• Johan Engelen (6/6) Mar 20 2016 The merge-2.070 branch is almost green on Travis:
• kinke (16/22) Mar 20 2016 Nice! So only runnable/evalorder.d fails [for 32-bit only -
• kinke (3/5) Mar 20 2016 By just deleting those 2 declarations in
• Johan Engelen (6/11) Mar 30 2016 kinke's PR was accepted into druntime to fix the double
• Kai Nacke (8/14) Mar 20 2016 From todays buildbot:

Johan Engelen <j j.nl> writes:

The merge-2.070 branch is almost green on Travis:
https://travis-ci.org/ldc-developers/ldc/builds/117300930

Now Windows needs some love (I have not looked at it at all):
https://ci.appveyor.com/project/kinke/ldc/build/1.0.1097/job/2fbe0xd3f4f1iu6s

cheers,
   Johan

kinke <noone nowhere.com> writes:

On Sunday, 20 March 2016 at 20:47:50 UTC, Johan Engelen wrote:
 The merge-2.070 branch is almost green on Travis:
 https://travis-ci.org/ldc-developers/ldc/builds/117300930

 Now Windows needs some love (I have not looked at it at all):
 https://ci.appveyor.com/project/kinke/ldc/build/1.0.1097/job/2fbe0xd3f4f1iu6s

 cheers,
   Johan

Nice! So only runnable/evalorder.d fails [for 32-bit only - 
that's interesting].
I just got the unittests (debug+release) to work on my Win64 box 
- `core.thread` needs a little mergefix, then there's this in 
`core.sys.windows.winuser`, extern(Windows):

BOOL GetMonitorInfoA(HMONITOR, LPMONITORINFO);
BOOL GetMonitorInfoA(HMONITOR, LPMONITORINFOEXA);
BOOL GetMonitorInfoW(HMONITOR, LPMONITORINFO);
BOOL GetMonitorInfoW(HMONITOR, LPMONITORINFOEXW);

LDC complains about `Error: Function type does not match 
previously declared function with the same mangled name: 
GetMonitorInfoA`. I tend to agree - no overloading for C 
functions, right? But dmd apparently swallows it. Note that the 
actual GetMonitorInfoA() WinAPI does support a pointer to both 
structs (detecting the actual type by a common size field).

kinke <noone nowhere.com> writes:

 BOOL GetMonitorInfoA(HMONITOR, LPMONITORINFOEXA);
 BOOL GetMonitorInfoW(HMONITOR, LPMONITORINFOEXW);

By just deleting those 2 declarations in 
core.sys.windows.winuser, all tests pass on Win64, incl. 
dmd-testsuite. :)

Johan Engelen <j j.nl> writes:

On Monday, 21 March 2016 at 01:47:38 UTC, kinke wrote:
 BOOL GetMonitorInfoA(HMONITOR, LPMONITORINFOEXA);
 BOOL GetMonitorInfoW(HMONITOR, LPMONITORINFOEXW);

 By just deleting those 2 declarations in 
 core.sys.windows.winuser, all tests pass on Win64, incl. 
 dmd-testsuite. :)

kinke's PR was accepted into druntime to fix the double 
definition problem.

With that change, merge-2.070 is green on Travis, and on AppVeyor 
it has the same issue that master has (std.math, line 4370, 
worked on here: https://github.com/ldc-developers/ldc/pull/1317).

Kai Nacke <kai redstar.de> writes:

On Sunday, 20 March 2016 at 20:47:50 UTC, Johan Engelen wrote:
 The merge-2.070 branch is almost green on Travis:
 https://travis-ci.org/ldc-developers/ldc/builds/117300930

 Now Windows needs some love (I have not looked at it at all):
 https://ci.appveyor.com/project/kinke/ldc/build/1.0.1097/job/2fbe0xd3f4f1iu6s

 cheers,
   Johan

 From todays buildbot:

LDC - the LLVM D compiler (fc93fc):
   based on DMD v2.070.2 and LLVM 3.7.1
   Default target: armv7a-hardfloat-linux-gnueabi
   Host CPU: krait
   http://dlang.org - http://wiki.dlang.org/LDC

:-)