search

ratt-ru / mt-imager

The Malta Imager -- a GPU-based imager from Daniel Muscat (U. Malta)
GNU General Public License v3.0
1 stars 2 forks source link

Cuda Error: invalid device function #13

Open xiaoluo1314 opened 10 years ago

xiaoluo1314 commented 10 years ago

[root@astronomy bin]# ./mt-imager 2014-05-29 09:12:50 SEVERE MeasTable::dUTC(Double) (file /root/casacore-1.5.0/measures/Measures/MeasTable.cc, line 6649) Leap second table TAI_UTC seems out-of-date. 2014-05-29 09:12:50 SEVERE MeasTable::dUTC(Double) (file /root/casacore-1.5.0/measures/Measures/MeasTable.cc, line 6649)+ Until table is updated (see aips++ manager) times and coordinates 2014-05-29 09:12:50 SEVERE MeasTable::dUTC(Double) (file /root/casacore-1.5.0/measures/Measures/MeasTable.cc, line 6649)+ derived from UTC could be wrong by 1s or more. terminate called after throwing an instance of 'GAFW::GPU::GPUCudaException' what():
Description: An error has been detected after submission of WProjectionConvolutionGenerator.calculator.taper Operator Name:Operator::gridsf Type: Cuda Exception Cuda Error: invalid device function File Name: /root/Desktop/CUDAWP/mt-imager-master/code/gafw/Implimentation/GPUImplimentation/CUDADeviceManager.cpp Line Number: 484

Aborted (core dumped). Could you tell me this is what causes? Can you help me to solve? Thank you very much.

twillis449 commented 10 years ago

The casacore Measures messages about Leap seconds etc are just telling you that one of your files in /opt/data (I think ephemerides) is not up-to-date with respect to the the date of observation in the measurement set you are working with. But that should not be causing the crash. The Measures errors are just warning messages.

This web page (http://casa.nrao.edu/docs/userman/UserManse82.html) has instructions how to update the 'data' files.

Tony

On 14-05-28 06:20 PM, xiaoluo1314 wrote:

[root@astronomy bin]# ./mt-imager 2014-05-29 09:12:50 SEVERE MeasTable::dUTC(Double) (file /root/casacore-1.5.0/measures/Measures/MeasTable.cc, line 6649) Leap second table TAI_UTC seems out-of-date. 2014-05-29 09:12:50 SEVERE MeasTable::dUTC(Double) (file /root/casacore-1.5.0/measures/Measures/MeasTable.cc, line 6649)+ Until table is updated (see aips++ manager) times and coordinates 2014-05-29 09:12:50 SEVERE MeasTable::dUTC(Double) (file /root/casacore-1.5.0/measures/Measures/MeasTable.cc, line 6649)+ derived from UTC could be wrong by 1s or more. terminate called after throwing an instance of 'GAFW::GPU::GPUCudaException' what():

Description: An error has been detected after submission of WProjectionConvolutionGenerator.calculator.taper Operator Name:Operator::gridsf Type: Cuda Exception Cuda Error: invalid device function File Name: /root/Desktop/CUDAWP/mt-imager-master/code/gafw/Implimentation/GPUImplimentation/CUDADeviceManager.cpp Line Number: 484

Aborted (core dumped). Could you tell me this is what causes? Can you help me to solve? Thank you very much.

— Reply to this email directly or view it on GitHub https://github.com/ska-sa/mt-imager/issues/13.

xiaoluo1314 commented 10 years ago

thanks .

[root@astronomy bin]# ./mt-imager terminate called after throwing an instance of 'mtimager::MTImagerException' what():
Description: Query to casa::MeasTable::Observatory() for telescope position was unsuccessful Type: Imager Exception Object Name: MSDataVisibilityAsync Object ObjectName: MS Name:/opt/data/sun0922_ms File Name: /root/Desktop/CUDAWP/mt-imager-master/code/mt-imager/lib/components/VisibilityManager/MSVisibilityDataSetAsync.cpp Line Number: 106

Aborted (core dumped).

Could you tell me this is what causes? Can you help me to solve? Thank you very much.

twillis449 commented 10 years ago

So which observatory has generated your data?

On 14-05-28 07:33 PM, xiaoluo1314 wrote:

thanks .

[root@astronomy bin]# ./mt-imager terminate called after throwing an instance of 'mtimager::MTImagerException' what():

Description: Query to casa::MeasTable::Observatory() for telescope position was unsuccessful Type: Imager Exception Object Name: MSDataVisibilityAsync Object ObjectName: MS Name:/opt/data/sun0922_ms File Name: /root/Desktop/CUDAWP/mt-imager-master/code/mt-imager/lib/components/VisibilityManager/MSVisibilityDataSetAsync.cpp Line Number: 106

Aborted (core dumped).

Could you tell me this is what causes? Can you help me to solve? Thank you very much.

— Reply to this email directly or view it on GitHub https://github.com/ska-sa/mt-imager/issues/13#issuecomment-44488324.

xiaoluo1314 commented 10 years ago

I don't quite understand what you mean.My data has not yet been generated. The following error:

[root@astronomy bin]# ./mt-imager terminate called after throwing an instance of 'mtimager::MTImagerException' what():

Description: Query to casa::MeasTable::Observatory() for telescope position was unsuccessful Type: Imager Exception Object Name: MSDataVisibilityAsync Object ObjectName: MS Name:/opt/data/sun0922_ms File Name: /root/Desktop/CUDAWP/mt-imager-master/code/mt-imager/lib/components/VisibilityManager/MSVisibilityDataSetAsync.cpp Line Number: 106

Aborted (core dumped).

Is my target file sun0922. ms right? Or for other reasons?

twillis449 commented 10 years ago

How can you be running an imager without having data to image?

On 14-05-28 09:20 PM, xiaoluo1314 wrote:

I don't quite understand what you mean.My data has not yet been generated. ...

— Reply to this email directly or view it on GitHub https://github.com/ska-sa/mt-imager/issues/13#issuecomment-44493035.

twillis449 commented 10 years ago

I would assume that mt-imager is trying to process your data file sun0922_ms. Unfortunately its crashing. Only the developer/programmer working on mt-imager can help you further.

On 14-05-28 09:26 PM, xiaoluo1314 wrote:

I don't quite understand what you mean.My data has not yet been generated. The following error:

[root@astronomy bin]# ./mt-imager terminate called after throwing an instance of 'mtimager::MTImagerException' what():

Description: Query to casa::MeasTable::Observatory() for telescope position was unsuccessful Type: Imager Exception Object Name: MSDataVisibilityAsync Object ObjectName: MS Name:/opt/data/sun0922_ms File Name: /root/Desktop/CUDAWP/mt-imager-master/code/mt-imager/lib/components/VisibilityManager/MSVisibilityDataSetAsync.cpp Line Number: 106

Aborted (core dumped).

Is my target file sun0922. ms right? Or for other reasons?

— Reply to this email directly or view it on GitHub https://github.com/ska-sa/mt-imager/issues/13#issuecomment-44493268.

ddmusc commented 10 years ago

..and here I am....

The last error is stating that he cannot find the observatory(telescope) in the Observatory table... Make sure you have the latest observatory table... (sometimes MeerKat and LOFAR are missing).

For the other error related to CUDA I suspect that you are not using architecture 3+ CUDA devices ie (K10,K20,K40 GTX6 series and GTX 7 series)

xiaoluo1314 if you wish we can set up a meeting, such that I can help you get over your problems.

Regards Daniel

On Thu, May 29, 2014 at 6:10 AM, twillis449 notifications@github.comwrote:

I would assume that mt-imager is trying to process your data file sun0922_ms. Unfortunately its crashing. Only the developer/programmer working on mt-imager can help you further.

On 14-05-28 09:26 PM, xiaoluo1314 wrote:

I don't quite understand what you mean.My data has not yet been generated. The following error:

[root@astronomy bin]# ./mt-imager terminate called after throwing an instance of 'mtimager::MTImagerException' what():

Description: Query to casa::MeasTable::Observatory() for telescope position was unsuccessful Type: Imager Exception Object Name: MSDataVisibilityAsync Object ObjectName: MS Name:/opt/data/sun0922_ms File Name:

/root/Desktop/CUDAWP/mt-imager-master/code/mt-imager/lib/components/VisibilityManager/MSVisibilityDataSetAsync.cpp Line Number: 106

Aborted (core dumped).

Is my target file sun0922. ms right? Or for other reasons?

— Reply to this email directly or view it on GitHub https://github.com/ska-sa/mt-imager/issues/13#issuecomment-44493268.

— Reply to this email directly or view it on GitHubhttps://github.com/ska-sa/mt-imager/issues/13#issuecomment-44495066 .

xiaoluo1314 commented 10 years ago

hi Daniel，

      thank you very much .I appreciate your,For me, I can't write, also can't think of such a good program. I am very willing to learn like you.So I hope you can give me a platform for the exchange of learning with you.I know you are very busy, I also don't want to disturb you.
       Only hope you have time to help me see I put those unsolved problems,So I will be very grateful.So I will be very grateful.To say thanks again.

sincerely xiaoluo1314

xiaoluo1314 commented 10 years ago

Hi,thanks again.I think it a lot . 1.Here I have a question about where to set the arguments of observatory table.You see, in the comand of casapy, input tb.browse,then we input our height,lattitude,longtitude. 2.Then ,I wonder to know the ms file is generated by casa or this program itself. Thanks very much.

twillis449 commented 10 years ago

Its not clear to me what you are trying to do.

1) Are you trying to add latitude, longitude and height for an Observatory that is not in the geodetic/Observatories data file? If so which observatory? If you want to add a new observatory you need to have a working 'pyrap'. Well, you could also do it with casacore C++ code, but that would give me a real headache!

2) Measurement sets are either created a) by an Observatory doing an actual observation or b) by some program which can simulate an observation. CASA, for example, has a simulator tool which will allow you to create simulated ALMA observations / measurement sets.

On 14-05-29 07:22 PM, xiaoluo1314 wrote:

Hi,thanks again.I think it a lot . 1.Here I have a question about where to set the arguments of observatory table.You see, in the comand of casapy, input tb.browse,then we input our height,lattitude,longtitude. 2.Then ,I wonder to know the ms file is generated by casa or this program itself. Thanks very much.

— Reply to this email directly or view it on GitHub https://github.com/ska-sa/mt-imager/issues/13#issuecomment-44608361.

ddmusc commented 10 years ago

It will be a pleasure helping you... Please communicate with me over support@mtimager.com such that we can set up a meeting and help each other.

On Fri, May 30, 2014 at 2:56 AM, xiaoluo1314 notifications@github.com wrote:

hi Daniel，

  thank you very much .I appreciate your,For me, I can't write, also can't think of such a good program. I am very willing to learn like you.So I hope you can give me a platform for the exchange of learning with you.I know you are very busy, I also don't want to disturb you.
   Only hope you have time to help me see I put those unsolved problems,So I will be very grateful.So I will be very grateful.To say thanks again.

sincerely xiaoluo1314

— Reply to this email directly or view it on GitHub https://github.com/ska-sa/mt-imager/issues/13#issuecomment-44603909.

xiaoluo1314 commented 10 years ago

Thanks very much.