Answer from Rom:

This probably means that some update was applied to the 32-bit virtual machine that changed the build environment on Ubuntu but hasn’t yet been applied to the 64-bit Ubuntu environment. Probably the next time I sign-on to the system it‘ll be applied.

I’ll have to revisit this issue after I have completed some work-items for the Progress Thru Processors BOINC client.


----- Rom

In the mean time the Linux versions of 6.6.36 have been demoted to development versions. 6.4.5 is recommended version there again. ;-)

/me begins to bang head on table for wasting so much time on this last night... sighhhhh

---

In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope

1) AP tasks influencing the anticipated running times of CUDA tasks is a design flaw in BOINC which we're going to have to live with until at least BOINC v6.10

The worst effects can be mitigated by careful fine-tuning of the FLOPs figures in app_info.xml: if you get it right (carefully balanced for the performance bias of your particular hardware), you can avoid EDF almost entirely.

2) Except that then, you wouldn't have discovered the mis-behaviour of CUDA under stress! From what I'm reading, there are problems:

a) When there are two or more CUDA devices in a single host (which rules me out for testing, sadly)

b) When a combination of cache size/DCF/deadlines brings 'shorties' forward in EDF - which to my mind should be flagged as 'High Priority', but doesn't seem to be.

I think someone - not me, I'm single GPU only - is going to have to do some logging with <coproc_debug> (see Client configuration), try and work out what's happening, and report the analysis and logs to boinc_alpha.

Thanks, Richard, that sums it up nicely.

The Lunatics Unified Installer was so much fun to play with, I was hoping to avoid doing the fpops/flops process on another computer. Now that I've done the fpops/flops thing, "To completion" times have settled down, so the odds of EDF mode will decrease.

I have also decreased the CUDA cache to 2 days.

Funny, isn't it, that the answer for another BOINC-exception-case on SETI@home once again is "keep your cache under 2 or 3 days"?

Bob

In ubuntu the file libpcre.so.3 exists in the package libpcre3.

---

1) AP tasks influencing the anticipated running times of CUDA tasks is a design flaw in BOINC which we're going to have to live with until at least BOINC v6.10

The worst effects can be mitigated by careful fine-tuning of the FLOPs figures in app_info.xml: if you get it right (carefully balanced for the performance bias of your particular hardware), you can avoid EDF almost entirely.

2) Except that then, you wouldn't have discovered the mis-behaviour of CUDA under stress! From what I'm reading, there are problems:

a) When there are two or more CUDA devices in a single host (which rules me out for testing, sadly)

b) When a combination of cache size/DCF/deadlines brings 'shorties' forward in EDF - which to my mind should be flagged as 'High Priority', but doesn't seem to be.

I think someone - not me, I'm single GPU only - is going to have to do some logging with <coproc_debug> (see Client configuration), try and work out what's happening, and report the analysis and logs to boinc_alpha.

Re: 2a, what did you have in mind? I have one machine that might be suitable (an i7 with dual GTX260's). Its still running 6.6.33 at the moment.

---

BOINC blog

Just curious why on the main download page for Boinc, it shows 6.6.36 for Windows 64-bit as the recommended download? Does that now cover the x32 too?

---

Your link leads me to the main BOINC page but when I click on the download it starts the one for X86. Maybe it reads our systems somehow.

---

PROUD MEMBER OF Team Starfire World BOINC

Your link leads me to the main BOINC page but when I click on the download it starts the one for X86. Maybe it reads our systems somehow.

Yep, you're right. I was working on my 64 bit sys at the time I wrote the post, but just tried it on my 32 bit sys and it took me to the 32 bit install. I'm guessing the download link has a script that directs it to the right page for the OS.

---

For the type of browser, it seems. So it can happen on a 64bit system using a 32bit browser, that you get the 32bit BOINC version.

---

Which browser? On my 64bit sys I get the 64bit download with either version of IE.

---

I'm using Firefox on a 32 bit OS.

---

PROUD MEMBER OF Team Starfire World BOINC