On Tue, May 20, 2008 at 12:26 AM, funofnet Funofnet <[EMAIL PROTECTED]> wrote:
> Hi,
>
> this is where I suspect the bug : LinkEstimatorP.nc
>
> **********************************
>
> j = 0;
> newPrevSentIdx = 0;
>
> NEIGHBOR_TABLE_SIZE = 10 // I added to have all the constants mentionned
> for (i = 0; i < NEIGHBOR_TABLE_SIZE && j < maxEntries; i++) {
> k = (prevSentIdx + i + 1) % NEIGHBOR_TABLE_SIZE;
> if ((NeighborTable[k].flags & VALID_ENTRY) &&
> (NeighborTable[k].flags & MATURE_ENTRY)) {
> footer->neighborList[j].ll_addr = NeighborTable[k].ll_addr;
> footer->neighborList[j].inquality = NeighborTable[k].inquality;
> newPrevSentIdx = k;
> dbg("LI", "Loaded on footer: %d %d %d\n", j,
> footer->neighborList[j].ll_addr,
> footer->neighborList[j].inquality);
> j++;
> }
> }
>
> ***********************************
>
> Normally footer->neighborList[] should contain distinct neighbour but in the
> case where the second and last element in the
>
> NeighborTable are mature and valid only element index 1 will be insert.
> Isn't it?
>
> I explain more:
>
>
>
> k = 0 + 0 + 1 % 10 = 1 (NeighborTable[1] is mature and valid) then
>
> footer->neighborList[0].ll_addr = NeighborTable[1].ll_addr;
> footer->neighborList[0].inquality = NeighborTable[1].inquality;
>
>
>
> then we continue until * i = 9 *
>
> k = 1 + 9 + 1 = 11 % 10 = 1 (NeighborTable[1] is mature and valid) then
You are updating prevSentIdx in the expression above to 1 - that
happens only after you exit the for loop. The correct expression
should be 0 + 9 + 1 = 10 % 10 = 0 so I don't think this is a bug.
Does this make sense?
- om_p
_______________________________________________
Tinyos-help mailing list
[email protected]
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
