postgresql/src/backend/storage/file/fd.c

1213 lines
28 KiB
C
Raw Normal View History

/*-------------------------------------------------------------------------
*
* fd.c
* Virtual file descriptor code.
*
2002-06-20 22:29:54 +02:00
* Portions Copyright (c) 1996-2002, PostgreSQL Global Development Group
* Portions Copyright (c) 1994, Regents of the University of California
*
* IDENTIFICATION
* $Header: /cvsroot/pgsql/src/backend/storage/file/fd.c,v 1.97 2003/04/04 20:42:12 momjian Exp $
*
* NOTES:
*
* This code manages a cache of 'virtual' file descriptors (VFDs).
* The server opens many file descriptors for a variety of reasons,
* including base tables, scratch files (e.g., sort and hash spool
* files), and random calls to C library routines like system(3); it
* is quite easy to exceed system limits on the number of open files a
* single process can have. (This is around 256 on many modern
* operating systems, but can be as low as 32 on others.)
*
* VFDs are managed as an LRU pool, with actual OS file descriptors
* being opened and closed as needed. Obviously, if a routine is
* opened using these interfaces, all subsequent operations must also
* be through these interfaces (the File type is not a real file
* descriptor).
*
* For this scheme to work, most (if not all) routines throughout the
* server should use these interfaces instead of calling the C library
* routines (e.g., open(2) and fopen(3)) themselves. Otherwise, we
* may find ourselves short of real file descriptors anyway.
*
* This file used to contain a bunch of stuff to support RAID levels 0
* (jbod), 1 (duplex) and 5 (xor parity). That stuff is all gone
* because the parallel query processing code that called it is all
* gone. If you really need it you could get it from the original
* POSTGRES source.
*-------------------------------------------------------------------------
*/
#include "postgres.h"
#include <sys/file.h>
#include <sys/param.h>
#include <sys/stat.h>
#include <dirent.h>
#include <errno.h>
#include <unistd.h>
#include <fcntl.h>
1999-07-16 05:14:30 +02:00
#include "miscadmin.h"
1996-11-08 07:02:30 +01:00
#include "storage/fd.h"
#include "storage/ipc.h"
/* Filename components for OpenTemporaryFile */
#define PG_TEMP_FILES_DIR "pgsql_tmp"
#define PG_TEMP_FILE_PREFIX "pgsql_tmp"
/*
* Problem: Postgres does a system(ld...) to do dynamic loading.
* This will open several extra files in addition to those used by
* Postgres. We need to guarantee that there are file descriptors free
* for ld to use.
*
* The current solution is to limit the number of file descriptors
* that this code will allocate at one time: it leaves RESERVE_FOR_LD free.
*
* (Even though most dynamic loaders now use dlopen(3) or the
* equivalent, the OS must still open several files to perform the
* dynamic loading. And stdin/stdout/stderr count too. Keep this here.)
*/
#ifndef RESERVE_FOR_LD
#define RESERVE_FOR_LD 10
#endif
/*
* We need to ensure that we have at least some file descriptors
* available to postgreSQL after we've reserved the ones for LD,
* so we set that value here.
*
* I think 10 is an appropriate value so that's what it'll be
* for now.
*/
#ifndef FD_MINFREE
#define FD_MINFREE 10
#endif
/*
* A number of platforms return values for sysconf(_SC_OPEN_MAX) that are
* far beyond what they can really support. This GUC parameter limits what
* we will believe.
*/
int max_files_per_process = 1000;
/* Debugging.... */
#ifdef FDDEBUG
#define DO_DB(A) A
#else
#define DO_DB(A) /* A */
#endif
#define VFD_CLOSED (-1)
#define FileIsValid(file) \
((file) > 0 && (file) < (int) SizeVfdCache && VfdCache[file].fileName != NULL)
#define FileIsNotOpen(file) (VfdCache[file].fd == VFD_CLOSED)
#define FileUnknownPos (-1L)
This patch implements holdable cursors, following the proposal (materialization into a tuple store) discussed on pgsql-hackers earlier. I've updated the documentation and the regression tests. Notes on the implementation: - I needed to change the tuple store API slightly -- it assumes that it won't be used to hold data across transaction boundaries, so the temp files that it uses for on-disk storage are automatically reclaimed at end-of-transaction. I added a flag to tuplestore_begin_heap() to control this behavior. Is changing the tuple store API in this fashion OK? - in order to store executor results in a tuple store, I added a new CommandDest. This works well for the most part, with one exception: the current DestFunction API doesn't provide enough information to allow the Executor to store results into an arbitrary tuple store (where the particular tuple store to use is chosen by the call site of ExecutorRun). To workaround this, I've temporarily hacked up a solution that works, but is not ideal: since the receiveTuple DestFunction is passed the portal name, we can use that to lookup the Portal data structure for the cursor and then use that to get at the tuple store the Portal is using. This unnecessarily ties the Portal code with the tupleReceiver code, but it works... The proper fix for this is probably to change the DestFunction API -- Tom suggested passing the full QueryDesc to the receiveTuple function. In that case, callers of ExecutorRun could "subclass" QueryDesc to add any additional fields that their particular CommandDest needed to get access to. This approach would work, but I'd like to think about it for a little bit longer before deciding which route to go. In the mean time, the code works fine, so I don't think a fix is urgent. - (semi-related) I added a NO SCROLL keyword to DECLARE CURSOR, and adjusted the behavior of SCROLL in accordance with the discussion on -hackers. - (unrelated) Cleaned up some SGML markup in sql.sgml, copy.sgml Neil Conway
2003-03-27 17:51:29 +01:00
/* these are the assigned bits in fdstate below: */
#define FD_TEMPORARY (1 << 0)
#define FD_TXN_TEMPORARY (1 << 1)
typedef struct vfd
{
signed short fd; /* current FD, or VFD_CLOSED if none */
unsigned short fdstate; /* bitflags for VFD's state */
File nextFree; /* link to next free VFD, if in freelist */
File lruMoreRecently; /* doubly linked recency-of-use list */
File lruLessRecently;
long seekPos; /* current logical file position */
char *fileName; /* name of file, or NULL for unused VFD */
/* NB: fileName is malloc'd, and must be free'd when closing the VFD */
int fileFlags; /* open(2) flags for (re)opening the file */
int fileMode; /* mode to pass to open(2) */
} Vfd;
/*
* Virtual File Descriptor array pointer and size. This grows as
1999-05-25 18:15:34 +02:00
* needed. 'File' values are indexes into this array.
* Note that VfdCache[0] is not a usable VFD, just a list header.
*/
static Vfd *VfdCache;
static Size SizeVfdCache = 0;
/*
* Number of file descriptors known to be in use by VFD entries.
*/
static int nfile = 0;
/*
* List of stdio FILEs opened with AllocateFile.
*
* Since we don't want to encourage heavy use of AllocateFile, it seems
* OK to put a pretty small maximum limit on the number of simultaneously
* allocated files.
*/
#define MAX_ALLOCATED_FILES 32
static int numAllocatedFiles = 0;
1999-05-25 18:15:34 +02:00
static FILE *allocatedFiles[MAX_ALLOCATED_FILES];
/*
* Number of temporary files opened during the current transaction;
* this is used in generation of tempfile names.
*/
static long tempFileCounter = 0;
/*--------------------
*
* Private Routines
*
* Delete - delete a file from the Lru ring
* LruDelete - remove a file from the Lru ring and close its FD
* Insert - put a file at the front of the Lru ring
* LruInsert - put a file at the front of the Lru ring and open it
* ReleaseLruFile - Release an fd by closing the last entry in the Lru ring
* AllocateVfd - grab a free (or new) file record (from VfdArray)
* FreeVfd - free a file record
*
* The Least Recently Used ring is a doubly linked list that begins and
* ends on element zero. Element zero is special -- it doesn't represent
* a file and its "fd" field always == VFD_CLOSED. Element zero is just an
* anchor that shows us the beginning/end of the ring.
* Only VFD elements that are currently really open (have an FD assigned) are
* in the Lru ring. Elements that are "virtually" open can be recognized
* by having a non-null fileName field.
*
* example:
*
* /--less----\ /---------\
* v \ v \
* #0 --more---> LeastRecentlyUsed --more-\ \
* ^\ | |
* \\less--> MostRecentlyUsedFile <---/ |
* \more---/ \--less--/
*
*--------------------
*/
static void Delete(File file);
static void LruDelete(File file);
static void Insert(File file);
static int LruInsert(File file);
static bool ReleaseLruFile(void);
static File AllocateVfd(void);
static void FreeVfd(File file);
static int FileAccess(File file);
static File fileNameOpenFile(FileName fileName, int fileFlags, int fileMode);
static char *filepath(const char *filename);
static long pg_nofile(void);
/*
* pg_fsync --- same as fsync except does nothing if enableFsync is off
*/
int
pg_fsync(int fd)
{
if (enableFsync)
return fsync(fd);
else
return 0;
}
/*
* pg_fdatasync --- same as fdatasync except does nothing if enableFsync is off
*
* Not all platforms have fdatasync; treat as fsync if not available.
*/
int
pg_fdatasync(int fd)
{
if (enableFsync)
{
#ifdef HAVE_FDATASYNC
return fdatasync(fd);
#else
return fsync(fd);
#endif
}
else
return 0;
}
/*
* BasicOpenFile --- same as open(2) except can free other FDs if needed
*
* This is exported for use by places that really want a plain kernel FD,
* but need to be proof against running out of FDs. Once an FD has been
* successfully returned, it is the caller's responsibility to ensure that
* it will not be leaked on elog()! Most users should *not* call this
* routine directly, but instead use the VFD abstraction level, which
* provides protection against descriptor leaks as well as management of
* files that need to be open for more than a short period of time.
*
* Ideally this should be the *only* direct call of open() in the backend.
* In practice, the postmaster calls open() directly, and there are some
* direct open() calls done early in backend startup. Those are OK since
* this module wouldn't have any open files to close at that point anyway.
*/
int
BasicOpenFile(FileName fileName, int fileFlags, int fileMode)
{
2001-03-22 05:01:46 +01:00
int fd;
tryAgain:
fd = open(fileName, fileFlags, fileMode);
if (fd >= 0)
return fd; /* success! */
if (errno == EMFILE || errno == ENFILE)
{
2001-03-22 05:01:46 +01:00
int save_errno = errno;
DO_DB(elog(LOG, "BasicOpenFile: not enough descs, retry, er= %d",
errno));
errno = 0;
if (ReleaseLruFile())
goto tryAgain;
errno = save_errno;
}
return -1; /* failure */
}
/*
* pg_nofile: determine number of filedescriptors that fd.c is allowed to use
*/
static long
pg_nofile(void)
{
static long no_files = 0;
/* need do this calculation only once */
if (no_files == 0)
{
/*
* Ask the system what its files-per-process limit is.
*/
#ifdef HAVE_SYSCONF
no_files = sysconf(_SC_OPEN_MAX);
if (no_files <= 0)
{
#ifdef NOFILE
no_files = (long) NOFILE;
#else
no_files = (long) max_files_per_process;
#endif
elog(LOG, "pg_nofile: sysconf(_SC_OPEN_MAX) failed; using %ld",
no_files);
}
#else /* !HAVE_SYSCONF */
#ifdef NOFILE
no_files = (long) NOFILE;
#else
no_files = (long) max_files_per_process;
#endif
#endif /* HAVE_SYSCONF */
/*
* Some platforms return hopelessly optimistic values. Apply a
* configurable upper limit.
*/
if (no_files > (long) max_files_per_process)
no_files = (long) max_files_per_process;
/*
* Make sure we have enough to get by after reserving some for LD.
*/
if ((no_files - RESERVE_FOR_LD) < FD_MINFREE)
elog(FATAL, "pg_nofile: insufficient file descriptors available to start backend.\n"
"\tSystem allows %ld, we need at least %d.",
no_files, RESERVE_FOR_LD + FD_MINFREE);
no_files -= RESERVE_FOR_LD;
}
return no_files;
}
#if defined(FDDEBUG)
static void
_dump_lru(void)
{
int mru = VfdCache[0].lruLessRecently;
Vfd *vfdP = &VfdCache[mru];
char buf[2048];
snprintf(buf, sizeof(buf), "LRU: MOST %d ", mru);
while (mru != 0)
{
mru = vfdP->lruLessRecently;
vfdP = &VfdCache[mru];
snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "%d ", mru);
}
snprintf(buf + strlen(buf), sizeof(buf) - strlen(buf), "LEAST");
elog(LOG, buf);
}
#endif /* FDDEBUG */
static void
Delete(File file)
{
Vfd *vfdP;
Assert(file != 0);
DO_DB(elog(LOG, "Delete %d (%s)",
file, VfdCache[file].fileName));
DO_DB(_dump_lru());
vfdP = &VfdCache[file];
VfdCache[vfdP->lruLessRecently].lruMoreRecently = vfdP->lruMoreRecently;
VfdCache[vfdP->lruMoreRecently].lruLessRecently = vfdP->lruLessRecently;
DO_DB(_dump_lru());
}
static void
LruDelete(File file)
{
Vfd *vfdP;
Assert(file != 0);
DO_DB(elog(LOG, "LruDelete %d (%s)",
file, VfdCache[file].fileName));
vfdP = &VfdCache[file];
/* delete the vfd record from the LRU ring */
Delete(file);
/* save the seek position */
vfdP->seekPos = (long) lseek(vfdP->fd, 0L, SEEK_CUR);
Assert(vfdP->seekPos != -1L);
/* close the file */
if (close(vfdP->fd))
elog(LOG, "LruDelete: failed to close %s: %m",
vfdP->fileName);
--nfile;
vfdP->fd = VFD_CLOSED;
}
static void
Insert(File file)
{
Vfd *vfdP;
Assert(file != 0);
DO_DB(elog(LOG, "Insert %d (%s)",
file, VfdCache[file].fileName));
DO_DB(_dump_lru());
vfdP = &VfdCache[file];
vfdP->lruMoreRecently = 0;
vfdP->lruLessRecently = VfdCache[0].lruLessRecently;
VfdCache[0].lruLessRecently = file;
VfdCache[vfdP->lruLessRecently].lruMoreRecently = file;
DO_DB(_dump_lru());
}
static int
LruInsert(File file)
{
Vfd *vfdP;
Assert(file != 0);
DO_DB(elog(LOG, "LruInsert %d (%s)",
file, VfdCache[file].fileName));
vfdP = &VfdCache[file];
if (FileIsNotOpen(file))
{
1999-05-25 18:15:34 +02:00
while (nfile + numAllocatedFiles >= pg_nofile())
{
2001-03-22 05:01:46 +01:00
if (!ReleaseLruFile())
break;
}
/*
* The open could still fail for lack of file descriptors, eg due
* to overall system file table being full. So, be prepared to
* release another FD if necessary...
*/
vfdP->fd = BasicOpenFile(vfdP->fileName, vfdP->fileFlags,
vfdP->fileMode);
if (vfdP->fd < 0)
{
DO_DB(elog(LOG, "RE_OPEN FAILED: %d", errno));
1998-09-01 05:29:17 +02:00
return vfdP->fd;
}
else
{
DO_DB(elog(LOG, "RE_OPEN SUCCESS"));
++nfile;
}
/* seek to the right position */
if (vfdP->seekPos != 0L)
{
long returnValue;
returnValue = (long) lseek(vfdP->fd, vfdP->seekPos, SEEK_SET);
Assert(returnValue != -1L);
}
}
/*
* put it at the head of the Lru ring
*/
Insert(file);
1998-09-01 05:29:17 +02:00
return 0;
}
static bool
ReleaseLruFile(void)
{
DO_DB(elog(LOG, "ReleaseLruFile. Opened %d", nfile));
if (nfile > 0)
{
/*
* There are opened files and so there should be at least one used
* vfd in the ring.
*/
Assert(VfdCache[0].lruMoreRecently != 0);
LruDelete(VfdCache[0].lruMoreRecently);
return true; /* freed a file */
}
return false; /* no files available to free */
}
static File
AllocateVfd(void)
{
Index i;
File file;
DO_DB(elog(LOG, "AllocateVfd. Size %d", SizeVfdCache));
if (SizeVfdCache == 0)
{
/* initialize header entry first time through */
VfdCache = (Vfd *) malloc(sizeof(Vfd));
if (VfdCache == NULL)
elog(FATAL, "AllocateVfd: no room for VFD array");
MemSet((char *) &(VfdCache[0]), 0, sizeof(Vfd));
VfdCache->fd = VFD_CLOSED;
SizeVfdCache = 1;
/*
* register proc-exit call to ensure temp files are dropped at
* exit
*/
on_proc_exit(AtEOXact_Files, 0);
}
if (VfdCache[0].nextFree == 0)
{
/*
* The free list is empty so it is time to increase the size of
* the array. We choose to double it each time this happens.
* However, there's not much point in starting *real* small.
*/
1999-05-25 18:15:34 +02:00
Size newCacheSize = SizeVfdCache * 2;
Vfd *newVfdCache;
if (newCacheSize < 32)
newCacheSize = 32;
/*
* Be careful not to clobber VfdCache ptr if realloc fails; we
* will need it during proc_exit cleanup!
*/
newVfdCache = (Vfd *) realloc(VfdCache, sizeof(Vfd) * newCacheSize);
if (newVfdCache == NULL)
elog(FATAL, "AllocateVfd: no room to enlarge VFD array");
VfdCache = newVfdCache;
/*
* Initialize the new entries and link them into the free list.
*/
for (i = SizeVfdCache; i < newCacheSize; i++)
{
MemSet((char *) &(VfdCache[i]), 0, sizeof(Vfd));
VfdCache[i].nextFree = i + 1;
VfdCache[i].fd = VFD_CLOSED;
}
VfdCache[newCacheSize - 1].nextFree = 0;
VfdCache[0].nextFree = SizeVfdCache;
/*
* Record the new size
*/
SizeVfdCache = newCacheSize;
}
file = VfdCache[0].nextFree;
VfdCache[0].nextFree = VfdCache[file].nextFree;
return file;
}
static void
FreeVfd(File file)
{
Vfd *vfdP = &VfdCache[file];
DO_DB(elog(LOG, "FreeVfd: %d (%s)",
file, vfdP->fileName ? vfdP->fileName : ""));
if (vfdP->fileName != NULL)
{
free(vfdP->fileName);
vfdP->fileName = NULL;
}
vfdP->fdstate = 0x0;
vfdP->nextFree = VfdCache[0].nextFree;
VfdCache[0].nextFree = file;
}
1997-11-07 07:38:51 +01:00
/* filepath()
* Convert given pathname to absolute.
*
* Result is a palloc'd string.
*
* (Generally, this isn't actually necessary, considering that we
* should be cd'd into the database directory. Presently it is only
* necessary to do it in "bootstrap" mode. Maybe we should change
* bootstrap mode to do the cd, and save a few cycles/bytes here.)
1997-11-07 07:38:51 +01:00
*/
static char *
filepath(const char *filename)
{
char *buf;
1997-11-07 07:38:51 +01:00
/* Not an absolute path name? Then fill in with database path... */
if (!is_absolute_path(filename))
{
buf = (char *) palloc(strlen(DatabasePath) + strlen(filename) + 2);
sprintf(buf, "%s/%s", DatabasePath, filename);
}
else
buf = pstrdup(filename);
1997-11-07 07:38:51 +01:00
#ifdef FILEDEBUG
printf("filepath: path is %s\n", buf);
1997-11-07 07:38:51 +01:00
#endif
1998-09-01 05:29:17 +02:00
return buf;
}
static int
FileAccess(File file)
{
int returnValue;
DO_DB(elog(LOG, "FileAccess %d (%s)",
file, VfdCache[file].fileName));
/*
1999-05-25 18:15:34 +02:00
* Is the file open? If not, open it and put it at the head of the
* LRU ring (possibly closing the least recently used file to get an
* FD).
*/
if (FileIsNotOpen(file))
{
returnValue = LruInsert(file);
if (returnValue != 0)
return returnValue;
}
else if (VfdCache[0].lruLessRecently != file)
{
/*
* We now know that the file is open and that it is not the last
* one accessed, so we need to move it to the head of the Lru
* ring.
*/
Delete(file);
Insert(file);
}
1998-09-01 05:29:17 +02:00
return 0;
}
/*
* Called when we get a shared invalidation message on some relation.
*/
#ifdef NOT_USED
void
FileInvalidate(File file)
{
Assert(FileIsValid(file));
if (!FileIsNotOpen(file))
LruDelete(file);
}
#endif
static File
fileNameOpenFile(FileName fileName,
int fileFlags,
int fileMode)
{
File file;
Vfd *vfdP;
if (fileName == NULL)
elog(ERROR, "fileNameOpenFile: NULL fname");
DO_DB(elog(LOG, "fileNameOpenFile: %s %x %o",
fileName, fileFlags, fileMode));
file = AllocateVfd();
vfdP = &VfdCache[file];
1999-05-25 18:15:34 +02:00
while (nfile + numAllocatedFiles >= pg_nofile())
{
2001-03-22 05:01:46 +01:00
if (!ReleaseLruFile())
break;
}
vfdP->fd = BasicOpenFile(fileName, fileFlags, fileMode);
if (vfdP->fd < 0)
{
FreeVfd(file);
return -1;
}
++nfile;
DO_DB(elog(LOG, "fileNameOpenFile: success %d",
vfdP->fd));
Insert(file);
vfdP->fileName = (char *) malloc(strlen(fileName) + 1);
if (vfdP->fileName == NULL)
elog(FATAL, "fileNameOpenFile: no room to save VFD filename");
strcpy(vfdP->fileName, fileName);
/* Saved flags are adjusted to be OK for re-opening file */
vfdP->fileFlags = fileFlags & ~(O_CREAT | O_TRUNC | O_EXCL);
vfdP->fileMode = fileMode;
vfdP->seekPos = 0;
vfdP->fdstate = 0x0;
return file;
}
/*
* open a file in the database directory ($PGDATA/base/...)
*/
File
FileNameOpenFile(FileName fileName, int fileFlags, int fileMode)
{
File fd;
char *fname;
fname = filepath(fileName);
fd = fileNameOpenFile(fname, fileFlags, fileMode);
pfree(fname);
1998-09-01 05:29:17 +02:00
return fd;
}
/*
* open a file in an arbitrary directory
*/
File
PathNameOpenFile(FileName fileName, int fileFlags, int fileMode)
{
1998-09-01 05:29:17 +02:00
return fileNameOpenFile(fileName, fileFlags, fileMode);
}
/*
* Open a temporary file that will disappear when we close it.
*
* This routine takes care of generating an appropriate tempfile name.
* There's no need to pass in fileFlags or fileMode either, since only
* one setting makes any sense for a temp file.
This patch implements holdable cursors, following the proposal (materialization into a tuple store) discussed on pgsql-hackers earlier. I've updated the documentation and the regression tests. Notes on the implementation: - I needed to change the tuple store API slightly -- it assumes that it won't be used to hold data across transaction boundaries, so the temp files that it uses for on-disk storage are automatically reclaimed at end-of-transaction. I added a flag to tuplestore_begin_heap() to control this behavior. Is changing the tuple store API in this fashion OK? - in order to store executor results in a tuple store, I added a new CommandDest. This works well for the most part, with one exception: the current DestFunction API doesn't provide enough information to allow the Executor to store results into an arbitrary tuple store (where the particular tuple store to use is chosen by the call site of ExecutorRun). To workaround this, I've temporarily hacked up a solution that works, but is not ideal: since the receiveTuple DestFunction is passed the portal name, we can use that to lookup the Portal data structure for the cursor and then use that to get at the tuple store the Portal is using. This unnecessarily ties the Portal code with the tupleReceiver code, but it works... The proper fix for this is probably to change the DestFunction API -- Tom suggested passing the full QueryDesc to the receiveTuple function. In that case, callers of ExecutorRun could "subclass" QueryDesc to add any additional fields that their particular CommandDest needed to get access to. This approach would work, but I'd like to think about it for a little bit longer before deciding which route to go. In the mean time, the code works fine, so I don't think a fix is urgent. - (semi-related) I added a NO SCROLL keyword to DECLARE CURSOR, and adjusted the behavior of SCROLL in accordance with the discussion on -hackers. - (unrelated) Cleaned up some SGML markup in sql.sgml, copy.sgml Neil Conway
2003-03-27 17:51:29 +01:00
*
* keepOverTxn: if true, don't close the file at end-of-transaction. In
* most cases, you don't want temporary files to outlive the transaction
* that created them, so this should be false -- but if you need
* "somewhat" temporary storage, this might be useful. In either case,
* the file is removed when the File is explicitely closed.
*/
File
This patch implements holdable cursors, following the proposal (materialization into a tuple store) discussed on pgsql-hackers earlier. I've updated the documentation and the regression tests. Notes on the implementation: - I needed to change the tuple store API slightly -- it assumes that it won't be used to hold data across transaction boundaries, so the temp files that it uses for on-disk storage are automatically reclaimed at end-of-transaction. I added a flag to tuplestore_begin_heap() to control this behavior. Is changing the tuple store API in this fashion OK? - in order to store executor results in a tuple store, I added a new CommandDest. This works well for the most part, with one exception: the current DestFunction API doesn't provide enough information to allow the Executor to store results into an arbitrary tuple store (where the particular tuple store to use is chosen by the call site of ExecutorRun). To workaround this, I've temporarily hacked up a solution that works, but is not ideal: since the receiveTuple DestFunction is passed the portal name, we can use that to lookup the Portal data structure for the cursor and then use that to get at the tuple store the Portal is using. This unnecessarily ties the Portal code with the tupleReceiver code, but it works... The proper fix for this is probably to change the DestFunction API -- Tom suggested passing the full QueryDesc to the receiveTuple function. In that case, callers of ExecutorRun could "subclass" QueryDesc to add any additional fields that their particular CommandDest needed to get access to. This approach would work, but I'd like to think about it for a little bit longer before deciding which route to go. In the mean time, the code works fine, so I don't think a fix is urgent. - (semi-related) I added a NO SCROLL keyword to DECLARE CURSOR, and adjusted the behavior of SCROLL in accordance with the discussion on -hackers. - (unrelated) Cleaned up some SGML markup in sql.sgml, copy.sgml Neil Conway
2003-03-27 17:51:29 +01:00
OpenTemporaryFile(bool keepOverTxn)
{
char tempfilepath[128];
1999-05-25 18:15:34 +02:00
File file;
1999-05-25 18:15:34 +02:00
/*
* Generate a tempfile name that's unique within the current
* transaction and database instance.
1999-05-25 18:15:34 +02:00
*/
snprintf(tempfilepath, sizeof(tempfilepath),
"%s/%s%d.%ld", PG_TEMP_FILES_DIR, PG_TEMP_FILE_PREFIX,
MyProcPid, tempFileCounter++);
/*
* Open the file. Note: we don't use O_EXCL, in case there is an
* orphaned temp file that can be reused.
*/
file = FileNameOpenFile(tempfilepath,
O_RDWR | O_CREAT | O_TRUNC | PG_BINARY,
0600);
if (file <= 0)
{
char *dirpath;
/*
* We might need to create the pg_tempfiles subdirectory, if no
* one has yet done so.
*
* Don't check for error from mkdir; it could fail if someone else
* just did the same thing. If it doesn't work then we'll bomb
* out on the second create attempt, instead.
*/
dirpath = filepath(PG_TEMP_FILES_DIR);
mkdir(dirpath, S_IRWXU);
pfree(dirpath);
file = FileNameOpenFile(tempfilepath,
O_RDWR | O_CREAT | O_TRUNC | PG_BINARY,
0600);
if (file <= 0)
elog(ERROR, "Failed to create temporary file %s", tempfilepath);
}
This patch implements holdable cursors, following the proposal (materialization into a tuple store) discussed on pgsql-hackers earlier. I've updated the documentation and the regression tests. Notes on the implementation: - I needed to change the tuple store API slightly -- it assumes that it won't be used to hold data across transaction boundaries, so the temp files that it uses for on-disk storage are automatically reclaimed at end-of-transaction. I added a flag to tuplestore_begin_heap() to control this behavior. Is changing the tuple store API in this fashion OK? - in order to store executor results in a tuple store, I added a new CommandDest. This works well for the most part, with one exception: the current DestFunction API doesn't provide enough information to allow the Executor to store results into an arbitrary tuple store (where the particular tuple store to use is chosen by the call site of ExecutorRun). To workaround this, I've temporarily hacked up a solution that works, but is not ideal: since the receiveTuple DestFunction is passed the portal name, we can use that to lookup the Portal data structure for the cursor and then use that to get at the tuple store the Portal is using. This unnecessarily ties the Portal code with the tupleReceiver code, but it works... The proper fix for this is probably to change the DestFunction API -- Tom suggested passing the full QueryDesc to the receiveTuple function. In that case, callers of ExecutorRun could "subclass" QueryDesc to add any additional fields that their particular CommandDest needed to get access to. This approach would work, but I'd like to think about it for a little bit longer before deciding which route to go. In the mean time, the code works fine, so I don't think a fix is urgent. - (semi-related) I added a NO SCROLL keyword to DECLARE CURSOR, and adjusted the behavior of SCROLL in accordance with the discussion on -hackers. - (unrelated) Cleaned up some SGML markup in sql.sgml, copy.sgml Neil Conway
2003-03-27 17:51:29 +01:00
/* Mark it for deletion at close */
VfdCache[file].fdstate |= FD_TEMPORARY;
This patch implements holdable cursors, following the proposal (materialization into a tuple store) discussed on pgsql-hackers earlier. I've updated the documentation and the regression tests. Notes on the implementation: - I needed to change the tuple store API slightly -- it assumes that it won't be used to hold data across transaction boundaries, so the temp files that it uses for on-disk storage are automatically reclaimed at end-of-transaction. I added a flag to tuplestore_begin_heap() to control this behavior. Is changing the tuple store API in this fashion OK? - in order to store executor results in a tuple store, I added a new CommandDest. This works well for the most part, with one exception: the current DestFunction API doesn't provide enough information to allow the Executor to store results into an arbitrary tuple store (where the particular tuple store to use is chosen by the call site of ExecutorRun). To workaround this, I've temporarily hacked up a solution that works, but is not ideal: since the receiveTuple DestFunction is passed the portal name, we can use that to lookup the Portal data structure for the cursor and then use that to get at the tuple store the Portal is using. This unnecessarily ties the Portal code with the tupleReceiver code, but it works... The proper fix for this is probably to change the DestFunction API -- Tom suggested passing the full QueryDesc to the receiveTuple function. In that case, callers of ExecutorRun could "subclass" QueryDesc to add any additional fields that their particular CommandDest needed to get access to. This approach would work, but I'd like to think about it for a little bit longer before deciding which route to go. In the mean time, the code works fine, so I don't think a fix is urgent. - (semi-related) I added a NO SCROLL keyword to DECLARE CURSOR, and adjusted the behavior of SCROLL in accordance with the discussion on -hackers. - (unrelated) Cleaned up some SGML markup in sql.sgml, copy.sgml Neil Conway
2003-03-27 17:51:29 +01:00
/* Mark it for deletion at EOXact */
if (!keepOverTxn)
VfdCache[file].fdstate |= FD_TXN_TEMPORARY;
return file;
}
/*
* close a file when done with it
*/
void
FileClose(File file)
{
Vfd *vfdP;
Assert(FileIsValid(file));
DO_DB(elog(LOG, "FileClose: %d (%s)",
file, VfdCache[file].fileName));
vfdP = &VfdCache[file];
if (!FileIsNotOpen(file))
{
/* remove the file from the lru ring */
Delete(file);
/* close the file */
if (close(vfdP->fd))
elog(LOG, "FileClose: failed to close %s: %m",
vfdP->fileName);
--nfile;
vfdP->fd = VFD_CLOSED;
}
/*
* Delete the file if it was temporary
*/
if (vfdP->fdstate & FD_TEMPORARY)
{
/* reset flag so that die() interrupt won't cause problems */
vfdP->fdstate &= ~FD_TEMPORARY;
if (unlink(vfdP->fileName))
elog(LOG, "FileClose: failed to unlink %s: %m",
vfdP->fileName);
}
/*
* Return the Vfd slot to the free list
*/
FreeVfd(file);
}
/*
* close a file and forcibly delete the underlying Unix file
*/
void
FileUnlink(File file)
{
Assert(FileIsValid(file));
DO_DB(elog(LOG, "FileUnlink: %d (%s)",
file, VfdCache[file].fileName));
/* force FileClose to delete it */
VfdCache[file].fdstate |= FD_TEMPORARY;
FileClose(file);
}
int
FileRead(File file, char *buffer, int amount)
{
int returnCode;
Assert(FileIsValid(file));
DO_DB(elog(LOG, "FileRead: %d (%s) %ld %d %p",
file, VfdCache[file].fileName,
VfdCache[file].seekPos, amount, buffer));
FileAccess(file);
returnCode = read(VfdCache[file].fd, buffer, amount);
if (returnCode > 0)
VfdCache[file].seekPos += returnCode;
else
VfdCache[file].seekPos = FileUnknownPos;
return returnCode;
}
int
FileWrite(File file, char *buffer, int amount)
{
int returnCode;
Assert(FileIsValid(file));
DO_DB(elog(LOG, "FileWrite: %d (%s) %ld %d %p",
file, VfdCache[file].fileName,
VfdCache[file].seekPos, amount, buffer));
FileAccess(file);
errno = 0;
returnCode = write(VfdCache[file].fd, buffer, amount);
/* if write didn't set errno, assume problem is no disk space */
if (returnCode != amount && errno == 0)
errno = ENOSPC;
if (returnCode > 0)
VfdCache[file].seekPos += returnCode;
else
VfdCache[file].seekPos = FileUnknownPos;
return returnCode;
}
long
FileSeek(File file, long offset, int whence)
{
Assert(FileIsValid(file));
DO_DB(elog(LOG, "FileSeek: %d (%s) %ld %ld %d",
file, VfdCache[file].fileName,
VfdCache[file].seekPos, offset, whence));
if (FileIsNotOpen(file))
{
switch (whence)
{
case SEEK_SET:
if (offset < 0)
elog(ERROR, "FileSeek: invalid offset: %ld", offset);
VfdCache[file].seekPos = offset;
break;
case SEEK_CUR:
VfdCache[file].seekPos += offset;
break;
case SEEK_END:
FileAccess(file);
VfdCache[file].seekPos = lseek(VfdCache[file].fd, offset, whence);
break;
default:
elog(ERROR, "FileSeek: invalid whence: %d", whence);
break;
}
}
else
{
switch (whence)
{
case SEEK_SET:
if (offset < 0)
elog(ERROR, "FileSeek: invalid offset: %ld", offset);
if (VfdCache[file].seekPos != offset)
VfdCache[file].seekPos = lseek(VfdCache[file].fd, offset, whence);
break;
case SEEK_CUR:
if (offset != 0 || VfdCache[file].seekPos == FileUnknownPos)
VfdCache[file].seekPos = lseek(VfdCache[file].fd, offset, whence);
break;
case SEEK_END:
VfdCache[file].seekPos = lseek(VfdCache[file].fd, offset, whence);
break;
default:
elog(ERROR, "FileSeek: invalid whence: %d", whence);
break;
}
}
return VfdCache[file].seekPos;
}
/*
* XXX not actually used but here for completeness
*/
#ifdef NOT_USED
long
FileTell(File file)
{
Assert(FileIsValid(file));
DO_DB(elog(LOG, "FileTell %d (%s)",
file, VfdCache[file].fileName));
return VfdCache[file].seekPos;
}
#endif
int
FileTruncate(File file, long offset)
{
int returnCode;
Assert(FileIsValid(file));
DO_DB(elog(LOG, "FileTruncate %d (%s)",
file, VfdCache[file].fileName));
FileAccess(file);
returnCode = ftruncate(VfdCache[file].fd, (size_t) offset);
1998-09-01 05:29:17 +02:00
return returnCode;
}
/*
* Routines that want to use stdio (ie, FILE*) should use AllocateFile
* rather than plain fopen(). This lets fd.c deal with freeing FDs if
1999-05-25 18:15:34 +02:00
* necessary to open the file. When done, call FreeFile rather than fclose.
*
* Note that files that will be open for any significant length of time
* should NOT be handled this way, since they cannot share kernel file
* descriptors with other files; there is grave risk of running out of FDs
* if anyone locks down too many FDs. Most callers of this routine are
* simply reading a config file that they will read and close immediately.
*
* fd.c will automatically close all files opened with AllocateFile at
* transaction commit or abort; this prevents FD leakage if a routine
* that calls AllocateFile is terminated prematurely by elog(ERROR).
*
* Ideally this should be the *only* direct call of fopen() in the backend.
*/
FILE *
AllocateFile(char *name, char *mode)
{
FILE *file;
DO_DB(elog(LOG, "AllocateFile: Allocated %d", numAllocatedFiles));
if (numAllocatedFiles >= MAX_ALLOCATED_FILES)
elog(ERROR, "AllocateFile: too many private FDs demanded");
TryAgain:
if ((file = fopen(name, mode)) != NULL)
{
allocatedFiles[numAllocatedFiles] = file;
numAllocatedFiles++;
return file;
}
if (errno == EMFILE || errno == ENFILE)
{
2001-03-22 05:01:46 +01:00
int save_errno = errno;
DO_DB(elog(LOG, "AllocateFile: not enough descs, retry, er= %d",
errno));
errno = 0;
if (ReleaseLruFile())
goto TryAgain;
errno = save_errno;
}
return NULL;
}
void
FreeFile(FILE *file)
{
1999-05-25 18:15:34 +02:00
int i;
DO_DB(elog(LOG, "FreeFile: Allocated %d", numAllocatedFiles));
/* Remove file from list of allocated files, if it's present */
1999-05-25 18:15:34 +02:00
for (i = numAllocatedFiles; --i >= 0;)
{
if (allocatedFiles[i] == file)
{
numAllocatedFiles--;
allocatedFiles[i] = allocatedFiles[numAllocatedFiles];
break;
}
}
if (i < 0)
elog(WARNING, "FreeFile: file was not obtained from AllocateFile");
fclose(file);
}
/*
* closeAllVfds
*
* Force all VFDs into the physically-closed state, so that the fewest
* possible number of kernel file descriptors are in use. There is no
* change in the logical state of the VFDs.
*/
void
closeAllVfds(void)
{
Index i;
if (SizeVfdCache > 0)
{
1999-05-25 18:15:34 +02:00
Assert(FileIsNotOpen(0)); /* Make sure ring not corrupted */
for (i = 1; i < SizeVfdCache; i++)
{
if (!FileIsNotOpen(i))
LruDelete(i);
}
}
}
/*
* AtEOXact_Files
*
* This routine is called during transaction commit or abort or backend
* exit (it doesn't particularly care which). All still-open temporary-file
* VFDs are closed, which also causes the underlying files to be deleted.
* Furthermore, all "allocated" stdio files are closed.
*/
void
AtEOXact_Files(void)
{
Index i;
if (SizeVfdCache > 0)
{
1999-05-25 18:15:34 +02:00
Assert(FileIsNotOpen(0)); /* Make sure ring not corrupted */
for (i = 1; i < SizeVfdCache; i++)
{
if ((VfdCache[i].fdstate & FD_TEMPORARY) &&
This patch implements holdable cursors, following the proposal (materialization into a tuple store) discussed on pgsql-hackers earlier. I've updated the documentation and the regression tests. Notes on the implementation: - I needed to change the tuple store API slightly -- it assumes that it won't be used to hold data across transaction boundaries, so the temp files that it uses for on-disk storage are automatically reclaimed at end-of-transaction. I added a flag to tuplestore_begin_heap() to control this behavior. Is changing the tuple store API in this fashion OK? - in order to store executor results in a tuple store, I added a new CommandDest. This works well for the most part, with one exception: the current DestFunction API doesn't provide enough information to allow the Executor to store results into an arbitrary tuple store (where the particular tuple store to use is chosen by the call site of ExecutorRun). To workaround this, I've temporarily hacked up a solution that works, but is not ideal: since the receiveTuple DestFunction is passed the portal name, we can use that to lookup the Portal data structure for the cursor and then use that to get at the tuple store the Portal is using. This unnecessarily ties the Portal code with the tupleReceiver code, but it works... The proper fix for this is probably to change the DestFunction API -- Tom suggested passing the full QueryDesc to the receiveTuple function. In that case, callers of ExecutorRun could "subclass" QueryDesc to add any additional fields that their particular CommandDest needed to get access to. This approach would work, but I'd like to think about it for a little bit longer before deciding which route to go. In the mean time, the code works fine, so I don't think a fix is urgent. - (semi-related) I added a NO SCROLL keyword to DECLARE CURSOR, and adjusted the behavior of SCROLL in accordance with the discussion on -hackers. - (unrelated) Cleaned up some SGML markup in sql.sgml, copy.sgml Neil Conway
2003-03-27 17:51:29 +01:00
(VfdCache[i].fdstate & FD_TXN_TEMPORARY) &&
VfdCache[i].fileName != NULL)
FileClose(i);
}
}
while (numAllocatedFiles > 0)
FreeFile(allocatedFiles[0]);
1999-05-25 18:15:34 +02:00
/*
* Reset the tempfile name counter to 0; not really necessary, but
* helps keep the names from growing unreasonably long.
*/
tempFileCounter = 0;
}
/*
* Remove old temporary files
*
* This should be called during postmaster startup. It will forcibly
* remove any leftover files created by OpenTemporaryFile.
*/
void
RemovePgTempFiles(void)
{
char db_path[MAXPGPATH];
char temp_path[MAXPGPATH];
char rm_path[MAXPGPATH];
DIR *db_dir;
DIR *temp_dir;
struct dirent *db_de;
struct dirent *temp_de;
/*
* Cycle through pg_tempfiles for all databases and remove old temp
* files.
*/
snprintf(db_path, sizeof(db_path), "%s/base", DataDir);
if ((db_dir = opendir(db_path)) != NULL)
{
while ((db_de = readdir(db_dir)) != NULL)
{
if (strcmp(db_de->d_name, ".") == 0 ||
strcmp(db_de->d_name, "..") == 0)
continue;
snprintf(temp_path, sizeof(temp_path),
"%s/%s/%s",
db_path, db_de->d_name,
PG_TEMP_FILES_DIR);
if ((temp_dir = opendir(temp_path)) != NULL)
{
while ((temp_de = readdir(temp_dir)) != NULL)
{
if (strcmp(temp_de->d_name, ".") == 0 ||
strcmp(temp_de->d_name, "..") == 0)
continue;
snprintf(rm_path, sizeof(temp_path),
"%s/%s/%s/%s",
db_path, db_de->d_name,
PG_TEMP_FILES_DIR,
temp_de->d_name);
if (strncmp(temp_de->d_name,
PG_TEMP_FILE_PREFIX,
strlen(PG_TEMP_FILE_PREFIX)) == 0)
unlink(rm_path);
else
{
/*
* would prefer to use elog here, but it's not up
* and running during postmaster startup...
*/
fprintf(stderr,
"Unexpected file found in temporary-files directory: %s\n",
rm_path);
}
}
closedir(temp_dir);
}
}
closedir(db_dir);
}
}