a proper modularisation job are several-fold:
791 * firstly, we will not be doing a full automated layout-and-hope
792 using alliance/coriolis2, we will be doing leaf-node thru tree node
793 half-automated half-manual layout, finally getting to the floorplan,
794 then revising and iteratively adjusting.
795 * secondly, examining modules at the gate level (or close to it) is just
796 good practice. poor design creeps in by *not* knowing what the tools
797 are actually doing (word to experienced developers: yes, we know that
798 the yosys graph != final netlist).
799 * thirdly, unit testing, particularly formal proofs, is far easier on
800 small sections of code, and complete in a reasonable time.
802 ## Special warning / alert to vim users!
804 Some time around the beginning of 2019 some bright spark decided that
805 an "auto-recommend-completion-of-stuff" option would be a nice, shiny
806 idea to enable by default from that point onwards.
808 This incredibly annoying "feature" results in tabs (or spaces) being
809 inserted "on your behalf" when you press return on one line, for your
810 "convenience" of not needing to type lots of spaces/tabs just to get
811 to the same indentation level.
813 Of course, this "feature", if you press return on one line in edit
814 mode and then press "escape", leaves a bundle-of-joy extraneous
815 whitespace **exactly** where you don't want it, and didn't ask for it,
816 pooped all over your file.
818 Therefore, *please*: **before** running "git commit", get into the
819 habit of always running "git diff", and at the very minimum
820 speed-skim the entire diff, looking for tell-tale "red squares"
821 (these show up under bash diff colour-syntax-highlighting) that
822 inform you that, without your knowledge or consent, vim has
823 "helpfully" inserted extraneous whitespace.
825 Remove them **before** git committing because they are not part
826 of the actual desired code-modifications, and committing them
827 is a major and constant distraction for reviewers about actual
828 important things like "the code that actually *usefully* was
829 modified for that commit"
831 This has the useful side-effect of ensuring that, right before
832 the commit, you've got the actual diff right in front of you
833 in the xterm window, on which you can base the "commit message".
837 For further reading, see the wikipedia page on
838 [Test-driven Development](https://en.wikipedia.org/wiki/Test-driven_development)
840 This deserves its own special section. It is extremely important to
841 appreciate that without unit tests, python projects are simply unviable.
842 Python itself has over 25,000 individual tests.
844 This can be quite overwhelming to a beginner developer, especially one
845 used to writing scripts of only 100 lines in length.
847 Thanks to Samuel Falvo we learned that writing unit tests as a formal
848 proof is not only shorter, it's also far more readable and also, if
849 written properly, provides 100% coverage of corner-cases that would
850 otherwise be overlooked or require tens to hundreds of thousands of
853 No this is not a joke or even remotely hypothetical, this is an actual
856 The ieee754fpu requires several hundreds of thousands of tests to be
857 run (currently needing several days to run them all), and even then we
858 cannot be absolutely certain that all possible combinations of input have
859 been tested. With 2^128 permutations to try with 2 64 bit FP numbers
860 it is simply impossible to even try.
862 This is where formal proofs come into play.
864 Samuel illustrated to us that "ordinary" unit tests can then be written
865 to *augment* the formal ones, serving the purpose of illustrating how
866 to use the module, more than anything.
868 However it is appreciated that writing formal proofs is a bit of a
869 black art. This is where team collaboration particularly kicks in,
870 so if you need help, ask on the mailing list.
872 ## Don't comment out unit tests: add them first (as failures) and fix code later
874 Unit tests serve an additional critical purpose of keeping track of code
875 that needs to be written. In many cases, you write the unit test *first*,
876 despite knowing full well that the code doesn't even exist or is completely
877 broken. The unit test then serves as a constant and important reminder
878 to actually fix (or write) the code.
880 Therefore, *do not* comment out unit tests just because they "don't work".
881 If you absolutely must stop a unit test from running, **do not delete it**.
882 Simply mark it with an appropriate
883 ["skip" decorator](https://docs.python.org/3/library/unittest.html#skipping-tests-and-expected-failures),
884 preferably with a link to a URL in the [bugtracker](https://bugs.libre-soc.org/)
885 with further details as to why the unit test should not be run.
889 Find appropriate tutorials for nmigen and yosys, as well as symbiyosys.
891 * Robert Baruch's nmigen tutorials look really good:
892 <https://github.com/RobertBaruch/nmigen-tutorial>
893 * Although a verilog example this is very useful to do
894 <https://symbiyosys.readthedocs.io/en/latest/quickstart.html#first-step-a-simple-bmc-example>
895 * This tutorial looks pretty good and will get you started
896 <http://blog.lambdaconcept.com/doku.php?id=nmigen:nmigen_install> and
897 walks not just through simulation, it takes you through using gtkwave
899 * There exist several nmigen examples which are also executable
900 <https://github.com/m-labs/nmigen/tree/master/examples/> exactly as
901 described in the above tutorial (python3 filename.py -h)