diff options
| author | Tom Lane | 2015-03-29 19:04:09 +0000 |
|---|---|---|
| committer | Tom Lane | 2015-03-29 19:04:09 +0000 |
| commit | e4cbfd673d530a3e841db26a74f22e11a991205a (patch) | |
| tree | 73b4a75797ef1f0bcd4ed3a348acbc47e3f227fc /src/backend/commands | |
| parent | 1601830ec20d56dc7bf6b60a34f69841429e4825 (diff) | |
Add vacuum_delay_point call in compute_index_stats's per-sample-row loop.
Slow functions in index expressions might cause this loop to take long
enough to make it worth being cancellable. Probably it would be enough
to call CHECK_FOR_INTERRUPTS here, but for consistency with other
per-sample-row loops in this file, let's use vacuum_delay_point.
Report and patch by Jeff Janes. Back-patch to all supported branches.
Diffstat (limited to 'src/backend/commands')
| -rw-r--r-- | src/backend/commands/analyze.c | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/src/backend/commands/analyze.c b/src/backend/commands/analyze.c index 366c4af27fa..d4d19148e57 100644 --- a/src/backend/commands/analyze.c +++ b/src/backend/commands/analyze.c @@ -742,6 +742,8 @@ compute_index_stats(Relation onerel, double totalrows, { HeapTuple heapTuple = rows[rowno]; + vacuum_delay_point(); + /* * Reset the per-tuple context each time, to reclaim any cruft * left behind by evaluating the predicate or index expressions. |