| Age | Commit message (Collapse) | Author |
|---|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13979 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
The recent experiment with -dont-load-proofs in the stdlib showed that
this options isn't fully safe: some axioms were generated (Include ?
functor application ? This is still to be fully understood).
Instead, I've implemented an idea of Yann: only load opaque proofs when
we need them. This is almost as fast as -dont-load-proofs (on the stdlib,
we're now 15% faster than before instead of 20% faster with -dont-load-proofs),
but fully compatible with Coq standard behavior.
Technically, the const_body field of Declarations.constant_body now regroup
const_body + const_opaque + const_inline in a ternary type. It is now either:
- Undef : an axiom or parameter, with an inline info
- Def : a transparent definition, with a constr_substituted
- OpaqueDef : an opaque definition, with a lazy constr_substitued
Accessing the lazy constr of an OpaqueDef might trigger the read on disk of
the final section of a .vo, where opaque proofs are located.
Some functions (body_of_constant, is_opaque, constant_has_body) emulate
the behavior of the old fields. The rest of Coq (including the checker)
has been adapted accordingly, either via direct access to the new const_body
or via these new functions. Many places look nicer now (ok, subjective notion).
There are now three options: -lazy-load-proofs (default), -force-load-proofs
(earlier semantics), -dont-load-proofs. Note that -outputstate now implies
-force-load-proofs (otherwise the marshaling fails on some delayed lazy).
On the way, I fixed what looked like a bug : a module type
(T with Definition x := c) was accepted even when x in T was opaque.
I also tried to clarify Subtyping.check_constant.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13952 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
- Avoid using Util which depends on Compat and hence Camlp4
- Instead, a small Minilib module specific to coqide, which
duplicate 5 functions from Util (50 lines)
- some dead code removal
- the coqlib variable is asked to coqtop
- remove obsolete Util.check_for_interrupt
This way, coqide only depends on 3 files outside ide/ :
Coq_config, Flags, Ide_intf. Makefile and ocamlbuild are adapted
accordingly.
TODO: how should we signal coqide error, warnings, etc ?
For the moment, some Printf.eprintf, some failwith.
To uniformize later...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13930 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
This was breaking the FireSquad extraction test
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13919 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
Since library files aren't meant to contain queries like Print Assumptions
nor extractions, it is safe (and quite quicker) to compile them with
-dont-load-proofs
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13918 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
This reverts pottier's commit r13849. It references a
ncring_plugin.cma for which there is no rule. I guess he forgot to
commit something...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13856 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13849 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
According to B. Gregoire, this stuff is obsolete. Fine control
on when to launch the VM in conversion problems is now provided
by VMcast. We were already almost never boxing definitions anymore
in stdlib files.
"(Un)Boxed Definition foo" will now trigger a parsing error,
same with Fixpoint. The option "(Un)Set Boxed Definitions"
aren't there anymore, but tolerated (as no-ops), since unknown
options raise a warning instead of an error by default.
Some more cleaning could be done in the vm.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13806 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13797 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
if `pkg-config --exists ige-mac-integration`, coqide.opt will be
able to open files by double-clik in finder on Darwin.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13779 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13777 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13776 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
in addition, a makefile entry pluginsopt for compiling only the .cmxs
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13509 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13325 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
If you want to try, it should be now as simple as:
make clean && ./configure -local -usecamlp4 && make
For the moment, the default stays camlp5, hence
./configure -usecamlp5 and ./configure are equivalent.
Thanks to a suggestion by N. Pouillard, the remaining
incompatibilities are now handled via some token filtering in
camlp4. See compat5*.mlp. Morally, these files should be named .ml4,
but I prefer having them not in $(...ML4) variables, it might confuse
the Makefile... The empty compat5*.ml are used to build empty .cmo
for making camlp5 happy. For camlp4,
- tools/compat5.cmo changes GEXTEND into EXTEND. Safe, always loaded
- tools/compat5b.cmo changes EXTEND into EXTEND Gram. Interact badly with
syntax such that VERNAC EXTEND, we only load it for a few files via
camlp4deps
TODO: check that my quick adaptation of camlp5-specific code in
tactics/extratactics.ml4 is ok. It seems the code by Chung-Kil Hur
is hiding information in the locations ?!
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13274 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
Looking at the source code of Sys.time reveals that it is exactly what
is computed by Profile.get_time. This can also be tested by evaluating:
Sys.time () -. Unix.(let x = times () in x.tms_utime +. x.tms_stime);;
in an OCaml toplevel with Unix.
This allows to put Profile in grammar.cma without the dependency to
unix.cma while preprocessing.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13233 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
Fixing tests 2145.v about Nsatz. Adding nsatz target in Makefile.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13203 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13178 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
Linking a system library into a .cma can lead to some system module
(such as Unix) being linked twice in an executable, resulting in
possibly hard-to-debug errors.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13054 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
The choice between camlp4/5 is done during configure with flags
-usecamlp5 (default for the moment) vs. -usecamlp4.
Currently, to have a full camlp4 compatibility, you need to change
all "EXTEND" and "GEXTEND Gram" into "EXTEND Gram", and change "EOI"
into "`EOI" in grammar entries. I've a sed script that does that
(actually the converse), but I prefer to re-think it and check a few
things before branching this sed into the build mechanism.
lib/compat.ml4 is heavily used to hide incompatibilities between camlp4/5
and try to propose a common interface (cf LexerSig / GrammarSig).
A few incompatible quotations have been turned into underlying code
manually, in order to make the IFDEF CAMLP5 THEN ... ELSE ... END
parsable by both camlp4 and 5. See in particular the fate of
<:str_item< declare ... end >>
Stdpp isn't used anymore, but rather Ploc (hidden behind local module Loc).
This forces to use camlp5 > 5.01.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13019 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@13016 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
- coq logo isn't destructed anymore
- Erase debug printers not implemented for new proofs
- ocamldoc compatible comments for pretyping/rawterm.mli
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12988 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12986 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
dev/ocamldoc/
old "make source-doc" that documents ml files and didn't work is now
"make ml-doc" but still don't work :-)
"make clean" cleans dev/ocamldoc/ properly
wierd? calls of dependency graph generation leave unchanged
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12978 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
- Many of them were broken, some of them after Pierre B's rework
of mli for ocamldoc, but not only (many bad annotation, many files
with no svn property about Id, etc)
- Useless for those of us that work with git-svn (and a fortiori
in a forthcoming git-only setting)
- Even in svn, they seem to be of little interest
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12972 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12926 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
camlp4deps/use lines
The build error Y. Makarov recently reported on coq-club was caused by some
.ml4 having dos-style end-of-line in his source-tree, and hence \r appearing in
the camlp4o commandline. We now simply discard what appears after camlp4xxx "...".
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12871 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
- avoid recomputing CAMLP4DEPS in the %.ml:%.ml4 rule
- a macro for compiling the tools by the best ocaml compiler
- use of $(if ...) rather that $ifdef
- some variables of Makefile.common were not that useful
(e.g. $(COQCCMX), which is $(COQCCCMO:.cmo=.cmx), used only once)
- the build of coqc.* should not depend upon coqtop, only its launch
(or I'm missing something)
- useless $(CAMLP4EXTENDFLAGS) variable
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12846 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
(thanks A. Frisch)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12841 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
dev/doc/build*.txt
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12840 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
- Instead of the separate stage mechanism, we let make handle the
build and inclusion of all .d. Some initial calls to camlp4o
will fail, but make tries again later, and this finally works
great. These initial error message are made nice to avoid bad
interaction with M-x next-error
- The only recursive call to a sub-make is Makefile calling Makefile.build
in which the includes of .d take place. This allows to avoid compiling
anything for a make clean or make tags
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12839 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
As said now in a comment of Makefile.build:
NB: for the moment ocamlopt erases and recreates .cmi if there's no .mli around.
This can lead to nasty things with make -j (e.g. another process accessing a
partial .cmi). To avoid that:
1) We make .cmx always depend on .cmi
2) This .cmi will be created from the .mli, or trigger the compilation of the
.cmo if there's no .mli (see rule below about MLWITHOUTMLI)
3) We tell ocamlopt to output to temporary names, remove the temp .cmi
(since the actual .cmi should already be there and up-to-date) and move
the temp .cmx and .o back in place
Ok, this is quite a hack. I'll make a proper bug report to ocaml asap...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12837 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
- We clarify their definition via some custom make function: find, diff...
- We avoid duplications via some $(sort ...)
- Some name changes:
* old $(MLFILES) now corresponds to $(MLSTATICFILES) (but .ml from
.mly and .mll aren't in it anymore).
* new $(MLFILES) contains all .ml, either static or generated from
.mly .mll .ml4 or _mod.ml made out of .mllib
* $(ML4FILESML) is now $(GENML4FILES)
...
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12836 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
We use some conditional variables $(if ...) to set the proper flags
instead of fully duplicating the rules according to the path
checker/%.cm? ide/%.cm? %.cm?
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12834 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
- This way, the Makefile.build gets shorter and simplier, with a few nasty
hacks removed.
- In particular, we stop creating dummy .ml of .ml4 early "to please ocamldep".
Instead, we now use ocamldep -modules, and process its output via coqdep_boot.
This ways, *.cm* of .ml4 are correctly located, even when some .ml files
aren't generated yet.
- There is no risk of editing the .ml of a .ml4 by mistake, since it is by
default in a binary format (cf pr_o.cmo and variable READABLE_ML4).
M-x next-error still open the right .ml4 at the right location.
- mltop.byteml is now mltop.ml, while mltop.optml keeps its name
- .ml of .ml4 are added to .gitignore
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12833 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
$(TOTARGET)
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12832 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
--fqdn. Changed this to -f (which has the same behavior) so it will work on
Mac OS 10.6 Note: Previous versions of Mac OS 10 don't have this option. Most
BSD's don't either. Their default behavior is to output the FQDN where Linux
defaults to outputting the short name.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12810 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
autodetection via ./configure, automated installation target
"install-im", and no more patching. Plus documentation of the procedure
in the reference manual.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12790 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12778 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12776 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12690 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
The macro-generated .v file is now NMake_gen.v, while NMake.v now
contain the static things (i.e. definition of gcd via mod).
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12687 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12666 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12628 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
in */*/vo.itarget
On the way: no more -fsets (yes|no) and -reals (yes|no) option of configure
if you want a partial build, make a specific rule such as theories-light
Beware: these vo.itarget should not contain comments. Even if this is legal
for ocamlbuild, the $(shell cat ...) we do in Makefile can't accept that.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12574 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
It has moved to the contribs (Sophia-Antipolis/Interface).
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12555 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12376 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
files embedded in grammar.cma can be profiled with the Profile module.
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12279 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12256 85f007b7-540e-0410-9357-904b9bb8a0f7
|
|
revisions 12063 and 12065).
git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@12169 85f007b7-540e-0410-9357-904b9bb8a0f7
|
