Commit graph

1120 commits

Author SHA1 Message Date
Adam Joseph
7ddea7340f fix(tvix/eval): catchable in type field of nix_eq()
Change-Id: I165ff77764e272cc94d18cb03ad6cbc9a8ebefde
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10348
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
2023-12-29 21:34:45 +00:00
Adam Joseph
2af8174e2e fix(tvix/eval): builtins.match: propagate catchables
Change-Id: I14c9e625c91369e10d0c00380dca992811ae9059
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10346
Reviewed-by: sterni <sternenseemann@systemli.org>
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
2023-12-29 21:34:45 +00:00
Adam Joseph
11e35a77a6 refactor(tvix/eval): let OpCoerceToString select the CoercionKind
Change-Id: I92d58ef216d7e0766af70f019b3dcd445284a95d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10344
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-12-29 21:34:45 +00:00
Adam Joseph
0c22454bb9 fix(tvix/eval): add stack depth assertion to OpReturn
I'm still trying to work out the exact stack invariants for tvix.
We really should add assertions for them; getting the stack messed
up is no fun.  This commit adds one simple assertion.  It also adds
a missing stack-push (my mistake) in one place, which was uncovered
by the assertion.

Change-Id: I9d8b4bd1702d954e325832c5935b0d7e3eb68422
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10369
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-12-29 21:34:45 +00:00
sterni
d07600dbca fix(tvix/eval/value): correctly emit spaces when coercing lists
r/7176 introduced an incorrect assumption was the benefit of the
nonrecursive coercion algorithm, namely that a coercion operation always
returns a non empty string. This allows to detect whether we are
coercing a list or not by checking if the intermediate result is empty
or not. Unfortunately, coercing null and false yields an empty string,
so we need to explicitly track whether we are coercing a list.

Updated the test case to hopefully catch similar bugs in the future. I'm
not a hundred percent certain I have not introduced a new edge case with
this, so it may be interesting to add a prop test case for this to
nix_oracle down the line. At least lists are the only nested data
structures that can be serialized as nested data structures, so the
problem is kind of limited.

Change-Id: Ia41e904356f1c41a9d35e4e65ec02f2fe5a4100e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10418
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Autosubmit: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
2023-12-25 15:06:23 +00:00
sterni
7165ebc43b fix(tvix/eval): remove incorrect imports when coercing
The default behavior of string coercion in C++ Nix is to weakly coerce
and import to store if necessary. There is a flag to make it strongly
coerce (coerceMore) and a flag that controls whether path values have
the corresponding file/directory imported into the store before
returning the (store) path as a string (copyToStore). We need to
implement our equivalent to the copyToStore (import_paths) flag for the
benefit of weak coercions that don't import into the store (dirOf,
baseNameOf, readFile, ...) and strong coercions that don't import into
the store (toString).

This makes coerce_to_string as well as CoercionKind weirder and more
versatile, but prevents us from reimplementing parts of the coercion
logic constantly as can be seen in the case of baseNameOf.

Note that it is not possible to test this properly in //tvix/eval tests
due to the lack of an appropriate EvalIO implementation being available.
Tests should be added to //tvix/glue down the line.

Change-Id: I8fb8ab99c7fe08e311d2ba1c36960746bf22f566
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10361
Autosubmit: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
Reviewed-by: Adam Joseph <adam@westernsemico.com>
2023-12-14 13:15:23 +00:00
sterni
a30dd0905a fix(tvix/eval): determine meaning of + exprs based on first type
This matches the behavior of C++ Nix more closely where the decision is
made based on the first type based to ExprConcatStrings:
1f93fa2ed2/src/libexpr/eval.cc (L1967-L2025)

Note that this doesn't make a difference in any successful
evaluation (at least to my knowledge), but ensures that our error
messages will match C++ Nix more closely, e.g. in the case of
`1 + "string"`.

Change-Id: I8059930788f9c8d98baf98e3d93d8a060ef961f2
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10360
Tested-by: BuildkiteCI
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: Adam Joseph <adam@westernsemico.com>
2023-12-14 02:59:58 +00:00
Adam Joseph
91456c3520 refactor(tvix/eval): vm::add_values(): be less verbose
Change-Id: Icf328649fd70bdf0fc3ba6cd7754ae29735f11f7
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10035
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 18:00:37 +00:00
Adam Joseph
243a4b5699 fix(tvix/eval): substring: propagate catchables
Change-Id: Ia9b7858c817fbc9c95a3d1c2855b2445f7830e8d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10326
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 17:51:29 +00:00
Adam Joseph
c5f58d4af3 fix(tvix/eval): baseNameOf: propagate catchables
Change-Id: Id8dc772ea8f338dfd243210f4108f79072570c3b
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10324
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 17:50:56 +00:00
Adam Joseph
2565a21aa9 fix(tvix/eval): builtins.length: propagate catchables
This commit fixes out builtins.length so it propagates catchables
like cppnix does.

Change-Id: I7670bec5eee1d4cd3f67a04c9a6808979fb56a8d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10315
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
2023-12-12 17:49:54 +00:00
Adam Joseph
1ddfed9825 fix(tvix/eval): builtins.filter: propagate catchables
This commit fixes builtins.filter so it propagates catchables
correctly.

Change-Id: Ib23a383bc5e272e42052205ffd1e94649a0ebc47
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10313
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 17:35:39 +00:00
Adam Joseph
25dd1458b9 feat(tvix/eval): builtins.hashString: add placeholder
This adds an unimplemented placeholder for builtins.hashString.

Change-Id: Ibc770103acf5dbc3ea7589ab5ca23fe6e07bd91a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10311
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 17:35:06 +00:00
Adam Joseph
8c8409c0d2 fix(tvix/eval): builtins.getAttr: propagate catchables
Change-Id: I84b6b8f8568d57614a03aff0d6069e0bc27357bf
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10310
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 17:34:34 +00:00
Adam Joseph
afba150036 fix(tvix/eval): builtins.elemAt: propagate catchables
This commit fixes builtins.elemAt so it propagates catchables like
cppnix does.

Change-Id: Ieca5e128da17e78af0b14dae4a28a1ff8796e4f2
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10308
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 17:34:02 +00:00
Adam Joseph
29464fc65b fix(tvix/eval): propagate catchables through builtins.splitVersion
This fixes our implementation of builtins.splitVersion so it
propagates catchables like cppnix does.

Change-Id: Id5d83ea76229f8c8f202aa42353cb609e67de43f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10305
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 17:21:53 +00:00
Adam Joseph
012b44244b feat(tvix/eval): builtins.hasContext: placeholder implementation
Currently this just `throw`s a message explaining that it is not
implemented.  This is necessary in order to allow enumerating the
nixpkgs release attrset (afaict only one package uses this builtin).

Change-Id: I45266d46af579ddb5856b192b6be4b481369543c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10302
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 17:21:53 +00:00
Adam Joseph
531f0c0c42 fix(tvix/eval): baseNameOf should not coerce paths into strings
... since this may import them to the store which changes their
basename.

Fixes b/350.

Change-Id: Iabd08ff4d6a424c66d6d7784d7a96b0c078f0a91
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10298
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 17:21:21 +00:00
Adam Joseph
8d4aa2c15c fix(tvix/eval): add unimplemented __curPos and builtins.filterSource
This commit adds __curPos (to the global scope, yuck) and
builtins.filterSource.  These are not implemented; forcing them will
produce the same result as `throw "message"`.

Unfortunately these two post-2.3 features are used throughout
nixpkgs.  Since an unresolved indentifier is a catchable error, this
breaks the entire release eval.  With this commit, it simply causes
those broken packages that use these features to appear as they are:
broken.

Change-Id: Ib43dea571f6a9fab4d54869349f80ee4ec5424c2
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10297
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 17:20:50 +00:00
Adam Joseph
bb1e79e5d1 fix(tvix/eval): propagate catchables through &&
Change-Id: I7bb5ac1ef47b41c47269e64cee0e90eb64c6fcce
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10322
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 17:04:07 +00:00
Adam Joseph
bcc1ea8552 fix(tvix/eval): make || propagate catchables
Change-Id: I42f994d7c9228368d5f6c30c4730c24666f7bc69
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10320
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 17:04:07 +00:00
Adam Joseph
7b9eea71d0 fix(tvix/eval): fix nested assertions b/340
This commit fixes our handling of `throw` within an `assert`
condition.

Fixes: b/340
Change-Id: I40a383639ec266da50a853f16216b1b7868495da
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10318
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 17:03:03 +00:00
Adam Joseph
fc963033ae fix(tvix/eval): ?: propagate catchables
This commit fixes out `?` operator so it correctly propagates
catchables.

Change-Id: Iebaa153a8492101ee3ddd29893c98730ff331547
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10317
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-12-12 16:57:55 +00:00
Adam Joseph
990c4e727f feat(tvix/eval): OpAttrsSelect should propagate catchables
Previously, using a catchable as either argument of OpAttrsSelect
would result in an unrecoverable error.  This commit matches cppnix
behavior by propagating the catchable.

Change-Id: I4877f4068ec2b823225f185290693c101d0b9c9e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10303
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 16:09:47 +00:00
Adam Joseph
2949ee08f1 fix(tvix/eval): calling a catchable is catchable
When attempting to call a Value, if it is a Value::Catchable we must
not cause an uncatchable failure.  This commit simply reuses the
Value::Catchable as the result of attempting to call it.  This is
safe because nix is designed so that nix code cannot distinguish
between different catchable failures -- they all look the same to
the interpreted code.

This fixes b/351.

Change-Id: Ibf763a08753e541843626182ff59fdbf15ea2959
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10300
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 16:09:15 +00:00
Adam Joseph
52b68c0539 fix(tvix/eval): fix catchables in named formals
Fixes b/348.

Change-Id: I5e8d56b5fd26a19eac32ec5e11baf93765691dc8
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10296
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-12-12 15:55:06 +00:00
Adam Joseph
e516046ed2 test(tvix/eval): test catchables in named formals
Relates to b/348.

$ /git/depot/result/bin/tvix -E '(builtins.tryEval (({ fred }: "bob") (throw "3"))).success'
note: while evaluating this Nix code
 --> [code]:1:1
  |
1 | (builtins.tryEval (({ fred }: "bob") (throw "3"))).success
  | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
note: while evaluating this as native code (force)
 --> [code]:1:1
  |
1 | (builtins.tryEval (({ fred }: "bob") (throw "3"))).success
  | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
note: while evaluating this as native code (tryEval)
 --> [code]:1:2
  |
1 | (builtins.tryEval (({ fred }: "bob") (throw "3"))).success
  |  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
note: while evaluating this Nix code
 --> [code]:1:20
  |
1 | (builtins.tryEval (({ fred }: "bob") (throw "3"))).success
  |                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
error[E006]: expected value of type 'set', but found a 'internal[catchable]'
 --> [code]:1:21
  |
1 | (builtins.tryEval (({ fred }: "bob") (throw "3"))).success
  |                     ^^^^^^^^

Change-Id: I730fdd996f7e1b81dbbf83dc1524104a8cad2f78
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10295
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 15:54:04 +00:00
Adam Joseph
663362f3df fix(tvix/eval): fix testing catchables for inequality
Fixes b/347.

Change-Id: Icad0251884d4d8adcdf8d690b91385bf4896f9c8
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10294
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 15:53:32 +00:00
Adam Joseph
1b2a1892cb test(tvix/eval): testing catchable for inequality
Relates to b/347.

    $ /git/depot/result/bin/tvix -E '(builtins.tryEval (throw "bob" != 3)).success'
    note: while evaluating this Nix code
     --> [code]:1:1
      |
    1 | (builtins.tryEval (throw "bob" != 3)).success
      | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    note: while evaluating this as native code (force)
     --> [code]:1:1
      |
    1 | (builtins.tryEval (throw "bob" != 3)).success
      | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    note: while evaluating this as native code (tryEval)
     --> [code]:1:2
      |
    1 | (builtins.tryEval (throw "bob" != 3)).success
      |  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    error[E006]: expected value of type 'bool', but found a 'internal[catchable]'
     --> [code]:1:20
      |
    1 | (builtins.tryEval (throw "bob" != 3)).success
      |                    ^^^^^^^^^^^^^^^^

Change-Id: Ide19b3ddaf314ef310efc2fe5ac36667e43011dc
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10293
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
2023-12-12 15:49:28 +00:00
Adam Joseph
289663cac4 fix(tvix/eval): handle catchables in attribute set updates
Fixes b/346.

Change-Id: I277121d2363e605ebe09651ed9440fe1bc126c8c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10292
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 15:48:25 +00:00
Adam Joseph
e54533518b test(tvix/eval): test for catchable in attribute merges
Relates to b/346.

    $ /git/depot/result/bin/tvix -E '(builtins.tryEval (throw "bob" // { })).success'
    note: while evaluating this Nix code
     --> [code]:1:1
      |
    1 | (builtins.tryEval (throw "bob" // { })).success
      | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    note: while evaluating this as native code (force)
     --> [code]:1:1
      |
    1 | (builtins.tryEval (throw "bob" // { })).success
      | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    note: while evaluating this as native code (tryEval)
     --> [code]:1:2
      |
    1 | (builtins.tryEval (throw "bob" // { })).success
      |  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    error[E006]: expected value of type 'set', but found a 'internal[catchable]'
     --> [code]:1:20
      |
    1 | (builtins.tryEval (throw "bob" // { })).success
      |                    ^^^^^^^^^^^^^^^^^^

Change-Id: Ib84c4ec6d2612d4f1f6066e66c3dd1bf04369b6e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10291
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 15:48:25 +00:00
Adam Joseph
edf411a86c fix(tvix/eval): fix recovering from throws in implications
This fixes b/345.

Change-Id: Ic0d3b6ffacd2a5e0050d22354d08320b69a4fe13
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10290
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 15:47:53 +00:00
Adam Joseph
24ff74d346 test(tvix/eval): test recovering from throw in implications
error[E006]: expected value of type 'bool', but found a 'internal[catchable]'
 --> src/tests/tvix_tests/notyetpassing/eval-okay-test-catchables-in-implications.nix:1:43
  |
1 | (builtins.tryEval (({ foo ? throw "up" }: foo -> true) { })).success
  |                                           ^^^^^^^^^^^

Relates to b/345

Change-Id: Ic331c32ea59bf67ae775f485b444dc6804ca13d5
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10289
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 15:47:21 +00:00
Adam Joseph
9792920f8c fix(tvix/eval): fix branching on catchable defaults (b/343)
This commit adds Opcode::OpJumpIfCatchable, which can be inserted
ahead of most VM operations which expect a boolean on the stack, in
order to handle catchables in branching position properly.

Other than remembering to patch the jump, no other changes should be
required.

This commit also fixes b/343 by emitting this new opcode when
compiling if-then-else.  There are probably other places where we
need to do the same thing.

Change-Id: I48de3010014c0bbeba15d34fc0d4800e0bb5a1ef
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10288
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 14:55:48 +00:00
Adam Joseph
e54eeda0ff test(tvix/eval): test branching on catchable defaults (b/343)
This is a test case for b/343, wherein tvix dies if you try to
branch on an argument whose defaulted value is a catchable.

Change-Id: I891ca825e39ad14dda9f220f06d9591874fcd45d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10287
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
2023-12-12 14:55:48 +00:00
Adam Joseph
1ac57b0d1c feat(tvix/eval): nonrecursive coerce_to_string()
After this commit, the only non-builtins uses of generators are:

  - coerce_to_string() uses generators::request_enter_lambda()
  - Thunk::force() uses generators::request_enter_lambda()

That's it!  Once those two are taken care of, GenCo can become an
implementation detail of `builtins::BuiltinGen`.  No more crazy
nonlocal flow control within the interpreter: if you've got a GenCo
floating around in your code it's because you're writing a builtin,
which isn't part of the core interpreter.  The interpreter won't
need GenCos to talk to itself anymore.

Technically generators::request_path_import() is also used by
coerce_to_string(), but that's just because the io_handle happens to
be part of the VM.  There's no recursion-depth issue there, so the
call doesn't need to go through the generator mechanism
(request_path_import() doesn't call back to the interpreter!)

Change-Id: I83ce5774d49b88fdafdd61160975b4937a435bb0
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10256
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 14:54:46 +00:00
Adam Joseph
3701379745 feat(tvix/eval): nonrecursive deep_force()
This commit implements deep_force() nonrecursively, by maintaining
an explicit stack rather than using the call stack for recursion.

As an added bonus, we don't need to pass around the SharedThunkSet
anymore, and can in fact completely eliminate SharedThunkSet.

Change-Id: I7c4f59f37834d451a28bf6be317eb0a90eac4ee6
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10252
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 14:36:02 +00:00
Adam Joseph
c956138ee6 docs(tvix/eval): clarify difference between ThunkSet and Blackhole
The comment explaining ThunkSet makes it seem like it does the same
think as ThunkRepr::Blackhole.  In fact neither one is a substitute
for the other.  Let's explain the difference.

Change-Id: I89ceaaa9d3c499edbc7d48f70ca5d11f97666c43
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10250
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 14:35:00 +00:00
Adam Joseph
ddba4d4e17 test(tvix/eval): nested assertions (b/340)
Change-Id: I898d7056877a6370d5720b633df416f54e7cf65f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10222
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: tazjin <tazjin@tvl.su>
2023-12-12 14:35:00 +00:00
Adam Joseph
c92d06271d feat(tvix/eval): drop LightSpan::Delayed
LightSpan::Delayed was introduced in commit
bf286a54bc which claimed that "This
reduces the eval time for `builtins.length (builtins.attrNames
(import <nixpkgs> {}))` by *one third*!"

I am unable to reproduce this result.  In fact, dropping the
LightSpan::Delayed variant of the enum makes eval of the same
expression slightly faster!  I also tried a large evaluation
(pkgsCross...hello) and got similar results: slightly faster,
slightly less memory.  See git footers.

I suspect that there was some unrelated horrific inefficiency that
has since been fixed.  The avoided computation in `get_span()` is
nothing more than a binary search!  If this were in fact a major
performance issue we could simply precompute the mapping from
CodeIdx to Span when the Chunk becomes immutable (i.e. at the end of
the compilation process, when compiler backtracking is no longer a
concern).  Since a Span is just 64 bits this is not a space issue,
and since binary search is much simpler than compiling Nix
expressions it isn't a performance issue either.

Technically there is no longer any reason to have LightSpan since it
is now a single-variant enum.  However there is no rush to remove
it, since Rust will optimize its representation into the same thing
you'd get if you replaced LightSpan by Span.

Prev-Benchmark: {"nixpkgs-attrnames":{"kbytes":"233824","system":"0.32","user":"2.02"}}
This-Benchmark: {"nixpkgs-attrnames":{"kbytes":"230192","system":"0.29","user":"2.00"}}
Prev-Benchmark: {"pkgsCross.aarch64-multiplatform.hello.outPath":{"kbytes":"458936","system":"0.73","user":"5.36"}}
This-Benchmark: {"pkgsCross.aarch64-multiplatform.hello.outPath":{"kbytes":"451808","system":"0.53","user":"5.10"}}

Change-Id: Ib9e04806850aa1fc4e66e2a042703986440a7b4e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10254
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
2023-12-12 14:34:28 +00:00
Florian Klink
27c07b72c6 refactor(tvix): use io::Result for EvalIO
This is just a alias for Result<_, io::Error>, but shorter.

Change-Id: I7c22f61b85e3014885a747b5c1e5abd11b0ef17d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10327
Tested-by: BuildkiteCI
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
2023-12-12 14:28:50 +00:00
Adam Joseph
ad566999ca fix(tvix/eval): preserve catchables in nix_cmp_ordering(), fix b/338
This commit fixes b/338 by properly propagating catchables through
comparison operations.

Change-Id: I6b0283a40f228ecf9a6398d24c060bdacb1077cf
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10221
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 14:26:46 +00:00
Adam Joseph
ae28dc3ca6 test(tvix/eval): test for b/338 catchable hygiene problem
Commit 05f42519b5 fixed b/281 by
establishing a hygiene regimen to partition *catchable* errors
(i.e. those which tryEval can detect) from all other errors, like
internal VM failures or I/O errors (which Nix must not be allowed to
detect, since these errors are fundamentally impure).

Unfotunately there are still cases where tvix assumes that anything
other than Value::Bool means it should panic!().  I found another
one, and added a test case for it in:

  eval_okay_src_tests_tvix_tests_eval_okay_compare_ordering_catchable_nix

Not yet passing.

Change-Id: I69c62ed9ea5c8f81870e8de5c5fe12dcde849763
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10220
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-12-12 14:26:46 +00:00
Adam Joseph
edbd5055a1 feat(tvix/eval): nonrecursive nix_cmp_ordering(), fixes b/339
This commit rewrites Value::nix_cmp_ordering() into an equivalent
nonrecursive form.  Except for calls to Thunk::force(), the new form
no longer uses generators, and is async only because of the fact
that it calls Thunk::force().

I originally believed that this commit would make evaluation faster.
In fact it is slightly slower.  I believe this is due to the added
vec![] allocation.  I am investigating.

Prev-Nixpkgs-Benchmark: {"attrpath":"pkgsCross.aarch64-multiplatform.hello.outPath","peak-kbytes":"460048","system-seconds":"0.68","user-seconds":"5.73"}
This-Nixpkgs-Benchmark: {"attrpath":"pkgsCross.aarch64-multiplatform.hello.outPath","peak-kbytes":"460224","system-seconds":"0.67","user-seconds":"5.84"}
Change-Id: Ic627bc220d9c5aa3c5e68b9b8bf199837cd55af5
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10212
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 14:26:46 +00:00
Adam Joseph
8a40f75c2d fix(tvix/eval): never use partial_cmp() (partial fix b/338)
This is part of a fix for b/338.

We should never use PartialOrd::partial_cmp().

All Nix types except floats are obviously totally-ordered.  In
addition, it turns out that because Nix treats division by zero
rather than producing a NaN, and because it does not support
"negative zero", even floats are in fact totally ordered in Nix.

Therefore, every call to PartialOrd::partial_cmp() in tvix is an
error.  We have to *implement* this function, but we should never
call it on built-in types.

Moreover, nix_cmp_ordering() currently returns an Option<Ordering>.
I'm not sure what was going on there, since it's impossible for it
to return None.  This commit fixes it to return simply Ordering
rather than Option<Ordering>.

Change-Id: If5c084164cf19cfb38c5a15554c0422faa5f895d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10218
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-12-12 14:26:46 +00:00
Adam Joseph
19d13eb070 test(tvix/eval): add test case for b/339
Not yet passing.

Change-Id: I1de3f72d8b3f46567fdba010fc3ab4bace3f1699
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10219
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
2023-12-12 14:26:46 +00:00
Adam Joseph
72ece2e518 feat(tvix/eval): nonrecursive nix_eq()
This commit rewrites Value::nix_eq() into an equivalent.  Except for
calls to Thunk::force(), the new form no longer uses generators, and
is async only because of the fact that it calls Thunk::force().

I believed that the nonrecursive form would be faster.  It is, in
fact, slightly slower.  I believe this is due to the vec![]
allocation; I am investigating.

Prev-Nixpkgs-Benchmark: {"attrpath":"pkgsCross.aarch64-multiplatform.hello.outPath","peak-kbytes":"459068","system-seconds":"0.71","user-seconds":"5.39"}
This-Nixpkgs-Benchmark: {"attrpath":"pkgsCross.aarch64-multiplatform.hello.outPath","peak-kbytes":"460048","system-seconds":"0.68","user-seconds":"5.73"}
Change-Id: I10f4868891e4b7475df13f0cbc41ec78dd985dd8
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10118
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-12-12 14:26:46 +00:00
Florian Klink
01a3c87a3a chore(tvix): bump criterion to 0.5
Change-Id: I28904ca23437b4bb745c0eb1f4eb9ae33e09eb5a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10244
Reviewed-by: grfn <grfn@gws.fyi>
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-12-11 03:59:30 +00:00
Florian Klink
d3ecef1a5b refactor(tvix/eval): address clippy lints
Change-Id: Ic2bd4e8291b30ceac9fa0e88a4f56e61ae99b603
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10227
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-12-09 12:58:39 +00:00
Adam Joseph
5fd42a9e70 feat(tvix/eval): impl DoubleEndedIter for OwnedAttrsIterator
Change-Id: I4bd85dbe9c27047f4abbdeff4e2b796e9bcab3a1
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10211
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
2023-12-09 11:35:19 +00:00
Adam Joseph
49b34183e3 feat(tvix/eval): rewrite Thunk::force() in nonrecursive form
This commit rewrites Thunk::force() so that it is not (directly)
self-recursive.  It maintains a Vec of all the
previously-encountered thunks which point to the one it is currently
forcing, rather than recursively calling itself.

Benefits:

- Short term:

  This commit saves the cost of a round-trip through the generator
  machinery for the generators::request_force() which is removed by
  this commit.

- Medium term:

  Once a similar transformation has been applied to nix_cmp(),
  nix_add(), nix_eq(), and coerce_to_string(), those four functions,
  along with Thunk::force(), will make non-tail calls only to each
  other.  They can then be merged into a single tail-recursive
  function which does not use the generator machinery at all:

    enum Task { Cmp, Add, Eq, CoerceToString, Force};

    fn Value::walk(task:Task, v1:Value, v2:Value) {
      // ...

- Long term:

  The long-term goal here is to use generators **only for builtins**
  and [Marionette]-style remote control of the VM.  In other words:
  use `async` for things that actually involve concurrency.  Calls
  from the VM to builtins can then be blocking calls, because even
  cppnix will overflow the stack if you make a MAX_STACK_DEPTH-deep
  recursive call which passes through a builtin at every stack frame
  (e.g. `{ func = builtins.sort (a: b: ... func ...) ...}`).

  This way the inner "tight loop" of the interpreter doesn't pay the
  costs of `async` and generators.  These costs manifest in terms
  of: performance, complex nonlocal control flow, and language
  impediments (async Rust is a restricted subset of real Rust, and
  is missing things like traits).

[Marionette]: https://firefox-source-docs.mozilla.org/testing/marionette/Intro.html

Change-Id: I6179b8abb2ea0492180fcb347f37595a14665777
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10039
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-12-06 06:53:01 +00:00
Aspen Smith
8135a8d38c fix(tvix/eval): Return error rather than panicking on bad substring
If builtins.substring is invoked with (byte!!) offsets that aren't at
codepoint boundaries, return an error rather than panicking. This is
still incorrect (see b/337) but pushes the incorrectness forward a step.

Change-Id: I5a4261f2ff250874cd36489ef598dcf886669d04
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10199
Tested-by: BuildkiteCI
Autosubmit: grfn <grfn@gws.fyi>
Reviewed-by: sterni <sternenseemann@systemli.org>
2023-12-05 23:12:23 +00:00
Vincent Ambo
e65fa82d74 refactor(tvix/eval): use or_default helper in entry API
This fixes a future clippy lint.

Change-Id: Ic830e94ef23595580c1037f10878c76bbb546dd9
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10110
Tested-by: BuildkiteCI
Reviewed-by: Adam Joseph <adam@westernsemico.com>
2023-11-25 15:48:46 +00:00
Vincent Ambo
5ffb997864 fix(tvix): ensure PartialOrd/Ord agree for StorePath & NixString
This fixes a *future* clippy lint:
https://rust-lang.github.io/rust-clippy/master/index.html#/incorrect_partial_ord_impl_on_ord_type

In essence, because the implementation of *both* Ord and PartialOrd
implies that ordering is not partial, all results of PartialOrd should
simply be those of Ord. This is to avoid subtle bugs in future
refactorings.

Change-Id: I8fc6694010208752dd47746a2aaaeca0c788d574
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10109
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
2023-11-25 15:48:46 +00:00
Adam Joseph
512346ba0b refactor(tvix/eval): add ThunkRepr::is_forced()
Change-Id: I4eab5c81fb82337da06327248845cd2f3a4490d3
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10038
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-11-25 02:55:57 +00:00
Adam Joseph
b5a15989cd feat(tvix/eval): add Thunk::unwrap_or_clone()
This commit adds Thunk::unwrap_or_clone(), which uses
Rc::try_unwrap() to avoid cloning the Value out of a an Rc which has
only one strong reference.

Change-Id: Icacefe0c823dcddf046d90c0c5cd5ed59fe976d4
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10037
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
2023-11-25 02:55:56 +00:00
sterni
c5961e7774 docs(tvix/eval): optimization potential for inherit (from) exprs
Change-Id: Ibddaa111a5b7a86c42dbe153ae8e53f9a5601a54
Reviewed-on: https://cl.tvl.fyi/c/depot/+/10112
Tested-by: BuildkiteCI
Reviewed-by: Adam Joseph <adam@westernsemico.com>
2023-11-24 23:36:15 +00:00
Vincent Ambo
7c32d85455 docs(tvix/eval): document where EvalIO methods are used
Change-Id: I335f2ba4420973861c2a22125995b45a34d3608c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9969
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Autosubmit: tazjin <tazjin@tvl.su>
2023-11-06 09:06:26 +00:00
Vincent Ambo
f325456cba chore(tvix): bump proptest dependency
This *might* contain a fix for a clippy lint thrown by that crate.

Relates to b/321.

Change-Id: Ia7ebd3e26e0feb8bcc7a6c811b1e583f9016fd9e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9966
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
Autosubmit: tazjin <tazjin@tvl.su>
2023-11-05 20:28:38 +00:00
Vincent Ambo
87d3fac676 chore(tvix): add missing clippy attributes & config
For cases where clippy lints don't apply to us, or something is
misfiring, add appropriate configuration.

Relates to b/321.

Change-Id: I0af453910b4a4112bf685b2a8e9a73de10ec87ea
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9965
Reviewed-by: flokli <flokli@flokli.de>
Autosubmit: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-11-05 20:28:38 +00:00
Vincent Ambo
2e1399698b refactor(tvix/eval): use IntoIterator trait for owned NixAttrs iter
Uses the standard library IntoIterator trait for the construction of
our iterators. Clippy complains about duplicating this.

While doing this, I opted to rename the `IntoIter` type into something
that is more useful to users, in case somebody ends up working with
these manually.

This fixes a clippy lint, and is related to b/321.

Change-Id: I851fde0d7b8b38d182343a0fd6d9f8dd2a33ee11
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9963
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
Autosubmit: tazjin <tazjin@tvl.su>
2023-11-05 20:28:38 +00:00
Vincent Ambo
b3b1f649d6 chore(tvix): fix trivial clippy lints
Relates to b/321.

Change-Id: I37284f89b186e469eb432e2bbedb37aa125a6ad4
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9961
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
Autosubmit: tazjin <tazjin@tvl.su>
2023-11-05 20:28:37 +00:00
Vincent Ambo
99f618bcb4 refactor(tvix/eval): delay allocation when comparing attr values
Delays allocation (through cloning) of the values to be compared
until *after* the keys have been compared.

Change-Id: I7d68c27d7a0fbcdcc387db7c092bce50ca4b94ea
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9900
Autosubmit: tazjin <tazjin@tvl.su>
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-11-03 09:24:31 +00:00
Vincent Ambo
2dd2b844c7 chore(tvix/eval): add a marker for sorted borrowed attrs iteration
Similar to `into_iter_sorted`, add a marker function for call sites
that want *borrowed* sorted iteration.

Change-Id: I7c6f14e1ac43fdb14b861b3da183eb5d12bba139
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9899
Autosubmit: tazjin <tazjin@tvl.su>
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-11-03 09:24:31 +00:00
Vincent Ambo
d91fcd4f28 refactor(tvix/eval): more efficiently intersect attributes
builtins.intersectAttrs is used a _lot_ in nixpkgs eval, for whatever
reason. We previously had a very inefficient implementation that would
allocate for each comparison. It stuck out like a sore thumb in perf
analysis.

This moves to a custom algorithm with two iterators, one for the left
and one for the right side, advancing them along the (borrowed) map
keys until a match is found and allocation is required.

I've not made any effort to reduce the verbosity of this code, I don't
think it's worth it.

On my machine this reduces the mean runtime of evaluating
`nixpkgs.emacs.outPath` by ~8%.

Change-Id: Ie506d82cb8d5f45909628f771a6b73e0eca16b27
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9898
Autosubmit: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
2023-11-03 09:24:30 +00:00
Florian Klink
1571d7195e fix(tvix/eval/benches): use black_box properly
The purpose of black_box is to actually prevent the compiler from being
able to optimize computation of the benchmarked function away.

To accomplish this, we need to actually *use* black_box to blackbox the
input data away, rather than the return type.

Change-Id: I5438982f57509fbf7b85034346a2739d76aef1fa
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9902
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-11-02 20:39:52 +00:00
Florian Klink
269ab866f1 refactor(tvix/eval/tests): migrate to tempfile
tempdir pulls in remove_dir_all 0.5.3 with
https://rustsec.org/advisories/RUSTSEC-2023-0018.html, and we use
tempfile everywhere else too, so let's just migrate to that.

Change-Id: I735ade7b65e12fc26e3d43ca95fcfa07fcc64642
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9565
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-10-08 22:47:33 +00:00
Adam Joseph
fac32d0a5f feat(tvix/eval): test case for b/281
This commit adds a test case for b/281.

Change-Id: I8dfbfc0ff636184d7882530d8aefb329a3af9e5c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9288
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: flokli <flokli@flokli.de>
2023-09-26 14:13:05 +00:00
Adam Joseph
05f42519b5 fix(tvix/eval): fix b/281 by adding Value::Catchable
This commit makes catchable errors a variant of Value.

The main downside of this approach is that we lose the ability to
use Rust's `?` syntax for propagating catchable errors.

Change-Id: Ibe89438d8a70dcec29e016df692b5bf88a5cad13
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9289
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
Tested-by: BuildkiteCI
2023-09-24 21:54:10 +00:00
Adam Joseph
926459ce69 refactor(tvix/eval): factor CatchableErrorKind out of ErrorKind
This commit creates a separate enum for "catchable" errors (the kind
that `builtins.tryEval` can detect).

Change-Id: Ie81d1112526d852255d9842f67045f88eab192af
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9287
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-09-24 21:54:09 +00:00
Florian Klink
731996fbfe docs(tvix/eval): fix some broken docstr references
There's some more left, but they've been renamed/refactored out of
sight.

Change-Id: I41579dedc74342b4c5f8cb39d2995b5b0c90b0f4
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9372
Tested-by: BuildkiteCI
Reviewed-by: Connor Brewster <cbrewster@hey.com>
Autosubmit: flokli <flokli@flokli.de>
2023-09-22 09:14:33 +00:00
Adam Joseph
9256621009 fix(tvix/eval): update identifier quoting to match cppnix 2.17
In cppnix 2.17, commit b72bc4a972fe568744d98b89d63adcd504cb586c, the
libexpr pretty-printing routine was fixed so that it would no longer
pretty-print attrsets with keywords in their attrnames incorrectly.

This commit implements the corresponding fix for tvix, fixes our
tests to work with cppnix>=2.17 oracles, and expands our test cases
to cover all the keywords.

Change-Id: I4b51389cd3a9c44babc8ab2a84b383b7b0b116ca
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9283
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
2023-09-15 00:00:59 +00:00
Adam Joseph
6fa80ecc3d fix(tvix/eval): make compare_strict_eval_tests use Strict
Previously, compare_strict_eval_tests() was using Strictness::Lazy.
This appears to be a minor oversight from
0ab6494286.

This commit corrects that, by changing Strictness::Lazy to
Strictness::Strict.

Change-Id: Ia2389a5d30481cd322ed55230731340b795e5d87
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9282
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: sterni <sternenseemann@systemli.org>
2023-09-09 16:11:57 +00:00
Vincent Ambo
d546b9a15d docs(tvix/eval): document remaining opcodes
Change-Id: Iad8a5f78930872719b6481ddf1bbad8532bfa888
Reviewed-on: https://cl.tvl.fyi/c/depot/+/7981
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-09-08 21:38:42 +00:00
Vincent Ambo
35e8dc97cf docs(tvix/eval): document attribute set related opcodes
Change-Id: Ib98a9fe8c9aa3f0e61c27d10285c0926cda7a969
Reviewed-on: https://cl.tvl.fyi/c/depot/+/7979
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
2023-09-08 21:38:42 +00:00
Vincent Ambo
3cf37c7cb6 docs(tvix/eval): add documentation strings for some OpCode variants
Change-Id: I42e610740b3687e1fd897d36694cce425751a8bc
Reviewed-on: https://cl.tvl.fyi/c/depot/+/7975
Tested-by: BuildkiteCI
Autosubmit: tazjin <tazjin@tvl.su>
Reviewed-by: flokli <flokli@flokli.de>
2023-09-08 21:38:42 +00:00
Linus Heckemann
a3dbb60213 fix(tvix/eval): off-by-one in replaceStrings
replaceStrings would previously fail to replace the last character
in a string.

Change-Id: I43a7c960945350b2e7a5b731b7fdb617723eb38f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9151
Tested-by: BuildkiteCI
Reviewed-by: sterni <sternenseemann@systemli.org>
2023-08-24 10:07:48 +00:00
Florian Klink
366e11b48a refactor(tvix/eval): don't use format! in write! args
Change-Id: I0b2fdb418c2e36280d5c551a81634e1742193903
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9105
Tested-by: BuildkiteCI
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
2023-08-20 21:52:21 +00:00
Florian Klink
c67b18bf45 refactor(tvix/eval/vm): don't put HashMap in a Box
HashMap already is on the heap.

Change-Id: I53763e17469359e85862f297b5c2e7c0d8c3a980
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9104
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
2023-08-20 21:52:21 +00:00
Florian Klink
aead2001d7 refactor(tvix/eval): impl Default for SourceCode
… instead of new(). Suggested by clippy.

Change-Id: Iac7be733392afefc2b4ff2e38386eee95f3bce94
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9103
Tested-by: BuildkiteCI
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
2023-08-20 21:52:21 +00:00
Florian Klink
cf1a884ec5 refactor(tvix/eval/observer): cargo clippy
Change-Id: I11d7d55f31ddd27952f201a0abbed7445013f434
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9102
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
2023-08-20 21:51:04 +00:00
Florian Klink
1a99b10b7e refactor(tvix/eval): cargo clippy
This passes a unit value to the function.

Change-Id: I4df3ad8fb0f35c0f110cee3349971ae28ce2878c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9101
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
2023-08-20 21:51:04 +00:00
Florian Klink
ded577a73b refactor(tvix/eval/io): cargo clippy &Path
Change-Id: Ifec2a4fc93c482e41ff5993183643b5126a7728b
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9100
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
2023-08-20 21:51:04 +00:00
Florian Klink
def773d4d5 refactor(tvix/eval): cargo clippy (len() is usize)
Change-Id: I9d931ffcc03c6df7c0392dbc1c9a4ae0e3804213
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9099
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
2023-08-20 21:51:04 +00:00
Florian Klink
1f02dc7eba refactor(tvix/eval): cargo clippy &GenCo
Change-Id: I6b1b902ccbc12bf2acdb0fdf406d6ef336ff0b2f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9098
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
Autosubmit: flokli <flokli@flokli.de>
2023-08-20 21:51:04 +00:00
Florian Klink
690884426b refactor(tvix/eval): derive default for value::AttrsRep enum
Change-Id: Iaeb9ed7024c2ce85373f8aec0d223f52e1a3a539
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9097
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
2023-08-20 21:51:04 +00:00
Vincent Ambo
8478448146 fix(tvix/eval): fix a comment position in value::json
Change-Id: Ic58653254b36694f1991a18db9ceea0d532c2e31
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9068
Autosubmit: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
2023-08-13 15:58:38 +00:00
sterni
8fb4c949dc test(tvix/eval): check truncation direction of builtins.div
builtins.div ought to truncate towards zero so that

    -(builtins.div a b) == builtins.div (-a) b
    -(builtins.div a b) == builtins.div a (-b)

Change-Id: I8b7c08cd7f4fa8a1363c786d42c8d484f6cd133d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/9006
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-08-11 22:22:30 +00:00
Florian Klink
432222f098 feat(tvix/store/proto): use Bytes instead of Vec<u8>
Makes use of https://github.com/tokio-rs/prost/pull/341, which makes our
bytes field cheaper to clone.

It's a bit annoying to configure due to
https://github.com/hyperium/tonic/issues/908, but the workaround does
get the job done.

Change-Id: I25714600b041bb5432d3adf5859b151e72b12778
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8975
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: flokli <flokli@flokli.de>
2023-07-22 09:03:02 +00:00
Florian Klink
72e82ffcb1 refactor(tvix/store): use bytes for node names and symlink targets
Some paths might use names that are not valid UTF-8. We should be able
to represent them.

We don't actually need to touch the PathInfo structures, as they need to
represent StorePaths, which come with their own harder restrictions,
which can't encode non-UTF8 data.

While this doesn't change any of the wire format of the gRPC messages,
it does however change the interface of tvix_eval::EvalIO - its
read_dir() method does now return a list of Vec<u8>, rather than
SmolStr. Maybe this should be OsString instead?

Change-Id: I821016d9a58ec441ee081b0b9f01c9240723af0b
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8974
Autosubmit: flokli <flokli@flokli.de>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Tested-by: BuildkiteCI
2023-07-21 19:01:49 +00:00
sterni
8adc9c56f2 docs(tvix): document when pointer equality is preserved in C++ Nix
This explicitly documents behavior of C++ Nix that goes against the
intuition you'd gather from this document: that e.g. a simple select
from an attribute set causes a value to no longer be pointer equal to
its former self.

The point of documenting this is that we can show in a to be written
section on the use of pointer equality in nixpkgs that pointer equality
is only needed in a limited sense for evaluating it (C++ Nix's exterior
pointer equality). Tvix's pointer equality is far more powerful since
value identity preserving operations also preserve pointer equality,
generally speaking (this is because we implement interior pointer
equality in my made up terminology). This should eventually also be
documented.

Change-Id: I6ce7ef2d67b012f5ebc92f9e81bba33fb9dce7d0
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8856
Tested-by: BuildkiteCI
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: tazjin <tazjin@tvl.su>
2023-07-11 16:17:42 +00:00
sterni
4ba624efae fix(tvix/eval): use byte, not codepoint index for slicing in escape
This fixes a subtle issue which would occasionally lead to a crash (e.g.
when evaluating (pkgs.systemd.outPath with --trace-runtime): With each
character in the string that has a multi byte representation in UTF-8,
the actual byte position and what tvix thought it was would get out of
sync. This could either lead to

* Tvix swallowing characters or jumbling characters if multi byte
  characters would cause the tracked index to become out of sync with
  the byte position before the first character to be escaped, or

* Tvix crashing if (in the same situation) the out of sync index would
  be within a UTF-8 byte sequence.

Luckily, std's `char_indices()` iterator implements exactly what
`nix_escape_char()`'s original author had in mind with
`.chars().enumerate()`. Using `i + 1` for continuing is safe, since all
characters that need (in fact, can) to be escaped in Nix are represented
as a single byte in UTF-8.

Change-Id: I1c836f70cde3d72db1c644e9112852f0d824715e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8952
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: sterni <sternenseemann@systemli.org>
2023-07-11 16:11:40 +00:00
Florian Klink
5796563f1f chore(tvix/eval/vm): drop unused import
Change-Id: Ia04778391c198fde21da217bf697aa40157898b0
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8846
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-06-30 12:42:59 +00:00
Evgeny Zemtsov
c8fcdca4eb feat(tvix/eval): allow extending builtins outside of tvix_eval
The change allows applications that use tvix_serde for parsing
nix-based configuration to extend the language with domain-specific
set of features.

Change-Id: Ia86612308a167c456ecf03e93fe0fbae55b876a6
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8848
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-06-22 20:19:06 +00:00
sterni
66047063e0 fix(tvix/eval): use realpaths for import cache
I've noticed this behavior when writing the admittedly cursed test case
included in this CL. Alternatively we could use some sort of machinery
using `builtins.trace`, but I don't think we capture stderr anywhere.

I've elected to put this into the eval cache itself while C++ Nix does
it in builtins.import already, namely via `realisePath`. We don't have
an equivalent for this yet, since we don't support any kind of IfD, but
we could revise that later. In any case, it seems good to encapsulate
`ImportCache` in this way, as it'll also allow using file hashes as
identifiers, for example.

C++ Nix also does our equivalent of canon_path in `builtins.import`
which we still don't, but I suspect it hardly makes a difference.

Change-Id: I05004737ca2458a4c67359d9e7d9a2f2154a0a0f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8839
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-06-21 07:48:52 +00:00
sterni
4516cd09c5 fix(tvix/eval): only finalise formal arguments if defaulting
When dealing with a formal argument in a function argument pattern that
has a default expression, there are two different things that can happen
at runtime: Either we select its value from the passed attribute
successfully or we need to use the default expression. Both of these may
be thunks and both of these may need finalisers. However, in the former
case this is taken care of elsewhere, the value will always be finalised
already if necessary. In the latter case we may need to finalise the
thunk resulting from the default expression. However, the thunk
corresponding to the expression may never end up in the local's stack
slot. Since finalisation goes by stack slot (and not constants), we need
to prevent a case where we don't fall back to the default expression,
but finalise anyways.

Previously, we worked around this by making `OpFinalise` ignore
non-thunks. Since finalisation of already evaluated thunks still
crashed, the faulty compilation of function pattern arguments could
still cause a crash.

As a new approach, we reinstate the old behavior of `OpFinalise` to
crash whenever encountering something that is either not a thunk or
doesn't need finalisation. This can also help catching (similar)
miscompilations in the future. To then prevent the crash, we need to
track whether we have fallen back or not at runtime. This is done using
an additional phantom on the stack that holds a new `FinaliseRequest`
value. When it comes to finalisation we check this value and
conditionally execute `OpFinalise` based on its value.

Resolves b/261 and b/265 (partially).

Change-Id: Ic04fb80ec671a2ba11fa645090769c335fb7f58b
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8705
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: sterni <sternenseemann@systemli.org>
2023-06-20 10:07:44 +00:00
sterni
77b0dddc3d chore(tvix/eval): fix markdown labeled link syntax
Change-Id: I639dc0801090eaba56b61858e28204b5a0e631b6
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8784
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: sterni <sternenseemann@systemli.org>
2023-06-15 19:40:08 +00:00
sterni
3b8c9ec9c8 test(tvix/eval): update nix_tests suite to C++ Nix master
Adds new tests for foldl', intersectAttrs as well as fills in missing
.exp files.

New test cases we don't pass:

- fromTOML timestamp capabilities
- path antiquotation
- replaceStrings is lazier on C++ Nix master

The C++ Nix revision used is 7066d21a0ddb421967980094222c4bc1f5a0f45a.

Change-Id: Ic619c96e2d41e6c5ea6fa93f9402b12e564af3c5
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8778
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
2023-06-15 19:28:16 +00:00
sterni
9278a0cd16 docs(tvix/eval): update test suite documentation
Change-Id: Ie9153c00b95ede4837a8eeab341e68bc90e97921
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8777
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-06-15 19:28:16 +00:00
sterni
0f71d8f813 test(tvix/eval): genericClosure (pointer) comparison support
genericClosure has very limited support for pointer equality: It relies
on comparison (not equality!) in C++ Nix, so as soon as C++ Nix supports
comparing lists (langVersion >= 6) we can rely on pointer equality for
key.

Since Tvix uses equality, not comparison for the insert, our behavior is
currently different, as documented by the notyetpassing tests.

Change-Id: Ifcd741ed4fc3ccc3825f7038875d56a9918b786a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8720
Tested-by: BuildkiteCI
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: tazjin <tazjin@tvl.su>
2023-06-15 11:01:06 +00:00
sterni
0005737f11 fix(tvix/eval): make tvix display values like nix-instantiate(1)
In order for the test suite we have currently to be comparable to C++
Nix, we need to display values in the same way. This was largely the
case except in some weird cases.

* <CODE> for thunks and <CYCLE> for repeated thunks (?) are already in
  use. <CODE> formatting is tested by the oracle test suite already.

* Instead of lambda, we need to use <LAMBDA>

* <<primop>> and <<primop-app>> (a formatting C++ Nix uses nowhere)
  now are <PRIMOP> and <PRIMOP-APP>.

We'll probably want to have a fancier display of values (in a separate
trait) down the line. This could be used for interactive usage, e.g. the
REPL or a potential debugger.

There is a peculiarity with C++ Nix 2.3 formatting primops: import is
considered a <<PRIMOP-APP>>, since it is internally implemented by means
of scopedImport. This implementation detail no longer leaks in C++ Nix
2.13 nor in Tvix.

<CYCLE> display is untested at the moment, since we exhibit a
discrepancy to C++ Nix 2.3. Our current detection is more similar to C++
Nix 2.13—luckily it is also the more consistent of the two. See also
b/245.

Change-Id: I1d534434b02e470bf5475b3758920ea81e3420dc
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8760
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
2023-06-15 11:01:06 +00:00
sterni
e409d9cc7e docs(tvix/eval): fix link to tvixbolt
Change-Id: Iff3f74ab6d5177246811bd3d58d171088915370f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8775
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-06-14 16:22:07 +00:00
sterni
80a71abb48 test(tvix/eval): move division by zero tests into tvix_tests
These were added by us in r/5276, so they should go into our test suite.

Change-Id: I6dc74fc242f33c22a17e0b4aee546ccae886ac85
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8774
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: sterni <sternenseemann@systemli.org>
2023-06-14 16:22:07 +00:00
sterni
1125b6b7b7 test(tvix/eval): add test case for builtins set pointer equality
Unsupported by Tvix at the moment. Documents b/280.

Change-Id: I48844feeefa9da8ed7e5d85300d52bb5650f82d2
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8772
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: sterni <sternenseemann@systemli.org>
2023-06-14 16:01:19 +00:00
sterni
d5b989ddc0 test(tvix/eval): re-enable blackhole test
Change-Id: Id933f3bd708aa3342b9fd6a5584e65ee11751ff8
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8773
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: sterni <sternenseemann@systemli.org>
2023-06-14 15:59:45 +00:00
sterni
8e5551329a fix(tvix/eval): don't thunk home relative paths
C++ Nix resolves home relative paths at [parse] time. This is not an
option for us, since it prevents being able to separate the compilation
and execution phase later (e.g. precompiled nix expressions). However, a
practical consequence of this is that paths expressions are always
literals (strict) and never thunks.

[parse]: 7066d21a0d/src/libexpr/parser.y (L518-L527)

Change-Id: Ie4b9dc68f62c86d6c7fd5f1c9460c850d97ed1ca
Reviewed-on: https://cl.tvl.fyi/c/depot/+/7041
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-06-14 10:57:28 +00:00
sterni
5139cc45c2 test(tvix/eval): builtins.substring's behavior with negative args
Change-Id: Ie8b97d174e9d58e33bf08c9b9e0afeeddd089ba8
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8700
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-06-12 09:38:08 +00:00
Linus Heckemann
5733274876 fix(tvix/eval): allow negative substring lengths
Nix uses string::substr without checking the sign of the length[1].
The NixOS testing infrastructure relies on this[2], and on the
implicit conversion of that to the maximum possible value for a
size_t.

[1]: ecae62020b/src/libexpr/primops.cc (L3597)
[2]: c7c2984716/nixos/lib/testing/driver.nix (L29)

Change-Id: I6d0caf6830b6bda3fdf44c40c81de6a1befeca7b
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8746
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-06-12 08:38:49 +00:00
Linus Heckemann
2b4ad47c16 fix(tvix/eval): emit only warnings on shadowed outputs
Unfortunately, nixpkgs has at least one case[1] where the out environment
variable is shadowed -- though it doesn't cause a problem, since it's
shadowed with the correct value, odd as this may be!

[1]: c7c2984716/pkgs/development/python-modules/pybind11/default.nix (L19)

Change-Id: Ibf6790d2484dc9cce8e424feeb5886664d498dc3
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8696
Autosubmit: tazjin <tazjin@tvl.su>
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-06-11 20:59:34 +00:00
sterni
a2303da01f fix(tvix/eval): use normal thunking behavior for default in formals
When comparing to C++ Nix, we notice that the thunking of default
expressions in function formals corresponds to their normal thunking,
e.g. literals are not thunked. This means that we can just invoke
compile() without much of a care and trust that it will sort it out
correctly.

If function formals blow up as a result of this, it likely indicates
that the expression is treated incorrectly by compile(), not
compile_param_pattern().

Change-Id: I64acbff2f251423eb72ce43e56a0603379305e1d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8704
Autosubmit: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-06-07 15:19:21 +00:00
sterni
10c6cb7251 fix(tvix/eval): type check function argument with set pattern
C++ Nix forces and typechecks the passed argument even if it is not
necessary in order to compute the return value of the function. I
discovered this when I thought our formals miscompilation might be that
we are too strict, but doesn't look like it in this case.

Change-Id: Ifb3c92592293052c489d1e3ae8c7c54e4b6b4dc6
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8701
Tested-by: BuildkiteCI
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: tazjin <tazjin@tvl.su>
2023-06-07 15:17:20 +00:00
sterni
617130b088 refactor(tvix/eval): don't track idx twice in compile_param_pattern
Change-Id: I27f9105ddb20d84342550b2a73b479a7764ee3fe
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8699
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
2023-06-07 15:17:19 +00:00
sterni
b492067c19 test(tvix/eval): check thunking behavior of basic exprs
nix_oracle.rs now gives us the possibility to check this by stuffing the
expressions in a list. In fact, the incorrect behavior fixed in

- cl/8656
- cl/8655
- cl/8662

was discovered using this test suite.

Change-Id: Id0ab01ee6be0b875791214e0a72a2ac941c46c96
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8658
Autosubmit: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-05-29 19:36:42 +00:00
sterni
0ab6494286 refactor(tvix/eval/nix_oracle): allow specifying eval strictness
This will be useful for comparing thunking behavior to C++ Nix. I
considered adding this capability to the tvix_tests/nix_tests
infrastructure, but as it would require changing the test file naming
scheme to do it in a clean way, I've postponed it–it's nice that our
tests are compatible with C++ Nix's test suite.

Change-Id: I60bcdd98ed25140e716f0858f8dc28f21ab957aa
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8657
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-05-29 19:36:41 +00:00
sterni
d09f333d0e fix(tvix/eval): thunk lambda expressions
As cl/8658 and b/274 reveal, lambda expressions are also wrapped in
thunks.

Resolves b/274.

Change-Id: I02fe5c8730ac76748d940e4f4427116587875275
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8662
Autosubmit: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-05-29 19:36:41 +00:00
sterni
2aab01ac29 fix(tvix/eval): thunk HasAttr expressions
HasAttrs was weird because with longer attribute paths it would
sometimes not turn out to be a thunk. If it was a thunk, it'd usually
still do some eval strictly which we'll want to avoid.

Verified against C++ Nix using a new test suite introduced in a later
CL.

Change-Id: I6d047ccc68d046bb268462f170a3c4f3c5ddeffe
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8656
Autosubmit: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-05-29 12:44:09 +00:00
sterni
9d0425acc0 fix(tvix/eval): thunk legacy let to match regular one
Probably no real world code broken by this overzealous evaluation, but
let's be thorough!

Change-Id: Ib405a677182eab7940ace940c68e107573473a54
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8655
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
2023-05-29 12:44:08 +00:00
sterni
385c797884 fix(tvix/eval): thunk unary operator applications
Unary operator applications are thunked which can easily be observed by

  nix-instantiate --eval -E '[ (!true) (-1) ]'

Unfortunately, there are few simple expressions where this makes a
difference in the end result. Thus it only cropped up when using nixpkgs
for cross compilation: Here we would compile the expression

  !(stdenv.cc.isGNU or false)

to assemble python3Minimal's passthru attribute set (at least this seems
to be the most likely explanation from the backtraces I've studied).
This means that an unthunked

    <stdenv.cc.isGNU or false>
    OpForce
    OpInvert

would be performed in order to assemble this attribute set, causing
stdenv.cc to be evaluated too early, causing an infinite recursion.

Resolves b/273.

It seems that having a test suite that doesn't use --strict and relies
on thunks rendered as <CODE> would be beneficial for catching such
issues. I've not been able to find a test case with --strict that
demonstrates the problem fixed in this CL.

Change-Id: I640a5827b963f5b9d0f86fa2142e75e3a6bbee78
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8654
Tested-by: BuildkiteCI
Autosubmit: sterni <sternenseemann@systemli.org>
Reviewed-by: tazjin <tazjin@tvl.su>
2023-05-29 12:44:08 +00:00
sterni
3b33c19a9c fix(tvix): don't call function eagerly in genList, map & mapAttrs
mapAttrs, map and genList call Nix functions provided by the caller and
store the result of applying them in a Nix data structure that does not
force all of its contents when forced itself. This means that when such
a builtin application is forced, the Nix function calls performed by the
builtin should not be forced: They may be forced later, but it is also
possible that they will never be forced, e.g. in

    builtins.length (builtins.map (builtins.add 2) [ 1 2 3 ])

it is not necessary to compute a single application of builtins.add.

Since request_call_with immediately performs the function call
requested, Tvix would compute function applications unnecessarily before
this change. Because this was not followed by a request_force, the
impact of this was relatively low in Nix code (most functions return a
new thunk after being applied), but it was enough to cause a lot of
bogus builtins.trace applications when evaluating anything from
`lib.modules`. The newly added test includes many cases where Tvix
previously incorrectly applied a builtin, breaking a working expression.

To fix this we add a new helper to construct a Thunk performing a
function application at runtime from a function and argument given as
`Value`s. This mimics the compiler's compile_apply(), but does itself
not require a compiler, since the necessary Lambda can be constructed
independently.

I also looked into other builtins that call a Nix function to verify
that they don't exhibit such a problem:

- Many builtins immediately use the resulting value in a way that makes
  it necessary to compute all the function calls they do as soon as
  the outer builtin application is forced:

  * all
  * any
  * filter
  * groupBy
  * partition

- concatMap needs to (shallowly) force the returned list for
  concatenation.

- foldl' is strict in the application of `op` (I added a comment that
  makes this explicit).

- genericClosure needs to (shallowly) force the resulting list and some
  keys of the attribute sets inside.

Resolves b/272.

Change-Id: I1fa53f744bcffc035da84c1f97ed25d146830446
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8651
Autosubmit: sterni <sternenseemann@systemli.org>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-05-26 22:35:39 +00:00
Vincent Ambo
22776280b5 feat(tvix/eval): unthunk empty lists and attribute sets
Change-Id: Ie66cb1b163a544d45d113fd0f866286f230b0188
Reviewed-on: https://cl.tvl.fyi/c/depot/+/7960
Tested-by: BuildkiteCI
Reviewed-by: grfn <grfn@gws.fyi>
2023-05-25 11:28:13 +00:00
Vincent Ambo
5d3fb33baa feat(tvix/eval): implement unthunking in compiler
This feature allows the compiler to detect situations where the
created thunk is useless and can be merged into the parent chunk
instead.

The only case where the compiler does this initially is when
statically optimising a select expression.

For example, previously the expression `builtins.length` compiled into
two thunks:

1. An "inner" thunk which contained an `OpConstant` that had the
   optimised `length` builtin in it.

2. An "outer" thunk which contained an `OpConstant` to access the
   inner thunk, and the trailing OpForce of the top-level program.

With this change, the inner thunk is skipped completely and the outer
chunk directly contains the `length` builtin access.

This can be applied in several situations, some easier than others,
and we will add them in as we go along.

Change-Id: Ie44521445fce1199f99b5b17712833faea9bc357
Reviewed-on: https://cl.tvl.fyi/c/depot/+/7959
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
2023-05-25 11:28:13 +00:00
Vincent Ambo
3aea9bf527 feat(tvix/eval): implement Chunk::extend method
This method extends the contents of one chunk with that of another,
effectively merging the thunks together.

This will be used for the upcoming "unthunking" functionality.

Change-Id: I6ad74232cd7f3eca198ed921e455205e00d76e6b
Reviewed-on: https://cl.tvl.fyi/c/depot/+/7958
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-05-25 11:28:13 +00:00
Florian Klink
d25962b9a4 refactor(tvix/eval): stop borrowing &mut self
This does undo cl/8571.

Change-Id: Ib14b4e7404f906e346304b6113860ae811afc94a
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8631
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: flokli <flokli@flokli.de>
2023-05-25 11:11:59 +00:00
Florian Klink
11771a06ae refactor(tvix/eval): use &Path instead of PathBuf
This allows getting rid of some clones in eval/src/vm/generators.rs.

Change-Id: I330390307d3bcfeef19c98954c753ee55b1ccee3
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8604
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-05-22 09:43:33 +00:00
Florian Klink
b4bb9062ea fix(tvix/eval): add path where useful to ErrorKind::IO
These two places didn't add the path from the context to the
ErrorKind, but simply relied on the
impl From<std::io::Error> for tvix_eval::ErrorKind, which doesn't add
the path.

Change-Id: Ifc7dbbe305d24242b0705de1dea34e8923e9d2cb
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8603
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
Autosubmit: flokli <flokli@flokli.de>
2023-05-22 09:43:21 +00:00
Florian Klink
b7ab6c0856 refactor(tvix/eval/io): use io::Error instead of tvix_eval errors
We didn't return anything useful other than ErrorKind::IO anyways.

We can use io::ErrorKind::Unsupported for DummyIO.

Fixes b/271.

Change-Id: Icb231e9b38168e8b6fa473bfa405d160357b317f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8602
Autosubmit: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-05-22 09:42:58 +00:00
Florian Klink
8bd7ced1fb feat(tvix/eval/io): allow &mut self in EvalIO
It's okay if these calls mutate some internal state inside an
implementation.

Change-Id: I12bb11bde0310778c3da1275696bf7de058863a3
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8571
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>
2023-05-14 18:59:55 +00:00
Vincent Ambo
969fbcd6d4 fix(tvix/eval): builtins.trace prints to stderr
Change-Id: Icf577396035474d6977e627058aba5805c61985e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8563
Tested-by: BuildkiteCI
Autosubmit: tazjin <tazjin@tvl.su>
Reviewed-by: flokli <flokli@flokli.de>
2023-05-12 12:50:59 +00:00
Florian Klink
6a30eb69e0 chore(tvix/*): bump to smol_str 0.2.0
Change-Id: Ic9ac1b6fecb564eafb41b265bf317cd385fdc170
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8560
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: flokli <flokli@flokli.de>
2023-05-11 18:19:20 +00:00
Vincent Ambo
ce502bdc89 refactor(tvix/eval): improve representation of chunk/span mapping
This switches out the previous compressed representation (count of
instructions per span) with a representation where the chunk's span
list stores the index of the first operation that belongs to a span,
and finds the right span by using a binary search when looking them
up.

This improves the lookup complexity from O(n) to O(log n).

This improvement was suggested and (mostly) implemented by GPT-4. I
only fixed up some names and updated the logic for deleting
spans (which it only did not do because I didn't tell it about that).

The code was verified by producing a complex error before/after the
change and ensuring that all spans in the error match exactly.

Co-Authored-By: GPT-4
Change-Id: Ibfa12cc6973af1c9b0ae55bb464d1975209771f5
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8385
Reviewed-by: ezemtsov <eugene.zemtsov@gmail.com>
Tested-by: BuildkiteCI
Autosubmit: tazjin <tazjin@tvl.su>
2023-03-31 16:46:19 +00:00
Vincent Ambo
2d305fd5b3 refactor(tvix/eval): retain call frames when entering calls
This grows the frame stack as the call stack grows, which yields *much*
better user-facing error messages.

I haven't measured the performance impact this has yet, for now I'm
still just trying to add more information to errors and then cut down
again where necessary.

Change-Id: I89f058ef31979edacf4667775d460b60704ce4d7
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8334
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Autosubmit: tazjin <tazjin@tvl.su>
2023-03-27 09:02:43 +00:00
Vincent Ambo
ba138712e4 feat(tvix/eval): add Evaluation::strict to toggle top-level deepseq
This makes it possible for callers to control whether they can receive
partially evaluated values from an evaluation or not.

We're actually flipping the default behaviour to non-strict top-level
evaluation, which means that callers have to set `strict = true` on
the Evaluation to get the previous behaviour.

Change-Id: Ic048e9ba09c88866d4c3177d5fa07db11c4eb20e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8325
Autosubmit: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Reviewed-by: sterni <sternenseemann@systemli.org>
2023-03-22 13:44:20 +00:00
Vincent Ambo
a5f28eea94 fix(tvix/eval): print unevaluated thunks like Nix does
Change-Id: Ie4c563e933f571f45cb4f4efe650d1b65f119e8d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8324
Tested-by: BuildkiteCI
Reviewed-by: sterni <sternenseemann@systemli.org>
Autosubmit: tazjin <tazjin@tvl.su>
2023-03-22 13:44:19 +00:00
Vincent Ambo
80aaadfc19 fix(tvix/eval): use span of set for OpForce in attribute access
Emits the span of the `set` that is being accessed in the `force`
operation of an attribute access.

Looking at traces, it's a lot more useful to get information about
*what* is being forced, as in cases like `foo.bar` it can be
misleading to have an error highlight `bar`, when the error occured
while forcing `foo` to be able to access `bar` in the first place.

Change-Id: Id46ff28f20c67cb4971727ac52cc4811795cea2d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8272
Reviewed-by: flokli <flokli@flokli.de>
Autosubmit: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-03-17 19:40:40 +00:00
Vincent Ambo
8c13f18d11 feat(tvix/eval): report all known spans on infinite recursion
This reports the span

1. of the code within a thunk,
2. of the place where the thunk was instantiated,
3. of the place where the thunk was first forced,
4. of the place where the thunk was forced again,

when yielding an infinite recursion error, which hopefully makes it
easier to debug them.

The spans are tracked in the ThunkRepr::Blackhole variant when putting
a thunk under evaluation.

Note that we currently have some loss of span precision in the VM loop
when switching between frame types, so spans 3/4 are currently a bit
wonky. Working on it.

Change-Id: Icbd2a9df903d00e8c2545b3fc46dcd2a9e3e3e55
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8270
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Autosubmit: tazjin <tazjin@tvl.su>
2023-03-17 19:32:38 +00:00
Vincent Ambo
3fa6b13c1e feat(tvix/eval): track span of first force in a thunk blackhole
This is step 1 towards being able to use all 4 spans that we know when
dealing with infinite recursion. It tracks the span at which the
force of a thunk was first requested when constructing a blackhole, so
that we can highlight the spans of the first and second forces.

These are actually the least relevant spans, but the easiest to put in
place, more coming soon.

Change-Id: I4c7e82f6211b98756439d4148a4191457cc46807
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8269
Autosubmit: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Reviewed-by: flokli <flokli@flokli.de>
2023-03-17 19:31:37 +00:00
Vincent Ambo
5095e4f269 feat(tvix/eval): add generator "name" to NativeError kind
This produces traces in which we can see what kind of native code was
run. Note that these "names" are named after the generator message, so
these aren't *really* intended for end-user consumption, but we can
give them saner names later.

Example:
https://gist.github.com/tazjin/82b24e92ace8e821008954867ee05057

This already makes the traces a little easier to parse.

Change-Id: Idcd601baf84f492211b732ea0f04b377112e10d0
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8268
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Autosubmit: tazjin <tazjin@tvl.su>
2023-03-17 19:31:37 +00:00
Vincent Ambo
ea80e0d3f8 feat(tvix/eval): enrich errors with VM's frame stack information
When emitting an error at runtime, the VM will now use the new
`NativeError` and `BytecodeError` error kinds (which just wrap inner
errors) to create a set of diagnostics to emit.

The primary diagnostic is emitted last, with `error` type (so it will
be coloured red in terminals), the other ones will be emitted with
`note` type, highlighting the causal chain.

Example:
https://gist.github.com/tazjin/25feba7d211702453c9ebd5f8fd378e4

This is currently quite verbose, and we can cut down on this further,
but the purpose of this commit is to surface more information first of
all before worrying about the exact display.

Change-Id: I058104a178c37031c0db6b4b3e4f4170cf76087d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8266
Autosubmit: tazjin <tazjin@tvl.su>
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-03-17 19:31:37 +00:00
Vincent Ambo
b78ae941a4 fix(tvix/eval): use coerce_to_string in builtins.substring
This actually uses coercion under the hood in C++ Nix. See the test
for an example.

Change-Id: Id56b364acf269225b6829d0b600e0222f8b3608d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8322
Reviewed-by: andi <andi@notmuch.email>
Tested-by: BuildkiteCI
2023-03-17 19:17:29 +00:00
Vincent Ambo
b5f5a1595d chore(tvix/eval): remove some dead code
This was commented out and forgotten during the generator refactor, oh
well.

Change-Id: I474b685159a955a846db462da0dd0067af177b04
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8321
Autosubmit: tazjin <tazjin@tvl.su>
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-03-17 11:01:07 +00:00
Adam Joseph
32999cb6f6 docs(tvix/eval): suggested layout adjustment to VM loop diagram
Change-Id: I5467cd66801ad8fe6c4ec0ae337763f1762cea1c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8252
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-03-14 09:56:28 +00:00
Vincent Ambo
466e6dc265 docs(tvix/eval): document inner workings of the new VM loop
Change-Id: I8224bf039f739c401900b5a2ddc839810c87cf6e
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8226
Tested-by: BuildkiteCI
Reviewed-by: Adam Joseph <adam@westernsemico.com>
2023-03-14 09:22:32 +00:00
Vincent Ambo
d456705352 chore(tvix): Generator{Request|Response} -> VM{Request|Response}
We settled on this being the most reasonable name for this construct.

Change-Id: Ic31c45461a842f22aa05f4446123fe3a61dfdbc0
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8291
Tested-by: BuildkiteCI
Autosubmit: tazjin <tazjin@tvl.su>
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
2023-03-14 09:22:22 +00:00
Adam Joseph
1e80b9ea8b chore(tvix/eval): mark async functions which are called by the VM
Given Rust's current lack of support for tail calls, we cannot avoid
using `async` for builtins.  This is the only way to avoid
overflowing the cpu stack when we have arbitrarily deep
builtin/interpreted/builtin/interpreted/... "sandwiches"

There are only five `async fn` functions which are not builtins
(some come in multiple "flavors"):

- add_values
- resolve_with
- force, final_deep_force
- nix_eq, nix_cmp_eq
- coerce_to_string

These can be written iteratively rather than recursively (and in
fact nix_eq used to be written that way!).  I volunteer to rewrite
them.  If written iteratively they would no longer need to be
`async`.

There are two motivations for limiting our reliance on `async` to
only the situation (builtins) where we have no other choice:

1. Performance.

   We don't really have any good measurement of the performance hit
   that the Box<dyn Future>s impose on us.  Right now all of our
   large (nixpkgs-eval) tests are swamped by the cost of other
   things (e.g. fork()ing `nix-store`) so we can't really measure
   it.  Builtins tend to be expensive operations anyways
   (regexp-matching, sorting, etc) that are likely to already cost
   more than the `async` overhead.

2. Preserving the ability to switch to `musttail` calls.

   Clang/LLVM recently got `musttail` (mandatory-elimination tail
   calls).  Rust has refused to add this mainly because WASM doesn't
   support, but WASM `tail_call` has been implemented and was
   recently moved to phase 4 (standardization).  It is very likely
   that Rust will get tail calls sometime in the next year; if it
   does, we won't need async anymore.  In the meantime, I'd like to
   avoid adding any further reliance on `async` in places where it
   wouldn't be straightforward to replace it with a tail call.

https://reviews.llvm.org/D99517

https://github.com/WebAssembly/proposals/pull/157

https: //github.com/rust-lang/rfcs/issues/2691#issuecomment-1462152908
Change-Id: Id15945d5a92bf52c16d93456e3437f91d93bdc57
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8290
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-03-13 21:33:58 +00:00
Adam Joseph
e7a534e0c6 refactor(tvix/eval): reduce fetch{forced|captured}_with visibility
This commit moves fetch_forced_with and fetch_captured_with into the
scope of their only caller (resolve_with).

Change-Id: I9a8bc27228888729d591e8cb021c431b2b6468f5
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8289
Autosubmit: Adam Joseph <adam@westernsemico.com>
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
2023-03-13 21:33:58 +00:00
Adam Joseph
47895c4c30 feat(tvix/eval): rewrite nix_cmp_ordering to be nonrecursive
This rewrites nix_cmp_ordering as an iterative loop, which
eliminates the extra pinned-boxing helper function.

Change-Id: I33d0ecc913e02affd8fd4c7bc1c9ecfdf4c7deb9
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8288
Reviewed-by: tazjin <tazjin@tvl.su>
Tested-by: BuildkiteCI
Autosubmit: Adam Joseph <adam@westernsemico.com>
2023-03-13 21:22:12 +00:00
Vincent Ambo
b3f8d66a6a chore(tvix/eval): prune some dependencies & features
* We no longer need backtrace-on-stack-overflow, as we no longer
  overflow the stack with the recent eval refactorings. This was weird
  voodoo anyways, introduced earlier to debug some cases where stack
  overflows occured.

* default features of genawaiter crate are not needed, as we don't use
  their proc macros

Change-Id: I346fc5a18d7f117ee805909a8be8f535b96be76c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8263
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
2023-03-13 20:30:59 +00:00
Vincent Ambo
94513525b9 refactor(tvix/eval): reorder bytecode operations match by frequency
This reorders the operations in the VM's main `match` statement while
evaluating bytecode according to the frequency with which these
operations appear in some nixpkgs evaluations.

I used raw data that looks like this:
https://gist.github.com/tazjin/63d0788a78eb8575b04defaad4ef610d

This has a small but noticeable impact on evaluation performance.

No operations have changed in any way, this is purely moving code
around.

Change-Id: Iaa4ef4f0577e98144e8905fec88149c41e8c315c
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8262
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
Reviewed-by: flokli <flokli@flokli.de>
Tested-by: BuildkiteCI
2023-03-13 20:30:59 +00:00
Vincent Ambo
e48a17a6b3 docs(tvix/eval): fix reference to Empty message in a comment
Change-Id: I3dc30cca33fbbd8e8686655635ee471f5937d9f8
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8257
Tested-by: BuildkiteCI
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
2023-03-13 20:30:59 +00:00
Vincent Ambo
cc59cbf3e2 refactor(tvix/eval): rename VM::tail_call_value -> VM::call_value
The name of this was not accurate anymore after all the recent
shuffling, as noted by amjoseph. Conceptual tail calls here only occur
for Nix bytecode calling Nix bytecode, but things like a builtin call
actually push a new native frame.

Change-Id: I1dea8c9663daf86482b8c7b5a23133254b5ca321
Reviewed-on: https://cl.tvl.fyi/c/depot/+/8256
Tested-by: BuildkiteCI
Reviewed-by: raitobezarius <tvl@lahfa.xyz>
2023-03-13 20:30:59 +00:00