Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>The most likely cause of the speed improvement is that:</p> <ul> <li>inserting a MOV shifts the subsequent instructions to different memory addresses</li> <li>one of those moved instructions was an important conditional branch</li> <li>that branch was being incorrectly predicted due to aliasing in the branch prediction table</li> <li>moving the branch eliminated the alias and allowed the branch to be predicted correctly</li> </ul> <p>Your Core2 doesn't keep a separate history record for each conditional jump. Instead it keeps a shared history of all conditional jumps. One disadvantage of <a href="http://en.wikipedia.org/wiki/Branch_predictor#Global_branch_prediction" rel="noreferrer">global branch prediction</a> is that the history is diluted by irrelevant information if the different conditional jumps are uncorrelated.</p> <p>This little <a href="http://www.ece.unm.edu/~jimp/611/slides/chap4_5.html" rel="noreferrer">branch prediction tutorial</a> shows how branch prediction buffers work. The cache buffer is indexed by the lower portion of the address of the branch instruction. This works well unless two important uncorrelated branches share the same lower bits. In that case, you end-up with aliasing which causes many mispredicted branches (which stalls the instruction pipeline and slowing your program).</p> <p>If you want to understand how branch mispredictions affect performance, take a look at this excellent answer: <a href="https://stackoverflow.com/a/11227902/1001643">https://stackoverflow.com/a/11227902/1001643</a></p> <p>Compilers typically don't have enough information to know which branches will alias and whether those aliases will be significant. However, that information can be determined at runtime with tools such as <a href="http://valgrind.org/docs/manual/cg-manual.html" rel="noreferrer">Cachegrind</a> and <a href="http://software.intel.com/en-us/forums/topic/392268" rel="noreferrer">VTune</a>.</p>
    singulars
    1. This table or related slice is empty.
    1. POWhy would introducing useless MOV instructions speed up a tight loop in x86_64 assembly?
      singulars
      1. PTQuestion
    1. PTAnswer
    1. USCommunity
    1. USRaymond Hettinger
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. POWhy would introducing useless MOV instructions speed up a tight loop in x86_64 assembly?
      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. 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. COHmm. This sounds promising. The only conditional branches in this sha256 implementation are the checks for the end of the FOR loops. At the time, I had tagged this revision as an oddity in git and continued optimizing. One of my next steps was to rewrite the pascal FOR loop myself in assembly, at which point these extra instructions no longer had a positive effect. Perhaps free pascal's generated code was harder for the processor to predict than the simple counter I replaced it with.
      singulars
      1. PO
      1. UStangentstorm
    2. CO@tangentstorm That sounds like a good summary. The branch prediction table isn't very large, so one table entry might refer to more than one branch. This can render some predictions useless. The problem is easily fixed if one of the conflicting branches moves to another part of the table. Almost any little change can make this happen :-)
      singulars
      1. PO
      1. USRaymond Hettinger
    3. COI think this is the most reasonable explanation of the specific behavior I observed, so I'm going to mark this as the answer. Thanks. :)
      singulars
      1. PO
      1. UStangentstorm
 

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