Introduction

The overlapped input and output buffer (overlapped I/O buffer) is a work block buffer that contains both input and output data. The input and output sections are dynamically designated for each work block. This buffer is especially useful when you want to maximize the use of accelerator memory and when the input buffer can be overwritten by the output data.

The data for the input buffer can come from distinct sections of a large data set in host memory. These distinct data segments are gathered into the input data buffer on the accelerators. The ALF framework minimizes performance overhead by not duplicating input data unnecessarily.

The output data buffer is a single, contiguous buffer in the memory of the accelerator. Output data can be transferred to distinct memory segments within a large output buffer in host memory. After the compute kernel returns from processing one work block, the data in this buffer is moved to the host memory locations specified by the alf_wb_dtl_entry_add routine when the work block is constructed.

The following two simple examples show the usage of overlapped I/O buffers. Both examples perform matrix addition.

• Implementation 1 implements C = A + B, in which A, B, and C are different matrices. There are three separate matrices on the host for matrices a, b, and c (Figure 1).
  
  Figure 1. Overlapped I/O buffer implementing C = A + B
  
  
  
• Implementation 2 implements A = A + B, in which matrix A is overwritten by the result. Storage is reserved on the host for matrix a and matrix b. The result of a + b is stored in matrix b (Figure 2).
  
  Figure 2. Overlapped I/O buffer implementing A = A + B
  
  
  

Setting up the matrix

The code is similar to the matrix_add example in Fun with ALF, Part 1: Adding large matrices together. Listing 1 shows only the relevant code.

/* ---------------------------------------------- */ 
/* matrix declaration for the two cases           */ 
/* ---------------------------------------------- */ 
#ifdef C_A_B // C = A + B 
       alf_data_int32_t mat_a[ROW_SIZE][COL_SIZE]; // the matrix a 
       alf_data_int32_t mat_b[ROW_SIZE][COL_SIZE]; // the matrix b 
       alf_data_int32_t mat_c[ROW_SIZE][COL_SIZE]; // the matrix c 
#else // A = A + B 
       alf_data_int32_t mat_a[ROW_SIZE][COL_SIZE]; // the matrix a 
       alf_data_int32_t mat_b[ROW_SIZE][COL_SIZE]; // the matrix b 
#endif

Setting up the work block

The Listing 2 code segment shows the work block creation process for the two cases.

for (i = 0; i < ROW_SIZE; i+=PART_SIZE){ 
     if(i+PART_SIZE <= ROW_SIZE) 
        wb_parm.num_data = PART_SIZE; 
     else 
        wb_parm.num_data = ROW_SIZE - i; 

     alf_wb_create(task_handle, ALF_WB_SINGLE, 0, &wb_handle); 

     #ifdef C_A_B // C = A + B 
            // the input data A and B 
            alf_wb_dtl_begin(wb_handle, ALF_BUF_OVL_IN, 0); // offset at 0 
            alf_wb_dtl_entry_add(wb_handle, &mat_a[i][0], wb_parm.num_data*COL_SIZE, 
             ALF_DATA_INT32); // A 
            alf_wb_dtl_entry_add(wb_handle, &mat_b[i][0], wb_parm.num_data*COL_SIZE, 
             ALF_DATA_INT32); // B 
            alf_wb_dtl_end(wb_handle); 

            // the output data C is overlapped with input data A 
            // offset at 0, this is overlapped with A 
            alf_wb_dtl_begin(wb_handle, ALF_BUF_OVL_OUT, 0); 
            alf_wb_dtl_entry_add(wb_handle, &mat_c[i][0], wb_parm.num_data*COL_SIZE, 
             ALF_DATA_INT32); // C 
            alf_wb_dtl_end(wb_handle); 
     
      #else // A = A + B 
            // the input and output data A 
            alf_wb_dtl_begin(wb_handle, ALF_BUF_OVL_INOUT, 0); // offset 0 
            alf_wb_dtl_entry_add(wb_handle, &mat_a[i][0], wb_parm.num_data*COL_SIZE, 
             ALF_DATA_INT32); // A 
            alf_wb_dtl_end(wb_handle); 

            // the input data B is placed after A 
            alf_wb_dtl_begin(wb_handle, ALF_BUF_OVL_IN, 
             wb_parm.num_data*COL_SIZE*sizeof(alf_data_int32_t)); 
            alf_wb_dtl_entry_add(wb_handle, &mat_b[i][0], wb_parm.num_data*COL_SIZE, 
             ALF_DATA_INT32); // B 
            alf_wb_dtl_end(wb_handle); 
      #endif 
      alf_wb_parm_add(wb_handle, (void *)&wb_parm, sizeof(wb_parm)/sizeof(unsigned int), 
       ALF_DATA_INT32, 0); 
      alf_wb_enqueue(wb_handle); 
}

Setting up the accelerator code

Listing 3 shows the accelerator code. In both cases, the output sc can be set to the same location in accelerator memory as sa and sb.

/* ---------------------------------------------- */ 
/* the accelerator side code                      */ 
/* ---------------------------------------------- */ 
/* the computation kernel function                */ 
int comp_kernel(void *p_task_context, void *p_parm_ctx_buffer, 
                void *p_input_buffer, void *p_output_buffer, 
                void *p_inout_buffer, unsigned int current_count, 
                unsigned int total_count) 
{ 
  unsigned int i, cnt; 
  int *sa, *sb, *sc; 
  my_wb_parms_t *p_parm = (my_wb_parms_t *) p_parm_context; 

  cnt = p_parm->num_data * COL_SIZE; 

  sa = (int *) p_inout_buffer; 
  sb = sa + cnt; 
  sc = sa; 

  for (i = 0; i < cnt; i ++) 
       sc[i] = sa[i] + sb[i]; 

  return 0; 
}

Conclusion

This article described two implementations you can use to do matrix addition with overlapped I/O buffers.

Resources

Learn

• Use an RSS feed to request notification for the upcoming articles in this series. (Find out more about RSS feeds of developerWorks content.)
  
  
• Refer to Accelerated Library Framework for Cell Broadband Engine Programmer’s Guide and API Reference for the source material from which this article was extracted.
  
  
• Check out other articles in this Fun with ALF series.
  
  
• Take a look at these other ALF-related quick-read guides:
  • "Introducing ALF."
  • "10 major ALF concepts."
  • "Programming with ALF: Basic ALF application structure."
  • "Programming with ALF: Double buffering."
  • "Programming with ALF: Handling ALF constraints."
  • "Programming with ALF: Optimizing ALF applications."
  • "ALF and hybrid x86."
  
  
• To learn more on Cell/B.E. programming, try the developerWorks series:
  • "Programming high-performance applications on the Cell/B.E. processor"
  • "PS3 fab-to-lab"
  • "The little broadband engine that could"
  
  
• Refer to the Cell Broadband Engine documentation section of the IBM Semiconductor Solutions Technical Library for a wealth of downloadable manuals, specifications, and more.
  
  
• Sign up for the developerWorks newsletter and get the latest developer news and Cell/B.E. happenings delivered to your inbox each week. Check Power Architecture® when you sign up to receive Cell/B.E. news in your newsletter.
  
  

Get products and technologies

• Find all Cell/B.E.-related articles, discussion forums, downloads, and more at the IBM developerWorks Cell Broadband Engine resource center: your definitive resource for all things Cell/B.E.
  
  
• Contact IBM about custom Cell/B.E.-based or custom-processor based solutions.
  
  
• Get your copy of the IBM SDK for Multicore Acceleration 3.0 or browse through the exhaustive library of Cell/B.E. documentation.
  
  

Discuss

• Participate in the discussion forum.
  
  
• Check out the Cell Broadband Engine Architecture forum to get your technical questions about the processor answered. Juicy problems and answers from the forums are rounded up periodically and highlighted in the "Forum watch" blog series.
  
  
• Go to the Power Architecture blog for news, downloads, instructional resources, and event notifications for Cell/B.E. and other Power Architecture-related technologies. You can find the popular "Forum watch" blog series (Q&A roundup), the "FixIt" technology updates, and the Infobomb quick-read technology introductions.
  
  

About the author

Kane Scarlett is a technology journalist/analyst with 20 years in the business, working for such publishers as National Geographic, Population Reference Bureau, Miller Freeman, and IDG, and managing, editing, and writing for such august journals as JavaWorld, LinuxWorld, and of course, developerWorks.

Rate this article

Comments

Back to top

Table of contents

• Introduction
• Setting up the matrix
• Setting up the work block
• Setting up the accelerator code
• Conclusion
• Resources
• About the author
• Comments

IBM SmartCloud trial. No charge.

Unleash the power of hybrid cloud computing today!

---

Public and private solutions in one trial.

Special offers