Annotation of wikisrc/kyua/import.mdwn, revision 1.5

1.1       jmmv        1: [[!meta title="Kyua: The way into NetBSD"]]
                      2: [[!toc levels=2]]
                      4: **Project owner: [Julio Merino](**  
1.5     ! jmmv        5: **Status: under review by tech-userlevel@ as of 2012-09-03.**
1.1       jmmv        6: 
                      7: The import of Kyua into NetBSD to replace the deprecated ATF tools is
                      8: planned to happen in NetBSD 7.0.  The ATF libraries will remain in place,
                      9: and as such no changes will happen to the tests that already live in
                     10: `src/tests`.
                     12: The transition from ATF to Kyua includes steps and tools to offer backwards
                     13: compatibility with users that may rely on the ATF tools shipped since
                     14: NetBSD 5.0.  These backwards compatibility tools cover the most common use
                     15: cases but might not be perfect initially.  Despite these provisions, this
                     16: plan should be executed well in advance the creation of the 7.0 branch to
                     17: ensure there is enough time in NetBSD-current to flesh out any major
                     18: problems.
1.3       jmmv       20: **Before moving on, please read the
1.2       jmmv       21: [[Kyua: An introduction for NetBSD users|/kyua]] page.  You should be
1.1       jmmv       22: familiar with the structure of Kyua and its major components to be able to
1.3       jmmv       23: review this plan.**
1.1       jmmv       24: 
                     25: # Proposed changes
                     27: The final user visible changes that this project will bring to NetBSD are
                     28: as follows.  Because these change the availability of tools that have
                     29: already been shipped since 5.0, the modifications are staged to happen
                     30: across two major releases:
                     32: ## For NetBSD 7.0
                     34: * Addition of the new `kyua` command-line tool.  This provides the runtime
                     35:   engine for the test programs (`atf-run`'s functionality) and the ability
                     36:   to generate reports of the results in plain text form and HTML form
                     37:   (`atf-report`'s functionality).  This tool is able to execute all the
                     38:   existing test programs without modifications.
                     40: * Replacement of the `Atffile`s files in `/usr/tests` with `Kyuafile`s.
                     42: * Ability to generate HTML reports right form the base system, with such
                     43:   reports hooked into the various continuous build systems.
                     45: * Replacement of the `atf-run` and `atf-report` tools with shell scripts
                     46:   that provide backwards compatibility implementations based on `kyua`.
                     47:   These can deal with old-style `Atffile`s so that users with custom test
                     48:   suites can continue to run them.
                     50: * Introduction of a new `atf2kyua` script to convert old `Atffile`s to
                     51:   `Kyuafile`s.  This is used internally by the compatibility `atf-run`
                     52:   script, but can also be invoked by the end user by hand to deal with his
                     53:   own test suites.  Could be placed into `libexec` if we do not want to
                     54:   make this public.
                     56: ## For NetBSD 8.0
                     58: * Removal of the backwards compatibility `atf-run` and `atf-report`
                     59:   scripts, as well as the supporting `atf2kyua` tool.
1.3       jmmv       61: # Why?
                     63: As mentioned in [[another page|/kyua]], Kyua should be seen as ATF 2.x even
                     64: though it carries a different name.  It is the evolution of the previous
                     65: ATF tools (*the tools only*), but written in a more modular and flexible
                     66: way, and with a more reliable codebase.  Therefore, you should consider
                     67: this project as the update of ATF to a newer version.
                     69: The `atf-run` and `atf-report` tools have effectively been in maintenance
                     70: mode for over a year already.  None of the desired features (see the list
                     71: of open PRs) have been implemented on top of them, mostly because doing so
                     72: is building upon a broken implementation.  Additionally, several developers
                     73: have had to implement their own test results dashboards due to
                     74: defficiencies in `atf-report`, effectively reinventing the wheel.
                     76: This update will permit the real removal of the obsolete tools, thus
                     77: allowing us to build additional features on top of Kyua without having to
                     78: worry about being compatible with `atf-run` (and thus adjusting this tool
                     79: to behave in the same manner).
                     81: Some possible answers to "Why not?" can be found later in this page.
                     83: ## And why now?
1.1       jmmv       84: 
                     85: Kyua has existed for almost 3 years already, so you may be wondering why
                     86: this import is being proposed now.  The major reasons are:
                     88: * Since 0.5, Kyua has now feature-parity with ATF.  Replacing the ATF tools
                     89:   with Kyua should not introduce functionality regressions.
                     91: * NetBSD-6 has been branched relatively recently, so we can expect NetBSD-7
                     92:   to be far away enough in the future to provide plenty of time to
                     93:   stabilize Kyua in NetBSD-current.
                     95: * The existence of the deprecated `atf-run` and `atf-report` tools hinders
                     96:   the development of new features that require changes to the inteface of
                     97:   ATF test programs.  As an example: a highly requested feature is the
                     98:   ability to change the timeout settings of the test programs (for the
                     99:   benefit of old, slow platforms); doing this in the ATF codebase is
                    100:   tricky.
                    102: * Some developers continue to add old-style tests to the tree.  While this
                    103:   is suboptimal and against the policy dictated by core@, Kyua brings in a
                    104:   mechanism to allow running these tests unmodified (i.e. without adding
                    105:   any ATF calls in their code).  This, in turn, will let such developers
                    106:   add their tests more easily, and thus increase the tests coverage.
                    108: # The plan
                    110: This section details what the transition plan looks like and an estimate
                    111: timeline.  As all things that depend on free time (opensource software
                    112: hacking, in this case), take the estimates with a grain of salt.
                    114: ## Discuss this plan in tech-userlevel@
                    116: If you are here it's possibly because a review request for this plan has
1.4       jmmv      117: already been published and thus the plan has already begun.
1.1       jmmv      118: 
                    119: This plan will be sent to the tech-userlevel@ mailing list asking for
                    120: comments.  **Two weeks shall be allowed for initial discussion.** Depending
                    121: on the outcome, the plan and/or the software will need to be adjusted
                    122: (which in turn may require significant amounts of time not yet accounted
                    123: for).  I'll be optimistic for now.
                    125: ## Get core@ approval
                    127: By policy, the import of any new software component into src requires core@
                    128: approval.  Even though this will have been discussed at length in
                    129: tech-userlevel@ (as per the previous step), the final decider on this issue
                    130: will be core@.  The corollary of this is that, if no consensus can be
                    131: reached in tech-userlevel@ regarding this plan, core@ will be asked to come
                    132: up with a decision.
                    134: If core@ approves the plan, the next steps shall start immediately.  If
                    135: core@ disagrees, core@ will be asked to provide advice on the corrections
                    136: that should be made before the plan can be approved.
1.4       jmmv      138: It is hard to tell how long this step will last, but possibly account for 2
                    139: to 4 weeks.
1.5     ! jmmv      141: ## Address feedback as a new release
        !           142: 
        !           143: Publish a new Kyua release that collects all the feedback from the reviews
        !           144: above.
        !           145: 
        !           146: The list of issues to be addressed can be found by querying the
        !           147: [bug tracker]( for the
        !           148: [Milestone=Release0.6](
        !           149: keyword.  In particular, the following are the issues that have arisen from
        !           150: the review:
        !           151: 
        !           152: * [Issue 40]( Provide
        !           153:   manpages instead of an info document.
        !           154: 
1.1       jmmv      155: ## Import Kyua into src
1.2       jmmv      157: As the [[introductory page to Kyua|/kyua]] describes, Kyua has been
                    158: available in pkgsrc for a while and can be readily installed and used to
                    159: run the tests from `/usr/tests`.
1.1       jmmv      160: 
                    161: However, because ATF lives in src, and because NetBSD aims to provide the
                    162: best environment for testing "out of the box", Kyua should be imported into
                    163: src just like ATF was.  The major reasons for this, as have been explained
                    164: in the past, are: first, to allow any new deployment of NetBSD to be
                    165: validated right after installation and continuously afterwards; and,
                    166: second, to permit the execution of tests during development without having
                    167: to install any additional software.
                    169: The specific steps to perform this import are:
                    171: 1. Import Lutok into `external/bsd/lutok/`.  This is a shared library that
                    172:    wraps the native Lua C interface in C++.  Lutok was originally part of
                    173:    Kyua, and was split into its own package per the request of some users
                    174:    that found this component useful on its own.
                    176: 1. Import Kyua into `external/bsd/kyua-cli/`.  This yields a new kyua
                    177:    binary in `/usr/bin`, a lot of test programs in `/usr/tests/kyua-cli`
                    178:    (around 100) and some auxiliary files in `/usr/share/kyua`.
1.5     ! jmmv      180: 1. Protect all products of Lutok and Kyua with a new `MKKYUA` knob.  Once
        !           181:    the ATF tools are removed, the existence of both the `MKATF` and
        !           182:    `MKKYUA` knows will probably be confusing.  When that happens, we can
        !           183:    revisit this decision by possibly replacing both with an `MKTESTS`.
1.1       jmmv      184: 
                    185: 1. Update `` to generate `Kyuafile`s *in addition to*
                    186:    `Atffile`s.
                    188: There is no real need to do this import in a branch given that this import
                    189: only adds new functionality without touching existing stuff.
                    191: All the preparatory work for the import can be done offline (in about two
                    192: weeks at most, given that I have mot of this ready).  Aside from the code
                    193: changes, this will involve the validation of NetBSD/amd64, NetBSD/i386 and
                    194: NetBSD/macppc builds (which are the ports I have access to).  If you
                    195: consider that some other tricky architecture should be build-tested
                    196: (sparc64?), let me know and I'll include it in the list.
                    198: The submission step to CVS, once all the code changes are ready locally,
                    199: and any post-commit validation checks will take a few hours.  Any build
                    200: breakage should be addressed in a timely manner (and can possibly be
1.5     ! jmmv      201: worked-around by setting `MKKYUA=no`).
1.1       jmmv      202: 
                    203: ## Update documentation
1.2       jmmv      205: The [[Kyua: An introduction for NetBSD users|/kyua]] wiki page will be
1.1       jmmv      206: updated to explain how Kyua is bundled in NetBSD and how to use the bundled
                    207: version.
1.2       jmmv      209: The [[Creating atf-based tests for NetBSD src|/tutorials/atf]] wiki page
                    210: will be updated to account for the differences in test programs execution
                    211: with Kyua instead of ATF.
1.1       jmmv      212: 
                    213: The afterboot(8) and tests(7) manpages will be adjusted to mention `kyua`
                    214: instead of the ATF tools.
                    216: ## Adjust continuous testing systems to use Kyua
                    218: With `kyua` being part of the release sets, it is possible to adjust the
                    219: continuous test systems to make use of this tool in the test environments
                    220: without having to take any additional step.
                    222: I'll work with gson@ and pgoyette@ to adapt their continuous testing
                    223: machines to use the new built-in `kyua` binary instead of `atf-run` and
                    224: `atf-report`.  I'm planning to do the necessary work to change `anita`
                    225: myself, and I expect to help them deploy the changes to their own systems.
                    226: Because Kyua and ATF will cohexist in the base system at this point,
                    227: migrating the continuous testing systems to Kyua can happen at its own
                    228: peace.
                    230: It might happen that Kyua misses some little detail needed by these
                    231: systems.  In that case, this may require a new release of Kyua and a
                    232: reimport into the tree.  Incremental imports with new features are much
                    233: easier than the original import described in here.  Also, it will be
                    234: possible to cherry-pick any external changes into the tree without a
                    235: reimport (as has often been done in ATF).
1.4       jmmv      237: This step can take a few weeks of time, mostly due to the back and forth
                    238: between different people in different timezones.
1.1       jmmv      240: ## User validation period
                    242: At this stage, **at least one month shall be given to the community** to
                    243: test the new tools and the new test results dashboards.  Collect feedback
1.5     ! jmmv      244: and address requests as appropriate (possibly by releasing and importing a
        !           245: new version of Kyua).
        !           246: 
        !           247: One important thing to validate is that the results reported by `atf-run`
        !           248: and `kyua test` match with each other.  I have already been validating this
        !           249: with every public release of Kyua, but it has been a manual process.  To
        !           250: make this more reliable, I will set up a continuous testing machine of my
        !           251: own in which I will execute `atf-run` and `kyua test` in sequence (possibly
        !           252: within anita) and will add an automatic comparison of the exit status of
        !           253: each other.
1.1       jmmv      254: 
1.4       jmmv      255: *Because the old `atf-run` and `atf-report` tools have not yet been dropped
                    256: at this point, we can spend as much time as necessary on this phase to get
                    257: things right.*
1.1       jmmv      259: ## Replace atf-run and atf-report with kyua-atf-compat
                    261: Import the Kyua-based `atf-run` and `atf-report` compatibility tools and
                    262: stop building the deprecated versions of these.  The compatibility versions
                    263: are shipped in a separate `kyua-atf-compat` package, and thus this will be
                    264: imported into `external/bsd/kyua-atf-compat/`.
                    266: Once this is done, change `` to not generate `Atffile`s any
                    267: more, as the compatibility tools do not need them.
                    269: ## Drop atf-run and atf-report's code
                    271: Delete the native `atf-run` and `atf-report` utilities from the source tree
                    272: (and from the upstream repository).  This also gets rid of a lot of
                    273: supporting helper code, which makes the various ATF libraries leaner (and
                    274: therefore benefits all test programs!).
                    276: ## Get rid of some ATF wrappers as a proof of concept
                    278: There are some test programs (e.g. the ones in the ipf test suite) that are
                    279: not ATF-based.  To plug them into the NetBSD test suite, we have
                    280: implemented ATF-based wrappers that invoke the original plain tests.  This
                    281: is certainly suboptimal, as this level of indirection hinders readability
                    282: and makes development harder.  (In particular, it prevents some key
                    283: developers from contributing tests in the first place.)
                    285: Because Kyua has the ability to run these "plain" (non-ATF) test programs
                    286: directly, and because a bunch of developers are really looking forward to
                    287: this feature, I will convert a few tests to not include ATF wrappers as a
                    288: proof of concept of this feature.  The ipf tests are probably a good choice
                    289: for this.
                    291: ## Remove atf-run and atf-report compatibility tools
                    293: This is still really far away in the future (needs to happen after NetBSD
                    294: 7.0 is branched), but I'm listing it for completeness of the plan.  The
                    295: idea is to get rid of any ATF compatibility scripts by NetBSD 8.0.  This
                    296: gives end users a full release cycle (that of 7) to adapt to Kyua while at
                    297: the same time being able to use the old-style tools.
                    299: # Possible concerns
                    301: This section attempts to collect the list of possible concerns and/or
                    302: objections that may come up during the review, together with an attempt of
                    303: rebuttal from my side.
                    305: ## Increased compilation time
                    307: The addition of any new component to src increases the build time of the
                    308: whole system.  In the case of Kyua, this increase might be noticeable
                    309: *because of the large amount of test programs provided (roughly 100)*, all
                    310: of which are in C++.
                    312: Note that ATF also had a relatively large codebase and a bunch of tests
                    313: (although not as many).  The "trick" was that the majority of these tests
                    314: were written in shell, and as such they did not increase the build time by
                    315: much.  However, they significantly increased the run time of the test
                    316: suite, and they were less detailed (mostly integration tests, few unit
                    317: tests) than the Kyua tests.
                    319: In order to mitigate this issue, the build of all pieces of Kyua will be
1.5     ! jmmv      320: protected by `MKKYUA` so that people allergic to C++ can avoid the whole
1.1       jmmv      321: thing.  Even more, there are some other additional provisions described
                    322: below.
                    324: ## Introduction of more C++ code in base
                    326: First of all, why did I use C++?  To make the implementation simpler and
                    327: safer (the RAII programming pattern is really useful in avoiding memory and
                    328: resource leaks with minimal effort).  And C++ is part of the base system
                    329: and a supported language, so there was no reason not to do so.  But that's
1.4       jmmv      330: not the point of this item: if you dislike like C++, this is not going to
                    331: make you think I did right.
1.1       jmmv      332: 
                    333: It's true that, if we count the number of lines, Kyua brings in more C++
                    334: code than what will eventually be dropped by the removal of the ATF tools.
                    335: However, because ATF was also C++, the import of Kyua itself does not make
                    336: the situation significantly worse.
                    338: Additionally, the two compilers we can really use (GCC and LLVM) already
                    339: use, or will soon use, C++ in their code base.  It is unlikely that we will
                    340: be able to remove all C++ support from base anytime soon due to this, while
                    341: at the same time keeping support for all the ports that NetBSD has.
1.4       jmmv      343: Long term, if the use of C++ proves to be a problem, there are a couple of
                    344: major things that can be done to slowly get rid of C++.  These ideas have
                    345: merits of their own (not only remove C++).  The first is to split the
                    346: execution engine of a single test case into a separate binary and, because
                    347: this would be performance-critical, write it in plain C.  The second is the
                    348: rewrite of most user-interface code in Lua, which in itself would bring
                    349: some extensibility advantages to the program; I haven't pursued this yet
                    350: because my knowledge of Lua is very limited.  Both options will be
                    351: investigated separtely when the time permits.  In the meantime, the
                    352: replacement of ATF with Kyua does not make things worse; it just changes
                    353: one chunk of code with another.
1.1       jmmv      354: 
                    355: ## No need for Lutok as a public library
                    357: Kyua depends on Lutok, which is a C++ interface to Lua.  The code of this
                    358: library was originally part of Kyua, but I split it into its own project
                    359: because some users asked for it.
                    361: If there is no desire to ship Lutok as a shared library in `/usr/lib`, we
                    362: can build Lutok as a static private library and link Kyua against it.
                    363: There is no need to install this as a shared library.

CVSweb for NetBSD wikisrc <> software: FreeBSD-CVSweb