Posts by skgiven

21) Message boards : GPUs : Caught SIGSEGV in ATI GPU detection (Message 39487)
Posted 6 Aug 2011 by skgiven
Post:
06/08/2011 20:25:37 Starting BOINC client version 6.10.60 for windows_x86_64
06/08/2011 20:25:37 Config: report completed tasks immediately
06/08/2011 20:25:37 Config: use all coprocessors
06/08/2011 20:25:37 Config: zero long-term debts on startup
06/08/2011 20:25:37 log flags: file_xfer, sched_ops, task, coproc_debug, cpu_sched_debug, dcf_debug
06/08/2011 20:25:37 log flags: debt_debug, rr_simulation, sched_op_debug, work_fetch_debug
06/08/2011 20:25:37 Libraries: libcurl/7.19.7 OpenSSL/0.9.8l zlib/1.2.3
06/08/2011 20:25:37 Data directory: E:\BOINC2
06/08/2011 20:25:37 Running under account Administrator
06/08/2011 20:25:37 Processor: 8 GenuineIntel Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz [Family 6 Model 42 Stepping 7]
06/08/2011 20:25:37 Processor: 256.00 KB cache
06/08/2011 20:25:37 Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 syscall nx lm vmx smx tm2 popcnt aes pbe
06/08/2011 20:25:37 OS: Microsoft Windows Server 2003 "R2": Standard Server x64 Edition, Service Pack 2, (05.02.3790.00)
06/08/2011 20:25:37 Memory: 7.98 GB physical, 23.53 GB virtual
06/08/2011 20:25:37 Disk: 931.51 GB total, 70.47 GB free
06/08/2011 20:25:37 Local time is UTC +1 hours
06/08/2011 20:25:38 No NVIDIA library found
06/08/2011 20:25:38 Caught SIGSEGV in ATI GPU detection
06/08/2011 20:25:38 No usable GPUs found
06/08/2011 20:25:38 Milkyway@home URL http://milkyway.cs.rpi.edu/milkyway/; Computer ID 278256; resource share 1

As you can see I'm going round in circles here. The drivers just won't work with Windows 2003. I'm going to pull the card and go back to an NVidia card.
22) Message boards : GPUs : Caught SIGSEGV in ATI GPU detection (Message 39484)
Posted 6 Aug 2011 by skgiven
Post:
Well as you thought Boinc 6.10.60 made no difference. Installed 11-7_vista64_win7_64_dd_ccc_ocl.exe again (after trying 11-7_xp64_dd_ccc_ocl.exe and a slightly earlier version), but made no difference.

I will try an earlier driver - already tried http://sites.amd.com/us/game/downloads/Pages/radeon_win7-64.aspx

06/08/2011 16:39:32 Starting BOINC client version 6.10.60 for windows_x86_64
06/08/2011 16:39:32 log flags: file_xfer, sched_ops, task
06/08/2011 16:39:32 Libraries: libcurl/7.19.7 OpenSSL/0.9.8l zlib/1.2.3
06/08/2011 16:39:32 Data directory: E:\BOINC2
06/08/2011 16:39:32 Running under account Administrator
06/08/2011 16:39:32 Processor: 8 GenuineIntel Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz [Family 6 Model 42 Stepping 7]
06/08/2011 16:39:32 Processor: 256.00 KB cache
06/08/2011 16:39:32 Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 syscall nx lm vmx smx tm2 popcnt aes pbe
06/08/2011 16:39:32 OS: Microsoft Windows Server 2003 "R2": Standard Server x64 Edition, Service Pack 2, (05.02.3790.00)
06/08/2011 16:39:32 Memory: 7.98 GB physical, 23.53 GB virtual
06/08/2011 16:39:32 Disk: 931.51 GB total, 70.69 GB free
06/08/2011 16:39:32 Local time is UTC +1 hours
06/08/2011 16:39:32 No usable GPUs found
06/08/2011 16:39:33 Milkyway@home URL http://milkyway.cs.rpi.edu/milkyway/; Computer ID 278256; resource share 1
06/08/2011 16:39:33 General prefs: from http://www.worldcommunitygrid.org/ (last modified 05-Aug-2011 20:52:06)
06/08/2011 16:39:33 Host location: none
06/08/2011 16:39:33 General prefs: using your defaults
06/08/2011 16:39:33 Preferences:
06/08/2011 16:39:33 max memory usage when active: 6540.06MB
06/08/2011 16:39:33 max memory usage when idle: 6540.06MB
06/08/2011 16:39:33 max disk usage: 10.00GB
06/08/2011 16:39:33 (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
06/08/2011 16:39:33 Not using a proxy

Cheers for the suggestions,
23) Message boards : GPUs : Caught SIGSEGV in ATI GPU detection (Message 39479)
Posted 6 Aug 2011 by skgiven
Post:
I was on 6.12.33, and had the same problem, so thought I would try the latest Boinc code. Yes I am an Alpha tester, and aware of some changes to GPU reporting, but I'm more of an intermittent tester than a regular. I'm also a CA, and test on several projects, so it's important to familiarize myself with the different Boinc versions.

I suspect this might be driver related. I'm on a 2003 x64 server and in the same way as motherboard producers dont like to offer drivers, ATI tend to bypass them. Most XP x64 drivers work, but not all. ATI offer a driver for the 2008 servers, but dont list 2003.

After a clean install of ATI drivers, Boinc is still reporting the same message. GPUz and Device manager show correct details without any problems.

I'm going to remove Boinc and then try a slightly earlier version of Boinc, 6.10.60.
24) Message boards : GPUs : Caught SIGSEGV in ATI GPU detection (Message 39475)
Posted 6 Aug 2011 by skgiven
Post:
Driver Sweeper did find more NVidia files and removed them. Had to restart and then found 4 more, but Boinc is still reporting the same Caught SIGSEGV in ATI GPU detection error message.

06/08/2011 10:53:29 | | Starting BOINC client version 6.13.1 for windows_x86_64
06/08/2011 10:53:29 | | This a development version of BOINC and may not function properly
06/08/2011 10:53:29 | | Config: report completed tasks immediately
06/08/2011 10:53:29 | | Config: use all coprocessors
06/08/2011 10:53:29 | | Config: zero long-term debts on startup
06/08/2011 10:53:29 | | log flags: file_xfer, sched_ops, task, coproc_debug, cpu_sched_debug, dcf_debug
06/08/2011 10:53:29 | | log flags: debt_debug, rr_simulation, sched_op_debug, work_fetch_debug
06/08/2011 10:53:29 | | Libraries: libcurl/7.21.6 OpenSSL/1.0.0d zlib/1.2.5
06/08/2011 10:53:29 | | Data directory: E:\BOINC
06/08/2011 10:53:29 | | Running under account Administrator
06/08/2011 10:53:29 | | Processor: 8 GenuineIntel Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz [Family 6 Model 42 Stepping 7]
06/08/2011 10:53:29 | | Processor: 256.00 KB cache
06/08/2011 10:53:29 | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 syscall nx lm vmx smx tm2 popcnt aes pbe
06/08/2011 10:53:29 | | OS: Microsoft Windows Server 2003 "R2": Standard Server x64 Edition, Service Pack 2, (05.02.3790.00)
06/08/2011 10:53:29 | | Memory: 7.98 GB physical, 23.53 GB virtual
06/08/2011 10:53:29 | | Disk: 931.51 GB total, 70.57 GB free
06/08/2011 10:53:29 | | Local time is UTC +1 hours
06/08/2011 10:53:29 | | No NVIDIA library found
06/08/2011 10:53:29 | | Caught SIGSEGV in ATI GPU detection
06/08/2011 10:53:29 | | clGetDeviceIDs for platform #0 returned error -1
06/08/2011 10:53:29 | | No usable GPUs found

From Device Manager I'm told I have 8.872.0.0 (release data 7/7/11), so that's up to date and visible to Windows.
When I tried to open GPU-Z I got an error message,
'ATI OpenCL driver bug detected, skipping OpenCL detection.
Uninstall the AMD Stream SDK / OpenCL driver to re-enable this functionality'.
When I click OK GPUZ reports that I do have an ATI Radeon HD 5800 Series GPU, driver version ati2mtag 8.872.0.0 / 2003R2. Under GPUZ's Sensors tab I can see all the usual details, Clock rates, temps, fan speeds, so this just seems to be an OpenCL bug. I take it Boinc tries to read this too.

So, does anyone know how to 'Uninstall the AMD Stream SDK / OpenCL driver', without the other drivers?
I'll do another reboot and if that does not work I will remove all ATI drivers, run driver sweeper to make sure, restart and try another fresh install again.
25) Message boards : GPUs : Caught SIGSEGV in ATI GPU detection (Message 39471)
Posted 6 Aug 2011 by skgiven
Post:
Thanks for the suggestion.
I already used CCleaner and did a Project reset and even updated to the latest Boinc Beta. Driver Sweeper (from your FAQ link) requires 'registration' which means giving them your details + £29.95.

Suggest replacing that with,
http://download.cnet.com/driver-sweeper/3000-2094_4-75219868.html
The Phyxion version of Driver Sweeper is Free, so I gave that a try.
26) Message boards : GPUs : Caught SIGSEGV in ATI GPU detection (Message 39466)
Posted 5 Aug 2011 by skgiven
Post:
Having problems with an AMD/ATI driver install.
Pulled a GTX470 and installed an HD 5850.

Tried the auto detect AMD install but had to chose XPx64, while using a Server (2003x86). After rebooting there was no indication that the drivers took, so I uninstalled all NVidia ware, rebooted and did a manual install of the driver (just the driver). While the driver is there, Boinc does not see it and reports, Caught SIGSEGV in ATI GPU detection:

06/08/2011 00:13:20 | | Starting BOINC client version 6.12.33 for windows_x86_64
06/08/2011 00:13:20 | | Config: report completed tasks immediately
06/08/2011 00:13:20 | | Config: use all coprocessors
06/08/2011 00:13:20 | | Config: zero long-term debts on startup
06/08/2011 00:13:20 | | log flags: file_xfer, sched_ops, task, coproc_debug, cpu_sched_debug, dcf_debug
06/08/2011 00:13:20 | | log flags: debt_debug, rr_simulation, sched_op_debug, work_fetch_debug
06/08/2011 00:13:20 | | Libraries: libcurl/7.19.7 OpenSSL/0.9.8l zlib/1.2.5
06/08/2011 00:13:20 | | Data directory: E:\BOINC
06/08/2011 00:13:20 | | Running under account Administrator
06/08/2011 00:13:20 | | Processor: 8 GenuineIntel Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz [Family 6 Model 42 Stepping 7]
06/08/2011 00:13:20 | | Processor: 256.00 KB cache
06/08/2011 00:13:20 | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 syscall nx lm vmx smx tm2 popcnt aes pbe
06/08/2011 00:13:20 | | OS: Microsoft Windows Server 2003 "R2": Standard Server x64 Edition, Service Pack 2, (05.02.3790.00)
06/08/2011 00:13:20 | | Memory: 7.98 GB physical, 23.53 GB virtual
06/08/2011 00:13:20 | | Disk: 931.51 GB total, 69.96 GB free
06/08/2011 00:13:20 | | Local time is UTC +1 hours
06/08/2011 00:13:20 | | No NVIDIA library found
06/08/2011 00:13:20 | | Caught SIGSEGV in ATI GPU detection
06/08/2011 00:13:20 | | No usable GPUs found

Any suggestions?
27) Message boards : BOINC client : My Wish List (Message 37781)
Posted 12 May 2011 by skgiven
Post:
Thanks Richard,

That would be perfect if it could be introduced into the Boinc Manager; an exit_when_idle button/option (Close Boinc When Tasks Finish), ideally with a text tip advising that this reports all tasks and closes Boinc.
Possibly with an additional, and shut down windows option.
28) Message boards : BOINC client : My Wish List (Message 37778)
Posted 12 May 2011 by skgiven
Post:
I'm reporting a minor default functionality problem here. This problem has popped up in several threads recently, when the average user discovers their tasks finished but did not report.

If for example, someone selects no new tasks because they are going on holidays for 3weeks and then just shuts the computer down when existing tasks have finished (100%) and uploaded, the tasks will still report too late by the time they get back from holiday (for most projects).

So I am asking that a new default setting be created to automatically report tasks when the last task is completed, in the situation where the cruncher chooses No new tasks.

For long tasks (over 12h) it might be a good idea that these report immediately, but that's another issue. Ideally each project could be individually configured to report immediately or not.

I am aware of the cc_config.xml file and how to use report tasks immediately. This setting is recommended for some GPU projects but not welcomed by some CPU projects.
29) Message boards : GPUs : Project GPU requirements (Message 37726)
Posted 9 May 2011 by skgiven
Post:
GPUGRID does not use ATI, only nvidia. It has apps for Win 32 bit, Win 64 bit and Linux 64 bit.

Yeah, for GPUGrid it's NVidia only, but there is only a Win 32bit app, and only a Linux 64bit app. The Win app works on X64 systems.
Since the majority of the work is on the GPU, there is little to be gained from a 64bit app (around 1% increase in performance last time I checked).

- Windows apps? Y
- Linux apps? Y (x64 only)
- Macintosh apps? N
- CAL/Brook+? N
- CUDA? Y
- OpenCL? N (could work on NVidia, but slower than CUDA)
- Memory Use on GPU? (Varies and depends on task type and length. Some long tasks use 650MB on a GTX470 which has 1.2GB onboard GDDR5, but they can still run on a GTX460 512MB card)
- Memory Use on PC? (Varies; Long Task uses Virtual Memory 113MB, Working set 98MB)
- Min. driver version? (197.45 for 6.12app, Fermi cards can only run the 6.13app and driver requirements vary but at least 257.21 for Win and 256.40 for Linux. Later for the 500series).
When we change to CUDA 4.0 the 6.12app will retire. This will mean updating any older cards drivers to at least 257.21 in order to support CUDA 3.1. Fermi's will hopefully work better on the CUDA 4 based app but people that don't update their Fermi driver will still be able to run the CUDA 3.1 app (6.13).
Upload sizes (worth adding) are about 28MB for the long tasks.

Hope that helps
30) Message boards : GPUs : GPU mis-identified in dual gpu system. (Message 37722)
Posted 9 May 2011 by skgiven
Post:
The most up to date GPU is reported twice in the system spec, based on compute capability, but in the task details you can see which GPU the task ran on and both GPU's real details.
31) Message boards : Documentation : FAQ thread discussion links (Message 37721)
Posted 9 May 2011 by skgiven
Post:
Any chance of adding discussion links to the FAQ's listed here, http://boincfaq.mundayweb.com/index.php

With changes in architecture and advances in Bionc, some of these might need to be revisited.
For example, Allow tasks to finish when they reach 99%:
A couple of years back the associated problems did not affect as many people, but with increasing CPU cores/threads it's more of an issue, and the problems associated with this (memory for example) might become more exasperated.

Might also reduce reposts.
32) Message boards : BOINC client : My Wish List (Message 37719)
Posted 9 May 2011 by skgiven
Post:
When a user selects No New Tasks for a project, it would be useful to have these tasks automatically report when they complete.
All complete or individually? Either would be better than the present situation; 6.10.60 does not automatically report tasks even when all tasks have finished and uploaded.
33) Message boards : BOINC client : Boinc for Tegra 2 (Message 36351)
Posted 6 Jan 2011 by skgiven
Post:
Microsoft's Windows chief Steven Sinofsky demonstrates Windows running on an Atom and 3 different Arm processors, including Tegra. Apparently, he also suggested that application and device compatibility won't be a major barrier.

Read more: Microsoft shows Windows running on ARM | News | PC Pro
http://www.pcpro.co.uk/news/364066/microsoft-shows-windows-running-on-arm

While some have dismissed these processors as being too weak or requiring too much energy, the opposite is in fact the truth. These are quite powerful CPU’s but not power hungry CPU’s. The Microsoft’s Tablet/Slate operating system requires the same hardware spec as Win7. If it’s powerful enough to run Win7 then it’s powerful enough to crunch on.

"When you think of mobile devices and slates and what we think of as PCs, they've all converged" - Sinofsky.

http://seattletimes.nwsource.com/html/microsoftpri0/2013854643_ces_2011_microsoft_shows_future_windows_running_on.html

Obviously it is not x86 but perhaps by the end of 2012 we will be able to run Boinc on a tablet.
34) Message boards : Projects : Ibercivis Projects (Message 35614)
Posted 5 Nov 2010 by skgiven
Post:
Ibercivis does not do research into Spanish :)

http://boinc.berkeley.edu/projects.php

http://www.ibercivis.es/index.php?module=public&section=channels&action=view&id_channel=3

Cheers,
35) Message boards : Questions and problems : 1 Task per GPU (Message 35173)
Posted 10 Oct 2010 by skgiven
Post:
Message reads as follows,
Config: ignoring NVIDIA GPU 0
36) Message boards : Questions and problems : 1 Task per GPU (Message 35163)
Posted 9 Oct 2010 by skgiven
Post:
OK, looks like running and queued tasks need to finish/get terminated, a project detatch or reset, several restarts and so on need to be done for this to work.
37) Message boards : Questions and problems : 1 Task per GPU (Message 35162)
Posted 9 Oct 2010 by skgiven
Post:
<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
<ignore_cuda_dev>0</ignore_cuda_dev>
</options>
</cc_config>

09/10/2010 21:32:13 Re-reading cc_config.xml
09/10/2010 21:32:13 Re-read config file
09/10/2010 21:32:13 Config: report completed tasks immediately
09/10/2010 21:32:13 Config: use all coprocessors
09/10/2010 21:32:13 log flags: file_xfer, sched_ops, task

boinc Exit and Start,

09/10/2010 21:33:25 Starting BOINC client version 6.10.58 for windows_intelx86
09/10/2010 21:33:25 Config: report completed tasks immediately
09/10/2010 21:33:25 Config: use all coprocessors
09/10/2010 21:33:25 log flags: file_xfer, sched_ops, task
09/10/2010 21:33:25 Libraries: libcurl/7.19.7 OpenSSL/0.9.8l zlib/1.2.3

After both a read and a restart nothing changed; still using 2 GPUs. Per chance, do existing tasks have to complete first?

- Or even, would I need to remove card 0 from the system, reboot, start Boinc and then shut down, install the card, reboot and start Boinc again?

Thanks,
38) Message boards : Questions and problems : 1 Task per GPU (Message 35159)
Posted 9 Oct 2010 by skgiven
Post:
Hi, is this cc_confiig.xml file sufficient to prevent GPU 0 from being used?

<cc_config>
<options>
<use_all_gpus>0</use_all_gpus>
<ignore_cuda_dev>0</ignore_cuda_dev>
</options>
</cc_config>

XPx86, i7-920, boinc 6.10.58, two GTX470's.

Thanks,
39) Message boards : The Lounge : GTS 450 and others (Message 34661)
Posted 13 Sep 2010 by skgiven
Post:
A new mid-range Fermi has been released (GTS 450).
This GF106 card differs in design from the other two types.
It has 192 shaders and uses 16 ROPs. A 12/1 ratio. The GF100 cards were 10/1 and the GF104 cards are 14/1.

Also out is a range of OEM/laptop/desktop cards:

GeForce GTX 470M
GeForce GTX 460M

GeForce GT 445M
GeForce GT 435M
GeForce GT 425M
GeForce GT 420M
GeForce GT 415M
40) Message boards : BOINC client : 6.11.1 Released to Beta - Notes? (Message 34063)
Posted 4 Aug 2010 by skgiven
Post:
BobCat13, do you mean that when a task has not been started it gets resent, or some form of acknowledgement/instruction from the server to say, continue crunching same task?
When I run Lattice tasks they all go well over the deadline. They don’t reset; at least the task is not dumped and reissued (same task or new one), and I don’t start crunching it from the start again.
Lattice advise crunchers to continue running tasks past their deadline, as most go over the estimated runtime, are just as useful and people get normal credit. I think in the case of Lattice it is just down to their app and their scheduler that the % complete is incorrect. Looking at running tasks, the expected runtime might now be more accurate, for some systems, but their deadlines are still too short; I have 2 tasks at 28h on a Q6600 (stock, 4GB XP system). The tasks are at 1% complete and the estimated time remaining is about 326h. These will both finish after their deadline of the 14thAug.
Lattice may deliberately set short runtime cutoffs to force Boinc to continuously run the Lattice tasks, rather than stopping to do other tasks because switching tasks would crash the Lattice tasks/reset them.
Whatever the situation, it would be better if the projects handle their own "abort task" messages, preferably without the word abort; crunching is not a rocket mission or a birth control campaign, so perhaps Boinc should not employ such brutally worded generic messages when individual tasks reach such deadlines.
Also, if the system loses its time, Boinc aborts everything! This is a clear situation and the user should be given the choice to end all tasks, or change the time.


Previous 20 · Next 20

Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.