www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - Failing DMD tests

reply =?ISO-8859-1?Q?Alex_R=F8nne_Petersen?= <xtzgzorex gmail.com> writes:
Hi,

I recently set up a Jenkins server which is running DMD builds and tests 
at: http://emoba.net:8888/jenkins/view/D/ It's running on an x86-64 
installation of Gentoo.

It seems that there is a test failing on both x86 and x86-64 builds of 
DMD; see: http://emoba.net:8888/jenkins/view/D/job/DMD-x64/1/console

Does anyone know why this might be happening? Is it a bug?

FYI, the shell commands used to compile:

 cd src;
 make -f posix.mak clean;
 make -f posix.mak MODEL=64;
 cd ../test;
 make clean;
 make;

Regards, Alex
Aug 20 2011
parent reply Jesse Phillips <jessekphillips+d gmail.com> writes:
On Sat, 20 Aug 2011 16:57:38 +0200, Alex Rønne Petersen wrote:

 Hi,
 
 I recently set up a Jenkins server which is running DMD builds and tests
 at: http://emoba.net:8888/jenkins/view/D/ It's running on an x86-64
 installation of Gentoo.

This tests are failing as evidence by: http://d.puremagic.com/test-results/ So the devs will need to work out the solution.
Aug 20 2011
parent reply Jonathan M Davis <jmdavisProg gmx.com> writes:
On Sunday, August 21, 2011 00:27:03 Jesse Phillips wrote:
 On Sat, 20 Aug 2011 16:57:38 +0200, Alex R=C3=B8nne Petersen wrote:
 Hi,
=20
 I recently set up a Jenkins server which is running DMD builds and =


 at: http://emoba.net:8888/jenkins/view/D/ It's running on an x86-64=


 installation of Gentoo.

This tests are failing as evidence by: =20 http://d.puremagic.com/test-results/ =20 So the devs will need to work out the solution.

It's a known issue. The deprecated std.random.rand was removed, and whe= n fixing=20 the dmd tests accordingly, it was revealed that a test which used rand = was=20 actually defective and that the optimizer bug that it was supposed to t= est for=20 wasn't actually fixed: http://www.digitalmars.com/d/archives/digitalmars/D/bugs/4766.html So, the optimizer bug needs to be properly fixed and the test adjusted=20= accordingly. Until then, the test is failing. - Jonathan M Davis
Aug 20 2011
parent =?UTF-8?B?QWxleCBSw7hubmUgUGV0ZXJzZW4=?= <xtzgzorex gmail.com> writes:
On 21-08-2011 02:35, Jonathan M Davis wrote:
 On Sunday, August 21, 2011 00:27:03 Jesse Phillips wrote:
 On Sat, 20 Aug 2011 16:57:38 +0200, Alex Rønne Petersen wrote:
 Hi,

 I recently set up a Jenkins server which is running DMD builds and tests
 at: http://emoba.net:8888/jenkins/view/D/ It's running on an x86-64
 installation of Gentoo.

This tests are failing as evidence by: http://d.puremagic.com/test-results/ So the devs will need to work out the solution.

It's a known issue. The deprecated std.random.rand was removed, and when fixing the dmd tests accordingly, it was revealed that a test which used rand was actually defective and that the optimizer bug that it was supposed to test for wasn't actually fixed: http://www.digitalmars.com/d/archives/digitalmars/D/bugs/4766.html So, the optimizer bug needs to be properly fixed and the test adjusted accordingly. Until then, the test is failing. - Jonathan M Davis

OK, thanks for the clarification! Regards, Alex
Aug 21 2011