database/postgres
database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-05-11 05:41:32 +03:00
Code

Update btree_xlog_delete() comments.

Browse Source 
Commit fe97c61c updated LP_DEAD item deletion comments, but missed a
minor discrepancy on the REDO side.  Fix it now.

In passing, don't talk about the btree_xlog_vacuum() behavior within
btree_xlog_delete().  The reliance on XLOG_HEAP2_CLEANUP_INFO records
for recovery conflicts is already discussed within btvacuumpage() and
mentioned again in passing above btree_xlog_vacuum(), which seems
sufficient.
This commit is contained in:
Peter Geoghegan 2020-01-01 11:32:07 -08:00
parent 7559d8ebfa
commit c5f3b53b0e
2 changed files with 6 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
src/backend/access/nbtree/nbtree.c
View File

@ -1208,9 +1208,10 @@ restart:
* as long as the callback function only considers whether the * as long as the callback function only considers whether the
* index tuple refers to pre-cutoff heap tuples that were * index tuple refers to pre-cutoff heap tuples that were
* certainly already pruned away during VACUUM's initial heap * certainly already pruned away during VACUUM's initial heap
* scan by the time we get here. (We can rely on conflicts * scan by the time we get here. (XLOG_HEAP2_CLEANUP_INFO
* produced by heap pruning, rather than producing our own * records produce conflicts using a latestRemovedXid value
* now.) * for the entire VACUUM, so there is no need to produce our
* own conflict now.)
* *
* Backends with snapshots acquired after a VACUUM starts but * Backends with snapshots acquired after a VACUUM starts but
* before it finishes could have a RecentGlobalXmin with a * before it finishes could have a RecentGlobalXmin with a

14
src/backend/access/nbtree/nbtxlog.c
View File

@ -428,14 +428,7 @@ btree_xlog_delete(XLogReaderState *record)
/* /*
* If we have any conflict processing to do, it must happen before we * If we have any conflict processing to do, it must happen before we
* update the page. * update the page
*
* Btree delete records can conflict with standby queries. You might
* think that vacuum records would conflict as well, but we've handled
* that already. XLOG_HEAP2_CLEANUP_INFO records provide the highest xid
* cleaned by the vacuum of the heap and so we can resolve any conflicts
* just once when that arrives. After that we know that no conflicts
* exist from individual btree vacuum records on that index.
*/ */
if (InHotStandby) if (InHotStandby)
{ {
@ -463,10 +456,7 @@ btree_xlog_delete(XLogReaderState *record)
PageIndexMultiDelete(page, unused, xlrec->nitems); PageIndexMultiDelete(page, unused, xlrec->nitems);
} }
/* /* Mark the page as not containing any LP_DEAD items */
* Mark the page as not containing any LP_DEAD items --- see comments
* in _bt_delitems_delete().
*/
opaque = (BTPageOpaque) PageGetSpecialPointer(page); opaque = (BTPageOpaque) PageGetSpecialPointer(page);
opaque->btpo_flags &= ~BTP_HAS_GARBAGE; opaque->btpo_flags &= ~BTP_HAS_GARBAGE;