Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <ol> <li>It will be better to call <code>CUT_CHECK_ERROR</code> after <code>cudaThreadSynchronize()</code>. Because kernel run asynchronous and you must wait for kernel ending to know about errors... Maybe in second iteration you receive an error from first kernel usage. </li> <li>Be sure that you have some valid number in the most interesting variable <code>pointsNumber[0]</code> (it might cause a long internal loop). </li> <li>You could also improve speed of your kernel function: <ul> <li>Use better blocks. Threads configuration 20x20 will cause very slow memory usage (see Programming Guide and Best Practices). Try to use blocks 16x16. </li> <li>Do not use <code>pow(..., 2.0)</code> function. It's faster to use SQR macro (<code>#define SQR(x) (x)*(x)</code>)</li> <li>You don't use shared mem, so <code>__syncthreads()</code> is not required.</li> </ul></li> </ol> <p>PS: You could also pass value parameters to CUDA functions, not only pointers. Speed will be the same.</p> <p>PPS: please improve code's readability... Now you must edit six places to change block configuration... Inside the kernel you could use <code>blockDim</code> variable and you could use constants in go2 function. You could also use <code>bool firstTime</code> - it will be MUCH better then <code>float</code>.</p>
    singulars
    1. This table or related slice is empty.
    1. POTimeout in CUDA? / fermi / gtx465
      singulars
      1. PTQuestion
    1. PTAnswer
    1. This table or related slice is empty.
    1. USKoppeKTop
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. POTimeout in CUDA? / fermi / gtx465
      singulars
      1. PTQuestion
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTAcceptedByOriginator
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    1. COI had mistake in pointNumber[0] value, moreover your clues speed up this alghoritm almost 2 times :D thanks :) I have second question.. it is possible that change of gpu (form gtx260 896MB to gtx465 1GB) could casue some mistakes in memory managment? On gtx260 I could allocate float array of 500*500*500 (this is kind of LUT array) floats and now on on gtx when I allocate more than 60*60*60 floats array I get error "unspecified launch error" in place where copy data from host to device memory (this memcopy isn't connected with that LUT array)...?
      singulars
      1. PO
      1. USTome
    2. COIt's very hard to say something without code... I had newer saw such error during memcpy. This error is pretty same with "Segmentation fault" in host programs and occurred in kernels. `CUT_CHECK_ERROR` read error message which can be caused by earlier kernel launches (if you don't use `cudaThreadSynchronize`). There are only 4 types async operatioins: 1. kernel launch 2. all cudaMemcpy*Async 3. Memcpy device <-> device 4. Memory initialization Some advice: - Check parameters in `cudaMemcpy`. Right order and right memory allocation (maybe you forgot `cudaMalloc`?) - Use emu mode to dbg.
      singulars
      1. PO
      1. USKoppeKTop
    3. COAnd 1 more guess. Do you add boundary check to kernel? `if (holo_x+holo_y*MAX_FINAL_X >= MAX_FINAL_X*MAX_FINAL_Y) return;` Actually, 2 times - is not a limit ) If you will use __constant__ memory for storing A, X, Y and Z + maybe move to fast math (carefully) you'll give some more speedup.
      singulars
      1. PO
      1. USKoppeKTop
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload