Re: [Monetdb-developers] [Monetdb-checkins] MonetDB4/src/modules/contrib malalgebra.mx, MonetDB_4-24, 1.8.2.2, 1.8.2.3
Did you see what I had done on the head branch for this problem?
On 2008-11-03 22:51, Stefan Manegold wrote:
Update of /cvsroot/monetdb/MonetDB4/src/modules/contrib In directory 23jxhf1.ch3.sourceforge.com:/tmp/cvs-serv6583/MonetDB4/src/modules/contrib
Modified Files: Tag: MonetDB_4-24 malalgebra.mx Log Message:
attempt to fix bug in Peter's new {band,left}mergejoin_limit() functions that was reported by Riham:
While {band,left}mergejoin_limit() internally use a limit of type size_t, the MIL level uses type lng; hence, we need to properly convert the types back and forth (and check for potential under-/overflows).
(For now, I did not dare to change the MIL interface to use the more suitable type wrd that (unlike lng) grows with system word size just like size_t.)
Riham,
could you please check, whether this fixes your problem? Thanks!
U malalgebra.mx Index: malalgebra.mx =================================================================== RCS file: /cvsroot/monetdb/MonetDB4/src/modules/contrib/malalgebra.mx,v retrieving revision 1.8.2.2 retrieving revision 1.8.2.3 diff -u -d -r1.8.2.2 -r1.8.2.3 --- malalgebra.mx 28 Oct 2008 17:49:34 -0000 1.8.2.2 +++ malalgebra.mx 3 Nov 2008 21:51:09 -0000 1.8.2.3 @@ -263,28 +263,50 @@ }
int -CMDbandmergejoin_limit(BAT **result, BAT *left, BAT *right, ptr minus, ptr plus, bit *li, bit *hi, lng* limit) +CMDbandmergejoin_limit(BAT **result, BAT *left, BAT *right, ptr minus, ptr plus, bit *li, bit *hi, lng* _limit) {
- size_t cutoff = *limit;
lng limit = *_limit;size_t cutoff = (size_t) limit;assert(limit >= 0);+#if SIZEOF_SIZE_T == SIZEOF_INT
assert(limit <= GDK_int_max);+#endif *result = BATbandmergejoin_limit(left, right, minus, plus, *li, *hi, &cutoff); if (*result) { bat bid = (*result)->batCacheid; *result = BATnew(TYPE_lng, TYPE_bat, 1);
if (*result) BUNins(*result, &cutoff, &bid, FALSE);
if (*result) {+#if SIZEOF_SIZE_T == SIZEOF_LNG
assert(cutoff <= (size_t) GDK_lng_max);+#endif
limit = (lng) cutoff;BUNins(*result, &limit, &bid, FALSE);} BBPunfix(bid); } return (*result)?GDK_SUCCEED:GDK_FAIL;}
int -CMDleftmergejoin_limit(BAT **result, BAT *left, BAT *right, lng* limit) +CMDleftmergejoin_limit(BAT **result, BAT *left, BAT *right, lng* _limit) {
- size_t cutoff = *limit;
lng limit = *_limit;size_t cutoff = (size_t) limit;assert(limit >= 0);+#if SIZEOF_SIZE_T == SIZEOF_INT
assert(limit <= GDK_int_max);+#endif *result = BATleftmergejoin_limit(left, right, cutoff, &cutoff); if (*result) { bat bid = (*result)->batCacheid; *result = BATnew(TYPE_lng, TYPE_bat, 1);
if (*result) BUNins(*result, &cutoff, &bid, FALSE);
if (*result) {+#if SIZEOF_SIZE_T == SIZEOF_LNG
assert(cutoff <= (size_t) GDK_lng_max);+#endif
limit = (lng) cutoff;BUNins(*result, &limit, &bid, FALSE);} BBPunfix(bid); } return (*result)?GDK_SUCCEED:GDK_FAIL;
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Monetdb-checkins mailing list Monetdb-checkins@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-checkins
On Mon, Nov 03, 2008 at 11:09:15PM +0100, Sjoerd Mullender wrote:
Did you see what I had done on the head branch for this problem?
yes --- but is does (IMHO) not help: in both cases, BUNins (into a [lng,BAT] BAT) was/is called with a *size_t or *BUN argument for the head value, which will than be dereferences as lng --- a 4 byte vs. 8 byte mismatch in case size_t (or BUN) is 32-bit ...
Stefan
ps: yes, my "fix" will require some (local "minor") manual conflict-fixing during propagattion --- I'll take care of that once Riham has checked my "fix" ...
On 2008-11-03 22:51, Stefan Manegold wrote:
Update of /cvsroot/monetdb/MonetDB4/src/modules/contrib In directory 23jxhf1.ch3.sourceforge.com:/tmp/cvs-serv6583/MonetDB4/src/modules/contrib
Modified Files: Tag: MonetDB_4-24 malalgebra.mx Log Message:
attempt to fix bug in Peter's new {band,left}mergejoin_limit() functions that was reported by Riham:
While {band,left}mergejoin_limit() internally use a limit of type size_t, the MIL level uses type lng; hence, we need to properly convert the types back and forth (and check for potential under-/overflows).
(For now, I did not dare to change the MIL interface to use the more suitable type wrd that (unlike lng) grows with system word size just like size_t.)
Riham,
could you please check, whether this fixes your problem? Thanks!
U malalgebra.mx Index: malalgebra.mx =================================================================== RCS file: /cvsroot/monetdb/MonetDB4/src/modules/contrib/malalgebra.mx,v retrieving revision 1.8.2.2 retrieving revision 1.8.2.3 diff -u -d -r1.8.2.2 -r1.8.2.3 --- malalgebra.mx 28 Oct 2008 17:49:34 -0000 1.8.2.2 +++ malalgebra.mx 3 Nov 2008 21:51:09 -0000 1.8.2.3 @@ -263,28 +263,50 @@ }
int -CMDbandmergejoin_limit(BAT **result, BAT *left, BAT *right, ptr minus, ptr plus, bit *li, bit *hi, lng* limit) +CMDbandmergejoin_limit(BAT **result, BAT *left, BAT *right, ptr minus, ptr plus, bit *li, bit *hi, lng* _limit) {
- size_t cutoff = *limit;
lng limit = *_limit;size_t cutoff = (size_t) limit;assert(limit >= 0);+#if SIZEOF_SIZE_T == SIZEOF_INT
assert(limit <= GDK_int_max);+#endif *result = BATbandmergejoin_limit(left, right, minus, plus, *li, *hi, &cutoff); if (*result) { bat bid = (*result)->batCacheid; *result = BATnew(TYPE_lng, TYPE_bat, 1);
if (*result) BUNins(*result, &cutoff, &bid, FALSE);
if (*result) {+#if SIZEOF_SIZE_T == SIZEOF_LNG
assert(cutoff <= (size_t) GDK_lng_max);+#endif
limit = (lng) cutoff;BUNins(*result, &limit, &bid, FALSE);} BBPunfix(bid); } return (*result)?GDK_SUCCEED:GDK_FAIL;}
int -CMDleftmergejoin_limit(BAT **result, BAT *left, BAT *right, lng* limit) +CMDleftmergejoin_limit(BAT **result, BAT *left, BAT *right, lng* _limit) {
- size_t cutoff = *limit;
lng limit = *_limit;size_t cutoff = (size_t) limit;assert(limit >= 0);+#if SIZEOF_SIZE_T == SIZEOF_INT
assert(limit <= GDK_int_max);+#endif *result = BATleftmergejoin_limit(left, right, cutoff, &cutoff); if (*result) { bat bid = (*result)->batCacheid; *result = BATnew(TYPE_lng, TYPE_bat, 1);
if (*result) BUNins(*result, &cutoff, &bid, FALSE);
if (*result) {+#if SIZEOF_SIZE_T == SIZEOF_LNG
assert(cutoff <= (size_t) GDK_lng_max);+#endif
limit = (lng) cutoff;BUNins(*result, &limit, &bid, FALSE);} BBPunfix(bid); } return (*result)?GDK_SUCCEED:GDK_FAIL;
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Monetdb-checkins mailing list Monetdb-checkins@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-checkins
-- Sjoerd Mullender
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Monetdb-checkins mailing list Monetdb-checkins@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-checkins
participants (2)
-
Sjoerd Mullender -
Stefan Manegold