Re: lynx-dev Re: [PATCH 2.8.5-dev14] *Really* large tables

[Ilya Zakharevich]

Leonid Pauzner wrote:

    >/* Experiments show that 2 is better than 1.5 is better than 1.25 (all by
    >   a small margin only)??? */
    >#define CELLS_GROWBY_FACTOR 2

Actually, this may be an artefact of the particular size of the table
(500K rows and 2 cells per row).  One needs to recheck with some other
sizes; there is no reason why 1.25 should not be much better in some
cases.

    IMHO, cell growby strategy should be as follows:
    the first row (maybe first and second) - cells_growby 16, init with 0;
    other rows - cells_growby 1, init with the previous row length.

There is no dependence between number of cells in a row and that of
the previous row.

Anyway, since cells are pooled, this is not relevant.  The macros you
quote are about pool allocation strategy, not cells-per-row array.

    Having in mind that number of cells in a row is not large, and limited by
    the screen width, the number of reallocs will be very small. And we will not
    allocate memory for 16 cells in a row if we need only a couple...

Number of cells in a row has nothing to do with the screen width.  It
is determined by HTML (row != line; rows retain the structure of HTML
table; same for cells, which may be multiline).

Hope this helps,
Ilya

; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to address@hidden