| Age | Commit message (Collapse) | Author |
|---|
|
Reviewed-by: SkySkimmer
Ack-by: ppedrot
Ack-by: ejgallego
|
|
called (fix #13849).
The libraries are eventually linked in native_norm and native_conv_gen,
just before mk_norm_code and mk_conv_code are called.
This commit also renames call_linker as execute_library to better reflect
its role. It also makes link_library independent from it, since their
error handling are completely opposite.
|
|
|
|
|
|
|
|
We store bound variable names instead of functions for both branches and
predicate, and we furthermore add the parameters in the node. Let bindings
are not taken into account and require an environment lookup for retrieval.
|
|
Instead we store that data in the native code that was generated in adapt
the compilation scheme accordingly. Less indirections and less imperative
tinkering makes the code safer.
The global symbol table was originally introduced in #10359 as a way not to
depend on the Global module in the generated code. By storing all the
native-related information in the cmxs file itself, this PR also makes other
changes easier, such as e.g. #13287.
|
|
Reviewed-by: ppedrot
|
|
|
|
Persistent arrays expose a functional interface but are implemented
using an imperative data structure. The OCaml implementation is based on
Jean-Christophe Filliâtre's.
Co-authored-by: Benjamin Grégoire <Benjamin.Gregoire@inria.fr>
Co-authored-by: Gaëtan Gilbert <gaetan.gilbert@skyskimmer.net>
|
|
|
|
Ack-by: SkySkimmer
Reviewed-by: maximedenes
|
|
User time is unreliable for `native_compute`.
Also output time spent in conversion to native code, just in case that
is ever slow.
Note that this also removes spurious newlines in the output.
Fixes #11962
|
|
- Provide new helper functions in `Goptions` on the model of
`declare_bool_option_and_ref`;
- Use these helper functions in many parts of the code base
(encapsulates the corresponding references);
- Move almost all options from `declare_string_option` to
`declare_stringopt_option` (only "Warnings" continue to use the
former). This means that these options now support `Unset` to get
back to the default setting. Note that there is a naming
misalignment since `declare_int_option` is similar to
`declare_stringopt_option` and supports `Unset`. When "Warning" is
eventually migrated to support `Unset` as well, we can remove
`declare_string_option` and rename `declare_stringopt_option` to
`declare_string_option`.
- For some vernac options and flags that have an equivalent
command-line option or flag, implement it like the standard `-set`
and `-unset`.
|
|
This corresponds more naturally to the use we make of them, as we don't need
fast indexation but we instead keep pushing terms on top of them.
|
|
Ack-by: SkySkimmer
Reviewed-by: maximedenes
|
|
Add headers to a few files which were missing them.
|
|
The command `Set NativeCompute Timing` causes calls to `native_compute`
(as well as kernel calls to the native compiler) to emit separate timing
information about compilation, execution, and reification. This allows
more fine-grained timing of the native compiler without needing to set
the `-debug` flag.
|
|
Instead of relying on the Coq_config immutable flag, we introduce an
initialization-only flag to govern the use of the native compiler. This
allows to make coqc passed with "-native-compiler no" behave as if it had
been configured without native compilation.
|
|
We also remove trailing whitespace.
Script used:
```bash
for i in `find . -name '*.ml' -or -name '*.mli' -or -name '*.mlg'`; do expand -i "$i" | sponge "$i"; sed -e's/[[:space:]]*$//' -i.bak "$i"; done
```
|
|
* Float added to is_value/get_value to avoid stack overflows
(cf. #7646)
* beware of the use of Array.map with floats (cf. comment in the
makeblock function)
NB: From here one, the configure option "-native-compiler no"
is no longer needed.
|
|
Reviewed-by: maximedenes
Reviewed-by: ppedrot
|
|
|
|
Instead we get the symbols from a Environ.env.
We make them accessible to the produced code through a reference
managed by the kernel, similar to the return values except inverting
when it's written and when it's read.
|
|
|
|
Fixes #6699
|
|
This was making an assertion fail on
https://github.com/coq/coq/issues/9684 after 23f84f37
|
|
Kernel should be mostly correct, higher levels do random stuff at
times.
|
|
It used to simply remember the normal form of the type of the constructor.
This is somewhat problematic as this is ambiguous in presence of
let-bindings. Rather, we store this data in a fully expanded way, relying
on rel_contexts.
Probably fixes a crapload of bugs with inductive types containing
let-bindings, but it seems that not many were reported in the bugtracker.
|
|
This work makes it possible to take advantage of a compact
representation for integers in the entire system, as opposed to only
in some reduction machines. It is useful for heavily computational
applications, where even constructing terms is not possible without such
a representation.
Concretely, it replaces part of the retroknowledge machinery with
a primitive construction for integers in terms, and introduces a kind of
FFI which maps constants to operators (on integers). Properties of these
operators are expressed as explicit axioms, whereas they were hidden in
the retroknowledge-based approach.
This has been presented at the Coq workshop and some Coq Working Groups,
and has been used by various groups for STM trace checking,
computational analysis, etc.
Contributions by Guillaume Bertholon and Pierre Roux <Pierre.Roux@onera.fr>
Co-authored-by: Benjamin Grégoire <Benjamin.Gregoire@inria.fr>
Co-authored-by: Vincent Laporte <Vincent.Laporte@fondation-inria.fr>
|
|
This is a pre-requisite to use automated formatting tools such as
`ocamlformat`, also, there were quite a few places where the comments
had basically no effect, thus it was confusing for the developer.
p.s: Reading some comments was a lot of fun :)
|
|
See commit [Simplify code for [Definition := Eval ...]] which without
this breaks test suite 7631.v
|
|
|
|
Même causes, mêmes effets, similar fix to #8119:
- Do not pass let-bound arguments to evars.
We seize the opportunity to remove the useless type information for Aevar.
Special fixes to native compilation:
- Evars are not handled correctly when iterating over lambda terms.
- Names.id_of_string is gone.
- Evar instances are not reified in the right order.
|
|
The upper layers still need a mapping constant -> projection, which is
provided by Recordops.
|
|
compilers
|
|
|
|
I believe this is legacy code due to a previous, more complex
representation of return predicates in the kernel.
|
|
This brings more compatibility with handling of mutual primitive records
in the kernel.
|
|
|
|
Instead of having a constant-based compilation of projections, we
generate them at the compilation time of the inductive block to which
they pertain.
|
|
Note that the normalization of the context of the return predicate was
not done by the native compilation but by the lazy machine. The patch
also "fixes" an anomaly in the case of an arity which was not in
canonical form as in:
Inductive A : nat -> id (nat->Type) := .
Eval native_compute in fun x => match x in A y z return y = z with end.
|
|
We now have only two notions of environments in the kernel: env and
safe_env.
|
|
We bootstrap the circular evar_map <-> econstr dependency by moving
the internal EConstr.API module to Evd.MiniEConstr. Then we make the
Evd functions use econstr.
|
|
|
|
|
|
|
|
The patch has three parts:
- Introduction of a configure flag `-bytecode-compiler (yes|no)`
(due to static initialization this is a configure-time option)
- Installing the hooks that register the VM with the pretyper and the
kernel conditionally on the flag.
- Replacing the normalization function in `Redexpr` by compute if the
VM is disabled.
We also rename `Coq_config.no_native_compiler` to `native_compiler`
and `Flags.native_compiler` to `output_native_objects` [see #4607].
|
|
We do up to `Term` which is the main bulk of the changes.
|
|
This is a first step towards some of the solutions proposed in #6008.
|
