Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

[Caml-list] OCaml 3.11.0 release candidate

62 views
Skip to first unread message

Damien Doligez

unread,
Nov 25, 2008, 12:09:39 PM11/25/08
to caml users
Dear OCaml users,

We are closing in on version 3.11.0. A Release Candidate is now
available.
If there are no show-stoppers in this RC, then 3.11.0 will be officially
released next week.

The release candidate is available here:
< http://caml.inria.fr/pub/distrib/ocaml-3.11/ >
(look for 3.11.0+rc1)

As usual, we need a few brave souls to try and install it on their
favourite architecture and report the result to me.

Happy hacking,

-- Damien Doligez for the OCaml team.

_______________________________________________
Caml-list mailing list. Subscription management:
http://yquem.inria.fr/cgi-bin/mailman/listinfo/caml-list
Archives: http://caml.inria.fr
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
Bug reports: http://caml.inria.fr/bin/caml-bugs

Dario Teixeira

unread,
Nov 25, 2008, 4:56:51 PM11/25/08
to caml users
Hi,

> We are closing in on version 3.11.0. A Release Candidate
> is now available.
> If there are no show-stoppers in this RC, then 3.11.0 will
> be officially released next week.

For GODI, what's the CVS branch one should use?
Am I correct in assuming "ocaml3110rc1"?

Also, using GODI and the "ocaml3110rc1" branch,
I get a compilation error:

make[8]: execvp: ./build/mixed-boot.sh: Permission denied
make[8]: *** [ocamlbuild-mixed-boot] Error 127

I reckon the problem is that the mixed-boot.sh script does not
have the executable flag set. Anyone else encountering this
problem?

Cheers,
Dario Teixeira

Xavier Leroy

unread,
Nov 26, 2008, 5:32:14 AM11/26/08
to caml users
Damien Doligez wrote:

> We are closing in on version 3.11.0. A Release Candidate is now available.

> The release candidate is available here:
> < http://caml.inria.fr/pub/distrib/ocaml-3.11/ >

The documentation for 3.11 is also available from the same place. The
HTML manual can be browsed online at

http://caml.inria.fr/pub/docs/manual-ocaml-311/index.html

Enjoy,

- Xavier Leroy

Richard Jones

unread,
Nov 26, 2008, 9:14:50 AM11/26/08
to Damien Doligez, caml users
On Tue, Nov 25, 2008 at 06:09:28PM +0100, Damien Doligez wrote:
> We are closing in on version 3.11.0. A Release Candidate is now
> available.
> If there are no show-stoppers in this RC, then 3.11.0 will be officially
> released next week.
>
> The release candidate is available here:
> < http://caml.inria.fr/pub/distrib/ocaml-3.11/ >
> (look for 3.11.0+rc1)
>
> As usual, we need a few brave souls to try and install it on their
> favourite architecture and report the result to me.

I'm just rebuilding all the Fedora 11 packages right now. I can
report so far that the base 3.11.0+rc1 package builds fine.

Rich.

--
Richard Jones
Red Hat

Richard Jones

unread,
Nov 26, 2008, 11:29:54 AM11/26/08
to Damien Doligez, caml users
Two camlp4 packages so far have given this error message, which is new
since I tried 3.11.0+beta1:

File "_none_", line 1, characters 0-1:
Error: Error while linking /usr/lib/ocaml/camlp4/camlp4lib.cma(Camlp4):
Reference to undefined global `Dynlink'

Andres Varon

unread,
Nov 26, 2008, 11:52:34 AM11/26/08
to Damien Doligez, caml users
Hello,

On Nov 25, 2008, at 12:09 PM, Damien Doligez wrote:

> The release candidate is available here:
> < http://caml.inria.fr/pub/distrib/ocaml-3.11/ >
> (look for 3.11.0+rc1)
>
> As usual, we need a few brave souls to try and install it on their
> favourite architecture and report the result to me.

So far our tests have compiled and run successfully in most platforms
[1] excepting Windows 64 bits (but it's the first time I attempt to
compile in this architecture). I am getting this error message:

Entering directory `/home/andres/ocaml-3.11.0+rc1/byterun`
flexlink -x64 -merge-manifest -exe -o ocamlrun.exe prims.obj
ws2_32.lib libcamlrun.lib
** Fatal error: Cannot resolve symbols for <my build path here>
\flexdll-0.10\flexdll_msvc.obj:
__security_check_cookie
__security_cookie

Any suggestions?

best,

Andres
[1] Linux x86, Linux amd64, Mac OS X PPC 32 bits, Mac OS X Intel 32
bits, , Mac OS X Intel 64 bits, Windows 32 bits.

Xavier Leroy

unread,
Nov 26, 2008, 12:05:51 PM11/26/08
to Andres Varon, caml users
> So far our tests have compiled and run successfully in most platforms
> [1] excepting Windows 64 bits (but it's the first time I attempt to
> compile in this architecture). I am getting this error message:

There are two problems with Windows 64 that I forgot to fix before
this RC1 release; this is one of them. A patch is available at

http://gallium.inria.fr/~xleroy/ocaml-3.11rc1-win64.patch

The CVS release branch contains this fix.

Regards,

- Xavier Leroy

Andres Varon

unread,
Nov 26, 2008, 12:38:45 PM11/26/08
to Xavier Leroy, caml users

On Nov 26, 2008, at 12:05 PM, Xavier Leroy wrote:

>> So far our tests have compiled and run successfully in most platforms
>> [1] excepting Windows 64 bits (but it's the first time I attempt to
>> compile in this architecture). I am getting this error message:
>
> There are two problems with Windows 64 that I forgot to fix before
> this RC1 release; this is one of them. A patch is available at
>
> http://gallium.inria.fr/~xleroy/ocaml-3.11rc1-win64.patch
>

Thanks. This time the compilation went further, but failed with the
errors below.


> andres@WIN-X85PMN663RN ~/ocaml-3.11.0+rc1
> $ make -f Makefile.nt world
> ....
> make[1]: Entering directory `/home/andres/ocaml-3.11.0+rc1/stdlib'
> make[1]: Nothing to be done for `all'.
> make[1]: Leaving directory `/home/andres/ocaml-3.11.0+rc1/stdlib'
> for i in win32unix systhreads str num win32graph dynlink bigarray;
> do make -f Ma
> kefile.nt -C otherlibs/$i all; done
> make[1]: Entering directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> win32unix'
> cl /nologo -D_CRT_SECURE_NO_DEPRECATE /I 'C:/Program Files/Microsoft
> Platform SD
> K/Include/crt' /I 'C:/Program Files/Microsoft Platform SDK/
> Include' /Ox /MD -I.
> ./../byterun -I../unix -c accept.c
> accept.c
> accept.c(58) : error C2146: syntax error : missing ';' before
> identifier 'fd'
> make[1]: *** [accept.obj] Error 2
> make[1]: Leaving directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> win32unix'
> make[1]: Entering directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> systhreads'
>
> cl /nologo -D_CRT_SECURE_NO_DEPRECATE /I 'C:/Program Files/Microsoft
> Platform SD
> K/Include/crt' /I 'C:/Program Files/Microsoft Platform SDK/
> Include' /Ox /MD -I.
> ./../byterun -c win32.c
> win32.c
> win32.c(217) : error C2037: left of 'mutex' specifies undefined
> struct/union 'ch
> annel'
> win32.c(225) : error C2037: left of 'mutex' specifies undefined
> struct/union 'ch
> annel'
> win32.c(228) : error C2037: left of 'mutex' specifies undefined
> struct/union 'ch
> annel'
> win32.c(231) : error C2037: left of 'mutex' specifies undefined
> struct/union 'ch
> annel'
> win32.c(231) : error C2198: 'WaitForSingleObject' : too few
> arguments for call t
> hrough pointer-to-function
> win32.c(236) : error C2037: left of 'mutex' specifies undefined
> struct/union 'ch
> annel'
> win32.c(236) : warning C4022: 'WaitForSingleObject' : pointer
> mismatch for actua
> l parameter 1
> win32.c(236) : error C2198: 'WaitForSingleObject' : too few
> arguments for call t
> hrough pointer-to-function
> win32.c(248) : error C2037: left of 'mutex' specifies undefined
> struct/union 'ch
> annel'
> win32.c(248) : error C2198: 'ReleaseMutex' : too few arguments for
> call through
> pointer-to-function
> win32.c(328) : error C2065: 'caml_channel_mutex_free' : undeclared
> identifier
> win32.c(328) : warning C4047: '=' : 'int' differs in levels of
> indirection from
> 'void (__cdecl *)(channel *)'
> win32.c(329) : error C2065: 'caml_channel_mutex_lock' : undeclared
> identifier
> win32.c(329) : warning C4047: '=' : 'int' differs in levels of
> indirection from
> 'void (__cdecl *)(channel *)'
> win32.c(330) : error C2065: 'caml_channel_mutex_unlock' : undeclared
> identifier
> win32.c(330) : warning C4047: '=' : 'int' differs in levels of
> indirection from
> 'void (__cdecl *)(channel *)'
> win32.c(331) : error C2065: 'caml_channel_mutex_unlock_exn' :
> undeclared identif
> ier
> win32.c(331) : warning C4047: '=' : 'int' differs in levels of
> indirection from
> 'void (__cdecl *)(void)'
> make[1]: *** [win32_b.obj] Error 2
> make[1]: Leaving directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> systhreads'
> make[1]: Entering directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> str'
> cl /nologo -D_CRT_SECURE_NO_DEPRECATE /I 'C:/Program Files/Microsoft
> Platform SD
> K/Include/crt' /I 'C:/Program Files/Microsoft Platform SDK/
> Include' /Ox /MD -I.
> ./../byterun -c strstubs.c
> strstubs.c
> strstubs.c(294) : warning C4047: 'initializing' : 'backtrack_stack
> *' differs in
> levels of indirection from 'int'
> strstubs.c(345) : error C2065: 'res' : undeclared identifier
> strstubs.c(346) : error C2143: syntax error : missing ';' before
> 'type'
> strstubs.c(347) : error C2143: syntax error : missing ';' before
> 'type'
> strstubs.c(348) : error C2143: syntax error : missing ';' before
> 'type'
> strstubs.c(349) : error C2143: syntax error : missing ';' before
> 'type'
> strstubs.c(351) : error C2065: 'n' : undeclared identifier
> strstubs.c(352) : error C2065: 'i' : undeclared identifier
> strstubs.c(353) : error C2065: 'group' : undeclared identifier
> strstubs.c(353) : warning C4047: '=' : 'int' differs in levels of
> indirection fr
> om 're_group *'
> strstubs.c(354) : error C2223: left of '->start' must point to
> struct/union
> strstubs.c(354) : error C2223: left of '->end' must point to struct/
> union
> strstubs.c(358) : error C2223: left of '->start' must point to
> struct/union
> strstubs.c(358) : error C2065: 'starttxt' : undeclared identifier
> strstubs.c(359) : error C2223: left of '->end' must point to struct/
> union
> strstubs.c(462) : error C2275: 'mlsize_t' : illegal use of this type
> as an expre
> ssion
> c:\cygwin\home\andres\ocaml-3.11.0+rc1\byterun
> \mlvalues.h(58) : see decl
> aration of 'mlsize_t'
> strstubs.c(462) : error C2146: syntax error : missing ';' before
> identifier 'sta
> rt'
> strstubs.c(462) : error C2065: 'start' : undeclared identifier
> strstubs.c(462) : error C2065: 'end' : undeclared identifier
> strstubs.c(462) : error C2065: 'len' : undeclared identifier
> strstubs.c(463) : error C2143: syntax error : missing ';' before
> 'type'
> strstubs.c(464) : error C2143: syntax error : missing ';' before
> 'type'
> strstubs.c(467) : error C2065: 'p' : undeclared identifier
> strstubs.c(467) : warning C4047: '=' : 'int' differs in levels of
> indirection fr
> om 'char *'
> strstubs.c(470) : error C2065: 'c' : undeclared identifier
> strstubs.c(470) : error C2100: illegal indirection
> strstubs.c(475) : error C2100: illegal indirection
> strstubs.c(496) : warning C4047: '=' : 'int' differs in levels of
> indirection fr
> om 'char *'
> strstubs.c(497) : error C2065: 'q' : undeclared identifier
> strstubs.c(497) : warning C4047: '=' : 'int' differs in levels of
> indirection fr
> om 'char *'
> strstubs.c(500) : error C2100: illegal indirection
> strstubs.c(502) : error C2100: illegal indirection
> strstubs.c(502) : error C2106: '=' : left operand must be l-value
> strstubs.c(504) : error C2100: illegal indirection
> strstubs.c(507) : error C2100: illegal indirection
> strstubs.c(507) : error C2106: '=' : left operand must be l-value
> strstubs.c(514) : warning C4022: 'memmove' : pointer mismatch for
> actual paramet
> er 1
> strstubs.c(518) : error C2100: illegal indirection
> strstubs.c(518) : error C2106: '=' : left operand must be l-value
> strstubs.c(518) : error C2100: illegal indirection
> strstubs.c(518) : error C2106: '=' : left operand must be l-value
> make[1]: *** [strstubs.obj] Error 2
> make[1]: Leaving directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> str'
> make[1]: Entering directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> num'
> make[1]: Nothing to be done for `all'.
> make[1]: Leaving directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> num'
> make[1]: Entering directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> win32graph'
>
> cl /nologo -D_CRT_SECURE_NO_DEPRECATE /I 'C:/Program Files/Microsoft
> Platform SD
> K/Include/crt' /I 'C:/Program Files/Microsoft Platform SDK/
> Include' /Ox /MD -I.
> ./../byterun -c draw.c
> draw.c
> draw.c(555) : error C2146: syntax error : missing ';' before
> identifier 'img'
> draw.c(618) : error C2146: syntax error : missing ';' before
> identifier 'matrix'
>
> make[1]: *** [draw.obj] Error 2
> make[1]: Leaving directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> win32graph'
> make[1]: Entering directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> dynlink'
> make[1]: Nothing to be done for `all'.
> make[1]: Leaving directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> dynlink'
> make[1]: Entering directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> bigarray'
> cl /nologo -D_CRT_SECURE_NO_DEPRECATE /I 'C:/Program Files/Microsoft
> Platform SD
> K/Include/crt' /I 'C:/Program Files/Microsoft Platform SDK/
> Include' /Ox /MD -I.
> ./../byterun -I../win32unix -DIN_OCAML_BIGARRAY -DCAML_NAME_SPACE -
> c bigarray_s
> tubs.c
> bigarray_stubs.c
> bigarray_stubs.c(840) : warning C4047: '=' : 'caml_ba_proxy *'
> differs in levels
> of indirection from 'int'
> bigarray_stubs.c(856) : error C2065: 'res' : undeclared identifier
> bigarray_stubs.c(857) : error C2275: 'intnat' : illegal use of this
> type as an e
> xpression
> c:\cygwin\home\andres\ocaml-3.11.0+rc1\byterun
> \config.h(47) : see declar
> ation of 'intnat'
> bigarray_stubs.c(857) : error C2146: syntax error : missing ';'
> before identifie
> r 'index'
> bigarray_stubs.c(857) : error C2065: 'index' : undeclared identifier
> bigarray_stubs.c(857) : error C2109: subscript requires array or
> pointer type
> bigarray_stubs.c(858) : error C2143: syntax error : missing ';'
> before 'type'
> bigarray_stubs.c(859) : error C2275: 'intnat' : illegal use of this
> type as an e
> xpression
> c:\cygwin\home\andres\ocaml-3.11.0+rc1\byterun
> \config.h(47) : see declar
> ation of 'intnat'
> bigarray_stubs.c(859) : error C2146: syntax error : missing ';'
> before identifie
> r 'offset'
> bigarray_stubs.c(859) : error C2065: 'offset' : undeclared identifier
> bigarray_stubs.c(860) : error C2275: 'intnat' : illegal use of this
> type as an e
> xpression
> c:\cygwin\home\andres\ocaml-3.11.0+rc1\byterun
> \config.h(47) : see declar
> ation of 'intnat'
> bigarray_stubs.c(860) : error C2065: 'sub_dims' : undeclared
> identifier
> bigarray_stubs.c(861) : error C2143: syntax error : missing ';'
> before 'type'
> bigarray_stubs.c(864) : error C2065: 'num_inds' : undeclared
> identifier
> bigarray_stubs.c(870) : error C2065: 'i' : undeclared identifier
> bigarray_stubs.c(870) : error C2109: subscript requires array or
> pointer type
> bigarray_stubs.c(871) : error C2109: subscript requires array or
> pointer type
> bigarray_stubs.c(872) : warning C4047: 'function' : 'intnat *'
> differs in levels
> of indirection from 'int'
> bigarray_stubs.c(872) : warning C4024: 'caml_ba_offset' : different
> types for fo
> rmal and actual parameter 2
> bigarray_stubs.c(873) : warning C4047: '=' : 'int' differs in levels
> of indirect
> ion from 'intnat *'
> bigarray_stubs.c(877) : error C2109: subscript requires array or
> pointer type
> bigarray_stubs.c(878) : error C2109: subscript requires array or
> pointer type
> bigarray_stubs.c(879) : warning C4047: 'function' : 'intnat *'
> differs in levels
> of indirection from 'int'
> bigarray_stubs.c(879) : warning C4024: 'caml_ba_offset' : different
> types for fo
> rmal and actual parameter 2
> bigarray_stubs.c(880) : warning C4047: '=' : 'int' differs in levels
> of indirect
> ion from 'intnat *'
> bigarray_stubs.c(882) : error C2065: 'sub_data' : undeclared
> identifier
> bigarray_stubs.c(884) : warning C4047: '=' : 'int' differs in levels
> of indirect
> ion from 'char *'
> bigarray_stubs.c(886) : warning C4022: 'caml_ba_alloc' : pointer
> mismatch for ac
> tual parameter 3
> bigarray_stubs.c(886) : warning C4047: 'function' : 'intnat *'
> differs in levels
> of indirection from 'int'
> bigarray_stubs.c(886) : warning C4024: 'caml_ba_alloc' : different
> types for for
> mal and actual parameter 4
> bigarray_stubs.c(902) : error C2275: 'intnat' : illegal use of this
> type as an e
> xpression
> c:\cygwin\home\andres\ocaml-3.11.0+rc1\byterun
> \config.h(47) : see declar
> ation of 'intnat'
> bigarray_stubs.c(902) : error C2146: syntax error : missing ';'
> before identifie
> r 'ofs'
> bigarray_stubs.c(902) : error C2065: 'ofs' : undeclared identifier
> bigarray_stubs.c(903) : error C2275: 'intnat' : illegal use of this
> type as an e
> xpression
> c:\cygwin\home\andres\ocaml-3.11.0+rc1\byterun
> \config.h(47) : see declar
> ation of 'intnat'
> bigarray_stubs.c(903) : error C2146: syntax error : missing ';'
> before identifie
> r 'len'
> bigarray_stubs.c(903) : error C2065: 'len' : undeclared identifier
> bigarray_stubs.c(904) : error C2143: syntax error : missing ';'
> before 'type'
> bigarray_stubs.c(905) : error C2275: 'intnat' : illegal use of this
> type as an e
> xpression
> c:\cygwin\home\andres\ocaml-3.11.0+rc1\byterun
> \config.h(47) : see declar
> ation of 'intnat'
> bigarray_stubs.c(905) : error C2146: syntax error : missing ';'
> before identifie
> r 'mul'
> bigarray_stubs.c(905) : error C2065: 'mul' : undeclared identifier
> bigarray_stubs.c(906) : error C2143: syntax error : missing ';'
> before 'type'
> bigarray_stubs.c(913) : error C2065: 'changed_dim' : undeclared
> identifier
> bigarray_stubs.c(925) : warning C4047: '=' : 'int' differs in levels
> of indirect
> ion from 'char *'
> bigarray_stubs.c(927) : warning C4022: 'caml_ba_alloc' : pointer
> mismatch for ac
> tual parameter 3
> bigarray_stubs.c(1049) : error C2275: 'intnat' : illegal use of this
> type as an
> expression
> c:\cygwin\home\andres\ocaml-3.11.0+rc1\byterun
> \config.h(47) : see declar
> ation of 'intnat'
> bigarray_stubs.c(1049) : error C2146: syntax error : missing ';'
> before identifi
> er 'dim'
> bigarray_stubs.c(1049) : error C2065: 'dim' : undeclared identifier
> bigarray_stubs.c(1049) : error C2109: subscript requires array or
> pointer type
> bigarray_stubs.c(1050) : error C2275: 'mlsize_t' : illegal use of
> this type as a
> n expression
> ../../byterun\mlvalues.h(58) : see declaration of 'mlsize_t'
> bigarray_stubs.c(1050) : error C2146: syntax error : missing ';'
> before identifi
> er 'num_dims'
> bigarray_stubs.c(1050) : error C2065: 'num_dims' : undeclared
> identifier
> bigarray_stubs.c(1051) : error C2275: 'uintnat' : illegal use of
> this type as an
> expression
> c:\cygwin\home\andres\ocaml-3.11.0+rc1\byterun
> \config.h(48) : see declar
> ation of 'uintnat'
> bigarray_stubs.c(1051) : error C2146: syntax error : missing ';'
> before identifi
> er 'num_elts'
> bigarray_stubs.c(1051) : error C2065: 'num_elts' : undeclared
> identifier
> bigarray_stubs.c(1052) : error C2143: syntax error : missing ';'
> before 'type'
> bigarray_stubs.c(1059) : error C2109: subscript requires array or
> pointer type
> bigarray_stubs.c(1060) : error C2109: subscript requires array or
> pointer type
> bigarray_stubs.c(1060) : error C2109: subscript requires array or
> pointer type
> bigarray_stubs.c(1062) : error C2109: subscript requires array or
> pointer type
> bigarray_stubs.c(1068) : warning C4047: 'function' : 'intnat *'
> differs in level
> s of indirection from 'int'
> bigarray_stubs.c(1068) : warning C4024: 'caml_ba_alloc' : different
> types for fo
> rmal and actual parameter 4
> make[1]: *** [bigarray_stubs.obj] Error 2
> make[1]: Leaving directory `/home/andres/ocaml-3.11.0+rc1/otherlibs/
> bigarray'
> make: *** [otherlibraries] Error 2
>
> andres@WIN-X85PMN663RN ~/ocaml-3.11.0+rc1
> $ this fix.

Andres Varon

unread,
Nov 26, 2008, 1:11:58 PM11/26/08
to Andres Varon, caml users, Xavier Leroy

On Nov 26, 2008, at 12:38 PM, Andres Varon wrote:

>
> On Nov 26, 2008, at 12:05 PM, Xavier Leroy wrote:
>
>>> So far our tests have compiled and run successfully in most
>>> platforms
>>> [1] excepting Windows 64 bits (but it's the first time I attempt to
>>> compile in this architecture). I am getting this error message:
>>
>> There are two problems with Windows 64 that I forgot to fix before
>> this RC1 release; this is one of them. A patch is available at
>>
>> http://gallium.inria.fr/~xleroy/ocaml-3.11rc1-win64.patch
>>
>
> Thanks. This time the compilation went further, but failed with the
> errors below.

Answering myself, the problem is my own specification of the cl
command line. Shouldn't have added those /I options. Clearly I'm not
comfortable in this environment ...

best,

Andres

Andres Varon

unread,
Nov 26, 2008, 1:43:21 PM11/26/08
to Xavier Leroy, caml users

On Nov 26, 2008, at 12:05 PM, Xavier Leroy wrote:

>> So far our tests have compiled and run successfully in most platforms
>> [1] excepting Windows 64 bits (but it's the first time I attempt to
>> compile in this architecture). I am getting this error message:
>
> There are two problems with Windows 64 that I forgot to fix before
> this RC1 release; this is one of them. A patch is available at
>
> http://gallium.inria.fr/~xleroy/ocaml-3.11rc1-win64.patch
>
> The CVS release branch contains this fix.

The patch was successful and OCaml has compiled cleanly.

best,

Andres

Stéphane Glondu

unread,
Nov 26, 2008, 5:29:16 PM11/26/08
to Richard Jones, Damien Doligez, caml users
Richard Jones wrote:
> Two camlp4 packages so far have given this error message, which is new
> since I tried 3.11.0+beta1:

You mean two packages using camlp4, don't you?

> File "_none_", line 1, characters 0-1:
> Error: Error while linking /usr/lib/ocaml/camlp4/camlp4lib.cma(Camlp4):
> Reference to undefined global `Dynlink'

That's probably a fix to be done in the packages using camlp4. See:

http://caml.inria.fr/mantis/view.php?id=4568

Maybe it should be documented as a "can possibly break existing code
(actually, Makefiles)" in the Changes file...?


Cheers,

--
Stéphane

Stefano Zacchiroli

unread,
Nov 27, 2008, 4:07:42 AM11/27/08
to caml users
On Tue, Nov 25, 2008 at 06:09:28PM +0100, Damien Doligez wrote:
> If there are no show-stoppers in this RC, then 3.11.0 will be
> officially released next week.

It is definitely no showstopper, but what about the request which has
circulated on this list to add .cmxs default target to ocamlbuild?

Are we still in time for that?

Cheers.

--
Stefano Zacchiroli -o- PhD in Computer Science \ PostDoc @ Univ. Paris 7
zack@{upsilon.cc,pps.jussieu.fr,debian.org} -<>- http://upsilon.cc/zack/
Dietro un grande uomo c'è ..| . |. Et ne m'en veux pas si je te tutoie
sempre uno zaino ...........| ..: |.... Je dis tu à tous ceux que j'aime

Richard Jones

unread,
Nov 27, 2008, 6:55:20 AM11/27/08
to Stéphane Glondu, Damien Doligez, caml users
On Wed, Nov 26, 2008 at 11:29:04PM +0100, Stéphane Glondu wrote:
> Richard Jones wrote:
> > Two camlp4 packages so far have given this error message, which is new
> > since I tried 3.11.0+beta1:
>
> You mean two packages using camlp4, don't you?

As it turns out, many packages using camlp4.

> > File "_none_", line 1, characters 0-1:
> > Error: Error while linking /usr/lib/ocaml/camlp4/camlp4lib.cma(Camlp4):
> > Reference to undefined global `Dynlink'
>
> That's probably a fix to be done in the packages using camlp4. See:
>
> http://caml.inria.fr/mantis/view.php?id=4568
>
> Maybe it should be documented as a "can possibly break existing code
> (actually, Makefiles)" in the Changes file...?

So as I understand this, every package which uses camlp4lib.cma has to
be changed to add 'dynlink.cma' before camlp4lib.cma?

Rich.

--
Richard Jones
Red Hat

_______________________________________________

Stéphane Glondu

unread,
Nov 27, 2008, 10:09:55 AM11/27/08
to Richard Jones, Damien Doligez, caml users
Richard Jones a écrit :

> So as I understand this, every package which uses camlp4lib.cma has to
> be changed to add 'dynlink.cma' before camlp4lib.cma?

Yes. For Findlib-aware build systems, the proper fix would be to make
camlp4 depend on dynlink.

--
Stéphane

Serge Aleynikov

unread,
Nov 27, 2008, 5:56:10 PM11/27/08
to Xavier Leroy, caml users
Is there a document listing new features / bug fixes between 3.10 and 3.11?

Serge

David Allsopp

unread,
Nov 27, 2008, 6:07:30 PM11/27/08
to Serge Aleynikov, caml users
Changes in the root of the tarball...


David

Andres Varon

unread,
Nov 28, 2008, 10:50:30 AM11/28/08
to Damien Doligez, caml users
Hello,

On Nov 25, 2008, at 12:09 PM, Damien Doligez wrote:

> Dear OCaml users,
>
> We are closing in on version 3.11.0. A Release Candidate is now
> available.
> If there are no show-stoppers in this RC, then 3.11.0 will be
> officially
> released next week.
>
> The release candidate is available here:
> < http://caml.inria.fr/pub/distrib/ocaml-3.11/ >
> (look for 3.11.0+rc1)

I have a question about the behavior of camlp4orf (and family) in
3.11.0. Now with the availability of native dynlink, it seems like
camlp4orf and camlp4orf.opt can load external modules inside the core.
If one looks at their respective help:

<file>.(cmo|cma) Load this module inside the Camlp4 core
<file>.cmx Load this module inside the Camlp4 core

This is a feature I've been looking for to speedup compilation times.
However, if I try the native version, here is what I get:

samson:src andres$ camlp4orf.opt poyExtension.cmx
Error: don't know what to do with poyExtension.cmx
Use option -help for usage

Is the usage different for the camlp4*.opt preprocessors?

best,

Andres

Nicolas Pouillard

unread,
Nov 28, 2008, 11:08:43 AM11/28/08
to Andres Varon, Damien Doligez, caml users
Excerpts from Andres Varon's message of Fri Nov 28 16:50:14 +0100 2008:

> Hello,
>
> On Nov 25, 2008, at 12:09 PM, Damien Doligez wrote:
>
> > Dear OCaml users,
> >
> > We are closing in on version 3.11.0. A Release Candidate is now
> > available.
> > If there are no show-stoppers in this RC, then 3.11.0 will be
> > officially
> > released next week.
> >
> > The release candidate is available here:
> > < http://caml.inria.fr/pub/distrib/ocaml-3.11/ >
> > (look for 3.11.0+rc1)
>
> I have a question about the behavior of camlp4orf (and family) in
> 3.11.0. Now with the availability of native dynlink, it seems like
> camlp4orf and camlp4orf.opt can load external modules inside the core.
> If one looks at their respective help:
>
> <file>.(cmo|cma) Load this module inside the Camlp4 core
> <file>.cmx Load this module inside the Camlp4 core
>
> This is a feature I've been looking for to speedup compilation times.
> However, if I try the native version, here is what I get:
>
> samson:src andres$ camlp4orf.opt poyExtension.cmx
> Error: don't know what to do with poyExtension.cmx
> Use option -help for usage
>
> Is the usage different for the camlp4*.opt preprocessors?

You should try with .cmxs files.

--
Nicolas Pouillard aka Ertai

Andres Varon

unread,
Nov 28, 2008, 11:30:36 AM11/28/08
to Nicolas Pouillard, Damien Doligez, caml users

Oh, of course! Thanks, I was blinded with the help message, maybe it
can be made explicit on it?

Andres

fo...@x9c.fr

unread,
Nov 29, 2008, 5:47:28 PM11/29/08
to caml users

Le 25 nov. 08 à 18:09, Damien Doligez a écrit :

> Dear OCaml users,
>
> We are closing in on version 3.11.0. A Release Candidate is now
> available.
> If there are no show-stoppers in this RC, then 3.11.0 will be
> officially
> released next week.
>
> The release candidate is available here:
> < http://caml.inria.fr/pub/distrib/ocaml-3.11/ >
> (look for 3.11.0+rc1)
>
> As usual, we need a few brave souls to try and install it on their
> favourite architecture and report the result to me.

Definitely not a show-stopper but I am quite puzzled by the semantics
of the newly-introduced "Printexc.get_backtrace" function (and hence
also by "Printexc.print_backtrace").

Things are not pleasant to me regarding programs involving multiple
threads.
By using "Printexc.get_backtrace" one may get an exception raised in
another thread. In the following program, there is no guarantee that the
two "print" expressions will output coherent results: the backtrace
printed
may be the one of an exception thrown by another thread.

try
(...)
with e ->
print_endline (Printexc.to_string e);
print_endline (Printexc.get_backtrace ())


Of course, coherence could be ensured by using a lock/mutex but I find
the
solution heavyweight (and even in some cases tricky because of existing
synchronization). Another option (involing changes to the OCaml runtime)
would be to attach each backtrace to either its exception or to its
thread.


Am I wrong ? Is the problem of no practical importance ?
Any comment or correction is very welcome to enlighten me.


Xavier Clerc

Christophe TROESTLER

unread,
Nov 30, 2008, 6:48:30 AM11/30/08
to Xavier LEROY, OCaml Mailing List
Hi,

On Wed, 26 Nov 2008 11:32:07 +0100, Xavier Leroy wrote:
>
> The documentation for 3.11 is also available from the same place. The
> HTML manual can be browsed online at
>
> http://caml.inria.fr/pub/docs/manual-ocaml-311/index.html

Maybe it is a good opportunity to replace the CAMLprim of chapter 18 by
CAMLexport (even though they are supposed not to be needed anymore IIRC).

Thanks for your work,
ChriS

Jacques Carette

unread,
Dec 2, 2008, 5:28:00 PM12/2/08
to caml users
In the CYGWIN32, SUPPORT_DYNAMIC_LINKING version of 3.11.0+rc1, file
unix.c has the following lines:

178: return flexdll_dlopen(libname, flags);
193: return flexdll_dlsym(flexdll_dlopen(NULL,0,1), name);

My version of gcc is extremely unhappy about line 193 [and says so
loudly by quitting], since it corresponds to a call to flexdll_dlopen
with an incorrect number of parameters.

Jacques

0 new messages