Additionally, add IsValidBase16() to restore the behavior of rejecting invalid base16, which absl's HexStringToBytes does not do.
Change-Id: I777a36f5dc787aa54a2aa316d6728f68da129768
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1484
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
LineFilter doesn't actually exist in this version of clang-tidy. It was only working because the config was ignored.
Change-Id: Ice5ddb5d1031dfc2cc4fee24674464f965323d8b
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1431
Reviewed-by: glittershark <grfn@gws.fyi>
Tested-by: BuildkiteCI
This eases debugging of live crashes.
Change-Id: Ie15a7f8fb3f091cae0fbe012e58862d416a42891
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1433
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
Add a stub class for wrapping a gRPC client to the new, proto-backed nix
store protocol, along with several methods implemented but several left
throwing a not implemented exception.
Paired-With: Vincent Ambo <mail@tazj.in>
Paired-With: Perry Lorier <isomer@tvl.fyi>
Change-Id: Id943d4f6d75084b8498786d580e6c9f7c92c104d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1436
Tested-by: BuildkiteCI
Reviewed-by: kanepyork <rikingcoding@gmail.com>
Implement the main function for the new, proto-based nix daemon. This
mostly replicates the behavior of the previous nix daemon - it starts a
grpc server listening on the same nix socket path and supports the same
set of options - with the exception of --stdio, which has been renamed
to --pipe and documented in the man page.
Change-Id: Ib729283c1d5d35c0e1b0a968bc1f052f5527f2d7
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1356
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Reviewed-by: kanepyork <rikingcoding@gmail.com>
Template the systemd unit file templates as part of the nix
installPhase, putting them in the same place that the upstream nix
derivation does.
Change-Id: I3ceabfc0c837564e33b9ae7f9eeb7185d6fbe907
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1429
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Reviewed-by: tazjin <mail@tazj.in>
This hardcodes eelco's home directory, among other things we no longer
care about.
Change-Id: Ide894107c091c37e30bba9daf473fc7dfc8b8563
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1427
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Including depotPath in the shellHook of the main derivation for tvix
was, unsurprisingly, causing spurious rebuilds. It's still useful when
developing locally, though, so I've extracted it to a `build-shell`
derivation as a passthru attribute and updated the documentation.
Fixes: #20
Change-Id: Ibc686b9f06ec68e79759ca2c989414bd5fbce696
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1426
Reviewed-by: tazjin <mail@tazj.in>
Tested-by: BuildkiteCI
Add an expression, based on the nixos qemu virtualisation framework, for
a basic system whose nix is tvix, to be used for testing tvix
Change-Id: I3c7422bb10d3ce05a3094671cb770f1f745d814c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1423
Reviewed-by: tazjin <mail@tazj.in>
Reviewed-by: Alyssa Ross <hi@alyssa.is>
Tested-by: BuildkiteCI
The full clang-tidy check suite is *very* slow. This reduced list is
some kind of middle-ground between running all checks, and having a
useful developer experience.
Crucially most of the static analyzer checks (except for the ones
related to security issues) have been disabled.
We should look into running the full suite in CI only.
Change-Id: I02b96ad3b4d1a43bd6aa90ffdcba800dad966714
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1422
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
Replace the custom, rather questionable base64 implementation with
absl::Base64{Une,E}scape. To make sure that the custom implementation
was doing the same thing I've also added a test covering
nix::Hash::to_string, which was one function that used it - the test
passed prior to the replacement, and continued to pass afterwards.
The previous base64Decode function threw an exception on failure - to
avoid going too far down the rabbit hole I've replicated that
functionality at all call sites, but this should be replaced with more
sensible error handling such as StatusOr eventually.
Also, before this change:
❯ nix eval -f . users.tazjin.emacs.outPath
"/nix/store/g6ri2q8nra96ix20bcsc734r1yyaylb1-tazjins-emacs"
And after:
❯ ./result/bin/nix eval -f . users.tazjin.emacs.outPath
"/nix/store/g6ri2q8nra96ix20bcsc734r1yyaylb1-tazjins-emacs"
Change-Id: Id292ffbb82fe808f3f1b34670afbe7b8c13ad615
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1385
Reviewed-by: tazjin <mail@tazj.in>
Tested-by: BuildkiteCI
CMake likes to remember that we overrode it with an empty string and not use the new env var.
Change-Id: I16587f27750c9ebd4f65349bb59b37e0f8117f18
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1406
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
The use of vfork() in Nix is entirely illegal. Quote:
If the process created by vfork() returns from the function in which vfork() was
called, or calls any other function before successfully calling _exit() or
one of the exec*() family of functions, the behavior is undefined.
-- Linux man-pages, release 5.05
Add a TODO to use the higher-performance variants of clone() on Linux when it
is available.
Change-Id: I42370e1568ad6e2d00d70d0b66c8aded8f1288bb
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1418
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Reviewed-by: Alyssa Ross <hi@alyssa.is>
nixutil depends on bzip2, lzma, boost::context, brotli{enc,dec}, and
openssl, but wasn't directly linking to them. This was causing linker
errors in a test that only depended on nixutil.
Change-Id: I60e77ea7b18b08e2946fcf9176ae0f355cd71844
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1384
Tested-by: BuildkiteCI
Reviewed-by: kanepyork <rikingcoding@gmail.com>
We need to -isystem the libcxx header files in order for clang-tidy to ignore them, as the Nix clang toolchain isn't doing that automatically.
Change-Id: I05b9e9bd522de4c0e2ad543214f6bf6ab66a306b
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1359
Reviewed-by: glittershark <grfn@gws.fyi>
Tested-by: BuildkiteCI
This formerly controlled access to builtins.exec and
builtins.importNative, but both of those are gone now, so there's no
need for this option any more.
Change-Id: I6850cbd6be264fbfb1b209a60026dadbd0ba1232
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1341
Reviewed-by: glittershark <grfn@gws.fyi>
Tested-by: BuildkiteCI
This is the shared object equivalent of builtins.exec, or a plugins
equivalent accessible from the Nix language. Either way, since we
don't have builtins.exec or plugins any more, I think it makes sense
to remove this builtin.
This will also allow us to drop the
allow-unsafe-native-code-during-evaluation option, which formerly
controlled whether builtins.exec and builtins.importNative were
enabled.
Cc: Griffin Smith <grfn@gws.fyi>
Cc: Profpatsch <mail@profpatsch.de>
Change-Id: I8993a8a79d559c102647308a2684c089bbc06713
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1340
Reviewed-by: glittershark <grfn@gws.fyi>
Reviewed-by: Profpatsch <mail@profpatsch.de>
Tested-by: BuildkiteCI
The SQL schemas are included as string constants which are
concatenated into a header file. In the previous Makefiles, this was
done with envsubst or something - we moved it to CMake.
There was a missing quote around the string to be interpolated, which
meant that CMake interpreted the semicolons as part of its language
syntax and did not emit them.
Change-Id: Ibb4512788b26b53f297db3535094dc0194614446
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1342
Reviewed-by: kanepyork <rikingcoding@gmail.com>
Reviewed-by: glittershark <grfn@gws.fyi>
Tested-by: BuildkiteCI
Add a forward-declared factory function for constructing and returning a
WorkerServiceImpl, for eventual use in the main function for the nix
daemon
Change-Id: I9032d69b6ee3bc3b1f39f3d5d55f951cffad8145
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1293
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Plugins seem to not really be used anywhere (I can find one plugin
that's actually defined, and it doesn't seem very useful, especially
since we got rid of builtins.exec) and their presence is adding
additional complexity and potential sources of bugs to an already
unsteady refactor. At some point we may want to bring back
something *like* plugins, but their design will likely be different and
it will definitely be after we have a functioning Nix again.
Change-Id: I3bc40e55917f70bf260fbc208c1705e2e6a7c626
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1291
Tested-by: BuildkiteCI
Reviewed-by: Alyssa Ross <hi@alyssa.is>
Reviewed-by: isomer <isomer@tvl.fyi>
This compiles under `-Wall -Werror`.
The largest chunk of this change is `final` qualifiers for the various
Nix CLI command structs, which inherit from a Command class that has
more virtual functions than are implemented by each command.
Change-Id: I0925e6e1a39013f026773db5816e4a77d50f3b4a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1294
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Reviewed-by: kanepyork <rikingcoding@gmail.com>
This setting was renamed to api.pure about 300 years ago and caused
warnings to be issued.
Change-Id: If883b0667c3afe67ae3d2a9950a796688cfbea7a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1290
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
Removes the verbosity enum, which is no longer actively used anywhere
other than a daemon protocol implementation bit that doesn't actually
work.
Since the verbosity was marked deprecated, this removes one of the
last remaining warnings.
Change-Id: Iaee9d1d6c14b30daac83bb44bcacff32a0e07fb0
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1289
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
The function is renamed to `SortedByKeys`, which is more descriptive,
and annotated with a comment about what it is used for.
The deprecation warning has been removed because this function is
currently functionally required.
Change-Id: I0ee3a76deff05f366feca9ddac8f38ab34bffbd0
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1288
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
This is the easiest way to get the checks up and running for now, but
we will probably want to separate out things like this into a separate
build step in the future.
Change-Id: I8e1a1095aef09b1eee97abad5b6240bc64d14b8c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1287
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
Enables loading of the expected output of evaluator tests from the
corresponding .exp files, and checks that the output matches.
This again leaves some tests behind in the disabled folder, but we now
have almost the entire suite up and running so I can get around to
cleaning up the disabled ones.
Other note: Some tests had XML output, despite not being related to
XML testing at all - I'm not sure why they chose to do this, but have
converted those test outputs to normal Nix instead.
We have a separate test suite for JSON & XML serialisation already,
which was contributed by andi-.
Change-Id: Id7c42c836edfec4c22db9d893e35489f3e6dd559
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1285
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Reviewed-by: glittershark <grfn@gws.fyi>
Enables the `eval-okay-` test suite, with some caveats:
* The output is not yet checked against the expected value, so the
tests only check that pure evaluation succeeds
* A handful of tests have been disabled as they are causing failures
that may be related to the DummyStore implementation.
Both of these will be addressed in followup commits, but there is
already some value in having the pure evaluation tests pass.
Change-Id: I62236c95ebffb06dc64a45455a8ee04344c400b7
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1284
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Reviewed-by: glittershark <grfn@gws.fyi>
Add a set of property tests for the attribute set (Bindings) class
checking that the Merge operation satisfies the monoid laws. This
will hopefully become useful to make sure we're not breaking the
language semantics as we work towards optimizing or replacing the
implementation, but also serves as a test bed for adding
rapidcheck-based property tests to the codebase.
Change-Id: I1b4b7b6503d08d80c1c5a8f9408fd4b787d00e8e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1283
Reviewed-by: isomer <isomer@tvl.fyi>
Tested-by: BuildkiteCI
Moves a function that is not dependent on the generated code over to
parser.hh. This function also looks like it could be improved, but
that is left as an exercise for the reader.
Code that remains in lexer.l has been reformatted, while we're here.
Change-Id: I9c26bb4eed0772a720d0715029e8bc10ab16ac38
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1279
Tested-by: BuildkiteCI
Reviewed-by: Kane York <rikingcoding@gmail.com>
This moves the language test suite into Googletest by constructing
parameterised tests out of the same language snippets used for the
previous lang.sh evaluation.
So far this includes support for about 3/4 of all tests, specifically:
* all parser success/failure tests
* all evaluator failure tests
The evaluator success tests will be implemented in a subsequent commit,
because the output comparison contains a whole bunch of additional
logic that I did not want to cram in here.
Change-Id: Icec9f368366cdbaa53b4c7e4472b8b6e8dd72eba
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1278
Reviewed-by: glittershark <grfn@gws.fyi>
Reviewed-by: isomer <isomer@tvl.fyi>
Tested-by: BuildkiteCI
Unfortunately, to guarantee correct behaviour of some evaluation (!)
tests, addToStore needs to actually check whether passed in source
files exist and fail appropriately.
There is a chance that the dependency on this behaviour is actually a
bug in the upstream test suite, but my attempts at finding out more
about this from, say, the git history have so far been unsuccessful.
Change-Id: I311999ea28fcedf5da13a4e627b1c1c8e4e59cbd
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1276
Reviewed-by: isomer <isomer@tvl.fyi>
Reviewed-by: glittershark <grfn@gws.fyi>
Tested-by: BuildkiteCI
These files will be integrated into the evaluator unit tests instead
of running separately via a shell script.
Change-Id: I1d229e73b1d862777f5108c86891689900edefbe
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1275
Tested-by: BuildkiteCI
Reviewed-by: Kane York <rikingcoding@gmail.com>
Reviewed-by: isomer <isomer@tvl.fyi>
These options only apply to nix-shell. The fact that nix-build
previously accepted them was an accident that resulted from the two
programs sharing an implementation.
Change-Id: I0047c98e2096010797316bff3ea4faf722fab86a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1273
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
I missed calling store_->assertStorePath on paths in several daemon
proto handlers, which the previous implementation did.
Change-Id: Ifad6eeb03b5a5babec7b4bcf7aca060813f15bb7
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1272
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
I got the message proto wrong on this one as well - it needs both a path
and a signatures.
Change-Id: I9a489b1285bda61c15b2a3b47d9cfc3b50e387da
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1270
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Implement the proto handler on the server side for
Worker::BuildDerivation. This includes several additions to the proto
which I had missed on the first pass, including the actual proto
definition for the Derivation itself and a few sequence number
reorderings which are fine because this is all provisional and not
deployed yet.
A couple things to note
- I implemented a couple constructors for nix classes that initialize
themselves based on their proto variants, which felt nice and didn't
end up causing any issues.
- I've made the conversions between the enum types in nix and in proto
explicit via switch statements rather than using a static_cast, out of
an abundance of caution that the error would get mismatched in the
future and we'd convert the wrong thing to the wrong thing - this is
verbose, but exceptionally future proof.
Change-Id: Iecf6b88e76bc37e49efa05fd65d6cd0cb0deffed
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1249
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Reviewed-by: Kane York <rikingcoding@gmail.com>
This silences a -Winconsistent-missing-override
Change-Id: I2492a8ab3c0b0a33a794f012bfee6bbd0c2f0a8e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1271
Reviewed-by: Kane York <rikingcoding@gmail.com>
Tested-by: BuildkiteCI
Interpolating a path into a string will copy the referenced path into
the Nix store, so this got a dependency on all of src. By first
constructing a path to the src/proto directory using the + operator,
and then interpolating it, we limit what is copied to the store, and
therefore what code triggers a nix-proto-srcs rebuild.
Change-Id: I8dd750f6bc5902b74ffb56470bc8a5f2c01c8cf1
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1263
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
Reviewed-by: tazjin <mail@tazj.in>
This store implementation is required in all unit tests that use the
evaluator.
Change-Id: I1cfe8cecab8722cd66dc803747821a2be2b2619f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1269
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
This change does away with the previous special-casing of lists of
certain element sizes, and the use of raw C-style arrays.
Lists are now backed by a std::vector of nix::Value*, which uses the
traceable GC allocator.
This change is unfortunately quite noisy because the accessor methods
were updated/removed accordingly, so all callsites of Nix-related
lists have changed.
For some operations in primops.cc where keeping the previous code
structure would have been more difficult with a "proper" vector, the
implementation has been replaced with std::vector methods. For
example, list concatenation now uses appropriate range inserts.
Anecdotally the performance of this is about equal, to even slightly
better, than the previous implementation.
All language tests pass and the depot paths I've used for testing
still evaluate.
Change-Id: Ib5eca6c0207429cb323a330c838c3a2200b2c693
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1266
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Reviewed-by: Kane York <rikingcoding@gmail.com>
Reviewed-by: glittershark <grfn@gws.fyi>
This has several advantages:
* we can ensure that the vector is traced by the GC
* we don't need to unsafely allocate memory to make an Env
Note that there was previously a check about the size of the
environment, but it's unclear why this was the case (git history
yielded nothing interesting) and it seems to have no effect.
Change-Id: I4998b879a728a6fb68e1bd187c521e2304e5047e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1265
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Reviewed-by: Kane York <rikingcoding@gmail.com>
Reviewed-by: glittershark <grfn@gws.fyi>
all of the executables that get built during regular development depend
on this being set to a directory that contains the nix directory -
previously we had been doing it manually every time, this automates it.
Change-Id: I4c957c0abf0a92ca7122a47d3b141a8ede280e13
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1258
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Configures the CMake build to load & run the GoogleTest tests.
I (grfn) also updated this to get the tests running as part of the nix
derivation, which required defining our own manual configurePhase and
installCheckPhase, rather than depending on the one provided by stdenv.
Not doing this would cause cmake to attempt to *run* the tests as part
of the buildPhase, which wouldn't work because the dynamic libraries
hadn't been put into a place where the test executables knew where to
find them. We're not sure *why* this fixes it, and for some reason
fixing this also breaks the automatic behavior of nixpkgs of passing
-j$NIX_BUILD_CORES -l$NIX_BUILD_CORES to make, but that's eaasy enough
to fix manually in a preBuild
Paired-With: Griffin Smith <grfn@gws.fyi>
Change-Id: I79d61854a3ff47301cdce8a40c76820a97bdf901
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1240
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
Also adds the missing check_contents field to the VerifyStoreRequest
proto message, since it was missed in the original pass. This is done
using a renumbering, which is fine in this case since the proto hasn't
been deployed yet
Change-Id: I92bf4e48a71a25ae02ae02b3deaf6e7c71fe5da7
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1237
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Reviewed-by: Kane York <rikingcoding@gmail.com>
Change-Id: I8b1d84799a608a516d0b4980022a7edd545a1ca1
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1235
Tested-by: BuildkiteCI
Reviewed-by: Kane York <rikingcoding@gmail.com>
Change-Id: I243d67b0bce29d54c7d6e08f5eee70bd395cf9a2
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1234
Tested-by: BuildkiteCI
Reviewed-by: Kane York <rikingcoding@gmail.com>
Backported from:
b3e5eea4a9fcd048a526
Intentionally skipped because we have not backported the JSON changes:
9f46f54de4
Did not apply changes ni primops.cc, because those look suspect and
are also based on something that we don't have in our tree.
Change-Id: I837787ce9f2c90267bc39fce15177980d209d4e9
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1253
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
We want to *trace* the 'Value *' arrays, not garbage-collect them!
Otherwise the vectors/maps can end up pointing to nowhere.
Backported from:
10e17eaa58
Change-Id: I30dc94caa80c9d982e7a14bc67ba2d065e8203aa
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1252
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Previously the memory would occasionally be collected during eval since
the GC doesn't consider the member variable as alive / doesn't scan the
region of memory where the pointer lives.
By using the traceable_allocator<T> allocator provided by Boehm GC we
can ensure the memory isn't collected. It should be properly freed when
SourceExprCommand goes out of scope.
Backported from:
d2c371927e
Change-Id: I1f7c745dbc66c7164bee50f4d9b0d437dbc7dd51
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1251
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
This is a blocker to enabling the linter.
Change-Id: I9f8d3cc19c7539086f53474a505362230fc56c04
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1245
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
This is a bad idea, it shouldn't exist, nixpkgs doesn't use it.
Change-Id: Ic4d1b936d8f059d5c40f0567af165b02427d7e36
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1241
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
These are the queries that are handled in the confusing case statement
in the old daemon implementation, because they have very similar
structure.
Change-Id: Ie7143354f66cef4336dff8072ede9a56271a7e89
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1228
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
What it does I will never know. It takes a path and returns a path.
Something happens in the middle.
Change-Id: I499a9df700e5b954c9aaf6d694753ff34e773dfd
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1210
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
These files were always implicitly depended upon by worker-protocol.hh, but just so happened also always be included in places where the other ones were already present.
We are likely getting rid of this file sooner rather than later, but in
the meantime this will cause clangd to shut up while I'm editing the
file.
Change-Id: I0d765f8b86828b6612e5483417e452221ea3c5b1
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1207
Reviewed-by: tazjin <mail@tazj.in>
Tested-by: BuildkiteCI
Add QueryValidPaths and QuerySubstitutablePaths, both of which filter a
list of paths based on a set of criteria.
Change-Id: I6aa4647efe82b82dc9582a311643d5f9b6d521d5
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1203
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Replaces the previous uses of the (ordered!) std::map and std::set
with absl::flat_hash_{map|set}.
After some careful reading it seems that there is actually no
ordering dependency on these types, and the (drop-in) replacements
perform slightly better.
Overall this is not fixing a bottleneck, just a driveby thing.
Change-Id: Ided695dc75676bd58515aa9382df0be0a09c565e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1220
Reviewed-by: glittershark <grfn@gws.fyi>
Reviewed-by: isomer <isomer@tvl.fyi>
Tested-by: BuildkiteCI
Several definitions of functions declared in eval.hh were previously
implemented in parser.y, this moves them over to parser.cc.
While this still isn't a reasonable place to keep them, the long-term
fix is more likely to be that eval.hh needs to be split up.
Before we get to that point however, this already gives us the ability
to use tooling with this code.
Change-Id: If06fb655325fe281564047ffab0a0a640428a0ee
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1219
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Moves several of the static helper functions into a new parser.cc
file.
Once the rest of the code is usefully extracted, these will be moved
to a private namespace.
Change-Id: I0d7b53dcefe31bb5c6bad3ad7f5fcb48276bf799
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1218
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
First step (of many?) towards extracting all the inline code from the
Yacc file and keeping it somewhere more accessible instead.
Note that none of this code has previously been touched by a linter or
formatter, pretty much ever, so as it is extracted it also undergoes
similar changes to the whole codebase after the initial fork.
Change-Id: If3b7181f22e3b3fd8c58dfa9befa7ee2896ea06d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1217
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Instead of manually iterating over the two bindings to be combined,
this adds a new static method on the Bindings class which merges two
attribute sets by calling the range insertion operator over them.
In some anecdotal tests, this can lead to a ~10% speed bump -
depending on the specific operation.
Change-Id: I5dea03b0589a83a789d3a8a0fc81d0d9e6598371
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1216
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
Changes the derivation name & README overview to say "Tvix" instead of
"tazjix".
The previous name was mostly intended as a joke, and a way for me to
distinguish output paths. It's certainly not the intention to have a
portmanteau with my name here, especially now that several people are
contributing to the fork.
Change-Id: Icface5484d52355111eca23b2f6bd3b9e5567275
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1212
Reviewed-by: glittershark <grfn@gws.fyi>
Tested-by: BuildkiteCI
They're not mentioned anywhere on the client or the server.
Change-Id: Ia78c8bde49326f6bf69b4aeded083cfd3235131f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1188
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
On the chopping block this time:
* requests/responses that return one or more store paths, and contain
nothing else, have been changed to use the same types (StorePath and
StorePaths, respectively)
* QuerySubstitutablePathInfos has been added. It should be noted that
legacy Nix has two versions of this call, one that only queries a
single info (deprecated) and one that queries multiple. We have only
implemented the latter.
* QueryDerivationOutputs has been added.
Change-Id: Iccc9041e7064e141cf593467ecdcc327581c4056
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1186
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
Slight performance optimisation of nix::printString by copying chunks
of the input string which do not need escaping as contiguous blocks.
Paired-With: Perry Lorier <isomer@tvl.fyi>
Change-Id: I48bad90c8f2831ae4524c814a12b1982989922f9
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1184
Tested-by: BuildkiteCI
Reviewed-by: edef <edef@edef.eu>
Reviewed-by: glittershark <grfn@gws.fyi>
A significant fraction of all created attribute sets are empty; hence
this is an easy optimisation to make.
Paired-With: Perry Lorier <isomer@tvl.fyi>
Change-Id: I0884194d04c1ee95b2b239a253515f2152bc0856
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1179
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
This reverts parts of the CLs splitting the backing implementation for
Bindings and moves back to only the BTreeMap-backed implementation.
Our evaluation has indicated that the Vector-backed implementation
does not match the performance of the plain array used upstream, and
in my view the complexity introduced by it is not worth the relatively
small (single-digit percentage) performance increase with a
pivot-point close to the number of attributes yielded by
stdenv.mkDerivation.
Going forward we will trial implementations of attribute sets backed
by HAMTs, and investigate other mechanisms of speeding up the language.
Some changes from the previous CLs are retained, for example the
removal of insert_or_assign and the passing of capacity.
Change-Id: I6eb4b075b453949583360755055c21a29d7ff642
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1172
Reviewed-by: glittershark <grfn@gws.fyi>
Tested-by: BuildkiteCI
Since one of the two implementations essentially uses the same shape
as the upstream Bindings, we backport their merge sort implementation
to ensure that we're doing the same thing semantically.
Paired-With: Luke Granger-Brown <git@lukegb.com>
Paired-With: Vincent Ambo <mail@tazj.in>
Paired-With: Perry Lorier <isomer@tvl.fyi>
Change-Id: I0d865897991eec0c4dd84d9bd0415cd1ca437792
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1162
Tested-by: BuildkiteCI
Reviewed-by: lukegb <lukegb@tvl.fyi>
Reviewed-by: glittershark <grfn@gws.fyi>
We accidentally returned the incremented iterator in the
post-increment, this fixes it.
Paired-With: Luke Granger-Brown <git@lukegb.com>
Paired-With: Vincent Ambo <mail@tazj.in>
Paired-With: Perry Lorier <isomer@tvl.fyi>
Change-Id: I36c79eb56359bb12a78ad3489e7d7d2eb2053510
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1140
Tested-by: BuildkiteCI
Reviewed-by: lukegb <lukegb@tvl.fyi>
Reviewed-by: glittershark <grfn@gws.fyi>
This function in never called, so let's just remove it
Paired-With: Luke Granger-Brown <git@lukegb.com>
Paired-With: Vincent Ambo <mail@tazj.in>
Change-Id: I79125866254d90dd0842bc86830d2103ac313cb6
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1125
Tested-by: BuildkiteCI
Reviewed-by: lukegb <lukegb@tvl.fyi>
Reviewed-by: isomer <isomer@tvl.fyi>
To aid in making the decision of where to (currently just statically)
use a vector or btree as the backing implementation, add an extra
constructor argument to Bindings::NewGC for a capacity, and use
a (currently hardcoded at 32, for no good reason other than it felt like
a reasonable number) pivot to switch between our possible backing
implementations. Then, update all the call sites where it feels
reasonable that we know the capacity statically to *pass* that capacity
to the constructor.
Paired-With: Luke Granger-Brown <git@lukegb.com>
Paired-With: Vincent Ambo <mail@tazj.in>
Paired-With: Perry Lorier <isomer@tvl.fyi>
Change-Id: I1858c161301a1cd0e83aeeb9a58839378869e71d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1124
Tested-by: BuildkiteCI
Reviewed-by: lukegb <lukegb@tvl.fyi>
Reviewed-by: isomer <isomer@tvl.fyi>
Add an alternative impl of the now-abstract Bindings base class that is
backed by a std::vector, somewhat similar but stylistically a little
superior to the array-backed implementation in upstream nix. The
underlying iterator type in BindingsIterator is now backed by a
std::variant that we std::visit an overload over in order to implement
the various bits of the iterator interface.
Paired-With: Luke Granger-Brown <git@lukegb.com>
Paired-With: Vincent Ambo <mail@tazj.in>
Paired-With: Perry Lorier <isomer@tvl.fyi>
Change-Id: I7fbd1f4d5c449e2f9b82102a701b0bacd5e80672
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1123
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
To pave the way for the thing we want to do eventually which is use a
linear-time array for bindings (aka attribute sets) that are statically
known to be small enough to get a performance benefit from doing so,
make the Bindings class abstract, and define a BTreeBindings class that
inherits from it and is (currently always) returned from the static
initializer. The idea is that we'll have an ArrayBindings class as well
later that we can dispatch to conditionally based on an optional
"capacity" parameter or something like that.
There was some difficulty here in getting the iterator to work - the
approach we settled on ended up making a concrete BindingsIterator class
which will wrap a std::variant of either a btree iterator or something
else later, but right now just wraps a btree iterator.
Paired-With: Luke Granger-Brown <git@lukegb.com>
Paired-With: Vincent Ambo <mail@tazj.in>
Paired-With: Perry Lorier <isomer@tvl.fyi>
Change-Id: Ie02ca5a1c55e8ebf99ab1e957110bd9284278907
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1121
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Having a default constructor for this causes a variety of annoying
situations across the codebase in which this is initialised to an
unexpected value, leading to constant guarding against those
conditions.
It turns out there's actually no intrinsic reason that this default
constructor needs to exist. The biggest one was addressed in CL/1138
and this commit cleans up the remaining bits.
Change-Id: I4a847f50bc90e72f028598196592a7d8730a4e01
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1139
Reviewed-by: isomer <isomer@tvl.fyi>
Tested-by: BuildkiteCI
This has been providing a warning and it's been bothering me.
Change-Id: I0548059950ec4250d7cf0938f9deae09eafe593c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1141
Reviewed-by: tazjin <mail@tazj.in>
Reviewed-by: isomer <isomer@tvl.fyi>
Tested-by: BuildkiteCI
nix:AttrName was one of the few classes that relied on the default
constructor of nix::Symbol (which I am trying to remove in a separate
change).
The class essentially represents the name of an attribute in a set,
which is either just a string expression or a dynamically evaluated
expression (e.g. string interpolation).
Previously it would be constructed by only setting one of the fields
and defaulting the other, now it is an explicit std::variant.
Note that there are several code paths where not all eventualities are
handled and this code is bug-for-bug compatible with those, except
that unknown conditions (which should never work) are now throwing
instead of silently doing ... something.
The language tests pass with this change, and the depot derivations
that I tested with evaluated successfully.
Change-Id: Icf1ee60a5f8308f4ab18a82749e00cf37a938a8f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1138
Reviewed-by: edef <edef@edef.eu>
Reviewed-by: glittershark <grfn@gws.fyi>
Tested-by: BuildkiteCI
Implements the fairly common lambda overload class used for std::visit
over variants and other things that require groups of callables.
Change-Id: Ia7448b7e1bd349b4909974758e6e6303a80d86d7
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1137
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
Reviewed-by: edef <edef@edef.eu>
Backport of b244e65cdb
Fixes https://github.com/NixOS/nix/issues/3175
------
Original fault description by puck:
I suspect I found the issue: The Nix Command structs are stored on the
heap (using ref<T>, a wrapper around std::shared_ptr<T>), which means
that any pointers that the NixRepl struct contains are eligible to be
reaped by the GC. This includes, but is not limited to, the Env
pointer, which seems to cause most of the random segfaults, or random
other values in the environment, which seems to be what @arianvp
experienced too.
Change-Id: I376d7cfd432daaa6f1fbbf77788ff048082f34e5
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1001
Tested-by: BuildkiteCI
Reviewed-by: Kane York <rikingcoding@gmail.com>
Reviewed-by: glittershark <grfn@gws.fyi>
Previously the nix-daemon would crash if a user fed it invalid store
paths for drv files. The crash was due to the changed assertion
triggering. Whenever that assertion would hit the nix-daemon process
along with all it's current childs (running builds from all users) would
be interrupted.
Before this patch:
$ nix-store --realise /nix/store/aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa.drv
don't know how to build these paths:
/nix/store/aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa.drv
error: unexpected end-of-file
< nix-daemon terminates >
With this patch:
$ nix-store --realise /nix/store/aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa.drv
don't know how to build these paths:
/nix/store/aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa.drv
error: path '/nix/store/aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa.drv' is not a valid store path
< nix-daemon does *NOT* terminate >
Change-Id: I01c5048c8a43a8b9154bdeb781d05b7744869ec0
Reviewed-on: https://cl.tvl.fyi/c/depot/+/981
Tested-by: BuildkiteCI
Reviewed-by: glittershark <grfn@gws.fyi>
Reviewed-by: tazjin <mail@tazj.in>
This takes us about halfway through worker-protocol.hh
I have left out the documentation strings for some of these items
because I don't feel that I can currently write an unambigous
description of them. For now I am just attempting to match the types.
Change-Id: Iae64b1676152fe4ea069e2021b75ad76465cf368
Reviewed-on: https://cl.tvl.fyi/c/depot/+/960
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
This isn't actually used yet, but forces the protos to be included in
the build which is useful for iteration.
Change-Id: I2abcaf297f34ae741f00ad0c929b226d5603c9d7
Reviewed-on: https://cl.tvl.fyi/c/depot/+/928
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Adds dependencies on the gRPC & protobuf libraries, and implements Nix
code to generate the C++ sources from the included proto definitions.
This is theoretically supported via CMake, but practically doesn't
work and I don't care to debug why.
Doing it like this lets us instead add a CMake library target for our
proto definitions based on the sources generated by Nix.
Pros:
* no need to deal with the gRPC CMake mess
* it works!
Cons: * iteration requires nix-shell restart
Change-Id: Ie1fe9807fc96c49cb8f7161ba59d093456062b15
Reviewed-on: https://cl.tvl.fyi/c/depot/+/927
Tested-by: BuildkiteCI
Reviewed-by: isomer <isomer@tvl.fyi>
Adds initial gRPC definitions for the Nix worker protocol, which is
currently defined messily across the following files:
src/libstore/worker-protocol.hh
src/libstore/remote-store.cc
src/nix-daemon/nix-daemon.cc
The protocol definition is basically a big enum with the signatures of
the calls being implicit in the various client/server implementation
functions.
The definitions in this file are slowly reversed from these implicit
signatures, and are likely to contain an error or two which will be
weeded out when this is taken into use.
Only a handful of the calls are included in this commit, it is
intended to get us up and running first.
Change-Id: Ibc9b2ab4b91a064c8935f09f7ac72bb8150fb476
Reviewed-on: https://cl.tvl.fyi/c/depot/+/926
Reviewed-by: isomer <isomer@tvl.fyi>
Tested-by: BuildkiteCI
Same as cl/921, it seems that the trick to making this work is indeed
overriding the C++ standard used in the dependency.
Change-Id: I3c5984d71014d774c161ecc283844f504fd44719
Reviewed-on: https://cl.tvl.fyi/c/depot/+/922
Tested-by: BuildkiteCI
Reviewed-by: lukegb <lukegb@tvl.fyi>
This didn't work previously ... but now it does. I think setting the
standard explicitly is what did the trick, but it's slightly unclear
to me why.
Either way this means that Abseil is no longer constantly getting
recompiled when building Nix, which is nice.
Change-Id: I377f7b68bf1ef9045df6a2eee8fdd0c92f243547
Reviewed-on: https://cl.tvl.fyi/c/depot/+/921
Tested-by: BuildkiteCI
Reviewed-by: lukegb <lukegb@tvl.fyi>
These bits are no longer required with the hashmap-backed
implementation of attribute sets.
Change-Id: I8b936d8d438a00bad4ccf8e0b4dd719c559ce8c2
Reviewed-on: https://cl.tvl.fyi/c/depot/+/912
Tested-by: BuildkiteCI
Reviewed-by: edef <edef@edef.eu>
This reverts commit 368e8d1edd.
Reason for revert: Didn't mean to submit, and the phase is currently failing (which breaks the otherwise-functional derivation)
Change-Id: I515b2fb45188dc90f09ae2458453192487c74d71
Reviewed-on: https://cl.tvl.fyi/c/depot/+/581
Reviewed-by: tazjin <mail@tazj.in>
Add an installCheckPhase that runs the appropriate substituteAll on
common.sh and runs the lang.sh tests with the build artifacts in the
PATH.
Change-Id: I2df5a93b8f3ffdfdc194a0e7d6b6669ef520c345
Reviewed-on: https://cl.tvl.fyi/c/depot/+/561
Reviewed-by: glittershark <grfn@gws.fyi>
We don't want traces compiled out since they're an actual language
feature that're used in userspace - also their absence is breaking the
tests
Change-Id: Icaefca8f52e94001785f724fdc0c10a7586b24e7
Reviewed-on: https://cl.tvl.fyi/c/depot/+/562
Reviewed-by: lukegb <lukegb@tvl.fyi>
Reviewed-by: Kane York <rikingcoding@gmail.com>
Reviewed-by: lukegbot <bot@lukegb.com>
This includes absl, which we install into the output, and boost and the boehm GC,
which are moved to propagated deps.
Change-Id: I8f9f9795ff92e26b2320359064241d7fd59c2d33
Reviewed-on: https://cl.tvl.fyi/c/depot/+/549
Reviewed-by: tazjin <mail@tazj.in>
This also installs the rest of corepkgs as a side-effect.
Change-Id: I67a42d45793d5e8fdad51c1f306eebf63e9c2868
Reviewed-on: https://cl.tvl.fyi/c/depot/+/548
Reviewed-by: tazjin <mail@tazj.in>
Also fixes the pkgconfig files to use the corresponding CMake variables.
Change-Id: I8095b8aff39ad91e592f3edc95555c9f1f1f153d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/545
Reviewed-by: tazjin <mail@tazj.in>
This ensures that we install both glog's .a and all the .so files we
generate into a single consistent output lib path (which is, err,
lib64, but whatever).
Change-Id: Ib6ac6eacf5f56e4b719cfb586db731efc122c31b
Reviewed-on: https://cl.tvl.fyi/c/depot/+/544
Reviewed-by: tazjin <mail@tazj.in>
Having a colon in the path may cause issues, and having the hash
function indicated isn't actually necessary. We now verify the path
format in the tests to prevent regressions.
(cherry picked from commit c65a6fa86aef7bdf51fb4fba7bd31d265619ba3f)
This makes the paths consistent without relying on ordering.
Co-authored-by: edef <edef@edef.eu>
(cherry picked from commit 515c0a263e137a00e82f7d981284dbe54db23247)
Completes the switch from Meson to CMake for the core build system in
Nix.
Meson was added originally because someone else had already done the
work for integrating it in Nix and it was an upgrade from the previous
setup.
However over time it became clear that Meson is not quite mature
enough for projects like Nix that have occasionally peculiar
configuration constraints.
Some issues encountered with Meson (some of these are due to the Meson
setup in Nix):
* Difficulty with generating correct compile_commands.json for
external tools like clangd
* Difficulty linking to libc++ when using clang
* Ugly shell invocations for certain parts of the build system (I want
these to be gone!!!)
This CMake setup mimics the Meson configuration, but there are some
differences (some temporary):
* headers are now included separately for each library (see a previous
commit that changes includes appropriately)
* autoheaders-style configuration is currently hardcoded. Before
blindly copying this I want to evaluate how much of it actually exists
for portability concerns that I don't have (such as support for OS
X).
* Nix is built with libc++ by default.
* [libstore] SQL schema is now inlined via a generated header, not an
included string literal
Abseil is still built as part of this build, rather than an external
dependency, because it chokes on differently configured compiler
invocations.
Note that because of the move to libc++ an unwanted behaviour is
introduced: glog log messages no longer have a body. I have yet to
debug what is going on there.
Previously all includes were anchored in one global mess of header
files. This moves the includes into filesystem "namespaces" (if you
will) for each sub-package of Nix.
Note: This commit does not introduce the relevant build system changes.
This function was a custom (and inefficient in the case of
single-character delimiters) string splitter which was used all over
the codebase. Abseil provides an appropriate replacement function.
Replaces these functions with corresponding functions from Abseil,
namely absl::StripAsciiWhitespace and absl::SimpleAtoi.
In the course of doing this some minor things I encountered along the
way were also refactored.
This also changes the signatures of the various custom readFile
functions to use absl::string_view types.
It is considered bad form to use things from includes in headers, as
these directives propagate to everywhere else and can make it
confusing.
types.hh (which is includes almost literally everywhere) had some of
these directives, which this commit removes.
Suppose I have a path /nix/store/[hash]-[name]/a/a/a/a/a/[...]/a,
long enough that everything after "/nix/store/" is longer than 4096
(MAX_PATH) bytes.
Nix will happily allow such a path to be inserted into the store,
because it doesn't look at all the nested structure. It just cares
about the /nix/store/[hash]-[name] part. But, when the path is deleted,
we encounter a problem. Nix will move the path to /nix/store/trash, but
then when it's trying to recursively delete the trash directory, it will
at some point try to unlink
/nix/store/trash/[hash]-[name]/a/a/a/a/a/[...]/a. This will fail,
because the path is too long. After this has failed, any store deletion
operation will never work again, because Nix needs to delete the trash
directory before recreating it to move new things to it. (I assume this
is because otherwise a path being deleted could already exist in the
trash, and then moving it would fail.)
This means that if I can trick somebody into just fetching a tarball
containing a path of the right length, they won't be able to delete
store paths or garbage collect ever again, until the offending path is
manually removed from /nix/store/trash. (And even fixing this manually
is quite difficult if you don't understand the issue, because the
absolute path that Nix says it failed to remove is also too long for
rm(1).)
This patch fixes the issue by making Nix's recursive delete operation
use unlinkat(2). This function takes a relative path and a directory
file descriptor. We ensure that the relative path is always just the
name of the directory entry, and therefore its length will never exceed
255 bytes. This means that it will never even come close to AX_PATH,
and Nix will therefore be able to handle removing arbitrarily deep
directory hierachies.
Since the directory file descriptor is used for recursion after being
used in readDirectory, I made a variant of readDirectory that takes an
already open directory stream, to avoid the directory being opened
multiple times. As we have seen from this issue, the less we have to
interact with paths, the better, and so it's good to reuse file
descriptors where possible.
I left _deletePath as succeeding even if the parent directory doesn't
exist, even though that feels wrong to me, because without that early
return, the linux-sandbox test failed.
Reported-by: Alyssa Ross <hi@alyssa.is>
Thanks-to: Puck Meerburg <puck@puckipedia.com>
Tested-by: Puck Meerburg <puck@puckipedia.com>
Reviewed-by: Puck Meerburg <puck@puckipedia.com>
(cherry picked from commit c05e20daa1abb3446e378331697938b78af2b3d7)
Replaces the previous implementations which performed sorting with one
that instead walks through the map (which is already sorted) and
yields values from it.
This fixes a handful of language tests because the previous
implementation did not actually yield useful values on the new implementation.
In the change to the backing structure of attribute sets, the
requirement to manually balance the capacity of the structure went
away.
This is a) because Abseil's data structures manage this on their own,
and b) because the new Bindings class is allocated using `new (GC)`
rather than writing into a predefined memory area.
As part of this change functions related to the capacity were
deprecated and set to 0 values, which in turn caused the creation of
new attribute sets to return the same (mutable!) default value in
various cases, leading to "side effects" that caused evaluation
failures.
FWIW, I'm not sure if this optimisation had noticeable performance
impact, but while untangling libexpr it definitely doesn't help trying
to follow what it's doing - so bye, bye!
This wrapper derivation (which assumes that the depot is available at
~/depot) can be used to actually get clangd working with
//third_party/nix.
In my setup I can launch this with M-x eglot, followed by
env
CLANGD_FLAGS='--compile-commands-dir=/home/tazjin/projects/nix-build'
nix-shell -A third_party.nix --run 'nix-clangd' /home/tazjin/depot
This is closer to bug-for-bug compatibility with the previous version,
which would put new elements at the end of the array and (due to the
linear scan) return previous ones.
The new attribute set API uses the iterators of the btree_map
directly. This requires changes in various files because the internals
of libexpr are very entangled.
This code runs and compiles, but there is a bug causing empty
attribute sets to be assigned incorrectly.
Instead of doing some sort of inline merge-sort of the two attribute
sets, use the attribute sets merge function.
This commit alone does not build and is not supposed to.
This is the first step towards replacing the implementation of
attribute sets with an absl::btree_map.
Currently many access are done using array offsets and pointer
arithmetic, so this change is currently causing Nix to fail in various
ways.
Replaces most uses of `string` with `std::string`.
This came up because I removed the "types.hh" import from
"symbol-table.hh", which percolated through a bunch of files where
`string` was suddenly no longer defined ... *sigh*
This replaces the previous use of std::unordered_set with
absl::node_hash_set.
This type was chosen because the current implementation requires
pointer stability.
This does not yet touch the 'Attr' struct.
As a bonus, the implementation of the SymbolTable struct is now
consolidated into a single header/implementation file pair.
Meson is unable to use CMake in Nix to determine the internal
structure of the Abseil libraries.
This commit adds an explicit list of most of the Abseil targets that
are relevant (so far) and bundles them into a list that is linked
together.
cmake automatically runs a configure hook which breaks the build,
since this isn't actually a cmake project. This hook is now disabled.
Additionally Abseil's sources are linked to an absolute derivation
path when the build launches, as opposed to the relative path used for
development builds.
This applies the modernization fixes listed here:
https://clang.llvm.org/extra/clang-tidy/checks/list.html
The 'modernize-use-trailing-return-type' fix was excluded due to my
personal preference (more specifically, I think the 'auto' keyword is
misleading in that position).