~percona-dev/percona-server/release-5.1.49-12-rnt-pre

# name       : innodb_lru_dump_restore.patch
# introduced : 11 or before
# maintainer : Yasufumi
#
#!!! notice !!!
# Any small change to this file in the main branch
# should be done or reviewed by the maintainer!
diff -Nur a/storage/innodb_plugin/buf/buf0lru.c b/storage/innodb_plugin/buf/buf0lru.c
--- a/storage/innodb_plugin/buf/buf0lru.c	2010-08-10 15:31:39.858241983 +0400
+++ b/storage/innodb_plugin/buf/buf0lru.c	2010-08-10 15:31:40.338242509 +0400
@@ -2122,6 +2122,218 @@
 	memset(&buf_LRU_stat_cur, 0, sizeof buf_LRU_stat_cur);
 }
 
+/********************************************************************//**
+Dump the LRU page list to the specific file. */
+#define LRU_DUMP_FILE "ib_lru_dump"
+
+UNIV_INTERN
+ibool
+buf_LRU_file_dump(void)
+/*===================*/
+{
+	os_file_t	dump_file = -1;
+	ibool		success;
+	byte*		buffer_base = NULL;
+	byte*		buffer = NULL;
+	buf_page_t*	bpage;
+	ulint		buffers;
+	ulint		offset;
+	ibool		ret = FALSE;
+	ulint		i;
+
+	for (i = 0; i < srv_n_data_files; i++) {
+		if (strstr(srv_data_file_names[i], LRU_DUMP_FILE) != NULL) {
+			fprintf(stderr,
+				" InnoDB: The name '%s' seems to be used for"
+				" innodb_data_file_path. Dumping LRU list is not"
+				" done for safeness.\n", LRU_DUMP_FILE);
+			goto end;
+		}
+	}
+
+	buffer_base = ut_malloc(2 * UNIV_PAGE_SIZE);
+	buffer = ut_align(buffer_base, UNIV_PAGE_SIZE);
+	if (!buffer) {
+		fprintf(stderr,
+			" InnoDB: cannot allocate buffer.\n");
+		goto end;
+	}
+
+	dump_file = os_file_create(LRU_DUMP_FILE, OS_FILE_OVERWRITE,
+				OS_FILE_NORMAL, OS_DATA_FILE, &success);
+	if (!success) {
+		os_file_get_last_error(TRUE);
+		fprintf(stderr,
+			" InnoDB: cannot open %s\n", LRU_DUMP_FILE);
+		goto end;
+	}
+
+	mutex_enter(&LRU_list_mutex);
+	bpage = UT_LIST_GET_LAST(buf_pool->LRU);
+
+	buffers = offset = 0;
+	while (bpage != NULL) {
+		if (offset == 0) {
+			memset(buffer, 0, UNIV_PAGE_SIZE);
+		}
+
+		mach_write_to_4(buffer + offset * 4, bpage->space);
+		offset++;
+		mach_write_to_4(buffer + offset * 4, bpage->offset);
+		offset++;
+
+		if (offset == UNIV_PAGE_SIZE/4) {
+			success = os_file_write(LRU_DUMP_FILE, dump_file, buffer,
+					(buffers << UNIV_PAGE_SIZE_SHIFT) & 0xFFFFFFFFUL,
+					(buffers >> (32 - UNIV_PAGE_SIZE_SHIFT)),
+					UNIV_PAGE_SIZE);
+			if (!success) {
+				mutex_exit(&LRU_list_mutex);
+				fprintf(stderr,
+					" InnoDB: cannot write page %lu of %s\n",
+					buffers, LRU_DUMP_FILE);
+				goto end;
+			}
+			buffers++;
+			offset = 0;
+		}
+
+		bpage = UT_LIST_GET_PREV(LRU, bpage);
+	}
+	mutex_exit(&LRU_list_mutex);
+
+	if (offset == 0) {
+		memset(buffer, 0, UNIV_PAGE_SIZE);
+	}
+
+	mach_write_to_4(buffer + offset * 4, 0xFFFFFFFFUL);
+	offset++;
+	mach_write_to_4(buffer + offset * 4, 0xFFFFFFFFUL);
+	offset++;
+
+	success = os_file_write(LRU_DUMP_FILE, dump_file, buffer,
+			(buffers << UNIV_PAGE_SIZE_SHIFT) & 0xFFFFFFFFUL,
+			(buffers >> (32 - UNIV_PAGE_SIZE_SHIFT)),
+			UNIV_PAGE_SIZE);
+	if (!success) {
+		goto end;
+	}
+
+	ret = TRUE;
+end:
+	if (dump_file != -1)
+		os_file_close(dump_file);
+	if (buffer_base)
+		ut_free(buffer_base);
+
+	return(ret);
+}
+/********************************************************************//**
+Read the pages based on the specific file.*/
+UNIV_INTERN
+ibool
+buf_LRU_file_restore(void)
+/*======================*/
+{
+	os_file_t	dump_file = -1;
+	ibool		success;
+	byte*		buffer_base = NULL;
+	byte*		buffer = NULL;
+	ulint		buffers;
+	ulint		offset;
+	ulint		reads = 0;
+	ulint		req = 0;
+	ibool		terminated = FALSE;
+	ibool		ret = FALSE;
+
+	buffer_base = ut_malloc(2 * UNIV_PAGE_SIZE);
+	buffer = ut_align(buffer_base, UNIV_PAGE_SIZE);
+	if (!buffer) {
+		fprintf(stderr,
+			" InnoDB: cannot allocate buffer.\n");
+		goto end;
+	}
+
+	dump_file = os_file_create_simple_no_error_handling(
+		LRU_DUMP_FILE, OS_FILE_OPEN, OS_FILE_READ_ONLY, &success);
+	if (!success) {
+		os_file_get_last_error(TRUE);
+		fprintf(stderr,
+			" InnoDB: cannot open %s\n", LRU_DUMP_FILE);
+		goto end;
+	}
+
+	buffers = 0;
+	while (!terminated) {
+		success = os_file_read(dump_file, buffer,
+				(buffers << UNIV_PAGE_SIZE_SHIFT) & 0xFFFFFFFFUL,
+				(buffers >> (32 - UNIV_PAGE_SIZE_SHIFT)),
+				UNIV_PAGE_SIZE);
+		if (!success) {
+			fprintf(stderr,
+				" InnoDB: cannot read page %lu of %s,"
+				" or meet unexpected terminal.",
+				buffers, LRU_DUMP_FILE);
+			goto end;
+		}
+
+		for (offset = 0; offset < UNIV_PAGE_SIZE/4; offset += 2) {
+			ulint	space_id, zip_size, page_no;
+			ulint	err;
+			ib_int64_t	tablespace_version;
+
+			space_id = mach_read_from_4(buffer + offset * 4);
+			page_no = mach_read_from_4(buffer + (offset + 1) * 4);
+			if (space_id == 0xFFFFFFFFUL
+			    || page_no == 0xFFFFFFFFUL) {
+				terminated = TRUE;
+				break;
+			}
+
+			if (offset % 16 == 15) {
+				os_aio_simulated_wake_handler_threads();
+				buf_flush_free_margin(FALSE);
+			}
+
+			zip_size = fil_space_get_zip_size(space_id);
+			if (UNIV_UNLIKELY(zip_size == ULINT_UNDEFINED)) {
+				continue;
+			}
+
+			if (fil_area_is_exist(space_id, zip_size, page_no, 0,
+					zip_size ? zip_size : UNIV_PAGE_SIZE)) {
+
+				tablespace_version = fil_space_get_version(space_id);
+
+				req++;
+				reads += buf_read_page_low(&err, FALSE, BUF_READ_ANY_PAGE
+						  | OS_AIO_SIMULATED_WAKE_LATER,
+						  space_id, zip_size, TRUE,
+						  tablespace_version, page_no, NULL);
+				buf_LRU_stat_inc_io();
+			}
+		}
+
+		buffers++;
+	}
+
+	os_aio_simulated_wake_handler_threads();
+	buf_flush_free_margin(FALSE);
+
+	ut_print_timestamp(stderr);
+	fprintf(stderr,
+		" InnoDB: reading pages based on the dumped LRU list was done."
+		" (requested: %lu, read: %lu)\n", req, reads);
+	ret = TRUE;
+end:
+	if (dump_file != -1)
+		os_file_close(dump_file);
+	if (buffer_base)
+		ut_free(buffer_base);
+
+	return(ret);
+}
+
 #if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
 /**********************************************************************//**
 Validates the LRU list.
diff -Nur a/storage/innodb_plugin/buf/buf0rea.c b/storage/innodb_plugin/buf/buf0rea.c
--- a/storage/innodb_plugin/buf/buf0rea.c	2010-08-10 15:31:39.858241983 +0400
+++ b/storage/innodb_plugin/buf/buf0rea.c	2010-08-10 15:31:40.338242509 +0400
@@ -56,7 +56,7 @@
 which case it is never read into the pool, or if the tablespace does
 not exist or is being dropped 
 @return 1 if read request is issued. 0 if it is not */
-static
+UNIV_INTERN
 ulint
 buf_read_page_low(
 /*==============*/
diff -Nur a/storage/innodb_plugin/fil/fil0fil.c b/storage/innodb_plugin/fil/fil0fil.c
--- a/storage/innodb_plugin/fil/fil0fil.c	2010-08-10 15:31:39.848242665 +0400
+++ b/storage/innodb_plugin/fil/fil0fil.c	2010-08-10 15:31:40.897004740 +0400
@@ -4735,6 +4735,78 @@
 	return(DB_SUCCESS);
 }
 
+/********************************************************************//**
+Confirm whether the parameters are valid or not */
+UNIV_INTERN
+ibool
+fil_area_is_exist(
+/*==============*/
+	ulint	space_id,	/*!< in: space id */
+	ulint	zip_size,	/*!< in: compressed page size in bytes;
+				0 for uncompressed pages */
+	ulint	block_offset,	/*!< in: offset in number of blocks */
+	ulint	byte_offset,	/*!< in: remainder of offset in bytes; in
+				aio this must be divisible by the OS block
+				size */
+	ulint	len)		/*!< in: how many bytes to read or write; this
+				must not cross a file boundary; in aio this
+				must be a block size multiple */
+{
+	fil_space_t*	space;
+	fil_node_t*	node;
+
+	/* Reserve the fil_system mutex and make sure that we can open at
+	least one file while holding it, if the file is not already open */
+
+	fil_mutex_enter_and_prepare_for_io(space_id);
+
+	space = fil_space_get_by_id(space_id);
+
+	if (!space) {
+		mutex_exit(&fil_system->mutex);
+		return(FALSE);
+	}
+
+	node = UT_LIST_GET_FIRST(space->chain);
+
+	for (;;) {
+		if (UNIV_UNLIKELY(node == NULL)) {
+			mutex_exit(&fil_system->mutex);
+			return(FALSE);
+		}
+
+		if (space->id != 0 && node->size == 0) {
+			/* We do not know the size of a single-table tablespace
+			before we open the file */
+
+			break;
+		}
+
+		if (node->size > block_offset) {
+			/* Found! */
+			break;
+		} else {
+			block_offset -= node->size;
+			node = UT_LIST_GET_NEXT(chain, node);
+		}
+	}
+
+	/* Open file if closed */
+	fil_node_prepare_for_io(node, fil_system, space);
+	fil_node_complete_io(node, fil_system, OS_FILE_READ);
+
+	/* Check that at least the start offset is within the bounds of a
+	single-table tablespace */
+	if (UNIV_UNLIKELY(node->size <= block_offset)
+	    && space->id != 0 && space->purpose == FIL_TABLESPACE) {
+		mutex_exit(&fil_system->mutex);
+		return(FALSE);
+	}
+
+	mutex_exit(&fil_system->mutex);
+	return(TRUE);
+}
+
 #ifndef UNIV_HOTBACKUP
 /**********************************************************************//**
 Waits for an aio operation to complete. This function is used to write the
diff -Nur a/storage/innodb_plugin/handler/innodb_patch_info.h b/storage/innodb_plugin/handler/innodb_patch_info.h
--- a/storage/innodb_plugin/handler/innodb_patch_info.h	2010-08-10 15:31:39.848242665 +0400
+++ b/storage/innodb_plugin/handler/innodb_patch_info.h	2010-08-10 15:31:40.907001824 +0400
@@ -41,5 +41,6 @@
 {"innodb_admin_command_base","XtraDB specific command interface through i_s","","http://www.percona.com/docs/wiki/percona-xtradb"},
 {"innodb_show_lock_name","Show mutex/lock name instead of crated file/line","","http://www.percona.com/docs/wiki/percona-xtradb"},
 {"innodb_extend_slow","Extended statistics in slow.log","It is InnoDB-part only. It needs to patch also to mysqld.","http://www.percona.com/docs/wiki/percona-xtradb"},
+{"innodb_lru_dump_restore","Dump and restore command for content of buffer pool","","http://www.percona.com/docs/wiki/percona-xtradb"},
 {NULL, NULL, NULL, NULL}
 };
diff -Nur a/storage/innodb_plugin/handler/i_s.cc b/storage/innodb_plugin/handler/i_s.cc
--- a/storage/innodb_plugin/handler/i_s.cc	2010-08-10 15:31:39.848242665 +0400
+++ b/storage/innodb_plugin/handler/i_s.cc	2010-08-10 15:31:40.907001824 +0400
@@ -46,6 +46,7 @@
 #include "trx0rseg.h" /* for trx_rseg_struct */
 #include "trx0sys.h" /* for trx_sys */
 #include "dict0dict.h" /* for dict_sys */
+#include "buf0lru.h" /* for XTRA_LRU_[DUMP/RESTORE] */
 }
 
 static const char plugin_author[] = "Innobase Oy";
@@ -3040,6 +3041,36 @@
 			"Hello!");
 		goto end_func;
 	}
+	else if (!strncasecmp("XTRA_LRU_DUMP", ptr, 13)) {
+		ut_print_timestamp(stderr);
+		fprintf(stderr, " InnoDB: administration command 'XTRA_LRU_DUMP'"
+				" was detected.\n");
+
+		if (buf_LRU_file_dump()) {
+			field_store_string(i_s_table->field[0],
+				"XTRA_LRU_DUMP was succeeded.");
+		} else {
+			field_store_string(i_s_table->field[0],
+				"XTRA_LRU_DUMP was failed.");
+		}
+
+		goto end_func;
+	}
+	else if (!strncasecmp("XTRA_LRU_RESTORE", ptr, 16)) {
+		ut_print_timestamp(stderr);
+		fprintf(stderr, " InnoDB: administration command 'XTRA_LRU_RESTORE'"
+				" was detected.\n");
+
+		if (buf_LRU_file_restore()) {
+			field_store_string(i_s_table->field[0],
+				"XTRA_LRU_RESTORE was succeeded.");
+		} else {
+			field_store_string(i_s_table->field[0],
+				"XTRA_LRU_RESTORE was failed.");
+		}
+
+		goto end_func;
+	}
 
 	field_store_string(i_s_table->field[0],
 		"Undefined XTRA_* command.");
diff -Nur a/storage/innodb_plugin/include/buf0lru.h b/storage/innodb_plugin/include/buf0lru.h
--- a/storage/innodb_plugin/include/buf0lru.h	2010-08-10 15:31:39.848242665 +0400
+++ b/storage/innodb_plugin/include/buf0lru.h	2010-08-10 15:31:40.907001824 +0400
@@ -213,6 +213,18 @@
 void
 buf_LRU_stat_update(void);
 /*=====================*/
+/********************************************************************//**
+Dump the LRU page list to the specific file. */
+UNIV_INTERN
+ibool
+buf_LRU_file_dump(void);
+/*===================*/
+/********************************************************************//**
+Read the pages based on the specific file.*/
+UNIV_INTERN
+ibool
+buf_LRU_file_restore(void);
+/*======================*/
 
 #if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
 /**********************************************************************//**
diff -Nur a/storage/innodb_plugin/include/buf0rea.h b/storage/innodb_plugin/include/buf0rea.h
--- a/storage/innodb_plugin/include/buf0rea.h	2010-08-10 15:31:39.848242665 +0400
+++ b/storage/innodb_plugin/include/buf0rea.h	2010-08-10 15:31:40.907001824 +0400
@@ -31,6 +31,37 @@
 #include "buf0types.h"
 
 /********************************************************************//**
+Low-level function which reads a page asynchronously from a file to the
+buffer buf_pool if it is not already there, in which case does nothing.
+Sets the io_fix flag and sets an exclusive lock on the buffer frame. The
+flag is cleared and the x-lock released by an i/o-handler thread.
+@return 1 if a read request was queued, 0 if the page already resided
+in buf_pool, or if the page is in the doublewrite buffer blocks in
+which case it is never read into the pool, or if the tablespace does
+not exist or is being dropped 
+@return 1 if read request is issued. 0 if it is not */
+UNIV_INTERN
+ulint
+buf_read_page_low(
+/*==============*/
+	ulint*	err,	/*!< out: DB_SUCCESS or DB_TABLESPACE_DELETED if we are
+			trying to read from a non-existent tablespace, or a
+			tablespace which is just now being dropped */
+	ibool	sync,	/*!< in: TRUE if synchronous aio is desired */
+	ulint	mode,	/*!< in: BUF_READ_IBUF_PAGES_ONLY, ...,
+			ORed to OS_AIO_SIMULATED_WAKE_LATER (see below
+			at read-ahead functions) */
+	ulint	space,	/*!< in: space id */
+	ulint	zip_size,/*!< in: compressed page size, or 0 */
+	ibool	unzip,	/*!< in: TRUE=request uncompressed page */
+	ib_int64_t tablespace_version, /*!< in: if the space memory object has
+			this timestamp different from what we are giving here,
+			treat the tablespace as dropped; this is a timestamp we
+			use to stop dangling page reads from a tablespace
+			which we have DISCARDed + IMPORTed back */
+	ulint	offset,	/*!< in: page number */
+	trx_t*	trx);
+/********************************************************************//**
 High-level function which reads a page asynchronously from a file to the
 buffer buf_pool if it is not already there. Sets the io_fix flag and sets
 an exclusive lock on the buffer frame. The flag is cleared and the x-lock
diff -Nur a/storage/innodb_plugin/include/fil0fil.h b/storage/innodb_plugin/include/fil0fil.h
--- a/storage/innodb_plugin/include/fil0fil.h	2010-08-10 15:31:39.848242665 +0400
+++ b/storage/innodb_plugin/include/fil0fil.h	2010-08-10 15:31:40.907001824 +0400
@@ -643,6 +643,22 @@
 	void*	message,	/*!< in: message for aio handler if non-sync
 				aio used, else ignored */
 	trx_t*	trx);
+/********************************************************************//**
+Confirm whether the parameters are valid or not */
+UNIV_INTERN
+ibool
+fil_area_is_exist(
+/*==============*/
+	ulint	space_id,	/*!< in: space id */
+	ulint	zip_size,	/*!< in: compressed page size in bytes;
+				0 for uncompressed pages */
+	ulint	block_offset,	/*!< in: offset in number of blocks */
+	ulint	byte_offset,	/*!< in: remainder of offset in bytes; in
+				aio this must be divisible by the OS block
+				size */
+	ulint	len);		/*!< in: how many bytes to read or write; this
+				must not cross a file boundary; in aio this
+				must be a block size multiple */
 /**********************************************************************//**
 Waits for an aio operation to complete. This function is used to write the
 handler for completed requests. The aio array of pending requests is divided