Commit graph

16355 commits

Author SHA1 Message Date
piegames 3ba5ef91bc benchmarks: Fix purity
Some stuff wasn't working on my machine because it depended on the host
environment in some ways, fixed those.

Change-Id: Iff4931a9a26c6827978f1ee6434710f406d67a96
2024-10-18 19:37:23 +02:00
piegames e5de1d13c4 libexpr: Optimize complex indented strings
The old behavior results in lots of concatenations happening for no good
reason and is an artifact of the technical limitations of the old parser
(combined with some lack of care for such details).

Change-Id: I0d78d6220ca6aeaa10bc437e48e08bf7922e0bb3
2024-10-18 19:37:23 +02:00
piegames 878e181882 libexpr: Print interpolations more accurately in show
This is only a minor semantical distinction, but we should be able to
properly test it, and the parser tests rely on show for that.

Change-Id: I25e868cf9544e30cdff17deb5fd50a434e0f367e
2024-10-18 11:40:04 +00:00
piegames c852ae60da libexpr: Rewrite stripIndentation for indented strings
This commit should faithfully reproduce the old behavior down to the
bugs. The new code is a lot more readable, all quirks are well
documented, and it is overall much more maintainable.

Change-Id: I629585918e4f2b7d296b6b8330235cdc90b7bade
2024-10-18 11:40:04 +00:00
piegames 765771a355 tests/functional-lang: Add test cases for indented strings
Accidentally came across them elsewhere, wanted them covered in the
ind-strings test as well.

Change-Id: Iba418a687388ba85516d13d4c9d815744297dc5c
2024-10-18 11:40:04 +00:00
piegames 14291856e4 tests/functional-lang: Test indented string parsing
There is already an eval test for it, but it doesn't test the parser's
output

Change-Id: I94809080f90ace8aa3f610bcd315e9df0cf5d12f
2024-10-18 11:40:04 +00:00
piegames e7d6212f77 libexpr: move parser semantics into separate file
Another preparation for forking off and versioning the parser

Change-Id: I7b1225a44a3b81486414c1d37bd3e76a3ab307f9
2024-10-18 11:40:04 +00:00
piegames f98ee07573 libexpr: rename grammar to grammar::v1
Let's make some space in the namespace for a v2

https://wiki.lix.systems/books/lix-contributors/page/nix-lang-v2
Change-Id: If56e6dbf680d931233aa822ef91c8832464471e4
2024-10-18 11:40:04 +00:00
piegames 580df9bdb3 tests/functional-lang: Small cleanup of flagfiles handling
Flag files now properly work for all tests.

Change-Id: If652235960e8cf4120270a1b974c17586a31a157
2024-10-18 11:40:04 +00:00
jade 52b64662a4 Merge "fix: macOS build broken by fatal lowdown CLI sandbox setup" into main 2024-10-17 21:17:49 +00:00
jade f6077314fa Merge "Fix std::terminate call in thread pool" into main 2024-10-15 00:11:59 +00:00
jade a020f5f6cb fix: macOS build broken by fatal lowdown CLI sandbox setup
This failed due to https://github.com/NixOS/nixpkgs/pull/346945, which
makes a second lowdown-unsandboxed that works in nix builds, and the
regular lowdown has executables that fail closed when the sandbox setup
fails.

The actual failure here is only visible on nixos-unstable at the moment,
not 24.05, but this commit should fix it up for all versions.

Fixes: #547
Change-Id: I50c0ecb59518ef01a7c0181114c1b4c5a7c6b78b
2024-10-14 16:49:16 -07:00
V. fbf7a8b440 Merge "Remove ancient let from 2 test files" into main 2024-10-14 22:06:48 +00:00
V. 31ff77b3f9 Remove ancient let from 2 test files
Change-Id: I992bc7f9e1cfcb1e4038fbe6ee04178bbf938556
2024-10-14 17:19:22 +04:00
Maximilian Bosch 4682e40183 ssh-ng: better way to keep SSH errors visible
A better fix than in 104448e75d, hence a
revert + the fix.

It turns out that this commit has the side-effect that when having e.g.
`StrictHostKeyChecking=accept-new` for a remote builder, the warnings à la

    Warning: Permanently added 'builder' (ED25519) to the list of known hosts.

actually end up in the derivation's log whereas hostkey verification
errors don't, but only in the stderr of the `nix-build` invocation
(which was the motivation for the patch).

This change writes the stderr from the build-hook to

* the daemon's stderr, so that the SSH errors appear in the journal
  (which was the case before 104448e75d)
* the client's stderr, as a log message
* NOT to the drv log (this is handled via `handleJSONLogMessage`)

I tried to fix the issue for legacy-ssh as well, but failed and
ultimately decided to not bother.

I know that we'll sooner or later replace the entire component, however
this is the part of the patch I have working for a while, so I figured I
might still submit it for the time being.

Change-Id: I21ca1aa0d8ae281d2eacddf26e0aa825272707e5
2024-10-14 06:01:18 +00:00
jade 326cbecb61 Merge changes I327db40f,If762efce into main
* changes:
  testsuite: use xdist for parallel test running
  testsuite: add a functional2 test suite based on pytest
2024-10-14 05:53:44 +00:00
Maximilian Bosch a322fcea4a
worker: respect C-c on sudo nix-build
While debugging something else I observed that latest `main` ignores
`Control-C` on `sudo nix-build`.

After reading through the capnproto docs, it seems as if the promise
must be fulfilled to actually terminate the `promise.wait()` below.

This also applies to scenarios such as stopping the client
(`nix-build`), but the builders on the daemon-side are still running,
i.e. closes #540

Co-authored-by: eldritch horrors <pennae@lix.systems>

Change-Id: I9634d14df4909fc1b65d05654aad0309bcca8a0a
2024-10-12 21:16:30 +02:00
jade a0fb52c0af Fix std::terminate call in thread pool
So we received a report that the thread pool crashed due to an
Interrupted exception.

Relevant log tail:

copying path '/nix/store/0kal2k73inviikxv9f1ciaj39lkl9a87-etc-os-release' to 'ssh://192.168.0.27'...
Lix crashed. This is a bug. We would appreciate if you report it along with what caused it at https://git.lix.systems/lix-project/lix/issues with the following information included:

error (ignored): error: interrupted by the user
Exception: nix::Interrupted: error: interrupted by the user

Relevant stack trace:

 4# __cxa_rethrow in /nix/store/22nxhmsfcv2q2rpkmfvzwg2w5z1l231z-gcc-13.3.0-lib/lib/libstdc++.so.6
 5# nix::ignoreExceptionExceptInterrupt(nix::Verbosity) in /nix/store/ghxr2ykqc3rrfcy8rzdys0rzx9ah5fqj-lix-2.92.0-dev-pre20241005-ed9b7f4/lib/liblixutil.so
 6# nix::ThreadPool::doWork(bool) in /nix/store/ghxr2ykqc3rrfcy8rzdys0rzx9ah5fqj-lix-2.92.0-dev-pre20241005-ed9b7f4/lib/liblixutil.so
 7# 0x00007FA7A00E86D3 in /nix/store/22nxhmsfcv2q2rpkmfvzwg2w5z1l231z-gcc-13.3.0-lib/lib/libstdc++.so.6
 8# 0x00007FA79FE99A42 in /nix/store/3dyw8dzj9ab4m8hv5dpyx7zii8d0w6fi-glibc-2.39-52/lib/libc.so.6
 9# 0x00007FA79FF1905C in /nix/store/3dyw8dzj9ab4m8hv5dpyx7zii8d0w6fi-glibc-2.39-52/lib/libc.so.6

Notably, this is *not* in the main thread, so this implies that the
thread didn't get joined properly before their destructors got called.
That, in turn, should have only possibly happened because join() threw
on a previous iteration of the loop joining threads, I think. Or if it
threw while in the ThreadPool destructor. Either way we had better stop
letting Interrupted fall out of our child threads!

If:
- Interrupted was thrown inside the action in the main thread: it would
  have fallen out of doWork if state->exception was already set and got
  caught by ThreadPool::process, calling shutdown() and the join loop
  which would crash the process entirely.
- Interrupted was thrown inside the action on a secondary thread: it
  would have been caught and put into the exception field and then
  possibly rethrown to fall out of the thread (since it was previously
  ignoreExceptionExceptInterrupt).

The one possible hole in this hypothesis is that there is an "error
(ignored)" line in there implying that at least one Interrupted got
eaten by an ignoreExceptionInDestructor. It's also unclear whether this
got reordered because of stderr buffering.

Fixes: #542
Change-Id: I322cf050da660af78f5cb0e08ec6e6d27d09ac76
2024-10-09 15:38:40 -07:00
jade 4180b84a67 testsuite: use xdist for parallel test running
This is capped at 12 because 3.7 seconds of startup is painful enough
and 5.5 seconds with 24 was more annoying.

Change-Id: I327db40fd98deaa5330cd9cf6de99fb07b2c1cb0
2024-10-09 14:47:39 -07:00
jade 3caf3e1e08 testsuite: add a functional2 test suite based on pytest
I am tired of bad shell scripts, let me write bad python quickly
instead. It's definitely, $100%, better.

This is not planned as an immediate replacement of the old test suite,
but we::jade would not oppose tests getting ported.

What is here is a mere starting point and there is a lot more
functionality that we need.

Fixes: #488

Change-Id: If762efce69030bb667491b263b874c36024bf7b6
2024-10-09 14:47:39 -07:00
jade 9865ebaaa6 Merge "Remove static initializers for RegisterLegacyCommand" into main 2024-10-09 20:37:58 +00:00
jade 7f7a38f278 Merge changes Ib27cb43d,I03687b8b into main
* changes:
  testsuite: override NIX_CONF_DIR and NIX_USER_CONF_FILES
  Remove some outdated `make test` invocation suggestions
2024-10-09 20:37:16 +00:00
Rebecca Turner 0012887310 Merge "Add release note for CTRL-C improvements" into main 2024-10-08 22:15:56 +00:00
Lulu 4ea8c9d643 Set c++ version to c++23
I followed @pennae's advice and moved the constructor definition of
`AttrName` from the header file `nixexpr.hh` to `nixexpr.cc`.

Change-Id: I733f56c25635b366b11ba332ccec38dd7444e793
2024-10-08 20:05:28 +02:00
Lulu 43e79f4434 Fix gcc warning -Wmissing-field-initializers
The approach that was taken here was to add default values to the type
definitions rather than specify them whenever they are missing.

Now the only remaining warning is '-Wunused-parameter' which @jade said
is usually counterproductive and that we can just disable it:
#456 (comment)

So this change adds the flags '-Wall', '-Wextra' and
'-Wno-unused-parameter', so that all warnings are enabled except for
'-Wunused-parameter'.

Change-Id: Ic223a964d67ab429e8da804c0721ba5e25d53012
2024-10-08 01:44:38 +00:00
Lulu 299813f324 Merge "Avoid calling memcpy when len == 0 in filetransfer.cc" into main 2024-10-08 01:41:41 +00:00
Lulu d6e1b11d3e Fix gcc warning -Wsign-compare
Add the compile flag '-Wsign-compare' and adapt the code to fix all
cases of this warning.

Change-Id: I26b08fa5a03e4ac294daf697d32cf9140d84350d
2024-10-08 01:32:12 +02:00
Lulu 51a5025913 Avoid calling memcpy when len == 0 in filetransfer.cc
There was a bug report about a potential call to `memcpy` with a null
pointer which is not reproducible:
#492

This occurred in `src/libstore/filetransfer.cc` in `InnerSource::read`.

To ensure that this doesn't happen, an early return is added before
calling `memcpy` if the length of the data to be copied is 0.

This change also adds a test that ensures that when `InnerSource::read`
is called with an empty file, it throws an `EndOfFile` exception.

Change-Id: Ia18149bee9a3488576c864f28475a3a0c9eadfbb
2024-10-08 01:26:30 +02:00
eldritch horrors ed9b7f4f84 libstore: remove Worker::{childStarted, goalFinished}
these two functions are now nearly trivial and much better inline into
makeGoalCommon. keeping them separate also separates information about
goal completion flows and how failure information ends up in `Worker`.

Change-Id: I6af86996e4a2346583371186595e3013c88fb082
2024-10-05 21:19:51 +00:00
eldritch horrors 649d8cd08f libstore: remove Worker::removeGoal
we can use our newfound powers of Goal::work Is A Real Promise to remove
completed goals from continuation promises. apart from being much easier
to follow it's also a lot more efficient because we have the iterator to
the item we are trying to remove, skipping a linear search of the cache.

Change-Id: Ie0190d051c5f4b81304d98db478348b20c209df5
2024-10-05 21:19:51 +00:00
eldritch horrors 9adf6f4568 libstore: remove Goal::notify
Goal::work() is a fully usable promise that does not rely on the worker
to report completion conditions. as such we no longer need the `notify`
field that enabled this interplay. we do have to clear goal caches when
destroying the worker though, otherwise goal promises may (incorrectly)
keep goals alive due to strong shared pointers created by childStarted.

Change-Id: Ie607209aafec064dbdf3464fe207d70ba9ee158a
2024-10-05 21:19:51 +00:00
eldritch horrors 03cbc0ecb9 libstore: move Goal::ex to WorkResult
yet another duplicated field. it's the last one though.

Change-Id: I352df8d306794d262d8c9066f3be78acd40e82cf
2024-10-05 21:19:51 +00:00
eldritch horrors 1caf2afb1d libstore: move Goal::buildResult to WorkResult
derivation goals still hold a BuildResult member variable since parts of
these results of accumulated in different places, but the Goal class now
no longer has such a field. substitution goals don't need it at all, and
derivation goals should also be refactored to not drop their buildResult

Change-Id: Ic6d3d471cdbe790a6e09a43445e25bedec6ed446
2024-10-05 20:53:39 +00:00
eldritch horrors 7ff60b7445 libstore: move Goal::exitCode to WorkResult
the field is simply duplicated between the two, and now that we can
return WorkResults from Worker::run we no longer need both of them.

Change-Id: I82fc47d050b39b7bb7d1656445630d271f6c9830
2024-10-05 20:17:20 +00:00
eldritch horrors fc6291e46d libstore: return goal results from Worker::run()
this will be needed to move all interesting result fields out of Goal
proper and into WorkResult. once that is done we can treat goals as a
totally internal construct of the worker mechanism, which also allows
us to fully stop exposing unclear intermediate state to Worker users.

Change-Id: I98d7778a4b5b2590b7b070bdfc164a22a0ef7190
2024-10-05 20:12:13 +00:00
eldritch horrors 40f154c0ed libstore: remove Worker::topGoals
since we now propagate goal exceptions properly we no longer need to
check topGoals for a reason to abort early. any early abort reasons,
whether by exception or a clean top goal failure, can now be handled
by inspecting the goal result in the main loop. this greatly reduces
goal-to-goal interactions that do not happen at the main loop level.

since the underscore-free name is now available for use as variables
we'll migrate to that where we currently use `_topGoals` for locals.

Change-Id: I5727c5ea7799647c0a69ab76975b1a03a6558aa6
2024-10-05 19:53:30 +00:00
eldritch horrors f389a54079 libstore: propagate goal exceptions using promises
drop childException since it's no longer needed. also makes
waitForInput, childFinished, and childTerminated redundant.

Change-Id: I05d88ffd323c5b5c909ac21056162f69ffb0eb9f
2024-10-05 19:44:47 +00:00
eldritch horrors 7ef4466018 libstore: have goals promise WorkResults, not void
Change-Id: Idd218ec1572eda84dc47accc0dcd8a954d36f098
2024-10-05 19:06:59 +00:00
eldritch horrors a9f2aab226 libstore: extract Worker::goalFinished specifics
there's no reason to have the worker set information on goals that the
goals themselves return from their entry point. doing this in the goal
`work()` function is much cleaner, and a prerequisite to removing more
implicit strong shared references to goals that are currently running.

Change-Id: Ibb3e953ab8482a6a21ce2ed659d5023a991e7923
2024-10-05 19:06:59 +00:00
eldritch horrors 99edc2ae38 libstore: check for interrupts in parallel promise
this simplifies the worker loop, and lets us remove it entirely later.
note that ideally only one promise waiting for interrupts should exist
in the entire system. not one per event loop, one per *process*. extra
interrupt waiters make interrupt response nondeterministic and as such
aren't great for user experience. if anything wants to react to aborts
caused by explicit interruptions, or anything else, those things would
be much better served using RAII guards such as Finally (or KJ_DEFER).

Change-Id: I41d035ff40172d536e098153c7375b0972110d51
2024-10-05 19:06:59 +00:00
eldritch horrors 896a123605 libstore: remove Goal::StillAlive
this was a triumph. i'm making a note here: huge success. it's hard to
overstate my satisfaction! i'm not even angry. i'm being so sincere ri

actually, no. we *are* angry. this was one dumbass odyssey. nobody has
asked for this. but not doing it would have locked us into old, broken
protocols forever or (possibly worse) forced us to write our own async
framework building on the old did-you-mean-continuations in Worker. if
we had done that we'd be locked into ever more, and ever more complex,
manual state management all over the place. this just could not stand.

Change-Id: I43a6de1035febff59d2eff83be9ad52af4659871
2024-10-05 18:21:02 +00:00
Rebecca Turner 0d484aa498
Add release note for CTRL-C improvements
I'm very excited for cl/2016, so others will probably be excited also!
Let's add a release note.

Change-Id: Ic84a4444241aafce4cb6d5a6d1dddb47e7a7dd7b
2024-10-05 10:40:51 -07:00
Rebecca Turner 86b213e632 Merge "Split ignoreException to avoid suppressing CTRL-C" into main 2024-10-05 17:33:00 +00:00
eldritch horrors a3dd07535c fix build test error count checks
with async runtime scheduling we can no longer guarantee exact error
counts for builds that do not set keepGoing. the old behavior can be
recovered with a number of hacks that affect scheduling, but none of
those are very easy to follow now advisable. exact error counts will
like not be needed for almost all uses except tests, and *those* had
better check the actual messages rather than how many they got. more
messages can even help to avoid unnecessary rebuilds for most users.

Change-Id: I1c9aa7a401227dcaf2e19975b8cb83c5d4f85d64
2024-10-05 16:21:19 +00:00
alois31 5df2cccc49
doc: install the HTML manual again
In 0e6b3435a1, installation of the HTML manual
was accidentally dropped: setting install_dir on a custom_target only sets the
directory where something is going to be installed if it is installed at all,
but does not itself trigger installation. The latter has to be explicitly
requested, which is just what we do here to get the manual back.

Change-Id: Iff8b791de7e7cb4c8d747c2a9b1154b5fcc32fe0
2024-10-05 10:49:34 +02:00
jade 345e3d068a testsuite: override NIX_CONF_DIR and NIX_USER_CONF_FILES
The test suite can load the global configuration files under certain
circumstances, and, though we would really rather it didn't ever do that
at all, we should at least break the mechanism.

Fixes: #474
Change-Id: Ib27cb43dd5dfaa70ac491c395b5ba308fd7bd289
2024-10-04 19:17:08 -07:00
jade 19edaed81b Remove some outdated make test invocation suggestions
These should be meson.

Change-Id: I03687b8b03f50fb1684e7ffcd487be855052d6c2
2024-10-04 18:55:52 -07:00
eldritch horrors 5b1715e633 libstore: forbid addWantedGoals when finished
due to event loop scheduling behavior it's possible for a derivation
goal to fully finish (having seen all paths it was asked to create),
but to not notify the worker of this in time to prevent another goal
asking the recently-finished goal for more outputs. if this happened
the finished goal would ignore the request for more outputs since it
considered itself fully done, and the delayed result reporting would
cause the requesting goal to assume its request had been honored. if
the requested goal had finished *properly* the worker would recreate
it instead of asking for more outputs, and this would succeed. it is
thus safe to always recreate goals once they are done, so we now do.

Change-Id: Ifedd69ca153372c623abe9a9b49cd1523588814f
2024-10-04 17:49:57 +00:00
Rebecca Turner 0b29859cfe Merge "editorconfig: Add meson.build" into main 2024-10-04 16:36:20 +00:00
Olivia Crain 1bfc37fea5 Merge "internal-api-docs: allow Doxygen to build regardless of workdir" into main 2024-10-04 09:59:01 +00:00