/*
* Flush dirty pages to disk during checkpoint or database shutdown
*/
void
SimpleLruFlush(SlruCtl ctl, bool checkpoint)
{
SlruShared shared = ctl->shared;
SlruFlushData fdata;
int slotno;
int pageno = 0;
int i;
bool ok;
/*
* Find and write dirty pages
*/
fdata.num_files = 0;
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
for (slotno = 0; slotno < shared->num_slots; slotno++)
{
SlruInternalWritePage(ctl, slotno, &fdata);
/*
* When called during a checkpoint, we cannot assert that the slot is
* clean now, since another process might have re-dirtied it already.
* That's okay.
*/
Assert(checkpoint ||
shared->page_status[slotno] == SLRU_PAGE_EMPTY ||
(shared->page_status[slotno] == SLRU_PAGE_VALID &&
!shared->page_dirty[slotno]));
}
LWLockRelease(shared->ControlLock);
/*
* Now fsync and close any files that were open
*/
ok = true;
for (i = 0; i < fdata.num_files; i++)
{
if (ctl->do_fsync && pg_fsync(fdata.fd[i]))
{
slru_errcause = SLRU_FSYNC_FAILED;
slru_errno = errno;
pageno = fdata.segno[i] * SLRU_PAGES_PER_SEGMENT;
ok = false;
}
if (close(fdata.fd[i]))
{
slru_errcause = SLRU_CLOSE_FAILED;
slru_errno = errno;
pageno = fdata.segno[i] * SLRU_PAGES_PER_SEGMENT;
ok = false;
}
}
if (!ok)
SlruReportIOError(ctl, pageno, InvalidTransactionId);
}
/*
* Return whether the given page exists on disk.
*
* A false return means that either the file does not exist, or that it's not
* large enough to contain the given page.
*/
bool
SimpleLruDoesPhysicalPageExist(SlruCtl ctl, int pageno)
{
int segno = pageno / SLRU_PAGES_PER_SEGMENT;
int rpageno = pageno % SLRU_PAGES_PER_SEGMENT;
int offset = rpageno * BLCKSZ;
char path[MAXPGPATH];
int fd;
bool result;
off_t endpos;
SlruFileName(ctl, path, segno);
fd = OpenTransientFile(path, O_RDWR | PG_BINARY, S_IRUSR | S_IWUSR);
if (fd < 0)
{
/* expected: file doesn't exist */
if (errno == ENOENT)
return false;
/* report error normally */
slru_errcause = SLRU_OPEN_FAILED;
slru_errno = errno;
SlruReportIOError(ctl, pageno, 0);
}
if ((endpos = lseek(fd, 0, SEEK_END)) < 0)
{
slru_errcause = SLRU_OPEN_FAILED;
slru_errno = errno;
SlruReportIOError(ctl, pageno, 0);
}
result = endpos >= (off_t) (offset + BLCKSZ);
CloseTransientFile(fd);
return result;
}
/*
* Write a page from a shared buffer, if necessary.
* Does nothing if the specified slot is not dirty.
*
* NOTE: only one write attempt is made here. Hence, it is possible that
* the page is still dirty at exit (if someone else re-dirtied it during
* the write). However, we *do* attempt a fresh write even if the page
* is already being written; this is for checkpoints.
*
* Control lock must be held at entry, and will be held at exit.
*/
static void
SlruInternalWritePage(SlruCtl ctl, int slotno, SlruFlush fdata)
{
SlruShared shared = ctl->shared;
int pageno = shared->page_number[slotno];
bool ok;
/* If a write is in progress, wait for it to finish */
while (shared->page_status[slotno] == SLRU_PAGE_WRITE_IN_PROGRESS &&
shared->page_number[slotno] == pageno)
{
SimpleLruWaitIO(ctl, slotno);
}
/*
* Do nothing if page is not dirty, or if buffer no longer contains the
* same page we were called for.
*/
if (!shared->page_dirty[slotno] ||
shared->page_status[slotno] != SLRU_PAGE_VALID ||
shared->page_number[slotno] != pageno)
return;
/*
* Mark the slot write-busy, and clear the dirtybit. After this point, a
* transaction status update on this page will mark it dirty again.
*/
shared->page_status[slotno] = SLRU_PAGE_WRITE_IN_PROGRESS;
shared->page_dirty[slotno] = false;
/* Acquire per-buffer lock (cannot deadlock, see notes at top) */
LWLockAcquire(shared->buffer_locks[slotno], LW_EXCLUSIVE);
/* Release control lock while doing I/O */
LWLockRelease(shared->ControlLock);
/* Do the write */
ok = SlruPhysicalWritePage(ctl, pageno, slotno, fdata);
/* If we failed, and we're in a flush, better close the files */
if (!ok && fdata)
{
int i;
for (i = 0; i < fdata->num_files; i++)
close(fdata->fd[i]);
}
/* Re-acquire control lock and update page state */
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
Assert(shared->page_number[slotno] == pageno &&
shared->page_status[slotno] == SLRU_PAGE_WRITE_IN_PROGRESS);
/* If we failed to write, mark the page dirty again */
if (!ok)
shared->page_dirty[slotno] = true;
shared->page_status[slotno] = SLRU_PAGE_VALID;
LWLockRelease(shared->buffer_locks[slotno]);
/* Now it's okay to ereport if we failed */
if (!ok)
SlruReportIOError(ctl, pageno, InvalidTransactionId);
}
/*
* Find a page in a shared buffer, reading it in if necessary.
* The page number must correspond to an already-initialized page.
*
* If write_ok is true then it is OK to return a page that is in
* WRITE_IN_PROGRESS state; it is the caller's responsibility to be sure
* that modification of the page is safe. If write_ok is false then we
* will not return the page until it is not undergoing active I/O.
*
* The passed-in xid is used only for error reporting, and may be
* InvalidTransactionId if no specific xid is associated with the action.
*
* Return value is the shared-buffer slot number now holding the page.
* The buffer's LRU access info is updated.
*
* Control lock must be held at entry, and will be held at exit.
*/
int
SimpleLruReadPage(SlruCtl ctl, int pageno, bool write_ok,
TransactionId xid)
{
SlruShared shared = ctl->shared;
/* Outer loop handles restart if we must wait for someone else's I/O */
for (;;)
{
int slotno;
bool ok;
/* See if page already is in memory; if not, pick victim slot */
slotno = SlruSelectLRUPage(ctl, pageno);
/* Did we find the page in memory? */
if (shared->page_number[slotno] == pageno &&
shared->page_status[slotno] != SLRU_PAGE_EMPTY)
{
/*
* If page is still being read in, we must wait for I/O. Likewise
* if the page is being written and the caller said that's not OK.
*/
if (shared->page_status[slotno] == SLRU_PAGE_READ_IN_PROGRESS ||
(shared->page_status[slotno] == SLRU_PAGE_WRITE_IN_PROGRESS &&
!write_ok))
{
SimpleLruWaitIO(ctl, slotno);
/* Now we must recheck state from the top */
continue;
}
/* Otherwise, it's ready to use */
SlruRecentlyUsed(shared, slotno);
return slotno;
}
/* We found no match; assert we selected a freeable slot */
Assert(shared->page_status[slotno] == SLRU_PAGE_EMPTY ||
(shared->page_status[slotno] == SLRU_PAGE_VALID &&
!shared->page_dirty[slotno]));
/* Mark the slot read-busy */
shared->page_number[slotno] = pageno;
shared->page_status[slotno] = SLRU_PAGE_READ_IN_PROGRESS;
shared->page_dirty[slotno] = false;
/* Acquire per-buffer lock (cannot deadlock, see notes at top) */
LWLockAcquire(shared->buffer_locks[slotno], LW_EXCLUSIVE);
/* Release control lock while doing I/O */
LWLockRelease(shared->ControlLock);
/* Do the read */
ok = SlruPhysicalReadPage(ctl, pageno, slotno);
/* Set the LSNs for this newly read-in page to zero */
SimpleLruZeroLSNs(ctl, slotno);
/* Re-acquire control lock and update page state */
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
Assert(shared->page_number[slotno] == pageno &&
shared->page_status[slotno] == SLRU_PAGE_READ_IN_PROGRESS &&
!shared->page_dirty[slotno]);
shared->page_status[slotno] = ok ? SLRU_PAGE_VALID : SLRU_PAGE_EMPTY;
LWLockRelease(shared->buffer_locks[slotno]);
/* Now it's okay to ereport if we failed */
if (!ok)
SlruReportIOError(ctl, pageno, xid);
SlruRecentlyUsed(shared, slotno);
return slotno;
}
}
/*
* Flush dirty pages to disk during checkpoint or database shutdown
*/
void
SimpleLruFlush(SlruCtl ctl, bool allow_redirtied)
{
SlruShared shared = ctl->shared;
SlruFlushData fdata;
int slotno;
int pageno = 0;
int i;
bool ok;
/*
* Find and write dirty pages
*/
fdata.num_files = 0;
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
for (slotno = 0; slotno < shared->num_slots; slotno++)
{
SlruInternalWritePage(ctl, slotno, &fdata);
/*
* In some places (e.g. checkpoints), we cannot assert that the slot
* is clean now, since another process might have re-dirtied it
* already. That's okay.
*/
Assert(allow_redirtied ||
shared->page_status[slotno] == SLRU_PAGE_EMPTY ||
(shared->page_status[slotno] == SLRU_PAGE_VALID &&
!shared->page_dirty[slotno]));
}
LWLockRelease(shared->ControlLock);
/*
* Now fsync and close any files that were open
*/
ok = true;
for (i = 0; i < fdata.num_files; i++)
{
pgstat_report_wait_start(WAIT_EVENT_SLRU_FLUSH_SYNC);
if (ctl->do_fsync && pg_fsync(fdata.fd[i]))
{
slru_errcause = SLRU_FSYNC_FAILED;
slru_errno = errno;
pageno = fdata.segno[i] * SLRU_PAGES_PER_SEGMENT;
ok = false;
}
pgstat_report_wait_end();
if (CloseTransientFile(fdata.fd[i]))
{
slru_errcause = SLRU_CLOSE_FAILED;
slru_errno = errno;
pageno = fdata.segno[i] * SLRU_PAGES_PER_SEGMENT;
ok = false;
}
}
if (!ok)
SlruReportIOError(ctl, pageno, InvalidTransactionId);
}
/*
* Find a page in a shared buffer, reading it in if necessary.
* The page number must correspond to an already-initialized page.
*
* The passed-in xid is used only for error reporting, and may be
* InvalidTransactionId if no specific xid is associated with the action.
*
* If the passed in pointer to valid is NULL, then log errors can be
* generated by this function. If valid is not NULL, then the function
* will not generate log errors, but will set the boolean value
* pointed to by valid to TRUE if it was able to read the page,
* or FALSE if the page read had error.
*
* Return value is the shared-buffer slot number now holding the page.
* The buffer's LRU access info is updated.
*
* Control lock must be held at entry, and will be held at exit.
*/
static int
SimpleLruReadPage_Internal(SlruCtl ctl, int pageno, TransactionId xid, bool *valid)
{
SlruShared shared = ctl->shared;
/* Outer loop handles restart if we must wait for someone else's I/O */
for (;;)
{
int slotno;
bool ok;
/* See if page already is in memory; if not, pick victim slot */
slotno = SlruSelectLRUPage(ctl, pageno);
/* Did we find the page in memory? */
if (shared->page_number[slotno] == pageno &&
shared->page_status[slotno] != SLRU_PAGE_EMPTY)
{
/* If page is still being read in, we must wait for I/O */
if (shared->page_status[slotno] == SLRU_PAGE_READ_IN_PROGRESS)
{
SimpleLruWaitIO(ctl, slotno);
/* Now we must recheck state from the top */
continue;
}
/* Otherwise, it's ready to use */
SlruRecentlyUsed(shared, slotno);
if (valid != NULL)
*valid = true;
return slotno;
}
/* We found no match; assert we selected a freeable slot */
Assert(shared->page_status[slotno] == SLRU_PAGE_EMPTY ||
(shared->page_status[slotno] == SLRU_PAGE_VALID &&
!shared->page_dirty[slotno]));
/* Mark the slot read-busy */
shared->page_number[slotno] = pageno;
shared->page_status[slotno] = SLRU_PAGE_READ_IN_PROGRESS;
shared->page_dirty[slotno] = false;
/* Acquire per-buffer lock (cannot deadlock, see notes at top) */
LWLockAcquire(shared->buffer_locks[slotno], LW_EXCLUSIVE);
/*
* Temporarily mark page as recently-used to discourage
* SlruSelectLRUPage from selecting it again for someone else.
*/
SlruRecentlyUsed(shared, slotno);
/* Release control lock while doing I/O */
LWLockRelease(shared->ControlLock);
/* Do the read */
ok = SlruPhysicalReadPage(ctl, pageno, slotno);
/* Re-acquire control lock and update page state */
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
Assert(shared->page_number[slotno] == pageno &&
shared->page_status[slotno] == SLRU_PAGE_READ_IN_PROGRESS &&
!shared->page_dirty[slotno]);
shared->page_status[slotno] = ok ? SLRU_PAGE_VALID : SLRU_PAGE_EMPTY;
LWLockRelease(shared->buffer_locks[slotno]);
/* Now it's okay to ereport if we failed */
if (!ok && valid == NULL)
SlruReportIOError(ctl, pageno, xid);
else if (valid != NULL)
{
if (!ok)
{
LWLockRelease(shared->ControlLock);
*valid = false;
return -1;
}
else
*valid = true;
}
SlruRecentlyUsed(shared, slotno);
return slotno;
}
}
/*
* Write a page from a shared buffer, if necessary.
* Does nothing if the specified slot is not dirty.
*
* NOTE: only one write attempt is made here. Hence, it is possible that
* the page is still dirty at exit (if someone else re-dirtied it during
* the write). However, we *do* attempt a fresh write even if the page
* is already being written; this is for checkpoints.
*
* Control lock must be held at entry, and will be held at exit.
*/
void
SimpleLruWritePage(SlruCtl ctl, int slotno, SlruFlush fdata)
{
SlruShared shared = ctl->shared;
int pageno;
bool ok;
/* Do nothing if page does not need writing */
if (shared->page_status[slotno] != SLRU_PAGE_DIRTY &&
shared->page_status[slotno] != SLRU_PAGE_WRITE_IN_PROGRESS)
return;
pageno = shared->page_number[slotno];
/*
* We must grab the per-buffer lock to do I/O. To avoid deadlock, must
* release ControlLock while waiting for per-buffer lock. Fortunately,
* most of the time the per-buffer lock shouldn't be already held, so we
* can do this:
*/
if (!LWLockConditionalAcquire(shared->buffer_locks[slotno],
LW_EXCLUSIVE))
{
LWLockRelease(shared->ControlLock);
LWLockAcquire(shared->buffer_locks[slotno], LW_EXCLUSIVE);
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
}
/*
* Check to see if someone else already did the write, or took the buffer
* away from us. If so, do nothing. NOTE: we really should never see
* WRITE_IN_PROGRESS here, since that state should only occur while the
* writer is holding the buffer lock. But accept it so that we have a
* recovery path if a writer aborts.
*/
if (shared->page_number[slotno] != pageno ||
(shared->page_status[slotno] != SLRU_PAGE_DIRTY &&
shared->page_status[slotno] != SLRU_PAGE_WRITE_IN_PROGRESS))
{
LWLockRelease(shared->buffer_locks[slotno]);
return;
}
/*
* Mark the slot write-busy. After this point, a transaction status
* update on this page will mark it dirty again.
*/
shared->page_status[slotno] = SLRU_PAGE_WRITE_IN_PROGRESS;
/* Okay, release the control lock and do the write */
LWLockRelease(shared->ControlLock);
ok = SlruPhysicalWritePage(ctl, pageno, slotno, fdata);
/* If we failed, and we're in a flush, better close the files */
if (!ok && fdata)
{
int i;
for (i = 0; i < fdata->num_files; i++)
close(fdata->fd[i]);
}
/* Re-acquire shared control lock and update page state */
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
Assert(shared->page_number[slotno] == pageno &&
(shared->page_status[slotno] == SLRU_PAGE_WRITE_IN_PROGRESS ||
shared->page_status[slotno] == SLRU_PAGE_DIRTY));
/* Cannot set CLEAN if someone re-dirtied page since write started */
if (shared->page_status[slotno] == SLRU_PAGE_WRITE_IN_PROGRESS)
shared->page_status[slotno] = ok ? SLRU_PAGE_CLEAN : SLRU_PAGE_DIRTY;
LWLockRelease(shared->buffer_locks[slotno]);
/* Now it's okay to ereport if we failed */
if (!ok)
SlruReportIOError(ctl, pageno, InvalidTransactionId);
}
/*
* Find a page in a shared buffer, reading it in if necessary.
* The page number must correspond to an already-initialized page.
*
* The passed-in xid is used only for error reporting, and may be
* InvalidTransactionId if no specific xid is associated with the action.
*
* Return value is the shared-buffer slot number now holding the page.
* The buffer's LRU access info is updated.
*
* Control lock must be held at entry, and will be held at exit.
*/
int
SimpleLruReadPage(SlruCtl ctl, int pageno, TransactionId xid)
{
SlruShared shared = ctl->shared;
/* Outer loop handles restart if we lose the buffer to someone else */
for (;;)
{
int slotno;
bool ok;
/* See if page already is in memory; if not, pick victim slot */
slotno = SlruSelectLRUPage(ctl, pageno);
/* Did we find the page in memory? */
if (shared->page_number[slotno] == pageno &&
shared->page_status[slotno] != SLRU_PAGE_EMPTY)
{
/* If page is still being read in, we cannot use it yet */
if (shared->page_status[slotno] != SLRU_PAGE_READ_IN_PROGRESS)
{
/* otherwise, it's ready to use */
SlruRecentlyUsed(shared, slotno);
return slotno;
}
}
else
{
/* We found no match; assert we selected a freeable slot */
Assert(shared->page_status[slotno] == SLRU_PAGE_EMPTY ||
shared->page_status[slotno] == SLRU_PAGE_CLEAN);
}
/* Mark the slot read-busy (no-op if it already was) */
shared->page_number[slotno] = pageno;
shared->page_status[slotno] = SLRU_PAGE_READ_IN_PROGRESS;
/*
* Temporarily mark page as recently-used to discourage
* SlruSelectLRUPage from selecting it again for someone else.
*/
SlruRecentlyUsed(shared, slotno);
/*
* We must grab the per-buffer lock to do I/O. To avoid deadlock,
* must release ControlLock while waiting for per-buffer lock.
* Fortunately, most of the time the per-buffer lock shouldn't be
* already held, so we can do this:
*/
if (!LWLockConditionalAcquire(shared->buffer_locks[slotno],
LW_EXCLUSIVE))
{
LWLockRelease(shared->ControlLock);
LWLockAcquire(shared->buffer_locks[slotno], LW_EXCLUSIVE);
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
}
/*
* Check to see if someone else already did the read, or took the
* buffer away from us. If so, restart from the top.
*/
if (shared->page_number[slotno] != pageno ||
shared->page_status[slotno] != SLRU_PAGE_READ_IN_PROGRESS)
{
LWLockRelease(shared->buffer_locks[slotno]);
continue;
}
/* Okay, release control lock and do the read */
LWLockRelease(shared->ControlLock);
ok = SlruPhysicalReadPage(ctl, pageno, slotno);
/* Re-acquire shared control lock and update page state */
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
Assert(shared->page_number[slotno] == pageno &&
shared->page_status[slotno] == SLRU_PAGE_READ_IN_PROGRESS);
shared->page_status[slotno] = ok ? SLRU_PAGE_CLEAN : SLRU_PAGE_EMPTY;
LWLockRelease(shared->buffer_locks[slotno]);
/* Now it's okay to ereport if we failed */
if (!ok)
SlruReportIOError(ctl, pageno, xid);
SlruRecentlyUsed(shared, slotno);
return slotno;
}
}
