- The problem was that Mrr_ordered_index_reader's interrupt_read() and resume_read() would 
  save and restore 1) index tuple  2) the rowid (as bytes returned by handler->position()).  Clustered 
  primary key columns were not saved/restored. 
  They are not explicitly present in the index tuple (i.e. table->key_info[secondary_key].key_parts 
  doesn't list them), but they are actually there, in particular 
  table->field[clustered_primary_key_member].part_of_key(secondary_key) == 1. Index condition pushdown
  code [correctly] uses the latter as inidication that pushed index condition can refer to clustered PK
  members. 

  The fix was to make interrupt_read()/resume_read() to save/restore clustered primary key members as well,
  so that we get correct values for them when evaluating pushed index condition.
[3rd attempt: remove the debugging aids, fix comments in testcase]

- Cause: handler::in_range_check_pushed_down was not reset when a 
  command would call handler->idx_cond_push() without later calling
  handler->index_end().
- Fix: reset the variable in handler->reset(), too (like we do with other
  Index Condition Pushdown members).

- Make Mrr_ordered_index_reader() save the rowid across scan interruptions

Also
- Fix compiler warning for setup_buffer_sizes()
- Add commented key_copy/key_restore for better handling of a similar issue
  with index record being destroyed by scan interruption (which causes 
  incorrect evaluation of pushed index condition later on).

- Testcase (but iself is no longer repeatable)

- Testcase (the bug itself was fixed by development on BKA side)

BUG#628785: multi_range_read.cc:430: int DsMrr_impl::dsmrr_init(): Assertion `do_sort_keys || do_rowid_fetch' failed 
- Make Ds_MrrImpl::check_cpk_scan() follow the execution code' logic: don't
  do MRR scans on clustered PK when mrr_sort_keys=off.

- Make index condition pushdown be controlled by an @@optimizer_switch flag,
  not by @@engine_condition_pushdown
- Make MRR buffer size be controlled by @@mrr_buffer_size, not 
  by @@read_rnd_buffer_size
- Move parts of code to separate files
- Code cleanup
- Add --sorted_result to some SELECTs in tests.

WL#2474 "Multi Range Read: Change the default MRR implementation to implement new MRR interface"
WL#2475 "Batched range read functions for MyISAM/InnoDb"
        "Index condition pushdown for MyISAM/InnoDB"
Igor's fix from sp1r-igor@olga.mysql.com-20080330055902-07614:
  There could be observed the following problems:
  1. EXPLAIN did not mention pushdown conditions from on expressions in the 
  'extra' column.  As a result if a query had no where conditions pushed 
  down to a table, but had on conditions pushed to this table the 'extra' 
  column in the EXPLAIN for the table missed 'using where'.
  2. Conditions for ref access were not eliminated from on expressions 
  though such conditions were eliminated from the where condition.