Commit Graph

138 Commits

Author SHA1 Message Date
Oli Scherer
e9a2f8fef1 Remove feed_local_crate in favor of creating the CrateNum via TyCtxt 2024-04-19 14:16:18 +00:00
Oli Scherer
fbc9b94064 Move stable_crate_ids from CrateStore to Untracked
This way it's like `Definitions`, which creates `DefId`s by interning `DefPathData`s, but for interning stable crate hashes
2024-04-19 14:16:18 +00:00
John Kåre Alsaker
12cd322062 Make incremental sessions identity no longer depend on the crate names provided by source code 2024-03-13 16:40:02 +01:00
John Kåre Alsaker
55ba7a7c62 Verify that query keys result in unique dep nodes 2024-03-12 05:31:41 +01:00
Guillaume Gomez
4de78d2a8d
Rollup merge of #121089 - oli-obk:create_def_feed, r=petrochenkov
Remove `feed_local_def_id`

best reviewed commit by commit

Basically I returned `TyCtxtFeed` from `create_def` and then preserved that in the local caches

based on https://github.com/rust-lang/rust/pull/121084

r? ````@petrochenkov````
2024-03-07 18:32:47 +01:00
Oli Scherer
3845be6b37 Prevent feeding CRATE_DEF_ID queries outside the resolver 2024-03-05 05:53:33 +00:00
Oli Scherer
5a0c46a22c Get rid of feed_local_def_id 2024-03-05 05:52:00 +00:00
Nicholas Nethercote
80d2bdb619 Rename all ParseSess variables/fields/lifetimes as psess.
Existing names for values of this type are `sess`, `parse_sess`,
`parse_session`, and `ps`. `sess` is particularly annoying because
that's also used for `Session` values, which are often co-located, and
it can be difficult to know which type a value named `sess` refers to.
(That annoyance is the main motivation for this change.) `psess` is nice
and short, which is good for a name used this much.

The commit also renames some `parse_sess_created` values as
`psess_created`.
2024-03-05 08:11:45 +11:00
Nicholas Nethercote
c2512a130f Inline and remove Session::compile_status.
Because it's now simple enough that it doesn't provide much benefit.
2024-02-22 08:03:47 +11:00
Nicholas Nethercote
72b172bdf6 Overhaul the handling of errors at the top-level.
Currently `emit_stashed_diagnostic` is called from four(!) different
places: `print_error_count`, `DiagCtxtInner::drop`, `abort_if_errors`,
and `compile_status`.

And `flush_delayed` is called from two different places:
`DiagCtxtInner::drop` and `Queries`.

This is pretty gross! Each one should really be called from a single
place, but there's a bunch of entanglements. This commit cleans up this
mess.

Specifically, it:
- Removes all the existing calls to `emit_stashed_diagnostic`, and adds
  a single new call in `finish_diagnostics`.
- Removes the early `flush_delayed` call in `codegen_and_build_linker`,
  replacing it with a simple early return if delayed bugs are present.
- Changes `DiagCtxtInner::drop` and `DiagCtxtInner::flush_delayed` so
  they both assert that the stashed diagnostics are empty (i.e.
  processed beforehand).
- Changes `interface::run_compiler` so that any errors emitted during
  `finish_diagnostics` (i.e. late-emitted stashed diagnostics) are
  counted and cannot be overlooked. This requires adding
  `ErrorGuaranteed` return values to several functions.
- Removes the `stashed_err_count` call in `analysis`. This is possible
  now that we don't have to worry about calling `flush_delayed` early
  from `codegen_and_build_linker` when stashed diagnostics are pending.
- Changes the `span_bug` case in `handle_tuple_field_pattern_match` to a
  `delayed_span_bug`, because it now can be reached due to the removal
  of the `stashed_err_count` call in `analysis`.
- Slightly changes the expected output of three tests. If no errors are
  emitted but there are delayed bugs, the error count is no longer
  printed. This is because delayed bugs are now always printed after the
  error count is printed (or not printed, if the error count is zero).

There is a lot going on in this commit. It's hard to break into smaller
pieces because the existing code is very tangled. It took me a long time
and a lot of effort to understand how the different pieces interact, and
I think the new code is a lot simpler and easier to understand.
2024-02-22 08:03:47 +11:00
Nicholas Nethercote
ede99234c4 Make CodegenBackend::join_codegen infallible.
Because they all are, in practice.
2024-02-17 10:51:35 +11:00
Nicholas Nethercote
d4b77f64e4 Tweak delayed bug mentions.
Now that we have both `delayed_bug` and `span_delayed_bug`, it makes
sense to use the generic term "delayed bug" more.
2024-02-12 18:39:20 +11:00
Nicholas Nethercote
3c4f1d85af Rename {create,emit}_warning as {create,emit}_warn.
For consistency with `warn`/`struct_warn`, and also `{create,emit}_err`,
all of which use an abbreviated form.
2024-01-10 07:33:06 +11:00
Nicholas Nethercote
b1b9278851 Make DiagnosticBuilder::emit consuming.
This works for most of its call sites. This is nice, because `emit` very
much makes sense as a consuming operation -- indeed,
`DiagnosticBuilderState` exists to ensure no diagnostic is emitted
twice, but it uses runtime checks.

For the small number of call sites where a consuming emit doesn't work,
the commit adds `DiagnosticBuilder::emit_without_consuming`. (This will
be removed in subsequent commits.)

Likewise, `emit_unless` becomes consuming. And `delay_as_bug` becomes
consuming, while `delay_as_bug_without_consuming` is added (which will
also be removed in subsequent commits.)

All this requires significant changes to `DiagnosticBuilder`'s chaining
methods. Currently `DiagnosticBuilder` method chaining uses a
non-consuming `&mut self -> &mut Self` style, which allows chaining to
be used when the chain ends in `emit()`, like so:
```
    struct_err(msg).span(span).emit();
```
But it doesn't work when producing a `DiagnosticBuilder` value,
requiring this:
```
    let mut err = self.struct_err(msg);
    err.span(span);
    err
```
This style of chaining won't work with consuming `emit` though. For
that, we need to use to a `self -> Self` style. That also would allow
`DiagnosticBuilder` production to be chained, e.g.:
```
    self.struct_err(msg).span(span)
```
However, removing the `&mut self -> &mut Self` style would require that
individual modifications of a `DiagnosticBuilder` go from this:
```
    err.span(span);
```
to this:
```
    err = err.span(span);
```
There are *many* such places. I have a high tolerance for tedious
refactorings, but even I gave up after a long time trying to convert
them all.

Instead, this commit has it both ways: the existing `&mut self -> Self`
chaining methods are kept, and new `self -> Self` chaining methods are
added, all of which have a `_mv` suffix (short for "move"). Changes to
the existing `forward!` macro lets this happen with very little
additional boilerplate code. I chose to add the suffix to the new
chaining methods rather than the existing ones, because the number of
changes required is much smaller that way.

This doubled chainging is a bit clumsy, but I think it is worthwhile
because it allows a *lot* of good things to subsequently happen. In this
commit, there are many `mut` qualifiers removed in places where
diagnostics are emitted without being modified. In subsequent commits:
- chaining can be used more, making the code more concise;
- more use of chaining also permits the removal of redundant diagnostic
  APIs like `struct_err_with_code`, which can be replaced easily with
  `struct_err` + `code_mv`;
- `emit_without_diagnostic` can be removed, which simplifies a lot of
  machinery, removing the need for `DiagnosticBuilderState`.
2024-01-08 15:24:49 +11:00
Ben Kimock
94c43ccd87 Report I/O errors with emit_fatal not emit_err 2024-01-02 07:31:02 -05:00
Nicholas Nethercote
99472c7049 Remove Session methods that duplicate DiagCtxt methods.
Also add some `dcx` methods to types that wrap `TyCtxt`, for easier
access.
2023-12-24 08:05:28 +11:00
Nicholas Nethercote
09af8a667c Rename Session::span_diagnostic as Session::dcx. 2023-12-18 16:06:21 +11:00
Nicholas Nethercote
5d1d384443 Rename HandlerInner::delay_span_bug as HandlerInner::span_delayed_bug.
Because the corresponding `Level` is `DelayedBug` and `span_delayed_bug`
follows the pattern used everywhere else: `span_err`, `span_warning`,
etc.
2023-12-02 09:01:19 +11:00
Vadim Petrochenkov
f0dc906319 resolve: Feed the def_kind query immediately on DefId creation 2023-11-28 15:39:31 +03:00
bjorn3
d7e9a30941 Turn write_dep_info into a regular function
It has side-effects and as such can't be cached.
2023-11-26 18:02:47 +00:00
bjorn3
365a580bc4 Mostly revert "Accept crate name instead of attributes in build_output_filenames" 2023-11-26 18:02:47 +00:00
bjorn3
4acaa0284e Feed the output filenames into the TyCtxt
Since the introduction of the crate attribute pre-expansion pass we
don't need access to the TyCtxt to compute it.
2023-11-26 18:02:47 +00:00
bjorn3
98a6eaa7f8 Serialize OutputFilenames into rmeta file
This ensures that linking will use the correct crate name even when
`#![crate_name = "..."]` is used to specify the crate name.
2023-11-26 18:02:42 +00:00
bjorn3
457dbbfc55 Inline and remove pre_configure 2023-11-26 17:43:49 +00:00
Ben Kimock
fbaa24ee35 Call FileEncoder::finish in rmeta encoding 2023-11-22 22:49:22 -05:00
Nicholas Nethercote
971010ea5a Merge Queries::{ongoing_codegen,linker}.
There is no real need for them to be separate.
2023-11-22 13:22:49 +11:00
Nicholas Nethercote
3a4798c92d Make Compiler::{sess,codegen_backend} public.
And remove the relevant getters on `Compiler` and `Queries`.
2023-11-22 13:22:41 +11:00
Nicholas Nethercote
cf561904db Add comments about a timer. 2023-11-22 13:20:56 +11:00
Nilstrieb
21a870515b Fix clippy::needless_borrow in the compiler
`x clippy compiler -Aclippy::all -Wclippy::needless_borrow --fix`.

Then I had to remove a few unnecessary parens and muts that were exposed
now.
2023-11-21 20:13:40 +01:00
Nicholas Nethercote
9582172964 Make Compiler::sess private.
Like `Compiler::codegen_backend`.
2023-11-17 17:30:41 +11:00
Nicholas Nethercote
4a57b80f3f Remove a low-value local variable. 2023-11-17 17:30:40 +11:00
Nicholas Nethercote
94c9075b27 Rename Linker::prepare_outputs as output_filenames.
It matches the type, and a noun makes more sense than a verb.

The `output_filenames` function still uses a profiling label named
`prepare_outputs`, but I think that makes sense as a verb and can be
left unchanged.
2023-11-17 17:30:38 +11:00
Nicholas Nethercote
aed8e1f3b6 Move CodegenBackend out of Linker.
It can easily be passed in. And that removes the single clone of
`Compiler::codegen_backend`, which means it no longer needs to be `Lrc`.
2023-11-17 17:30:36 +11:00
Nicholas Nethercote
de91b6d249 Move Session out of Linker.
It can easily be passed in. And that removes the single clone of
`Compiler::session`, which means it no longer needs to be `Lrc`.
2023-11-17 17:14:17 +11:00
Nicholas Nethercote
3560122bfc Streamline Queries::linker. 2023-11-17 17:14:17 +11:00
Nicholas Nethercote
a3b4961d5f Move lint_store from GlobalCtxt to Session.
This was made possible by the removal of plugin support, which
simplified lint store creation.

This simplifies the places in rustc and rustdoc that call
`describe_lints`, which are early on. The lint store is now built before
those places, so they don't have to create their own lint store for
temporary use, they can just use the main one.
2023-11-17 10:39:18 +11:00
Nicholas Nethercote
5dfe47acd4 Inline and remove create_lint_store. 2023-11-17 10:39:18 +11:00
Nicholas Nethercote
5c462a32bd Remove support for compiler plugins.
They've been deprecated for four years.

This commit includes the following changes.
- It eliminates the `rustc_plugin_impl` crate.
- It changes the language used for lints in
  `compiler/rustc_driver_impl/src/lib.rs` and
  `compiler/rustc_lint/src/context.rs`. External lints are now called
  "loaded" lints, rather than "plugins" to avoid confusion with the old
  plugins. This only has a tiny effect on the output of `-W help`.
- E0457 and E0498 are no longer used.
- E0463 is narrowed, now only relating to unfound crates, not plugins.
- The `plugin` feature was moved from "active" to "removed".
- It removes the entire plugins chapter from the unstable book.
- It removes quite a few tests, mostly all of those in
  `tests/ui-fulldeps/plugin/`.

Closes #29597.
2023-11-04 08:50:46 +11:00
Nilstrieb
9d42b1e268 Stop telling people to submit bugs for internal feature ICEs
This keeps track of usage of internal features, and changes the message
to instead tell them that using internal features is not supported.

See MCP 620.
2023-10-25 23:23:04 +02:00
John Kåre Alsaker
ba8d53dc8f Don't use a thread to load the dep graph 2023-09-24 00:30:43 +02:00
Matthias Krüger
c943ec2fba
Rollup merge of #115730 - bjorn3:some_driver_refactors, r=compiler-errors
Some more small driver refactors

To improve clarity and simplify some code.
2023-09-11 21:16:22 +02:00
bjorn3
90e9053189 Deprecate the pre_configure query
Only deprecating it rather than making it private to just in case
someone has a use case for it.
2023-09-10 09:41:03 +00:00
John Kåre Alsaker
1d8fdc0332 Use FreezeLock for CStore 2023-09-09 16:02:11 +02:00
John Kåre Alsaker
a3ad045ea4 Rename Freeze to FreezeLock 2023-09-02 08:14:06 +02:00
John Kåre Alsaker
0c96a9260b Add Freeze type and use it to store Definitions 2023-09-02 08:13:03 +02:00
John Kåre Alsaker
90f5f94699 Use OnceLock for SingleCache 2023-09-01 03:11:51 +02:00
bjorn3
223c43b154 Fix review comment 2023-08-14 16:16:51 +00:00
bjorn3
c6247387b4 Remove metadata_loader query
It is only used by CrateLoader. We can store the metadata loader in
CStore instead which CrateLoader has access to.
2023-08-13 16:38:50 +00:00
bjorn3
980143b50c Pass WorkProductMap to build_dep_graph instead of FxIndexMap
Constructing an FxIndexMap is useless work as the iteration order never
matters.
2023-08-13 16:07:48 +00:00
bjorn3
4a6de8e0dd Inline queries for crate_name, crate_types and stable_crate_id
All of them are not exported from rustc_interface and used only during
global_ctxt(). Inlining them makes it easier to follow the order of
queries and slightly reduces line count.
2023-08-13 13:33:37 +00:00