[arch-commits] Commit in mariadb/trunk (0007-MDEV-20646.patch PKGBUILD)
Christian Hesse
eworm at archlinux.org
Wed Dec 11 08:36:47 UTC 2019
Date: Wednesday, December 11, 2019 @ 08:36:46
Author: eworm
Revision: 370654
upgpkg: mariadb 10.4.11-1
new upstream release
Modified:
mariadb/trunk/PKGBUILD
Deleted:
mariadb/trunk/0007-MDEV-20646.patch
-----------------------+
0007-MDEV-20646.patch | 45 ---------------------------------------------
PKGBUILD | 15 +++++----------
2 files changed, 5 insertions(+), 55 deletions(-)
Deleted: 0007-MDEV-20646.patch
===================================================================
--- 0007-MDEV-20646.patch 2019-12-11 08:22:05 UTC (rev 370653)
+++ 0007-MDEV-20646.patch 2019-12-11 08:36:46 UTC (rev 370654)
@@ -1,45 +0,0 @@
-commit d4edb0510ec1189f65850bb47977e94ed98b1f71
-Author: Sergei Petrunia <psergey at askmonty.org>
-Date: Wed Nov 13 18:53:59 2019 +0300
-
- MDEV-20646: 10.3.18 is slower than 10.3.17
-
- Fix incorrect change introduced in the fix for MDEV-20109.
-
- The patch tried to compute a more precise estimate for the record_count
- value in SJ-Materialization-Scan strategy (in
- Sj_materialization_picker::check_qep). However the new formula is worse
- as it produces extremely optimistic results in common cases where
- SJ-Materialization-Scan should be used)
-
- The old formula produces pessimistic results in cases when Sj-Materialization-
- Scan is unlikely to be a good choice anyway. So, the old behavior is better.
-
-diff --git a/sql/opt_subselect.cc b/sql/opt_subselect.cc
-index aeafc13998a..a8afd952a4d 100644
---- a/sql/opt_subselect.cc
-+++ b/sql/opt_subselect.cc
-@@ -3029,7 +3029,22 @@ bool Sj_materialization_picker::check_qep(JOIN *join,
-
- *strategy= SJ_OPT_MATERIALIZE_SCAN;
- *read_time= prefix_cost;
-- *record_count= prefix_rec_count / mat_info->rows_with_duplicates;
-+ /*
-+ Note: the next line means we did not remove the subquery's fanout from
-+ *record_count. It needs to be removed, as the join prefix is
-+
-+ ntX SJM-SCAN(it1 ... itN) | (ot1 ... otN) ...
-+
-+ here, the SJM-SCAN may have introduced subquery's fanout (duplicate rows,
-+ rows that don't have matches in ot1_i). All this fanout is gone after
-+ table otN (or earlier) but taking it into account is hard.
-+
-+ Some consolation here is that SJM-Scan strategy is applicable when the
-+ subquery is smaller than tables otX. If the subquery has large cardinality,
-+ we can greatly overestimate *record_count here, but it doesn't matter as
-+ SJ-Materialization-Lookup is a better strategy anyway.
-+ */
-+ *record_count= prefix_rec_count;
- *handled_fanout= mat_nest->sj_inner_tables;
- return TRUE;
- }
Modified: PKGBUILD
===================================================================
--- PKGBUILD 2019-12-11 08:22:05 UTC (rev 370653)
+++ PKGBUILD 2019-12-11 08:36:46 UTC (rev 370654)
@@ -4,8 +4,8 @@
pkgbase=mariadb
pkgname=('mariadb-libs' 'mariadb-clients' 'mariadb' 'mytop')
pkgdesc='Fast SQL database server, derived from MySQL'
-pkgver=10.4.10
-pkgrel=2
+pkgver=10.4.11
+pkgrel=1
arch=('x86_64')
license=('GPL')
url='https://mariadb.org/'
@@ -19,14 +19,12 @@
source=("rsync://rsync.osuosl.org/mariadb/mariadb-${pkgver}/source/mariadb-${pkgver}.tar.gz"{,.asc}
'0001-arch-specific.patch'
'0002-systemd-sysusers-tmpfiles.patch'
- '0005-fix-galera_recovery-with-fs.protected_regular-enabled.patch'
- '0007-MDEV-20646.patch')
-sha256sums=('cd50fddf86c2a47405737e342f78ebd40d5716f0fb32b976245de713bed01421'
+ '0005-fix-galera_recovery-with-fs.protected_regular-enabled.patch')
+sha256sums=('4c076232b99433b09eb3c6d62f607192b3474d022703699b8f6aef4e79de3fb9'
'SKIP'
'ce72ea1563ad773e00e8b1c299babea176abae1102827c2f743921e9de615041'
'3e83467af80fbd53400a201a34fc858b88509ea8e88b10709947eb66545f9457'
- 'c8c801f80924ccb97b499552fe1c532b3ebf8f86cdfc0d23715d4adb1a8810f0'
- '825d4ab1601c9e97bf24857bcfd8bed2b6d8ad15a4c2411c7867bff2333b09d8')
+ 'c8c801f80924ccb97b499552fe1c532b3ebf8f86cdfc0d23715d4adb1a8810f0')
prepare() {
cd $pkgbase-$pkgver/
@@ -46,9 +44,6 @@
# fix galera_recovery with fs.protected_regular enabled
# https://github.com/MariaDB/server/pull/1137
patch -Np1 < ../0005-fix-galera_recovery-with-fs.protected_regular-enabled.patch
-
- # MDEV-20646: 10.3.18 is slower than 10.3.17
- patch -Np1 < ../0007-MDEV-20646.patch
}
build() {
More information about the arch-commits
mailing list