Rather than storing the leaked allocation for the string as the key in
the interner, store the hash (using NoHashHashBuilder). I thought this
would improve performance, but it doesn't:
hello outpath time: [736.85 ms 748.42 ms 760.42 ms]
change: [-2.0754% +0.4798% +2.7096%] (p = 0.72 > 0.05)
No change in performance detected.
but it at least doesn't *hurt* performance, and it *does* avoid an
`unsafe`, so it's probably net good.
Change-Id: Ie413955bdb6f04b1f468f511e5ebce56e329fa37
Reviewed-on: https://cl.tvl.fyi/c/depot/+/12049
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
Autosubmit: aspen <root@gws.fyi>
The problem with using runCommand and recreating the src directory with
lndir is that it changes the file types of individual files, they will
now be a symlink instead of a regular file. If you have a crate that tests
that a file is of regular type then it will fail inside the crate2nix
derivation.
Also regenerate Cargo.nix for //tvix as it will be needed in the next
commit.
Change-Id: I9275602cc17a428f9fdf0e55daf12cd673bbc030
Reviewed-on: https://cl.tvl.fyi/c/depot/+/12131
Autosubmit: Ilan Joselevich <personal@ilanjoselevich.com>
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Use the faster and newer MiMalloc memory allocator for all endpoints in
the workspace.
Change-Id: Ic60237284ed168e46ec6e8f28e2710bae4385c6f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/12149
Tested-by: BuildkiteCI
Reviewed-by: aspen <root@gws.fyi>
Per https://nnethercote.github.io/perf-book/hashing.html, we have
basically no reason to use the default hasher over a faster,
non-DoS-resistant hasher. This gives a nice perf boost basically for
free:
hello outpath time: [704.76 ms 714.91 ms 725.63 ms]
change: [-7.2391% -6.1018% -4.9189%] (p = 0.00 < 0.05)
Performance has improved.
Change-Id: If5587f444ed3af69f8af4eead6af3ea303b4ae68
Reviewed-on: https://cl.tvl.fyi/c/depot/+/12046
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
Reviewed-by: Ilan Joselevich <personal@ilanjoselevich.com>
Autosubmit: aspen <root@gws.fyi>
Default to always leaking strings, and copying strings by copying
pointers rather than cloning the underlying allocation. This (somewhat
bafflingly) doesn't seem to affect any benchmarks, but paves the way for
some tricks around string allocation that do.
Unfortunately, we can't do this (yet?) for contextful strings, for
reasons I don't currently understand but which I will address later,
when I address contextful strings more holistically.
I've left a flag in here to disable this, both to test the cloning logic
for strings for when/if we decide to bring this back, and to allow
people who care more about memory usage than perf to disable leaking.
Change-Id: Iec44bcbfe9b3d20389d2450b9a551792a79b9b26
Reviewed-on: https://cl.tvl.fyi/c/depot/+/12045
Autosubmit: aspen <root@gws.fyi>
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
This provides a PathInfoService implementation using redb
(https://github.com/cberner/redb) as the underlying storage engine.
Both an in-memory variant, as well as a filesystem one is provided,
similar how it's done with the sled implementation.
Supersedes: https://cl.tvl.fyi/c/depot/+/11692
Change-Id: I744619c51bf2efd0fb63659b12a27cbe0b2fd6fc
Signed-off-by: Ilan Joselevich <personal@ilanjoselevich.com>
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11995
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
It introduces a new accept_trace function for axum0.7 which can be used
to accept a header trace from a received request. This function can be
used for tonic 0.12 once that version is released, and the specific
`accept_trace` function within `tvix_tracing::propagate::tonic` can then
be removed.
This also integrates http propagation into the nar_bridge crate.
Change-Id: I46dcc797d494bb3977c2633753e7060d88d29129
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11925
Reviewed-by: Brian Olsen <me@griff.name>
Tested-by: BuildkiteCI
Reviewed-by: Simon Hauser <simon.hauser@helsinki-systems.de>
Reviewed-by: flokli <flokli@flokli.de>
This ingests NAR files into the {Blob,Directory}Service, which are
already part of the AppState.
As we then need to correlate the root node to the uploaded PathInfo, we
need to keep a (short-lived) lookup table from NARHash to root node
around. We insert it into a `LruCache` after the NAR is uploaded, and
use `peek()` to do the lookup, which doesn't update the LRU list.
Change-Id: I48a4c6246bacf76559c5a4ccad2a0bc25c1b7900
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11986
Tested-by: BuildkiteCI
Reviewed-by: Brian Olsen <me@griff.name>
This adds an implementation of nar-bridge in Rust.
Currently, only the GET parts are implemented.
Contrary to the Go variant, this doesn't try to keep a mapping from nar
hashes to root node in memory, it simply encodes the root node itself
(stripped by its basename) into the URL.
This pulls in a more recent version of axum than what we use in
tonic, causing two versions of http and hyper, however dealing with
`Body::from_stream` in axum 0.6 is much more annoying, and
https://github.com/hyperium/tonic/pull/1740 suggests this will be fixed
soon.
Change-Id: Ia4c2dbda7cd3fdbe47a75f3e33544d19eac6e44e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11898
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: Brian Olsen <me@griff.name>
Tested-by: BuildkiteCI
Juggle around the internals of the tvix-cli crate so that we expose the
Repl as a public type with a `send` method, that sends a string to the
repl and *captures all output* so that it can be subsequently asserted
on in tests. Then, demonstrate that this works with a single (for now)
REPL test using expect-test to assert on the output of a single command
sent to the REPL.
As the REPL gets more complicated, this will allow us to make tests that
cover that complex behavior.
Change-Id: I88175bd72d8760c79faade95ebb1d956f08a7b83
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11958
Autosubmit: aspen <root@gws.fyi>
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
Allow binding variables at the REPL's toplevel in the same way the Nix
REPL does, using the syntax <ident> = <expr>. This fully, strictly
evaluates the value and sets it in the repl's "env", which gets passed
in at the toplevel when evaluating expressions.
The laziness behavior differs from Nix's, but I think this is good:
❯ nix repl
Welcome to Nix version 2.3.18. Type :? for help.
nix-repl> x = builtins.trace "x" 1
nix-repl> x
trace: x
1
nix-repl> x
1
vs tvix:
tvix-repl> x = builtins.trace "x" 1
trace: "x" :: string
tvix-repl> x
=> 1 :: int
tvix-repl> x
=> 1 :: int
Bug: https://b.tvl.fyi/issues/371
Change-Id: Ieb2d626b7195fa87be638c9a4dae2eee45eb9ab1
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11954
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Autosubmit: aspen <root@gws.fyi>
Introduces a helper function within tvix-tracing that returns a reqwest
tracing middleware that will ingest the traceparent if otlp is enabled.
It is feature flagged in tvix-tracing so not every consumer of that
library automatically has reqwest in its dependencies.
Tested using netcat to verify that the `traceparent` header is there if
otlp is enabled and missing if otlp feature is disabled.
Change-Id: I5abccae777b725f5ff7382e3686165383c477a39
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11886
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
See https://github.com/NixOS/nixpkgs/pull/323753 for details.
Changes:
* git: temporarily comment out dottime patch (it doesn't apply, but it's not critical)
* third-party/cgit: use an older git version where dottime patch still applies
* 3p/crate2nix: remove crate2nix patches included in latest release
* tvix: remove unneeded defaultCrateOverrides (upstreamed to nixpkgs)
* tvix: regenerate Cargo.nix
* tvix/nix-compat: remove unnused AtermWriteable::aterm_bytes pub(crate) function
* tvix/nix-compat: remove redundant trait bounds
* tvix/glue: use clone_into() to set drv.{builder,system}
* tools/crate2nix: apply workaround for https://github.com/numtide/treefmt/issues/327
* toold/depotfmt: expose treefmt config as passthru
* tools/crate2nix: undo some more hacks in the crate2nix-check drv
Change-Id: Ifbcedeb3e8f81b2f6ec1dbf10189bfa6dfd9c75c
Co-Authored-By: Florian Klink <flokli@flokli.de>
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11907
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
This allows binding on unix sockets, as well as systemd socket
activation.
Change-Id: Icf648c4fd0895468c52607deb6397b8b5928102b
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11901
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Reviewed-by: Connor Brewster <cbrewster@hey.com>
This introduces optional helper function in tvix/tracing for trace
propagation and uses these helper in the `tvix-store`.
The GRPCBlobService, GRPCDirectoryService and GRPCPathInfoService now
accept a generic client, meaning the client can be generated with either
`::new` or `::with_interceptor`.
This was tested and validated by starting a `tvix-store daemon` and
`tvix-store import`.
Change-Id: I4b194483bf09266820104b4b56e4a135dca2b77a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11863
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
This patch has already been applied on upstream but there hasn't been an
official release yet so we patch it ourselves.
Upstreamed patch: 0209f258cd
Follow-up of: https://cl.tvl.fyi/c/depot/+/11856
Change-Id: If56071ce5753fd26e2b4c203cd831bbe5d329009
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11858
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Don't show an empty spinner for daemon commands.
Move the bar to the right, so the text is better aligned between spinner
progress and bar progress styles.
Generally, push progress bars a bit more down to the place where we can
track progress. This includes adding one in the upload_blob span.
Introduce another progress style template for transfers, which
interprets the counter as bytes (not just a plain integer), and also a data rate.
Use it for here and in the fetching code, and also make the progress bar
itself a bit less wide.
Change-Id: I15c2ea3d2b24b5186cec19cd3dbd706638497f40
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11845
Tested-by: BuildkiteCI
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: Simon Hauser <simon.hauser@helsinki-systems.de>
Both umounts happening from another process, as well as tvix-store
itself calling umount() on FuseDaemon will cause the FUSE worker threads
to terminate.
So far there was no nice way to wait on these threads to be terminated
from multiple places, causing the `tvix-store mount` command to only be
terminated if interrupted via ctrl-c, not via an external umount.
Update FuseDaemon to use a ThreadPool, which gives us a join primitive
over all threads, that can also be called from multiple places.
Await on a join() from there to end the program, not the ctrl-c signal
handler as it was before.
Using FuseDaemon from multiple tasks requires Arc<>-ing both the
ThreadPool as well as the inner FuseSession (which also needs to be
inside a Mutex if we want to unmount), but now we can clone FuseDaemon
around and use it in two places. We could probably also have used an
Option and drop the FuseSession after the first umount, but this looks
cleaner.
Change-Id: Id635ef59b560c111db52ad0b3ca3d12bc7ae28ca
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11825
Reviewed-by: Brian Olsen <me@griff.name>
Tested-by: BuildkiteCI
If compiled with this features, this emits packets compatible with the
[Tracy](https://github.com/wolfpld/tracy) format.
Change-Id: I330f5d85ab290abe51f2df38dc55464f3ccfc6cd
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11815
Tested-by: BuildkiteCI
Reviewed-by: Connor Brewster <cbrewster@hey.com>
This introduces another feature flag, "tracy" to the `tvix-tracing` crate.
If enabled (not enabled by default), it'll add an additional layer
emitting packets in a format that https://github.com/wolfpld/tracy can
display.
I had to be a bit tricky with the combinatorial complexity when adding
this, but the resulting code still seems manageable.
Change-Id: Ica824496728fa276ceae3f7a9754be0166e6558f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10952
Tested-by: BuildkiteCI
Reviewed-by: Simon Hauser <simon.hauser@helsinki-systems.de>
Reviewed-by: flokli <flokli@flokli.de>
Provide a new interface for forcing a flush of otlp traces and use this
interface to shutdown otlp prior to exiting tvix-store, either if the
tool was stopped with a SIGTERM or ended regularly.
This also fixes an issue where traces were not even exported if for
example we just imported 10 paths and never even emitted more than 256
traces. The implementation uses a mpsc channel so a flush can be done
without having to wait for it to complete. If you want to wait for a
flush to complete you can provide a oneshot channel which will receive a
message once flushing is complete.
Because of a otlp bug `force_flush` as well as
`shutdown_tracer_provider` need to be executed using `spawn_blocking`
otherwise the function will deadlock. See
https://github.com/open-telemetry/opentelemetry-rust/issues/1395#issuecomment-1953280335
Change-Id: I0a828391adfb1f72dc8305f62ced8cba0515847c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11803
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Autosubmit: Simon Hauser <simon.hauser@helsinki-systems.de>
This allows explicitly opting in to get DEBUG-level log lines, by
setting RUST_LOG.
It currently also causes traces to be emitted in all cases, so we might
do some runtime filtering there too, as discussed in cl/11791.
Change-Id: I2865bb06a62465836d63196422f5f734f7165386
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11801
Tested-by: BuildkiteCI
Reviewed-by: aspen <root@gws.fyi>
Autosubmit: flokli <flokli@flokli.de>
This gives some better feedback something is going on, gives a (named)
root for all the progress children we're drawing, and also counts the
time we are in eval.
Change-Id: Ibe81dcebf0a2b59bb0680da62e206eb5270f9d3c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11798
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Reviewed-by: aspen <root@gws.fyi>
In `store_path_to_node`, in case we need to build or fetch something,
render a progress bar, using the spinner for now.
We can upgrade this to a progress *bar* later.
Change-Id: I4a7cf5ef8f639076f176af9b39d276be3f37c8ff
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11793
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Tested-by: BuildkiteCI
Introduce a new common crate that contains tracing boilerplate which then
can be used in the cli, tvix-store and tvix-build crates.
It has otlp as an optional feature, which is currently only used by
tvix-store.
Change-Id: I41468ac4d9c65174515d721513b96fea463d6ed2
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11758
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
Autosubmit: Simon Hauser <simon.hauser@helsinki-systems.de>
We don't use this in tvix-glue currently. A previous version of cl/11747
did, and I forgot to remove it afterwards.
Change-Id: I6191ffc60450f3e79f22a3aca55246e3956ee9cc
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11756
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
Reviewed-by: Simon Hauser <simon.hauser@helsinki-systems.de>
Autosubmit: flokli <flokli@flokli.de>
This adds the tracing-indicatif crate, and configures it as a layer in
our tracing_subscriber pipeline to emit progress for every span that's
configured so.
It also moves from using std::io::stderr to write logs to using their
writer, to avoid clobbering output.
Progress bar styles are defined in a lazy_static, moving this into a
general tracing is left for later.
This adds some usage of this to the `imports` and `copy` commands.
The output can still be improved a bit - we should probably split each
task up into a smaller (instrumented) helper functions, so we can create
a progress bar for each task.
Change-Id: I59a1915aa4e0caa89c911632dec59c4cbeba1b89
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11747
Reviewed-by: flokli <flokli@flokli.de>
Reviewed-by: Simon Hauser <simon.hauser@helsinki-systems.de>
Tested-by: BuildkiteCI
Autosubmit: flokli <flokli@flokli.de>
Transparently support multiline input in the Tvix REPL, by handling the
UnexpectedEOF error returned by the parser and using it to progressively
build up an input expr over multiple iterations of the REPL's outer
loop.
This works quite nicely:
❯ cargo r --bin tvix
Compiling tvix-cli v0.1.0 (/home/aspen/code/depot/tvix/cli)
Finished dev [unoptimized + debuginfo] target(s) in 1.72s
Running `target/debug/tvix`
tvix-repl> { foo
> =
> 1;
> }
=> { foo = 1; } :: set
<press up arrow>
tvix-repl> { foo
=
1;
}
=> { foo = 1; } :: set
Change-Id: Ib0ed4766b13e8231d696cdc27281ac158e20a777
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11732
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: aspen <root@gws.fyi>
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
If building tvix-build with the tonic-reflection feature, it needs
to import `tvix_castore::proto::FILE_DESCRIPTOR_SET`, which is only
available if tvix-castore is built with the `tonic-reflection` feature.
Change-Id: I355b4c5b4c1333d5cc56335de47ad5d2f1db6337
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11716
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: flokli <flokli@flokli.de>
In order to be compatible with the nix XML generator, it’s easier to
generate the XML directly, instead of going through a library which we
have to bend to do what we need.
Removes dependency on `xml-rs`, which came with a full XML parser that
we didn’t use. Only takes a tiny bit of code for the XML escaping,
somewhat simplified.
I add a little escaping value, to make sure we have the same behaviour
as nix proper.
Interestingly enough, we never need to escape XML attribute names,
because the `builtins.toXML` format encodes user-defined values as
attribute keys only. So we only escape attribute values.
Fixes: https://b.tvl.fyi/issues/399
Change-Id: If4d407d324864b3bb9aa3160e2ec6889f7727127
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11697
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
Autosubmit: Profpatsch <mail@profpatsch.de>
Drop the (unused) feature flags here, and get rid of some crate
dependencies.
Change-Id: Id64852b498725467d56abb12b548301cfba6a760
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11678
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
This has already been dropped for tvix-store, drop it here as well.
Change-Id: Ib3aa37bbe9fd4c498b5ba1213f7d922d0c64ffc7
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11677
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Tested-by: BuildkiteCI
This gives multi-listener support, as well as listening on named socket
FDs.
Change-Id: I92b441f854e2faf80074463d0ca6bdc23cbd890a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11663
Tested-by: BuildkiteCI
Reviewed-by: yuka <yuka@yuka.dev>
Autosubmit: flokli <flokli@flokli.de>
This aligns the tonic version we use directly in Tvix to the one pulled
in by the OTLP stack.
Change-Id: I658528c8dabb7cd6948f1207ddcdef1984e82037
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11666
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: yuka <yuka@yuka.dev>
Tested-by: BuildkiteCI
This one doesn't require us to deal with poisoning, is upgradeable and
the right thing to use when locking access to data, not IO resources.
Change-Id: I78634953a73404500d28f51f1d93a87e215c8149
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11612
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Reviewed-by: Connor Brewster <cbrewster@hey.com>
This provides an implementation of PathInfoService storing PathInfo in
memory up to a certain capacity, then evicting these that have been used
the least recently.
Change-Id: I9d738687caf4f181a957f72245f26b92832313cd
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11622
Tested-by: BuildkiteCI
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Autosubmit: flokli <flokli@flokli.de>
Rust 1.77 supports async recursion as long as there is some form of
indirection (ie. `Box::pin`). This removes the need to use the
async-recursion crate.
Change-Id: Ic9613ab7f32016f0103032a861edff92e2fb8b41
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11596
Reviewed-by: flokli <flokli@flokli.de>
Autosubmit: Connor Brewster <cbrewster@hey.com>
Tested-by: BuildkiteCI
Rename read_nar to ingest_nar, and have it use the async nar reader
version, and the ingest_entries machinery.
This means we can now drop all code dealing with manually assembling
castore nodes.
Update our consumer, NixHTTPPathInfoService to use the new API.
As we now accept an AsyncRead, we don't need to do any blocking here
anymore, and can use the same async-compression crate as in the fetching
logic (and support some more compression formats out of the box).
Change-Id: I8646d20bd8603f8da47b5c84bc9e4ac236eb7f1a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11580
Tested-by: BuildkiteCI
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: Connor Brewster <cbrewster@hey.com>
The emulator and bigtable client are quite big. Remove them from the
default //tvix:shell.
Put the tests behind a `integration` feature flag, and add a variant
with that enabled to CI, and drop the bigtable tools from //tvix:shell.
Change-Id: Ie042097a0d6fc26542faa96c139b77298ccb160a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11582
Reviewed-by: edef <edef@edef.eu>
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Since a recent nixpkgs bump bringing a version of cargo with
https://github.com/rust-lang/cargo/pull/12914,
crate2nix creates a crate-hashes.json with all crate hashes from
Cargo.lock (and downloads a lot of stuff while producing it).
https://github.com/nix-community/crate2nix/pull/341 prevents this from
happening, but our hardcoded crate2nix pin prevented us from getting the
fix included in 0.14.0, which did land in nixpkgs.
Replace the pin with a simply override, carrying our only leftover patch
on top of it, and link to that PR.
Change-Id: I9503898e15d61fa6a2b1589d141bec1b4ed3d616
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11581
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Reviewed-by: sterni <sternenseemann@systemli.org>
Nix supports file:// - URLs for `fetchurl` and `fetchTarball`.
Convert the enums and function arguments to hold a URL type.
reqwest::Url is a re-export of the url crate, but they don't re-export
the parsing errors, and as we want to hold these in our Error types, add
it to Cargo.toml explicitly.
The Fetcher::download function now checks on the scheme, and either
opens the file locally, or does do a HTTP request as before.
Fetch gets its custom debug impl, removing potentially sensitive
username and password out of URLs.
Change-Id: I777db1fe487370e822cbfec4624034aca5e08045
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11504
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
We actually want to delay fetching until we actually need the file. A
simple evaluation asking for `.outPath` or `.drvPath` should work even
in a pure offline environment.
Before this CL, the fetching logic was quite distributed between
tvix_store_io, and builtins/fetchers.rs.
Rather than having various functions and conversions between structs,
describe a Fetch as an enum type, with the fields describing the fetch.
Define a store_path() function on top of `Fetch` which can be used to
ask for the calculated store path (if the digest has been provided
upfront).
Have a `Fetcher` struct, and give it a `fetch_and_persist` function,
taking a `Fetch` as well as a desired name, and have it deal with all
the logic of persisting the PathInfos. It also returns a StorePathRef,
similar to the `.store_path()` method on a `Fetch` struct.
In a followup CL, we can extend KnownPaths to track fetches AND
derivations, and then use `Fetcher` when we need to do IO into that
store path.
Change-Id: Ib39a96baeb661750a8706b461f8ba4abb342e777
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11500
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Implement a first pass at the fetchTarball builtin.
This uses much of the same machinery as fetchUrl, but has the extra
complexity that tarballs have to be extracted and imported as store
paths (into the directory- and blob-services) before hashing. That's
reasonably involved due to the structure of those two services.
This is (unfortunately) not easy to test in an automated way, but I've
tested it manually for now and it seems to work:
tvix-repl> (import ../. {}).third_party.nixpkgs.hello.outPath
=> "/nix/store/dbghhbq1x39yxgkv3vkgfwbxrmw9nfzi-hello-2.12.1" :: string
Co-authored-by: Connor Brewster <cbrewster@hey.com>
Change-Id: I57afc6b91bad617a608a35bb357861e782a864c8
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11020
Autosubmit: aspen <root@gws.fyi>
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
This is especially nice, as it allows us getting rid of the other
versions of proc-macro2 and quote.
Change-Id: I9fdd012ee6c0ded3e18ec30504b83ac2032d1390
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11474
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Use a proper graph library to ensure all nodes are reachable from the
root.
We had a bit of that handrolled during add(), as well as later, which
had an annoying bug:
Redundant nodes were omitted during insert, but when returning the list
during finalize, we did not properly account they need to be introduced
before their parents are sent.
We now simply populate a petgraph DiGraph during insert (skipping
inserting nodes we already saw), and use petgraph's DfsPostOrder to
traverse the graph during finalize.
If the number of returned indices equals the total number of nodes in
the graph, all nodes are reachable from the root, we can consume the
graph and return the nodes as a vec, in the same order as the traversal
(and insertion).
Providing a regression test for the initial bug is challenging, as the
current code uses a bunch of HashSets. I manually tested ingesting a
full NixOS closure using this mechanism (via gRPC, which exposes this
problem, as it validates twice), and it now works.
Change-Id: Ic1d5e3e981f2993cc08c5c6b60ad895e578326dc
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11418
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Tested-by: BuildkiteCI
This exposes `user.tvix.castore.{blob,directory}.digest` xattr keys for
files and directories:
```
❯ getfattr -d /tmp/tvix/06jrrv6wwp0nc1m7fr5bgdw012rfzfx2-nano-7.2-info
getfattr: Removing leading '/' from absolute path names
user.tvix.castore.directory.digest="b3:SuYDcUM9RpWcnA40tYB1BtYpR0xw72v3ymhKDQbBfe4="
❯ getfattr -d /tmp/tvix/156a89x10c3kaby9rgf3fi4k0p6r9wl1-etc-shells
getfattr: Removing leading '/' from absolute path names
user.tvix.castore.blob.digest="b3:pZkwZoHN+/VQ8wkaX0wYVXZ0tV/HhtKlSqiaWDK7uRs="
```
It's currently mostly used for debugging, though it might be useful for
tvix-castore-aware syncing programs using the filesystem too.
Change-Id: I26ac3cb9fe51ffbf7f880519f26741549cb5ab6a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11422
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
Reviewed-by: Brian Olsen <me@griff.name>
This allows copying individual store paths from the local /nix/store
into tvix-store.
As we don't support getting this information from Nix yet, we currently
expect metadata to be provided externally:
Nix' `exportReferencesGraph` feature contains pretty much all data we
need for this. Expect a list of this information at a well-known key
(`closure`, similar to nixpkgs' `pkgs/build-support/binary-cache/
default.nix`).
We currently simply upload all store paths sequentially, without any
parallelism or awareness in how the reference graph looks like.
As long as the connected stores don't enforce this, this is fine, at
least for now.
Change-Id: Ib83c998465adddfdb110db994843c44e26b3d3d8
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11397
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
Autosubmit: flokli <flokli@flokli.de>
There's little reason to keep the nar writer using Async{Read,Write}
traits from futures, while everything else async in tvix (and
nix-compat) uses tokio.
Change-Id: I8cd1efcd0dd5bb76471de997603c7b701a5095de
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11391
Tested-by: BuildkiteCI
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Reviewed-by: Brian Olsen <me@griff.name>
Move everything bytes-related into its own module, and re-export
both bytes and primitive in a flat space from wire/mod.rs.
Expose this if a `wire` feature flag is set. We only have `async` stuff
in here.
Change-Id: Ia4ce4791f13a5759901cc9d6ce6bd6bbcca587c7
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11389
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
Reviewed-by: Brian Olsen <me@griff.name>
Put behind the "cloud" backend, like in the `tvix-castore` crate.
Change-Id: Ib38d198baf11ab2a4b6dc405121676147c424611
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11362
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Tested-by: BuildkiteCI
This adds a Directory service using
https://cloud.google.com/bigtable/docs/ as a K/V store.
Directory (closures) are put in individual keys.
We don't do any bucketed upload of directory closures (yet), as castore/
fs does query individually, does not request recursively (and buffers).
This will be addressed by store composition at some point.
Change-Id: I7fada45bf386a78b7ec93be38c5f03879a2a6e22
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11212
Tested-by: BuildkiteCI
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Autosubmit: flokli <flokli@flokli.de>
This deals with writing byte packets of larger sizes to an underlying
AsyncWrite.
Its constructor receives the expected size. It also deals with writing
padding if flush/shutdown is called after writing all the payload.
Change-Id: I8acbf992467f3862ffb8c7d669e8c0c8eced14c1
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11355
Reviewed-by: picnoir picnoir <picnoir@alternativebit.fr>
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: Brian Olsen <me@griff.name>
Tested-by: BuildkiteCI
This introduces rstest-based tests. We also add fixtures for creating
some BlobService / DirectoryService out of thin air.
To test a PathInfoService, we don't really care too much about its
internal storage - ensuring they work is up to the castore tests.
Change-Id: Ia62af076ef9c9fbfcf8b020a781454ad299d972e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11272
Tested-by: BuildkiteCI
Reviewed-by: Connor Brewster <cbrewster@hey.com>
This creates test scenarios (using the DirectoryService trait) that we
want all DirectoryService implementations to pass.
Some of these tests are ported from proto::tests::grpc_directoryservice,
which tested this on the gRPC interface (rather than the trait),
some others ensure certain behaviour for which we only recently
introduced general checking logic (through ClosureValidator).
We also borrow some code related to setting up a gRPC DirectoryService
client (connecting to a server exposing a in-memory DiretoryService)
from castore::utils, this will be deleted once it's all ported over.
Change-Id: I6810215a76101f908e2aaecafa803c70d85bc552
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11247
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Tested-by: BuildkiteCI
We had max_level_debug, not max_level_trace, so it was not possible to
have `RUST_LOG` emit trace-level values (at least for our crates).
release_max_level_info still controls the level for release builds, so
that's not affected.
Change-Id: I1f127ab63f13b1622d6616e06759a9414d9bb201
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11208
Tested-by: BuildkiteCI
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Autosubmit: flokli <flokli@flokli.de>
This controls whether tvix-castore has support for various cloud
backends or not.
Use this to control the set of feature flags for the object_store
backend, and only enable the aws, azure and gcp ones if it's set.
In the future this can be used to enable/disable other cloud backends
too.
Without feature flags, `object_store` already supports the `InMemory`
and `LocalFilesystem` backends, and we also want to unconditionally
enable the `http` one. Make sure at least the construction of these
services is covered in the tests.
Similarly, the tvix-store crate, which provides the tvix-store CLI has a
`cloud` feature flag too (defaulting to enabled).
Change-Id: I9fb9c87b740e7dc83f8ff7a0862905d036d513f2
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11204
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
This allows selectively increasing the log level for only parts of the
stack.
For example, the following RUST_LOG env var enables "tracing" level
logging for `tvix_store` and `tvix_castore`, while keeping it at "info"
for the rest of the stack:
export RUST_LOG='info,tvix_store=trace,tvix_castore=trace'
It only affects logs, not traces (if enabled).
Change-Id: Ib936bd132a405f216e75c843db83fbd71d20a18a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11182
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Reviewed-by: Connor Brewster <cbrewster@hey.com>
This bumps tonic and surrounding crates to 0.11.x.
We added support for tonic 0.11.x into tokio-listener
(https://github.com/vi/tokio-listener/pull/4), so that's bumped as well.
Change-Id: Icfade5894403228299836fefb21b2f9ae59dbebb
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11156
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
This brings some initial Nix wire format parsing code, used in the nix
daemon protocol, remote store/builder protocol, as well as the NAR
format itself (note we already have more specialized code for the last
one).
Thanks to embr, this code already exists, in
https://codeberg.org/gorgon/gorgon/src/branch/main/nix-daemon/src/wire.rs,
and we can vendor it into here, as EUPL is compatible with GPL (in that
direction).
The code uses the tokio::io Reader and Writer traits, not the ones from
the `futures` crate, as they provide some more convenient `read_u64_le`
functions.
More application-specific parsing code, as well as code to read strings,
or bytes are left out for now, as we want to be be more restrictive
w.r.t allowed max sizes, and need to parse bytes, not strings.
The code slightly diverges, as we have clippy looped into CI.
`Ok(…?)` can be turned into just the inner expression, and
some .and_then can be expressed in a simpler fashion.
Change-Id: Ie3adcb485e9d66786673b1962a08d4e5df3781d9
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11148
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Reviewed-by: picnoir picnoir <picnoir@alternativebit.fr>
This uses the `object_store` crate to expose a tvix-castore BlobService
backed by object storage.
It's using FastCDC to chunk blobs into smaller chunks when writing to
it.
These are exposed at the .chunks() method.
Change-Id: I2858c403d4d6490cdca73ebef03c26290b2b3c8e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11076
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Tested-by: BuildkiteCI
Reviewed-by: Brian Olsen <me@griff.name>
Add a new AsyncRead wrapper, DecompressedReader, that wraps an
underlying AsyncRead, but sniffs the magic bytes at the start of the
stream to determine which compression format is being used out of the
three that are supported by builtins.fetchTarball, and switches to the
correct decompression algorithm adapter dynamically.
This will be used in the implementation of builtins.fetchTarball
Change-Id: I892a4683d5c93e67d4c173f3d21199bdc6605922
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11019
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Implement the fetchurl builtin, and lay the groundwork for implementing
the fetchTarball builtin (which works very similarly, and is implemented
using almost the same code in C++ nix).
An overview of how this works:
1. First, we check if the store path that *would* result from the
download already exists in the store - if it does, we just return
that
2. If we need to download the URL, TvixStoreIO has an `http_client:
reqwest::Client` field now which we use to make the request
3. As we're downloading the blob, we hash the data incrementally into a
SHA256 hasher
4. We compare the hash against the expected hash (if any) and bail out
if it doesn't match
5. Finally, we put the blob in the store and return the store path
Since the logic is very similar, this commit also implements a *chunk*
of `fetchTarball` (though the actual implementation will likely include
a refactor to some of the code reuse here).
The main thing that's missing here is caching of downloaded blobs when
fetchurl is called without a hash - I've opened b/381 to track the TODO
there.
Adding the `SSL_CERT_FILE` here is necessary to teach reqwest how to
load it during tests - see 1c16dee20 (feat(tvix/store): use reqwests'
rustls-native-roots feature, 2024-03-03) for more info.
Change-Id: I83c4abbc7c0c3bfe92461917e23d6d3430fbf137
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11017
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
Autosubmit: aspen <root@gws.fyi>
Otherwise this causes conflicts when adding a more recent zstd version
to another crate.
Change-Id: I02d1b1a37f18711c0969cb04e761e967fc4655dc
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11100
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Tested-by: BuildkiteCI
This makes reqwest honor `SSL_CERT_FILE` - previously it was using the
chain bundled in webpki-roots.
`object_store` pulls in `reqwest` with this feature, and the cargo
solver will enable that feature globally as soon as we pull it in, as it
assumes features are additive.
This requires setting `SSL_CERT_FILE` when running tests, otherwise
they'll fail with the unhelpful "NotFound" error.
This was quite some fun to debug, why adding `object_store` to
tvix-castore suddenly made tvix-store tests fail!
Change-Id: I64fc82b4d994715480efdb1ffecb279716456ab9
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11090
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
This allows calling .into() to get a B3Digest.
Change-Id: I6e63b496413cd00d84acfcd15c7de0f64c79721f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11086
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
Implements md5, sha1, sha256 and sha512 using the related crates from
the RustCrypto hashes project (https://github.com/RustCrypto/hashes)
Change-Id: I00730dea44ec9ef85309edc27addab0ae88814b8
Reviewed-on: https://cl.tvl.fyi/c/depot/+/11005
Tested-by: BuildkiteCI
Reviewed-by: aspen <root@gws.fyi>
We add a new set of builtins called `import_builtins`, which
will contain import-related builtins, such as `builtins.path` and
`builtins.filterSource`. Both can import paths into the store, with
various knobs to alter the result, e.g. filtering, renaming, expected
hashes.
We introduce `filtered_ingest` which will drive the filtered ingestion
via the Nix function via the generator machinery, and then we register
the root node to the path info service inside the store.
`builtins.filterSource` is very simple, `builtins.path` is a more
complicated model requiring the same logic albeit more sophisticated
with name customization, file ingestion method and expected SHA-256.
Change-Id: I1083f37808b35f7b37818c8ffb9543d9682b2de2
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10654
Autosubmit: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
That's one possible abstraction to drive builds.
Whenever IO into a store path is requested, we look up the root node,
and in case we don't have it in PathInfoService, but KnownPaths gives us
a Derivation for that output path, trigger a build and await the result.
This recursively might trigger builds for parent paths if they haven't
been built yet.
Another option would be to simply expose a PathInfoService interface for
a builder too, and loop all building into IO via PathInfoService
composition - but let's start with something.
Note tvix-cli doesn't have a configurable BuildService yet, it's plugged
to the DummyBuildService, so whenever it needs to do a build, it'll fail,
but that's how it can be provoked:
```
(builtins.readFile (import <nixpkgs> {}).hello.outPath + "/bin/hello")
[…]
error[E029]: I/O error: /nix/store/cg8a576pz2yfc1wbhxm1zy4x7lrk8pix-hello-2.12.1: builds are not supported with DummyBuildService
--> [code]:1:2
|
1 | (builtins.readFile (import <nixpkgs> {}).hello.outPath + "/bin/hello")
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
```
Note how this fails, while pure output path calculation
(`(import <nixpkgs> {}).hello.outPath + "/bin/hello")`) still succeeds.
Change-Id: Id2075d8a2b18554d0dd608b4b29146a8cd411e7f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10793
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
No otlp yet, this simply gives us structured log output and a cli
argument for the log level.
Change-Id: Ifaa60bae419640e92baebb6ee59eedd775c769c2
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10853
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
Autosubmit: flokli <flokli@flokli.de>
This is was dropped accidentially, but we actually want to run tests in
debug mode, to check for overflows.
Change-Id: Ic0cf2d4a263bfd71f3453ed335d8be9dacd945e8
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10932
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
C++ nix uses C-style zero-terminated char pointers to represent strings
internally - however, up to this point, tvix has used Rust `String` and
`str` for string values. Since those are required to be valid utf-8, we
haven't been able to properly represent all the string values that Nix
supports.
To fix that, this change converts the internal representation of the
NixString struct from `Box<str>` to `BString`, from the `bstr` crate -
this is a wrapper around a `Vec<u8>` with extra functions for treating
that byte vector as a "morally string-like" value, which is basically
exactly what we need.
Since this changes a pretty fundamental assumption about a pretty core
type, there are a *lot* of changes in a lot of places to make this work,
but I've tried to keep the general philosophy and intent of most of the
code in most places intact. Most notably, there's nothing that's been
done to make the derivation stuff in //tvix/glue work with non-utf8
strings everywhere, instead opting to just convert to String/str when
passing things into that - there *might* be something to be done there,
but I don't know what the rules should be and I don't want to figure
them out in this change.
To deal with OS-native paths in a way that also works in WASM for
tvixbolt, this also adds a dependency on the "os_str_bytes" crate.
Fixes: b/189
Fixes: b/337
Change-Id: I5e6eb29c62f47dd91af954f5e12bfc3d186f5526
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10200
Reviewed-by: tazjin <tazjin@tvl.su>
Reviewed-by: flokli <flokli@flokli.de>
Reviewed-by: sterni <sternenseemann@systemli.org>
Autosubmit: aspen <root@gws.fyi>
Tested-by: BuildkiteCI
TvixStoreIO triggers builds whenever IO into a not-yet-built store path
is requested, if it knows how to build that path.
Change-Id: If30e9db6be2f2a30cbc9d0576f357f3ecfa0d35a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10645
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
This is an additional test suite on the top of the Nix ones
for context strings matters.
It already smoked out multiple mistakes and potential bugs and non-deterministic result from the evaluator.
It uses a similar technology as the one in the tvix-eval albeit we instantiate a fully fledged evaluator
with in-memory store.
We copy the files instead of symlinking them because crates are built in
isolation, so symlinks cannot work.
Change-Id: I63ae225ce4f83c6e2c8ccd60d779c2f8eb9d08fb
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10619
Autosubmit: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
Previously, we were assembling very naively an attribute set composed of context we saw.
But it was forgetting that `"${drv}${drv.drvPath}"` would contain 2 contexts with the same key, but
with different values, one with `outputs = [ "out" ];` and `allOutputs = true;`.
Following this reasoning and comparing with what Nix does, we ought to merge underlying values systematically.
Hence, I bring `itertools` to perform a group by on the key and merge everything on the fly, it's not
beautiful but it's the best I could find, notice that I don't use
`group_by` but I talk about group by, that is, because `group_by` is a
`group_by_consecutive`, see
https://github.com/rust-itertools/itertools/issues/374.
Initially, I tried to do it without a `into_grouping_map_by`, it was akin to assemble the final `NixAttrs` directly,
it was less readable and harder to pull out because we don't have a lot of in-place mutable functions on
our data structures.
Change-Id: I9933c9bd88ffe04de50dda14f21879b60d8b8cd4
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10620
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
