bzr push into mysql-trunk branch (hemant.hk.kumar:3707 to 3708)

[Hemant Kumar]

 3708 Hemant Kumar	2012-04-20
      making main.bug12427262 test experimental on windows.

    modified:
      mysql-test/collections/default.experimental
 3707 Vasil Dimov	2012-04-20
      Fix Bug#13946118 STATS RELATED, ASSERT (RW_LATCH == 1) || (RW_LATCH == 2) || (RW_LATCH == 3)
      
      This bug was introduced in vasil.dimov@stripped
      
      In btr_pcur_open_low() we shold clear the newly added flag
      BTR_ALREADY_S_LATCHED from latch_mode because later in buf_page_get_gen()
      we assert that latch_mode is one of RW_S_LATCH, RW_X_LATCH or RW_NO_LATCH.
      
      This was discussed with Marko and is fixed in WL#6175 (which is not pushed
      yet). I am not sure why it was fixed there instead of mysql-trunk, since
      the problem exists in mysql-trunk.

    modified:
      storage/innobase/btr/btr0cur.cc
      storage/innobase/include/btr0btr.h
      storage/innobase/include/btr0pcur.ic
=== modified file 'mysql-test/collections/default.experimental'
--- a/mysql-test/collections/default.experimental	2012-03-27 09:55:06 +0000
+++ b/mysql-test/collections/default.experimental	2012-04-20 11:32:32 +0000
@@ -13,6 +13,7 @@ main.sp @solaris
 main.bug33509 @solaris                   # Bug#11753919 2012-02-16 didrik Several test cases fail on Solaris with error Thread stack overrun
 
 main.kill @freebsd                       # Bug#12619719 2011-08-04 Occasional failure in PB2
+main.bug12427262 @windows                # Bug#12427262 2012-04-20 Hemant Fails on mysql-trunk on Windows after Mayank push
 
 innodb.innodb_monitor                    # Bug#12320827 2011-08-04 Occasional failure in PB2
 

No bundle (reason: useless for push emails).