It appears as though there are some more faulty wu's out there....

ps_sgr_208_3s_1_20922_1244137807

Plus others...

seems to me that:

only "3s" type are involved !

GPU goes on "compute error" but CPU works well

http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=72202438

At the moment i have only 2s running. 1s and 3s are producing complete stillstand on both of my GPUs.

ok 1s start working again, took them only a while to recover after i terminated the 3s.

Grrrr.
____________
Dragons can fly because they don't fit into pirate ships!

Same problem here

...1s ... complete stillstand

Ok just had a closer look.
The stillstand here is caused by the "3s" but if everything stood still, the "1s" and "2s" do need a complete restart of boinc to get going again.
____________
Dragons can fly because they don't fit into pirate ships!

yes - got the "3s" problem too (Gipsel 0.19e GPU App)

"1s" and "2s" ok

I thought I had a haul when it said '48 new tasks'... =P
____________

6/4/2009 7:58:26 PM|Milkyway@home|Scheduler request completed: got 5 new tasks
6/4/2009 8:07:51 PM|Milkyway@home|Scheduler request completed: got 19 new tasks
6/4/2009 8:14:51 PM|Milkyway@home|Scheduler request completed: got 21 new tasks
6/4/2009 8:16:03 PM|Milkyway@home|Scheduler request completed: got 16 new tasks
6/4/2009 8:20:46 PM|Milkyway@home|Scheduler request completed: got 16 new tasks
6/4/2009 8:24:14 PM|Milkyway@home|Scheduler request completed: got 7 new tasks

Seems like the s3 WUs are shutting down the automated machines? I'm getting only s1 and s2 at the minute after a small batch of s3
____________

..Seems like the s3 WUs are shutting down the automated machines?

Same here, wu's failing to run then causing errors.

ps_sgr_208_3s_2_9411_1244143134_0
ps_sgr_208_3s_2_9410_1244143134_0
ps_sgr_208_3s_2_9409_1244143134_0
ps_sgr_208_3s_2_9408_1244143134_0
ps_sgr_208_3s_2_9407_1244143134_0
ps_sgr_208_3s_2_9386_1244143134_0
ps_sgr_208_3s_2_9385_1244143134_0
ps_sgr_208_3s_2_9384_1244143134_0

<core_client_version>6.6.28</core_client_version>
<![CDATA[
<message>
Incorrect function. (0x1) - exit code 1 (0x1)
</message>

and, seems to be using a lot of processor power for a wu unit running on an ATI
and, once downloaded they fail to start processing, a shut down and restart of BOINC does not remedy this either.
I've started to abort these units but the next batch of downloads just sends more of the same!
____________

........... just seems to be hanging WUs with a side order of computation errors at the moment on the ps_sgr_208_xs series of WUs.

Now,

On one of my ATi clients, I have to cancel the s3s, but s1 and s2 work fine.
On another ATi client, s3 WUs come out as comp errors, but s1 and s2 process and are marked invalid.
____________

I'm getting errors on s1, s2 & s3's

I'm having to micromange at the moment as none are completeing, therefore, none are being returned.....so unless i abort them i don't get any new units.
the circle of life...............
____________

I have the samme problems. All my Ati's hang and errors out:-(
____________

I have had about 60 or so error or freeze on the 3s units.
____________

My 2 ATI machines error immediately on the 3s WUs, my CPU machine finishes these just fine.
____________

Calm Chaos Forum...Join Calm Chaos Now

Ok just had a closer look.
The stillstand here is caused by the "3s" but if everything stood still, the "1s" and "2s" do need a complete restart of boinc to get going again.

Ok just had a closer look.
The stillstand here is caused by the "3s" but if everything stood still, the "1s" and "2s" do need a complete restart of boinc to get going again.

I think this is the correct diagnosys for ATI GPU clients

if I'm lucky, I erase all "3s" WUs before BOINC starts crunching them. In this way I save "1s" and "2s" without restarting BOINC

Just checked my task manager, though i have no wu's currently downloaded or running, task manager shows : astronomy_0.19_ATI_SSE2e.exe running in the background?

I have re-installed the cpu version of the opt. app. and can confirm that these same units work as expected under the CPU and as I will be unable to micro manage throughout the night, will run the units this way until tomorrow when I will try again, hopefully it may be sorted by then.
____________

I can confirm that "3s" cause errors on my machine too. They are hanging without any output. Restarting BOINC causes "computation error". "2s" and "1s" are crunched as ever.

ATI 48700
Vista Ultimate SP1
Catalyst 9.1
astronomy_0.19_ati_SSE2e

I've had the 2s wu's validate OK on my 4870.

Well Just for a notation here, I have 14 hd 4870 that all are hung up and erroring out. Some one please push the oops button please and fix them...

Thanks..

DD,
____________

All WUs of the ....3s.... lot got errors on my GPU (HD4870). If I cancel them, the .....2s.... will not be calculated til I restart the whole PC, a restart of BOINC isn't enough. What is changed?

Sorry guys!

I just have seen this mess and but it is already midnight here. So I can't do much about it before tomorrow.
What has changed with the ps_208_s3 WUs compared to all earlier ones is that these WUs do computations on three streams (opposed to only one ore two in all previous WUs). The GPU app was supposed to handle that (Travis told me that probably three streams will coming already in february or so), but as there were no appropriate WUs it was never tested. Obviously there is a bug in the GPU app (as the CPU app is working). So I need to find the bug, fix it and distribute a new version of the GPU application.

I hope it will be ready tomorrow.

If you abort the 3s, and the 2s or 1s don't start up, rather than restart your computer:

Close BOINC
Open Task Manager
End all Astronomy sse2 processes
Reopen BOINC

The WUs start crunching again

Sorry guys!

I just have seen this mess and but it is already midnight here. So I can't do much about it before tomorrow.
What has changed with the ps_208_s3 WUs compared to all earlier ones is that these WUs do computations on three streams (opposed to only one ore two in all previous WUs). The GPU app was supposed to handle that (Travis told me that probably three streams will coming already in february or so), but as there were no appropriate WUs it was never tested. Obviously there is a bug in the GPU app (as the CPU app is working). So I need to find the bug, fix it and distribute a new version of the GPU application.

I hope it will be ready tomorrow.

The plural of CPU is CPUs
The plural of GPU is GPUs

THERE IS NO APOSTROPHE UNLESS YOU ARE DESCRIBING A FEATURE!!!

>"Bad English is the language of the science ..."
No, bad english is the language of the net. :)
____________
Cheers,

PeterV

.

<offtopic> Meh I was fed up with people and businesses putting apostrophes in CPUs, GPUs, CDs, TVs, DVDs and PCs, so decided to make a facebook fan page about Spelling, Punctuation and Apostophe Use. Not too bad getting 10k fans in a month - I'm not the only soldier in the crusade against bad English. Not bad for a computational chemist who actually dropped out of English A-Level during Wuthering Heights.

Also, science rocks my socks.
</offtopic>
____________

If you abort the 3s, and the 2s or 1s don't start up, rather than restart your computer:

Close BOINC
Open Task Manager
End all Astronomy sse2 processes
Reopen BOINC

The WUs start crunching again

Sorry guys!

I just have seen this mess and but it is already midnight here. So I can't do much about it before tomorrow.
What has changed with the ps_208_s3 WUs compared to all earlier ones is that these WUs do computations on three streams (opposed to only one ore two in all previous WUs). The GPU app was supposed to handle that (Travis told me that probably three streams will coming already in february or so), but as there were no appropriate WUs it was never tested. Obviously there is a bug in the GPU app (as the CPU app is working). So I need to find the bug, fix it and distribute a new version of the GPU application.

I hope it will be ready tomorrow.

Thanks for the update :)

I have 6 208_3s_2's now which seem to be running ok so far with ~15-30% increase in crunch time.
____________
Doesn't expecting the unexpected make the unexpected the expected?
If it makes sense, DON'T do it.

Hello

Are 3s WUs a test (either voluntary or involuntary) to encourage GPU ATI cruncheurs to wait for milky_GPU applications?

I switched yesterday to ubuntu where the CPU optimized application runs well without errors.

On Vista64 and ATI HD4870 it doesn't run correctly, there are errors... snif

Having just got home from work [GMT+10] and then spent 15mins aborting stalled 3s WUs [on 3 PCs] and then restarting boinc to get things going again, might I suggest that until the 3s problem is sorted the project stop serving up 3s WUs. Of course if the project can detect which PCs are or are not running Cluster's GPU client [individual PC racs could be a hint] then problem solved - just send 3s WUs to CPU-only PCs and 1 and 2s to the GPUs.

Someone may be ahead of me here - the last couple of batches of WUs that I got contained only 1s and 2s WUs.
____________
Cheers,

PeterV

.

............. I'm still getting the 3s in mixed batches.

I have 4 gpu running in circles for nothing because I am not close to deal with this problem computers ares far away from me, why not stop send gpu wu on cpu site? problem résolve yaiting for gpu site opening lol

Please, stop sending out 3s wu's please.

____________

Yea - stop sending "3s" WU for about 2 days - Gipsel is working on it !

I got 48new WUs at once - all "3s" - 60s later - all crashed ...

The Milkyway-Project-Team is waisting WUs ... !

Still the same ... "3s" still coming ...

Just had this 3s invalidate on my P4 cpu. So not just a gpu problem it seems.
____________

And same here. I was thinking that it was related to my instalation of Folding@home so I removed it, to bad. This project used to be greate but now it's just crap with an alpha stus that seams to go on for ever.

Hi everyone!
Í've got same problem with my ATI 4850.
With any version of aplication.

And same here. I was thinking that it was related to my instalation of Folding@home so I removed it, to bad. This project used to be greate but now it's just crap with an alpha stus that seams to go on for ever.

But why wasn't it fixed yesterday! :)
____________
Cheers,

PeterV

.

I've aborted 40-50 of the 3s units this morning to go along with the 100 or so of them last night. All GPU crunching grinds to a halt when these hang up. Please stop sending these things out until there is a fix. What happened to testing the work before sending it out to everybody?
____________

4870 GPU
4870 GPU

Oh my god, Is anybody home hear. Stop sending out 3s wu's Now!!!!!

____________

Like many others posting in to this thread I have a stalled ATI HD3850 due, I surmise, to the problem ps_sgr_208_3s_etc WUs.

Interestingly, the HD3850 seems to be getting lots of work, but cannot crunch them as the _3s_ cause the system to stall for some reason or other. When the GPU does crunch it ends up terminating as a computer error.

I am going to detach and reattach to allow Milkyway to do CPU crunching as I noticed there is no problem with these _3s_ WUs on them. I can continue this way, at 10% of the GPU RAC, until these _3s_ WUs stop.

I hope Travis or Dave can see a way to sort this problem out in this project soon. It has been at least 2 days now living (or not) with this problem.

:((
____________
Go away, I was asleep

I didn't have any problems with the six 3s_2 I got. Is it only the 3s_1 giving problems? Could be some lingering still that need canceled.
____________
Doesn't expecting the unexpected make the unexpected the expected?
If it makes sense, DON'T do it.

The current workaround is to abort all the _3s_ WUs, close and restart boinc, and hope you get some _1s_ or _2s_ WUs next time. Its horribly manual and requires keeping an eye on boinc, which explains all the requests/demands in this thread for Travis to fix it.
____________
Cheers,

PeterV

.

The current workaround is to abort all the _3s_ WUs, close and restart boinc, and hope you get some _1s_ or _2s_ WUs next time. Its horribly manual and requires keeping an eye on boinc, which explains all the requests/demands in this thread for Travis to fix it.

The current workaround is to abort all the _3s_ WUs, close and restart boinc, and hope you get some _1s_ or _2s_ WUs next time. Its horribly manual and requires keeping an eye on boinc, which explains all the requests/demands in this thread for Travis to fix it.

Come on guys, deep breaths :)
The problem isn't with the workunits, they're fine, so Travis can't fix it.
It's a bug in Cluster Physiks GPU application as he has already pointed out.
Give him a chance and he'll sort it

It's a bug in Cluster Physiks GPU application as he has already pointed out.

Hi John - agreed. However GPU users cannot choose their WUs - and once downloaded these WUs can't be run - therefore abort or leave stalled is the only option. Granted I'm sure that CP will get it sorted, but the project may be best suspending the generation of this type of WU until it is addressed ......... can't be good for the retruned data to have so may aborted WUs or project resets.

Hello MW@Home,

These *_3s_* runs are 3 stream runs that I started. I will tell Travis that there is a problem with them on the GPUs but not the CPUs and abort said run asap.

Sorry for the inconvenience,
John Vickers

Looks like the searches are stopped, we'll not do 3 stream runs until the ATI code is fixed :)

No 3s to be had, everything seems ok at present.
____________

I detached and reattached my PC, with the AGP HD3850, to run the WUs on it's CPUs.

I now have 12 _3s_2_ WUs to crunch. So, there may be a few of thse around still.

I only moved to this about an hour ago, so when this picks up new work I will look see if I only get the _1s_2's and the _2s_1s and _2s_2's first before moving back to the GPU.
____________
Go away, I was asleep

Gipsel got it fixed with version 0.19f:

http://www.file-upload.net/download-1684038/Milkyway_0.19f_ATI.zip.html


Vielen Dank Gipsel !

Unrelated to the 3s workunits, I've noticed some odd behavior since updating my BOINC client to version 6.6.31. When I get some work from here on the ATI app it sits at ready to start instead of beginning immediately as it used to. I have to pause / resume the other project in progress for it to begin working GPU tasks in parallel with other CPU project work. Re-installing the ATI app did not fix the problem. I'll watch how it behaves for a little while and may have to roll back to an earlier BOINC version. I've left the supplied app_info.xml as default and it's worked fine in the past. Perhaps it may be necessary to specify some values now.

Gipsel got it fixed with version 0.19f

Gipsel got it fixed with version 0.19f:

http://www.file-upload.net/download-1684038/Milkyway_0.19f_ATI.zip.html


Vielen Dank Gipsel !

"booger"

Gipsel got it fixed with version 0.19f:

http://www.file-upload.net/download-1684038/Milkyway_0.19f_ATI.zip.html


Vielen Dank Gipsel !

Yes! thank you.
You guys are just great and second to non in the BOINC community.
Apparently you and the guys at Stanford are the only ones that know how to make programs that can put the ATI card to good use.

Gipsel got it fixed with version 0.19f

I will make a link out of it.
Version 0.19f of the ATI GPU application
is ready for download. Besides the fix for the three stream WUs it reports now the GPU time as WU time.

It was a really absurd bug, a missing line break in the GPU assembly.

I just had an odd one. I updated to the new version. I finished ok but it got 0 credit. Here it is The odd thing it was a ps_sgr_208_2s_2 unit.
____________

4870 GPU
4870 GPU

I just had an odd one. I updated to the new version. I finished ok but it got 0 credit. Here it is The odd thing it was a ps_sgr_208_2s_2 unit.

Me either, I'm keeping an eye on it for a while to see if any others do the same. That's the first invalid unit i've had here.
____________

4870 GPU
4870 GPU

I've just had another one on a diffrent system. Here it is, I'm not sure what's happening. Both are very stable systems.
____________

4870 GPU
4870 GPU

Here's another
____________

4870 GPU
4870 GPU

Someone mentioned that the new 0.19f app runs the GPU a little harder and a little hotter. Are your fans clean?


____________

Calm Chaos Forum...Join Calm Chaos Now

Aah. Not crap anymore then?

Someone mentioned that the new 0.19f app runs the GPU a little harder and a little hotter. Are your fans clean?

Yes, the fans are clean. One is a recent build and the other was just changed over into a new case and everything was cleaned before installing. Under load the cards are reporting temps of ~75C. Neither card is overclocked. Both systems are in Antec900 cases. Plenty of airflow there :-).
I had several more return as invalid overnight. I didn't do a reboot after going to 19f yesterday at 2230. I just rebooted both systems at 1400 UTC today. I'm going to watch them for a bit and see if that makes any difference.

<edit>
Reboot made no difference. Still getting the occasional invalid result. Is anybody else having the same problem? I might suspect it was a computer issue if it wasn't happening on both systems.
____________

4870 GPU
4870 GPU

Yes, the fans are clean. One is a recent build and the other was just changed over into a new case and everything was cleaned before installing. Under load the cards are reporting temps of ~75C. Neither card is overclocked. Both systems are in Antec900 cases. Plenty of airflow there :-).
I had several more return as invalid overnight. I didn't do a reboot after going to 19f yesterday at 2230. I just rebooted both systems at 1400 UTC today. I'm going to watch them for a bit and see if that makes any difference.

<edit>
Reboot made no difference. Still getting the occasional invalid result. Is anybody else having the same problem? I might suspect it was a computer issue if it wasn't happening on both systems.

I didn't have any last night, but now I just looked and have 6 out of almost 600 results on two machines.


____________

Calm Chaos Forum...Join Calm Chaos Now

I've had a few between my two GPU boxes. But *maybe* 1 in 50 or less. Example:

Task ID 74360632
Name ps_sgr_235_2s_1_674470_1244317010_0
Workunit 73253956
Created 6 Jun 2009 19:36:53 UTC
Sent 6 Jun 2009 19:39:25 UTC
Received 6 Jun 2009 19:52:15 UTC
Server state Over
Outcome Success
Client state Done
Exit status 0 (0x0)
Computer ID 24918
Report deadline 9 Jun 2009 19:39:25 UTC
CPU time 84.34375
stderr out

<core_client_version>6.4.6</core_client_version>
<![CDATA[
<stderr_txt>
Running Milkyway@home ATI GPU application version 0.19f by Gipsel
CPU: Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz (4 cores/threads) 2.40002 GHz (259ms)

CAL Runtime: 1.3.145
Found 1 CAL device

Device 0: ATI Radeon HD 3800 (RV670) 512 MB local RAM (remote 28 MB cached + 512 MB uncached)
GPU core clock: 669 MHz, memory clock: 829 MHz
320 shader units organized in 4 SIMDs with 16 VLIW units (5-issue), wavefront size 64 threads
supporting double precision

3 WUs already running on GPU 0
No free GPU! Waiting ... 461.688 seconds.
Starting WU on GPU 0

main integral, 160 iterations
predicted runtime per iteration is 408 ms (33.3333 ms are allowed), dividing each iteration in 13 parts
borders of the domains at 0 128 248 376 496 616 744 864 984 1112 1232 1360 1480 1600
Calculated about 3.70012e+012 floatingpoint ops on GPU, 6.34181e+007 on FPU. Approximate GPU time 84.3438 seconds.

probability calculation (stars)
Calculated about 1.20373e+009 floatingpoint ops on FPU.

WU completed.
CPU time: 19.9844 seconds, GPU time: 84.3438 seconds, wall clock time: 716.234 seconds, CPU frequency: 2.40012 GHz

</stderr_txt>
]]>

Validate state Invalid
Claimed credit 0.364573240852738
Granted credit 0
application version 0.19

-----

-jim

I have had a few over 4 different 3850 cards, but am not really worrying about it...Probably over 99% have completed sucessully...
____________
Click to help Seti City.

CPUs are having invalid results as well. My old dual P3 has 1 in 4 results returned here.
____________
Go away, I was asleep

I'm still getting quite a few. Maybe around 5%, that's still alot as far as the project goes. Somebody may want to look at why it's happening. That could be quite alot of work on the grand scale that will be missing if not corrected. Insta purge isn't helping any. If they could see why the're failing it would help.
____________

4870 GPU
4870 GPU

I'm now showing 7 on my CPU machine that were not there this morning.


____________

Calm Chaos Forum...Join Calm Chaos Now

If the rare invalid result is occurring on both CPU and GPU crunched tasks, perhaps there is still a fundamental issue with some of the new work we're getting.

I see a small percentage of my work report as invalid as well.

I'm seeing approx 10% failure rate on my 4870.

It seems overnight all my 2s_2 have been marked as invalid (Have had 5 of them). All on cpu. No problems with the other searches.
____________

My work E6850 (cpu) has about a 5% to 10% error rate.

The error rate on my 3850 seems to have disappeared, but risen on the CPUs to about 5%.
____________
Go away, I was asleep

My card is not getting any errors since updating to 0.19f, but my CPU's are hardly getting any work at all so I can't comment on them. The last 2s unit that errored out resulted in 3 screens worth of debugging information in the output file. If anyone wants it I'll pass it on. Hopefully when the GPU project is able to start up all this will go away.

Just a few of the invalid wu's from my 4870

http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73584677
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73571669
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73544740
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73544745
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73571652
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73571666
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73571667
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73544733
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73501652
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73501661
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73544733
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73544740
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73544745
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73571652
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73571667
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73571666
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73571669


And a couple from my cpu

http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73422247
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=73422257

Yep I've got a few invalid Wu's too, all are 2_2's.
;-(
____________

Yep I've got a few invalid Wu's too, all are 2_2's.
;-(

Hello MW@Home,

Can you please tell me if these _2s_2 runs that were returning errors are all "ps_sgr_208_2s_2", all "ps_sgr_235_2s_2" or a mix of both?

Thank You,
John Vickers

Can you please tell me if these _2s_2 runs that were returning errors are all "ps_sgr_208_2s_2", all "ps_sgr_235_2s_2" or a mix of both?

Hello MW@Home,
Can you please tell me if these _2s_2 runs that were returning errors are all "ps_sgr_208_2s_2", all "ps_sgr_235_2s_1" or a mix of both?
Thank You,
John Vickers
Edit: typo: "ps_sgr_235_2s_2" - > "ps_sgr_235_2s_1"

These are the ones I could get before insta purge took care of them.

Host 39176 GPU
ps_sgr_208_2s_2_1637698_1244467419
ps_sgr_208_2s_2_1637697_1244467419
ps_sgr_208_2s_2_1637695_1244467419_0
ps_sgr_208_2s_2_1624691_1244465356_0
ps_sgr_208_2s_2_1615716_1244463950
ps_sgr_208_2s_2_1615702_1244463950_0

Host 60779 GPU
ps_sgr_208_2s_2_1641829_1244468066
ps_sgr_208_2s_2_1628695_1244465990
ps_sgr_208_2s_2_1628692_1244465990
ps_sgr_208_2s_2_1623219_1244465117
ps_sgr_235_2s_1_1572923_1244457187
ps_sgr_208_2s_2_308822_1244201732

Host 39247 CPU
ps_sgr_208_2s_2_1598226_1244461206
____________

4870 GPU
4870 GPU

These are the ones I could get before insta purge took care of them.

Host 39176 GPU
ps_sgr_208_2s_2_1637698_1244467419
ps_sgr_208_2s_2_1637697_1244467419
ps_sgr_208_2s_2_1637695_1244467419_0
ps_sgr_208_2s_2_1624691_1244465356_0
ps_sgr_208_2s_2_1615716_1244463950
ps_sgr_208_2s_2_1615702_1244463950_0

Host 60779 GPU
ps_sgr_208_2s_2_1641829_1244468066
ps_sgr_208_2s_2_1628695_1244465990
ps_sgr_208_2s_2_1628692_1244465990
ps_sgr_208_2s_2_1623219_1244465117
ps_sgr_235_2s_1_1572923_1244457187
ps_sgr_208_2s_2_308822_1244201732

Host 39247 CPU
ps_sgr_208_2s_2_1598226_1244461206

Looks like the searches are stopped, we'll not do 3 stream runs until the ATI code is fixed :)

I'm getting a lot of sigsev errors on 2s-4 and 2s-6.

Is this a related problem?

_3s_ WUs crunching madly with Cluster's v.0.19f.
____________
Cheers,

PeterV

.

I'm getting a lot of sigsev errors on 2s-4 and 2s-6.

Is this a related problem?

No. It's Ubuntu causing it. Get a proper linux distribution or downgrade to 8.xx.

Something caused both of my gpu crunchers to freeze up around 5:30 UTC. I say that time because that was the time the last result remaining on my systems was sent out to me.I have no idea what it was. That was 12:30 AM my time. I just got up this AM and found it. I don't see any errors but insta purge probably took the units away before I could see them. Might be something worth looking into.

<edit>
It was only milkyway that froze. Also running Prime Grid on both systems. I was still running.
<edit 2> It just froze up on one machine again. It was running ps_sgr_208_3s_6 and a ps_sgr_210_3s_5 and a ps_sgr_235_2s_6 When it froze. This is definately worth looking into.
<edit 3> It was either the 210_3 or the 235_2 that locked it up.
____________

4870 GPU
4870 GPU

I'm not so much getting a system freeze but a GPU reset! These seem to be happenong with the 3s, specifically the '3s 6', I have visually seen this happen, have'nt noticed it on '3s 5', but will be watching - difficult to catch as it has only happened on three seperate occasions in the last 36 hours.
____________

Yes, I am using the latest version 0.19f. I was able to play around with it a little more last night. It seems that absolutely everything is running high priority for some reason. I made no changes to my BOINC prefrences either. I also found that if I suspend my other project (prime grid) everthing starts back up. That will however have a negative impact on PG. None of this started happening until a recent windows update. I'm very much open to suggestions on how to correct it. I thought I might try and reinstall 19f as soon as I can get a chance in case something got messed up with the update. If that doesn't work maybe reinstalling BOINC. The two systems are running 6.4.7
____________

4870 GPU
4870 GPU

Yes, I am using the latest version 0.19f. I was able to play around with it a little more last night. It seems that absolutely everything is running high priority for some reason. I made no changes to my BOINC prefrences either. I also found that if I suspend my other project (prime grid) everthing starts back up. That will however have a negative impact on PG. None of this started happening until a recent windows update. I'm very much open to suggestions on how to correct it. I thought I might try and reinstall 19f as soon as I can get a chance in case something got messed up with the update. If that doesn't work maybe reinstalling BOINC. The two systems are running 6.4.7

I am getting the 0.19f GPU lock up as well on 2 different pc's this am. MW GPU app locks up but the SETI AP CPU keeps going without problems. Stop BOINC and start BOINC again gets the GPU app going again. I am also getting High Priority running on GPU WU's as well.

Look at this one. Note the GPU time and the wall clock time.
<EDIT>
I would definately call that a hang!

Task ID 77725727
Name ps_sgr_235_2s_6_1603847_1244722735_0
Workunit 76446733
Created 11 Jun 2009 12:18:59 UTC
Sent 11 Jun 2009 12:20:07 UTC
Received 11 Jun 2009 15:50:10 UTC
Server state Over
Outcome Success
Client state Done
Exit status 0 (0x0)
Computer ID 39176
Report deadline 14 Jun 2009 12:20:07 UTC
CPU time 11878.53
stderr out <core_client_version>6.4.7</core_client_version>
<![CDATA[
<stderr_txt>
Running Milkyway@home ATI GPU application version 0.19f by Gipsel
CPU: Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz (4 cores/threads) 3.54598 GHz (227ms)

CAL Runtime: 1.3.145
Found 1 CAL device

Device 0: ATI Radeon HD 4800 (RV770) 1024 MB local RAM (remote 28 MB cached + 1024 MB uncached)
GPU core clock: 750 MHz, memory clock: 900 MHz
800 shader units organized in 10 SIMDs with 16 VLIW units (5-issue), wavefront size 64 threads
supporting double precision

3 WUs already running on GPU 0
No free GPU! Waiting ... 93.7969 seconds.
Starting WU on GPU 0

main integral, 160 iterations
predicted runtime per iteration is 145 ms (33.3333 ms are allowed), dividing each iteration in 5 parts
borders of the domains at 0 320 640 960 1280 1600
Calculated about 3.70012e+012 floatingpoint ops on GPU, 6.34181e+007 on FPU. Approximate GPU time 11878.5 seconds.

probability calculation (stars)
Calculated about 1.20373e+009 floatingpoint ops on FPU.

WU completed.
CPU time: 1.35938 seconds, GPU time: 11878.5 seconds, wall clock time: 12032.2 seconds, CPU frequency: 3.546 GHz

</stderr_txt>
]]>

Validate state Valid
Claimed credit 82.9045335151938
Granted credit 27.75994
application version 0.19
____________

4870 GPU
4870 GPU

Imcrazy,

This happens quit a bit on hosts that are shared with other projects.

It seems that the shorter the other projects WU's the more MW hangs.

I believe it has something to do with the way BOINC handles debt.

I think you mentioned you are also crunching Prime Grid and Aqua. The shorter WU's will suspend your MW WU's until your short term, long term debt is cleared.

The new Multi Thread aqua can play havoc on the ATI app since a Aqua WU now wants to use multiple CPU's and will occasionally put MW in suspend mode.



To test this, when you see MW hung up, just suspend the other projects, MW should take off and start crunching again without having to reset or reboot your box
____________
.

Imcrazy,

This happens quit a bit on hosts that are shared with other projects.

It seems that the shorter the other projects WU's the more MW hangs.

I believe it has something to do with the way BOINC handles debt.

I think you mentioned you are also crunching Prime Grid and Aqua. The shorter WU's will suspend your MW WU's until your short term, long term debt is cleared.

The new Multi Thread aqua can play havoc on the ATI app since a Aqua WU now wants to use multiple CPU's and will occasionally put MW in suspend mode.



To test this, when you see MW hung up, just suspend the other projects, MW should take off and start crunching again without having to reset or reboot your box

I've had 3 systems hang up also...and I'm not running any other project.

Do you think the shorter WU that Travis took care of was causing this?

The only other project that I have running on that system is prime grid. After the windows update it seems like everthing PG and MW went into high priority. It did that on both quads. Last night I reset my debts to 0 on one system using BOINC DV. I'll try that on the other(the one this long winded WU came from) tonight.
____________

4870 GPU
4870 GPU

Thanks for the info! I did notice that i could suspend the Prime Grid PSP Sieve units and MW would imediately start back up and run for an extended period. Now it's happening with Prime Grid PSP LLR units (Prime Grid Challenge). Not exactly a short running task 30+ hours for each one. I'm only running 3 at a time to leave 1 core free for MW. I did start a new thread "Hanging Work Units" Please look there for any new developments or suggestions.
____________

4870 GPU
4870 GPU

Can you people using 3850 tell what settings you are using in app_info.xml? I have tried many different settings, but still get VPU Recovery events..

I am running on Win7 64bit, ATI 0.19f and BOINC 6.6.36

Everything is fine and the WU's are processed at peak efficiency IF I don't do anything on the computer. But if it is used normally (watch videos, browse web with Firefox etc) then I constantly get blank screen+VPU recovery+jammed WU that I have to either kill with task manager or restart BOINC. I don't seem to have this problem on my 48xx series cards only the 3850.

I get somewhat better functionality (still problems but less) with f60 w1.7 n1, but then GPU utilization is only 50-60%.

Hi
this might be of only little help as both my pcs where I have a 3850 are old and dedicated to crunching but anyway..
I have two AGP bus 3850 on MW, one is on a AMD Athlon and the other Intel Celeron, both running XP Home.
Both are running on standard settings except n2 as the cards are 512 mb and the motherboards have only 256 mb each.
Basic functionality (web etc) is ok even if as said these are dedicated to MW and pretty much taken back to use as I noted that I can fit a AGP 3850 into them and do some crunching