Re: [Boost-testing] Huang-WinXP-x64_32: link errors

From: Sean Huang (huangsean_at_[hidden])
Date: 2006-09-20 12:06:53

----- Original Message ----- 
From: "Sean Huang" <huangsean_at_[hidden]>
To: "Running Boost regression tests" <boost-testing_at_[hidden]>
Sent: Sunday, September 17, 2006 1:14 PM
Subject: Re: [Boost-testing] Huang-WinXP-x64_32: link errors

> I checked the eventlog and found this:
>
> An unhandled win32 exception occurred in mt.exe [1700]. Just-In-Time
> debugging this exception failed with the following error: Debugger could 
> not
> be started because no user is logged on.
>
> Followed by this:
>
> The program or feature
> "\??\F:\BoostRegressions\RC_1_34_0\all_32\results\boost\bin.v2\libs\assign\test\email_example.test\intel-vc8-win-9.1\debug\email_example.exe"
> cannot start or run due to incompatibility with 64-bit versions of 
> Windows.
> Please contact the software vendor to ask if a 64-bit Windows compatible
> version is available.

I disabled the "monitored" feature of the regreession test and all these 
errors are gone so it seems that linkers are not the ones to blame.

It turned out that there were two processes that had been running well more 
than 4 hours using full cpu before I killed them manually. They are:

F:\BoostRegressions\RC_1_34_0\all_32\results\boost\bin.v2\libs\graph\test\csr_graph_test.test\intel-vc8-win-9.1\debug\csr_graph_test.exe 
hung

F:\BoostRegressions\RC_1_34_0\all_64\results\boost\bin.v2\libs\regex\test\recursion_test.test\intel-vc8-win-9.1_x86_64\debug\address-model-64\instruction-set-nocona\threading-multi\recursion_test.exe

I ran recursion_test.exe again and it finished in no time. However, 
csr_graph_test.exe either takes really long to finish or is in a dead-loop.

I'll send out another email asking either to disable csr_graph_test or 
change to release mode (is it possible?).

Just to help me under this issue better, can somebody explain to me how 
"build_monitor.exe" works?

Thanks,

Sean