Boost Testing :
From: Rene Rivera (grafikrobot_at_[hidden])
Date: 2008-05-21 00:13:07
K. Noel Belcourt wrote:
> On May 20, 2008, at 9:21 PM, Rene Rivera wrote:
>> K. Noel Belcourt wrote:
>>> On May 20, 2008, at 2:14 PM, Rene Rivera wrote:
>>>> Boris Gubenko wrote:
>>>>> Rene Rivera wrote:
>>>>>> We can still get the 1.35.0 results. We just need volunteers to
>>>>>> run the
>>>>>> regression results against 'tags/release/1_35_0'. And if we have
>>>>>> results, I could turn around and generate the
>>>>> I can do it for HP platforms.
>>>> Ok, that would be wonderful :-) So please go ahead, all it takes, at
>>>> minimum is to do a single full run with
>>>> "--tag=tags/release/Boost_1_35_0". And of course if you don't want
>>>> it to
>>>> interrupt results from trunk runs you might also want to run it in a
>>>> separate directory than the regular results.
>>>> Any one else willing to do a 1.35.0 release test run?
>>> Hi Rene,
>>> I can do this for the Sandia results (it'll take a few days before I
>>> can get to it).
>> That's good... Especially since I failed at running the darwin ones I
>> mentioned earlier. Either BBv2 stalls, afaict, when running 2
>> Or PJL.cpp crashes, with a core dump, when I run 1 toolset :-(
> Stalls as in takes a parallel build request and serializes it or
> stalls as in stops building completely? Are you on Leopard (all my
> tests run under Tiger)? What exactly happens when it stalls?
Stalls as in when it runs bjam it eats up the usual huge amount of
memory and cpu to start up. Then when it should start doing action
commands it just pauses, both stops eating memory and stops using the
cpu, and it just sits there in idle. Oh, yeah, and it's Leopard. So an
OS change might be suspect (it's a new Mac Mini x86 2GB RAM).
-- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org (msn) - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim,yahoo,skype,efnet,gmail