Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>You can exploit some minor parallelism for large register lengths and small sums. For instance, adding up 16 values of 1 byte (which happen to fit into one sse register) requires only log<sub>2</sub>16 additions and an equal number of shifts.<br> Not much, but faster than 15 depended additions and the additional memory accesses.</p> <pre><code>__m128i x = _mm_set_epi8(3,1,7,0,4,1,6,3,3,1,7,0,4,1,6,3); x = _mm_add_epi8(x, _mm_srli_si128(x, 1)); x = _mm_add_epi8(x, _mm_srli_si128(x, 2)); x = _mm_add_epi8(x, _mm_srli_si128(x, 4)); x = _mm_add_epi8(x, _mm_srli_si128(x, 8)); // x == 3, 4, 11, 11, 15, 16, 22, 25, 28, 29, 36, 36, 40, 41, 47, 50 </code></pre> <p>If you have longer sums, the dependencies could be hidden by exploiting instruction level parallelism and taking advantage of instruction reordering.</p> <p>Edit: something like</p> <pre><code>__m128i x0 = _mm_set_epi8(3,1,7,0,4,1,6,3,3,1,7,0,4,1,6,3); __m128i x1 = _mm_set_epi8(3,1,7,0,4,1,6,3,3,1,7,0,4,1,6,3); __m128i x2 = _mm_set_epi8(3,1,7,0,4,1,6,3,3,1,7,0,4,1,6,3); __m128i x3 = _mm_set_epi8(3,1,7,0,4,1,6,3,3,1,7,0,4,1,6,3); __m128i mask = _mm_set_epi8(0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0); x0 = _mm_add_epi8(x0, _mm_srli_si128(x0, 1)); x1 = _mm_add_epi8(x1, _mm_srli_si128(x1, 1)); x2 = _mm_add_epi8(x2, _mm_srli_si128(x2, 1)); x3 = _mm_add_epi8(x3, _mm_srli_si128(x3, 1)); x0 = _mm_add_epi8(x0, _mm_srli_si128(x0, 2)); x1 = _mm_add_epi8(x1, _mm_srli_si128(x1, 2)); x2 = _mm_add_epi8(x2, _mm_srli_si128(x2, 2)); x3 = _mm_add_epi8(x3, _mm_srli_si128(x3, 2)); x0 = _mm_add_epi8(x0, _mm_srli_si128(x0, 4)); x1 = _mm_add_epi8(x1, _mm_srli_si128(x1, 4)); x2 = _mm_add_epi8(x2, _mm_srli_si128(x2, 4)); x3 = _mm_add_epi8(x3, _mm_srli_si128(x3, 4)); x0 = _mm_add_epi8(x0, _mm_srli_si128(x0, 8)); x1 = _mm_add_epi8(x1, _mm_srli_si128(x1, 8)); x2 = _mm_add_epi8(x2, _mm_srli_si128(x2, 8)); x3 = _mm_add_epi8(x3, _mm_srli_si128(x3, 8)); x1 = _mm_add_epi8(_mm_shuffle_epi8(x0, mask), x1); x2 = _mm_add_epi8(_mm_shuffle_epi8(x1, mask), x2); x3 = _mm_add_epi8(_mm_shuffle_epi8(x2, mask), x3); </code></pre>
    singulars
    1. This table or related slice is empty.
    1. POSIMD prefix sum on Intel cpu
      singulars
      1. PTQuestion
    1. PTAnswer
    1. USGunther Piez
    1. USGunther Piez
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    1. COI'd love to hear more about the "long sums" scenario. How can you exploit instruction level parallelism?
      singulars
      1. PO
      1. USIgor ostrovsky
    2. CO@hirschhornsalz I don't understand the last three adds. I printed out the results. `x0 = [3 4 11 11 15 16 22 25 28 29 36 36 40 41 47 50]`. x1 should = x0 + 50 (the last element of x0). However your code is not doing that. It gives x1 = [6 8 22 22 30 32 44 50 56 58 72 72 80 82 94 100]. I think you want to broadcast the last element and add those.
      singulars
      1. PO
      1. USZ boson
    3. CO@redrum Yes, of course you are right. I edited the broadcast in (without testing, I hope I got it right^^).
      singulars
      1. PO
      1. USGunther Piez
 

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