Skip to content

profcheck: implement duplicate mappings check #108

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
florianl wants to merge 5 commits into
base: main
Choose a base branch
from
+322 −227
Open

profcheck: implement duplicate mappings check #108

Show file tree
Hide file tree
Changes from 4 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
72fa6e1
profcheck: implement duplicate mappings check
florianl Apr 18, 2026
6418b2b
fix bug: all fields must be equal
florianl Apr 20, 2026
306497f
use sorted slices
florianl Apr 21, 2026
c1acfdf
add fileoffset
florianl Apr 21, 2026
4da74dc
sort only once
florianl Apr 23, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
36 changes: 34 additions & 2 deletions profcheck/check.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,7 @@ package profcheck
import (
"errors"
"fmt"
"slices"

profiles "go.opentelemetry.io/proto/otlp/profiles/v1development"
"google.golang.org/protobuf/proto"
Expand Down Expand Up @@ -212,7 +213,17 @@ func (c ConformanceChecker) checkMappingTable(mappingTable []*profiles.Mapping,
if err := checkZeroVal(mappingTable); err != nil {
errs = errors.Join(errs, err)
}
for idx, m := range mappingTable {

type uniqMapping struct {
filenameStrIdx int32
attrIdxs []int32
memStart uint64
memLimit uint64
fileOffset uint64
}
var uniqMappings []uniqMapping

for idx, m := range mappingTable[1:] {
if err := c.checkIndex(len(dict.StringTable), m.FilenameStrindex); err != nil {
errs = errors.Join(errs, prefixErrorf(err, "[%d].filename_strindex", idx))
}
Expand All @@ -222,8 +233,29 @@ func (c ConformanceChecker) checkMappingTable(mappingTable []*profiles.Mapping,
if !(m.MemoryStart == 0 && m.MemoryLimit == 0) && !(m.MemoryStart < m.MemoryLimit) {
errs = errors.Join(errs, fmt.Errorf("[%d]: memory_start=%016x, memory_limit=%016x: must be both zero or start < limit", idx, m.MemoryStart, m.MemoryLimit))
}
if c.CheckDictionaryDuplicates {
Copy link
Copy Markdown
Member Author

@florianl florianl Apr 18, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If this style of duplicate entry check is fine for everyone, I can apply it to the other tables as well.

newMapping := uniqMapping{
filenameStrIdx: m.FilenameStrindex,
attrIdxs: m.AttributeIndices,
memStart: m.MemoryStart,
memLimit: m.MemoryLimit,
Comment thread
florianl marked this conversation as resolved.
fileOffset: m.FileOffset,
}
if slices.ContainsFunc(uniqMappings, func(e uniqMapping) bool {
slices.Sort(newMapping.attrIdxs)
slices.Sort(e.attrIdxs)
Copy link
Copy Markdown
Member

@christos68k christos68k Apr 22, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We're doing a lot of redundant operations here (like iterating uniqMappings for every mapping, sorting e.attrIdxs repeatedly instead of once) and depending on number of mappings, it could get expensive.

Can't we switch uniqMappings to be a map instead? We can keep uniqMapping as the key if we switch attrIdxs to be a string of sorted indices.

Copy link
Copy Markdown
Member Author

@florianl florianl Apr 23, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can't we switch uniqMappings to be a map instead? We can keep uniqMapping as the key if we switch attrIdxs to be a string of sorted indices.

Unfortunatly it is not possible to use uniqMapping as key to a map, as uniqMapping contains a slice as a field. See https://go.dev/play/p/4-wpRA-KV3Q.

Copy link
Copy Markdown
Member

@christos68k christos68k Apr 23, 2026 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That's why I mentioned switching attrIdxs to be a string of a sorted indices. Alternatively, we can sort and hash the slice like we do in the eBPF profiler reporter for location indices.

return newMapping.filenameStrIdx == e.filenameStrIdx &&
slices.Equal(newMapping.attrIdxs, e.attrIdxs) &&
Comment thread
florianl marked this conversation as resolved.
newMapping.memStart == e.memStart &&
newMapping.memLimit == e.memLimit &&
newMapping.fileOffset == e.fileOffset
}) {
errs = errors.Join(errs, fmt.Errorf("duplicate mapping at index %d: %v", idx, m))
continue
}
uniqMappings = append(uniqMappings, newMapping)
}
}
// TODO: Add optional uniqueness check.
// TODO: Add optional unreferenced entries check.
return errs
}
Expand Down
Loading