Good. I'm not surprised, but glad to have confirmation. See bug 13821
that I filed earlier. It references druntime pull 150 where there was
code added specifically for win 2008 support. See the two pull requests
I created a little while ago. I'll bet good money that applying them in
your vm will result in passing tests (obviously just hiding the
underlying issue).
On 12/4/2014 8:08 PM, Daniel Murphy wrote:
Looks like I can reproduce the crash (or at least a crash) in the
druntime unittests running on a 2012 vm.
On Fri, Dec 5, 2014 at 8:57 AM, Brad Roberts via dmd-internals
<[email protected]> wrote:
Ok, the two tests that fail are the fiber + exception tests:
https://github.com/D-Programming-Language/druntime/blob/master/src/core/thread.d#L4897
https://github.com/D-Programming-Language/druntime/blob/master/src/core/thread.d#L4943
On 12/4/2014 12:48 PM, Brad Roberts via dmd-internals wrote:
Does anyone else build dmd 32 bit environment on windows server 2012?
I've been having trouble doing so and it's blocking my efforts to help
get ddmd moving forward on windows.
The problem I'm seeing is that the druntime unittest fails. Digging
into them it looks like it's the core.thread tests. My windows
debugging experience is close to 0, and windbg is.. interesting.
Before I spend more time, is anyone else running successfully in the
same environment? Or even better, can anyone else duplicate my results?
The details of my env:
ec2 c1.medium windows instance in a vpc
running ami-a13d6891 (Windows_Server_2012-R2 ... 2014.11.19)
dm857c.zip -- from downloads.dlang.org/other/
dmd, druntime, phobos from tip of master on Nov 26
(the tree was nice and green)
It shouldn't matter, but for completeness, I'm in a cygwin environment,
but with dmc's toolchain at the head of the path and properly setup in
sc.ini.
The win/32 tester is a windows 2008 server. The win/64 tester is
windows 2012. I see this problem on both the existing win/64 tester
(started with a much older ami but current with windows updates so
should be very similar to the above ami) and a brand new instance spun
up specifically to dig into this problem (with exactly the above ami).
I'm going to keep digging, but some confirmation of success or failure
by anyone else would be useful info.
Thanks,
Brad
_______________________________________________
dmd-internals mailing list
[email protected]
http://lists.puremagic.com/mailman/listinfo/dmd-internals
_______________________________________________
dmd-internals mailing list
[email protected]
http://lists.puremagic.com/mailman/listinfo/dmd-internals
_______________________________________________
dmd-internals mailing list
[email protected]
http://lists.puremagic.com/mailman/listinfo/dmd-internals
