Re: [Mingw-w64-public] time guard

2015-01-29 Thread Kai Tietz
This change makes sense.  Right now Eric is checking this change on
Fedora.  So we should wait for his results.  This might be something
we would like to back-merge later to the 4.x branch.
Btw Martell, the fopen_s emulation-function would be something good
IMO.  We should consider then emulation also for freopen_s, _wfopen_s,
_wfreopen_s, and fread_s.  We might want the unlocked-version there
too.

Kai

2015-01-28 22:00 GMT+01:00 Martell Malone martellmal...@gmail.com:
 This should fix the issue of windows programs assuming localtime_r and
 gmtime_r unavailable.

 if _POSIX_C_SOURCE is defined then we are specifically asking for them.

 Quote from
 http://pubs.opengroup.org/onlinepubs/007904975/functions/xsh_chap02_02.html

 A POSIX-conforming application should ensure that the feature test macro
 _POSIX_C_SOURCE is defined before inclusion of any header.

 Any posix application that does not do this needs to be patched todo this
 and not mingw-w64 to accommodate them

 --
 Dive into the World of Parallel Programming. The Go Parallel Website,
 sponsored by Intel and developed in partnership with Slashdot Media, is your
 hub for all things parallel software development, from weekly thought
 leadership blogs to news, videos, case studies, tutorials and more. Take a
 look and join the conversation now. http://goparallel.sourceforge.net/
 ___
 Mingw-w64-public mailing list
 Mingw-w64-public@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/mingw-w64-public


--
Dive into the World of Parallel Programming. The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net/
___
Mingw-w64-public mailing list
Mingw-w64-public@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mingw-w64-public


Re: [Mingw-w64-public] time guard

2015-01-29 Thread Martell Malone

 This change makes sense.  Right now Eric is checking this change on
 Fedora.  So we should wait for his results.

Yes but I think he should also build on rc1 without the patch to make sure
that if there is a breakage this is the cause.

Btw Martell, the fopen_s emulation-function would be something good
 IMO.  We should consider then emulation also for freopen_s, _wfopen_s,
 _wfreopen_s, and fread_s.  We might want the unlocked-version there
 too.

Okay in that case I'll try to prepare a patch for that sometime next week.
I want to get some dx11 changes and hopefully this inline issue fixed
before 4.x so that has my full attention atm.


On Thu, Jan 29, 2015 at 8:04 AM, Kai Tietz ktiet...@googlemail.com wrote:

 This change makes sense.  Right now Eric is checking this change on
 Fedora.  So we should wait for his results.  This might be something
 we would like to back-merge later to the 4.x branch.
 Btw Martell, the fopen_s emulation-function would be something good
 IMO.  We should consider then emulation also for freopen_s, _wfopen_s,
 _wfreopen_s, and fread_s.  We might want the unlocked-version there
 too.

 Kai

 2015-01-28 22:00 GMT+01:00 Martell Malone martellmal...@gmail.com:
  This should fix the issue of windows programs assuming localtime_r and
  gmtime_r unavailable.
 
  if _POSIX_C_SOURCE is defined then we are specifically asking for them.
 
  Quote from
 
 http://pubs.opengroup.org/onlinepubs/007904975/functions/xsh_chap02_02.html
 
  A POSIX-conforming application should ensure that the feature test macro
  _POSIX_C_SOURCE is defined before inclusion of any header.
 
  Any posix application that does not do this needs to be patched todo this
  and not mingw-w64 to accommodate them
 
 
 --
  Dive into the World of Parallel Programming. The Go Parallel Website,
  sponsored by Intel and developed in partnership with Slashdot Media, is
 your
  hub for all things parallel software development, from weekly thought
  leadership blogs to news, videos, case studies, tutorials and more. Take
 a
  look and join the conversation now. http://goparallel.sourceforge.net/
  ___
  Mingw-w64-public mailing list
  Mingw-w64-public@lists.sourceforge.net
  https://lists.sourceforge.net/lists/listinfo/mingw-w64-public
 


 --
 Dive into the World of Parallel Programming. The Go Parallel Website,
 sponsored by Intel and developed in partnership with Slashdot Media, is
 your
 hub for all things parallel software development, from weekly thought
 leadership blogs to news, videos, case studies, tutorials and more. Take a
 look and join the conversation now. http://goparallel.sourceforge.net/
 ___
 Mingw-w64-public mailing list
 Mingw-w64-public@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/mingw-w64-public

--
Dive into the World of Parallel Programming. The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net/___
Mingw-w64-public mailing list
Mingw-w64-public@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mingw-w64-public


Re: [Mingw-w64-public] time guard

2015-01-29 Thread Erik van Pienbroek
Martell Malone schreef op do 29-01-2015 om 15:23 [+]:
 This change makes sense.  Right now Eric is checking this
 change on
 Fedora.  So we should wait for his results. 
 Yes but I think he should also build on rc1 without the patch to make
 sure that if there is a breakage this is the cause.

I'm currently doing a mass rebuild test run with mingw-w64 v4.0rc1
including the patch mentioned in this thread. It takes around 24hours to
prepare and complete such a test mass rebuild. I wasn't really planning
on also doing a test mass rebuild with plain v4.0rc1 (without the
patch). However, I do was planning to introduce GCC 5 in Fedora soon
(the native gcc in Fedora was also just updated to the latest GCC 5
snapshot) and use this to do another test mass rebuild.

Regards,

Erik





--
Dive into the World of Parallel Programming. The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net/
___
Mingw-w64-public mailing list
Mingw-w64-public@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mingw-w64-public


[Mingw-w64-public] time guard

2015-01-28 Thread Martell Malone
This should fix the issue of windows programs assuming localtime_r and
gmtime_r unavailable.

if _POSIX_C_SOURCE is defined then we are specifically asking for them.

Quote from
http://pubs.opengroup.org/onlinepubs/007904975/functions/xsh_chap02_02.html

A POSIX-conforming application should ensure that the feature test macro
_POSIX_C_SOURCE is defined before inclusion of any header.

Any posix application that does not do this needs to be patched todo this
and not mingw-w64 to accommodate them


0001-Use-a-Guard-defined-by-the-spec-and-not-_POSIX.-This.patch
Description: Binary data
--
Dive into the World of Parallel Programming. The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net/___
Mingw-w64-public mailing list
Mingw-w64-public@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mingw-w64-public