This provides an explicit API for call-fail-retry-succeed evaluation
flows, such as currently used in NixOps4.
An alternative design would simply reset the `Value` to the original
thunk instead of `tFailed` under the condition of catching a
`RecoverableEvalError`.
That is somewhat simpler, but I believe the presence of `tFailed` is
beneficial for possible use in the repl; being able to show the error
sooner, without re-evaluation.
The hasPos method and isEmpty function are required in order to avoid
an include loop.
This uses `gc_cleanup` to call the exception_ptr destructor when the
`Failed` is collected.
I don't know exactly how bad it is to deallocate `std::exception_ptr`
without destruction, but I guess it ranges from small leak to UB and crash.
In the multithreaded evaluator, it's possible for multiple threads to
wait on the same thunk. If evaluation of the thunk results in an
exception, the waiting threads shouldn't try to re-force the thunk.
Instead, they should rethrow the same exception, without duplicating
any work.
Therefore, there is now a new value type `tFailed` that stores an
std::exception_ptr. If evaluation of a thunk/app results in an
exception, `forceValue()` overwrites the value with a `tFailed`. If
`forceValue()` encounters a `tFailed`, it rethrows the exception. So
you normally never need to check for failed values (since forcing them
causes a rethrow).
This implements a special back-compat shim to specifically allow
unbracketed IPv6 addresses in store references. This is something
that is relied upon in the wild and the old parsing logic accepted
both ways (brackets were optional). This patch restores this behavior.
As always, we didn't have any tests for this.
Addresses #13937.
`perf c2c` shows a lot of cacheline conflicts between purely read-only
Store methods (like `parseStorePath()`) and the Sync classes. So
allocate pathInfoCache separately to avoid that.
Calling `drainFD()` will hang if another process has the write side
open, since then the child won't get an EOF. This can happen if we
have multiple threads doing a build, since in that case another thread
may fork a child process that inherits the write side of the first
thread.
We could set O_CLOEXEC on the write side (using pipe2()) but it won't
help here since we don't always do an exec() in the child, e.g. in the
case of builtin builders. (We need a "close-on-fork", not a
"close-on-exec".)
Since the only construction and push_back() calls
to Bindings happen through the `BindingsBuilder` [1] we don't
need to keep `capacity` around on the heap anymore. This saves 8 bytes
(because of the member alignment padding)
per one Bindings allocation. This isn't that much, but it does
save significant memory.
This also shows that the Bindings don't necessarily have to
be mutable, which opens up opportunities for doing small bindings
optimization and storing a 1-element Bindings directly in Value.
For the following scenario:
nix-env --query --available --out-path --file ../nixpkgs --eval-system x86_64-linux
(nixpkgs revision: ddcddd7b09a417ca9a88899f4bd43a8edb72308d)
This patch results in reduction of `sets.bytes` 13115104016 -> 12653087640,
which amounts to 462 MB less bytes allocated for Bindings.
[1]: Not actually, `getBuiltins` does mutate bindings, but this is pretty
inconsequential and doesn't lead to problems.
This is relied upon (specifically the `local` store) by existing
tooling [1] and we broke this in 3e7879e6df (which
was first released in 2.31).
To lessen the scope of the breakage we should not normalize "auto" references
and explicitly specified references like "local" or "daemon". It also makes
sense to canonicalize local://,daemon:// to be more compatible with prior
behavior.
[1]: 05e1b3cba2/lib/NOM/Builds.hs (L60-L64)
Exactly why is is correct is a little subtle, because sometimes the
worker is owned by the worker. But the commit message in
e437b08250 explained the situation well
enough: I made that commit message part of the ABI docs, and now it
should be understandable to the next person.
Do this with a new `useHook` boolean we carefully make sure is set in
all cases. This change isn't really worthwhile by itself, but it allows
us to make further refactors (see later commits) which are
well-motivated.
When useMaster is true, startMaster() acquires the state lock, then
calls isMasterRunning(), which calls addCommonSSHOpts(), which tries
to acquire the state lock again, causing a deadlock.
The solution is to move tmpDir out of the state. It doesn't need to be
there in the first place because it never changes.
On macOS, poll() is fundamentally broken for HUP detection. It loses event
subscriptions when EVFILT_READ fires without matching the requested events
in the pollfd. This causes daemon processes to linger after client disconnect.
This commit replaces poll() with kqueue on macOS, which is what poll()
uses internally but without the bugs. The kqueue implementation uses
EVFILT_READ which works for both sockets and pipes, avoiding EVFILT_SOCK
which only works for sockets.
On Linux and other platforms, we continue using poll() with the standard
POSIX behavior where POLLHUP is always reported regardless of requested events.
Based on work from the Lix project (https://git.lix.systems/lix-project/lix)
commit 69ba3c92db3ecca468bcd5ff7849fa8e8e0fc6c0
Fixes: https://github.com/NixOS/nix/issues/13847
Related: https://git.lix.systems/lix-project/lix/issues/729
Apple bugs: rdar://37537852 (poll), FB17447257 (poll)
Co-authored-by: Jade Lovelace <jadel@mercury.com>
Now that Symbols are statically allocated at compile time with known IDs,
we can use switch statements instead of if-else chains for Symbol comparisons.
This provides better performance through compiler optimizations like jump tables.
Changes:
- Add public getId() method to Symbol class to access the internal ID
- Convert if-else chains comparing Symbol values to switch statements
in primops.cc's derivationStrictInternal function
- Simplify control flow by removing the 'handled' flag and moving the
default attribute handling into the switch's default case
The static and runtime Symbol IDs are guaranteed to match by the
copyIntoSymbolTable implementation which asserts this invariant.
Co-authored-by: John Ericson <git@JohnEricson.me>
In b70d22b `mkStringNoCopy()` was renamed to
`mkString()`, but this is a bit risky since in code like
vStringRegular.mkString("regular");
we want to be sure that the right overload is picked. (This is
especially problematic since the overload that takes an
`std::string_view` *does* allocate.) So let's be explicit.
(Rebased from https://github.com/NixOS/nix/pull/11551)
I (@Ericson2314) messed up. We were supposed to test the status quo
before landing any new chnages, and also there is one change that is not
quite right (relative paths).
I am reverting for now, and then backporting the test suite to the old
situation.
This reverts commit 04ad66af5f.
Git URI can also support scp style links similar to git itself.
This change augments the function fixGitURL to better handle the scp
style urls through a minimal parser rather than regex which has been
found to be brittle.
* Support for IPV6 added
* New test cases added for fixGitURL
* Clearer documentation on purpose and goal of function
* More `std::string_view` for performance
* A few more URL tests
Fixes#5958
The URL should not be normalized before handing it off to cURL, because
builtin fetchers like fetchTarball/fetchurl are expected to work with
arbitrary URLs, that might not be RFC3986 compliant. For those cases
Nix should not normalize URLs, though validation is fine. ParseURL and
cURL are supposed to match the set of acceptable URLs, since they implement
the same RFC.
This adds regression tests for fromTOML overflow/underflow behavior.
Previous versions of toml11 used to saturate, but this was never an
intended behavior (and Snix/Nix 2.3/toml11 >= 4.0 validate this).
(cherry picked from Lix [1,2])
[1]: 7ee442079d
[2]: 4de09b6b54
The motivation for this change is two-fold:
1. Commonly used Symbol values can be referred to
quite often and they can be assigned at compile-time
rather than runtime.
2. This also unclutters EvalState constructor, which was
getting very long and unreadable.
Spiritually similar to https://gerrit.lix.systems/c/lix/+/2218,
though that patch doesn't allocate the Symbol at compile time.
Co-authored-by: eldritch horrors <pennae@lix.systems>
Looking at perf:
0.21 │ push %rbp
0.99 │ mov %rsp,%rbp
│ push %r15
0.25 │ push %r14
│ push %r13
0.49 │ push %r12
0.66 │ push %rbx
1.23 │ lea -0x10000(%rsp),%r11
0.23 │ 15: sub $0x1000,%rsp
1.01 │ orq $0x0,(%rsp)
59.12 │ cmp %r11,%rsp
0.27 │ ↑ jne 15
Seems like 64K is too much to have on the stack for each invocation, considering
that only a minuscule number of allocations are actually larger than 4K.
There's actually no good reason this function should use so much stack space. Or
use small_string at all. Everything can be done in small chunks that don't require
any memory allocations and use up 2K bytes on the stack.
This patch also adds a microbenchmark for tracking the unparsing performance. Here
are the results for this change:
(Before)
BM_UnparseRealDerivationFile/hello 7275 ns 7247 ns 96093 bytes_per_second=232.136Mi/s
BM_UnparseRealDerivationFile/firefox 40538 ns 40376 ns 17327 bytes_per_second=378.534Mi/s
(After)
BM_UnparseRealDerivationFile/hello 3228 ns 3218 ns 215671 bytes_per_second=522.775Mi/s
BM_UnparseRealDerivationFile/firefox 39724 ns 39584 ns 17617 bytes_per_second=386.101Mi/s
This translates into nice evaluation performance improvements (compared to 18c3d2348f):
Benchmark 1: GC_INITIAL_HEAP_SIZE=8G old-nix/bin/nix-instantiate ../nixpkgs -A nixosTests.gnome --readonly-mode
Time (mean ± σ): 3.111 s ± 0.021 s [User: 2.513 s, System: 0.580 s]
Range (min … max): 3.083 s … 3.143 s 10 runs
Benchmark 2: GC_INITIAL_HEAP_SIZE=8G result/bin/nix-instantiate ../nixpkgs -A nixosTests.gnome --readonly-mode
Time (mean ± σ): 3.037 s ± 0.038 s [User: 2.461 s, System: 0.558 s]
Range (min … max): 2.960 s … 3.086 s 10 runs
Old versions of nix happily accepted a lot of weird flake references,
which we didn't have tests for, so this was accidentally broken in
c436b7a32a.
This patch restores previous behavior and adds a plethora of tests
to ensure we don't break this in the future.
These test cases are aligned with how 2.18/2.28 parsed flake references.
Starting from c436b7a32a
this used to lead to assertion failures like:
> std::string nix::ParsedURL::renderAuthorityAndPath() const: Assertion `path.empty() || path.front().empty()' failed.
This has the bugfix for the issue and regressions tests
so that this gets properly tested in the future.
This would print erroneous and misleading diagnostics like:
> error (ignored): error: '--arg' and '--argstr' are incompatible with flakes
When run with --expr/--file. Since this installable is used to get the
bash package it doesn't make sense to check this.
It is only done in the `force = true` case, and the only
`cleanupBuild(true)` call is right after where it used to be, so this
has the exact same behavior as before.
Calling `reset` on this `std::optional` field of `DerivationBuilderImpl`
is also what the (automatically created) destructor of
`DerivationBuilderImpl` will do. We should be making sure that the
derivation builder is cleaned up by the goal anyways, and if we do that,
then this `Finally` is no longer needed.
Before, had a very ugly `appendLogTailErrorMsg` callback. Now, we
instead have a `fixupBuilderFailureErrorMessage` that is just used by
`DerivationBuildingGoal`, and `DerivationBuilder` just returns the raw
data needed by this.
Now we have better separation of the core logic --- an integral part of
the store layer spec even --- from the goal mechanism and other
minutiae.
Co-authored-by: Jeremy Kolb <kjeremy@gmail.com>
See the new extensive doxygen in `url.hh`.
This fixes fetching gitlab: flakes.
Paths are now stored as a std::vector of individual path
segments, which can themselves contain path separators '/' (%2F).
This is necessary to make the Gitlab's /projects/ API work.
Co-authored-by: John Ericson <John.Ericson@Obsidian.Systems>
Co-authored-by: Sergei Zimmerman <sergei@zimmerman.foo>
I think this should be fine for repairing. If anything, it is better,
because it would be weird to "mark and output good" only for it to then
fail output checks.
Sadly we cannot unexpose `DerivationBuilder::killChild` yet, because
`DerivationBuildingGoal` calls it elsewhere, but we can at least haave a
better division of labor between the two destructors.
It's hard to tell if I changed any behavior, but if I did, I think I
made it better, because now we explicitly move stuff out of the chroot
(if we were going to) before trying to delete the chroot.
Do this to match `DerivationBuilder::deleteTmpDir`, which we'll want to
combine it with next.
Also chenge one caller from `deleteTmpDir(true)` to `cleanupBuild(true)`
now that this is done, because it will not make a difference.
This should be a pure refactor with no behavioral change.
Aftet the previous simplifications, there is no reason to catch the
error and immediately return it with a `std::variant` --- just let the
caller catch it instead.
Instead of that funny business, the fixed output checks are not put in
`checkOutputs`, with the other (newer) output checks, where they also
better belong. The control flow is reworked (with comments!) so that
`checkOutputs` also runs in the `bmCheck` case.
Not only does this preserve existing behavior of `bmCheck`
double-checking fixed output hashes with less tricky code, it also makes
`bmCheck` better by also double-checking the other output checks, rather
than just assuming they pass if the derivation is deterministic.
It's fine to set these worker flags a little later in the control flow,
since we'll be sure to reach those points in the error cases. And doing
that is much nicer than having these tangled callbacks.
I originally made the callbacks to meticulously recreate the exact
behavior which I didn't quite understand. Now, thanks to cleaning up the
error handling, I do understand what is going on, so I can be confident
that this change is safe to make.
Instead of passing them around separately, or doing finicky logic in a
try-catch block to recover them, just make `BuildError` always contain a
status, and make it the thrower's responsibility to set it. This is much
more simple and explicit.
Once that change is done, split the `done` functions of `DerivationGoal`
and `DerivationBuildingGoal` into separate success and failure
functions, which ends up being easier to understand and hardly any
duplication.
Also, change the handling of failures in resolved cases to use
`BuildResult::DependencyFailed` and a new message. This is because the
underlying derivation will also get its message printed --- which is
good, because in general the resolved derivation is not unique. One dyn
drv test had to be updated, but CA (and dyn drv) is experimental, so I
do not mind.
Finally, delete `SubstError` because it is unused.
The commit says it was added for CA testing --- manual I assume, since
there is no use of this in the test suite. I don't think we need it any
more, and I am not sure whether it was ever supposed to have made it to
`master` either.
This reverts commit 2eec2f765a.
We currently just use this during the build of a derivation, but there is no
reason we wouldn't want to use it elsewhere, e.g. to check the outputs
of someone else's build after the fact.
Moreover, I like pulling things out of `DerivationBuilder` that are
simple and don't need access to all that state. While
`DerivationBuilder` is unix-only, this refactor also make the code more
portable "for free".
The header is private, at Eelco's request.
With the migration to /nix/var/nix/builds we now have failing builds
when the derivation name is too long.
This change removes the derivation name from the temporary build to have
a predictable prefix length:
Also see: https://github.com/NixOS/infra/pull/764
for context.
This allows us to replace some very hacky and not correct string
concatentation in `HttpBinaryCacheStore`. It will especially be useful
with #13752, when today's hacks started to cause problems in practice,
not just theory.
Also make `fixGitURL` returned a `ParsedURL`.
• Updated input 'nixpkgs':
'github:NixOS/nixpkgs/cd32a774ac52caaa03bcfc9e7591ac8c18617ced?narHash=sha256-VtMQg02B3kt1oejwwrGn50U9Xbjgzfbb5TV5Wtx8dKI%3D' (2025-08-17)
→ 'github:NixOS/nixpkgs/d98ce345cdab58477ca61855540999c86577d19d?narHash=sha256-O2CIn7HjZwEGqBrwu9EU76zlmA5dbmna7jL1XUmAId8%3D' (2025-08-26)
This update contains d1266642a8722f2a05e311fa151c1413d2b9653c, which
is necessary for the TOML timestamps to get tested via nixpkgsLibTests job.
I need this for some `ParseURL` improvements, but I figure this is
better to send as its own PR.
I changed the tests willy-nilly to sometimes use
`std::list<std::string_view>` instead of `Strings` (which is
`std::list<std::string>`).
Co-Authored-By: Sergei Zimmerman <sergei@zimmerman.foo>
It is suppposed to be "post build" not "during the build" after all. Its
location now matches that for the hook case (see elsewhere in
`DerivationdBuildingGoal`).
It was in a try-catch before, and now it isn't, but I believe that it is
impossible for it to throw `BuildError`, which is sufficient for this
code motion to be correct.
Update src/libutil/windows/current-process.cc
Prefer `nullptr` over `NULL`
Co-authored-by: Sergei Zimmerman <sergei@zimmerman.foo>
Update src/libutil/unix/current-process.cc
Prefer C++ type casts
Co-authored-by: Sergei Zimmerman <sergei@zimmerman.foo>
Update src/libutil/windows/current-process.cc
Prefer C++ type casts
Co-authored-by: Sergei Zimmerman <sergei@zimmerman.foo>
Update src/libutil/unix/current-process.cc
Don't allocate exception
Co-authored-by: Sergei Zimmerman <sergei@zimmerman.foo>
It would have been nice to use std::views::enumerate here, but
it uses a signed difference type for the value_type:
> value_type = std::tuple<difference_type, ranges::range_value_t<Base>>
zip + iota has the same semantics as the code used to have, so there's
no behavior change here.
This is a nicer separation of concerns --- `DerivationBuilder` just
mounts the extra paths you tell it too, and the outside world is
responsible for making sure those extra paths make sense.
Since the closure only depends on global settings, and not
per-derivation information, we also have the option of moving this up
further and caching it across all local builds. (I only just realized
this after having done this refactor. I am not doing that change at this
time, however.)
Now, `DerivationBuilder` only concerns itself with `finalEnv` and
`extraFiles`, in straightforward unconditional code. All the fancy
desugaring logic is consolidated in `DerivationBuildingGoal`.
We should better share the pulled-out logic with `nix-shell`/`nix
develop`, which would fill in some missing features, arguably fixing
bugs.
I think this is a better separation of concerns. `DerivationBuilder`
doesn't need to to the final, query-heavy details about how these things
are constructed. It just operates on the level of "simple, stupid" files
and environment variables.
As much as I prefer rewriting the parsed rather than unparsed JSON for
elegance, this gets in the way of the separation of concerns that I am
trying to do.
As a practical matter, any rewriting that this did will also be done by
the second round of rewriting that remains below, so removing this code
should have no effect.
This is needed to rearrange include order, but I also think it is a good
thing anyways, as we seek to reduce the use of global settings variables
over time.
This avoids problems with older versions of Nix that don't put the
caches in WAL mode. That's generally not a problem, until you do something like
nix build --print-out-paths ... | cachix
which deadlocks because cachix tries to switch the caches to truncate
mode, which requires exclusive access. But the first process cannot
make progress because the cachix process isn't reading from the pipe.
With "truncate" mode, if we try to write to the database while another
process has an active write transaction, we'll block until the other
transaction finishes. This is a problem for the evaluation cache in
particular, since it uses long-running transactions.
WAL mode does not have this issue: it just returns "busy" right away,
so Nix will print
error (ignored): SQLite database '/home/eelco/.cache/nix/eval-cache-v5/...' is busy
and stop trying to write to the evaluation cache. (This was the
intended/original behaviour, see AttrDb::doSQLite().)
Compilers in nixpkgs have caught up and major distros
should also have recent enough compilers. It would be
nice to have newer features like more full featured
ranges and deducing this.
About time we upgraded our nixpkgs flake input. Ideally
we'd have automation to do this.
Flake lock file updates:
• Updated input 'nixpkgs':
'github:NixOS/nixpkgs/adaa24fbf46737f3f1b5497bf64bae750f82942e?narHash=sha256-qhFMmDkeJX9KJwr5H32f1r7Prs7XbQWtO0h3V0a0rFY%3D' (2025-05-13)
→ 'github:NixOS/nixpkgs/cd32a774ac52caaa03bcfc9e7591ac8c18617ced?narHash=sha256-VtMQg02B3kt1oejwwrGn50U9Xbjgzfbb5TV5Wtx8dKI%3D' (2025-08-17)
This caused RemoteStore::queryPathInfoUncached() to mark the
connection as invalid (see
RemoteStore::ConnectionHandle::~ConnectionHandle()), causing it to
disconnect and reconnect after every lookup of an invalid path. This
caused huge slowdowns in conjunction with
19f89eb684 and lazy-trees.
Period '.' is a special branch name in the gitsubmodule file which
represents the branch of the parent repository [1].
We add support for this by registering the ref of the InputAccessor to
be that of the parent input if '.' is encountered.
Fixes#13215
[1]: man gitmodules
lowdown >= 1.4.0 supports LOWDOWN_TERM_NORELLINK to render
absolute urls. This is useful, since we want to keep links to
web resources and such intact.
Fix description of `NIX_CONF_DIR`. It currently say that it defaults to `prefix/etc/nix`, which would mean `/nix/etc/nix` on default installations, and contradicts the description in `conf-file-prefix.md`.
This fix makes the description of `NIX_CONF_DIR` consistent with `conf-file-prefix.md`, assuming that the latter is correct.
Turns out we didn't have tests for some of the important behavior introduced
for flake reference fragments and url queries [1]. This is rather important
and is relied upon by existing tooling. This fixes up these exact cases before
handing off the URL to the Boost.URL parser.
To the best of my knowledge this implements the same behavior as prior regex-based
parser did [2]:
> fragmentRegex = "(?:" + pcharRegex + "|[/? \"^])*";
> queryRegex = "(?:" + pcharRegex + "|[/? \"])*";
[1]: 9c0a09f09f
[2]: https://github.com/NixOS/nix/blob/2.30.2/src/libutil/include/nix/util/url-parts.hh
It's about time we added debuggers to the dev-shell. Having it in build inputs
does some magic so pretty printers for standard library types work better.
Since this goal has no (goal-wide) notion of "wanted outputs" (we're
building the derivation, and thus making all outputs), we should have
`initialOutputs` for all outputs, and if we're missing one that's an
internal error caused by a bug in Nix.
Concretely, `DerivationBuildingGoal::gaveUpOnSubstitution` now clearly
does create `initialOutputs` for all outputs, whereas a few commits ago
that was not obvious, so I feel confident in saying that this invariant
that should be upheld, in fact is upheld.
`scatchOutputs` is initialized for every initial output, so the same
change to it follows for the same reasons.
This is just more honest, since we downcasted it to `LocalStore` in many
places. We had the downcast before because it wasn't needed in the hook
case, just the local building case, but now that `DerivationBuilder` is
separated and just does the building case, we have formalized the
boundary where the single downcast should occur.
No derivation goal type has a notion of variable wanted outputs any
more. They either want them all, or they just care about a single
output, in which case we would just store this information for the one
output in question.
We can cut out some gratuitous inhertence as follows:
- `MixStoreDirMethods` -> `StoreDirConfig`
- `StoreDirConfig` deleted because no longer needed. It is just folded
into `StoreConfig`.
- `StoreDirConfigBase` -> `StoreConfigBase` same trick still needed, but
now is for `StoreConfig` not `StoreDirConfig`
Here's how we got here:
1. I once factored out `StoreDirConfig` in #6236.
2. I factored out `MixStoreDirMethods` in #13154.
But, I didn't realize at point (2) that we didn't need `StoreDirConfig`
anymore, all uses of `StoreDirConfig` could instead be uses of
`MixStoreDirMethods`. Now I am doing that, and renaming
`MixStoreDirMethods` to just `StoreDirConfig` to reduce churn.
This leads to a use-after free, because staticOutputHashes returns a temporary
object that dies before we can do a `return *mOutputHash`.
This is most likely the cause for random failures in Hydra [1].
[1]: https://hydra.nixos.org/build/305091330/nixlog/2
Old code completely ignored query parameters and it seems ok to keep
that behavior. There's a lot of code out there that parses nix code
like nix-output-monitor and it can't parse messages like:
> copying path '/nix/store/wha2hi4yhkjmccqhivxavbfspsg1wrsj-source' from 'https://cache.nixos.org' to 'local://'...
Let's not break these tools without a good reason. This goes in line
with what other code does by ignoring parameters in logs.
The issue is just in detecting the shorthand notations for the store
reference - not in printing the url in logs.
By default the daemon opens a local store with ?path-info-cache-size=0,
so that leads to the erronenous 'local://'.
The problem with old code was that it used getUri for both the `diskCache`
as well as logging. This is really bad because it mixes the textual human
readable representation with the caching.
Also using getUri for the cache key is really problematic for the S3 store,
since it doesn't include the `endpoint` in the cache key, so it's totally broken.
This starts separating the logging / cache concerns by introducing a
`getHumanReadableURI` that should only be used for logging. The caching
logic now instead uses `getReference().render(/*withParams=*/false)` exclusively.
This would need to be fixed in follow-ups, because that's really fragile and
broken for some store types (but it was already broken before).
Move output result filtering logic and assert just into the branch where
it is not obviously a no op / meeting the assertion.
Add a comment too, while we are at it.
Now that `DerivationGoal::checkPathValidity` is legible, we can see that
it only sets `outputKnown`, and doesn't read it. Likewise, with
co-routines, we don't have tiny scopes that make local variables
difficult. Between these two things, we can simply have
`checkPathValidity` return what it finds, rather than mutate some state,
and update everyting to use local variables.
The same transformation could probably be done to the other derivation
goal types (which currently, unfortunately, contain their own
`checkPathValidity`s, though they are diverging, and we hope and believe
that they continue to diverge).
`Store::queryPartialDerivationOutputMap` is nothing but checking
statically-known output paths, and then `Store::queryRealisation`, and
we were doing both of those things already. Inline that and simplify,
again taking advantage of the fact that we only care about one output.
Rather than having store implementations return a free-form URI string,
have them return a `StoreReference`. This reflects that fact that this
method is supposed to invert `resolveStoreConfig`, which goes from a
`StoreReference` to some `StoreConfig` concrete derived class (based on
the registry).
`StoreConfig::getUri` is kept only as a convenience for the common case
that we want to immediately render the `StoreReference`.
A few tests were changed to use `local://` not `local`, since
`StoreReference` does not encode the `local` and `daemon` shorthands
(and instead desugars them to `local://` and `unix://` right away). I
think that is fine. `local` and `daemon` still work as input.
* Adds support for NIX_SSHOPTS
* Properly uses the parsed port from URL (fixes#13337)
* Don't guess the HTTP endpoint, use the response of git-lfs-authenticate
* Add an SSH Git LFS test
* Removed some unused test code
Clang refused to do a narrowing conversion in an initializer list:
```
local-keys.cc:56:90: note: insert an explicit cast to silence this issue
return name + ":" + base64::encode(std::as_bytes(std::span<const unsigned char>{sig, sigLen}));
^~~~~~
static_cast<size_type>( )
```
This addresses several changes from toml11 4.0 bump in
nixpkgs [1].
1. Added more regression tests for timestamp formats.
Special attention needs to be paid to the precision
of the subsecond range for local-time. Prior versions select the closest
(upwards) multiple of 3 with a hard cap of 9 digits.
2. Normalize local datetime and offset datetime to always
use the uppercase separator `T`. This is actually the issue
surfaced in [2]. This canonicalization is basically a requirement
by (a certain reading) of rfc3339 section 5.6 [3].
3. If using toml11 >= 4.0 also keep the old behavior wrt
to the number of digits used for subsecond part of the local-time.
Newer versions cap it at 6 digits unconditionally.
[1]: https://www.github.com/NixOS/nixpkgs/pull/331649
[2]: https://www.github.com/NixOS/nix/issues/11441
[3]: https://datatracker.ietf.org/doc/html/rfc3339
Current test suite doesn't cover the subsecond formatting at
all and toml11 is quite finicky with that. We should at the very
least test its behavior to avoid silent breakages on updates.
Previously `getUri` didn't include store query parameters,
`ssh-ng` didn't include any information at all and the local
store didn't have the path:
```
$ nix store info --store "local?root=/tmp/aaa&require-sigs=false"
Store URL: local
Version: 2.31.0
Trusted: 1
$ nix store info --store "ssh-ng://localhost?remote-program=nix-daemon"
Store URL: ssh-ng://
Version: 2.31.0
Trusted: 1
$ nix store info --store "ssh://localhost?remote-program=nix-store"
Store URL: ssh://localhost
```
This commit changes this to:
```
$ nix store info --store "local?root=/tmp/aaa&require-sigs=false"
Store URL: local?require-sigs=false&root=/tmp/aaa
Version: 2.31.0
Trusted: 1
$ nix store info --store "ssh-ng://localhost?remote-program=nix-daemon"
Store URL: ssh-ng://localhost?remote-program=nix-daemon
Version: 2.31.0
Trusted: 1
$ nix store info --store "ssh://localhost?remote-program=nix-store"
Store URL: ssh://localhost?remote-program=nix-store
```
The clang-analyzer incorrectly flags a use-after-free for GC-managed objects
when used with std::unique_ptr. Since NixRepl inherits from gc, its memory
is properly managed by Boehm GC and this is a false positive.
Added NOLINTNEXTLINE directive to suppress the warning.
Fixes usage of `#` symbol in the reference name.
This also seems to identify several deficiencies in the libgit2 refname
validation code wrt to DEL symbol and a singular `@` symbol [1].
[1]: https://git-scm.com/docs/git-check-ref-format#_description
This change was necessary when we were using `nix flake check` for CI
(see 6db6b269ed). Now this is not really
necessary, because we are running the checks in a much saner way, that
doesn't use up too much memory for evaluation.
Sometime ago we lost the coverage job in the midst of
meson migration. Until we have something like codecov
it'd be very useful to restore this job with the html
reports and historical metrics.
As a bonus we get more coverage metrics by switching to
LLVM tooling from LCOV.
The implicit dependency on refLength (which is the StorePath::HashLen)
is not good. Also the companion tests and benchmarks are already in libstore-tests.
This doesn't seem to be used anywhere at the moment.
It might be used out-of-tree, but this is a small convenience
function that is not worth keeping without in-tree usage.
With this I'm able to do a fresh config + meson test with all dependencies
correctly propagated.
Co-authored-by: Sergei Zimmerman <sergei@zimmerman.foo>
This was removed in https://github.com/NixOS/nix/pull/11152. However,
we need it for the multi-threaded evaluator, because otherwise Boehm
GC will crash while scanning the thread stack:
#0 GC_push_all_eager (bottom=<optimized out>, top=<optimized out>) at extra/../mark.c:1488
#1 0x00007ffff74691d5 in GC_push_all_stack_sections (lo=<optimized out>, hi=<optimized out>, traced_stack_sect=0x0) at extra/../mark_rts.c:704
#2 GC_push_all_stacks () at extra/../pthread_stop_world.c:876
#3 GC_default_push_other_roots () at extra/../os_dep.c:2893
#4 0x00007ffff746235c in GC_mark_some (cold_gc_frame=0x7ffee8ecaa50 "`\304G\367\377\177") at extra/../mark.c:374
#5 0x00007ffff7465a8d in GC_stopped_mark (stop_func=stop_func@entry=0x7ffff7453c80 <GC_never_stop_func>) at extra/../alloc.c:875
#6 0x00007ffff7466724 in GC_try_to_collect_inner (stop_func=0x7ffff7453c80 <GC_never_stop_func>) at extra/../alloc.c:624
#7 0x00007ffff7466a22 in GC_collect_or_expand (needed_blocks=needed_blocks@entry=1, ignore_off_page=ignore_off_page@entry=0, retry=retry@entry=0) at extra/../alloc.c:1688
#8 0x00007ffff746878f in GC_allocobj (gran=<optimized out>, kind=<optimized out>) at extra/../alloc.c:1798
#9 GC_generic_malloc_inner (lb=<optimized out>, k=k@entry=1) at extra/../malloc.c:193
#10 0x00007ffff746cd40 in GC_generic_malloc_many (lb=<optimized out>, k=<optimized out>, result=<optimized out>) at extra/../mallocx.c:477
#11 0x00007ffff746cf35 in GC_malloc_kind (bytes=120, kind=1) at extra/../thread_local_alloc.c:187
#12 0x00007ffff796ede5 in nix::allocBytes (n=<optimized out>, n=<optimized out>) at ../src/libexpr/include/nix/expr/eval-inline.hh:19
This is because it will use the stack pointer of the coroutine, so it
will scan a region of memory that doesn't exist, e.g.
Stack for thread 0x7ffea4ff96c0 is [0x7ffe80197af0w,0x7ffea4ffa000)
(where 0x7ffe80197af0w is the sp of the coroutine and 0x7ffea4ffa000
is the base of the thread stack).
We don't scan coroutine stacks, because currently they don't have GC
roots (there is no evaluation happening in coroutines). So there is
currently no need to restore the other parts of the original patch,
such as BoehmGCStackAllocator.
I had not wanted to cause unncessary churn before, but now that we've
bitten the bullet with the Big Reformat, I feel it is the right time.
Future readers will appreciate that the declarations and definitions
files are one-to-one as they should be, and `store-api.cc` is good to
shrink in any event.
I don't think there are outstanding PRs changing this code either. (I
had some for a while, but they are all merged.)
- No more private constructor that is kinda weird
- Two new static functions, `baseFromSize` and `baseFromSize`, that do
one thing, and one thing only (simple).
- Two `Hash::parse*` that previously used the private constructor now
can use these two functions directly.
- The remaining `Hash::parseAny*` methods, which are inherently more
complex, are written in terms of a `parseAnyHelper` static function
which is also complex, but keeps the complexity in one spot.
This patch allows users to specify the connection port
in the store URLS like so:
```
nix store info --store "ssh-ng://localhost:22" --json
```
Previously this failed with: `error: failed to start SSH connection to 'localhost:22'`,
because the code did not distinguish the port from the hostname. This
patch remedies that problem by introducing a ParsedURL::Authority type
for working with parsed authority components of URIs.
Now that the URL parsing code is less ad-hoc we can
add more long-awaited fixes for specifying SSH connection
ports in store URIs.
Builds upon the work from bd1d2d1041.
Co-authored-by: Sergei Zimmerman <sergei@zimmerman.foo>
Co-authored-by: John Ericson <John.Ericson@Obsidian.Systems>
Prior to this patch, mode 0444 is not updated to 0555 for directories.
That means for instance 0554 is canonicalized, but not 0444.
We don't believe this has any implications for backwards compatibility,
because directories do not have permissions in NAR format and so are
always 0555 after deserialization, and store paths with wrong
permissions can’t be copied to another host.
Co-authored-by: Robert Hensing <robert@roberthensing.nl>
Make it separate from Hash, since other things can be base-encoded too.
This isn't really needed for Nix, but it makes the code easier to read
e.g. for someone reimplementing this stuff in a different language. (Of
course, Base16/Base64 should be gotten off-the-shelf, but now the hash
code, which is more bespoke, is less cluttered with the parts that would
be from some library.)
Many reimplementations of "Nix32" and our hash type already exist, so
this cleanup is coming years too late, but I say better late than never
/ it is always good to nudge the code in the direction of being a
"living spec".
Co-authored-by: Sergei Zimmerman <sergei@zimmerman.foo>
Having a State class in the nix namespace is asking
for ODR trouble. This class is already private to the
translation unit, let's move it into an anonymous namespace.
There have been prior concerns about reogranizing the repo, but
this seems like a trivial simplification which will not interfere
with either our packaging or the modular builds in nixpkgs.
This adds the necessary infrastructure to collect
code coverage in CI, which could be useful to look
at munually or track consistently via something like
codecov.
Co-authored-by: Jade Lovelace <lix@jade.fyi>
This keeps things fast by making the function inline, but also prevents
people from having to know about the `0xFF` implementation detail
directly, instead making one go through a `std::optional` (which could be
fused away with a sufficiently smart compiler).
Additionally, the base "nix32" implementation is moved to its own header
file pair, as it is logically distinct and prior to the `Hash` data
type. It would probably be nice to do this with all the hash format
implementations.
Some binary caches (incorrectly) use this header to indicate lack of
compression, inspired by the valid "identity" token in the
"Accept-Encoding" header.
This benchmark should provide a relatively realistic
scenario for reference scanning.
As shown by the following results, reference scanning code
is already plenty fast and is definitely not a bottleneck:
```
BM_RefScanSinkRandom/10000 1672 ns 1682 ns 413354 bytes_per_second=5.53691Gi/s
BM_RefScanSinkRandom/100000 11217 ns 11124 ns 64341 bytes_per_second=8.37231Gi/s
BM_RefScanSinkRandom/1000000 205745 ns 204421 ns 3360 bytes_per_second=4.55591Gi/s
BM_RefScanSinkRandom/5000000 1208407 ns 1201046 ns 597 bytes_per_second=3.87713Gi/s
BM_RefScanSinkRandom/10000000 2534397 ns 2523344 ns 273 bytes_per_second=3.69083Gi/s
```
(Measurements on Ryzen 5900X via `nix build --file ci/gha/tests componentTests.nix-store-tests-run -L`)
The changes include:
* Defining nix32Chars as a constexpr char[].
* Adding a constexpr std::array<unsigned char, 256> (reverseNix32Map) to map characters to their base-32 digit values at compile time.
* Replacing the slow character search loop with a direct lookup using reverseNix32Map.
* Removing std::once_flag/isBase32 logic in references.cc in favor of reverseNix32Map
Signed-off-by: Alexander V. Nikolaev <avn@avnik.info>
This changes our GHA CI and nix-store-tests packaging
to build and run the benchmarks. This does not affect
the default packaging - the overrides apply only for the
GHA CI.
GCC doesn't really benefit as much as Clang does from
using precompiled headers. Another aspect to consider is that
clangd doesn't really like GCC's PCH flags in the compilation database,
so GCC based devshells would continue to work with clangd.
This also has the slight advantage of ensuring that our includes are in
order, since we build with both Clang and GCC.
Instead of parsing a structured attrs at some later point, we parsed it
right away when parsing the A-Term format, and likewise serialize it to
`__json = <JSON dump>` when serializing a derivation to A-Term.
The JSON format can directly contain the JSON structured attrs without
so encoding it, so we just do that.
Add a new setting to warn about path literals that don't start with "." or "/". When enabled,
expressions like `foo/bar` will emit a warning suggesting to use `./foo/bar` instead.
A functional test is included.
The setting defaults to false for backward compatibility but could eventually default to true in
the future.
Closes: #13374
Co-authored-by: Robert Hensing <roberth@users.noreply.github.com>
All of the existing tests only run on x86_64-linux and
the whole `nixpkgsFor` makes it hard to reuse those for
e.g. running the nixos tests with a sanitizer build of nix.
This rips off the bandaid and removes the `nixpkgsFor` parameter
in favor of a single instance of nixpkgs with a separate `nixComponents`.
This command fetches all inputs of a flake in parallel.
Example runtime for
$ chmod -R u+w /tmp/nix2; rm -rf /tmp/nix2; rm ~/.cache/nix/fetcher-cache-v3.sqlite*; rm -rf ~/.cache/nix/tarball-cache/ ~/.cache/nix/gitv3/; time nix flake prefetch-inputs --store /tmp/nix2 https://api.flakehub.com/f/pinned/informalsystems/cosmos.nix/0.3.0/018ce9ed-d0be-7ce5-81b6-a3c6e3ae1187/source.tar.gz
with http-connections = 1:
real 4m11.859s
user 2m6.931s
sys 0m25.619s
and http-connections = 25 (the default):
real 0m57.146s
user 2m49.506s
sys 0m36.008s
This moves out the checks that get run in GHA CI into ci/gha/tests
folder and splits those into `topLevel` and `componentTests` attributes.
The idea behind this is to make it easier to parametrize tests that can
be run with sanitizers in order to run those as a matrix of jobs. The same
can be said for static builds.
Existing stdenv selection infrastructure via `lib.makeComponents` would
also allow us to switch over to using `clangStdenv` to significantly speed
up pre-merge CI (though the default stdenv would still be used for non-overridable
topLevel checks, like installer artifacts).
Prior patches in 54dc5314e8
and 6db6190002 fixed the default
system double for i686 and ppc/ppc64. This also patch also covers
32 bit arm and mips. ARM cpu names are taken from host_machine.cpu()
for a lack of a better option, but host_machine.cpu_family() is
preferred, since that is supposed to be somewhat standard for cross
files. Endianness is handled correctly by looking at host_machine.endian().
This also updates the documentation to be up to date to how system cpu
is translated from the host_machine specification.
Before this change, if you were cross compiling Nix, then the nix-manual
subproject would never get built. In some situations, it makes sense to
not build the nix-manual subproject when cross compiling. For example,
if the build system is x86_64 and the host system is riscv64, then it
makes sense to not build the manual. Building the manual requires
executing certain build artifacts, and you can’t run x86_64 executables
on riscv64 systems.
That being said, there are some situations where it does make sense to
build the nix-manual subproject when cross compiling. For example, if
the build system is x86_64 and the host system is i686, then it doesn’t
make sense to not build the manual. You can run i686 executables on
x86_64 systems just fine.
This change makes it so that the nix-manual subproject will sometimes
get built when cross compiling. Specifically, the nix-manual subproject
will get built as long as the doc-gen option is enabled and the build
system is capable of running host binaries.
---
The main motivation behind this change is to fix this Nixpkgs issue [1].
Building pkgs.nixStatic counts as cross compiling Nix, and
pkgs.nixStatic is supposed to produce a man output. Building
pkgs.nixStatic currently fails because it isn’t actually producing a man
output. That issue will go away once this commit gets backported to Nix
2.28.x.
[1]: <https://github.com/NixOS/nixpkgs/issues/426410>
SHA-256 is Git's next hash algorithm. The world is still basically stuck
on SHA-1 with git, but shouldn't be. We can at least do our part to get
ready.
On the C++ implementation side, only a little bit of generalization was
needed, and that was fairly straight-forward. The tests (unit and
system) were actually bigger, and care was taken to make sure they were
all cover both algorithms equally.
For regular, non-executable files, there is no str("") between str("regular")
and str("contents"). Note that str("") is exactly 8 zero bytes, while just ""
is actual empty string (0 bytes).
libfetchers uses `git_mempack_write_thin_pack` which was introduced in libgit2-1.9.0
This avoids error like:
../src/libfetchers/git-utils.cc: In member function ‘virtual void nix::GitRepoImpl::flush()’:
../src/libfetchers/git-utils.cc:270:13: error: ‘git_mempack_write_thin_pack’ was not declared in this scope
270 | git_mempack_write_thin_pack(mempack_backend, packBuilder.get())
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~
on older libgit2 (like 1.7.2 in Centos Stream 10)
If the reference for git+file is an annotated tag, the revision will
differ than when it's fetched using other fetchers such as `github:`
since Github seems to automatiacally peel to the underlying commit.
Turns out that rev-parse has the capability through it's syntax to
request the underlying commit by "peeling" using the `^{commit}` syntax.
This is safe to apply in all scenarios where the goal is to get an
underlying commit.
fixes#11266
I've missed this while reviewing 6db6190002.
I only built big endian ppc64, so that didn't occur to me.
From meson manual:
> Those porting from autotools should note that Meson does not add
> endianness to the name of the cpu_family. For example, autotools will
> call little endian PPC64 "ppc64le", Meson will not, you must also check
> the .endian() value of the machine for this information.
This code should handle that correctly.
builtins.fetchGit is not using the cached Git directory if
packed-references are used.
This is because the ref file for the fetchGit `refs/heads/master` is
used to check the mtime for whether to cache or not.
Let's at least codify this failure in a test case.
This is a follow-up to 6ec50ba736, which
also almost halves the compile times on clang for subprojects that use PCH.
`-fpch-instantiate-templates` is a clang-only option to force the instantiation
of templates once in the PCH itself, not all of the translation units that
it gets included to. This really cuts down on the overhead from nlohmann::json
and std::format code:
48244 ms: nlohmann::basic_json<>::parse<const char *> (76 times, avg 634 ms)
36193 ms: nlohmann::basic_json<>::basic_json (310 times, avg 116 ms)
28307 ms: nlohmann::detail::parser<nlohmann::basic_json<>, nlohmann::detail::i... (76 times, avg 372 ms)
20334 ms: nlohmann::detail::parser<nlohmann::basic_json<>, nlohmann::detail::i... (76 times, avg 267 ms)
17387 ms: nlohmann::basic_json<>::json_value::json_value (389 times, avg 44 ms)
16822 ms: std::vformat_to<std::__format::_Sink_iter<char>> (76 times, avg 221 ms)
16771 ms: std::__format::__do_vformat_to<std::__format::_Sink_iter<char>, char... (76 times, avg 220 ms)
12160 ms: std::vformat_to<std::__format::_Sink_iter<wchar_t>> (76 times, avg 160 ms)
12127 ms: std::__format::__do_vformat_to<std::__format::_Sink_iter<wchar_t>, w... (76 times, avg 159 ms)
10397 ms: nlohmann::detail::json_sax_dom_callback_parser<nlohmann::basic_json<... (76 times, avg 136 ms)
9118 ms: nlohmann::basic_json<>::data::data (76 times, avg 119 ms)
Initially done by Jade Lovelace <lix@jade.fyi> in https://gerrit.lix.systems/c/lix/+/1842.
We are doing basically the same, but unconditionally. It would be
a huge pain to add a pch option for all subprojects to just support the
usecase of using clangd in a gcc devshell.
In total, this basically halves the frontend times for nix-util and nix-store
to the point that the most expensive part of the build is linking.
(nix-store before):
```
**** Time summary:
Compilation (77 times):
Parsing (frontend): 243.4 s
Codegen & opts (backend): 140.3 s
```
(nix-store after):
```
**** Time summary:
Compilation (77 times):
Parsing (frontend): 120.2 s
Codegen & opts (backend): 141.2 s
```
This adds a GHA jobs to help analyze build times
and its regressions. It is based on `clangStdenv` with `-ftime-trace`
together with `ClangBuildAnalyzer` to prepare markdown summary for
individual components.
This also has the minor benefit of dogfooding CA and impure derivations.
Users can access these through the `lib.makeComponents` return value,
so it's helpful to briefly explain some of them.
This doesn't replace `meta.description`, but supplements it.
(TODO: improve `meta.description`)
This was carefully refactored by inlining some Nixpkgs definitions
like `generateSplicesForMkScope`, so the memoization properties
should remain the same.
Boost.URL is a significantly more RFC-compliant parser
than what libutil currently has a bundle of incomprehensible
regexes.
One aspect of this change is that RFC4007 ZoneId IPv6 literals
are represented in URIs according to RFC6874 [1].
Previously they were represented naively like so: [fe80::818c:da4d:8975:415c\%enp0s25].
This is not entirely correct, because the percent itself has to be pct-encoded:
> "%" is always treated as
an escape character in a URI, so, according to the established URI
syntax [RFC3986] any occurrences of literal "%" symbols in a URI MUST
be percent-encoded and represented in the form "%25". Thus, the
scoped address fe80::a%en1 would appear in a URI as
http://[fe80::a%25en1].
[1]: https://datatracker.ietf.org/doc/html/rfc6874
Co-authored-by: Jörg Thalheim <joerg@thalheim.io>
The myriad of hand-rolled URL parsing and validation code
is a constant source of problems. Regexes are not a great way
of writing parsers and there's a history of getting them wrong.
Boost.URL is a good library we can outsource most of the heavy
lifting to.
* It is tough to contribute to a project that doesn't use a formatter,
* It is extra hard to contribute to a project which has configured the formatter, but ignores it for some files
* Code formatting makes it harder to hide obscure / weird bugs by accident or on purpose,
Let's rip the bandaid off?
Note that PRs currently in flight should be able to be merged relatively easily by applying `clang-format` to their tip prior to merge.
Previously, `nix help shell` failed with "Nix has no subcommand 'shell'" despite `nix shell --help`
working correctly. This happened because the `shell` command is actually an alias for `env shell`,
and the help system wasn't resolving aliases when looking up documentation.
This patch modifies the `showHelp` function to check for and resolve aliases before generating the
manpage name, ensuring that shorthand commands like `shell` get proper help documentation.
Closes: #13431
The default constructor for Attr was not initializing the value pointer,
which could lead to undefined behavior when the uninitialized pointer is
accessed. This was caught by clang-tidy's UninitializedObject check.
This fixes the warning:
1 uninitialized field at the end of the constructor call
[clang-analyzer-optin.cplusplus.UninitializedObject]
The s3.hh public header was incorrectly including store-config-private.hh
instead of the public config.hh. Since NIX_WITH_S3_SUPPORT is defined in
the public config, this caused clang-tidy to report it as undefined.
Move init() call from constructor to openStore() method to avoid calling
virtual methods during object construction. This prevents undefined
behavior when virtual methods are called before the object is fully
constructed.
Move init() call from constructor to openStore() method to avoid calling
virtual methods during object construction. This prevents undefined
behavior when virtual methods are called before the object is fully
constructed.
When `file://` is used accidentally in a flake as the source it is
expected to be a tarball by default.
Add some friendlier error messages to either inform the user this is not
in fact a tarball or if it's a git directory, let them know they can use
`git+file`.
fixes#12935
Users have complained that fetchGit is flaky however the culprit is
likely that `git fetch` was unable itself to download the repository for
whatever reason (i.e. poor network etc..)
Nothing was checking the status of `git fetch` and the error message
that would eventually surface to the users were that the commit was not
found.
Add explicit error checking for status code from `git fetch` and return
a message earlier on to indicate that the failure was from that point.
fixes#10431
We’re already allowing `/tmp` anyway, so this should be harmless,
and it fixes a regression in the default configuration caused by
moving the build directories out of `temp-dir`. (For instance, that
broke the Lix `guessOrInventPath.sockets` test.)
Note that removing `/tmp` breaks quite a few builds, so although it may
be a good idea in general it would require work on the Nixpkgs side.
Fixes: 749afbbe99fd7b45f828b72628252feba9241362
Change-Id: I6a6a69645f429bc50d4cb24283feda3d3091f534
(This is a cherry-pick of commit d1db3e5fa3faa43b3d2f2e2e843e9cfc1e6e1b71)
Lix patch: https://gerrit.lix.systems/c/lix/+/3500
When AWS credentials expired, in some scenarios they led to the
nix process aborting with an error similar to ' Unable to parse ExceptionName: ExpiredToken'.
This change updates the S3 handling code such that those errors are treated like 403s or 404s.
Closes#13459
c39cc00404 has added assertions for
all Value accesses and the following case has started failing with
an `unreachable`:
(/tmp/fun.nix):
```nix
{a}: a
```
```
$ nix eval --impure --expr 'import /tmp/fun.nix {a="a";b="b";}'
```
This would crash:
```
terminating due to unexpected unrecoverable internal error: Unexpected condition in getStorage at ../include/nix/expr/value.hh:844
```
This is not a regression, but rather surfaces an existing problem, which previously
was left undiagnosed. In the case of an import `fun` is the `import` primOp, so that read is invalid
and previously this resulted in an access into an inactive union member, which is UB.
The correct thing to use is `vCur`. Identical problem also affected the case of a missing argument.
Add previously failing test cases to the functional/lang test suite.
Fixes#13448.
Fixes:
[261/394] Linking target src/libexpr/libnixexpr.so
In function ‘copy’,
inlined from ‘__ct ’ at /nix/store/24sdvjs6rfqs69d21gdn437mb3vc0svh-gcc-14.2.1.20250322/include/c++/14.2.1.20250322/bits/basic_string.h:688:23,
inlined from ‘operator+’ at /nix/store/24sdvjs6rfqs69d21gdn437mb3vc0svh-gcc-14.2.1.20250322/include/c++/14.2.1.20250322/bits/basic_string.h:3735:43,
inlined from ‘operator()’ at ../src/libexpr/primops/fetchClosure.cc:127:58,
inlined from ‘prim_fetchClosure’ at ../src/libexpr/primops/fetchClosure.cc:132:88:
/nix/store/24sdvjs6rfqs69d21gdn437mb3vc0svh-gcc-14.2.1.20250322/include/c++/14.2.1.20250322/bits/char_traits.h:427:56: warning: ‘__builtin_memcpy’ writing 74 bytes into a region of size 16 overflows the destination [-Wstringop-overflow=]
427 | return static_cast<char_type*>(__builtin_memcpy(__s1, __s2, __n));
| ^
../src/libexpr/primops/fetchClosure.cc: In function ‘prim_fetchClosure’:
../src/libexpr/primops/fetchClosure.cc:132:88: note: at offset 16 into destination object ‘<anonymous>’ of size 32
132 | fromPath = state.coerceToStorePath(attr.pos, *attr.value, context, attrHint());
| ^
-p preserves xattrs and acls which can be incompatible between
filesystems
Unfortunately keep -p on darwin because the bsd coreutils do not
support --preserve.
Fixes#13426
This failed on macOS:
nix-store-tests-run> C++ exception with description "../nix_api_store.cc:33: nix_err_code(ctx) != NIX_OK, message: error: getting status of '/nix/var/nix/daemon-socket/socket': Operation not permitted" thrown in the test body.
This changes makes nix detect a machines available cores automatically whenever build-cores is set to 0.
So far, nix simply passed NIX_BUILD_CORES=0 whenever build-cores is set to 0. (only when build-cores is unset it was detecting cores automatically)
The behavior of passing NIX_BUILD_CORES=0 leads to a performance penalty when sourcing nixpkgs' generic builder's `setup.sh`, as setup.sh has to execute `nproc`. This significantly slows down sourcing of setup.sh
The use of R"(...)" added a bunch of unnecessary whitespace, e.g.
error:
Unable to start any build;
either increase '--max-jobs' or enable remote builds.
For more information run 'man nix.conf' and search for '/machines'.
Now we get
error: Unable to start any build; either increase '--max-jobs' or enable remote builds.
For more information run 'man nix.conf' and search for '/machines'.
This shaves off a very significand amount of memory used
for evaluation as well as reduces the GC-managed heap.
Previously the union discriminator (InternalType) was
stored as a separate field in the Value, which takes up
whole 8 bytes due to padding needed for member alignment.
This effectively wasted 7 whole bytes of memory. Instead
of doing that InternalType is instead packed into pointer
alignment niches. As it turns out, there's more than enough
unused bits there for the bit packing to be effective.
See the doxygen comment in the ValueStorage specialization
for more details.
This does not add any performance overhead, even though
we now consistently assert the InternalType in all getters.
This can also be made atomic with a double width compare-and-swap
instruction on x86_64 (CMPXCHG16B instruction) for parallel evaluation.
This factors out most of the value representation into a mixin class.
`finishValue` is now gone for good and replaced with a simple template
function `setStorage` which derives the type information/disriminator from
the type of the argument. Likewise, reading of the value goes through function
template `getStorage`.
An empty type `Null` is introduced to make the bijection InternalType <-> C++ type
complete.
Previous use of symlink_status() always translated into a stat call, leading
to huge performance penalties for by-name-overlay in nixpkgs. The comment
below references the possible caching, but that seemed to be erroneous, since
the correct way to make use of the caching API is by calling a bunch of `is_*`
functions [1]. For example, here's how libstdc++ does that [2], [3].
This translates to great nixpkgs eval performance improvements:
```
Benchmark 1: GC_INITIAL_HEAP_SIZE=4G result/bin/nix-instantiate ../nixpkgs -A hello --readonly-mode
Time (mean ± σ): 186.7 ms ± 6.7 ms [User: 121.3 ms, System: 64.9 ms]
Range (min … max): 179.4 ms … 201.6 ms 16 runs
Benchmark 2: GC_INITIAL_HEAP_SIZE=4G nix-instantiate ../nixpkgs -A hello --readonly-mode
Time (mean ± σ): 230.6 ms ± 5.0 ms [User: 126.9 ms, System: 103.1 ms]
Range (min … max): 225.1 ms … 241.4 ms 13 runs
```
[1]: https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2016/p0317r1.html
[2]: 8ea555b7b4/libstdc%2B%2B-v3/include/bits/fs_dir.h (L341-L348)
[3]: 8ea555b7b4/libstdc%2B%2B-v3/include/bits/fs_dir.h (L161-L163)
These wrappers don't seem to be used anywhere in and out of tree.
Also the declaration in the header has an incorrect function name `maybeParseFlake`.
Closes#11948
Fixes
error:
… while processing sandbox path '/private/tmp/nix-shell.0MDgyx/nix-test/ca/repl/store/nix/var/nix/builds/nix-build-simple.drv-65916-3910734210' (/private/tmp/nix-shell.0MDgyx/nix-test/ca/repl/store)
error: 'nix' is too short to be a valid store path
which happened because we were now putting the build directory
underneath the store directory.
If a build directory is accessible to other users it is possible to
smuggle data in and out of build directories. Usually this is only
a build purity problem, but in combination with other issues it can
be used to break out of a build sandbox. to prevent this we default
to using a subdirectory of nixStateDir (which is more restrictive).
(cherry picked from pennae Lix commit 55b416f6897fb0d8a9315a530a9b7f0914458ded)
(store setting done by roberth)
In the case the deletion fails, we should ensure that the temporary
directory cannot be used for nefarious purposes.
Change-Id: I498a2dd0999a74195d13642f44a5de1e69d46120
Signed-off-by: Raito Bezarius <raito@lix.systems>
When calling `_deletePath` with a parent file descriptor, `openat` is
made effective by using relative paths to the directory file descriptor.
To avoid the problem, the signature is changed to resist misuse with an
assert in the prologue of the function.
Change-Id: I6b3fc766bad2afe54dc27d47d1df3873e188de96
Signed-off-by: Raito Bezarius <raito@lix.systems>
This ensures that `passAsFile` data is created inside the expected
temporary build directory by `openat()` from the parent directory file
descriptor.
This avoids a TOCTOU which is part of the attack chain of CVE-????.
Change-Id: Ie5273446c4a19403088d0389ae8e3f473af8879a
Signed-off-by: Raito Bezarius <raito@lix.systems>
`writeFile` lose its `sync` boolean flag to make things simpler.
A new `writeFileAndSync` function is created and all call sites are
converted to it.
Change-Id: Ib871a5283a9c047db1e4fe48a241506e4aab9192
Signed-off-by: Raito Bezarius <raito@lix.systems>
We use it immediately for the build temporary directory.
Change-Id: I180193c63a2b98721f5fb8e542c4e39c099bb947
Signed-off-by: Raito Bezarius <raito@lix.systems>
We now keep around a proper AutoCloseFD around the temporary directory
which we plan to use for openat operations and avoiding the build
directory being swapped out while we are doing something else.
Change-Id: I18d387b0f123ebf2d20c6405cd47ebadc5505f2a
Signed-off-by: Raito Bezarius <raito@lix.systems>
This is useful for certain error recovery paths (no pun intended) that
does not thread through the original path name.
Change-Id: I2d800740cb4f9912e64c923120d3f977c58ccb7e
Signed-off-by: Raito Bezarius <raito@lix.systems>
This is just a code cleanup; it should not be behavior change.
`addWantedOutputs` is removed by introducing `DerivationTrampolineGoal`.
`DerivationGoal` now only tracks a single output, and is back to
tracking a plain store path `drvPath`, not a deriving path one. Its
`addWantedOutputs` method is gone. These changes will allow subsequent
PRs to simplify it greatly.
Because the purpose of each goal is back to being immutable, we can also
once again make `Goal::buildResult` a public field, and get rid of the
`getBuildResult` method. This simplifies things also.
`DerivationTrampolineGoal` is, as the nane is supposed to indicate, a
cheap "trampoline" goal. It takes immutable sets of wanted outputs, and
just kicks of `DerivationGoal`s for them. Since now "actual work" is
done in these goals, it is not wasteful to have separate ones for
separate sets of outputs, even if those outputs (and the derivations
they are from) overlap.
This design is described in more detail in the doc comments on the goal
types, which I've now greatly expanded.
---
This separation of concerns will make it possible for future work on
issues like #11928, and to continue the path of having more goal types,
but each goal type does fewer things (issue #12628).
---
This commit in some sense reverts
f4f28cdd0e, but that one kept around
`addWantedOutputs`. I am quite sure it was having two layers of goals
with `addWantedOutputs` that caused the issues --- restarting logic like
`addWantedOutputs` has is very tempermental! In this version of the
change, we have *zero* layers of `addWantedOutputs` --- no goal type
needs it, or otherwise has a mutable objective --- and so I think this
change is safe.
Co-authored-by: Sergei Zimmerman <145775305+xokdvium@users.noreply.github.com>
Running parallel nix in nix can lead to multiple instances trying to
create the state directories and failing on the `createSymlink` step,
because the link already exists.
`replaceSymlink` is already idempotent, so let's use that.
Resolves#2706
libstdc++'s std::stable_sort and new builtins.sort implementation
special-case ranges with length less than or equal to 16 and delegate
to insertionsort.
Having a larger e2e test would allow catching sort stability issues
at functional level as well.
This prevents C++ level undefined behavior from affecting
the evaluator. Stdlib implementation details should not affect
eval, regardless of the build platform. Even erroneous usage
of `builtins.sort` should not make it possible to crash the
evaluator or produce results that depend on the host platform.
Unlike std::sort and std::stable_sort, this implementation
does not lead to out-of-bounds memory reads or other undefined
behavior when the predicate is not strict weak ordering.
This makes it possible to use this function in libexpr for
builtins.sort, where an incorrectly implemented comparator
in the user nix code currently can crash and burn the evaluator
by invoking C++ UB.
Makes the behavoral change of #13263 without the underlying refactor.
Hopefully this clearly safe from a perf and GC perspective, and will
make it easier to benchmark #13263.
This overload isn't actually necessary anywhere and
doesn't make much sense. The pointers to `Value`s are
themselves const, but the `Value`s are mutable.
A non-const member function implies that the object itself
can be modified but this doesn't make much sense considering
the return type: `Value * const * `, which is a pointer
to a constant array of pointers to mutable values.
It seems obvious that erasing any output paths from pathsInChroot needs
to happen after getPathsInSandbox(), not before.
Signed-off-by: Samuli Thomasson <samuli.thomasson@pm.me>
This adds a meson.format file that mostly mirrors the projects
meson style and a pre-commit hook to enforce this style.
Some low-diff files are formatted.
Seen in https://github.com/DeterminateSystems/nix-src/actions/runs/15590867877/job/43909540271:
nix-functional-tests> grep: repl_output: No such file or directory
nix-functional-tests> +(repl.sh:174) cat repl_output
This is because there is a small possibility that the `nix repl` child
process hasn't created `repl_output` yet. So make sure it exists.
Add helpful context when opening the Nix database lock fails due to
permission errors. Instead of just showing "Permission denied", now
provides guidance about possible causes:
- Running as non-root in a single-user Nix installation
- Nix daemon may have crashed
Invoking `:ll` will start a pager with all variables which have just
been loaded by `:lf`, `:l`, or by a flake provided to `nix repl` as an
argument.
https://github.com/NixOS/nix/issues/11404
When we run `nix repl nixpkgs` we get "Added 6 variables". This is not
useful as it doesn't tell us which variables the flake has exported to
our global repl scope.
This patch prints the name of each variable that was just loaded. We
currently cap printing to 20 variables in order to avoid excessive
prints.
https://github.com/NixOS/nix/issues/11404
`getPrimOp` function was basically identical to existing
`Value::primOpAppPrimOp` modulo some trivial differences.
Makes sense to reuse existing code for that.
There's actually no mutation happening so there's no point in using
a mutable shared_ptr. Furthermore, this makes it much more evident to
the reader that no actual mutation (especially in multithreaded case)
is happening.
Also get rid of redundant constructor that isn't actually used anywhere
other than `Pos::operator std::shared_ptr<Pos>` which just passes in &*this,
(identical to just `this`), which can't be nullptr.
Judging by the comment for `makeEmptySourceAccessor` the prefix has
to be empty:
> Return a source accessor that contains only an empty root directory.
Fixes#13295.
Since the migration to meson precompiled-headers.h isn't actually used
anymore and is just confusing. Meson can't handle shared pch across
subprojects [1] and without that there's no performances benefit of PCH
at all. Also rolling our own support for that isn't trivial. See [2] for
an example of how that would look like.
[1]: https://github.com/mesonbuild/meson/issues/4350
[2]: 22bc8b6473/plugins/meson.build
Clang doesn't like the double indent that is needed for the `if...else`
that is CPP'd away. Adding braces is fine in the `if...else...` case,
and fine as a naked block in the CPP'd away case, and properly-indented
both ways.
These tests have been collected from nixpkgs f870c6ccc8951fc48aeb293cf3e98ade6ac42668
usage of builtins.match for x86_64-linux eval system. At most 2 matching and
non-matching cases are included for each encountered regex. This should
hopefully add more confidence when possibly trying to switch the regex implementation
in the future.
This is the utility changes from #9968, which were easier to rebase
first.
I (@Ericson2314) didn't write this code; I just rebased it.
Co-Authored-By: Artemis Tosini <me@artem.ist>
Co-Authored-By: Audrey Dutcher <audrey@rhelmot.io>
Allows to copy the archive to a remote host and not get
error: cannot add path '/nix/store/01x2k4nlxcpyd85nnr0b9gm89rm8ff4x-source' because it lacks a signature by a trusted key
Before we got something like this but only inside the VM test:
vm-test-run-functional-tests-on-nixos_user> machine # fetchGit.sh: line 286: unexpected EOF while looking for matching `)'
We now try to do not too much in a single line, so that the bash parser does not get confused.
This also seems more readable and better quoted.
This makes the profiler much more useful by actually distiguishing
different derivations being evaluated. This does make the implementation
a bit more convoluted, but I think it's worth it.
Sometimes the profiler might want to do evaluation (e.g. for getting
derivation names). This is not ideal, but is really necessary
to make the profiler stack traces useful for end users.
...and also NIX_STATE_HOME in nix-profile.fish. This is directly
translated from the bash scripts and makes the fish scripts equivalent
in functionality to the bash scripts.
Note that nix-profile.fish checks for NIX_STATE_HOME and
nix-profile-daemon.fish does not, so the two scripts are no longer
identical.
This was discussed in https://github.com/NixOS/nix/issues/8034. I
personally like `PRJ_ROOT`, which hopefully avoids some ambiguity
around with subflakes.
I only implemented this for `nix fmt` because it doesn't let you point
at a flake not on your filesystem.
macOS compilation fixes
This method does *not* create a new type of goal. We instead just make
`DerivationGoal` more sophisticated, which is much easier to do now that
`DerivationBuildingGoal` has been split from it (and so many fields are
gone, or or local variables instead).
This avoids the need for a secondarily trampoline goal that interacted
poorly with `addWantedOutputs`. That, I hope, will mean the bugs from
before do not reappear.
There may in fact be a reason to introduce such a trampoline in the
future, but it would only happen in conjunction with getting rid of
`addWantedOutputs`.
Restores the functionality (and tests) that was reverted in
f4f28cdd0e.
This patch adds support for a native stack sampling
profiler to the evaluator, which saves a collapsed stack
profile information to a configurable location.
Introduced options (in `EvalSettings`):
- `eval-profile-file` - path to the collected profile file.
- `eval-profiler-frequency` - sampling frequency.
- `eval-profiler` - enumeration option for enabling the profiler.
Currently only `flamegraph` is supported, but having this an
enumeration rather than a boolean switch leaves the door open
for other profiler variants (e.g. tracy).
Profile includes the following information on best-effort basis (e.g. some lambdas might
have an undefined name). Callstack information contains:
- Call site location (where the function gets called).
- Primop/lambda name of the function being called.
- Functors/partial applications don't have a name attached to them unlike special-cased primops and lambads.
For cases where callsite location isn't available we have to resort to providing
the location where the lambda itself is defined. This removes some of the confusing
`«none»:0` locations in the profile from previous attempts.
Example usage with piping directly into zstd for compression:
```
nix eval --no-eval-cache nixpkgs#nixosTests.gnome \
--eval-profiler flamegraph \
--eval-profile-file >(zstd -of nix.profile.zstd)
```
Co-authored-by: Jörg Thalheim <joerg@thalheim.io>
Commit b36637c8f7 set
`__ETC_PROFILE_NIX_SOURCED` globally, but this is not enough to prevent
the script from being run again by child shells, because the
variable was not exported and thus not inherited by any child process.
Exporting the variable also agrees with the bash scripts.
Notably, the old behavior broke `nix develop -c fish` in some cases,
because the profile bin directory got prepended to the path, causing
binaries from the profile to override binareis from the devshell.
This wires up the {pre,post}FunctionCallHook machinery
in EvalState::callFunction and migrates FunctionCallTrace
to use the new EvalProfiler mechanisms for tracing.
Note that branches when the hook gets called are marked with [[unlikely]]
as a hint to the compiler that this is not a hot path. For non-tracing
evaluation this should be a 100% predictable branch, so the performance
cost is nonexistent.
Some measurements to prove support this point:
```
nix build .#nix-cli
nix build github:nixos/nix/d692729759e4e370361cc5105fbeb0e33137ca9e#nix-cli --out-link before
```
(Before)
```
$ taskset -c 2,3 hyperfine "GC_INITIAL_HEAP_SIZE=16g before/bin/nix eval nixpkgs#gnome --no-eval-cache" --warmup 4
Benchmark 1: GC_INITIAL_HEAP_SIZE=16g before/bin/nix eval nixpkgs#gnome --no-eval-cache
Time (mean ± σ): 2.517 s ± 0.032 s [User: 1.464 s, System: 0.476 s]
Range (min … max): 2.464 s … 2.557 s 10 runs
```
(After)
```
$ taskset -c 2,3 hyperfine "GC_INITIAL_HEAP_SIZE=16g result/bin/nix eval nixpkgs#gnome --no-eval-cache" --warmup 4
Benchmark 1: GC_INITIAL_HEAP_SIZE=16g result/bin/nix eval nixpkgs#gnome --no-eval-cache
Time (mean ± σ): 2.499 s ± 0.022 s [User: 1.448 s, System: 0.478 s]
Range (min … max): 2.472 s … 2.537 s 10 runs
```
This patch adds an EvalProfiler and MultiEvalProfiler that can be used
to insert hooks into the evaluation for the purposes of function tracing
(what function-trace currently does) or for flamegraph/tracy profilers.
See the following commits for how this is supposed to be integrated into
the evaluator and performance considerations.
Ensure relative path inputs are relative to the parent node's _actual_
`outPath`, instead of the subtly different `sourceInfo.outPath`.
Additionally, non-flake inputs now also have a `sourceInfo` attribute.
This fixes the relationship between `self.outPath` and
`self.sourceInfo.outPath` in some edge cases.
Fixes#13164
Previous code had a sneaky bug due to which no caching
actually happened:
```cpp
auto linesForInput = (*lines)[origin->offset];
```
That should have been:
```cpp
auto & linesForInput = (*lines)[origin->offset];
```
See [1].
Now that it also makes sense to make the cache bound in side
in order not to memoize all the sources without freeing any memory.
The default cache size has been chosen somewhat arbitrarily to be ~64k
origins. For reference, 25.05 nixpkgs has ~50k .nix files.
Simple benchmark:
```nix
let
pkgs = import <nixpkgs> { };
in
builtins.foldl' (acc: el: acc + el.line) 0 (
builtins.genList (x: builtins.unsafeGetAttrPos "gcc" pkgs) 10000
)
```
(After)
```
$ hyperfine "result/bin/nix eval -f ./test.nix"
Benchmark 1: result/bin/nix eval -f ./test.nix
Time (mean ± σ): 292.7 ms ± 3.9 ms [User: 131.0 ms, System: 120.5 ms]
Range (min … max): 288.1 ms … 300.5 ms 10 runs
```
(Before)
```
hyperfine "nix eval -f ./test.nix"
Benchmark 1: nix eval -f ./test.nix
Time (mean ± σ): 666.7 ms ± 6.4 ms [User: 428.3 ms, System: 191.2 ms]
Range (min … max): 659.7 ms … 681.3 ms 10 runs
```
If the origin happens to be a `all-packages.nix` or similar in size then the
difference is much more dramatic.
[1]: 22e3f0e987
Try to make `DerivationGoal` care less whether we're working from an
in-memory derivation or not.
It's a clean-up in its own right, but it will also help with other
cleanups under the umbrella of #12628.
Now, each class provides the initial coroutine by value. This avoids
some sketchy virtual function stuff, and will also be further put to
good use in the next commit.
As summarized in
https://github.com/NixOS/nix/issues/77#issuecomment-2843228280 the
motivation is that the complicated retry logic this introduced was
making the cleanup task #12628 harder to accomplish. It was not easy to
ascertain just what policy / semantics the extra control-flow was
implementing, in order to figure out a different way to implementing it
either.
After talking to Eelco about it, he decided we could just....get rid of
the feature entirely! It's a bit scary removing a decade+ old feature,
but I think he is right. See the release notes for more explanation.
This reverts commit 299141ecbd.
Co-authored-by: Eelco Dolstra <edolstra@gmail.com>
Leverage #10766 to show how we can now resolve a store configuration
without actually opening the store for that resolved configuration.
Co-authored-by: Robert Hensing <roberth@users.noreply.github.com>
Splicing the list element to the back can be done in
a much simpler and concise way without the need for
erasing and re-inserting the element. Doing it this
way is equivalent to just moving node pointers around,
whereas inserting/erasing allocates/deallocates new nodes.
This renders the rl-next notes when `officialRelease = false`, which
corresponds to the case where we're not on a release branch.
Previously we had disabled this behavior because changelog-d is
somewhat of a heavy dependency, being the only Haskell package.
However, we now have new circumstances that topple the tradeoff.
- We render `master` docs to https://nix.dev/manual/nix/development/release-notes/rl-next.html
- `.#manual` is a separate build now, so `nix build nix/foo` is
not affected by the increased closure of build input outputs.
Because of these factors, I believe adding this functionality back
is more valuable, as we can use it to
- Previous release notes
- Showcase the upcoming release to the community
The existing header is a bit too big. Now the following use-cases are
separated, and get their own headers:
- Using or implementing an arbitrary store: remaining `store-api.hh`
This is closer to just being about the `Store` (and `StoreConfig`)
classes, as one would expect.
- Opening a store from a textual description: `store-open.hh`
Opening an aribtrary store implementation like this requires some sort
of store registration mechanism to exists, but the caller doesn't need
to know how it works. This just exposes the functions which use such a
mechanism, without exposing the mechanism itself
- Registering a store implementation: `store-registration.hh`
This requires understanding how the mechanism actually works, and the
mechanism in question involves templated machinery in headers we
rather not expose to things that don't need it, as it would slow down
compilation for no reason.
I can't find a good way to benchmark in isolation from the
git cache, but common sense dictates that creating (and destroying)
a 131KiB std::vector for each regular file from the archive imposes
quite a significant overhead regardless of the IO bound git cache.
AFAICT there is no reason to keep a copy of the data since
it always gets fed into the sink and there are no coroutines/threads
in sight.
warning: installing Nix as root is not supported by this script!
performing a single-user installation of Nix...
copying Nix to /nix/store.......................................................
......
warning: the group 'nixbld' specified in 'build-users-group' does not exist
warning: the group 'nixbld' specified in 'build-users-group' does not exist
installing 'nix-2.28.3'
error: the group 'nixbld' specified in 'build-users-group' does not exist
/tmp/nix-binary-tarball-unpack.2j3lCU0A89/unpack/nix-2.28.3-x86_64-linux/install: unable to install Nix into your default profile
The nixos/nix docker image is built using `buildLayeredImage`, which spreads the nix store over a configured number of layers. This number was set to create an image with 100 layers. Because there is a limit of (typically) 127 layers in AUFS, this only left 27 layers to build on top. At the same time, nearly half of the created layers were only <100kb in size, many even <10kb, negating the intended advantage in cachability.
This commit moves the tradeoff a bit by reducing the number of layers to 70.
Layer sizes for the 2.28.3 nixos/nix image: https://hub.docker.com/layers/nixos/nix/2.28.3/images/sha256-d078d7153763895fce17c5fbbdeb86fcfcac414ca0ba875d413c1df57be19931
As it turns out using `std::regex` is actually the bottleneck
for root discovery. Just substituting `std::` -> `boost::`
makes root discovery twice as fast (3x if counting only userspace time).
Some rather ad-hoc measurements to motivate the switch:
(On master)
```
nix build github:nixos/nix/1e822bd4149a8bce1da81ee2ad9404986b07914c#nix-cli --out-link result-1e822bd4149a8bce1da81ee2ad9404986b07914c
taskset -c 2,3 hyperfine "result-1e822bd4149a8bce1da81ee2ad9404986b07914c/bin/nix store gc --dry-run --max 0"
Benchmark 1: result-1e822bd4149a8bce1da81ee2ad9404986b07914c/bin/nix store gc --dry-run --max 0
Time (mean ± σ): 481.6 ms ± 3.9 ms [User: 336.2 ms, System: 142.0 ms]
Range (min … max): 474.6 ms … 487.7 ms 10 runs
```
(After this patch)
```
taskset -c 2,3 hyperfine "result/bin/nix store gc --dry-run --max 0"
Benchmark 1: result/bin/nix store gc --dry-run --max 0
Time (mean ± σ): 254.7 ms ± 9.7 ms [User: 111.1 ms, System: 141.3 ms]
Range (min … max): 246.5 ms … 281.3 ms 10 runs
```
`boost::regex` is a drop-in replacement for `std::regex`, but much faster.
Doing a simple before/after comparison doesn't surface any change in behavior:
```
result/bin/nix store gc --dry-run -vvvvv --max 0 |& grep "got additional" | wc -l
result-1e822bd4149a8bce1da81ee2ad9404986b07914c/bin/nix store gc --dry-run -vvvvv --max 0 |& grep "got additional" | wc -l
```
Remove outdated and no longer relevant TODO. It's more confusing
now, since symbol table must now be addressed by uint32_t indices
in order to keep Attr size down to 16 bytes on 64 bit machines.
This patch finally applies the transition to std::less<>,
which is a transparent comparator. There's no functional
change and string lookups in sets are now more efficient
and don't produce temporaries (e.g. set.find(std::string_view{"key"})).
Unfortunately Feature is just an alias to `std::string`
and not a new-type, so a ton of code relies on it being
exactly a `std::string`.
Using transparent comparators just for StringSet necessitates
using it here as well.
The intention is to switch to transparent comparators from N3657 for
ordered set containers for strings and using the alias consistently
would simplify things.
This trades off some executable size for measurable lexer performance
improvements.
Note on the explicitly enabling 8bit scanner.
This is needed due to the default behavior of flex (excerpt from the manual [1]):
> Flex’s default behavior is to generate an 8-bit scanner unless you
> use the ‘-Cf’ or ‘-CF’, in which case flex defaults to generating
> 7-bit scanners unless your site was always configured to generate 8-bit
> scanners.
Some quantifyable metrics:
Nixpkgs revision: a6e3f45acf4e817532a861ab0eda4ab5485fecc1
Parsing the largest file in nixpkgs: pkgs/development/haskell-modules/hackage-packages.nix.
(Before this patch)
```
$ nix build github:nixos/nix/9fe3077d4#nix-expr
$ du --apparent-size result/lib/libnixexpr.so
2518 result/lib/libnixexpr.so
$ nix build github:nixos/nix/9fe3077d4#nix-cli
$ taskset -c 2,3 hyperfine "GC_INITIAL_HEAP_SIZE=16g \
result/bin/nix-instantiate --parse \
../nixpkgs/pkgs/development/haskell-modules/hackage-packages.nix > /dev/null"
Time (mean ± σ): 375.5 ms ± 6.3 ms [User: 316.9 ms, System: 56.7 ms]
Range (min … max): 368.5 ms … 388.3 ms 10 runs
```
(After the patch)
```
$ nix build .#nix-expr
$ du --apparent-size result/lib/libnixexpr.so
2685 result/lib/libnixexpr.so
$ nix build .#nix-cli
$ taskset -c 2,3 hyperfine "GC_INITIAL_HEAP_SIZE=16g \
result/bin/nix-instantiate --parse \
../nixpkgs/pkgs/development/haskell-modules/hackage-packages.nix > /dev/null"
Time (mean ± σ): 326.8 ms ± 4.9 ms [User: 269.5 ms, System: 55.3 ms]
Range (min … max): 319.7 ms … 335.5 ms 10 runs
```
Overall, the change is roughly:
- 2518KiB -> 2685KiB ~ 150 KiB of machine code
- 375ms -> 325ms ~ 50ms
The perf uplift for eval-heavy test cases is obviously less noticeable,
but it doesn't make sense not to take this free perf win.
[1]: https://westes.github.io/flex/manual/Options-Affecting-Scanner-Behavior.html#Options-Affecting-Scanner-Behavior
When a PUT is redirected, some of the data can be sent by curl before headers are read. This means the subsequent PUT operation needs to seek back to origin.
The current example relies upon [nixfmt's deprecated tree traversal
behavior](https://github.com/NixOS/nixfmt/pull/240). The simplest
alternative is the new `nixfmt-tree` wrapper for `nixfmt`/`treefmt`.
Since we dropped fs::symlink_exists, we no longer have a need for the fs
namespace. Having less abstractions makes it easier to lookup the
functions in reference documentations.
As it turns out the orignal implementation of symlink_exists cannot be
used in Nix because it did now std::filesystem::filesystem_error.
The new implementation fixes that but is now actually the same as
pathExists except for the path type.
`nix formatter build` is sort of like `nix build`: it builds, links, and
prints a path to the formatter program:
$ nix formatter build
/nix/store/cb9w44vkhk2x4adfxwgdkkf5gjmm856j-treefmt/bin/treefmt
Note that unlike `nix build`, this prints the full path to the program,
not just the store path (in the example above that would be
`/nix/store/cb9w44vkhk2x4adfxwgdkkf5gjmm856j-treefmt`).
Motivation
----------
I maintain a vim plugin that automatically runs `nix fmt` on files on
save. Since `nix fmt` can be quite slow due to nix evaluation, I choose
to cache the `nix fmt `entrypoint. This was very awkward to do, see the
implementation for details:
7864607231/lua/null-ls/builtins/formatting/nix_flake_fmt.lua (L83-L110).
I recently discovered that my implementation was buggy (it didn't handle
flakes that expose a `formatter` package, such as nixpkgs), so I had to
rework the implementation:
https://github.com/nvimtools/none-ls.nvim/pull/272.
With the new `nix formatter build` command, I can delete all this akward
code, and it will be easier for other folks to build performant editor
integrations for `nix fmt`.
This refactor shouldn't change much except add a new `nix formatter run`
command. This creates space for the new `nix formatter build` command,
which I'll be introducing in the next commit.
Before the change `nix` was stripping warning flags
reported by `gcc-14` too eagerly:
$ nix build -f. texinfo4
error: builder for '/nix/store/i9948l91s3df44ip5jlpp6imbrcs646x-texinfo-4.13a.drv' failed with exit code 2;
last 25 log lines:
> 1495 | info_tag (mbi_iterator_t iter, int handle, size_t *plen)
> | ~~~~~~~~^~~~
> window.c:1887:39: error: passing argument 4 of 'printed_representation' from incompatible pointer type []
> 1887 | &replen);
> | ^~~~~~~
> | |
> | int *
After the change the compiler flag remains:
$ ~/patched.nix build -f. texinfo4
error: builder for '/nix/store/i9948l91s3df44ip5jlpp6imbrcs646x-texinfo-4.13a.drv' failed with exit code 2;
last 25 log lines:
> 1495 | info_tag (mbi_iterator_t iter, int handle, size_t *plen)
> | ~~~~~~~~^~~~
> window.c:1887:39: error: passing argument 4 of 'printed_representation' from incompatible pointer type [-Wincompatible-pointer-types]
> 1887 | &replen);
> | ^~~~~~~
> | |
> | int *
Note the difference in flag rendering around the warning.
https://gist.github.com/egmontkob/eb114294efbcd5adb1944c9f3cb5feda has a
good sumamry of why it happens. Befomre the change `nix` was handling
just one form or URL separator:
$ printf '\e]8;;http://example.com\e\\This is a link\e]8;;\e\\\n'
Now it also handled another for (used by gcc-14`):
printf '\e]8;;http://example.com\aThis is a link\e]8;;\a\n'
While at it fixed accumulation of trailing escape `\e\\` symbol.
sometimes it's these little things that let beginners stumble at the
first step...
mentioning one potentially foreign concept while introducing an entirely
new concept is asking enough already.
The explicit include is needed for clangd to not get confused somehow,
which is also what threw me off initially and made me pick the wrong
constructor.
The (pointer, number, number) constructor first constructs a C string
and then takes a substring from that, but we didn't specify that the
buffer needs to be NUL-terminated, and then what would be the point of
the size argument anyway...
basic_string.h:
> basic_string(const _Tp& __t, size_type __pos, size_type __n,
> const _Alloc& __a = _Alloc())
> : basic_string(_S_to_string_view(__t).substr(__pos, __n), __a) { }
Valgrind on nixops4/rust/nix-flake tests:
==1344422== Conditional jump or move depends on uninitialised value(s)
==1344422== at 0x48513E8: strlen (vg_replace_strmem.c:505)
==1344422== by 0x488E941: UnknownInlinedFun (char_traits.h:391)
==1344422== by 0x488E941: UnknownInlinedFun (string_view:141)
==1344422== by 0x488E941: UnknownInlinedFun (basic_string.h:790)
==1344422== by 0x488E941: nix_flake_reference_and_fragment_from_string (nix_api_flake.cc:81)
==1344422== by 0x127332: nix_flake::FlakeReference::parse_with_fragment (lib.rs:123)
It seems that the intention was to format a number in base 8 (as
suggested by the %o format specifier), but `perms` is a `std::string`
and not a number. Looks like `rawMode` is the correct thing to use here.
This name is close to the Nixpkgs lib function `escapeShellArg`,
making it easier to find.
A friendlier function with the same behavior as lib could be added
later.
Fixes
../src/libstore/unix/build/derivation-builder.cc:1130:86: warning: the compiler can assume that the address of ‘nix::DerivationBuilderParams::drv’ will never be NULL [-Waddress]
1130 | if (useChroot && settings.preBuildHook != "" && dynamic_cast<const Derivation *>(&drv)) {
| ^~~~
Assuming this check was left over from the time `drv` could be a
`BasicDerivation`.
I split it out before to try to separate the building logic, but now we
have the much better `DerivationBuilder` abstraction for that. With that
change, I think `LocalDerivationGoal` has outlived its usefulness.
We just inline it back into `DerivationGoal`, and do so with minimal
`#ifdef` for Windows.
Note that the order of statements in `~DerivationGoal` is different than
it was after the `~LocalDerivationGoal` split, but it is *restored* to
the way it original was before --- evidently I did the split slightly
wrong, but nobody noticed, probably because the order doesn't actually
matter.
Co-authored-by: Robert Hensing <roberth@users.noreply.github.com>
... by moving our stuff out of the way from upstream's
`nixComponents` and `nixDependencies` attrsets.
(I prefer not to use overlays, but let's make it work this way
first)
Now that `DerivationBuilder` is created after the underlying data has
already been initialized, we can just refer this data normally, with a
direct reference.
Only `parsedDrv` takes a (borrowing) pointer, because independent of
initialization the derivation may or may not have structured attrs.
While it seems unlikely that `$USER` will be unset while `$HOME` is set,
as `$USER` is not used in the script and as `nix-profile-daemon.fish` is
not checking `$USER`, it seems better to remove this check.
`nix-profile.fish` and `nix-profile-daemon.fish` now become identical.
It seems reasonable to add both `$HOME/.profile/bin` and
`@localstatedir@/nix/profiles/default/bin` to `$PATH` for both user
local and daemon based nix execution. Nix daemon execution mode does
not affect these path.
The building logic is now free of the scheduling logic!
(The interface between them is just what is in the new header. This
makes it much easier to audit, and shrink over time.)
We have a new `DerivationBuilder` struct, and `DerivationBuilderParams`
`DerivationBuilderCallbacks` supporting it.
`LocalDerivationGoal` doesn't subclass any of these, so we are ready to
now move them out to a new file!
Now, most of it is in two new functions:
`LocalDerivationGoal::{,un}repareBuild`.
This might seems like a step backwards from coroutines --- now we have
more functions, and are stuck with class vars --- but I don't think it
needs to be.
There's a few options here:
- (Re)introduce coroutines for the isolated building logic. We could use the
same coroutines types, or simpler ones specialized to this use-case.
The `tryLocalBuild` caller can still use `Goal::Co`, and just will
manually "pump" this inner coroutine.
- Return closures from each step. This is sort of like coroutines by
hand, but it still allows us to stop writing down the local variables
in each type.
Being able to fully-use RAII again would be very nice!
- Keep top-level first-order functions like now, but make more
functional. Instead of having one state object (`DerivationBuilder`)
for all steps (setup, run, teardown), we can have separate structs for
the live variables at each point we consume and return.
This at least avoids "are these variables active at this time?"
questions, but doesn't give us the full benefit of RAII as we must
manually ensure FIFO create/destroy orders still.
One thing to note is that by keeping the `outputLock` unlocking in
`tryLocalBuild`, we are arguably uncovering a rebuild scheduling vs
building distinction, as the output locks are pretty squarely a
scheduling concern. It's nice that the builder doesn't need to know
about them at all.
We had fields set to the same values before in our test data. This is
not a problem per-se, but does mean we wouldn't catch certain mixups.
Now, the fields are set to distinct values (where possible), which makes
the test more robust.
This requires that we refer to the `sourceInfo` instead of the
`result`. However, `sourceInfo` does not create a chain of basedir
resolution, so we add that back with `flakeDir`.
Only a much smaller `StructuredAttrs` remains, the rest is is now moved
to `DerivationOptions`.
This gets us quite close to `std::optional<StructuredAttrs>` and
`DerivationOptions` being included in `Derivation` as fields.
Now, both the unit and functional tests relating to derivation options
are tested both ways -- with input addressing and content-addressing
derivations.
This is a small optimization used when we're signing a narinfo for
multiple keys in one go. Using this sign variant, we only compute the
NAR fingerprint once, then sign it with all the keys.
It is just use for adding context to errors, but we have `addTrace` to
do that. Let the callers do that instead.
The callers doing so is a bit duplicated, yes, but this will get better
once `DerivationOptions` is included in `Derivation`.
This fixes evaluation for Windows. There are unfortunately deps that
still don't build, but this can be fixed next.
Flake lock file updates:
• Updated input 'nixpkgs':
'github:NixOS/nixpkgs/52faf482a3889b7619003c0daec593a1912fddc1?narHash=sha256-6hl6L/tRnwubHcA4pfUUtk542wn2Om%2BD4UnDhlDW9BE%3D' (2025-03-30)
→ 'github:NixOS/nixpkgs/f675531bc7e6657c10a18b565cfebd8aa9e24c14?narHash=sha256-gbl9hE39nQRpZaLjhWKmEu5ejtQsgI5TWYrIVVJn30U%3D' (2025-04-09)
The STSProfileCredentialsProviders allows to assume a specific IAM role
when accessing an S3 repository. Sometimes this is needed to obtain the
permissions to operate on the bucket.
The "derivation" is one of the key concepts and captures the most distinctive aspect of Nix:
that we work with a certain type data (linked files) in a certain manner (using pure functions).
Here we finally arrange all the important pieces to show how they belong
together, while referring to the respective reference documentation for details.
This change also unbreaks downstream links to `//glossary#gloss-derivation`,
which had been broken by removing the glossary entry
Nix shipping with Yet Another Programming Language is often questioned
among beginners. This change highlights distinctive aspects of the Nix
language to ease the learning curve and better orient readers around
what really matters for using Nix.
Since it's on topic, this change also polishes the wording on the motivation for string contexts.
This change follows the definition from aptitude, but using precise notions from Nix:
> package managers deal with packages: collections of files that are
> bundled together and can be installed and removed as a group.
> [...]
> If a package A depends upon another package B, then B is required
> for A to operate properly.
> [...]
> The job of a package manager is to present an interface which assists
> the user in managing the collection of packages installed on his or her system.
>
> -- <https://www.debian.org/doc/manuals/aptitude/pr01s02.en.html>
An interesting addition:
> Packages are abstractions defining the granularity at which users can act
> (add, remove, upgrade, etc.) on available software.
> A distribution is a collection of packages maintained (hopefully) coherently.
>
> -- Package Upgrades in FOSS Distributions: Details and Challenges
> (Roberto Di Cosmo, Stefano Zacchiroli; 2009) <https://arxiv.org/pdf/0902.1610>
Notably these quotes and this change don't say anything about installation,
or what it means for software to be available. In practice, this is
handled downstream, e.g. in NixOS or Home Manager. Nix historically
provides rudimentary facilities for package management such as
`nix-env`, but I claim they are widely agreed upon being discouraged,
with plenty of arguments provided in <https://stop-using-nix-env.privatevoid.net>.
Similarly, the specific structure of packages is determined downstream,
since Nix is policy-free:
> Nix is policy-free; it provides mechanisms to implement various deployment policies, but does not enforce a specific one.
>
> -- The Purely Functional Software Deployment Model (Eelco Dolstra; 2006) <https://edolstra.github.io/pubs/phd-thesis.pdf>
Specifically, Nix mechanisms do not define what a package is supposed to be:
> It's worth noting that the Nix language is intended as a DSL for package and configuration management, but it has no notions of "packages" or "configurations".
>
> -- <https://gist.github.com/edolstra/29ce9d8ea399b703a7023073b0dbc00d>
This is why we say, Nix *allows* denoting packages in a certain way, but
doesn't enforce any particular way.
Rather than "mounting" the store inside an empty virtual filesystem,
just return the store as a virtual filesystem. This is more modular.
(FWIW, it also supports two long term hopes of mind:
1. More capability-based Nix language mode. I dream of a "super pure
eval" where you can only use relative path literals (See #8738), and
any `fetchTree`-fetched stuff + the store are all disjoint (none is
mounted in another) file systems.
2. Windows, where the store dir may include drive letters, etc., and is
thus unsuitable to be the prefix of any `CanonPath`s.
)
Co-authored-by: Eelco Dolstra <edolstra@gmail.com>
Add a `secretKeyFiles` URI parameter in the store URIs receiving a
coma-separated list of Nix signing keyfiles.
For instance:
nix copy --to "file:///tmp/store?secret-keys=/tmp/key1,/tmp/key2" \
"$(nix build --print-out-paths nixpkgs#hello)"
The keys passed through this new store URI parameter are merged with
the key specified in the `secretKeyFile` parameter, if any.
We'd like to rotate the signing key for cache.nixos.org. To simplify
the transition, we'd like to sign the new paths with two keys: the new
one and the current one. With this, the cache can support nix
configurations only trusting the new key and legacy configurations
only trusting the current key.
See https://github.com/NixOS/rfcs/pull/149 for more informations
behind the motivation.
nix> meson.build:216: WARNING: Project targets '>= 1.1' but uses feature introduced in '1.4.0': fs.name with build_tgt, custom_tgt, and custom_idx.
nix> meson.build:222: WARNING: Project targets '>= 1.1' but uses feature introduced in '1.4.0': fs.name with build_tgt, custom_tgt, and custom_idx.
nix> meson.build:235: WARNING: Project targets '>= 1.1' but uses feature introduced in '1.4.0': fs.name with build_tgt, custom_tgt, and custom_idx.
nix> meson.build:236: WARNING: Project targets '>= 1.1' but uses feature introduced in '1.4.0': fs.name with build_tgt, custom_tgt, and custom_idx.
nix> meson.build:242: WARNING: Project targets '>= 1.1' but uses feature introduced in '1.4.0': fs.name with build_tgt, custom_tgt, and custom_idx.
Revert most of "Hack together a fix for the public headers"
- The `libmain` change is kept, and one more libmain change is made.
(Need to update Meson and Nix per the package alike).
- The S3 situation is fixed in a different way: the variable is public
now, used in the header, and fixed accordingly.
- Fix TODO for `HAVE_EMBEDDED_SANDBOX_SHELL`
This reverts commit 2b51250534.
Previously, when lock file contained invalid JSON nix reported a parser
error without specifying the file it came from.
This change adds flake.lock file path to the error message to avoid
confusion.
Some of the maintainer attribute names got changed in nixos-unstable
(e.g. "edolstra" is now "eelco") but we want this flake to work on
nixos-24.11. So just get rid of them.
This is required for other bindings like nix-flake-c to hook into
nix-expr-c appropriately.
The `_internal` part should be a sufficient deterrent normally,
and it may also be useful for bindings that migrate from the C++
interface.
For people self-hosting caches that can be occasionally down, the
default timeout is very long. This is annoying if you are trying to
update your binary cache at the same time you are trying to update
another machine. Same if cachix has one of its rare hiccups.
We tested this value of 5s in srvos now for years and we like to travel
around the world with shitty internet, so it should be still reasonable
high.
For example, instead of doing
#include "nix/store-config.hh"
#include "nix/derived-path.hh"
Now do
#include "nix/store/config.hh"
#include "nix/store/derived-path.hh"
This was originally planned in the issue, and also recent requested by
Eelco.
Most of the change is purely mechanical. There is just one small
additional issue. See how, in the example above, we took this
opportunity to also turn `<comp>-config.hh` into `<comp>/config.hh`.
Well, there was already a `nix/util/config.{cc,hh}`. Even though there
is not a public configuration header for libutil (which also would be
called `nix/util/config.{cc,hh}`) that's still confusing, To avoid any
such confusion, we renamed that to `nix/util/configuration.{cc,hh}`.
Finally, note that the libflake headers already did this, so we didn't
need to do anything to them. We wouldn't want to mistakenly get
`nix/flake/flake/flake.hh`!
Progress on #7876
Previously, when users entered an incomplete expression in the REPL,
the continuation prompt was just 10 blank spaces, which looked invisible
and gave the impression that the REPL had stalled.
This change updates the prompt to " > ", aligning it visually
with 'nix-repl> ' and clearly indicating that the REPL is waiting for
more input.
Fixes: https://github.com/NixOS/nix/issues/12702
There are two big changes:
1. Public and private config is now separated. Configuration variables
that are only used internally do not go in a header which is
installed.
(Additionally, libutil has a unix-specific private config header,
which should only be used in unix-specific code. This keeps things a
bit more organized, in a purely private implementation-internal way.)
2. Secondly, there is no more `-include`. There are very few config
items that need to be publically exposed, so now it is feasible to
just make the headers that need them just including the (public)
configuration header.
And there are also a few more small cleanups on top of those:
- The configuration files have better names.
- The few CPP variables that remain exposed in the public headers are
now also renamed to always start with `NIX_`. This ensures they should
not conflict with variables defined elsewhere.
- We now always use `#if` and not `#ifdef`/`#ifndef` for our
configuration variables, which helps avoid bugs by requiring that
variables must be defined in all cases.
See comments on the script; this is supposed to avoid breaking muscle
memory without complicating the build system (which proved harder than I
thought too) or not doing the header hygiene change at all.
link-headers: use pathlib consistenly and fix type errors
The short answer for why we need to do this is so we can consistently do
`#include "nix/..."`. Without this change, there are ways to still make
that work, but they are hacky, and they have downsides such as making it
harder to make sure headers from the wrong Nix library (e..g.
`libnixexpr` headers in `libnixutil`) aren't being used.
The C API alraedy used `nix_api_*`, so its headers are *not* put in
subdirectories accordingly.
Progress on #7876
We resisted doing this for a while because it would be annoying to not
have the header source file pairs close by / easy to change file
path/name from one to the other. But I am ameliorating that with
symlinks in the next commit.
- Since it's now private, give it a rename. Note that I want to switch the
word order on the public ones too.
- Since it is only needed by two files, just include there rather than
the nasty blanket-forced thing.
On my system (Ubuntu 24.04 with nix installed using
https://zero-to-nix.com/), I noticed that my PATH
contained multiple times the following entries:
/home/thomas/.nix-profile/bin
/nix/var/nix/profiles/default/bin
Fix it by inserting a missing `export`, to make
sure `nix-daemon.sh` is really only executed once.
This makes it behave the same as nix-daemon. Opening the store in the
parent can cause a SIGBUS in libsqlite in the child:
#0 0x00007f141cf6f789 in __memset_avx2_unaligned_erms () from /nix/store/wn7v2vhyyyi6clcyn0s9ixvl7d4d87ic-glibc-2.40-36/lib/libc.so.6
#1 0x00007f141c322fe8 in walIndexAppend () from /nix/store/bbd59cqw259149r2ddk4w1q0lr2fch8c-sqlite-3.46.1/lib/libsqlite3.so.0
#2 0x00007f141c3711a2 in pagerWalFrames () from /nix/store/bbd59cqw259149r2ddk4w1q0lr2fch8c-sqlite-3.46.1/lib/libsqlite3.so.0
#3 0x00007f141c38317e in sqlite3PagerCommitPhaseOne.part.0 () from /nix/store/bbd59cqw259149r2ddk4w1q0lr2fch8c-sqlite-3.46.1/lib/libsqlite3.so.0
#4 0x00007f141c383555 in sqlite3BtreeCommitPhaseOne.part.0 () from /nix/store/bbd59cqw259149r2ddk4w1q0lr2fch8c-sqlite-3.46.1/lib/libsqlite3.so.0
#5 0x00007f141c384797 in sqlite3VdbeHalt () from /nix/store/bbd59cqw259149r2ddk4w1q0lr2fch8c-sqlite-3.46.1/lib/libsqlite3.so.0
#6 0x00007f141c3b8f60 in sqlite3VdbeExec () from /nix/store/bbd59cqw259149r2ddk4w1q0lr2fch8c-sqlite-3.46.1/lib/libsqlite3.so.0
#7 0x00007f141c3bbfef in sqlite3_step () from /nix/store/bbd59cqw259149r2ddk4w1q0lr2fch8c-sqlite-3.46.1/lib/libsqlite3.so.0
#8 0x00007f141c3bd0e5 in sqlite3_exec () from /nix/store/bbd59cqw259149r2ddk4w1q0lr2fch8c-sqlite-3.46.1/lib/libsqlite3.so.0
#9 0x00007f141da140e0 in nix::SQLiteTxn::commit() () from /nix/store/1m4r8s7s1v54zq9isncvjgia02bffxlz-determinate-nix-store-3.1.0/lib/libnixstore.so
#10 0x00007f141d9ce69c in nix::LocalStore::registerValidPaths(std::map<nix::StorePath, nix::ValidPathInfo, std::less<nix::StorePath>, std::allocator<std::pair<nix::StorePath const, nix::ValidPathInfo> > > const&)::{lambda()#1}::operator()() const () from /nix/store/1m4r8s7s1v54zq9isncvjgia02bffxlz-determinate-nix-store-3.1.0/lib/libnixstore.so
- Some headers were completely redundant and have been removed.
- Other headers have been turned private.
- Unnecessary meson.build code has been removed.
- libutil-tests now has a private config header, where previously
it had none. This removes the need to expose a package version
macro publicly.
We rely on `yyerror()` instead.
> The variable yynerrs contains the number of syntax errors reported so
> far.
> Normally this variable is global; but if you request a pure parser
> (see A Pure (Reentrant) Parser) then it is a local variable which only
> the actions can access.
https://www.gnu.org/software/bison/manual/html_node/Error-Reporting-Function.html
at least clang-tidy is not convinced that this initialized.
If this is not the case, the impact should be small and hopefully also
more robust if changed.
Otherwise without the change the test fails on nix-2.26 as:
error: derivation contains an illegal reference specifier 'dev'
Note: the error message does not match intended change.
Before the change "illegal reference" was hard to interpret as it did
not mention what derivation actually hits it.
Today's `nixpkgs` example:
Before the change:
$ nix build --no-link -f. postgresql_14
...
error: derivation contains an illegal reference specifier 'man'
After the change:
$ nix build --no-link -f. postgresql_14
...
error: derivation '/nix/store/bxp6g57limvwiga61vdlyvhy7i8rp6wd-postgresql-14.15.drv' output check for 'lib' contains an illegal reference specifier 'man', expected store path or output name (one of [debug, dev, doc, lib, out])
After the previous commit it should not be necessary. Furthermore, if we
*do* sleep, we'll exacerbate a race condition (in conjunction with
getting rid of the thread cancellation) that will cause test failures.
This was filed as https://github.com/nixos/nix/issues/7584, but as far
as I can tell, the previous solution of POLLHUP works just fine on macOS
14. I've also tested on an ancient machine with macOS 10.15.7, which
also has POLLHUP work correctly.
It's possible this might regress some older versions of macOS that have
a kernel bug, but I went looking through the history on the sources and
didn't find anything that looked terribly convincingly like a bug fix
between 2020 and today. If such a broken version exists, it seems pretty
reasonable to suggest simply updating the OS.
Change-Id: I178a038baa000f927ea2cbc4587d69d8ab786843
Based off of commit 69e2ee5b25752ba5fd8644cef56fb9d627ca4a64. Ericson2314 added
additional other information.
On https://github.com/NixOS/nix/issues/8946, we faced a surprising
behaviour wrt. exception when using pthread_cancel. In a nutshell when
a thread is inside a catch block and it's getting pthread_cancel by
another one, then the original exception is bubbled up and crashes the
process.
We now poll on the notification pipe from the thread and exit when the
main thread closes its end. This solution does not exhibit surprising
behaviour wrt. exceptions.
Co-authored-by: Mic92 <joerg@thalheim.io>
Fixes https://github.com/NixOS/nix/issues/8946
See also Lix https://gerrit.lix.systems/c/lix/+/1605 which is very
similar by coincidence. Pulled a comment from that.
Without this :u, :sh and :i repl commands fail with:
> Cannot run 'nix-shell'/`nix-env` because no method of calling the Nix
> CLI was provided. This is a configuration problem pertaining to how
> this program was built.
Remove the default ctor argument as it evidently makes catching
refactoring bugs much harder. `NixRepl` implementation lives completely
in `repl.cc`, so we can be as explicit as necessary.
We want the $doc, $man outputs to be symlinks pointing to nix-manual and
nix-manual.man. Creating the directories first makes the `ln` command
produce symlink $doc/${nix-manual} instead.
```
$file /nix/store/q4dwlnd36gpfajgfcp6hca2xwy068wjq-nix-2.27.1-man/rwh8ky3k040wyrywl8k2v5b3csdfbdg7-nix-manual-2.27.1-man
/nix/store/q4dwlnd36gpfajgfcp6hca2xwy068wjq-nix-2.27.1-man/rwh8ky3k040wyrywl8k2v5b3csdfbdg7-nix-manual-2.27.1-man:
symbolic link to /nix/store/rwh8ky3k040wyrywl8k2v5b3csdfbdg7-nix-manual-2.27.1-man
```
This is the reason `nix-env --help` is once again broken on 2.26/2.27/master
after 4108529.
Instead of calling `worker.waitForAWhile(shared_from_this())` etc.,
the subclasses of Goal instead call protected functions defined in Goal
that abstract over these.
The code for awaiting has also been heavily simplified.
Instead of calling `addWaitee`, then suspending,
`co_await await(waitees)` is called once, which also handles the suspend.
The end-goal is to remove all manual `co_await Suspend{}`s.
We now see exception beeing thrown when remote building in master
because of writing to a non-blocking file descriptor from our json logger.
> #0 0x00007f2ea97aea9c in __pthread_kill_implementation () from /nix/store/wn7v2vhyyyi6clcyn0s9ixvl7d4d87ic-glibc-2.40-36/lib/libc.so.6
> #1 0x00007f2ea975c576 in raise () from /nix/store/wn7v2vhyyyi6clcyn0s9ixvl7d4d87ic-glibc-2.40-36/lib/libc.so.6
> #2 0x00007f2ea9744935 in abort () from /nix/store/wn7v2vhyyyi6clcyn0s9ixvl7d4d87ic-glibc-2.40-36/lib/libc.so.6
> #3 0x00007f2ea99e8c2b in __gnu_cxx::__verbose_terminate_handler() [clone .cold] () from /nix/store/ybjcla5bhj8g1y84998pn4a2drfxybkv-gcc-13.3.0-lib/lib/libstdc++.so.6
> #4 0x00007f2ea99f820a in __cxxabiv1::__terminate(void (*)()) () from /nix/store/ybjcla5bhj8g1y84998pn4a2drfxybkv-gcc-13.3.0-lib/lib/libstdc++.so.6
> #5 0x00007f2ea99f8275 in std::terminate() () from /nix/store/ybjcla5bhj8g1y84998pn4a2drfxybkv-gcc-13.3.0-lib/lib/libstdc++.so.6
> #6 0x00007f2ea99f84c7 in __cxa_throw () from /nix/store/ybjcla5bhj8g1y84998pn4a2drfxybkv-gcc-13.3.0-lib/lib/libstdc++.so.6
> #7 0x00007f2eaa5035c2 in nix::writeFull (fd=2, s=..., allowInterrupts=true) at ../unix/file-descriptor.cc:43
> #8 0x00007f2eaa5633c4 in nix::JSONLogger::write (this=this@entry=0x249a7d40, json=...) at /nix/store/4krab2h0hd4wvxxmscxrw21pl77j4i7j-gcc-13.3.0/include/c++/13.3.0/bits/char_traits.h:358
> #9 0x00007f2eaa5658d7 in nix::JSONLogger::logEI (this=<optimized out>, ei=...) at ../logging.cc:242
> #10 0x00007f2ea9c5d048 in nix::Logger::logEI (ei=..., lvl=nix::lvlError, this=0x249a7d40) at /nix/store/a7cq5bqh0ryvnkv4m19ffchnvi8l9qx6-nix-util-2.27.0-dev/include/nix/logging.hh:108
> #11 nix::handleExceptions (programName="nix", fun=...) at ../shared.cc:343
> #12 0x0000000000465b1f in main (argc=<optimized out>, argv=<optimized out>) at /nix/store/4krab2h0hd4wvxxmscxrw21pl77j4i7j-gcc-13.3.0/include/c++/13.3.0/bits/allocator.h:163
> (gdb) frame 10
> #10 0x00007f2ea9c5d048 in nix::Logger::logEI (ei=..., lvl=nix::lvlError, this=0x249a7d40) at /nix/store/a7cq5bqh0ryvnkv4m19ffchnvi8l9qx6-nix-util-2.27.0-dev/include/nix/logging.hh:108
> 108 logEI(ei);
So far only drainFD sets the non-blocking flag on a "readable" file descriptor,
while this is a "writeable" file descriptor.
It's not clear to me yet, why we see logs after that point, but it's
also not that bad to handle EAGAIN in read/write functions after all.
Now that we have coroutines, we can go back to loops and regular RAII,
which is must less error-proone!
I look forward to removing the other instances!
Perhaps more significantly, it no longer knows about
`LocalDerivationGoal`, and without any effort it also compiles on
Windows just fine. (`local-derivation-goal.{cc,hh}` is currently skipped
on Windows.)
Default: istty(stdout)
This refactors `nix develop` internals a bit to use the `json` type
more. The assertion now operates in the in-memory json instead of
re-parsing it. While this is technically a weaker guarantee, we
should be able to rely on the library to get this right. It's its
most essential purpose.
The underlying issue is that debugger code path was
calling PosTable::operator[] in each eval method.
This has become incredibly expensive since 5d9fdab3de.
While we are it it, I've reworked the code to
not use std::shared_ptr where it really isn't necessary.
As I've documented in previous commits, this is actually
more a workaround for recursive header dependencies now
and is only necessary in `error.hh` code.
Some ad-hoc benchmarking:
After this commit:
```
Benchmark 1: nix eval nixpkgs#hello --impure --ignore-try --no-eval-cache --debugger
Time (mean ± σ): 784.2 ms ± 7.1 ms [User: 561.4 ms, System: 147.7 ms]
Range (min … max): 773.5 ms … 792.6 ms 10 runs
```
On master 3604c7c51:
```
Benchmark 1: nix eval nixpkgs#hello --impure --ignore-try --no-eval-cache --debugger
Time (mean ± σ): 22.914 s ± 0.178 s [User: 18.524 s, System: 4.151 s]
Range (min … max): 22.738 s … 23.290 s 10 runs
```
All of this code doesn't actually depend on anything from
libexpr. Because Pos is so tigtly coupled with Error, it
makes sense to have in the same library.
The simplification here is due to a long-standing bug, but it is not
worth fixing the bug at this time. Instead we've finally written up an
issue for the bug, and referenced the issue number in the code.
In the local building case, there is many things which can through
`BuildError`, but in the hook case there is just this one. We can
therefore simplify the code by "cinching" down the logic just to the
spot the error is thrown.
There is other code outside `libstore/build` which also uses
`BuildError`, but I believe those cases are mistakes. The point of
`BuildError` is the narrow technical use-cases of "errors which should
not be fatal with `--keep-going`". Using it outside the
building/scheduling code doesn't really make sense in that regard. It
seems likely that those usages were instead merely because "oh, this
error has something to do with building, so I guess `BuildError` is
better than `Error`".
It is quite likely that I myself used `BuildError` incorrectly as
described above :).
The variables are only set by CGroup mechanisms in `killSandbox` in the
local build. In the build hook case, these variables will not be set, so
there is nothing to do.
We can move this method from `LocalStore` to `Store` --- even if we only
want the actual builder to sign things in many cases, there is no reason
to try to enforce this policy by spurious moving the method to a
subclass.
Now, we might technically sign class, but CA derivations is
experimental, and @Ericson2314 is going to revisit all this stuff with
issue #11896 anyways.
- `chrootParentDir` can be a local variable instead of a class variable.
- `getChildStatus` can be inlined. Again, we have the `assert(!hook);`
in the local building case, which makes for a simpler thing inlined.
Thanks to the previous commit, we can inline all these small callbacks.
In the build-hook case, they were empty, and now they disappear
entirely.
While `LocalDerivationGoal` can be used in the hook case (we use it
based on whether we have a local store, not based on whether we are
using the build hook, a decision which comes later), the previous
commit's inline moved the code into a spot where we know we are cleaning
up after local building, *not* after running the build hook. This allows
for much more simplification.
The basic idea is that while we have duplicated this function, we now
have one call-site in the local build case, and one call site in the
build hook case. This unlocks big opportunities to specialize each copy,
since they really shouldn't be doing the same things. By the time we are
are done, there should not be much duplication left.
See #12628 for further info.
This is supposed to firstly improve the docs as they are, and secondly
hint at how the core conceptual information ought to be moved to the
store derivation section of the manual.
Co-authored-by: Jörg Thalheim <Mic92@users.noreply.github.com>
This variable was once used to communicate already acquired store path
locks between Nix and the build hook, but this hasn't been the case
since 9bcb4d2dd9. So let's get rid of
it.
It was first introduced in 19e0ce2c03
In Nix we only register the crash handler in main instead of initNix,
because library user may want to use their own crash handler.
Sample output:
Mar 12 08:38:06 eve nix[2303762]: Nix crashed. This is a bug. Please report this at https://github.com/NixOS/nix/issues with the following information included:
Mar 12 08:38:06 eve nix[2303762]: Exception: nix::SysError: error: writing to file: Resource temporarily unavailable
Mar 12 08:38:06 eve nix[2303762]: Stack trace:
Mar 12 08:38:06 eve nix[2303762]: 0# 0x000000000076876A in nix
1# 0x00007FDA40E9F20A in /nix/store/2lhklm5aizx30qbw49acnrrzkj9lbmij-gcc-14-20241116-lib/lib/libstdc++.so.6
2# std::unexpected() in /nix/store/2lhklm5aizx30qbw49acnrrzkj9lbmij-gcc-14-20241116-lib/lib/libstdc++.so.6
3# 0x00007FDA40E9F487 in /nix/store/2lhklm5aizx30qbw49acnrrzkj9lbmij-gcc-14-20241116-lib/lib/libstdc++.so.6
4# nix::writeFull(int, std::basic_string_view<char, std::char_traits<char> >, bool) in /home/joerg/git/nix/inst/lib/libnixutil.so
5# nix::writeLine(int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >) in /home/joerg/git/nix/inst/lib/libnixutil.so
6# nix::JSONLogger::write(nlohmann::json_abi_v3_11_3::basic_json<std::map, std::vector, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, bool, long, unsigned long, double, std::allocator, nlohmann::json_abi_v3_11_3::adl_serializer, std::vector<unsigned char, std::allocator<unsigned char> >, void> const&) in /home/joerg/git/nix/inst/lib/libnixutil.so
7# nix::JSONLogger::logEI(nix::ErrorInfo const&) in /home/joerg/git/nix/inst/lib/libnixutil.so
8# nix::Logger::logEI(nix::Verbosity, nix::ErrorInfo) in nix
9# nix::handleExceptions(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::function<void ()>) in /home/joerg/git/nix/inst/lib/libnixmain.so
10# 0x000000000087A563 in nix
11# 0x00007FDA40BD41FE in /nix/store/6q2mknq81cyscjmkv72fpcsvan56qhmg-glibc-2.40-66/lib/libc.so.6
12# __libc_start_main in /nix/store/6q2mknq81cyscjmkv72fpcsvan56qhmg-glibc-2.40-66/lib/libc.so.6
13# 0x00000000006F4DF5 in nix
Co-authored-by: eldritch horrors <pennae@lix.systems>
1. Fix confusing wording that might imply unnecessary double-hashing.
2. Add references to specifics of base-32 encoding.
3. Fix incorrect description that sha256 hash of `fingerprint` is
truncated. "Truncated" is actual wording used in Nix theses, but it has
unusual meaning, that is better conveyed by word "compressed", which is
used by the reference C++ implementation.
4. Clarify details of base16 encoding.
Doing this makes catching non-obvious bugs easier. GHA CI workload is
already a concern and there isn't much benefit in running the tests with
and without sanitizers at the same time, so UBSAN is enabled for default
checks.
This change doesn't affect production builds in any way, but is rather a
step in the direction of improving automated testing during development.
Relates to #10969.
This fixes a few of the property tests, now that the property tests
are actually generating arbitrary data - some of that data now
requiring experimental features to function properly.
Here we're switching to combinators instead of dereference operator.
It turns out the dereference operator was being executed upon test
setup, meaning that we were only using a only single value for each of
the executions of the property tests! Really not good.
And on Windows, we instead get:
operator* is not allowed in this context
ff6af6fc68/src/gen/detail/GenerationHandler.cpp (L16C31-L16C71)
Now a few of the property tests fail, because we're generating cases
which haven't been exercised before.
I refactored the way that input resolution works in `DerivationGoal`. To
be honest, it is probably unclear to the reader whether this new way is
better or worse. I suppose *intrinsic* motivation, I can say that
- the more structured use of `inputGoal` (a local variable) is better
than the shotgrun approach with `inputDrvOutputs`
- A virtual `waiteeDone` was a hack, and now it's gone.
However, the *real* motivation of this is not the above things, but that
it is needed for my mammoth refactor fixing #11897 and #11928.
It is nice that this step could come first, rather than making that
refactor even bigger.
The bug reappeared after all, and the fix introduced a different bug. I
just reverted on 2.27 first, in #12576, but upon further introspection
and discussion with @roberth, with preparing for and travelling to
Planet Nix I will not be able to fix it on `master` soon enough for a
revert to not be warranted here in the meantime also.
This reverts commit c98525235f.
This should fix a few packaging regressions.
`dev` also includes a merged `includes/`, which may be helpful until
inter-component includes are fixed properly.
It's not very clear what the ownership model is here, but one thing
is certain: `.up` can't be destroyed before the StaticEnv that refers
to it is.
Changing a non-owning pointer to taking shared ownership of the parent
`StaticEnv` prevents the `.up` from being freed.
I'm not a huge fan of the inverted ownership, where child `StaticEnv`
takes a refcount of the parent, but this seems like the least intrusive
way to fix the use-after-free.
This shouldn't cause any shared_ptr cycles to appear (hopefully).
When #9863 converted the `Nix::Store` free functions into member functions, the
implicit `this` argument was not accounted for when iterating over the variable
number of arguments in some functions.
Seems like this got dropped at some point during meson migration, so
put it back in the build system.
Drop all tests for `parseGitUrl`, since that function doesn't exist
and migrating doesn't look sensible because git-lfs stuff seems to use
`ParsedURL`.
Note that in pure mode, we don't need to use the union FS even when
using a chroot store, since the user shouldn't have access to the
physical /nix/store.
This makes `nix.version` quicker to evaluate, which should speed up
package listing operations.
If you want an accurate count, use `lib.optionals` in your override
instead of `null` values.
By appending a colon to MANPATH NIX_MAN_DIR gets prepended to the
final MANPATH before default search paths.
This makes man still consider default search paths, but prefers
NIX_MAN_DIR (if it exists).
It still makes sense to point NIX_MAN_DIR to a correct location
by moving man pages build from nix-manual.man to nix-cli.man, but
this should fix most common use-cases where nix is installed globally.
... as intended.
Requirements:
- don't build fresh libraries for each git commit
- have git commit in the CLI
Bug:
- echo ${version} went into the wrong file => use the fact that it's
a symlink, not just for reading but also for writing.
Logging to another Logger was kind of nonsensical - it was really just
an easy way to get it to write its output to stderr, but that only
works if the underlying logger writes to stderr.
This change is needed to make it easy to log JSON output somewhere
else (like a file or socket).
It is unused in Nix currently, but will be used in Hydra. This reflects
what Hydra does in https://github.com/NixOS/hydra/pull/1387.
We may probably to use it more widely for better SSH store performance,
but this needs to be subject to more testing before we do that.
This is a first step towards PR #10760, and the issues it addresses.
See the Doxygen for details.
Thanks to these changes, we are able to drastically restrict how the
rest of the code-base uses `ParseDerivation`.
Co-Authored-By: HaeNoe <git@haenoe.party>
It's best we teach users that the "foo" derivation is less than pure in the sense that it cannot be built just on any system, in particular that builders cannot be selected arbitrarily but based on their system-features. The `"recursive-nix"` system-feature is automatically defined by `--extra-experimental-features recursive-nix`
I do not believe there is any problem with computing
`hashDerivationModulo` the normal way with impure derivations.
Conversely, the way this used to work is very suspicious because two
almost-equal derivations that only differ in depending on different
impure derivations could have the same drv hash modulo. That is very
suspicious because there is no reason to think those two different
impure derivations will end up producing the same content-addressed
data!
Co-authored-by: Alain Zscheile <zseri.devel@ytrizja.de>
E.g. in a derivation attribute `foo = ./bar`, if ./bar is a symlink,
we should copy the symlink to the store, not its target. This restores
the behaviour of Nix <= 2.19.
"content-address*ed*" derivation is misleading because all derivations
are *themselves* content-addressed. What may or may not be
content-addressed is not derivation itself, but the *output* of the
derivation.
The outputs are not *part* of the derivation (for then the derivation
wouldn't be complete before we built it) but rather separate entities
produced by the derivation.
"content-adddress*ed*" is not correctly because it can only describe
what the derivation *is*, and that is not what we are trying to do.
"content-address*ing*" is correct because it describes what the
derivation *does* --- it produces content-addressed data.
This is a big step documenting the store layer on its own, separately from the evaluator (and `builtins.derivation`).
Co-authored-by: Robert Hensing <roberth@users.noreply.github.com>
Curl creates sockets without setting FD_CLOEXEC/SOCK_CLOEXEC, this can
cause connections to remain open forever when using commands like `nix
shell`
This change sets the FD_CLOEXEC flag using a CURLOPT_SOCKOPTFUNCTION
callback.
This seems to be the way to do it now, even though I can't run them
without setting at least one env var.
I'll only fix shellcheck for now. Don't shoot the messenger.
It isn't quite clear to me why the previous commit masked this problem,
but I'm glad shellcheck has an effect or more effect now.
Note that this is just a script that is meant to run outside a
derivation (but also can be called by a derivation builder).
`touch $out` does not belong in it.
`touch $out` worked accidentally in the derivation-based check,
and also in the dev shell, but if pre-commit is invoked without
the dev shell it would fail.
This uses the single-threaded C-based routines from libblake3.
This is not optimal performance-wise but should be a good starting point
for nix compatibility with BLAKE3 hashing until a more performant
implementation based on the multi-threaded BLAKE3 routines
(written in Rust) can be developed.
This fixes dynamic derivations, reverting #9081.
I believe that this time around, #9052 is fixed. When I first rebased
this, tests were failing (which wasn't the case before). The cause of
those test failures were due to the crude job in which the outer goal
tried to exit with the inner goal's status.
Now, that error handling has been reworked to be more faithful. The exit
exit status and exception of the inner goal is returned by the outer
goal. The exception was what was causing the test failures, but I
believe it was not having the right error code (there is more than one
for failure) that caused #9081.
The only cost of doing things the "right way" was that I had to
introduce a hacky `preserveException` boolean. I don't like this, but,
then again, none of us like anything about how the scheduler works.
Issue #11927 is still there to clean everything up, subsuming the need
for any `preserveException` because I doubt we will be fishing
information out of state machines like this at all.
This reverts commit 8440afbed7.
Co-Authored-By: Eelco Dolstra <edolstra@gmail.com>
There seems to be no good reason for `nix-profile.fish` and
`nix-profile-daemon.fish` to differ in how they look for the location of
the `ca-bundle.crt` that might be installed by one of the packages.
As `$NIX_PROFILES` points to user local paths, not checking there is
strictly less useful, it seems?
Using `set --local` is better than using `set`/`set --erase`. `--local`
will preserve any existing `NIX_LINK` value. And the local variable is
automatically removed for any execution path.
This allows a flake to specify that it needs Git submodules to be
enabled (or disabled, if we ever change the default) on the top-level
flake. This requires the input to be refetched, but since the first
fetch is lazy, this shouldn't be expensive.
Currently the only attribute allowed by `inputs.self` is `submodules`,
but more can be added in the future (e.g. a `lazy` attribute to opt in
to lazy tree behaviour).
Fixes#5312, #9842.
It seems reasonable to add the `share` folder from the user profile into
`$XDG_DATA_DIRS` both for daemon and profile execution. Nix could add
package shared files into this folder regardless of how the nix daemon
itself is running.
If we previously fetched by revision, the output of "git ls-remote"
won't start with the expected line like
ref: refs/heads/master HEAD
but will be something like
5c4410e3b9891c05ab40d723de78c6f0be45ad30 refs/heads/5c4410e3b9891c05ab40d723de78c6f0be45ad30
This then causes Nix to treat that revision as a refname, which then
leads to warnings like
warning: could not update cached head '5c4410e3b9891c05ab40d723de78c6f0be45ad30' for 'file:///tmp/repo'
This causes Git to create a local ref named refs/head/<rev>, e.g.
$ git -C ~/.cache/nix/gitv3/11irpim06vj4h6c0w8yls6kx4hvl0qd0gr1fvk47n76g6wf1s1vk ls-remote --symref .
5c4410e3b9891c05ab40d723de78c6f0be45ad30 refs/heads/5c4410e3b9891c05ab40d723de78c6f0be45ad30
7f6bde8a20de4cccc2256f088bc5af9dbe38881d refs/heads/7f6bde8a20de4cccc2256f088bc5af9dbe38881d
which confuses readHead(), leading to errors like
fatal: Refusing to point HEAD outside of refs/
warning: could not update cached head 'd275d93aa0bb8a004939b2f1e87f559f989453be' for 'file:///tmp/repo'
Git interprets them as part of the file name, so passing parameters
like 'rev' breaks. Only relevant for testing (when _NIX_FORCE_HTTP is
set) and local bare repos.
Fixes messages like 'copying /tmp/repo/tmp/repo to the store'. The
PosixSourceAccessor already sets the prefix. Setting the prefix twice
shouldn't be a problem, but GitRepoImpl::getAccessor() returns a
wrapped accessor so it's not actually idempotent.
It seems that `meson test --print-errorlogs` only captures stderr,
so this makes it forward the logs as intended.
We might want to redirect stdout in our common setup script instead.
The main improvement is that the new message gives an example of a path
that is referenced, which should make it easier to track down. While
there, I also clarified the wording, saying exactly why the paths in
question were illegal.
Not sure what the intent was expecting help.sh to fail in the main suite, but it caused `meson test` to fail inside a `nix develop` shell:
$ meson test help --print-errorlogs
ninja: Entering directory `/home/eelco/Dev/nix-master/build'
1/1 nix-functional-tests:main / help UNEXPECTEDPASS 4.02s
I started getting these warnings `warning: download buffer is full; consider increasing the 'download-buffer-size' setting` but the documentation does not make it obvious what unit of measurement it accepts.
1. Fix this eval error:
https://hydra.nixos.org/jobset/nix/master#tabs-errors
The dev package output (actually a separate derivation) needs to skip
this for cross just as the main package output does.
2. Deduplicate libs attrset and list.
3. Move `nix-functional-tests` to `checkInputs`.
With the Meson build system, we no longer need a `check` vs
`install-check` distinction, so it is simpler to just keeep
everything in one place.
This label will be useful for constructing queries to find backportable PRs.
Specifically, those should omit both automatic backports and
"backports reviewed" PRs.
This allows RemoteStore::addMultipleToStore() to free the Source
objects early (and in particular the associated sinkToSource()
buffers). This should fix#7359. For example, memory consumption of
nix copy --derivation --to ssh-ng://localhost?remote-store=/tmp/nix --derivation --no-check-sigs \
/nix/store/4p9xmfgnvclqpii8pxqcwcvl9bxqy2xf-nixos-system-...drv
went from 353 MB to 74 MB.
Fixes
$ nix copy --derivation --to /tmp/nix /nix/store/...
error: cannot enqueue a work item while the thread pool is shutting down
The ThreadPoolShutDown exception was hiding the reason for the thread
pool shut down, e.g.
error: cannot add path '/nix/store/03sl46khd8gmjpsad7223m32ma965vy9-fix-static.patch' because it lacks a signature by a trusted key
In order for the script not be sourced multiple times by the same shell
instance, `__ETC_PROFILE_NIX_SOURCED` needs to be set with a `--global`
flag.
Both files are almost identical. And style differences make it harder
to see what is actually different and keep them in sync, when it is
required.
`nix-profile.fish` and part of `nix-profile-daemon.fish` use 4 space
indentation. Which is also the indentation that the fish shell
documentation is using.
Reformatting a chunk of `nix-profile-daemon.fish` from 2 space
indentation to 4 space indentation for consistency.
- Multiple choices of stdenv are handled more consistently, especially for the dev
shells which were previously not done correctly.
- Some stray nix code was moving into the `packaging` directory
nix-env can read priorities from a derivations meta attributes, but this
only works when installing a nix expression.
nix-env can also install bare store paths, however meta attributes are
not readable in that case. This means that a store path can not be
installed with a specific priority.
Some cases where it is advantageous to install a store path: a remote
host following a `nix copy`, or any time you want to save some
evaluation time and happen to already know the store path.
This PR addresses this shortcoming by adding a --priority flag to
nix-env --install.
Since ff8e2fe84e, 'path:' URLs on the
CLI are interpreted as relative to the current directory of the user,
not the path of the flake we're overriding.
During garbage collection we cache several things -- a set of known-dead
paths, a set of known-alive paths, and a map of paths to their derivers.
Currently they use STL maps and sets, which are ordered structures that
typically are backed by binary trees. Since we are putting pseudorandom
paths into these and looking them up by exact key, we don't need the
ordering, and we're paying a nontrivial cost per insertion.
The existing maps require O(n log n) memory and have O(log n) insertion
and lookup time.
We could instead use unordered maps, which are typically backed by
hashmaps. These require O(n) memory and have O(1) insertion and lookup
time.
On my system this appears to result in a dramatic speedup -- prior to
this patch I was able to delete 400k paths out of 9.5 million over the
course of 34.5 hours. After this patch the same result took 89 minutes.
This result should NOT be taken at face value because the two runs
aren't really comparable; in particular the first started when I had 9.5
million store paths and the seconcd started with 7.8 million, so we are
deleting a different set of paths starting from a much cleaner
filesystem. But I do think it's indicative.
Related: https://github.com/NixOS/nix/issues/9581
This fixes segfaults with nix copy when there was an error processing
addMultipleToStore.
Running with ASAN/TSAN pointed at an use-after-free with threads from
the pool accessing the graph declared in processGraph after the function
was exiting and destructing the variables.
It turns out that if there is an error before pool.process() is called,
for example while we are still enqueuing tasks, then pool.process()
isn't called and threads are still left to run.
By creating the pool last we ensure that it is stopped first before
running other destructors even if an exception happens early.
[ lix porting note: nix does not name threads so the patch has been
adapted to not pass thread name ]
Link: https://git.lix.systems/lix-project/lix/issues/618
Link: https://gerrit.lix.systems/c/lix/+/2355
This allows writing lock files with dirty inputs, so long as they have
a NAR hash. (Currently they always have a NAR hash, but with lazy
trees that may not always be the case.)
Generally dirty locks are bad for reproducibility (we can detect if
the dirty input has changed, but we have no way to fetch it except
substitution). Hence we don't allow them by default.
Fixes#11181.
Commands like `nix flake metadata '.?submodules=1'` ignored the query
part of the URL, while `nix build '.?submodules=1#foo'` did work
correctly because of the presence of the fragment part.
The current instructions for building the Nix manual include a command that doesn't work as described. Specifically:
```
nix build .#nix^doc
```
Running this command results in the error:
```
error: derivation '/nix/store/hddqxzfqgx2fhj8q66ss3idym7pk7aj1-nix-2.26.0pre20250107_383ab87.drv' does not have wanted outputs 'doc'
```
However, this command works if you specify the Nix version explicitly, such as:
```
nix build nix/2.24.11#nix^doc
```
Additionally, these commands are run within the Nix root directory.
However, the nix build .#nix^doc command does work when run from the nixpkgs directory and generates the NixOS manual.
I'm not sure if I'm missing something. Is the `nix^doc` supposed to be added somehow to flake outputs?
The incremental build section does not work since as make has been decommissioned in favor of Meson. Should this be simply deleted?
I'd messed up a rebase in my previous iteration, causing `weakly_canonical` to reappear,
but not trigger a test failure.
These two functions behave similarly when the argument is a path that points to a broken
symlink. `weakly_canonical` would not resolve it because the target doesn't exist, and
`makeParentCanonical` would not resolve it, because it never resolves the final path
element.
This new test case now also tests a valid symlink, "differentiating" the two.
The experimental `nix eval` command already supports a `--raw` flag.
This commit implements the same flag for the stable nix-instantiate command.
Until now instructions and scripts that didn't want to rely on experimental
features had to use workarounds such as:
nix-instantiate --eval <something> | tr -d \"
(which also undesirably also removes double quotation marks within the string), or
nix-instantiate --eval <something> | jq -j
(which undesirably depends on another package).
Co-authored-by: Silvan Mosberger <silvan.mosberger@tweag.io>
Upstream change
bab1d75079
moved a few fields from `lowdown_opts` toa new `lowdown_opts_term`
struct. As a result the build started failing as:
nix-cmd> [2/17] Compiling C++ object libnixcmd.so.p/markdown.cc.o
nix-cmd> FAILED: libnixcmd.so.p/markdown.cc.o
nix-cmd> g++ -Ilibnixcmd.so.p -I. -I.. -I/nix/store/b0bnrk5lacxbpgxgnc28r8q3wcazrgxj-nix-util-2.26.0pre-dev/include/nix -I/nix/store/cxnynq9ykyj4xxv6wf6dw7r0aw5x6n9k-libarchive-3.7.7-dev/include -I/nix/store/bfgjwkcb8snkizx578rzdahi75m8zyh4-nlohmann_json-3.11.3/include -I/nix/store/3sx8bq3sip6j2nv1m5xx4gbdp33v7iy6-nix-store-2.26.0pre-dev/include/nix -I/nix/store/sih2dgqzvsbv7p510lkfmas7s7wbsl4j-nix-fetchers-2.26.0pre-dev/include/nix -I/nix/store/68p8s20fsiiakj7nys7grbaixfnhsdzs-nix-expr-2.26.0pre-dev/include/nix -I/nix/store/gw7wknhzhfzzj9zww2kyi5xrzgf1ndki-boehm-gc-8.2.8-dev/include -I/nix/store/3jwb9j4vnsk5saq3wfyyp9il3mhs41l9-nix-flake-2.26.0pre-dev/include/nix -I/nix/store/8nwjvmq7m48v8g646jrxkikv6x47bc3m-nix-main-2.26.0pre-dev/include/nix -I/nix/store/rb0hzsw5wc1a7daizhpj824mbxlvijrq-lowdown-1.4.0-dev/include -I/nix/store/m388ywpk53fsp8r98brfd7nf1f5sskv0-editline-1.17.1-dev/include -fdiagnostics-color=always -D_GLIBCXX_ASSERTIONS=1 -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -std=c++2a -include config-util.hh -include config-store.hh -include config-expr.hh -include config-main.hh -include config-cmd.hh -Wdeprecated-copy -Werror=suggest-override -Werror=switch -Werror=switch-enum -Werror=unused-result -Wignored-qualifiers -Wimplicit-fallthrough -Wno-deprecated-declarations -O3 -fPIC -pthread -std=c++2a -std=c++2a -std=c++2a -std=c++2a -std=c++2a -std=c++2a -MD -MQ libnixcmd.so.p/markdown.cc.o -MF libnixcmd.so.p/markdown.cc.o.d -o libnixcmd.so.p/markdown.cc.o -c ../markdown.cc
nix-cmd> ../markdown.cc: In function 'std::string nix::doRenderMarkdownToTerminal(std::string_view)':
nix-cmd> ../markdown.cc:28:5: error: 'lowdown_opts' has no non-static data member named 'cols'
nix-cmd> 28 | };
nix-cmd> | ^
The change adds version-based conditional to support both pre-1.4 and
1.4 forms of the initialization.
Closes: https://github.com/NixOS/nix/issues/12113
The primitive `readFileType p` has a list of acceptable types, and so does `readDir path`
This edit makes the formatting of the list consistent between themselves, and other parts of the documentation.
Paths are already quoted:
error:
… while fetching the input 'path:/nix/store/rs2s2ca7xs87v82aps54m1p3sqrfz6c8-source'
error: chmod '"/nix/store/rs2s2ca7xs87v82aps54m1p3sqrfz6c8-source"': Read-only file system
src/nix/flake.md describes the format of flake.lock files. Before this
change, it said that the original field was “The original input
specification from `flake.lock`[…]” The original input specification is
in flake.nix, not flake.lock.
When resolving indirect flake references like `nixpkgs` in `flake.nix`
files, Nix will no longer use the system and user flake registries. It
will only use the global flake registry and overrides given on the
command line via `--override-flake`.
Just now there is a dependency on cachix, which means we cannot test
the installer in CI if forks do not have the necessary secrets set up.
We replace this with a simple http server that serves the installer and
can be both used in CI and locally.
Commit cfe66dbec updated `nix upgrade-nix` to use
`ExecutablePath::load().find`, which broke the logic for finding the
profile associated with the nix executable. The error looks something
like:
```
$ sudo -i nix upgrade-nix --debug
found Nix in '"/nix/store/46p1z0w9ad605kky62dr53z4h24k2a5r-nix-2.25.2/bin/nix"'
found profile '/nix/store/46p1z0w9ad605kky62dr53z4h24k2a5r-nix-2.25.2/bin'
error: directory '"/nix/store/46p1z0w9ad605kky62dr53z4h24k2a5r-nix-2.25.2/bin/nix"' does not appear to be part of a Nix profile
```
This seems to happen for two reasons:
1. The original PATH search resulted in a directory, but `find` returns
the path to the executable. Fixed by getting the path's parent.
2. The profile symlink cannot be found because
`ExecutablePath::load().find` canonicalizes the executable path. I
updated find to normalize the path instead, which seems more in line
with how other programs resolve paths. I'm not sure if this affects
other callers though.
I manually tested this on macOS and Linux, and it seemed to fix
upgrading from 2.25.2 to 2.25.3.
- This speeds up macOS builds from 30 minutes to 11 minutes (3x faster).
- Also improve error reporting e.g. printing out what actually failed to build.
- As a result we also no longer need swap.
Subflakes are flakes in the same tree, accessed in flake inputs via
relative paths (e.g. `inputs.foo.url = "path:./subdir"`). Previously
these didn't work very well because they would be separately copied to
the store, which is inefficient and makes references to parent
directories tricky or impossible. Furthermore, they had their own NAR
hash in the lock file, which is superfluous since the parent is
already locked.
Now subflakes are accessed via the accessor of the calling flake. This
avoids the unnecessary copy and makes it possible for subflakes to
depend on flakes in a parent directory (so long as they're in the same
tree).
Lock file nodes for relative flake inputs now have a new `parent` field:
{
"locked": {
"path": "./subdir",
"type": "path"
},
"original": {
"path": "./subdir",
"type": "path"
},
"parent": [
"foo",
"bar"
]
}
which denotes that `./subdir` is to be interpreted relative to the
directory of the `bar` input of the `foo` input of the root flake.
Extracted from the lazy-trees branch.
- To remove the stale label, just leave a new comment.
- _How to find the right people to ping?_ → [`git blame`](https://git-scm.com/docs/git-blame) to the rescue! (or GitHub's history and blame buttons.)
- You can always ask for help on [our Discourse Forum](https://discourse.nixos.org/) or on [Matrix - #nix:nixos.org](https://matrix.to/#/#nix:nixos.org).
- You can always ask for help on [our Discourse Forum](https://discourse.nixos.org/) or on [Matrix - #users:nixos.org](https://matrix.to/#/#users:nixos.org).
- run:nix --experimental-features 'nix-command flakes' flake show --all-systems --json
# Steps to test CI automation in your own fork.
# Cachix:
# 1. Sign-up for https://www.cachix.org/
# 2. Create a cache for $githubuser-nix-install-tests
# 3. Create a cachix auth token and save it in https://github.com/$githubuser/nix/settings/secrets/actions in "Repository secrets" as CACHIX_AUTH_TOKEN
# Dockerhub:
# 1. Sign-up for https://hub.docker.com/
# 2. Store your dockerhub username as DOCKERHUB_USERNAME in "Repository secrets" of your fork repository settings (https://github.com/$githubuser/nix/settings/secrets/actions)
# 3. Create an access token in https://hub.docker.com/settings/security and store it as DOCKERHUB_TOKEN in "Repository secrets" of your fork
check_secrets:
permissions:
contents:none
name:Check Cachix and Docker secrets present for installer tests
# 2. Store your dockerhub username as DOCKERHUB_USERNAME in "Repository secrets" of your fork repository settings (https://github.com/$githubuser/nix/settings/secrets/actions)
# 3. Create an access token in https://hub.docker.com/settings/security and store it as DOCKERHUB_TOKEN in "Repository secrets" of your fork
@@ -31,7 +31,7 @@ Today, a world-wide developer community contributes to Nix and the ecosystem tha
- [Nixpkgs](https://github.com/NixOS/nixpkgs) is [the largest, most up-to-date free software repository in the world](https://repology.org/repositories/graphs)
- [NixOS](https://github.com/NixOS/nixpkgs/tree/master/nixos) is a Linux distribution that can be configured fully declaratively
- [Discourse](https://discourse.nixos.org/)
-[Matrix](https://matrix.to/#/#nix:nixos.org)
- Matrix: [#users:nixos.org](https://matrix.to/#/#users:nixos.org) for user support and [#nix-dev:nixos.org](https://matrix.to/#/#nix-dev:nixos.org) for development
synopsis: "Removed support for daemons and clients older than Nix 2.0"
prs: [13951]
---
We have dropped support in the daemon worker protocol for daemons and clients that don't speak at least version 18 of the protocol. This first Nix release that supports this version is Nix 2.0, released in February 2018.
The advantage is that this avoids a time window where *path* is not a garbage collector root, and so could be deleted by a concurrent `nix store gc` process.
synopsis: "Temporary build directories no longer include derivation names"
prs: [13839]
---
Temporary build directories created during derivation builds no longer include the derivation name in their path to avoid build failures when the derivation name is too long. This change ensures predictable prefix lengths for build directories under `/nix/var/nix/builds`.
@@ -53,6 +53,11 @@ This command has the following operations:
Download the Nix expressions of subscribed channels and create a new generation.
Update all channels if none is specified, and only those included in *names* otherwise.
> **Note**
>
> Downloaded channel contents are cached.
> Use `--tarball-ttl` or the [`tarball-ttl` configuration option](@docroot@/command-ref/conf-file.md#conf-tarball-ttl) to change the validity period of cached downloads.
-`--list-generations`
Prints a list of all the current existing generations for the
@@ -62,6 +62,15 @@ These options are for deleting old [profiles] prior to deleting unreachable [sto
This is the equivalent of invoking [`nix-env --delete-generations <period>`](@docroot@/command-ref/nix-env/delete-generations.md#generations-time) on each found profile.
See the documentation of that command for additional information about the *period* argument.
<!-- duplication from https://github.com/NixOS/nix/blob/442a2623e48357ff72c77bb11cf2cf06d94d2f90/doc/manual/source/command-ref/nix-store/gc.md?plain=1#L39-L44 -->
Keep deleting paths until at least *bytes* bytes have been deleted,
then stop. The argument *bytes* can be followed by the
@@ -21,11 +22,11 @@ It is based on the current generation of the active [profile](@docroot@/command-
The arguments *args* map to store paths in a number of possible ways:
- By default, *args* is a set of [derivation] names denoting derivations in the [default Nix expression].
- By default, *args* is a set of names denoting derivations in the [default Nix expression].
These are [realised], and the resulting output paths are installed.
Currently installed derivations with a name equal to the name of a derivation being added are removed unless the option `--preserve-installed` is specified.
@@ -61,11 +62,15 @@ The arguments *args* map to store paths in a number of possible ways:
The derivations returned by those function calls are installed.
This allows derivations to be specified in an unambiguous way, which is necessary if there are multiple derivations with the same name.
- If *args* are [store derivations](@docroot@/glossary.md#gloss-store-derivation), then these are [realised], and the resulting output paths are installed.
- If `--priority` *priority* is given, the priority of the derivations being installed is set to *priority*.
This can be used to override the priority of the derivations being installed.
This is useful if *args* are [store paths], which don't have any priority information.
- If *args* are [store paths] that are not store derivations, then these are [realised] and installed.
- If *args* are [store paths] that point to [store derivations][store derivation], then those store derivations are [realised], and the resulting output paths are installed.
- By default all [outputs](@docroot@/language/derivations.md#attr-outputs) are installed for each [derivation].
- If *args* are [store paths] that do not point to store derivations, then these are [realised] and installed.
- By default all [outputs](@docroot@/language/derivations.md#attr-outputs) are installed for each [store derivation].
This can be overridden by adding a `meta.outputsToInstall` attribute on the derivation listing a subset of the output names.
Example:
@@ -117,6 +122,8 @@ The arguments *args* map to store paths in a number of possible ways:
This guide explains how to build and run performance benchmarks in the Nix codebase.
## Overview
Nix uses the [Google Benchmark](https://github.com/google/benchmark) framework for performance testing. Benchmarks help measure and track the performance of critical operations like derivation parsing.
## Building Benchmarks
Benchmarks are disabled by default and must be explicitly enabled during the build configuration. For accurate results, use a debug-optimized release build.
### Development Environment Setup
First, enter the development shell which includes the necessary dependencies:
```bash
nix develop .#native-ccacheStdenv
```
### Configure Build with Benchmarks
From the project root, configure the build with benchmarks enabled and optimization:
@@ -79,7 +79,7 @@ This shell also adds `./outputs/bin/nix` to your `$PATH` so you can run `nix` im
To get a shell with one of the other [supported compilation environments](#compilation-environments):
```console
$ nix develop .#native-clangStdenvPackages
$ nix develop .#native-clangStdenv
```
> **Note**
@@ -167,11 +167,13 @@ It is useful to perform multiple cross and native builds on the same source tree
for example to ensure that better support for one platform doesn't break the build for another.
Meson thankfully makes this very easy by confining all build products to the build directory --- one simple shares the source directory between multiple build directories, each of which contains the build for Nix to a different platform.
Nixpkgs's `configurePhase` always chooses `build` in the current directory as the name and location of the build.
This makes having multiple build directories slightly more inconvenient.
The good news is that Meson/Ninja seem to cope well with relocating the build directory after it is created.
Here's how to do that:
Here's how to do that
1. Instruct Nixpkgs's infra where we want Meson to put its build directory
```bash
mesonBuildDir=build-my-variant-name
```
1. Configure as usual
@@ -179,24 +181,12 @@ Here's how to do that
configurePhase
```
2. Rename the build directory
```bash
cd .. # since `configurePhase` cd'd inside
mv build build-linux # or whatever name we want
cd build-linux
```
3. Build as usual
```bash
buildPhase
```
> **N.B.**
> [`nixpkgs#335818`](https://github.com/NixOS/nixpkgs/issues/335818) tracks giving `mesonConfigurePhase` proper support for custom build directories.
> When it is fixed, we can simplify these instructions and then remove this notice.
## System type
Nix uses a string with the following format to identify the *system type* or *platform* it runs on:
@@ -205,28 +195,38 @@ Nix uses a string with the following format to identify the *system type* or *pl
<cpu>-<os>[-<abi>]
```
It is set when Nix is compiled for the given system, and based on the output of [`config.guess`](https://github.com/nixos/nix/blob/master/config/config.guess) ([upstream](https://git.savannah.gnu.org/cgit/config.git/tree/config.guess)):
It is set when Nix is compiled for the given system, and based on the output of Meson's [`host_machine` information](https://mesonbuild.com/Reference-manual_builtin_host_machine.html)>
```
<cpu>-<vendor>-<os>[<version>][-<abi>]
```
When Nix is built such that `./configure` is passed any of the `--host`, `--build`, `--target` options, the value is based on the output of [`config.sub`](https://github.com/nixos/nix/blob/master/config/config.sub) ([upstream](https://git.savannah.gnu.org/cgit/config.git/tree/config.sub)):
When cross-compiling Nix with Meson for local development, you need to specify a [cross-file](https://mesonbuild.com/Cross-compilation.html) using the `--cross-file` option. Cross-files define the target architecture and toolchain. When cross-compiling Nix with Nix, Nixpkgs takes care of this for you.
In the nix flake we also have some cross-compilation targets available:
For historic reasons and backward-compatibility, some CPU and OS identifiers are translated from the GNU Autotools naming convention in [`configure.ac`](https://github.com/nixos/nix/blob/master/configure.ac) as follows:
For historic reasons and backward-compatibility, some CPU and OS identifiers are translated as follows:
@@ -240,18 +240,18 @@ Nix can be compiled using multiple environments:
To build with one of those environments, you can use
```console
$ nix build .#nix-ccacheStdenv
$ nix build .#nix-cli-ccacheStdenv
```
for flake-enabled Nix, or
```console
$ nix-build --attr nix-ccacheStdenv
$ nix-build --attr nix-cli-ccacheStdenv
```
for classic Nix.
You can use any of the other supported environments in place of `nix-ccacheStdenv`.
You can use any of the other supported environments in place of `nix-cli-ccacheStdenv`.
## Editor integration
@@ -261,7 +261,8 @@ See [supported compilation environments](#compilation-environments) and instruct
To use the LSP with your editor, you will want a `compile_commands.json` file telling `clangd` how we are compiling the code.
Meson's configure always produces this inside the build directory.
Configure your editor to use the `clangd` from the `.#native-clangStdenvPackages` shell. You can do that either by running it inside the development shell, or by using [nix-direnv](https://github.com/nix-community/nix-direnv) and [the appropriate editor plugin](https://github.com/direnv/direnv/wiki#editor-integration).
Configure your editor to use the `clangd` from the `.#native-clangStdenv` shell.
You can do that either by running it inside the development shell, or by using [nix-direnv](https://github.com/nix-community/nix-direnv) and [the appropriate editor plugin](https://github.com/direnv/direnv/wiki#editor-integration).
> **Note**
>
@@ -277,6 +278,8 @@ You may run the formatters as a one-off using:
./maintainers/format.sh
```
### Pre-commit hooks
If you'd like to run the formatters before every commit, install the hooks:
```
@@ -291,3 +294,30 @@ If it fails, run `git add --patch` to approve the suggestions _and commit again_
To refresh pre-commit hook's config file, do the following:
1. Exit the development shell and start it again by running `nix develop`.
2. If you also use the pre-commit hook, also run `pre-commit-hooks-install` again.
### VSCode
Insert the following json into your `.vscode/settings.json` file to configure `nixfmt`.
This will be picked up by the _Format Document_ command, `"editor.formatOnSave"`, etc.
@@ -30,7 +30,7 @@ The unit tests are defined using the [googletest] and [rapidcheck] frameworks.
> src
> ├── libexpr
> │ ├── meson.build
> │ ├── value/context.hh
> │ ├── include/nix/expr/value/context.hh
> │ ├── value/context.cc
> │ …
> │
@@ -46,8 +46,12 @@ The unit tests are defined using the [googletest] and [rapidcheck] frameworks.
> │ │
> │ ├── libexpr-test-support
> │ │ ├── meson.build
> │ │ ├── include/nix/expr
> │ │ │ ├── meson.build
> │ │ │ └── tests
> │ │ │ ├── value/context.hh
> │ │ │ …
> │ │ └── tests
> │ │ ├── value/context.hh
> │ │ ├── value/context.cc
> │ │ …
> │ │
@@ -59,7 +63,7 @@ The unit tests are defined using the [googletest] and [rapidcheck] frameworks.
> ```
The tests for each Nix library (`libnixexpr`, `libnixstore`, etc..) live inside a directory `src/${library_name_without-nix}-test`.
Given an interface (header) and implementation pair in the original library, say, `src/libexpr/value/context.{hh,cc}`, we write tests for it in `src/libexpr-tests/value/context.cc`, and (possibly) declare/define additional interfaces for testing purposes in `src/libexpr-test-support/tests/value/context.{hh,cc}`.
Given an interface (header) and implementation pair in the original library, say, `src/libexpr/include/nix/expr/value/context.hh` and `src/libexpr/value/context.cc`, we write tests for it in `src/libexpr-tests/value/context.cc`, and (possibly) declare/define additional interfaces for testing purposes in `src/libexpr-test-support/include/nix/expr/tests/value/context.hh` and `src/libexpr-test-support/tests/value/context.cc`.
Data for unit tests is stored in a `data` subdir of the directory for each unit test executable.
For example, `libnixstore` code is in `src/libstore`, and its test data is in `src/libstore-tests/data`.
@@ -67,7 +71,7 @@ The path to the `src/${library_name_without-nix}-test/data` directory is passed
Note that each executable only gets the data for its tests.
The unit test libraries are in `src/${library_name_without-nix}-test-support`.
All headers are in a `tests` subdirectory so they are included with `#include "tests/"`.
All headers are in a `tests` subdirectory so they are included with `#include "nix/tests/"`.
The use of all these separate directories for the unit tests might seem inconvenient, as for example the tests are not "right next to" the part of the code they are testing.
But organizing the tests this way has one big benefit:
@@ -87,7 +91,11 @@ A environment variables that Google Test accepts are also worth knowing:
See [functional characterisation testing](#characterisation-testing-functional) for a broader discussion of characterisation testing.
@@ -144,7 +168,7 @@ $ checkPhase
Sometimes it is useful to group related tests so they can be easily run together without running the entire test suite.
Each test group is in a subdirectory of `tests`.
For example, `tests/functional/ca/meson.build` defines a `ca` test group for content-addressed derivation outputs.
For example, `tests/functional/ca/meson.build` defines a `ca` test group for content-addressing derivation outputs.
That test group can be run like this:
@@ -213,10 +237,10 @@ edit it like so:
bar
```
Then, running the test with `./mk/debug-test.sh` will drop you into GDB once the script reaches that point:
Then, running the test with [`--interactive`](https://mesonbuild.com/Unit-tests.html#other-test-options) will prevent Meson from hijacking the terminal so you can drop you into GDB once the script reaches that point:
@@ -297,7 +321,7 @@ Creating a Cachix cache for your installer tests and adding its authorisation to
- `armv7l-linux`
- `x86_64-darwin`
- The `installer_test` job (which runs on `ubuntu-latest` and `macos-latest`) will try to install Nix with the cached installer and run a trivial Nix command.
- The `installer_test` job (which runs on `ubuntu-24.04` and `macos-14`) will try to install Nix with the cached installer and run a trivial Nix command.
Software Heritage's writing on [*Intrinsic and Extrinsic identifiers*](https://www.softwareheritage.org/2020/07/09/intrinsic-vs-extrinsic-identifiers) is also a good introduction to the value of content-addressing over other referencing schemes.
Besides content addressing, the Nix store also uses [input addressing](#gloss-input-addressed-store-object).
The industry term for storage and retrieval systems using [content addressing](#gloss-content-address). A Nix store also has [input addressing](#gloss-input-addressed-store-object), and metadata.
- [derivation]{#gloss-derivation}
A description of a build task. The result of a derivation is a
store object. Derivations declared in Nix expressions are specified
using the [`derivation` primitive](./language/derivations.md). These are
translated into low-level *store derivations* (implicitly by
`nix-build`, or explicitly by `nix-instantiate`).
A derivation can be thought of as a [pure function](https://en.wikipedia.org/wiki/Pure_function) that produces new [store objects][store object] from existing store objects.
Derivations are implemented as [operating system processes that run in a sandbox](@docroot@/store/building.md#builder-execution).
This sandbox by default only allows reading from store objects specified as inputs, and only allows writing to designated [outputs][output] to be [captured as store objects](@docroot@/store/building.md#processing-outputs).
A derivation is typically specified as a [derivation expression] in the [Nix language], and [instantiated][instantiate] to a [store derivation].
There are multiple ways of obtaining store objects from store derivatons, collectively called [realisation][realise].
[derivation]: #gloss-derivation
- [store derivation]{#gloss-store-derivation}
A [derivation] represented as a `.drv` file in the [store].
It has a [store path], like any [store object].
It is the [instantiated][instantiate] form of a derivation.
A [directed acyclic graph](https://en.wikipedia.org/wiki/Directed_acyclic_graph) (DAG) is graph whose edges are given a direction ("a to b" is not the same edge as "b to a"), and for which no possible path (created by joining together edges) forms a cycle.
DAGs are very important to Nix.
In particular, the non-self-[references][reference] of [store object][store object] form a cycle.
- [derivation path]{#gloss-derivation-path}
A [store path] which uniquely identifies a [store derivation].
See [Referencing Store Derivations](@docroot@/store/derivation/index.md#derivation-path) for details.
A [derivation] where a cryptographic hash of the [output] is determined in advance using the [`outputHash`](./language/advanced-attributes.md#adv-attr-outputHash) attribute, and where the [`builder`](@docroot@/language/derivations.md#attr-builder) executable has access to the network.
A [store derivation] where a cryptographic hash of the [output] is determined in advance using the [`outputHash`](./language/advanced-attributes.md#adv-attr-outputHash) attribute, and where the [`builder`](@docroot@/language/derivations.md#attr-builder) executable has access to the network.
- [store]{#gloss-store}
@@ -84,6 +116,12 @@
[store]: #gloss-store
- [Nix instance]{#gloss-nix-instance}
<!-- ambiguous -->
1. An installation of Nix, which includes the presence of a [store], and the Nix package manager which operates on that store.
A local Nix installation and a [remote builder](@docroot@/advanced-topics/distributed-builds.md) are two examples of Nix instances.
2. A running Nix process, such as the `nix` command.
- [binary cache]{#gloss-binary-cache}
A *binary cache* is a Nix store which uses a different format: its
@@ -130,15 +168,17 @@
- [input-addressed store object]{#gloss-input-addressed-store-object}
See [input-addressing derivation outputs](store/derivation/outputs/input-address.md) for details.
- [content-addressed store object]{#gloss-content-addressed-store-object}
A [store object] which is [content-addressed](#gloss-content-address),
i.e. whose [store path] is determined by its contents.
This includes derivations, the outputs of [content-addressed derivations](#gloss-content-addressed-derivation), and the outputs of [fixed-output derivations](#gloss-fixed-output-derivation).
This includes derivations, the outputs of [content-addressing derivations](#gloss-content-addressing-derivation), and the outputs of [fixed-output derivations](#gloss-fixed-output-derivation).
See [Content-Addressing Store Objects](@docroot@/store/store-object/content-address.md) for details.
@@ -188,35 +228,37 @@
>
> The contents of a `.nix` file form a Nix expression.
Nix expressions specify [derivations][derivation], which are [instantiated][instantiate] into the Nix store as [store derivations][store derivation].
Nix expressions specify [derivation expressions][derivation expression], which are [instantiated][instantiate] into the Nix store as [store derivations][store derivation].
These derivations can then be [realised][realise] to produce [outputs][output].
> **Example**
>
> Building and deploying software using Nix entails writing Nix expressions as a high-level description of packages and compositions thereof.
> Building and deploying software using Nix entails writing Nix expressions to describe [packages][package] and compositions thereof.
- [reference]{#gloss-reference}
A [store object] `O` is said to have a *reference* to a store object `P` if a [store path] to `P` appears in the contents of `O`.
An edge from one [store object] to another.
Store objects can refer to both other storeobjects and themselves.
References from a store object to itself are called *self-references*.
References other than a self-reference must not form a cycle.
See [References](@docroot@/store/store-object.md#references) for details.
[reference]: #gloss-reference
See [References](@docroot@/store/store-object.md#references) for details.
- [reachable]{#gloss-reachable}
A store path `Q` is reachable from another store path `P` if `Q`
is in the *closure* of the *references* relation.
See [References](@docroot@/store/store-object.md#references) for details.
- [closure]{#gloss-closure}
The closure of a store path is the set of store paths that are
directly or indirectly “reachable” from that store path; that is,
it’s the closure of the path under the *references* relation. For
a package, the closure of its derivation is equivalent to the
build-time dependencies, while the closure of its output path is
build-time dependencies, while the closure of its [output path] is
equivalent to its runtime dependencies. For correct deployment it
is necessary to deploy whole closures, since otherwise at runtime
files could be missing. The command `nix-store --query --requisites ` prints out
@@ -226,18 +268,31 @@
to a store object at path `Q`, then `Q` is in the closure of `P`. Further, if `Q`
references `R` then `R` is also in the closure of `P`.
See [References](@docroot@/store/store-object.md#references) for details.
[closure]: #gloss-closure
- [requisite]{#gloss-requisite}
A store object [reachable] by a path (chain of references) from a given [store object].
The [closure] is the set of requisites.
See [References](@docroot@/store/store-object.md#references) for details.
- [referrer]{#gloss-reference}
A reversed edge from one [store object] to another.
- [output]{#gloss-output}
A [store object] produced by a [derivation].
A [store object] produced by a [store derivation].
See [the `outputs` argument to the `derivation` function](@docroot@/language/derivations.md#attr-outputs) for details.
[output]: #gloss-output
- [output path]{#gloss-output-path}
The [store path] to the [output] of a [derivation].
The [store path] to the [output] of a [store derivation].
[output path]: #gloss-output-path
@@ -246,14 +301,11 @@
- [deriving path]{#gloss-deriving-path}
Deriving paths are a way to refer to [store objects][store object] that ar not yet [realised][realise].
This is necessary because, in general and particularly for [content-addressed derivations][content-addressed derivation], the [output path] of an [output] is not known in advance.
There are two forms:
Deriving paths are a way to refer to [store objects][store object] that might not yet be [realised][realise].
- *constant*: just a [store path]
It can be made [valid][validity] by copying it into the store: from the evaluator, command line interface or another store.
See [Deriving Path](./store/derivation/index.md#deriving-path) for details.
- *output*: a pair of a [store path] to a [derivation] and an [output] name.
Not to be confused with [derivation path].
- [deriver]{#gloss-deriver}
@@ -301,7 +353,7 @@
See [Nix Archive](store/file-system-object/content-address.html#serial-nix-archive) for details.
- [`∅`]{#gloss-emtpy-set}
- [`∅`]{#gloss-empty-set}
The empty set symbol. In the context of profile history, this denotes a package is not present in a particular version of the profile.
@@ -311,18 +363,17 @@
- [package]{#package}
1. A software package; a collection of files and other data.
A software package; files that belong together for a particular purpose, and metadata.
2. A [package attribute set].
Nix represents files as [file system objects][file system object], and how they belong together is encoded as [references][reference] between [store objects][store object] that contain these file system objects.
- [package attribute set]{#package-attribute-set}
The [Nix language] allows denoting packages in terms of [attribute sets](@docroot@/language/types.md#attribute-set) containing:
- attributes that refer to the files of a package, typically in the form of [derivation outputs](#output),
- attributes with metadata, such as information about how the package is supposed to be used.
An [attribute set](@docroot@/language/types.md#attribute-set) containing the attribute `type = "derivation";` (derivation for historical reasons), as well as other attributes, such as
- attributes that refer to the files of a [package], typically in the form of [derivation outputs](#output),
- attributes that declare something about how the package is supposed to be installed or used,
- other metadata or arbitrary attributes.
The exact shape of these attribute sets is up to convention.
If the special attribute `__structuredAttrs` is set to `true`, the other derivation
attributes are serialised into a file in JSON format.
This obviates the need for [`passAsFile`](#adv-attr-passAsFile) since JSON files have no size restrictions, unlike process environments.
It also makes it possible to tweak derivation settings in a structured way;
see [`outputChecks`](#adv-attr-outputChecks) for example.
See the [corresponding section in the derivation page](@docroot@/store/derivation/index.md#structured-attrs) for further details.
> **Warning**
>
> If set to `true`, other advanced attributes such as [`allowedReferences`](#adv-attr-allowedReferences), [`allowedRequisites`](#adv-attr-allowedRequisites),
[`disallowedReferences`](#adv-attr-disallowedReferences) and [`disallowedRequisites`](#adv-attr-disallowedRequisites), maxSize, and maxClosureSize.
will have no effect.
## Output checks
See the [corresponding section in the derivation output page](@docroot@/store/derivation/outputs/index.md).
These attributes declare that the derivation is a so-called *fixed-output derivation* (FOD), which means that a cryptographic hash of the output is already known in advance.
As opposed to regular derivations, the [`builder`] executable of a fixed-output derivation has access to the network.
Nix computes a cryptographic hash of its output and compares that to the hash declared with these attributes.
If there is a mismatch, the derivation fails.
The rationale for fixed-output derivations is derivations such as
those produced by the `fetchurl` function. This function downloads a
file from a given URL. To ensure that the downloaded file has not
been modified, the caller must also specify a cryptographic hash of
If a `fetchurl` derivation was treated like a normal derivation, the
output paths of the derivation and *all derivations depending on it*
would change. For instance, if we were to change the URL of the
Glibc source distribution in Nixpkgs (a package on which almost all
other packages depend) massive rebuilds would be needed. This is
unfortunate for a change which we know cannot have a real effect as
it propagates upwards through the dependency graph.
For fixed-output derivations, on the other hand, the name of the
output path only depends on the `outputHash*` and `name` attributes,
while all other attributes are ignored for the purpose of computing
the output path. (The `name` attribute is included because it is
part of the path.)
As an example, here is the (simplified) Nix expression for
`fetchurl`:
```nix
{ stdenv, curl }: # The curl program is used for downloading.
{ url, sha256 }:
stdenv.mkDerivation {
name = baseNameOf (toString url);
builder = ./builder.sh;
buildInputs = [ curl ];
# This is a fixed-output derivation; the output must be a regular
# file with SHA256 hash sha256.
outputHashMode = "flat";
outputHashAlgo = "sha256";
outputHash = sha256;
inherit url;
}
```
The `outputHash` attribute must be a string containing the hash in either hexadecimal or "nix32" encoding, or following the format for integrity metadata as defined by [SRI](https://www.w3.org/TR/SRI/).
The "nix32" encoding is an adaptation of base-32 encoding.
The [`convertHash`](@docroot@/language/builtins.md#builtins-convertHash) function shows how to convert between different encodings, and the [`nix-hash` command](../command-ref/nix-hash.md) has information about obtaining the hash for some contents, as well as converting to and from encodings.
The `outputHashAlgo` attribute specifies the hash algorithm used to compute the hash.
It can currently be `"sha1"`, `"sha256"`, `"sha512"`, or `null`.
`outputHashAlgo` can only be `null` when `outputHash` follows the SRI format.
The `outputHashMode` attribute determines how the hash is computed.
> For example, in [nix.conf](../command-ref/conf-file.md) you could add:
>
> ```
> extra-experimental-features = ca-derivations
> ```
If this attribute is set to `true`, then the derivation
outputs will be stored in a content-addressed location rather than the
traditional input-addressed one.
Setting this attribute also requires setting
[`outputHashMode`](#adv-attr-outputHashMode)
and
[`outputHashAlgo`](#adv-attr-outputHashAlgo)
like for *fixed-output derivations* (see above).
It also implicitly requires that the machine to build the derivation must have the `ca-derivations` [system feature](@docroot@/command-ref/conf-file.md#conf-system-features).
- [`passAsFile`]{#adv-attr-passAsFile}\
A list of names of attributes that should be passed via files rather
than environment variables. For example, if you have
```nix
passAsFile = ["big"];
big = "a very long string";
```
then when the builder runs, the environment variable `bigPath`
will contain the absolute path to a temporary file containing `a
very long string`. That is, for any attribute *x* listed in
`passAsFile`, Nix will pass an environment variable `xPath`
holding the path of the file containing the value of attribute
*x*. This is useful when you need to pass large strings to a
builder, since most operating systems impose a limit on the size
of the environment (typically, a few hundred kilobyte).
If this attribute is set to `true` and [distributed building is enabled](@docroot@/command-ref/conf-file.md#conf-builders), then, if possible, the derivation will be built locally instead of being forwarded to a remote machine.
This is useful for derivations that are cheapest to build locally.
If this attribute is set to `false`, then Nix will always build this derivation (locally or remotely); it will not try to substitute its outputs.
This is useful for derivations that are cheaper to build than to substitute.
This attribute can be ignored by setting [`always-allow-substitutes`](@docroot@/command-ref/conf-file.md#conf-always-allow-substitutes) to `true`.
> **Note**
>
> If set to `false`, the [`builder`] should be able to run on the system type specified in the [`system` attribute](./derivations.md#attr-system), since the derivation cannot be substituted.
If the special attribute `__structuredAttrs` is set to `true`, the other derivation
attributes are serialised into a file in JSON format. The environment variable
`NIX_ATTRS_JSON_FILE` points to the exact location of that file both in a build
and a [`nix-shell`](../command-ref/nix-shell.md). This obviates the need for
[`passAsFile`](#adv-attr-passAsFile) since JSON files have no size restrictions,
unlike process environments.
It also makes it possible to tweak derivation settings in a structured way; see
[`outputChecks`](#adv-attr-outputChecks) for example.
As a convenience to Bash builders,
Nix writes a script that initialises shell variables
corresponding to all attributes that are representable in Bash. The
environment variable `NIX_ATTRS_SH_FILE` points to the exact
location of the script, both in a build and a
[`nix-shell`](../command-ref/nix-shell.md). This includes non-nested
(associative) arrays. For example, the attribute `hardening.format = true`
ends up as the Bash associative array element `${hardening[format]}`.
> **Warning**
>
> If set to `true`, other advanced attributes such as [`allowedReferences`](#adv-attr-allowedReferences), [`allowedReferences`](#adv-attr-allowedReferences), [`allowedRequisites`](#adv-attr-allowedRequisites),
[`disallowedReferences`](#adv-attr-disallowedReferences) and [`disallowedRequisites`](#adv-attr-disallowedRequisites), maxSize, and maxClosureSize.
will have no effect.
- [`outputChecks`]{#adv-attr-outputChecks}\
When using [structured attributes](#adv-attr-structuredAttrs), the `outputChecks`
attribute allows defining checks per-output.
@@ -341,8 +157,9 @@ Derivations can declare some infrequently used optional attributes.
If this attribute is set to `true` and [distributed building is enabled](@docroot@/command-ref/conf-file.md#conf-builders), then, if possible, the derivation will be built locally instead of being forwarded to a remote machine.
This is useful for derivations that are cheapest to build locally.
If this attribute is set to `false`, then Nix will always build this derivation (locally or remotely); it will not try to substitute its outputs.
This is useful for derivations that are cheaper to build than to substitute.
This attribute can be ignored by setting [`always-allow-substitutes`](@docroot@/command-ref/conf-file.md#conf-always-allow-substitutes) to `true`.
> **Note**
>
> If set to `false`, the [`builder`] should be able to run on the system type specified in the [`system` attribute](./derivations.md#attr-system), since the derivation cannot be substituted.
If a derivation has the `requiredSystemFeatures` attribute, then Nix will only build it on a machine that has the corresponding features set in its [`system-features` configuration](@docroot@/command-ref/conf-file.md#conf-system-features).
For example, setting
@@ -370,6 +204,171 @@ Derivations can declare some infrequently used optional attributes.
ensures that the derivation can only be built on a machine with the `kvm` feature.
As discussed in [Derivation Outputs and Types of Derivations](@docroot@/store/derivation/outputs/index.md), there are multiples kinds of derivations / kinds of derivation outputs.
The choice of the following attributes determines which kind of derivation we are making.
- [`__contentAddressed`]
- [`outputHash`]
- [`outputHashAlgo`]
- [`outputHashMode`]
The three types of derivations are chosen based on the following combinations of these attributes.
> This method is part of the [`git-hashing`][xp-feature-git-hashing] experimental feature.
See [content-addressing store objects](@docroot@/store/store-object/content-address.md) for more information about the process this flag controls.
- [`outputHashAlgo`]{#adv-attr-outputHashAlgo}
This specifies the hash algorithm used to digest the [file system object] data of a content-addressing derivation output.
This works in conjunction with [`outputHashMode`](#adv-attr-outputHashAlgo).
Specifying one without the other is an error (unless `outputHash` is also specified and includes its own hash algorithm as described below).
The `outputHashAlgo` attribute specifies the hash algorithm used to compute the hash.
It can currently be `"blake3"`, `"sha1"`, `"sha256"`, `"sha512"`, or `null`.
`outputHashAlgo` can only be `null` when `outputHash` follows the SRI format, because in that case the choice of hash algorithm is determined by `outputHash`.
This will specify the output hash of the single output of a [fixed-output derivation].
The `outputHash` attribute must be a string containing the hash in either hexadecimal or "nix32" encoding, or following the format for integrity metadata as defined by [SRI](https://www.w3.org/TR/SRI/).
The "nix32" encoding is an adaptation of base-32 encoding.
> **Note**
>
> The [`convertHash`](@docroot@/language/builtins.md#builtins-convertHash) function shows how to convert between different encodings.
> The [`nix-hash` command](../command-ref/nix-hash.md) has information about obtaining the hash for some contents, as well as converting to and from encodings.
Evaluation is the process of turning a Nix expression into a [Nix value](types.md).
This happens by a number of rules, such as:
- Constructing values from literals.
For example the number literal `1` is turned into the number value `1`.
- Applying operators
For example the addition operator `+` is applied to two number values to produce a new number value.
- Applying built-in functions
For example the expression `builtins.isInt 1` is evaluated to `true`.
- Applying user-defined functions
For example the expression `(x: x + 1) 10` can[*](#laziness) be thought of rewriting `x` in the function body to the argument, `10 + 1`, which is then evaluated to `11`.
These rules are applied as needed, driven by the specific use of the expression. For example, this can occur in the Nix command line interface or interactively with the [repl (read-eval-print loop)](@docroot@/command-ref/new-cli/nix3-repl.md), which is a useful tool when learning about evaluation.
# Details
## Values {#values}
Nix values can be thought of as a subset of Nix expressions.
For example, the expression `1 + 2` is not a value, because it can be reduced to `3`. The expression `3` is a value, because it cannot be reduced any further.
Evaluation normally happens by applying rules to the "head" of the expression, which is the outermost part of the expression. The head of an expression like `[ 1 2 ]` is the list literal (`[ a1 a2 ]`), for `1 + 2` it is the addition operator (`+`), and for `f 1` it is the function application "operator" (` `).
After applying all possible rules to the head until no rules can be applied, the expression is in "weak head normal form" (WHNF). This means that the outermost constructor of the expression is evaluated, but the inner values may or may not be. "Weak" only signifies that the expression may be a function. This is an historical or academic artifact, and Nix has no use for the non-weak "head normal form".
## Laziness and thunks {#laziness}
The Nix language implements _call by need_ (as opposed to _call by value_ or _call by reference_). <!-- No wikipedia link, which would be a huge distraction. --> Call by need is commonly known as laziness in functional programming, as it is a specific implementation of the concept where evaluation is deferred until the result is required, aiming to only evaluate the parts of an expression that are needed to produce the final result.
Furthermore, the result of evaluation is preserved, in values, in `let` bindings, in function _parameters_, which behave a lot like `let` bindings, but with the notable exception of function _calls_. Results of function calls rely on being put into `let` bindings, etc to be reused. <!-- which would be prohibitively expensive and too strict, or we wouldn't have a cache key for the argument -->
When discussing the process of evaluation in lower level terms, we may define values not as a subset of expressions, but separately, where each "value" is either a data constructor, a function or a _thunk_. A thunk is a delayed computation, represented by an expression reference and a "closure" – the values for the lexical scope around the delayed expression.
As a user of the language, you generally don't have to think about thunks, as they are not part of the language semantics, but you may encounter them in the repl, in the [C API] or in discussions.
## Strictness
Instead of thinking about thunks, it is often more productive to think in terms of _strictness_.
This term is used in functional programming to refer to the opposite of laziness, i.e. not just for something like error propagation. It refers to the need to evaluate certain expressions before evaluation can produce any result.
Statements about strictness usually implicitly refer to weak head normal form.
For example, we can say that the following function is strict in its argument:
```nix
x:isAttrsx||isFunctionx
```
The above function must be strict in its argument `x` because determining its type requires evaluating `x` to at least some degree.
The following function is not strict in its argument:
```nix
x:{isOk=isAttrsx||isFunctionx;}
```
It is not strict, because it can return the attribute set before evaluating `x`.
The attribute value for `isOk`_is_ strict in `x`.
A function with a _set pattern_ is always strict in its argument, as a consequence of checking the argument's type and/or attribute names:
```nix
letf={...}:"ok";
inf(throw"kablam")
=>error:kablam
```
However, a set pattern does not add any strictness beyond WHNF of the attribute set argument.
The Nix language is designed for conveniently creating and composing *derivations*– precise descriptions of how contents of existing files are used to derive new files.
The Nix language is designed for conveniently creating and composing [derivations](@docroot@/glossary.md#gloss-derivation)– precise descriptions of how contents of existing files are used to derive new files.
> **Tip**
>
@@ -11,7 +11,14 @@ The language is:
- *domain-specific*
It comes with [built-in functions](@docroot@/language/builtins.md) to integrate with the Nix store, which manages files and performs the derivations declared in the Nix language.
The Nix language is purpose-built for working with text files.
Its most characteristic features are:
- [File system path primitives](@docroot@/language/types.md#type-path), for accessing source files
- [Indented strings](@docroot@/language/string-literals.md) and [string interpolation](@docroot@/language/string-interpolation.md), for creating file contents
- [Strings with contexts](@docroot@/language/string-context.md), for transparently linking files
It comes with [built-in functions](@docroot@/language/builtins.md) to integrate with the [Nix store](@docroot@/store/index.md), which manages files and enables [realising](@docroot@/glossary.md#gloss-realise) derivations declared in the Nix language.
The idea is that a user can combine together values to create a build instructions for derivations without manually keeping track of where they come from.
Then the Nix language implicitly does that bookkeeping to efficiently obtain the closure of derivation inputs.
The idea is that a user can reference other files when creating text files through Nix expressions, without manually keeping track of the exact paths.
Nix will ensure that the all referenced files are accessible – that all [store paths](@docroot@/glossary.md#gloss-store-path) are [valid](@docroot@/glossary.md#gloss-validity).
> **Note**
>
@@ -115,7 +115,7 @@ It creates an [attribute set] representing the string context, which can be insp
## Clearing string contexts
[`buitins.unsafeDiscardStringContext`](./builtins.md#builtins-unsafeDiscardStringContext) will make a copy of a string, but with an empty string context.
[`builtins.unsafeDiscardStringContext`](./builtins.md#builtins-unsafeDiscardStringContext) will make a copy of a string, but with an empty string context.
The returned string can be used in more ways, e.g. by operators that require the string context to be empty.
The requirement to explicitly discard the string context in such use cases helps ensure that string context elements are not lost by mistake.
The "unsafe" marker is only there to remind that Nix normally guarantees that dependencies are tracked, whereas the returned string has lost them.
For historical reasons, [derivations](@docroot@/glossary.md#gloss-store-derivation) are stored on-disk in [ATerm](https://homepages.cwi.nl/~daybuild/daily-books/technology/aterm-guide/aterm-guide.html) format.
For historical reasons, [store derivations][storederivation] are stored on-disk in [ATerm](https://homepages.cwi.nl/~daybuild/daily-books/technology/aterm-guide/aterm-guide.html) format.
## The ATerm format used
Derivations are serialised in one of the following formats:
@@ -17,3 +19,20 @@ Derivations are serialised in one of the following formats:
The only `version-string`s that are in use today are for [experimental features](@docroot@/development/experimental-features.md):
- `"xp-dyn-drv"` for the [`dynamic-derivations`](@docroot@/development/experimental-features.md#xp-feature-dynamic-derivations) experimental feature.
## Use for encoding to store object
When derivation is encoded to a [store object] we make the following choices:
- The store path name is the derivation name with `.drv` suffixed at the end
Indeed, the ATerm format above does *not* contain the name of the derivation, on the assumption that a store path will also be provided out-of-band.
- The derivation is content-addressed using the ["Text" method] of content-addressing derivations
Currently we always encode derivations to store object using the ATerm format (and the previous two choices),
but we reserve the option to encode new sorts of derivations differently in the future.
@@ -24,7 +24,7 @@ is a JSON object with the following fields:
*`method`:
For an output which will be [content addresed], a string representing the [method](@docroot@/store/store-object/content-address.md) of content addressing that is chosen.
For an output which will be [content addressed], a string representing the [method](@docroot@/store/store-object/content-address.md) of content addressing that is chosen.
@@ -7,7 +7,7 @@ The format of this specification is close to [Extended Backus–Naur form](https
Regular users do *not* need to know this information --- store paths can be treated as black boxes computed from the properties of the store objects they refer to.
But for those interested in exactly how Nix works, e.g. if they are reimplementing it, this information can be useful.
[store path](@docroot@/store/store-path.md)
[store path]: @docroot@/store/store-path.md
## Store path proper
@@ -20,14 +20,17 @@ where
-`store-dir` = the [store directory](@docroot@/store/store-path.md#store-directory)
-`digest` = base-32 representation of the first 160 bits of a [SHA-256] hash of `fingerprint`
-`digest` = base-32 representation of the compressed to 160 bits [SHA-256] hash of `fingerprint`
This the hash part of the store name
For the definition of the hash compression algorithm, please refer to the section 5.1 of
the [Nix thesis](https://edolstra.github.io/pubs/phd-thesis.pdf), which also defines the
specifics of base-32 encoding. Note that base-32 encoding processes the hash bytestring from
the end, while base-16 processes in from the beginning.
## Fingerprint
- ```ebnf
fingerprint = type ":" sha256 ":" inner-digest ":" store ":" name
fingerprint = type ":sha256:" inner-digest ":" store ":" name
```
Note that it includes the location of the store as well as the name to make sure that changes to either of those are reflected in the hash
@@ -53,7 +56,7 @@ where
method of content addressing store objects,
if the hash algorithm is [SHA-256].
Just like in the "Text" case, we can have the store objects referenced by their paths.
Additionally, we can have an optional `:self` label to denote selfreference.
Additionally, we can have an optional `:self` label to denote self-reference.
- ```ebnf
| "output:" id
@@ -70,7 +73,8 @@ where
`id` is the name of the output (usually, "out").
For content-addressed store objects, `id`, is always "out".
- `inner-digest` = base-16 representation of a SHA-256 hash of `inner-fingerprint`
- `inner-digest` = base-16 representation of a SHA-256 hash of `inner-fingerprint`.
The base-16 encoding uses lower-cased hex digits.
## Inner fingerprint
@@ -82,7 +86,7 @@ where
- if `type` = `"source:" ...`:
the hash of the [Nix Archive (NAR)] serialization of the [file system object](@docroot@/store/file-system-object.md) of the store object.
the [Nix Archive (NAR)] serialization of the [file system object](@docroot@/store/file-system-object.md) of the store object.
- To operate on a flake outside the current directory, you must now pass `--flake path/to/flake`.
- The flake-specific flags `--recreate-lock-file` and `--update-input` have been removed from all commands operating on installables.
They are superceded by `nix flake update`.
They are superseded by `nix flake update`.
- Commit signature verification for the [`builtins.fetchGit`](@docroot@/language/builtins.md#builtins-fetchGit) is added as the new [`verified-fetches` experimental feature](@docroot@/development/experimental-features.md#xp-feature-verified-fetches).
- Modify `nix derivation {add,show}` JSON format [#9866](https://github.com/NixOS/nix/issues/9866) [#10722](https://github.com/NixOS/nix/pull/10722)
The JSON format for derivations has been slightly revised to better conform to our [JSON guidelines](@docroot@/development/cli-guideline.md#returning-future-proof-json).
In particular, the hash algorithm and content addressing method of content-addresed derivation outputs are now separated into two fields `hashAlgo` and `method`,
In particular, the hash algorithm and content addressing method of content-addressed derivation outputs are now separated into two fields `hashAlgo` and `method`,
rather than one field with an arcane `:`-separated format.
This JSON format is only used by the experimental `nix derivation` family of commands, at this time.
`<nix/fetchurl.nix>` is also known as the builtin derivation builder `builtin:fetchurl`. It's not to be confused with the evaluation-time function `builtins.fetchurl`, which was not affected by this issue.
# Contributors
## Contributors
This release was made possible by the following 58 contributors:
- Support for relative path inputs [#10089](https://github.com/NixOS/nix/pull/10089)
Flakes can now refer to other flakes in the same repository using relative paths, e.g.
```nix
inputs.foo.url = "path:./foo";
```
uses the flake in the `foo` subdirectory of the referring flake. For more information, see the documentation on [the `path` flake input type](@docroot@/command-ref/new-cli/nix3-flake.md#path-fetcher).
This feature required a change to the lock file format. Previous Nix versions will not be able to use lock files that have locks for relative path inputs in them.
- Flake lock file generation now ignores local registries [#12019](https://github.com/NixOS/nix/pull/12019)
When resolving indirect flake references like `nixpkgs` in `flake.nix` files, Nix will no longer use the system and user flake registries. It will only use the global flake registry and overrides given on the command line via `--override-flake`.
This avoids accidents where users have local registry overrides that map `nixpkgs` to a `path:` flake in the local file system, which then end up in committed lock files pushed to other users.
In the future, we may remove the use of the registry during lock file generation altogether. It's better to explicitly specify the URL of a flake input. For example, instead of
The advantage is that this avoids a time window where *path* is not a garbage collector root, and so could be deleted by a concurrent `nix store gc` process.
- `nix-instantiate --eval` now supports `--raw` [#12119](https://github.com/NixOS/nix/pull/12119)
The `nix-instantiate --eval` command now supports a `--raw` flag, when used
the evaluation result must be a string, which is printed verbatim without
This update improves the reliability of SSH-related operations using `NIX_SSHOPTS` across Nix CLIs.
- Nix is now built using Meson
As proposed in [RFC 132](https://github.com/NixOS/rfcs/pull/132), Nix's build system now uses Meson/Ninja. The old Make-based build system has been removed.
- Evaluation caching now works for dirty Git workdirs [#11992](https://github.com/NixOS/nix/pull/11992)
## Contributors
This release was made possible by the following 45 contributors:
- Anatoli Babenia [**(@abitrolly)**](https://github.com/abitrolly)
- Handle the case where a chroot store is used and some inputs are in the "host" `/nix/store` [#12512](https://github.com/NixOS/nix/pull/12512)
The evaluator now presents a "union" filesystem view of the `/nix/store` in the host and the chroot.
This change also removes some hacks that broke `builtins.{path,filterSource}` in chroot stores [#11503](https://github.com/NixOS/nix/issues/11503).
- `nix flake prefetch` now has a `--out-link` option [#12443](https://github.com/NixOS/nix/pull/12443)
- Set `FD_CLOEXEC` on sockets created by curl [#12439](https://github.com/NixOS/nix/pull/12439)
Curl created sockets without setting `FD_CLOEXEC`/`SOCK_CLOEXEC`. This could previously cause connections to remain open forever when using commands like `nix shell`. This change sets the `FD_CLOEXEC` flag using a `CURLOPT_SOCKOPTFUNCTION` callback.
This is an atypical release, and for almost all intents and purposes, it is just a continuation of 2.27; not a feature release.
We had originally set the goal of making 2.27 the Nixpkgs default for NixOS 25.05, but dependents that link to Nix need certain _interface breaking_ changes in the C++ headers. This is not something we should do in a patch release, so this is why we branched 2.28 right off 2.27 instead of `master`.
This completes the infrastructure overhaul for the [RFC 132](https://github.com/NixOS/rfcs/blob/master/rfcs/0132-meson-builds-nix.md) switchover to meson as our build system.
## Major changes
- Unstable C++ API reworked
[#12836](https://github.com/NixOS/nix/pull/12836)
[#12798](https://github.com/NixOS/nix/pull/12798)
[#12773](https://github.com/NixOS/nix/pull/12773)
Now the C++ interface confirms to common conventions much better than before:
- All headers are expected to be included with the initial `nix/`, e.g. as `#include "nix/....hh"` (what Nix's headers now do) or `#include <nix/....hh>` (what downstream projects may choose to do).
Likewise, the pkg-config files have `-I${includedir}` not `-I${includedir}/nix` or similar.
Including without the `nix/` like before sometimes worked because of how for `#include` C pre-process checks the directory containing the current file, not just the lookup path, but this was not reliable.
- All configuration headers are included explicitly by the (regular) headers that need them.
There is no more need to pass `-include` to force additional files to be included.
- The public, installed configuration headers no longer contain implementation-specific details that are not relevant to the API.
The vast majority of definitions that were previously in there are now moved to new headers that are not installed, but used during Nix's own compilation only.
The remaining macro definitions are renamed to have `NIX_` as a prefix.
- The name of the Nix component the header comes from
(e.g. `util`, `store`, `expr`, `flake`, etc.)
is now part of the path to the header, coming after `nix` and before the header name
(or rest of the header path, if it is already in a directory).
Here is a contrived diff showing a few of these changes at once:
```diff
@@ @@
-#include "derived-path.hh"
+#include "nix/store/derived-path.hh"
@@ @@
+// Would include for the variables used before. But when other headers
+// need these variables. those will include these config themselves.
+#include "nix/store/config.hh"
+#include "nix/expr/config.hh"
@@ @@
-#include "config.hh"
+// Additionally renamed to distinguish from components' config headers.
+#include "nix/util/configuration.hh"
@@ @@
-#if HAVE_ACL_SUPPORT
+#if NIX_SUPPORT_ACL
@@ @@
-#if HAVE_BOEHMGC
+#if NIX_USE_BOEHMGC
@@ @@
#endif
#endif
@@ @@
-const char *s = "hi from " SYSTEM;
+const char *s = "hi from " NIX_LOCAL_SYSTEM;
```
- C API `nix_flake_init_global` removed [#5638](https://github.com/NixOS/nix/issues/5638) [#12759](https://github.com/NixOS/nix/pull/12759)
In order to improve the modularity of the code base, we are removing a use of global state, and therefore the `nix_flake_init_global` function.
Instead, use `nix_flake_settings_add_to_eval_state_builder`.
Although this change is not as critical, we figured it would be good to do this API change at the same time, also.
Also note that we try to keep the C API compatible, but we decided to break this function because it was young and likely not in widespread use yet. This frees up time to make important progress on the rest of the C API.
## Contributors
This earlier-than-usual release was made possible by the following 16 contributors:
After the special backport-based release of Nix 2.28 (timed to coincide with Nixpkgs 25.05), the release process is back to normal with 2.29.
As such, we have slightly more weeks of work from `master` (since 2.28 was branched from 2.27) than usual.
This fact is counterbalanced by the fact that most of those changes are bug fixes rather than larger new features.
- Prettified JSON output on the terminal [#12555](https://github.com/NixOS/nix/issues/12555) [#12652](https://github.com/NixOS/nix/pull/12652)
This makes the output easier to read.
Scripts are mostly unaffected because for those, stdout will be a file or a pipe, not a terminal, and for those, the old single-line behavior applies.
`--json --pretty` can be passed to enable it even if the output is not a terminal.
If your script creates a pseudoterminal for Nix's stdout, you can pass `--no-pretty` to disable the new behavior.
- Repl: improve continuation prompt for incomplete expressions [#12846](https://github.com/NixOS/nix/pull/12846)
Improved REPL user experience by updating the continuation prompt from invisible blank spaces to a visible `" > "`, enhancing clarity when entering multi-line expressions.
- REPL `:load-flake` and `:reload` now work together [#8753](https://github.com/NixOS/nix/issues/8753) [#13180](https://github.com/NixOS/nix/pull/13180)
Previously, `:reload` only reloaded the files specified with `:load` (or on the command line).
Now, it also works with the flakes specified with `:load-flake` (or on the command line).
This makes it correctly reload everything that was previously loaded, regardless of what sort of thing (plain file or flake) each item is.
- Increase retry delays on HTTP 429 Too Many Requests [#13052](https://github.com/NixOS/nix/pull/13052)
When downloading Nix, the retry delay was previously set to 0.25 seconds. It has now been increased to 1 minute to better handle transient CI errors, particularly on GitHub.
- S3: opt-in the STSProfileCredentialsProvider [#12646](https://github.com/NixOS/nix/pull/12646)
Added support for STS-based authentication for S3-based binary caches, i.e. enabling seamless integration with `aws sso login`.
- Reduce connect timeout for http substituter [#12876](https://github.com/NixOS/nix/pull/12876)
Previously, the Nix setting `connect-timeout` had no limit. It is now set to `5s`, offering a more practical default for users self-hosting binary caches, which may occasionally become unavailable, such as during updates.
- C API: functions for locking and loading a flake [#10435](https://github.com/NixOS/nix/issues/10435) [#12877](https://github.com/NixOS/nix/pull/12877) [#13098](https://github.com/NixOS/nix/pull/13098)
This release adds functions to the C API for handling the loading of flakes. Previously, this had to be worked around by using `builtins.getFlake`.
C API consumers and language bindings now have access to basic locking functionality.
It does not expose the full locking API, so that the implementation can evolve more freely.
Locking is controlled with the functions, which cover the common use cases for consuming a flake:
-`nix_flake_lock_flags_set_mode_check`
-`nix_flake_lock_flags_set_mode_virtual`
-`nix_flake_lock_flags_set_mode_write_as_needed`
-`nix_flake_lock_flags_add_input_override`, which also enables `virtual`
This change also introduces the new `nix-fetchers-c` library, whose single purpose for now is to manage the (`nix.conf`) settings for the built-in fetchers.
More details can be found in the [C API documentation](@docroot@/c-api.md).
- No longer copy flakes that are in the nix store [#10435](https://github.com/NixOS/nix/issues/10435) [#12877](https://github.com/NixOS/nix/pull/12877) [#13098](https://github.com/NixOS/nix/pull/13098)
Previously, we would duplicate entries like `path:/nix/store/*` back into the Nix store.
This was prominently visible for pinned system flake registry entries in NixOS, e.g., when running `nix run nixpkgs#hello`.
- Consistently preserve error messages from cached evaluation [#12762](https://github.com/NixOS/nix/issues/12762) [#12809](https://github.com/NixOS/nix/pull/12809)
In one code path, we are not returning the errors cached from prior evaluation, but instead throwing generic errors stemming from the lack of value (due to the error).
These generic error messages were far less informative.
Now we consistently return the original error message.
The implementation for blake3 hashing is now multi-threaded and used memory-mapped IO.
Benchmark results can be found the [pull request](https://github.com/NixOS/nix/pull/12676).
- Fix progress bar for S3 binary caches and make file transfers interruptible [#12877](https://github.com/NixOS/nix/issues/12877) [#13098](https://github.com/NixOS/nix/issues/13098) [#12538](https://github.com/NixOS/nix/pull/12538)
The progress bar now correctly display upload/download progress for S3 up/downloads. S3 uploads are now interruptible.
- Add host attribute of github/gitlab flakerefs to URL serialization [#12580](https://github.com/NixOS/nix/pull/12580)
Resolved an issue where `github:` or `gitlab:` URLs lost their `host` attribute when written to a lockfile, resulting in invalid URLs.
- Multiple signatures support in store urls [#12976](https://github.com/NixOS/nix/pull/12976)
Added support for a `secretKeyFiles` URI parameter in Nix store URIs, allowing multiple signing key files to be specified as a comma-separated list.
This enables signing paths with multiple keys. This helps with [RFC #149](https://github.com/NixOS/rfcs/pull/149) to enable binary cache key rotation in the NixOS infra.
- nix flake show now skips over import-from-derivation [#4265](https://github.com/NixOS/nix/issues/4265) [#12583](https://github.com/NixOS/nix/pull/12583)
Previously, if a flake contained outputs relying on [import from derivation](@docroot@/language/import-from-derivation.md) during evaluation, `nix flake show` would fail to display the rest of the flake. The updated behavior skips such outputs, allowing the rest of the flake to be shown.
Note that unlike `nix build`, this prints the full path to the program, not just the store path (in the example above that would be `/nix/store/cb9w44vkhk2x4adfxwgdkkf5gjmm856j-treefmt`).
- Amend OSC 8 escape stripping for xterm-style separator [#13109](https://github.com/NixOS/nix/pull/13109)
Improve terminal escape code filtering to understand a second type of hyperlink escape codes.
This in particular prevents parts of GCC 14's diagnostics from being improperly filtered away.
## Contributors
This release was made possible by the following 40 contributors:
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
