greg/stasis-aries-wal
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

The linked list implementation now uses nested top actions. (Also, many bugfixes regarding nested top actions were introduced.)

Browse source
This commit is contained in:
Sears Russell 2005-01-14 10:08:10 +00:00
parent 360f0d15e2
commit 7a990886d2
13 changed files with 205 additions and 44 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
lladd/constants.h
View file

@ -78,8 +78,8 @@ terms specified in this license.
/* #define MAX_BUFFER_SIZE 100003 */
/*#define MAX_BUFFER_SIZE 20029 */
/*#define MAX_BUFFER_SIZE 10007 */
#define MAX_BUFFER_SIZE 5003
/*#define MAX_BUFFER_SIZE 2003 */
/*#define MAX_BUFFER_SIZE 5003*/
#define MAX_BUFFER_SIZE 2003
/* #define MAX_BUFFER_SIZE 71 */
/*#define MAX_BUFFER_SIZE 7 */
/*#define BUFFER_ASOOCIATIVE 2 */
@ -120,6 +120,8 @@ terms specified in this license.
#define OPERATION_UNDO_LINEAR_DELETE 26
#define OPERATION_SET_RANGE 27
#define OPERATION_SET_RANGE_INVERSE 28
#define OPERATION_LINKED_LIST_INSERT 29
#define OPERATION_LINKED_LIST_REMOVE 30
/* number above should be less than number below */
#define MAX_OPERATIONS 40

11
lladd/operations.h
View file

@ -76,6 +76,17 @@ typedef int (*Function)(int xid, Page * p, lsn_t lsn, recordid r, const void *d)
argument passed into the operation.
*/
#define SIZEOF_RECORD -1
/**
Logical log entries (such as those used by nested top actions
have a null recordid, as they are not assoicated with a specific page
If a log entry is not associated with a specific page, the page id can
be overloaded to hold the size of the associated log entry. Contrast
this with the description of SIZEOF_RECORD, which is used when the
operation uses a variable length argument, but is associated with
a specfic page.
*/
#define SIZEIS_PAGEID -2
/** If the Operation struct's undo field is set to this value, then
physical logging is used in lieu of logical logging.
*/

3
lladd/operations/linkedListNTA.h
View file

@ -32,5 +32,6 @@ lladd_linkedList_iterator * TlinkedListIterator(int xid, recordid list, int keyS
int TlinkedListNext(int xid, lladd_linkedList_iterator * it, byte ** key, int * keySize, byte ** value, int * valueSize);
recordid TlinkedListCreate(int xid, int keySize, int ValueSize);
void TlinkedListDelete(int xid, recordid list);
Operation getLinkedListInsert();
Operation getLinkedListRemove();
#endif //__LINKED_LIST_NTA_H

2
lladd/operations/nestedTopActions.h
View file

@ -61,6 +61,6 @@ terms specified in this license.
#include <lladd/operations.h>
void initNestedTopActions();
void TbeginNestedTopAction(int xid);
void TbeginNestedTopAction(int xid, int op, const byte* log_arguments, int log_arguments_length);
lsn_t TendNestedTopAction(int xid);
#endif

9
src/lladd/logger/logger2.c
View file

@ -48,6 +48,7 @@ terms specified in this license.
#include "page.h"
/*#include <lladd/bufferManager.h>*/
#include <stdio.h>
#include <assert.h>
TransactionLog LogTransBegin(int xid) {
TransactionLog tl;
tl.xid = xid;
@ -91,6 +92,9 @@ LogEntry * LogUpdate(TransactionLog * l, Page * p, recordid rid, int operation,
if(operationsTable[operation].sizeofData == SIZEOF_RECORD) {
argSize = rid.size;
} else if(operationsTable[operation].sizeofData == SIZEIS_PAGEID) {
argSize = rid.page;
// printf("argsize (page) %d, %d\n", argSize, sizeof(recordid) * 2 + sizeof(int) * 3);
} else {
argSize = operationsTable[operation].sizeofData;
}
@ -109,11 +113,6 @@ LogEntry * LogUpdate(TransactionLog * l, Page * p, recordid rid, int operation,
DEBUG("got preimage");
}
e = allocUpdateLogEntry(l->prevLSN, l->xid, operation, rid, args, argSize, preImage);
writeLogEntry(e);

10
src/lladd/operations.c
View file

@ -111,16 +111,20 @@ void undoUpdate(const LogEntry * e, Page * p, lsn_t clr_lsn) {
recordid rid = e->contents.update.rid;
#endif
/* lsn_t page_lsn = readLSN(e->contents.update.rid.page); */
lsn_t page_lsn = pageReadLSN(p);
if(e->LSN <= page_lsn) {
lsn_t page_lsn = -1;
if(p) {
page_lsn = pageReadLSN(p);
}
if(e->LSN <= page_lsn || !p) {
/* Actually execute the undo */
if(undo == NO_INVERSE) {
/* Physical undo */
assert(p); // Must be provided wiht a page in order to perform a physical undo!
DEBUG("OPERATION %d Physical undo, %ld {%d %d %ld}\n", undo, e->LSN, e->contents.update.rid.page, e->contents.update.rid.slot, e->contents.update.rid.size);
writeRecord(e->xid, p, clr_lsn, e->contents.update.rid, getUpdatePreImage(e));
} else if(undo == NO_INVERSE_WHOLE_PAGE) {
assert(p);
DEBUG("OPERATION %d Whole page physical undo, %ld {%d}\n", undo, e->LSN, e->contents.update.rid.page);
memcpy(p->memAddr, getUpdatePreImage(e), PAGE_SIZE);
pageWriteLSN(p, clr_lsn);

123
src/lladd/operations/linkedListNTA.c
View file

@ -28,8 +28,94 @@
@file
*/
static void __TlinkedListInsert(int xid, recordid list, const byte * key, int keySize, const byte * value, int valueSize);
typedef struct {
recordid list;
int keySize;
} lladd_linkedListInsert_log;
typedef struct {
recordid list;
int keySize;
int valueSize;
} lladd_linkedListRemove_log;
static int operateInsert(int xid, Page *p, lsn_t lsn, recordid rid, const void *dat) {
assert(!p);
lladd_linkedListRemove_log * log = (lladd_linkedListRemove_log*)dat;
byte * key;
byte * value;
int keySize, valueSize;
keySize = log->keySize;
valueSize = log->valueSize;
key = (byte*)(log+1);
value = ((byte*)(log+1))+keySize;
// printf("Operate insert called: rid.page = %d keysize = %d valuesize = %d %d {%d %d %d}\n", rid.page, log->keySize, log->valueSize, *(int*)key, value->page, value->slot, value->size);
// Skip writing the undo! Recovery will write a CLR after we're done, effectively
// wrapping this in a nested top action, so we needn't worry about that either.
__TlinkedListInsert(xid, log->list, key, keySize, value, valueSize);
return 0;
}
static int operateRemove(int xid, Page *p, lsn_t lsn, recordid rid, const void *dat) {
assert(!p);
lladd_linkedListRemove_log * log = (lladd_linkedListRemove_log*)dat;
byte * key;
int keySize;
keySize = log->keySize;
key = (byte*)(log+1);
// printf("Operate remove called: %d\n", *(int*)key);
// Don't call the version that writes an undo entry!
__TlinkedListRemove(xid, log->list, key, keySize);
return 0;
}
int TlinkedListInsert(int xid, recordid list, const byte * key, int keySize, const byte * value, int valueSize) {
int ret = TlinkedListRemove(xid, list, key, keySize);
lladd_linkedListInsert_log * undoLog = malloc(sizeof(lladd_linkedListInsert_log) + keySize);
undoLog->list = list;
undoLog->keySize = keySize;
memcpy(undoLog+1, key, keySize);
TbeginNestedTopAction(xid, OPERATION_LINKED_LIST_INSERT,
(byte*)undoLog, sizeof(lladd_linkedListInsert_log) + keySize);
__TlinkedListInsert(xid, list, key, keySize, value, valueSize);
TendNestedTopAction(xid);
return ret;
}
Operation getLinkedListInsert() {
Operation o = {
OPERATION_NOOP,
SIZEIS_PAGEID,
OPERATION_LINKED_LIST_REMOVE,
&operateInsert
};
return o;
}
Operation getLinkedListRemove() {
Operation o = {
OPERATION_NOOP,
SIZEIS_PAGEID,
OPERATION_LINKED_LIST_INSERT,
&operateRemove
};
return o;
}
static void __TlinkedListInsert(int xid, recordid list, const byte * key, int keySize, const byte * value, int valueSize) {
//int ret = TlinkedListRemove(xid, list, key, keySize);
lladd_linkedList_entry * entry = malloc(sizeof(lladd_linkedList_entry) + keySize + valueSize);
Tread(xid, list, entry);
if(!entry->next.size) {
@ -51,7 +137,7 @@ int TlinkedListInsert(int xid, recordid list, const byte * key, int keySize, con
free(newEntry);
}
free(entry);
return ret;
//return ret;
}
int TlinkedListFind(int xid, recordid list, const byte * key, int keySize, byte ** value) {
@ -81,7 +167,42 @@ int TlinkedListFind(int xid, recordid list, const byte * key, int keySize, byte
free(entry);
return -1;
}
static int __TlinkedListRemove(int xid, recordid list, const byte * key, int keySize);
int TlinkedListRemove(int xid, recordid list, const byte * key, int keySize) {
byte * value;
int valueSize;
int ret = TlinkedListFind(xid, list, key, keySize, &value);
if(ret != -1) {
valueSize = ret;
} else {
return 0;
}
int entrySize = sizeof(lladd_linkedListRemove_log) + keySize + valueSize;
lladd_linkedListRemove_log * undoLog = malloc(entrySize);
undoLog->list = list;
undoLog->keySize = keySize;
undoLog->valueSize = valueSize;
memcpy(undoLog+1, key, keySize);
memcpy(((byte*)(undoLog+1))+keySize, value, valueSize);
// printf("entry size %d sizeof(remove_log)%d keysize %d valuesize %d sizeof(rid) %d key %d value {%d %d %ld}\n",
// entrySize, sizeof(lladd_linkedListRemove_log), keySize, valueSize, sizeof(recordid), key, value->page, value->slot, value->size);
TbeginNestedTopAction(xid, OPERATION_LINKED_LIST_REMOVE,
(byte*)undoLog, entrySize);
free(value);
__TlinkedListRemove(xid, list, key, keySize);
TendNestedTopAction(xid);
return 1;
}
static int __TlinkedListRemove(int xid, recordid list, const byte * key, int keySize) {
lladd_linkedList_entry * entry = malloc(list.size);
Tread(xid, list, entry);

19
src/lladd/operations/nestedTopActions.c
View file

@ -53,7 +53,7 @@ terms specified in this license.
#include <string.h>
#include <malloc.h>
#include <pthread.h>
#include <assert.h>
/** @todo Remove extern declaration of transactional_2_mutex from nestedTopActions.c */
extern pthread_mutex_t transactional_2_mutex;
@ -72,13 +72,21 @@ void initNestedTopActions() {
NULLRID.slot = 0;
NULLRID.size = -1;
}
void TbeginNestedTopAction(int xid) {
/** @todo TbeginNestedTopAction's API might not be quite right.
Are there cases where we need to pass a recordid in?
*/
void TbeginNestedTopAction(int xid, int op, const byte * dat, int datSize) {
recordid rid = NULLRID;
rid.page = datSize;
LogEntry * e = LogUpdate(&XactionTable[xid % MAX_TRANSACTIONS], NULL, rid, op, dat);
DEBUG("Begin Nested Top Action e->LSN: %ld\n", e->LSN);
lsn_t * prevLSN = malloc(sizeof(lsn_t));
*prevLSN = e->LSN;
pthread_mutex_lock(&transactional_2_mutex);
*prevLSN = XactionTable[xid].prevLSN;
assert(!pblHtLookup(nestedTopActions, &xid, sizeof(int)));
pblHtInsert(nestedTopActions, &xid, sizeof(int), prevLSN);
pthread_mutex_unlock(&transactional_2_mutex);
free(e);
}
/**
@ -94,7 +102,7 @@ lsn_t TendNestedTopAction(int xid) {
pthread_mutex_lock(&transactional_2_mutex);
lsn_t * prevLSN = pblHtLookup(nestedTopActions, &xid, sizeof(int));
pblHtRemove(nestedTopActions, &xid, sizeof(int));
// This action wasn't really undone -- This is a nested top action!
lsn_t undoneLSN = XactionTable[xid].prevLSN;
recordid undoneRID = NULLRID; // Not correct, but this field is unused anyway. ;)
@ -111,6 +119,7 @@ lsn_t TendNestedTopAction(int xid) {
lsn_t ret = e->LSN;
free(e);
free(prevLSN);
pthread_mutex_unlock(&transactional_2_mutex);
return ret;

5
src/lladd/operations/noop.c
View file

@ -49,7 +49,10 @@ terms specified in this license.
#include "../page.h"
static int operate(int xid, Page *p, lsn_t lsn, recordid rid, const void *dat) {
pageWriteLSN(p, lsn);
/* If p is null, then this is a logical no-op that spans pages, so do nothing.
Otherwise, write the LSN to the appropriate page (to keep recovery happy)
and return */
if(p) pageWriteLSN(p, lsn);
return 0;
}

54
src/lladd/recovery2.c
View file

@ -210,29 +210,37 @@ static void Undo(int recovery) {
case UPDATELOG:
{
/* Need write lock for undo.. (Why??) */
Page * p = getPage(e->contents.update.rid.page, RO);
this_lsn= pageReadLSN(p); /* e->contents.update.rid.page); */
/* printf("1"); fflush(NULL); */
/* Sanity check. If this fails, something is wrong with the
redo phase or normal operation. */
assert(e->LSN <= this_lsn);
/* printf("1a"); fflush(NULL); */
/* Need to log a clr here. */
clr_lsn = LogCLR(e);
/* Undo update is a no-op if the page does not reflect this
update, but it will write the new clr_lsn if necessary. */
undoUpdate(e, p, clr_lsn);
/* printf("1b"); fflush(NULL); */
releasePage(p);
if(e->contents.update.rid.size != -1) {
Page * p = getPage(e->contents.update.rid.page, RO);
this_lsn= pageReadLSN(p); /* e->contents.update.rid.page); */
/* printf("1"); fflush(NULL); */
/* Sanity check. If this fails, something is wrong with the
redo phase or normal operation. */
assert(e->LSN <= this_lsn);
/* printf("1a"); fflush(NULL); */
/* Need to log a clr here. */
clr_lsn = LogCLR(e);
/* Undo update is a no-op if the page does not reflect this
update, but it will write the new clr_lsn if necessary. */
undoUpdate(e, p, clr_lsn);
/* printf("1b"); fflush(NULL); */
releasePage(p);
} else {
// The log entry is not associated with a particular page.
// (Therefore, it must be an idempotent logical log entry.)
clr_lsn = LogCLR(e);
undoUpdate(e, NULL, clr_lsn);
}
break;
}
case CLRLOG:

3
src/lladd/transactional2.c
View file

@ -67,6 +67,9 @@ void setupOperationsTable() {
operationsTable[OPERATION_SET_RANGE] = getSetRange();
operationsTable[OPERATION_SET_RANGE_INVERSE] = getSetRangeInverse();
operationsTable[OPERATION_LINKED_LIST_INSERT] = getLinkedListInsert();
operationsTable[OPERATION_LINKED_LIST_REMOVE] = getLinkedListRemove();
}

2
test/lladd/Makefile.am
View file

@ -6,6 +6,6 @@ else
TESTS =
endif
noinst_PROGRAMS = $(TESTS)
LDADD = @CHECK_LIBS@ $(top_builddir)/src/lladd/liblladd.a $(top_builddir)/src/pbl/libpbl.a $(top_builddir)/src/libdfa/librw.a #-lefence
LDADD = @CHECK_LIBS@ $(top_builddir)/src/lladd/liblladd.a $(top_builddir)/src/pbl/libpbl.a $(top_builddir)/src/libdfa/librw.a # -lefence
CLEANFILES = check_lht.log check_logEntry.log storefile.txt logfile.txt blob0_file.txt blob1_file.txt check_blobRecovery.log check_logWriter.log check_operations.log check_recovery.log check_transactional2.log check_page.log check_bufferManager.log check_indirect.log check_bufferMananger.log check_lladdhash.log check_pageOperations.log check_linearhash.log check_linkedListNTA.log check_linearHashNTA.log
AM_CFLAGS= -g -Wall -pedantic -std=gnu99

2
test/lladd/check_operations.c
View file

@ -368,7 +368,7 @@ START_TEST(operation_nestedTopAction) {
*dat = 10;
Tset(xid, rid1, dat);
TbeginNestedTopAction(xid);
TbeginNestedTopAction(xid, OPERATION_NOOP, NULL, 0);
*dat = 20;
Tset(xid, rid2, dat);