- Notifications
You must be signed in to change notification settings - Fork5k
Commitb452900
committed
Revert "Add single-item cache when looking at topmost XID of a subtrans XID"
This reverts commit06f5295 as per issues with this approach, both interms of efficiency impact and stability. First, contrary to thesingle-item cache for transaction IDs in transam.c, the cache may finishby not be hit for a long time, and without an invalidation mechanism toclear it, it would cause inconsistent results on wraparound forexample. Second, the use of SubTransGetTopmostTransaction() for thecaching has a limited impact on performance. SubTransGetParent() couldhave more impact, though the benchmarking of the single-item approachstill needs to be proved, particularly under the conditions where SLRUlookups are stressed in parallel with overflowed snapshots (aka morethan 64 subxids generated, for example).After discussion with Andres Freund.Discussion:https://postgr.es/m/20220524235250.gtt3uu5zktfkr4hv@alap3.anarazel.de1 parentf1431f3 commitb452900
1 file changed
+0
-18
lines changedLines changed: 0 additions & 18 deletions
Original file line number | Diff line number | Diff line change | |
---|---|---|---|
| |||
54 | 54 |
| |
55 | 55 |
| |
56 | 56 |
| |
57 |
| - | |
58 |
| - | |
59 |
| - | |
60 |
| - | |
61 |
| - | |
62 |
| - | |
63 |
| - | |
64 |
| - | |
65 | 57 |
| |
66 | 58 |
| |
67 | 59 |
| |
| |||
163 | 155 |
| |
164 | 156 |
| |
165 | 157 |
| |
166 |
| - | |
167 |
| - | |
168 |
| - | |
169 |
| - | |
170 |
| - | |
171 |
| - | |
172 |
| - | |
173 | 158 |
| |
174 | 159 |
| |
175 | 160 |
| |
| |||
189 | 174 |
| |
190 | 175 |
| |
191 | 176 |
| |
192 |
| - | |
193 |
| - | |
194 |
| - | |
195 | 177 |
| |
196 | 178 |
| |
197 | 179 |
| |
|
0 commit comments
Comments
(0)