void
smgr_redo(XLogRecPtr lsn, XLogRecord *record)
{
uint8 info = record->xl_info & ~XLR_INFO_MASK;
/* Backup blocks are not used in smgr records */
Assert(!(record->xl_info & XLR_BKP_BLOCK_MASK));
if (info == XLOG_SMGR_CREATE)
{
xl_smgr_create *xlrec = (xl_smgr_create *) XLogRecGetData(record);
SMgrRelation reln;
reln = smgropen(xlrec->rnode);
smgrcreate(reln, MAIN_FORKNUM, true);
}
else if (info == XLOG_SMGR_TRUNCATE)
{
xl_smgr_truncate *xlrec = (xl_smgr_truncate *) XLogRecGetData(record);
SMgrRelation reln;
Relation rel;
reln = smgropen(xlrec->rnode);
/*
* Forcibly create relation if it doesn't exist (which suggests that
* it was dropped somewhere later in the WAL sequence). As in
* XLogOpenRelation, we prefer to recreate the rel and replay the log
* as best we can until the drop is seen.
*/
smgrcreate(reln, MAIN_FORKNUM, true);
smgrtruncate(reln, MAIN_FORKNUM, xlrec->blkno, false);
/* Also tell xlogutils.c about it */
XLogTruncateRelation(xlrec->rnode, MAIN_FORKNUM, xlrec->blkno);
/* Truncate FSM and VM too */
rel = CreateFakeRelcacheEntry(xlrec->rnode);
if (smgrexists(reln, FSM_FORKNUM))
FreeSpaceMapTruncateRel(rel, xlrec->blkno);
if (smgrexists(reln, VISIBILITYMAP_FORKNUM))
visibilitymap_truncate(rel, xlrec->blkno);
FreeFakeRelcacheEntry(rel);
}
else
elog(PANIC, "smgr_redo: unknown op code %u", info);
}
/*
* Ensure that the visibility map fork is at least vm_nblocks long, extending
* it if necessary with zeroed pages.
*/
static void
vm_extend(Relation rel, BlockNumber vm_nblocks)
{
BlockNumber vm_nblocks_now;
Page pg;
pg = (Page) palloc(BLCKSZ);
PageInit(pg, BLCKSZ, 0);
/*
* We use the relation extension lock to lock out other backends trying to
* extend the visibility map at the same time. It also locks out extension
* of the main fork, unnecessarily, but extending the visibility map
* happens seldom enough that it doesn't seem worthwhile to have a
* separate lock tag type for it.
*
* Note that another backend might have extended or created the relation
* by the time we get the lock.
*/
LockRelationForExtension(rel, ExclusiveLock);
/* Might have to re-open if a cache flush happened */
RelationOpenSmgr(rel);
/*
* Create the file first if it doesn't exist. If smgr_vm_nblocks is
* positive then it must exist, no need for an smgrexists call.
*/
if ((rel->rd_smgr->smgr_vm_nblocks == 0 ||
rel->rd_smgr->smgr_vm_nblocks == InvalidBlockNumber) &&
!smgrexists(rel->rd_smgr, VISIBILITYMAP_FORKNUM))
smgrcreate(rel->rd_smgr, VISIBILITYMAP_FORKNUM, false);
vm_nblocks_now = smgrnblocks(rel->rd_smgr, VISIBILITYMAP_FORKNUM);
/* Now extend the file */
while (vm_nblocks_now < vm_nblocks)
{
smgrextend(rel->rd_smgr, VISIBILITYMAP_FORKNUM, vm_nblocks_now,
(char *) pg, false);
vm_nblocks_now++;
}
/*
* Send a shared-inval message to force other backends to close any smgr
* references they may have for this rel, which we are about to change.
* This is a useful optimization because it means that backends don't have
* to keep checking for creation or extension of the file, which happens
* infrequently.
*/
CacheInvalidateSmgr(rel->rd_smgr->smgr_rnode);
/* Update local cache with the up-to-date size */
rel->rd_smgr->smgr_vm_nblocks = vm_nblocks_now;
UnlockRelationForExtension(rel, ExclusiveLock);
pfree(pg);
}
/*
* FreeSpaceMapTruncateRel - adjust for truncation of a relation.
*
* The caller must hold AccessExclusiveLock on the relation, to ensure that
* other backends receive the smgr invalidation event that this function sends
* before they access the FSM again.
*
* nblocks is the new___ size of the heap.
*/
void
FreeSpaceMapTruncateRel(Relation rel, BlockNumber nblocks)
{
BlockNumber new_nfsmblocks;
FSMAddress first_removed_address;
uint16 first_removed_slot;
Buffer buf;
RelationOpenSmgr(rel);
/*
* If no FSM has been created yet for this relation, there's nothing to
* truncate.
*/
if (!smgrexists(rel->rd_smgr, FSM_FORKNUM))
return;
/* Get the location in the FSM of the first removed heap block */
first_removed_address = fsm_get_location(nblocks, &first_removed_slot);
/*
* Zero out the tail of the last remaining FSM page. If the slot
* representing the first removed heap block is at a page boundary, as the
* first slot on the FSM page that first_removed_address points to, we can
* just truncate that page altogether.
*/
if (first_removed_slot > 0)
{
buf = fsm_readbuf(rel, first_removed_address, false);
if (!BufferIsValid(buf))
return; /* nothing to do; the FSM was already smaller */
LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
fsm_truncate_avail(BufferGetPage(buf), first_removed_slot);
MarkBufferDirtyHint(buf, false);
UnlockReleaseBuffer(buf);
new_nfsmblocks = fsm_logical_to_physical(first_removed_address) + 1;
}
else
{
new_nfsmblocks = fsm_logical_to_physical(first_removed_address);
if (smgrnblocks(rel->rd_smgr, FSM_FORKNUM) <= new_nfsmblocks)
return; /* nothing to do; the FSM was already smaller */
}
/* Truncate the unused FSM pages, and send smgr inval message */
smgrtruncate(rel->rd_smgr, FSM_FORKNUM, new_nfsmblocks);
/*
* We might as well update the local smgr_fsm_nblocks setting.
* smgrtruncate sent an smgr cache inval message, which will cause other
* backends to invalidate their copy of smgr_fsm_nblocks, and this one too
* at the next command boundary. But this ensures it isn't outright wrong
* until then.
*/
if (rel->rd_smgr)
rel->rd_smgr->smgr_fsm_nblocks = new_nfsmblocks;
}
/*
* XLogRecordPageWithFreeSpace - like RecordPageWithFreeSpace, for use in
* WAL replay
*/
void
XLogRecordPageWithFreeSpace(RelFileNode rnode, BlockNumber heapBlk,
Size spaceAvail)
{
int new_cat = fsm_space_avail_to_cat(spaceAvail);
FSMAddress addr;
uint16 slot;
BlockNumber blkno;
Buffer buf;
Page page;
bool write_to_fsm;
/* This is meant to mirror the logic in fsm_allow_writes() */
if (heapBlk >= HEAP_FSM_CREATION_THRESHOLD)
write_to_fsm = true;
else
{
/* Open the relation at smgr level */
SMgrRelation smgr = smgropen(rnode, InvalidBackendId);
if (smgrexists(smgr, FSM_FORKNUM))
write_to_fsm = true;
else
{
BlockNumber heap_nblocks = smgrnblocks(smgr, MAIN_FORKNUM);
if (heap_nblocks > HEAP_FSM_CREATION_THRESHOLD)
write_to_fsm = true;
else
write_to_fsm = false;
}
}
if (!write_to_fsm)
return;
/* Get the location of the FSM byte representing the heap block */
addr = fsm_get_location(heapBlk, &slot);
blkno = fsm_logical_to_physical(addr);
/* If the page doesn't exist already, extend */
buf = XLogReadBufferExtended(rnode, FSM_FORKNUM, blkno, RBM_ZERO_ON_ERROR);
LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
page = BufferGetPage(buf);
if (PageIsNew(page))
PageInit(page, BLCKSZ, 0);
if (fsm_set_avail(page, slot, new_cat))
MarkBufferDirtyHint(buf, false);
UnlockReleaseBuffer(buf);
}
/*
* Ensure that the FSM fork is at least fsm_nblocks long, extending
* it if necessary with empty pages. And by empty, I mean pages filled
* with zeros, meaning there's no free space.
*/
static void
fsm_extend(Relation rel, BlockNumber fsm_nblocks)
{
BlockNumber fsm_nblocks_now;
Page pg;
pg = (Page) palloc(BLCKSZ);
PageInit(pg, BLCKSZ, 0);
/*
* We use the relation extension lock to lock out other backends trying to
* extend the FSM at the same time. It also locks out extension of the
* main fork, unnecessarily, but extending the FSM happens seldom enough
* that it doesn't seem worthwhile to have a separate lock tag type for
* it.
*
* Note that another backend might have extended or created the relation
* by the time we get the lock.
*/
LockRelationForExtension(rel, ExclusiveLock);
/* Might have to re-open if a cache flush happened */
RelationOpenSmgr(rel);
/*
* Create the FSM file first if it doesn't exist. If smgr_fsm_nblocks is
* positive then it must exist, no need for an smgrexists call.
*/
if ((rel->rd_smgr->smgr_fsm_nblocks == 0 ||
rel->rd_smgr->smgr_fsm_nblocks == InvalidBlockNumber) &&
!smgrexists(rel->rd_smgr, FSM_FORKNUM))
smgrcreate(rel->rd_smgr, FSM_FORKNUM, false);
fsm_nblocks_now = smgrnblocks(rel->rd_smgr, FSM_FORKNUM);
while (fsm_nblocks_now < fsm_nblocks)
{
PageSetChecksumInplace(pg, fsm_nblocks_now);
smgrextend(rel->rd_smgr, FSM_FORKNUM, fsm_nblocks_now,
(char *) pg, false);
fsm_nblocks_now++;
}
/* Update local cache with the up-to-date size */
rel->rd_smgr->smgr_fsm_nblocks = fsm_nblocks_now;
UnlockRelationForExtension(rel, ExclusiveLock);
pfree(pg);
}
/*
* For heaps, we prevent creation of the FSM unless the number of pages
* exceeds HEAP_FSM_CREATION_THRESHOLD. For tables that don't already have
* a FSM, this will save an inode and a few kB of space.
*
* XXX The API is a little awkward -- if the caller passes a valid nblocks
* value, it can avoid invoking a system call. If the caller passes
* InvalidBlockNumber and receives a false return value, it can get an
* up-to-date relation size from get_nblocks. This saves a few cycles in
* the caller, which would otherwise need to get the relation size by itself.
*/
static bool
fsm_allow_writes(Relation rel, BlockNumber heapblk,
BlockNumber nblocks, BlockNumber *get_nblocks)
{
bool skip_get_nblocks;
if (heapblk >= HEAP_FSM_CREATION_THRESHOLD)
return true;
/* Non-heap rels can always create a FSM. */
if (rel->rd_rel->relkind != RELKIND_RELATION &&
rel->rd_rel->relkind != RELKIND_TOASTVALUE)
return true;
/*
* If the caller knows nblocks, we can avoid a system call later. If it
* doesn't, maybe we have relpages from a previous VACUUM. Since the table
* may have extended since then, we still have to count the pages later if
* we can't return now.
*/
if (nblocks != InvalidBlockNumber)
{
if (nblocks > HEAP_FSM_CREATION_THRESHOLD)
return true;
else
skip_get_nblocks = true;
}
else
{
if (rel->rd_rel->relpages != InvalidBlockNumber &&
rel->rd_rel->relpages > HEAP_FSM_CREATION_THRESHOLD)
return true;
else
skip_get_nblocks = false;
}
RelationOpenSmgr(rel);
if (smgrexists(rel->rd_smgr, FSM_FORKNUM))
return true;
if (skip_get_nblocks)
return false;
/* last resort */
*get_nblocks = RelationGetNumberOfBlocks(rel);
if (*get_nblocks > HEAP_FSM_CREATION_THRESHOLD)
return true;
else
return false;
}
/*
* smgrDoPendingDeletes() -- Take care of relation deletes at end of xact.
*
* This also runs when aborting a subxact; we want to clean up a failed
* subxact immediately.
*/
void
smgrDoPendingDeletes(bool isCommit)
{
int nestLevel = GetCurrentTransactionNestLevel();
PendingRelDelete *pending;
PendingRelDelete *prev;
PendingRelDelete *next;
prev = NULL;
for (pending = pendingDeletes; pending != NULL; pending = next)
{
next = pending->next;
if (pending->nestLevel < nestLevel)
{
/* outer-level entries should not be processed yet */
prev = pending;
}
else
{
/* unlink list entry first, so we don't retry on failure */
if (prev)
prev->next = next;
else
pendingDeletes = next;
/* do deletion if called for */
if (pending->atCommit == isCommit)
{
int i;
/* schedule unlinking old files */
SMgrRelation srel;
srel = smgropen(pending->relnode);
for (i = 0; i <= MAX_FORKNUM; i++)
{
if (smgrexists(srel, i))
smgrdounlink(srel,
i,
pending->isTemp,
false);
}
smgrclose(srel);
}
/* must explicitly free the list entry */
pfree(pending);
/* prev does not change */
}
}
}
/*
* Read a visibility map page.
*
* If the page doesn't exist, InvalidBuffer is returned, or if 'extend' is
* true, the visibility map file is extended.
*/
static Buffer
vm_readbuf(Relation rel, BlockNumber blkno, bool extend)
{
Buffer buf;
/*
* We might not have opened the relation at the smgr level yet, or we
* might have been forced to close it by a sinval message. The code below
* won't necessarily notice relation extension immediately when extend =
* false, so we rely on sinval messages to ensure that our ideas about the
* size of the map aren't too far out of date.
*/
RelationOpenSmgr(rel);
/*
* If we haven't cached the size of the visibility map fork yet, check it
* first.
*/
if (rel->rd_smgr->smgr_vm_nblocks == InvalidBlockNumber)
{
if (smgrexists(rel->rd_smgr, VISIBILITYMAP_FORKNUM))
rel->rd_smgr->smgr_vm_nblocks = smgrnblocks(rel->rd_smgr,
VISIBILITYMAP_FORKNUM);
else
rel->rd_smgr->smgr_vm_nblocks = 0;
}
/* Handle requests beyond EOF */
if (blkno >= rel->rd_smgr->smgr_vm_nblocks)
{
if (extend)
vm_extend(rel, blkno + 1);
else
return InvalidBuffer;
}
/*
* Use ZERO_ON_ERROR mode, and initialize the page if necessary. It's
* always safe to clear bits, so it's better to clear corrupt pages than
* error out.
*/
buf = ReadBufferExtended(rel, VISIBILITYMAP_FORKNUM, blkno,
RBM_ZERO_ON_ERROR, NULL);
if (PageIsNew(BufferGetPage(buf)))
PageInit(BufferGetPage(buf), BLCKSZ, 0);
return buf;
}
/*
* Ensure that the visibility map fork is at least vm_nblocks long, extending
* it if necessary with zeroed pages.
*/
static void
vm_extend(Relation rel, BlockNumber vm_nblocks)
{
BlockNumber vm_nblocks_now;
Page pg;
pg = (Page) palloc(BLCKSZ);
PageInit(pg, BLCKSZ, 0);
/*
* We use the relation extension lock to lock out other backends trying to
* extend the visibility map at the same time. It also locks out extension
* of the main fork, unnecessarily, but extending the visibility map
* happens seldom enough that it doesn't seem worthwhile to have a
* separate lock tag type for it.
*
* Note that another backend might have extended or created the relation
* before we get the lock.
*/
LockRelationForExtension(rel, ExclusiveLock);
/* Create the file first if it doesn't exist */
if ((rel->rd_vm_nblocks == 0 || rel->rd_vm_nblocks == InvalidBlockNumber)
&& !smgrexists(rel->rd_smgr, VISIBILITYMAP_FORKNUM))
{
smgrcreate(rel->rd_smgr, VISIBILITYMAP_FORKNUM, false);
vm_nblocks_now = 0;
}
else
vm_nblocks_now = smgrnblocks(rel->rd_smgr, VISIBILITYMAP_FORKNUM);
while (vm_nblocks_now < vm_nblocks)
{
smgrextend(rel->rd_smgr, VISIBILITYMAP_FORKNUM, vm_nblocks_now,
(char *) pg, rel->rd_istemp);
vm_nblocks_now++;
}
UnlockRelationForExtension(rel, ExclusiveLock);
pfree(pg);
/* Update the relcache with the up-to-date size */
if (!InRecovery)
CacheInvalidateRelcache(rel);
rel->rd_vm_nblocks = vm_nblocks_now;
}
/*
* Read a FSM page.
*
* If the page doesn't exist, InvalidBuffer is returned, or if 'extend' is
* true, the FSM file is extended.
*/
static Buffer
fsm_readbuf(Relation rel, FSMAddress addr, bool extend)
{
BlockNumber blkno = fsm_logical_to_physical(addr);
Buffer buf;
RelationOpenSmgr(rel);
/*
* If we haven't cached the size of the FSM yet, check it first. Also
* recheck if the requested block seems to be past end, since our cached
* value might be stale. (We send smgr inval messages on truncation, but
* not on extension.)
*/
if (rel->rd_smgr->smgr_fsm_nblocks == InvalidBlockNumber ||
blkno >= rel->rd_smgr->smgr_fsm_nblocks)
{
if (smgrexists(rel->rd_smgr, FSM_FORKNUM))
rel->rd_smgr->smgr_fsm_nblocks = smgrnblocks(rel->rd_smgr,
FSM_FORKNUM);
else
rel->rd_smgr->smgr_fsm_nblocks = 0;
}
/* Handle requests beyond EOF */
if (blkno >= rel->rd_smgr->smgr_fsm_nblocks)
{
if (extend)
fsm_extend(rel, blkno + 1);
else
return InvalidBuffer;
}
/*
* Use ZERO_ON_ERROR mode, and initialize the page if necessary. The FSM
* information is not accurate anyway, so it's better to clear corrupt
* pages than error out. Since the FSM changes are not WAL-logged, the
* so-called torn page problem on crash can lead to pages with corrupt
* headers, for example.
*/
buf = ReadBufferExtended(rel, FSM_FORKNUM, blkno, RBM_ZERO_ON_ERROR, NULL);
if (PageIsNew(BufferGetPage(buf)))
PageInit(BufferGetPage(buf), BLCKSZ, 0);
return buf;
}
/*
* Read a visibility map page.
*
* If the page doesn't exist, InvalidBuffer is returned, or if 'extend' is
* true, the visibility map file is extended.
*/
static Buffer
vm_readbuf(Relation rel, BlockNumber blkno, bool extend)
{
Buffer buf;
RelationOpenSmgr(rel);
/*
* If we haven't cached the size of the visibility map fork yet, check it
* first. Also recheck if the requested block seems to be past end, since
* our cached value might be stale. (We send smgr inval messages on
* truncation, but not on extension.)
*/
if (rel->rd_smgr->smgr_vm_nblocks == InvalidBlockNumber ||
blkno >= rel->rd_smgr->smgr_vm_nblocks)
{
if (smgrexists(rel->rd_smgr, VISIBILITYMAP_FORKNUM))
rel->rd_smgr->smgr_vm_nblocks = smgrnblocks(rel->rd_smgr,
VISIBILITYMAP_FORKNUM);
else
rel->rd_smgr->smgr_vm_nblocks = 0;
}
/* Handle requests beyond EOF */
if (blkno >= rel->rd_smgr->smgr_vm_nblocks)
{
if (extend)
vm_extend(rel, blkno + 1);
else
return InvalidBuffer;
}
/*
* Use ZERO_ON_ERROR mode, and initialize the page if necessary. It's
* always safe to clear bits, so it's better to clear corrupt pages than
* error out.
*/
buf = ReadBufferExtended(rel, VISIBILITYMAP_FORKNUM, blkno,
RBM_ZERO_ON_ERROR, NULL);
if (PageIsNew(BufferGetPage(buf)))
PageInit(BufferGetPage(buf), BLCKSZ, 0);
return buf;
}
/*
* Read a visibility map page.
*
* If the page doesn't exist, InvalidBuffer is returned, or if 'extend' is
* true, the visibility map file is extended.
*/
static Buffer
vm_readbuf(Relation rel, BlockNumber blkno, bool extend)
{
Buffer buf;
RelationOpenSmgr(rel);
/*
* The current size of the visibility map fork is kept in relcache, to
* avoid reading beyond EOF. If we haven't cached the size of the map yet,
* do that first.
*/
if (rel->rd_vm_nblocks == InvalidBlockNumber)
{
if (smgrexists(rel->rd_smgr, VISIBILITYMAP_FORKNUM))
rel->rd_vm_nblocks = smgrnblocks(rel->rd_smgr,
VISIBILITYMAP_FORKNUM);
else
rel->rd_vm_nblocks = 0;
}
/* Handle requests beyond EOF */
if (blkno >= rel->rd_vm_nblocks)
{
if (extend)
vm_extend(rel, blkno + 1);
else
return InvalidBuffer;
}
/*
* Use ZERO_ON_ERROR mode, and initialize the page if necessary. It's
* always safe to clear bits, so it's better to clear corrupt pages than
* error out.
*/
buf = ReadBufferExtended(rel, VISIBILITYMAP_FORKNUM, blkno,
RBM_ZERO_ON_ERROR, NULL);
if (PageIsNew(BufferGetPage(buf)))
PageInit(BufferGetPage(buf), BLCKSZ, 0);
return buf;
}
/*
* visibilitymap_truncate - truncate the visibility map
*
* The caller must hold AccessExclusiveLock on the relation, to ensure that
* other backends receive the smgr invalidation event that this function sends
* before they access the VM again.
*
* nheapblocks is the new size of the heap.
*/
void
visibilitymap_truncate(Relation rel, BlockNumber nheapblocks)
{
BlockNumber newnblocks;
/* last remaining block, byte, and bit */
BlockNumber truncBlock = HEAPBLK_TO_MAPBLOCK(nheapblocks);
uint32 truncByte = HEAPBLK_TO_MAPBYTE(nheapblocks);
uint8 truncBit = HEAPBLK_TO_MAPBIT(nheapblocks);
#ifdef TRACE_VISIBILITYMAP
elog(DEBUG1, "vm_truncate %s %d", RelationGetRelationName(rel), nheapblocks);
#endif
RelationOpenSmgr(rel);
/*
* If no visibility map has been created yet for this relation, there's
* nothing to truncate.
*/
if (!smgrexists(rel->rd_smgr, VISIBILITYMAP_FORKNUM))
return;
/*
* Unless the new size is exactly at a visibility map page boundary, the
* tail bits in the last remaining map page, representing truncated heap
* blocks, need to be cleared. This is not only tidy, but also necessary
* because we don't get a chance to clear the bits if the heap is extended
* again.
*/
if (truncByte != 0 || truncBit != 0)
{
Buffer mapBuffer;
Page page;
char *map;
newnblocks = truncBlock + 1;
mapBuffer = vm_readbuf(rel, truncBlock, false);
if (!BufferIsValid(mapBuffer))
{
/* nothing to do, the file was already smaller */
return;
}
page = BufferGetPage(mapBuffer);
map = PageGetContents(page);
LockBuffer(mapBuffer, BUFFER_LOCK_EXCLUSIVE);
/* Clear out the unwanted bytes. */
MemSet(&map[truncByte + 1], 0, MAPSIZE - (truncByte + 1));
/*
* Mask out the unwanted bits of the last remaining byte.
*
* ((1 << 0) - 1) = 00000000 ((1 << 1) - 1) = 00000001 ... ((1 << 6) -
* 1) = 00111111 ((1 << 7) - 1) = 01111111
*/
map[truncByte] &= (1 << truncBit) - 1;
MarkBufferDirty(mapBuffer);
UnlockReleaseBuffer(mapBuffer);
}
else
newnblocks = truncBlock;
if (smgrnblocks(rel->rd_smgr, VISIBILITYMAP_FORKNUM) <= newnblocks)
{
/* nothing to do, the file was already smaller than requested size */
return;
}
/* Truncate the unused VM pages, and send smgr inval message */
smgrtruncate(rel->rd_smgr, VISIBILITYMAP_FORKNUM, newnblocks);
/*
* We might as well update the local smgr_vm_nblocks setting. smgrtruncate
* sent an smgr cache inval message, which will cause other backends to
* invalidate their copy of smgr_vm_nblocks, and this one too at the next
* command boundary. But this ensures it isn't outright wrong until then.
*/
if (rel->rd_smgr)
rel->rd_smgr->smgr_vm_nblocks = newnblocks;
}
/*
* visibilitymap_test - truncate the visibility map
*/
void
visibilitymap_truncate(Relation rel, BlockNumber nheapblocks)
{
BlockNumber newnblocks;
/* last remaining block, byte, and bit */
BlockNumber truncBlock = HEAPBLK_TO_MAPBLOCK(nheapblocks);
uint32 truncByte = HEAPBLK_TO_MAPBYTE(nheapblocks);
uint8 truncBit = HEAPBLK_TO_MAPBIT(nheapblocks);
#ifdef TRACE_VISIBILITYMAP
elog(DEBUG1, "vm_truncate %s %d", RelationGetRelationName(rel), nheapblocks);
#endif
/*
* If no visibility map has been created yet for this relation, there's
* nothing to truncate.
*/
if (!smgrexists(rel->rd_smgr, VISIBILITYMAP_FORKNUM))
return;
/*
* Unless the new size is exactly at a visibility map page boundary, the
* tail bits in the last remaining map page, representing truncated heap
* blocks, need to be cleared. This is not only tidy, but also necessary
* because we don't get a chance to clear the bits if the heap is extended
* again.
*/
if (truncByte != 0 || truncBit != 0)
{
Buffer mapBuffer;
Page page;
char *map;
newnblocks = truncBlock + 1;
mapBuffer = vm_readbuf(rel, truncBlock, false);
if (!BufferIsValid(mapBuffer))
{
/* nothing to do, the file was already smaller */
return;
}
page = BufferGetPage(mapBuffer);
map = PageGetContents(page);
LockBuffer(mapBuffer, BUFFER_LOCK_EXCLUSIVE);
/* Clear out the unwanted bytes. */
MemSet(&map[truncByte + 1], 0, MAPSIZE - (truncByte + 1));
/*
* Mask out the unwanted bits of the last remaining byte.
*
* ((1 << 0) - 1) = 00000000 ((1 << 1) - 1) = 00000001 ... ((1 << 6) -
* 1) = 00111111 ((1 << 7) - 1) = 01111111
*/
map[truncByte] &= (1 << truncBit) - 1;
MarkBufferDirty(mapBuffer);
UnlockReleaseBuffer(mapBuffer);
}
else
newnblocks = truncBlock;
if (smgrnblocks(rel->rd_smgr, VISIBILITYMAP_FORKNUM) < newnblocks)
{
/* nothing to do, the file was already smaller than requested size */
return;
}
smgrtruncate(rel->rd_smgr, VISIBILITYMAP_FORKNUM, newnblocks,
rel->rd_istemp);
/*
* Need to invalidate the relcache entry, because rd_vm_nblocks seen by
* other backends is no longer valid.
*/
if (!InRecovery)
CacheInvalidateRelcache(rel);
rel->rd_vm_nblocks = newnblocks;
}
/*
* FinishPreparedTransaction: execute COMMIT PREPARED or ROLLBACK PREPARED
*/
void
FinishPreparedTransaction(const char *gid, bool isCommit)
{
GlobalTransaction gxact;
TransactionId xid;
char *buf;
char *bufptr;
TwoPhaseFileHeader *hdr;
TransactionId latestXid;
TransactionId *children;
RelFileNode *commitrels;
RelFileNode *abortrels;
RelFileNode *delrels;
int ndelrels;
int i;
/*
* Validate the GID, and lock the GXACT to ensure that two backends do not
* try to commit the same GID at once.
*/
gxact = LockGXact(gid, GetUserId());
xid = gxact->proc.xid;
/*
* Read and validate the state file
*/
buf = ReadTwoPhaseFile(xid);
if (buf == NULL)
ereport(ERROR,
(errcode(ERRCODE_DATA_CORRUPTED),
errmsg("two-phase state file for transaction %u is corrupt",
xid)));
/*
* Disassemble the header area
*/
hdr = (TwoPhaseFileHeader *) buf;
Assert(TransactionIdEquals(hdr->xid, xid));
bufptr = buf + MAXALIGN(sizeof(TwoPhaseFileHeader));
children = (TransactionId *) bufptr;
bufptr += MAXALIGN(hdr->nsubxacts * sizeof(TransactionId));
commitrels = (RelFileNode *) bufptr;
bufptr += MAXALIGN(hdr->ncommitrels * sizeof(RelFileNode));
abortrels = (RelFileNode *) bufptr;
bufptr += MAXALIGN(hdr->nabortrels * sizeof(RelFileNode));
/* compute latestXid among all children */
latestXid = TransactionIdLatest(xid, hdr->nsubxacts, children);
/*
* The order of operations here is critical: make the XLOG entry for
* commit or abort, then mark the transaction committed or aborted in
* pg_clog, then remove its PGPROC from the global ProcArray (which means
* TransactionIdIsInProgress will stop saying the prepared xact is in
* progress), then run the post-commit or post-abort callbacks. The
* callbacks will release the locks the transaction held.
*/
if (isCommit)
RecordTransactionCommitPrepared(xid,
hdr->nsubxacts, children,
hdr->ncommitrels, commitrels);
else
RecordTransactionAbortPrepared(xid,
hdr->nsubxacts, children,
hdr->nabortrels, abortrels);
ProcArrayRemove(&gxact->proc, latestXid);
/*
* In case we fail while running the callbacks, mark the gxact invalid so
* no one else will try to commit/rollback, and so it can be recycled
* properly later. It is still locked by our XID so it won't go away yet.
*
* (We assume it's safe to do this without taking TwoPhaseStateLock.)
*/
gxact->valid = false;
/*
* We have to remove any files that were supposed to be dropped. For
* consistency with the regular xact.c code paths, must do this before
* releasing locks, so do it before running the callbacks.
*
* NB: this code knows that we couldn't be dropping any temp rels ...
*/
if (isCommit)
{
delrels = commitrels;
ndelrels = hdr->ncommitrels;
}
else
{
delrels = abortrels;
ndelrels = hdr->nabortrels;
}
for (i = 0; i < ndelrels; i++)
{
SMgrRelation srel = smgropen(delrels[i]);
ForkNumber fork;
for (fork = 0; fork <= MAX_FORKNUM; fork++)
{
if (smgrexists(srel, fork))
smgrdounlink(srel, fork, false, false);
}
smgrclose(srel);
}
/* And now do the callbacks */
if (isCommit)
ProcessRecords(bufptr, xid, twophase_postcommit_callbacks);
else
ProcessRecords(bufptr, xid, twophase_postabort_callbacks);
/* Count the prepared xact as committed or aborted */
AtEOXact_PgStat(isCommit);
/*
* And now we can clean up our mess.
*/
RemoveTwoPhaseFile(xid, true);
RemoveGXact(gxact);
pfree(buf);
}
/*
* pg_prewarm(regclass, mode text, fork text,
* first_block int8, last_block int8)
*
* The first argument is the relation to be prewarmed; the second controls
* how prewarming is done; legal options are 'prefetch', 'read', and 'buffer'.
* The third is the name of the relation fork to be prewarmed. The fourth
* and fifth arguments specify the first and last block to be prewarmed.
* If the fourth argument is NULL, it will be taken as 0; if the fifth argument
* is NULL, it will be taken as the number of blocks in the relation. The
* return value is the number of blocks successfully prewarmed.
*/
Datum
pg_prewarm(PG_FUNCTION_ARGS)
{
Oid relOid;
text *forkName;
text *type;
int64 first_block;
int64 last_block;
int64 nblocks;
int64 blocks_done = 0;
int64 block;
Relation rel;
ForkNumber forkNumber;
char *forkString;
char *ttype;
PrewarmType ptype;
AclResult aclresult;
/* Basic sanity checking. */
if (PG_ARGISNULL(0))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("relation cannot be null")));
relOid = PG_GETARG_OID(0);
if (PG_ARGISNULL(1))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
(errmsg("prewarm type cannot be null"))));
type = PG_GETARG_TEXT_P(1);
ttype = text_to_cstring(type);
if (strcmp(ttype, "prefetch") == 0)
ptype = PREWARM_PREFETCH;
else if (strcmp(ttype, "read") == 0)
ptype = PREWARM_READ;
else if (strcmp(ttype, "buffer") == 0)
ptype = PREWARM_BUFFER;
else
{
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("invalid prewarm type"),
errhint("Valid prewarm types are \"prefetch\", \"read\", and \"buffer\".")));
PG_RETURN_INT64(0); /* Placate compiler. */
}
if (PG_ARGISNULL(2))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
(errmsg("relation fork cannot be null"))));
forkName = PG_GETARG_TEXT_P(2);
forkString = text_to_cstring(forkName);
forkNumber = forkname_to_number(forkString);
/* Open relation and check privileges. */
rel = relation_open(relOid, AccessShareLock);
aclresult = pg_class_aclcheck(relOid, GetUserId(), ACL_SELECT);
if (aclresult != ACLCHECK_OK)
aclcheck_error(aclresult, ACL_KIND_CLASS, get_rel_name(relOid));
/* Check that the fork exists. */
RelationOpenSmgr(rel);
if (!smgrexists(rel->rd_smgr, forkNumber))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("fork \"%s\" does not exist for this relation",
forkString)));
/* Validate block numbers, or handle nulls. */
nblocks = RelationGetNumberOfBlocksInFork(rel, forkNumber);
if (PG_ARGISNULL(3))
first_block = 0;
else
{
first_block = PG_GETARG_INT64(3);
if (first_block < 0 || first_block >= nblocks)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("starting block number must be between 0 and " INT64_FORMAT,
nblocks - 1)));
}
if (PG_ARGISNULL(4))
last_block = nblocks - 1;
else
{
last_block = PG_GETARG_INT64(4);
if (last_block < 0 || last_block >= nblocks)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("ending block number must be between 0 and " INT64_FORMAT,
nblocks - 1)));
}
/* Now we're ready to do the real work. */
if (ptype == PREWARM_PREFETCH)
{
#ifdef USE_PREFETCH
/*
* In prefetch mode, we just hint the OS to read the blocks, but we
* don't know whether it really does it, and we don't wait for it to
* finish.
*
* It would probably be better to pass our prefetch requests in chunks
* of a megabyte or maybe even a whole segment at a time, but there's
* no practical way to do that at present without a gross modularity
* violation, so we just do this.
*/
for (block = first_block; block <= last_block; ++block)
{
CHECK_FOR_INTERRUPTS();
PrefetchBuffer(rel, forkNumber, block);
++blocks_done;
}
#else
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("prefetch is not supported by this build")));
#endif
}
else if (ptype == PREWARM_READ)
{
/*
* In read mode, we actually read the blocks, but not into shared
* buffers. This is more portable than prefetch mode (it works
* everywhere) and is synchronous.
*/
for (block = first_block; block <= last_block; ++block)
{
CHECK_FOR_INTERRUPTS();
smgrread(rel->rd_smgr, forkNumber, block, blockbuffer);
++blocks_done;
}
}
else if (ptype == PREWARM_BUFFER)
{
/*
* In buffer mode, we actually pull the data into shared_buffers.
*/
for (block = first_block; block <= last_block; ++block)
{
Buffer buf;
CHECK_FOR_INTERRUPTS();
buf = ReadBufferExtended(rel, forkNumber, block, RBM_NORMAL, NULL);
ReleaseBuffer(buf);
++blocks_done;
}
}
/* Close relation, release lock. */
relation_close(rel, AccessShareLock);
PG_RETURN_INT64(blocks_done);
}
/*
* RelationTruncate
* Physically truncate a relation to the specified number of blocks.
*
* This includes getting rid of any buffers for the blocks that are to be
* dropped.
*/
void
RelationTruncate(Relation rel, BlockNumber nblocks)
{
bool fsm;
bool vm;
/* Open it at the smgr level if not already done */
RelationOpenSmgr(rel);
/*
* Make sure smgr_targblock etc aren't pointing somewhere past new end
*/
rel->rd_smgr->smgr_targblock = InvalidBlockNumber;
rel->rd_smgr->smgr_fsm_nblocks = InvalidBlockNumber;
rel->rd_smgr->smgr_vm_nblocks = InvalidBlockNumber;
/* Truncate the FSM first if it exists */
fsm = smgrexists(rel->rd_smgr, FSM_FORKNUM);
if (fsm)
FreeSpaceMapTruncateRel(rel, nblocks);
/* Truncate the visibility map too if it exists. */
vm = smgrexists(rel->rd_smgr, VISIBILITYMAP_FORKNUM);
if (vm)
visibilitymap_truncate(rel, nblocks);
/*
* We WAL-log the truncation before actually truncating, which means
* trouble if the truncation fails. If we then crash, the WAL replay
* likely isn't going to succeed in the truncation either, and cause a
* PANIC. It's tempting to put a critical section here, but that cure
* would be worse than the disease. It would turn a usually harmless
* failure to truncate, that might spell trouble at WAL replay, into a
* certain PANIC.
*/
if (!rel->rd_istemp)
{
/*
* Make an XLOG entry reporting the file truncation.
*/
XLogRecPtr lsn;
XLogRecData rdata;
xl_smgr_truncate xlrec;
xlrec.blkno = nblocks;
xlrec.rnode = rel->rd_node;
rdata.data = (char *) &xlrec;
rdata.len = sizeof(xlrec);
rdata.buffer = InvalidBuffer;
rdata.next = NULL;
lsn = XLogInsert(RM_SMGR_ID, XLOG_SMGR_TRUNCATE, &rdata);
/*
* Flush, because otherwise the truncation of the main relation might
* hit the disk before the WAL record, and the truncation of the FSM
* or visibility map. If we crashed during that window, we'd be left
* with a truncated heap, but the FSM or visibility map would still
* contain entries for the non-existent heap pages.
*/
if (fsm || vm)
XLogFlush(lsn);
}
/* Do the real work */
smgrtruncate(rel->rd_smgr, MAIN_FORKNUM, nblocks, rel->rd_istemp);
}
static void
ReadBlocks(int filenum)
{
FILE *file;
char record_type;
char *dbname;
Oid record_filenode;
ForkNumber record_forknum;
BlockNumber record_blocknum;
BlockNumber record_range;
int log_level = DEBUG3;
Oid relOid = InvalidOid;
Relation rel = NULL;
bool skip_relation = false;
bool skip_fork = false;
bool skip_block = false;
BlockNumber nblocks = 0;
BlockNumber blocks_restored = 0;
const char *filepath;
/*
* If this condition changes, then this code, and the code in the writer
* will need to be changed; especially the format specifiers in log and
* error messages.
*/
StaticAssertStmt(MaxBlockNumber == 0xFFFFFFFE, "Code may need review.");
filepath = getSavefileName(filenum);
file = fileOpen(filepath, PG_BINARY_R);
dbname = readDBName(file, filepath);
/*
* When restoring global objects, the dbname is zero-length string, and non-
* zero length otherwise. And filenum is never expected to be smaller than 1.
*/
Assert(filenum >= 1);
Assert(filenum == 1 ? strlen(dbname) == 0 : strlen(dbname) > 0);
/* To restore the global objects, use default database */
BackgroundWorkerInitializeConnection(filenum == 1 ? guc_default_database : dbname, NULL);
SetCurrentStatementStartTimestamp();
StartTransactionCommand();
SPI_connect();
PushActiveSnapshot(GetTransactionSnapshot());
pgstat_report_activity(STATE_RUNNING, "restoring buffers");
/*
* Note that in case of a read error, we will leak relcache entry that we may
* currently have open. In case of EOF, we close the relation after the loop.
*/
while (fileRead(&record_type, 1, file, true, filepath))
{
/*
* If we want to process the signals, this seems to be the best place
* to do it. Generally the backends refrain from processing config file
* while in transaction, but that's more for the fear of allowing GUC
* changes to affect expression evaluation, causing different results
* for the same expression in a transaction. Since this worker is not
* processing any queries, it is okay to process the config file here.
*
* Even though it's okay to process SIGHUP here, doing so doesn't add
* any value. The only reason we might want to process config file here
* would be to allow the user to interrupt the BlockReader's operation
* by changing this extenstion's GUC parameter. But the user can do that
* anyway, using SIGTERM or pg_terminate_backend().
*/
/* Stop processing the save-file if the Postmaster wants us to die. */
if (got_sigterm)
break;
ereport(log_level,
(errmsg("record type %x - %c", record_type, record_type)));
switch (record_type)
{
case 'r':
{
/* Close the previous relation, if any. */
if (rel)
{
relation_close(rel, AccessShareLock);
rel = NULL;
}
record_forknum = InvalidForkNumber;
record_blocknum = InvalidBlockNumber;
nblocks = 0;
fileRead(&record_filenode, sizeof(Oid), file, false, filepath);
relOid = GetRelOid(record_filenode);
ereport(log_level, (errmsg("processing filenode %u, relation %u",
record_filenode, relOid)));
/*
* If the relation has been rewritten/dropped since we saved it,
* just skip it and process the next relation.
*/
if (relOid == InvalidOid)
skip_relation = true;
else
{
skip_relation = false;
/* Open the relation */
rel = relation_open(relOid, AccessShareLock);
RelationOpenSmgr(rel);
}
}
break;
case 'f':
{
record_blocknum = InvalidBlockNumber;
nblocks = 0;
fileRead(&record_forknum, sizeof(ForkNumber), file, false, filepath);
if (skip_relation)
continue;
if (rel == NULL)
ereport(ERROR,
(errmsg("found a fork record without a preceeding relation record")));
ereport(log_level, (errmsg("processing fork %d", record_forknum)));
if (!smgrexists(rel->rd_smgr, record_forknum))
skip_fork = true;
else
{
skip_fork = false;
nblocks = RelationGetNumberOfBlocksInFork(rel, record_forknum);
}
}
break;
case 'b':
{
if (record_forknum == InvalidForkNumber)
ereport(ERROR,
(errmsg("found a block record without a preceeding fork record")));
fileRead(&record_blocknum, sizeof(BlockNumber), file, false, filepath);
if (skip_relation || skip_fork)
continue;
/*
* Don't try to read past the file; the file may have been shrunk
* by a vaccum/truncate operation.
*/
if (record_blocknum >= nblocks)
{
ereport(log_level,
(errmsg("reader %d skipping block filenode %u forknum %d blocknum %u",
filenum, record_filenode, record_forknum, record_blocknum)));
skip_block = true;
continue;
}
else
{
Buffer buf;
skip_block = false;
ereport(log_level,
(errmsg("reader %d reading block filenode %u forknum %d blocknum %u",
filenum, record_filenode, record_forknum, record_blocknum)));
buf = ReadBufferExtended(rel, record_forknum, record_blocknum, RBM_NORMAL, NULL);
ReleaseBuffer(buf);
++blocks_restored;
}
}
break;
case 'N':
{
BlockNumber block;
Assert(record_blocknum != InvalidBlockNumber);
if (record_blocknum == InvalidBlockNumber)
ereport(ERROR,
(errmsg("found a block range record without a preceeding block record")));
fileRead(&record_range, sizeof(int), file, false, filepath);
if (skip_relation || skip_fork || skip_block)
continue;
ereport(log_level,
(errmsg("reader %d reading range filenode %u forknum %d blocknum %u range %u",
filenum, record_filenode, record_forknum, record_blocknum, record_range)));
for (block = record_blocknum + 1; block <= (record_blocknum + record_range); ++block)
{
Buffer buf;
/*
* Don't try to read past the file; the file may have been
* shrunk by a vaccum operation.
*/
if (block >= nblocks)
{
ereport(log_level,
(errmsg("reader %d skipping block range filenode %u forknum %d start %u end %u",
filenum, record_filenode, record_forknum,
block, record_blocknum + record_range)));
break;
}
buf = ReadBufferExtended(rel, record_forknum, block, RBM_NORMAL, NULL);
ReleaseBuffer(buf);
++blocks_restored;
}
}
break;
default:
{
ereport(ERROR,
(errmsg("found unexpected save-file marker %x - %c)", record_type, record_type)));
Assert(false);
}
break;
}
}
if (rel)
relation_close(rel, AccessShareLock);
ereport(LOG,
(errmsg("Block Reader %d: restored %u blocks",
filenum, blocks_restored)));
SPI_finish();
PopActiveSnapshot();
CommitTransactionCommand();
pgstat_report_activity(STATE_IDLE, NULL);
fileClose(file, filepath);
/* Remove the save-file */
if (remove(filepath) != 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("error removing file \"%s\" : %m", filepath)));
}
/*
* Read a FSM page.
*
* If the page doesn't exist, InvalidBuffer is returned, or if 'extend' is
* true, the FSM file is extended.
*/
static Buffer
fsm_readbuf(Relation rel, FSMAddress addr, bool extend)
{
BlockNumber blkno = fsm_logical_to_physical(addr);
Buffer buf;
RelationOpenSmgr(rel);
/*
* If we haven't cached the size of the FSM yet, check it first. Also
* recheck if the requested block seems to be past end, since our cached
* value might be stale. (We send smgr inval messages on truncation, but
* not on extension.)
*/
if (rel->rd_smgr->smgr_fsm_nblocks == InvalidBlockNumber ||
blkno >= rel->rd_smgr->smgr_fsm_nblocks)
{
if (smgrexists(rel->rd_smgr, FSM_FORKNUM))
rel->rd_smgr->smgr_fsm_nblocks = smgrnblocks(rel->rd_smgr,
FSM_FORKNUM);
else
rel->rd_smgr->smgr_fsm_nblocks = 0;
}
/* Handle requests beyond EOF */
if (blkno >= rel->rd_smgr->smgr_fsm_nblocks)
{
if (extend)
fsm_extend(rel, blkno + 1);
else
return InvalidBuffer;
}
/*
* Use ZERO_ON_ERROR mode, and initialize the page if necessary. The FSM
* information is not accurate anyway, so it's better to clear corrupt
* pages than error out. Since the FSM changes are not WAL-logged, the
* so-called torn page problem on crash can lead to pages with corrupt
* headers, for example.
*
* The initialize-the-page part is trickier than it looks, because of the
* possibility of multiple backends doing this concurrently, and our
* desire to not uselessly take the buffer lock in the normal path where
* the page is OK. We must take the lock to initialize the page, so
* recheck page newness after we have the lock, in case someone else
* already did it. Also, because we initially check PageIsNew with no
* lock, it's possible to fall through and return the buffer while someone
* else is still initializing the page (i.e., we might see pd_upper as set
* but other page header fields are still zeroes). This is harmless for
* callers that will take a buffer lock themselves, but some callers
* inspect the page without any lock at all. The latter is OK only so
* long as it doesn't depend on the page header having correct contents.
* Current usage is safe because PageGetContents() does not require that.
*/
buf = ReadBufferExtended(rel, FSM_FORKNUM, blkno, RBM_ZERO_ON_ERROR, NULL);
if (PageIsNew(BufferGetPage(buf)))
{
LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
if (PageIsNew(BufferGetPage(buf)))
PageInit(BufferGetPage(buf), BLCKSZ, 0);
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
}
return buf;
}
/*
* FreeSpaceMapTruncateRel - adjust for truncation of a relation.
*
* The caller must hold AccessExclusiveLock on the relation, to ensure that
* other backends receive the smgr invalidation event that this function sends
* before they access the FSM again.
*
* nblocks is the new size of the heap.
*/
void
FreeSpaceMapTruncateRel(Relation rel, BlockNumber nblocks)
{
BlockNumber new_nfsmblocks;
FSMAddress first_removed_address;
uint16 first_removed_slot;
Buffer buf;
RelationOpenSmgr(rel);
/*
* If no FSM has been created yet for this relation, there's nothing to
* truncate.
*/
if (!smgrexists(rel->rd_smgr, FSM_FORKNUM))
return;
/* Get the location in the FSM of the first removed heap block */
first_removed_address = fsm_get_location(nblocks, &first_removed_slot);
/*
* Zero out the tail of the last remaining FSM page. If the slot
* representing the first removed heap block is at a page boundary, as the
* first slot on the FSM page that first_removed_address points to, we can
* just truncate that page altogether.
*/
if (first_removed_slot > 0)
{
buf = fsm_readbuf(rel, first_removed_address, false);
if (!BufferIsValid(buf))
return; /* nothing to do; the FSM was already smaller */
LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
/* NO EREPORT(ERROR) from here till changes are logged */
START_CRIT_SECTION();
fsm_truncate_avail(BufferGetPage(buf), first_removed_slot);
/*
* Truncation of a relation is WAL-logged at a higher-level, and we
* will be called at WAL replay. But if checksums are enabled, we need
* to still write a WAL record to protect against a torn page, if the
* page is flushed to disk before the truncation WAL record. We cannot
* use MarkBufferDirtyHint here, because that will not dirty the page
* during recovery.
*/
MarkBufferDirty(buf);
if (!InRecovery && RelationNeedsWAL(rel) && XLogHintBitIsNeeded())
log_newpage_buffer(buf, false);
END_CRIT_SECTION();
UnlockReleaseBuffer(buf);
new_nfsmblocks = fsm_logical_to_physical(first_removed_address) + 1;
}
else
{
new_nfsmblocks = fsm_logical_to_physical(first_removed_address);
if (smgrnblocks(rel->rd_smgr, FSM_FORKNUM) <= new_nfsmblocks)
return; /* nothing to do; the FSM was already smaller */
}
/* Truncate the unused FSM pages, and send smgr inval message */
smgrtruncate(rel->rd_smgr, FSM_FORKNUM, new_nfsmblocks);
/*
* We might as well update the local smgr_fsm_nblocks setting.
* smgrtruncate sent an smgr cache inval message, which will cause other
* backends to invalidate their copy of smgr_fsm_nblocks, and this one too
* at the next command boundary. But this ensures it isn't outright wrong
* until then.
*/
if (rel->rd_smgr)
rel->rd_smgr->smgr_fsm_nblocks = new_nfsmblocks;
/*
* Update upper-level FSM pages to account for the truncation. This is
* important because the just-truncated pages were likely marked as
* all-free, and would be preferentially selected.
*/
FreeSpaceMapVacuumRange(rel, nblocks, InvalidBlockNumber);
}
void
smgr_redo(XLogRecPtr lsn, XLogRecord *record)
{
uint8 info = record->xl_info & ~XLR_INFO_MASK;
/* Backup blocks are not used in smgr records */
Assert(!(record->xl_info & XLR_BKP_BLOCK_MASK));
if (info == XLOG_SMGR_CREATE)
{
xl_smgr_create *xlrec = (xl_smgr_create *) XLogRecGetData(record);
SMgrRelation reln;
reln = smgropen(xlrec->rnode, InvalidBackendId);
smgrcreate(reln, xlrec->forkNum, true);
}
else if (info == XLOG_SMGR_TRUNCATE)
{
xl_smgr_truncate *xlrec = (xl_smgr_truncate *) XLogRecGetData(record);
SMgrRelation reln;
Relation rel;
reln = smgropen(xlrec->rnode, InvalidBackendId);
/*
* Forcibly create relation if it doesn't exist (which suggests that
* it was dropped somewhere later in the WAL sequence). As in
* XLogReadBuffer, we prefer to recreate the rel and replay the log as
* best we can until the drop is seen.
*/
smgrcreate(reln, MAIN_FORKNUM, true);
/*
* Before we perform the truncation, update minimum recovery point
* to cover this WAL record. Once the relation is truncated, there's
* no going back. The buffer manager enforces the WAL-first rule
* for normal updates to relation files, so that the minimum recovery
* point is always updated before the corresponding change in the
* data file is flushed to disk. We have to do the same manually
* here.
*
* Doing this before the truncation means that if the truncation fails
* for some reason, you cannot start up the system even after restart,
* until you fix the underlying situation so that the truncation will
* succeed. Alternatively, we could update the minimum recovery point
* after truncation, but that would leave a small window where the
* WAL-first rule could be violated.
*/
XLogFlush(lsn);
smgrtruncate(reln, MAIN_FORKNUM, xlrec->blkno);
/* Also tell xlogutils.c about it */
XLogTruncateRelation(xlrec->rnode, MAIN_FORKNUM, xlrec->blkno);
/* Truncate FSM and VM too */
rel = CreateFakeRelcacheEntry(xlrec->rnode);
if (smgrexists(reln, FSM_FORKNUM))
FreeSpaceMapTruncateRel(rel, xlrec->blkno);
if (smgrexists(reln, VISIBILITYMAP_FORKNUM))
visibilitymap_truncate(rel, xlrec->blkno);
FreeFakeRelcacheEntry(rel);
}
else
elog(PANIC, "smgr_redo: unknown op code %u", info);
}
