pgsql: Change the way we decide whether to give up on abbreviated text

Change the way we decide whether to give up on abbreviated text keys.

Be more aggressive about aborting early on if it looks like it's not
helping, but be less aggressive about aborting later on, since it's
more expensive at that point, and also since we're currently aborting
in some cases where abbreviation can still deliver a substantial win.

Peter Geoghegan. Extensive testing by Tomas Vondra.

Branch
------
master

Details
-------
http://git.postgresql.org/pg/commitdiff/f85155e18cb71a599724536e598e8d6f5e140454

Modified Files
--------------
src/backend/utils/adt/varlena.c |   45 +++++++++++++++++++++++++++++++++------
1 file changed, 39 insertions(+), 6 deletions(-)