Commit 80f27ff3 authored by mmakela's avatar mmakela

branches/innodb+: Merge revisions 6949:6970 from branches/zip,

skipping the revisions that do not make sense in branches/innodb+:
  ------------------------------------------------------------------------
  r6957 | vdimov | 2010-04-06 22:05:54 +0300 (Tue, 06 Apr 2010) | 2 lines
  Changed paths:
     M /branches/zip/trx/trx0trx.c

  branches/zip: Whitespace fixup
  ------------------------------------------------------------------------
  r6962 | mmakela | 2010-04-07 15:03:54 +0300 (Wed, 07 Apr 2010) | 2 lines
  Changed paths:
     M /branches/zip/handler/ha_innodb.cc

  branches/zip: innobase_init(): Correct the error message
  about wrong innodb_change_buffering value.  Reported by Ranger.
  ------------------------------------------------------------------------
  r6967 | vdimov | 2010-04-07 22:40:38 +0300 (Wed, 07 Apr 2010) | 2 lines
  Changed paths:
     M /branches/zip/include/univ.i

  branches/zip: Whitespace fixup in univ.i
  ------------------------------------------------------------------------
parent bd77a36e
...@@ -2281,7 +2281,7 @@ mem_free_and_error: ...@@ -2281,7 +2281,7 @@ mem_free_and_error:
} }
sql_print_error("InnoDB: invalid value " sql_print_error("InnoDB: invalid value "
"innodb_file_format_check=%s", "innodb_change_buffering=%s",
innobase_change_buffering); innobase_change_buffering);
goto mem_free_and_error; goto mem_free_and_error;
} }
......
...@@ -115,7 +115,7 @@ if we are compiling on Windows. */ ...@@ -115,7 +115,7 @@ if we are compiling on Windows. */
/* Include <sys/stat.h> to get S_I... macros defined for os0file.c */ /* Include <sys/stat.h> to get S_I... macros defined for os0file.c */
# include <sys/stat.h> # include <sys/stat.h>
# if !defined(__NETWARE__) && !defined(__WIN__) # if !defined(__NETWARE__) && !defined(__WIN__)
# include <sys/mman.h> /* mmap() for os0proc.c */ # include <sys/mman.h> /* mmap() for os0proc.c */
# endif # endif
......
...@@ -847,7 +847,7 @@ trx_commit_off_kernel( ...@@ -847,7 +847,7 @@ trx_commit_off_kernel(
recovery i.e.: back ground rollback thread is still active recovery i.e.: back ground rollback thread is still active
then there is a chance that the rollback thread may see then there is a chance that the rollback thread may see
this trx as COMMITTED_IN_MEMORY and goes adhead to clean it this trx as COMMITTED_IN_MEMORY and goes adhead to clean it
up calling trx_cleanup_at_db_startup(). This can happen up calling trx_cleanup_at_db_startup(). This can happen
in the case we are committing a trx here that is left in in the case we are committing a trx here that is left in
PREPARED state during the crash. Note that commit of the PREPARED state during the crash. Note that commit of the
rollback of a PREPARED trx happens in the recovery thread rollback of a PREPARED trx happens in the recovery thread
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment