Commit Graph

2106 Commits (c530b74e5e4c7bb8f44531a3ff21eecfa03d66f8)

Author SHA1 Message Date
Xavier Leroy 791fe017df
Merge pull request #9655 from xavierleroy/obj-bits
Introduce type Obj.raw_data and functions Obj.raw_field, Obj.set_raw_field to manipulate out-of-heap pointers in no-naked-pointers mode, and more generally  all other data that is not a well-formed OCaml value
2020-06-14 11:44:04 +02:00
Gabriel Scherer f333db8b0f
Merge pull request #9662 from gadmm/fix-dune-build
Fix dune build after using Atomic in Stdlib
2020-06-11 21:43:09 +02:00
Sébastien Hinderer b94a3776a0
Remove temporary installation hacks (#9659)
* Remove temporary hack related to the Num library

This is a follow-up to commit 3de0115bfe

* Also do some cleanup in stdlib/Makefiile
2020-06-10 16:42:53 +02:00
Xavier Leroy ec33006c0a Add type Obj.raw_data and functions Obj.raw_field, Obj.set_raw_field
Some OCaml objects contain data that cannot be safely represented
as an OCaml value (type Obj.t).  For example, in no-naked-pointers mode,
this is the case for code pointers inside closures, and for the
"custom operations" pointers inside custom blocks.

This PR introduces a type Obj.raw_data (an alias for nativeint)
to encapsulate this data, and functions
Obj.raw_field / Obj.set_raw_field to read and write the "raw" contents
of fields of blocks.

Note: just like it is wrong to access code pointers and custom operations
using Obj.field / Obj.set_field, it is wrong to access regular fields
possibly containing pointers into the OCaml heap using
Obj.raw_field / Obj.set_raw_field.  The OCaml heap block can be
reclaimed or moved after its address was captured by Obj.raw_field.
Symmetrically, Obj.set_raw_field on a regular field bypasses the
write barrier of the GC.
2020-06-10 16:29:37 +02:00
Guillaume Munch-Maccagnoni 3a1901d873 Fix dune build
Introduce CamlinternalAtomic since Stdlib cannot refer to
Stdlib__-prefixed modules without breaking the dune build.

No change entry needed.
2020-06-10 14:29:21 +02:00
octachron 8875357638 review and change entry for #9618 2020-06-08 08:59:01 +02:00
octachron 99f0910081 format: document structural box quirk 2020-06-08 08:58:25 +02:00
octachron b6ba1823c0 format: comment max indent induced line splits 2020-06-08 08:47:39 +02:00
octachron 982d4a378e Format: margin documentation precision 2020-06-08 08:47:39 +02:00
Gabriel Scherer 63eb3dad0d Arg: a new Rest_all spec, similar to Rest, that passes all arguments at once
In particular this allow to detect a Rest_all keyword followed by no
arguments at all (an empty list).
2020-06-05 22:06:21 +02:00
Xavier Leroy 5890df0ad1 Update documentation of Digest module
State more clearly that MD5 is not cryptographically secure.
2020-05-18 11:33:06 +02:00
Guillaume Munch-Maccagnoni 1e4eea19ef Make at_exit forward-compatible with multicore 2020-05-16 22:58:07 +02:00
Guillaume Munch-Maccagnoni a469b4d0af Implement at_exit with Atomics
Atomic becomes a dependency of Stdlib
2020-05-16 22:58:07 +02:00
Guillaume Munch-Maccagnoni abe982165b Fix data race in at_exit and Printexc.register_printer with systhreads 2020-05-16 22:57:34 +02:00
Gabriel Scherer 04d9c425f3 stdlib: provide a sequential interface of the Atomic module from OCaml Multicore
This module provides a purely sequential implementation of the
concurrent atomic references provided by the Multicore OCaml
standard library:

https://github.com/ocaml-multicore/ocaml-multicore/blob/parallel_minor_gc/stdlib/atomic.mli

This sequential implementation is provided in the interest of
compatibility: when people will start writing code to run on
Multicore, it would be nice if their use of Atomic was
backward-compatible with older versions of OCaml without having to
import additional compatibility layers. *)
2020-05-16 17:51:54 +02:00
Sébastien Briais f52fdc2068
Define to_rev_seq in Set and Map module (#9075) 2020-05-15 17:38:45 +02:00
Nicolás Ojeda Bär 4a44bf1767 Expose %loc_FUNCTION as __FUNCTION__ 2020-05-13 20:57:21 +02:00
Christophe Troestler c920ea142b
Fix comment about NaN propagation (#9536)
Fix documentation comment about NaN propagation

Closes: #7891

Co-authored-by: David Allsopp <david.allsopp@metastack.com>
2020-05-12 15:35:04 +02:00
Florian Angeletti c86a5d8d11
#9189: avoid one-letter make variables (#9281) 2020-05-11 12:21:59 +02:00
Gabriel Scherer fd789b5cec [minor] stdlib/Makefile: remove redundant clean action
There is already a more precise

    clean::
            rm -f $(CAMLHEADERS)

rule above.

(Change suggested by David Allsopp.)
2020-05-06 12:10:02 +02:00
Gabriel Scherer fab58a938a makefiles: move the inclusion of Makefile.build_config in Makefile.common 2020-05-06 12:10:02 +02:00
Florian Angeletti 24806a0223
Merge pull request #9228 from yallop/map-documentation
Fix some issues in the Map documentation
2020-05-04 09:49:53 +02:00
Josh Berdine a282c170f9
Update stdlib/ephemeron.mli
Co-authored-by: David Allsopp <david.allsopp@metastack.com>
2020-05-03 00:25:54 +01:00
Josh Berdine 5e2b28decc Fix ephemeron documentation typos
Signed-off-by: Josh Berdine <josh@berdine.net>
2020-05-02 22:46:48 +01:00
Jeremy Yallop afe681cc16 Fix some issues in the Map documentation:
- Add the key argument in the description of 'merge'
  - Note that the keys of 'union f m1 m2' are a subset of the
    input keys, not all the keys, since

        bindings (union (fun _ _ _ -> None) m m) = []

  - Fix grammar in the descriptions of 'filter', 'union', 'merge'
  - Fix mismatched variable name in the description of 'partition'
  - Note that 'find' and 'find_opt' return values, not bindings
2020-04-30 16:10:00 +01:00
Enguerrand Decorne b7f0494df5 Rewrite the instrumented runtime to store traces in the CTF format.
The instrumentation code in the instrumented runtime was replaced
with new APIs to gather runtime statistics and output them in a new format
(Common Trace Format).
This commit also exposes new functions in the Gc module to pause or resume
instrumentation during a program execution (Gc.eventlog_pause and
Gc.eventlog_resume).
2020-04-30 10:32:01 +02:00
Edwin Török f66e395c70 Fix docs: runtime warnings are not enabled by default
The docs on this are out of date: they say runtime warnings are enabled by
default, but the source code has a commit 6c90da4 pointing to
https://github.com/ocaml/ocaml/pull/210 that disables them by default.

It seems like enabling runtime warnings by default never made its way into a
release and the docs could just be updated to say that runtime warnings are
disabled by default.

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
2020-04-27 21:38:58 +01:00
Stephen Dolan 0040c5d783 Print function names (derived from Lambda.scoped_location) in backtraces
Function names now appear in backtraces and are available via Printexc.
2020-04-27 12:58:53 +01:00
Mark Shinwell df20ccf838 Fix recursive module initialisation that does not terminate (#9497)
This fixes the bug reported in issue 9494 and adds a test case.

Closes: #9494 
(cherry picked from commit ff98901d11fb01ba224772c402763df20d246635)
2020-04-24 17:50:19 +02:00
Greta Yorsh b823c8d8fc Remove stray newlines 2020-04-20 13:56:53 +01:00
Greta Yorsh 7271919a47 Check FLAMBDA in Makefile.common.in to update OPTCOMPFLAGS 2020-04-20 13:56:53 +01:00
Greta Yorsh 44af4aea7c Check FUNCTION_SECTIONS in Makefile.common.in to update OPTCOMPFLAGS 2020-04-20 13:56:53 +01:00
Greta Yorsh 7465e9f2c7 Define OPTCOMPFLAGS in Makefile.common.in 2020-04-20 13:56:53 +01:00
David Allsopp 3a40b2fd94 Introduce Makefile.build_config.in
This moves the configure-generated parts of Makefile.common to a
separate (generated) Makefile, allowing Makefile.common to be a normal
Makefile.

OCaml's build system Makefile's now include Makefile.build_config (which
itself includes Makefile.config) but Makefile.config is still installed
as before. This allows configure to generate variables which are
specific to the build process and are not intended to be exported to the
installation.
2020-04-17 13:53:49 +01:00
David Allsopp 13786d7d12 Ensure make distclean works on an unconfigured tree 2019-10-15 11:46:36 +01:00
David Allsopp d4a566573f Allow make to be invoked before configure
This should be improved to give better warnings for when Makefile.config
and Makefile.common are required.
2018-12-04 10:28:36 +00:00
Guillaume Munch-Maccagnoni b1fdc44547
Install a pretty printer for the Fun.Finally_raised exception (#9266) 2020-04-13 19:09:47 +02:00
Gabriel Scherer 1d76e2ca91
Merge pull request #9365 from gasche/Set-Map-filter_map
Map.filter_map and Set.filter_map
2020-04-02 17:59:51 +02:00
Gabriel Scherer ff6b20098f Gc.Memprof.start: take a record instead of 5 optional parameters
The Gc.Memprof module provides a low-level API, that will hopefully be
paired with user libraries that provide high-level instrumentation
choices.

A natural question is: how are the higher-level API going to expose
their choice of instrumentation to their users? With the current
Memprof.start API (before this patch), they would have to either
provide their own `start` function wrapping Memprof.start, or provide
a tuple of callbacks for to their users to pass to Memprof.start
themselves.

    val start : params -> unit
    (* or *)
    val callback : params ->
      ((allocation -> foo option) * (allocation -> bar option) * ... )

With an explicit record, it is easier for libraries to expose an
instrumentation choice (possibility parametrized over
user-provided settings):

    val tracker : params -> (foo, bar) Gc.Memprof.tracker

In addition, providing a record instead of optional parameters makes
it much easier to provide "default settings" (helper functions) that
instantiates the types `'minor` and `'ḿajor`, see for example
`simple_tracker` in this patch (which stores the same information for
the minor and major heap, and does not observe promotion), or to later
define checking predicates that can verify that a given choice of
callbacks is sensible (for example: providing a major-dealloc callback
but no promotion callback (dropping all tracked value on promotion) is
probably not a good idea).

Bootstrap: to avoid requiring an awkward bootstrap, this commit keeps
the (now unused) function caml_memprof_start_byt unchanged -- it is
used in the bootstrap binaries, so removing it would break the
build. The intention is to remove it in the following commit.
2020-04-01 16:37:28 +02:00
Gabriel Scherer 1cd6e4451f Map.filter_map and Set.filter_map 2020-03-31 15:01:41 +02:00
muskangarg21 4019c33aaa
Stdlib: added Printf.ikbprintf, Printf.ibprintf (#9364) 2020-03-26 14:28:00 +01:00
David Allsopp e4f3aa8c81 Don't assume . in AWKPATH
Ensure all script invocations explicitly refer to a directory.
2020-03-20 09:38:33 +00:00
zapashcanon 75bc37d4d9
fix doc comment for seq type 2020-03-19 21:39:06 +01:00
zapashcanon a8a3c55b18
fix doc comment for unit type 2020-03-18 20:22:08 +01:00
Sébastien Briais cc4a075e9b
Define Seq.cons and Seq.append. (#9077)
* Define Seq.cons and Seq.append.
2020-03-06 09:26:38 +01:00
Glenn Slotte 4823934182
Document base case of List.for_all and List.exists (#9325)
* Document the base case (`[ ]`) of List.for_all and List.exists
2020-02-26 09:51:01 +01:00
Gabriel Scherer b237282e97 Merge pull request #9270 from gadmm/doc_stack_overflow
Documentation: Stack_overflow is reliable, Out_of_memory is not.
(cherry picked from commit 84f62eee5d26efb936cde0e799b7aaa39e46c010)
2020-01-29 11:08:57 +01:00
Raphael Sousa Santos c9d7cc7da4 Add Printexc.default_uncaught_exception_handler (issue #9248)
Printexc.uncaught_exception_handler ceases to be an option ref and becomes
a ref to the handler function initialized to
Printexc.default_uncaught_exception_handler.
2020-01-26 12:00:27 +01:00
Gabriel Scherer 82d7971b4f
Merge pull request #9237 from gasche/format-update-geometry
Format.pp_update_geometry: formatter -> (geometry -> geometry) -> unit
2020-01-15 22:37:37 +01:00
Jeremy Yallop 73a37d45f8 Add Seq.unfold. 2020-01-15 15:38:46 +00:00