small preview of 'what can happen if you run' seti-cuda
driver version: nv4_disp 6.14.11.8120 - nVIDIA ForceWare 181.20
OS type: WinXPPro SP3 32bit
gfx: 9600GT
HOST

a) gpu is not overclocked
b) ~
c) using RivaTuner to regulate fan speed, temp never goes 50°C+
default is 35%, when crunhing its at 45% speed
d) ~
e) Everest Ultimate Edition / RivaTuner v2.22

- using stock app 6.06_windows_intelx86

lastest errors:
-----------------------------
this one kicked my gfx to hell (snow on screen), i suspended work and did a instant reboot
05no08ab.24748.20931.8.8.16_3
WU true angle range is : 0.009233
Cuda error 'cudaMemcpy(best_PoT, dev_tmp_pot, max_nb_of_elems * sizeof(float), cudaMemcpyDeviceToHost)' in file 'c:/sw/gpgpu/seti/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1265 : unknown error.
-----------------------------
this one runs 3 times into a stuck / idle
05no08ab.8510.4571.10.8.9_1
WU true angle range is : 0.011551
Cuda error 'cudaMemcpy(best_PoT, dev_tmp_pot, max_nb_of_elems * sizeof(float), cudaMemcpyDeviceToHost)' in file 'c:/sw/gpgpu/seti/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1265 : unknown error.
-----------------------------
this one causes a bluescreen with instant reboot
05no08ab.8510.4571.10.8.8_0
WU true angle range is : 0.011551
Cuda error 'cudaMemcpy(best_PoT, dev_tmp_pot, max_nb_of_elems * sizeof(float), cudaMemcpyDeviceToHost)' in file 'c:/sw/gpgpu/seti/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1265 : unknown error.
-----------------------------
this one runs into a stuck / idle
05no08ab.8510.4571.10.8.11_1
Cuda error 'cudaMemcpy(best_PoT, dev_tmp_pot, max_nb_of_elems * sizeof(float), cudaMemcpyDeviceToHost)' in file 'c:/sw/gpgpu/seti/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1265 : unknown error.
-----------------------------
start - error - done
05no08ab.8510.4571.10.8.6_0
WU true angle range is : 0.011551
Cuda error 'cudaMemcpy(best_PoT, dev_tmp_pot, max_nb_of_elems * sizeof(float), cudaMemcpyDeviceToHost)' in file 'c:/sw/gpgpu/seti/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1265 : unknown error.

small request to seti-staff: please dont provide VeryLowAngelRange Task's to Seti-Cuda-Users. Thx

---

small preview of 'what can happen if you run' seti-cuda
driver version: nv4_disp 6.14.11.8120 - nVIDIA ForceWare 181.20
OS type: WinXPPro SP3 32bit
gfx: 9600GT
HOST

a) gpu is not overclocked
b) ~
c) using RivaTuner to regulate fan speed, temp never goes 50°C+
default is 35%, when crunhing its at 45% speed
d) ~
e) Everest Ultimate Edition / RivaTuner v2.22

- using MB_r396mod based on 6.06 stock app

this one kicked my gfx to hell (snow on screen), i suspended work and did a instant reboot
06no08ab.23244.1708.13.8.38_2
WU true angle range is : 0.013119
Cuda error 'cudaMemcpy(best_PoT, dev_tmp_pot, max_nb_of_elems * sizeof(float), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1265 : unknown error.
Cuda error 'cudaMemcpy(best_PoT, dev_tmp_pot, max_nb_of_elems * sizeof(float), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1265 : unknown error.
Cuda error 'cudaMemcpy(best_PoT, dev_tmp_pot, max_nb_of_elems * sizeof(float), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1265 : unknown error.
Cuda error 'cudaMemcpy(best_PoT, dev_tmp_pot, max_nb_of_elems * sizeof(float), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1265 : unknown error.

Unhandled Exception Detected...

- Unhandled Exception Record -
Reason: Breakpoint Encountered (0x80000003) at address 0x7C91120E

Engaging BOINC Windows Runtime Debugger...

I know its VLAR,
i know i should abort them before starting,
i know i use a mod-version,
but maybe the devs want to have a look at the debugging message.

---

i know i use a mod-version,
but maybe the devs want to have a look at the debugging message.

Raistmer is sending them to the developers already and working together with them to get this fixed, no worries about that.

---

1119287178
Name 08no08ad.9844.9480.5.8.123_0
Workunit 393724769
Created 11 Jan 2009 9:02:40 UTC
Sent 11 Jan 2009 10:41:17 UTC
Received 11 Jan 2009 20:05:12 UTC
Server state Over
Outcome Success
Client state Done
Exit status 0 (0x0)
Computer ID 4751479
Report deadline 3 Feb 2009 16:44:37 UTC
CPU time 24.88216
stderr out

<core_client_version>6.5.0</core_client_version>
<![CDATA[
<stderr_txt>
setiathome_CUDA: Found 1 CUDA device(s):
Device 1 : GeForce 8600 GT
totalGlobalMem = 268435456
sharedMemPerBlock = 16384
regsPerBlock = 8192
warpSize = 32
memPitch = 262144
maxThreadsPerBlock = 512
clockRate = 1188000
totalConstMem = 65536
major = 1
minor = 1
textureAlignment = 256
deviceOverlap = 0
multiProcessorCount = 4
setiathome_CUDA: CUDA Device 1 specified, checking...
Device 1: GeForce 8600 GT is okay
SETI@home using CUDA accelerated device GeForce 8600 GT
Rise priority modification by Raistmer based on rev396 of SETI@home sources
Priority of worker thread rised successfully
Total GPU memory 268435456 free GPU memory 225050624
setiathome_enhanced 6.02 Visual Studio/Microsoft C++
libboinc: 6.3.22

Work Unit Info:
...............
WU true angle range is : 0.008259
Optimal function choices:
-----------------------------------------------------
name
-----------------------------------------------------
v_BaseLineSmooth (no other)
v_GetPowerSpectrum 0.00018 0.00000
v_ChirpData 0.02580 0.00000
v_Transpose4 0.00670 0.00000
FPU opt folding 0.00677 0.00000
Cuda error 'find_pulse_kernel2<3, false>' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1166 : unknown error.
Cuda error 'find_pulse_kernel2<4, true>' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1172 : unknown error.
Cuda error 'find_pulse_kernel2<4, true>' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1172 : unknown error.
Cuda error 'find_pulse_kernel2<5, true>' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1178 : unknown error.
Cuda error 'find_pulse_kernel2<5, true>' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1178 : unknown error.
Cuda error 'cudaMemcpy(&flags, dev_find_pulse_flag, sizeof(*dev_find_pulse_flag), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1250 : unknown error.
Cuda error 'cudaMemcpy(PulseResults, dev_PulseResults, 4 * (cudaAcc_NumDataPoints / AdvanceBy + 1) * sizeof(*dev_PulseResults), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1262 : unknown error.
Cuda error 'cudaAcc_transpose' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_transpose.cu' in line 74 : unknown error.
Cuda error 'cudaAcc_transpose' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_transpose.cu' in line 74 : unknown error.
Cuda error 'cudaMemcpy(best_PoT, dev_tmp_pot, max_nb_of_elems * sizeof(float), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1265 : unknown error.
Cuda error 'cudaAcc_transpose' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_transpose.cu' in line 74 : unknown error.
Cuda error 'cudaAcc_transpose' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_transpose.cu' in line 74 : unknown error.
Cuda error 'cudaMemcpy(tmp_PoT, dev_tmp_pot, max_nb_of_elems * sizeof(float), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1269 : unknown error.
Cuda error 'cufftExecC2C' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_fft.cu' in line 63 : unknown error.
Cuda error 'cudaAcc_GetPowerSpectrum_kernel' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_PowerSpectrum.cu' in line 56 : unknown error.
Cuda error 'cudaAcc_GetPowerSpectrum_kernel' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_PowerSpectrum.cu' in line 56 : unknown error.
Cuda error 'cudaAcc_summax32_kernel' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_summax.cu' in line 147 : unknown error.
Cuda error 'cudaAcc_summax32_kernel' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_summax.cu' in line 147 : unknown error.
Cuda error 'cudaMemcpy(PowerSpectrumSumMax, dev_PowerSpectrumSumMax, cudaAcc_NumDataPoints / fftlen * sizeof(*dev_PowerSpectrumSumMax), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_summax.cu' in line 160 : unknown error.
Cuda error 'find_triplets_kernel' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 224 : unknown error.
Cuda error 'find_triplets_kernel' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 224 : unknown error.
Cuda error 'cudaMemcpy(&flags, dev_flag, sizeof(*dev_flag), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 228 : unknown error.
Cuda error 'cudaMemcpy(TripletResults, dev_TripletResults, 2 * grid.x * block.x * grid.y * block.y * sizeof(*dev_TripletResults), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 243 : unknown error.
Cuda error 'cudaAcc_transpose' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_transpose.cu' in line 74 : unknown error.
Cuda error 'cudaAcc_transpose' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_transpose.cu' in line 74 : unknown error.
Cuda error 'cudaMemcpy(tmp_PoT, dev_t_PowerSpectrum, cudaAcc_NumDataPoints * sizeof(*dev_t_PowerSpectrum), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 245 : unknown error.
Cuda error 'find_pulse_kernel2<3, false>' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1166 : unknown error.
Cuda error 'find_pulse_kernel2<3, false>' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1166 : unknown error.
Cuda error 'find_pulse_kernel2<4, true>' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1172 : unknown error.
Cuda error 'find_pulse_kernel2<4, true>' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1172 : unknown error.
Cuda error 'find_pulse_kernel2<5, true>' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1178 : unknown error.
Cuda error 'find_pulse_kernel2<5, true>' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1178 : unknown error.
Cuda error 'cudaMemcpy(&flags, dev_find_pulse_flag, sizeof(*dev_find_pulse_flag), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1250 : unknown error.
Cuda error 'cudaMemcpy(PulseResults, dev_PulseResults, 4 * (cudaAcc_NumDataPoints / AdvanceBy + 1) * sizeof(*dev_PulseResults), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1262 : unknown error.
Cuda error 'cudaAcc_transpose' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_transpose.cu' in line 74 : unknown error.
Cuda error 'cudaAcc_transpose' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_transpose.cu' in line 74 : unknown error.
Cuda error 'cudaMemcpy(best_PoT, dev_tmp_pot, max_nb_of_elems * sizeof(float), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu' in line 1265 : unknown error.
Cuda error 'cufftExecC2C' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_fft.cu' in line 63 : unknown error.
Cuda error 'cudaAcc_GetPowerSpectrum_kernel' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_PowerSpectrum.cu' in line 56 : unknown error.
Cuda error 'cudaAcc_GetPowerSpectrum_kernel' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_PowerSpectrum.cu' in line 56 : unknown error.
Cuda error 'cudaAcc_summax32_kernel' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_summax.cu' in line 147 : unknown error.
Cuda error 'cudaAcc_summax32_kernel' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_summax.cu' in line 147 : unknown error.
Cuda error 'cudaMemcpy(PowerSpectrumSumMax, dev_PowerSpectrumSumMax, cudaAcc_NumDataPoints / fftlen * sizeof(*dev_PowerSpectrumSumMax), cudaMemcpyDeviceToHost)' in file 'd:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_summax.cu' in line 160 : unknown error.
SETI@Home Informational message -9 result_overflow
NOTE: The number of results detected exceeds the storage space allocated.

Flopcounter: 20886855729.456421

Spike count: 30
Pulse count: 0
Triplet count: 0
Gaussian count: 0
called boinc_finish

</stderr_txt>
]]>

Validate state Initial
Claimed credit 0.0717780525135726
Granted credit 0
application version 6.06

...............
WU true angle range is : 0.008259

It's known bug with VLAR. You should abort such tasks until new version of CUDA MB will arrive.
Doing such tasks on current version can (and sometimes will) lead to BSoD, driver restart and invalid processing of other tasks until whole OS will be rebooted.
Much better to abort them before they do harm to system.

driver version: nv4_disp 6.14.11.8120 - nVIDIA ForceWare 181.20
OS type: WinXPPro SP3 32bit
gfx: 9600GT
HOST

a) gpu is not overclocked
b) ~
c) using RivaTuner to regulate fan speed, temp never goes 50°C+
default is 35%, when crunhing its at 45% speed
d) ~
e) Everest Ultimate Edition / RivaTuner v2.22

- using MB_r396mod based on 6.06 stock app

error:
-----------------------------
05no08aa.1851.267182.3.8.193_1
WU true angle range is : 10.349294
SETI@home error -12 Unknown error
cudaAcc_find_triplets erroneously found a triplet twice in find_triplets_kernel
File: d:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu
Line: 235

maybe it is known anyway but i lost overview at reported errors ;)

---

driver version: nv4_disp 6.14.11.8120 - nVIDIA ForceWare 181.20
OS type: WinXPPro SP3 32bit
gfx: 9600GT
HOST

a) gpu is not overclocked
b) ~
c) using RivaTuner to regulate fan speed, temp never goes 50°C+
default is 35%, when crunhing its at 45% speed
d) ~
e) Everest Ultimate Edition / RivaTuner v2.22

- using MB_r396mod based on 6.06 stock app

error:
-----------------------------
05no08aa.1851.267182.3.8.193_1
WU true angle range is : 10.349294
SETI@home error -12 Unknown error
cudaAcc_find_triplets erroneously found a triplet twice in find_triplets_kernel
File: d:/BTR/seticuda/seti_boinc/client/cuda/cudaAcc_pulsefind.cu
Line: 235

maybe it is known anyway but i lost overview at reported errors ;)

Yes, it's known error.
Report:
http://setiathome.berkeley.edu/beta/forum_thread.php?id=1482

Hi there,

I've got several Cuda errors. Follows the required details.

Dell Laptop XPS M1530
Windows XP SP3 32 Bit
NVIDIA GeForce 8600M GT - Driver version 6.14.11.7431

List of the last few WU with errors:

http://setiathome.berkeley.edu/result.php?resultid=1118901983
http://setiathome.berkeley.edu/result.php?resultid=1118901967
http://setiathome.berkeley.edu/result.php?resultid=1118901907
http://setiathome.berkeley.edu/result.php?resultid=1118901906
http://setiathome.berkeley.edu/result.php?resultid=1118896567
http://setiathome.berkeley.edu/result.php?resultid=1118896552
http://setiathome.berkeley.edu/result.php?resultid=1118896549
http://setiathome.berkeley.edu/result.php?resultid=1118896545
http://setiathome.berkeley.edu/result.php?resultid=1118896544

Questions from the development group and my answers.

Q -- State if you overclock the GPU and if so by how much?
A -- No I did not. This is a brand new laptop (5 days old)

Q -- tell if you have tried to clock the GPU to default speeds and still see the problem?
A -- No I did not. The only thing I've done was to "upgrade" from Vista to Windows XP SP3.

Q -- Did you set the fan speed manually to anything else than default?
A -- Nope, it was never touched.

Q -- Which program(s) do you use to overclock the GPU, fan etc.?
A -- Not applicable

Q -- Which program(s) do you use to keep check of the GPU?
A -- None, I never checked it.

---

NVIDIA GeForce 8600M GT - Driver version 6.14.11.7431

Try driver update from nvidia.com, lastest built is 6.14.11.8120

---

I am getting artifacts on my monitor when running with cuda enable/disabled with the latest BOINC release (6.4.5).

- I am running a fresh install of XP (SP3)
- AMD Phenon II X4 940 overclocked to 3.31GHz
- 4G of RAM
- New Bfg Tech GeForce GTX 280 H2OC (this is their overclocked version I have underclocked) temp is around 100 degrees F (water cooled).
- 1G of video RAM on the card.
- Latest NVidia Driver (6.14.11.8120)
- Dual Monitors

I am getting both artifacts on my monitor plus the system will seem to pause for several minutes. I get this only when I run BOINC. I have tried running the CPU and its regular setting. I have under clocked the GPU to its 'normal' setting. I set my preferences to not run CUDA and I still would have the same problem. Sometimes the system will run for a day (and produce a but load of results) and some times I will only get errors running BOINC. I finally down graded BOINC to 6.2.19 and I no longer have this issue.

---

...
- Dual Monitors
...

Dual Monitors - Is this a problem ?

---

- 1G of video RAM on the card.
- Latest NVidia Driver (6.14.11.8120)
- Dual Monitors

1GB of RAM should be enough to run dual monitors with.

I set my preferences to not run CUDA and I still would have the same problem. Sometimes the system will run for a day (and produce a but load of results) and some times I will only get errors running BOINC. I finally down graded BOINC to 6.2.19 and I no longer have this issue.

Hold on, you disabled running CUDA through your preferences and you still get errors. I just looked at those errors and it looks like you are running the graphics or screen saver, right?

I see these errors:


<core_client_version>6.4.5</core_client_version>
<![CDATA[
<message>
- exit code -1073741819 (0xc0000005)
</message>
<stderr_txt>
In ap_gfx_main.cpp: in ap_graphics_init(): Starting client.
In ap_client_main.cpp: in mainloop(): at dm_chunk_large 896

</stderr_txt>

Please look at this FAQ which will give you some pointers as to what might be broken. Do know that between 5.10.45 and the latest 6.6.0 version nothing changed in the way BOINC shows the graphics or screen saver, so it may really (have been) be something else.

I must say, I like this task, for its claimed credit of 15,068,949,224,180,199,424.00 .. I hope you get it. ;-)

---

...I set my preferences to not run CUDA and I still would have the same problem...

You cannot set your preferences to not run CUDA, you can only set your preferences to not download new CUDA tasks. So, already downloaded CUDA tasks in your queue will still run (and eventually cause problems).

Gruß,
Gundolf

---

Computer sind nicht alles im Leben. (Kleiner Scherz)

SETI@home classic workunits 3,758
SETI@home classic CPU time 66,520 hours

They're Astropulse 5.00 and Set 6.03's.

---

I am pretty sure I don't have the screen saver enabled. That's usually the first thing I turn off. I still received the screen artifacts with BOINC not running any CUDA results. I ran all day yesterday without any problems on the older download.

Thanks

---

It's possible one of the libraries that BOINC uses went a little bad and that the install of 6.2.19 fixed that.

Keep an eye on things. I've seen the graphics errors on other people's AP tasks with errors as well.

---

You were right, the screen saver was on to Seti@home.

---