From eef644e57c38a79eb29bf9f3f05efbcee8fbdfce Mon Sep 17 00:00:00 2001 From: Tom Lane Date: Fri, 3 Jul 2026 13:50:14 -0400 Subject: [PATCH] Fix btree_gist's NotEqual strategy on internal index pages. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit gbt_var_consistent() handled the <> (BtreeGistNotEqual) strategy without distinguishing leaf from internal pages, unlike every other strategy. In particular, it tried to apply the datatype-specific f_eq method, which is completely wrong since internal keys might not have the same representation as leaf keys. This led to OOB reads and potentially crashes, and most likely to wrong query results as well. On leaf pages we can apply the inverse of what the Equal strategy does. On internal pages, use a correct implementation of what the previous code intended: we can descend if the query value equals both bounds, *so long as the bounds aren't truncated*. With truncated bounds we don't quite know the range of what's below, so we must always descend. Adjust the code in gbt_num_consistent() to look similar, too. This fixes a performance buglet in that there's no need to do two comparisons on a leaf entry, but the main point is just to keep code consistency. Reported-by: 王跃林 Author: Ayush Tiwari Reviewed-by: Tom Lane Discussion: https://postgr.es/m/AH*AvQCYKhQGVvPWi1GiU4oY.8.1781609375063.Hmail.3020001251@tju.edu.cn Backpatch-through: 14 --- contrib/btree_gist/btree_utils_num.c | 15 +++++++++++++-- contrib/btree_gist/btree_utils_var.c | 23 +++++++++++++++++++++-- 2 files changed, 34 insertions(+), 4 deletions(-) diff --git a/contrib/btree_gist/btree_utils_num.c b/contrib/btree_gist/btree_utils_num.c index 3affe4c2c46..b0e41de00e1 100644 --- a/contrib/btree_gist/btree_utils_num.c +++ b/contrib/btree_gist/btree_utils_num.c @@ -297,8 +297,19 @@ gbt_num_consistent(const GBT_NUMKEY_R *key, retval = tinfo->f_le(query, key->upper, flinfo); break; case BtreeGistNotEqualStrategyNumber: - retval = (!(tinfo->f_eq(query, key->lower, flinfo) && - tinfo->f_eq(query, key->upper, flinfo))); + if (is_leaf) + retval = !(tinfo->f_eq(query, key->lower, flinfo)); + else + { + /* + * If the upper/lower bounds are equal, then all entries below + * this node must have exactly that value. So we can avoid + * descending if the query equals both bounds. In all other + * cases, we must descend. + */ + retval = !(tinfo->f_eq(query, key->lower, flinfo) && + tinfo->f_eq(query, key->upper, flinfo)); + } break; default: retval = false; diff --git a/contrib/btree_gist/btree_utils_var.c b/contrib/btree_gist/btree_utils_var.c index bceadf5ae60..945fa67de3e 100644 --- a/contrib/btree_gist/btree_utils_var.c +++ b/contrib/btree_gist/btree_utils_var.c @@ -594,6 +594,13 @@ gbt_var_consistent(GBT_VARKEY_R *key, { bool retval = false; + /* + * Remember that f_cmp is for internal pages, f_eq etc for leaf pages, and + * on internal pages we need to check gbt_var_node_pf_match too. + * + * The leaf-page tests use swapped operands (e.g., f_gt(query, lower) + * means "lower < query"), which is why they look reversed. + */ switch (strategy) { case BTLessEqualStrategyNumber: @@ -634,8 +641,20 @@ gbt_var_consistent(GBT_VARKEY_R *key, || gbt_var_node_pf_match(key, query, tinfo); break; case BtreeGistNotEqualStrategyNumber: - retval = !(tinfo->f_eq(query, key->lower, collation, flinfo) && - tinfo->f_eq(query, key->upper, collation, flinfo)); + if (is_leaf) + retval = !(tinfo->f_eq(query, key->lower, collation, flinfo)); + else + { + /* + * If the upper/lower bounds are equal and not truncated, then + * all entries below this node must have exactly that value. + * So we can avoid descending if the query equals both bounds. + * In all other cases, we must descend. + */ + retval = tinfo->trnc || + !(tinfo->f_cmp(query, key->lower, collation, flinfo) == 0 && + tinfo->f_cmp(query, key->upper, collation, flinfo) == 0); + } break; default: retval = false;