planner, statistics: use the correct column ID when recording stats loading status #52208

time-and-fate · 2024-03-28T11:52:04Z

What problem does this PR solve?

Issue Number: close #52207

What changed and how does it work?

Previously, this place incorrectly used the UniqueID, which is only valid during this query.

We should use the column ID from the metadata so that we can correctly fetch the column name from TableInfo.

Check List

Tests

Unit test
Integration test
Manual test (add detailed scripts or steps below)
No need to test
- I checked and no code files have been changed.

Side effects

Performance regression: Consumes more CPU
Performance regression: Consumes more Memory
Breaking backward compatibility

Documentation

Release note

Please refer to Release Notes Language Style Guide to write a quality release note.

None

pkg/statistics/table.go

codecov · 2024-03-28T23:12:22Z

Codecov Report

Merging #52208 (688b40d) into master (c9a250a) will increase coverage by 2.1758%.
Report is 12 commits behind head on master.
The diff coverage is 88.8888%.

Additional details and impacted files

@@               Coverage Diff                @@
##             master     #52208        +/-   ##
================================================
+ Coverage   71.1432%   73.3190%   +2.1758%     
================================================
  Files          1468       1468                
  Lines        432193     433639      +1446     
================================================
+ Hits         307476     317940     +10464     
+ Misses       105490      95852      -9638     
- Partials      19227      19847       +620

Flag	Coverage Δ
integration	`48.8438% <85.1851%> (?)`
unit	`70.9403% <87.0370%> (-0.0186%)`	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Components	Coverage Δ
dumpling	`53.9957% <ø> (ø)`
parser	`∅ <ø> (∅)`
br	`41.4176% <ø> (+7.3304%)`	⬆️

elsa0520 · 2024-04-01T08:58:18Z

pkg/planner/cardinality/cross_estimation.go

@@ -152,7 +152,7 @@ func crossEstimateRowCount(sctx context.PlanContext,
 		return 0, err == nil, corr
 	}
 	idxID := int64(-1)
-	idxIDs, idxExists := dsStatsInfo.HistColl.ColID2IdxIDs[colID]


It is better to use the clearly name 'colUniqueID' instead of 'colID' to avoid the wrong used.

elsa0520 · 2024-04-01T09:00:25Z

pkg/planner/cardinality/row_count_column.go

@@ -45,7 +45,11 @@ func GetRowCountByColumnRanges(sctx context.PlanContext, coll *statistics.HistCo
 	}
 	sc := sctx.GetSessionVars().StmtCtx
 	c, ok := coll.Columns[colID]
-	recordUsedItemStatsStatus(sctx, c, coll.PhysicalID, colID)
+	colInfoID := colID


Same as above

elsa0520 · 2024-04-01T09:05:10Z

pkg/planner/cardinality/row_count_column.go

+	if len(coll.UniqueID2colInfoID) > 0 {
+		colInfoID = coll.UniqueID2colInfoID[colID]
+	}
+	recordUsedItemStatsStatus(sctx, c, coll.PhysicalID, colInfoID)


Maybe we should distinguish colId and colUniqueId in explain such as "colID: xxx" , "colUniqueID: XXX"

We want the column name in the EXPLAIN result, which should be fetched using the column ID from the metadata instead of the UniqueID.

elsa0520 · 2024-04-01T09:05:31Z

pkg/planner/cardinality/row_count_index.go

@@ -170,15 +170,15 @@ func getIndexRowCountForStatsV1(sctx context.PlanContext, coll *statistics.HistC
 			}
 			var count float64
 			var err error
-			colIDs := coll.Idx2ColumnIDs[idxID]
+			colIDs := coll.Idx2ColUniqueIDs[idxID]


same as above

Index ID is not relevant here, only Column ID has this difference.

Rustin170506

Looks good to me. Thanks! 👍

/hold

Feel free to update the testing code or ignore it.

Rustin170506 · 2024-04-02T06:16:21Z

build/nogo_config.json

@@ -175,6 +175,7 @@
  "fieldalignment": {
    "exclude_files": {
      "pkg/parser/parser.go": "parser/parser.go code",
+      "pkg/statistics/table.go": "disable this limitation that prevents us from splitting struct fields for clarity",


I always want to do this, but I can never convince myself. 😃

😂 I think performance benefits and memory efficiency from this rule are small. So I think it's totally OK to disable it when it causes problems or inconvenience.

Rustin170506 · 2024-04-02T06:21:24Z

pkg/statistics/table.go

@@ -215,17 +215,11 @@ const (

 // HistColl is a collection of histogram. It collects enough information for plan to calculate the selectivity.
 type HistColl struct {
-	Columns map[int64]*Column


Big thanks for refactoring this structure!

Rustin170506 · 2024-04-02T06:22:59Z

pkg/statistics/table.go

-	Idx2ColumnIDs map[int64][]int64
-	// ColID2IdxIDs maps the column id to a list index ids whose first column is it. It's used to calculate the selectivity in planner.
-	ColID2IdxIDs map[int64][]int64
-	// MVIdx2Columns maps the index id to its columns by expression.Column.
-	// For normal index, the column id is enough, as we already have in Idx2ColumnIDs. But currently, mv index needs more
-	// information to match the filter against the mv index columns, and we need this map to provide this information.
-	MVIdx2Columns map[int64][]*expression.Column


nit: It seems we still using some of these names in our testing code. Feel free to update it or ignore it.

From what I found, that's used inside a function, I think that's OK.

ti-chi-bot · 2024-04-02T06:23:40Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: hi-rustin, winoros

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

~~OWNERS~~ [hi-rustin,winoros]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

ti-chi-bot · 2024-04-02T06:23:44Z

[LGTM Timeline notifier]

Timeline:

2024-03-28 23:08:01.616151721 +0000 UTC m=+2195708.638398110: ☑️ agreed by winoros.
2024-04-02 06:23:43.792525531 +0000 UTC m=+338685.320066077: ☑️ agreed by hi-rustin.

time-and-fate · 2024-04-02T08:10:32Z

/unhold

time-and-fate · 2024-04-02T09:00:37Z

/cherry-pick release-7.5
/cherry-pick release-7.1

Signed-off-by: ti-chi-bot <[email protected]>

ti-chi-bot · 2024-04-02T09:01:23Z

@time-and-fate: new pull request created to branch release-7.1: #52308.

In response to this:

/cherry-pick release-7.5
/cherry-pick release-7.1

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the ti-community-infra/tichi repository.

ti-chi-bot · 2024-04-02T09:02:05Z

@time-and-fate: new pull request created to branch release-7.5: #52309.

In response to this:

/cherry-pick release-7.5
/cherry-pick release-7.1

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the ti-community-infra/tichi repository.

Signed-off-by: ti-chi-bot <[email protected]>

…oading status (#52208) (#52320) close #52207

…oading status (#52208) (#52309) close #52207

time-and-fate added 3 commits March 28, 2024 19:40

fix

c35daf5

add comments and update bazel

e60c752

Merge remote-tracking branch 'upstream/master' into issue52207

72dfc6a

ti-chi-bot bot added release-note-none Denotes a PR that doesn't merit a release note. size/L Denotes a PR that changes 100-499 lines, ignoring generated files. labels Mar 28, 2024

fix fmt

33291ed

winoros approved these changes Mar 28, 2024

View reviewed changes

ti-chi-bot bot added approved needs-1-more-lgtm Indicates a PR needs 1 more LGTM. labels Mar 28, 2024

winoros reviewed Mar 28, 2024

View reviewed changes

pkg/statistics/table.go Outdated Show resolved Hide resolved

time-and-fate added 7 commits March 29, 2024 16:44

rename the fields as the comments suggest

a44e2b2

Merge remote-tracking branch 'upstream/master' into issue52207

93eb1e6

update comments

51918df

Merge remote-tracking branch 'upstream/master' into issue52207

a7eab99

fix lint

9808c88

fix lint

a320f42

disable fieldalignment check for table.go

65c5878

elsa0520 reviewed Apr 1, 2024

View reviewed changes

time-and-fate added 2 commits April 1, 2024 17:25

update var name according to comments

6758bd7

Merge remote-tracking branch 'upstream/master' into issue52207

688b40d

time-and-fate requested review from Rustin170506 and elsa0520 April 2, 2024 04:34

Rustin170506 approved these changes Apr 2, 2024

View reviewed changes

ti-chi-bot bot added the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Apr 2, 2024

ti-chi-bot bot added lgtm and removed needs-1-more-lgtm Indicates a PR needs 1 more LGTM. labels Apr 2, 2024

ti-chi-bot bot removed the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Apr 2, 2024

ti-chi-bot bot merged commit 21e9d3c into pingcap:master Apr 2, 2024
23 checks passed

ti-chi-bot pushed a commit to ti-chi-bot/tidb that referenced this pull request Apr 2, 2024

This is an automated cherry-pick of pingcap#52208

9789623

Signed-off-by: ti-chi-bot <[email protected]>

ti-chi-bot mentioned this pull request Apr 2, 2024

planner, statistics: use the correct column ID when recording stats loading status (#52208) #52308

Closed

13 tasks

ti-chi-bot pushed a commit to ti-chi-bot/tidb that referenced this pull request Apr 2, 2024

This is an automated cherry-pick of pingcap#52208

05e413e

Signed-off-by: ti-chi-bot <[email protected]>

ti-chi-bot mentioned this pull request Apr 2, 2024

planner, statistics: use the correct column ID when recording stats loading status (#52208) #52309

Merged

13 tasks

time-and-fate mentioned this pull request Apr 2, 2024

planner, statistics: use the correct column ID when recording stats loading status (#52208) #52320

Merged

13 tasks

ti-chi-bot bot pushed a commit that referenced this pull request Apr 8, 2024

planner, statistics: use the correct column ID when recording stats l…

fd9583b

…oading status (#52208) (#52320) close #52207

ti-chi-bot bot pushed a commit that referenced this pull request Apr 8, 2024

planner, statistics: use the correct column ID when recording stats l…

b2d6b97

…oading status (#52208) (#52309) close #52207

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

planner, statistics: use the correct column ID when recording stats loading status #52208

planner, statistics: use the correct column ID when recording stats loading status #52208

time-and-fate commented Mar 28, 2024

codecov bot commented Mar 28, 2024 •

edited

Loading

elsa0520 Apr 1, 2024

elsa0520 Apr 1, 2024

elsa0520 Apr 1, 2024

time-and-fate Apr 1, 2024

elsa0520 Apr 1, 2024

time-and-fate Apr 1, 2024

Rustin170506 left a comment

Rustin170506 Apr 2, 2024

time-and-fate Apr 2, 2024

Rustin170506 Apr 2, 2024

Rustin170506 Apr 2, 2024

time-and-fate Apr 2, 2024

ti-chi-bot bot commented Apr 2, 2024

ti-chi-bot bot commented Apr 2, 2024

time-and-fate commented Apr 2, 2024

time-and-fate commented Apr 2, 2024

ti-chi-bot commented Apr 2, 2024

ti-chi-bot commented Apr 2, 2024

planner, statistics: use the correct column ID when recording stats loading status #52208

planner, statistics: use the correct column ID when recording stats loading status #52208

Conversation

time-and-fate commented Mar 28, 2024

What problem does this PR solve?

What changed and how does it work?

Check List

Release note

codecov bot commented Mar 28, 2024 • edited Loading

Codecov Report

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Rustin170506 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ti-chi-bot bot commented Apr 2, 2024

ti-chi-bot bot commented Apr 2, 2024

[LGTM Timeline notifier]

time-and-fate commented Apr 2, 2024

time-and-fate commented Apr 2, 2024

ti-chi-bot commented Apr 2, 2024

ti-chi-bot commented Apr 2, 2024

codecov bot commented Mar 28, 2024 •

edited

Loading