151 merged changes. Check any of them.

Since June, the maintainers of 36 projects have merged 151 changes that started with Perfloop. Go, PostgreSQL®, Redis® and Elasticsearch are among them.

36 projects132 pull requests, 6 commits and 13 project fixes of issues Perfloop reported

Cumulative merged changes from Jun 21 to Nov 18, 2026, reaching 151. The list at the end of the page holds every change.

Each square is one merged change, in the colour of its pattern. Point at one to read it; select it to open the merge on GitHub.

36projects merged a change
36maintainers pressed merge
2.6daysmedian time from a pull request opening to its merge
44pull requests merged within a day, of 132

Merged per month

Through September, 51 of the 53 merged changes were performance gains. Since October 1, 71 bug fixes have merged.

2June, from the 21st

Agents write the software. Perfloop makes it self-improving. It finds what is slow, wrong or unsafe and proves every change.

It learns from the answers it gets. Perfloop reads maintainer review comments, closed pull requests and failed checks, and keeps each lesson with the code it came from.

Ten projects you know

Ten of the 36, each with one of its merged changes.

Go

Programming language, 139,957 stars on GitHub

A recover in a range-over-func loop body returned nil after the goroutine stack grew or shrank. Merged Nov 10, by Cuong Manh Le (their fix, Perfloop's report)

bug fixed

PostgreSQL®

Relational database, 22,843 stars on GitHub

Freeing a WAL reader leaked separately allocated records still in its decode queue. Merged Nov 4, by Fujii Masao

bug fixed

Redis®

In-memory data store, 77,520 stars on GitHub

After the IDMP expiry cron pruned a stream, CLUSTER SLOT-STATS kept counting the freed memory. Merged Oct 28, by Sergei Georgiev

bug fixed

Elasticsearch

Search engine, 78,689 stars on GitHub

Async cross-cluster searches returned the wrong page of hits while a remote cluster was still running. Merged Nov 11, by Eugene Rizhkov

bug fixed

Prometheus®

Monitoring system, 67,647 stars on GitHub

PromQL® range functions built a fresh one-sample vector for every scalar argument, series and query step. 6 changes merged. This one: Nov 2, by Bartlomiej Plotka

2.1× lower

curl

Data transfer tool and library, 43,527 stars on GitHub

Since curl 8.18.0, cookie-file imports could overwrite cookies added with CURLOPT_COOKIELIST before the first transfer, contrary to their documented priority. 2 changes merged. This one: Nov 14, by Daniel Stenberg

bug fixed

CoreDNS

The default DNS server in Kubernetes®, 14,433 stars on GitHub

The cache cut fractional SERVFAIL lifetimes to whole seconds: a 500ms failure expired at once, and a 1.5s failure could be retried after one second. 4 changes merged. This one: Nov 15, by Yong Tang

bug fixed

LiveKit

Real-time audio and video server, 22,235 stars on GitHub

Forwarding packets with common 4-byte dependency descriptors allocated memory for their headers. 2 changes merged. This one: Sep 30, by cnderrauber

allocations to zero

Distribution

Container image registry, 10,690 stars on GitHub

A cold read of a small blob reserved a 4 MiB buffer. 6 changes merged. This one: Nov 10, by Milos Gajdos

754× lower

Dolt

SQL database with version control, 24,908 stars on GitHub

Undrained keyless secondary-index scans left producer goroutines blocked on full channels. 2 changes merged. This one: Oct 13, by Nick Tobey

bug fixed

What it finds

127 of the 151 changes are in Go. The rest are in Rust, C, Python, C++, C# and Java.

How large the measured gains are, one square for each change

The largest measured gains in open source projects

Distribution #5070Memory per direct download, 575-byte image config754× lowerRoaringBitmap #532BenchmarkBatchEqual31× lowerZoekt #1110Sharded search, 128 repos x 64 matches, no display limit26× lowerZoekt #1127Branch shard filtering, 10,000 shards, 90% excluded24× lowerRoaringBitmap #539CompareValue result serialization, 50k contiguous matches14× lowerRoaringBitmap #535Bitmap.Or fresh-interleaved-40966.9× lowerDistribution #4906Schema 2 manifest verification, 10 ms driver latency6.0× lowerAxiom HyperLogLog #75Populated NewNoSparse p14 Estimate, ns/op5.7× lower

Before against after, on the measurement the change itself reports. Bar length is on a log scale. Most are component benchmarks, and the line under each name says what was measured.

One change, from finding to merge

parseablehq/parseable#1846, merged Nov 9 by Nikhil Sinha.

found

Each streamed SQL batch was written to JSON by Arrow, parsed back into a value tree and serialized a second time.

changed

Perfloop returns Arrow's JSON buffer for each streamed batch directly, keeping the newline framing and field-name prefix.

measured

Streamed SELECT * response time, 1,000,000 rows4.5× lower
before4.05 safter0.87 s
Time per 20,000-row streamed batch, in-process handler12× lower
before26,717,460 nsafter2,340,867 ns
Allocation calls per 20,000-row streamed batch227× lower
before180,832 callsafter796 calls

Streamed SELECT * over 1,000,000 rows (346 MB, 53 batches) · 7 interleaved pairs

On your code

The same process that produced this record, on the repositories you choose.

  1. You connect the repositories you choose.

    Perfloop works only on the repositories an admin selects.
  2. It looks for what is slow, wrong or unsafe.

    Performance problems, bugs and security flaws.
  3. It proves each change before you see it.

    By a measurement, or by a test that asserts the fix.
  4. It opens a pull request.

    The evidence is in its description. sign up →

Tigris Data runs Perfloop on its storage systems

Tigris Data met Perfloop as a pull request in its open source gateway, tigrisdata/tag. Ovais Tariq merged it and asked for access.

Tigris is now a customer. It has merged 48 of Perfloop's pull requests in its public repositories, 34 in tag and 14 in ocache, and wrote 4 fixes of its own for problems Perfloop reported.

52

merged changes in its public repositories, one square each

The people who merged them

36 maintainers pressed merge. The number is how many each one merged; a fix they wrote themselves for a problem Perfloop reported is named as their own.

  • Cuong Manh LeGo own fix
  • Eugene RizhkovElasticsearch 1
  • Sergei GeorgievRedis® 1
  • Arve KnudsenPrometheus® 3
  • Bartlomiej PlotkaPrometheus® 3
  • Micha ReiserRuff 1
  • Alexey MilovidovClickHouse 1
  • Daniel Stenbergcurl 2
  • Nick TobeyDolt 1
  • Aaron SonDolt 1
  • Fujii MasaoPostgreSQL® 1
  • cnderrauberLiveKit 1
  • Raja SubramanianLiveKit 1
  • Ruben FiszelWindmill own fix
  • Aleix Conchillo FlaquéPipecat 2 + own fix
  • Asdine El HrychyWeaviate 1
  • Yong TangCoreDNS 4
  • Yann ColletLZ4 1
  • Milos GajdosDistribution 6
  • Florian RapplAngleSharp 1
  • Laurent DemaillyFortio, grol 4
  • Mustafa AbdelrahmanSkipper 2
  • Sandor SzücsSkipper 1
  • Daniel LemireRoaringBitmap 8 + own fix
  • Guy A MolinariRoaringBitmap 1
  • Nikhil SinhaParseable 1
  • Keegan Carruthers-SmithZoekt 7
  • Seif LotfyAxiom FSST, Axiom HyperLogLog 3
  • Liz Fong-Jonespierrec/lz4 1
  • Achilleparquet-go, kway-go 13
  • Josh Stevensrindexer 1
  • Torbjörn Einarssonmp4ff 5 + 4 own fixes
  • Jitendra Adhikarigronx 1
  • Nuno Lopescpp2rust 8
  • AnoopDataFusion DuckLake 1
  • Ovais TariqTigris TAG, Tigris ocache 48 + 4 own fixes

Read each merged change

Newest first, with the person who merged it.

MergedChangeWhat was wrongResultMerged by
Nov 18Weaviate #13940A recoverable queue write failure could leave an HFresh posting marked pending with no task, so later searches did not schedule its merge.bug fixedAsdine El Hrychy
Nov 18CoreDNS #8774CoreDNS exited at startup when a valid zone file had an SOA serial of zero, and the auto plugin skipped such a zone.bug fixedYong Tang
Nov 18CoreDNS #8776A secondary whose zone had expired still answered zone transfers with the obsolete records.bug fixedYong Tang
Nov 18CoreDNS #8778With a hostname upstream, forward could refuse to start or configure an incomplete upstream list when the resolver truncated its UDP answer.bug fixedYong Tang
Nov 18parquet-go #685After two row seeks in a row, a file reader could return the wrong row or skip later pages.bug fixedAchille
Nov 17parquet-go #683A descending sort of an optional column put the nulls on the wrong side, so the rows disagreed with the sorting columns the file declares.bug fixedAchille
Nov 17mp4ff #721Streamed sample getters read every sample from the file on its own, even in a contiguous run.2.5× lowerTorbjörn Einarsson
Nov 17LiveKit #5213A receiver restart during a retransmission could pair old outgoing packet headers with new media.bug fixedRaja Subramanian
Nov 17mp4ff #720CBCS encryption succeeded with an IV different from the init segment's constant IV, and readers then recovered different media bytes.bug fixedTorbjörn Einarsson
Nov 16curl #23945A failed allocation while replacing a cookie could discard the old cookie and leave the cookie count one too high.bug fixedDaniel Stenberg
Nov 16cpp2rust #583Translated try_emplace calls skipped the side effects of their arguments when the key already existed.bug fixedNuno Lopes
Nov 16Zoekt #1250A failed blob read during git indexing could publish a partial or empty shard over the complete index, so files dropped out of search.bug fixedKeegan Carruthers-Smith
Nov 15parquet-go #640Schema.Reconstruct could fill a missing column with a previous call's values, or let a stale value satisfy a required field.bug fixedAchille
Nov 15parquet-go #632Packing many small Parquet files allocated a 1,024-value scratch buffer for every segment and column.1.6× lowerAchille
Nov 15parquet-go #638variant.Marshal took the pooled encoder path even for a single integer, a short string or an empty container.2.4× lowerAchille
Nov 15cpp2rust #616A translated C++ string left no terminator after a move, so size() wrapped and an unsafe substring length could underflow.bug fixedNuno Lopes
Nov 15CoreDNS #8756The cache cut fractional SERVFAIL lifetimes to whole seconds: a 500ms failure expired at once, and a 1.5s failure could be retried after one second.bug fixedYong Tang
Nov 14curl #23942Since curl 8.18.0, cookie-file imports could overwrite cookies added with CURLOPT_COOKIELIST before the first transfer, contrary to their documented priority.bug fixedDaniel Stenberg
Nov 14rindexer #474Address-filtered log requests that spanned several address chunks could rewind the historical cursor and dispatch the same logs again.bug fixedJosh Stevens
Nov 14mp4ff #706HEVC slice headers that reused the SPS's only short-term reference picture set or long-term picture returned the wrong reference pictures.bug fixedTorbjörn Einarsson
Nov 13mp4ff a72c55dBoxes decoded while streaming, such as emsg message data and avcC parameter sets, were overwritten by the next box read.Reported by Perfloop in Eyevinn/mp4ff#712.bug fixedTorbjörn Einarsson (their fix, Perfloop's report)
Nov 13mp4ff b308adcStreamed sample access gave a track the first track's trex defaults, such as the video duration 3000 instead of 1024 for audio samples.Reported by Perfloop in Eyevinn/mp4ff#711.bug fixedTorbjörn Einarsson (their fix, Perfloop's report)
Nov 13mp4ff a95a8f5SttsBox.GetTimeCode wrapped the elapsed time of a sample that started more than 2^32 ticks into a track.Reported by Perfloop in Eyevinn/mp4ff#708.bug fixedTorbjörn Einarsson (their fix, Perfloop's report)
Nov 13Dolt #12427GC admission could drop a cancellation signal, because the broadcast could land before the listener registered.bug fixedAaron Son
Nov 13Zoekt #1230Case-insensitive substring verification could lose matches that differ only within a Unicode fold class, such as sss and ssſ or σσσ and σσς.bug fixedKeegan Carruthers-Smith
Nov 12Zoekt #1227A failed shard rename during index publication could delete the prior index data or hide the error.bug fixedKeegan Carruthers-Smith
Nov 12RoaringBitmap #591On arm64, batch iteration over a bitmap container found each set bit with a population count, which has no scalar instruction there.Reported by Perfloop in RoaringBitmap/roaring#553.gain, not sizedDaniel Lemire (fix by Liz Fong-Jones, Perfloop's report)
Nov 12Distribution #5069A blob upload could answer 202 Accepted with an offset the stored file did not reach, so the next request got 416.bug fixedMilos Gajdos
Nov 12gronx #81A task admitted just before Stop or the cutoff left its result goroutine waiting forever.bug fixedJitendra Adhikari
Nov 12mp4ff #705HEVC slice headers whose reference picture set was predicted from another set failed to parse or returned the wrong reference pictures.bug fixedTorbjörn Einarsson
Nov 11pierrec/lz4 #297lz4c left an old file's tail after a shorter output and could write its output over its own input.bug fixedLiz Fong-Jones
Nov 11Distribution #4906Manifest uploads checked each referenced layer in storage one at a time.6.0× lowerMilos Gajdos
Nov 11Distribution #5072A pull-through cache sent the full blob again to clients that already held it, because its first ETag was unquoted.gain, not sizedMilos Gajdos
Nov 11Distribution #5071Each resumed upload chunk saved the same digest checkpoint to storage twice.gain, not sizedMilos Gajdos
Nov 11Elasticsearch #163290Async cross-cluster searches returned the wrong page of hits while a remote cluster was still running.bug fixedEugene Rizhkov
Nov 10Go f75d503A recover in a range-over-func loop body returned nil after the goroutine stack grew or shrank.Reported by Perfloop in golang/go#82930.bug fixedCuong Manh Le (their fix, Perfloop's report)
Nov 10Distribution #5070A cold read of a small blob reserved a 4 MiB buffer.754× lowerMilos Gajdos
Nov 9Parseable #1846Streamed queries decoded every batch's JSON into a value tree and encoded it again before sending it.4.5× lowerNikhil Sinha
Nov 8cpp2rust #614Pointers to fields of temporary objects outlived their owners in refcount translations, causing dangling-pointer panics.bug fixedNuno Lopes
Nov 7LZ4 #2066LZ4F_decompress reported zero bytes consumed and produced when it failed, even after it had written output.bug fixedYann Collet
Nov 4PostgreSQL® 6a146d2Freeing a WAL reader leaked separately allocated records still in its decode queue.bug fixedFujii Masao
Nov 3cpp2rust #584Translated getsockname on Unix sockets returned success with the caller's address buffer unchanged.bug fixedNuno Lopes
Nov 3mp4ff #690aac.NewADTSHeader accepted a payload of 8185 bytes and encoded a frame length of zero.Reported by Perfloop in Eyevinn/mp4ff#689.bug fixedTorbjörn Einarsson (their fix, Perfloop's report)
Nov 3mp4ff #686HEVC parameter-set extraction dropped the final set when an Annex-B stream ended before video, and appending to one returned set could overwrite another.bug fixedTorbjörn Einarsson
Nov 3Skipper #4396Warm HTTP cache reads copied each serialized response before decoding it.−20%Mustafa Abdelrahman
Nov 3Prometheus® #20180PromQL® quantiles multiplied equal samples by interpolation weights, so the smallest subnormal value came back as zero.bug fixedBartlomiej Plotka
Nov 3Prometheus® #20184Streamed Remote Read returned in-order head samples that its isolation snapshot should have hidden.bug fixedBartlomiej Plotka
Nov 2cpp2rust #581Unlabelled breaks inside nested goto and switch macros left a generated loop instead of their intended target.bug fixedNuno Lopes
Nov 2cpp2rust #580Translated repeated-character string append panicked on a checked borrow for strings whose data pointer was used.bug fixedNuno Lopes
Nov 2Prometheus® #19389PromQL® range functions built a fresh one-sample vector for every scalar argument, series and query step.2.1× lowerBartlomiej Plotka
Nov 1cpp2rust #577Translated substring calls evaluated the receiver and arguments more than once, changing the selected bytes and caller state.bug fixedNuno Lopes
Oct 29Zoekt #1202Chunk search reuses resolved lines when building ranges and scoring matches.−20%Keegan Carruthers-Smith
Oct 28Windmill #12087Every code job deep-copied the cached script's module map: path keys, module contents and lock strings.Reported by Perfloop in windmill-labs/windmill#12084.1.8× lowerRuben Fiszel (their fix, Perfloop's report)
Oct 28Redis® #15887After the IDMP expiry cron pruned a stream, CLUSTER SLOT-STATS kept counting the freed memory.bug fixedSergei Georgiev
Oct 24Axiom HyperLogLog #75Every dense estimate called a general math.Pow on every register.5.7× lowerSeif Lotfy
Oct 23AngleSharp #1388HTML serialization allocated pooled builders to rebuild strings even when no characters needed escaping.−23%Florian Rappl
Oct 22E2B #3713Killing a sandbox that its node had already dropped subtracted the sandbox's resources from the node's allocations a second time.bug fixedthe project
Oct 22Tigris TAG #274Cached multipart-completion replays were not bound to the access key that completed the upload.Reported by Perfloop in tigrisdata/tag#242.bug fixedOvais Tariq (their fix, Perfloop's report)
Oct 22Tigris TAG #269A truncated whole-object cache entry was served with the full Content-Length until its TTL ran out.Reported by Perfloop in tigrisdata/tag#249.bug fixedOvais Tariq (their fix, Perfloop's report)
Oct 22Tigris TAG #271In signing mode, TAG forwarded signed streaming uploads without a check of each chunk's signature.Reported by Perfloop in tigrisdata/tag#243.bug fixedOvais Tariq (their fix, Perfloop's report)
Oct 21Tigris TAG #248Transparent mode treated successful writes and non-read responses as permission to read private cached data.bug fixedOvais Tariq
Oct 21Tigris TAG #256Embedded storage closed while detached cache writers were still running during shutdown.bug fixedOvais Tariq
Oct 21Tigris TAG #219Broadcasting copied every upstream body chunk even when serving a single listener.2.2× lowerOvais Tariq
Oct 21Tigris TAG #238Bulk delete responses could finish before cache invalidation cleared stale racing refills.bug fixedOvais Tariq
Oct 21Tigris TAG #223Warm block-cached multi-block range requests ran serial block presence probes before reading payload data.13× lowerOvais Tariq
Oct 21Tigris TAG #266Transparent local auth rejected URL-only SigV4 presigned GET requests lacking an X-Amz-Content-Sha256 header.bug fixedOvais Tariq
Oct 21Tigris TAG #270A stalled cache read from a remote peer held a request until the client disconnected or the server's five-minute write timeout fired.Reported by Perfloop in tigrisdata/tag#222.bug fixedOvais Tariq (their fix, Perfloop's report)
Oct 21Tigris TAG #251Positive short cache range reads were accepted as complete hits before reaching advertised Content-Length.bug fixedOvais Tariq
Oct 21Tigris TAG #247Unreadable or malformed legacy tombstones allowed stale metadata and ETags to overwrite newer completions.bug fixedOvais Tariq
Oct 21Tigris TAG #234Sparse range requests probed every covering block in the cache even after exceeding fan-out limits.3.3× lowerOvais Tariq
Oct 21Tigris TAG #258The embedded cache closed during shutdown before admitted HTTP handlers finished synchronous cleanup.bug fixedOvais Tariq
Oct 21Tigris TAG #263Broadcaster terminal error loss published incomplete response bodies under complete-object cache metadata.bug fixedOvais Tariq
Oct 21Tigris TAG #245Incomplete upstream streamed responses finished framing around received prefixes instead of aborting.bug fixedOvais Tariq
Oct 21Tigris TAG #206Copying response headers lowercased every ordinary header key and allocated memory for each one.−5.5%Ovais Tariq
Oct 21Tigris TAG #259Creating the HTTP server inside Start allowed early Stop calls to miss the uninitialized server.bug fixedOvais Tariq
Oct 21Tigris TAG #265Every multi-block Range hit ran a serial presence preflight probe before serving cached data.2.2× lowerOvais Tariq
Oct 21Tigris TAG #239Parquet footer warming stopped when a metadata-only HEAD entry was already visible after multipart completion.to zeroOvais Tariq
Oct 21Tigris TAG #237Read-triggered Parquet footer scans lacked concurrency bounds, creating unrestricted background workers during burst tail reads.bug fixedOvais Tariq
Oct 21Tigris TAG #240Block fetches continued admitting tasks and creating state after parent context cancellation.bug fixedOvais Tariq
Oct 21Tigris TAG #267Signing mode rejected valid SDK SigV4 requests when metadata headers contained repeated ASCII spaces.bug fixedOvais Tariq
Oct 21Tigris TAG #257Tiered PUTs routed to the local engine acknowledged and served bodies with mismatched payload hashes.bug fixedOvais Tariq
Oct 21Tigris TAG #215Single-range cache hits relayed chunks through an intermediate pipe before writing to the response.2.2× lowerOvais Tariq
Oct 21Tigris TAG #264Failed whole-body cache writes left relay goroutines and populate reservations alive.bug fixedOvais Tariq
Oct 21Tigris TAG #253Delayed tiered background re-tier tasks could overwrite newer metadata from local writes.bug fixedOvais Tariq
Oct 21Tigris TAG #260Full-object GET requests hung when upstream transport failed before sending response headers.bug fixedOvais Tariq
Oct 21Tigris TAG #261Late GET requests that missed the subscription window could mix upstream response headers and bodies.bug fixedOvais Tariq
Oct 19grol #544Negative range bounds that underflowed caused panics instead of returning bounded slices.bug fixedLaurent Demailly
Oct 19grol #545Failed compound assignments stored arithmetic errors into identifiers, array elements, or map fields.bug fixedLaurent Demailly
Oct 19cpp2rust #544Translated map subscripting used equality instead of ordering equivalence, returning wrong values or replacing entries.bug fixedNuno Lopes
Oct 14Distribution #4921Canonical full-hash fallback validation executed a duplicate SHA-256 pass over stored blob bytes.1.9× lowerMilos Gajdos
Oct 13Dolt #12066Undrained keyless secondary-index scans left producer goroutines blocked on full channels.bug fixedNick Tobey
Oct 12Tigris TAG #252CopyObject requests with no-store directives still triggered post-copy cache warming and origin reads.15× lowerOvais Tariq
Oct 11Fortio #1167FastClient reused connections after truncating oversized responses, parsing remaining body bytes as subsequent responses.bug fixedLaurent Demailly
Oct 11Fortio #1168FastClient spun in empty-buffer read loops when chunks exceeded the response buffer.bug fixedLaurent Demailly
Oct 9DataFusion DuckLake #368Every row-ID scan filtered out deletes that did not exist.2.5× lowerAnoop
Oct 8Skipper #4326Default text access logs built and copied field maps even without supplemental fields.−15%Sandor Szücs
Oct 6E2B #3671A sandbox create kept waiting for a fresh network slot while a returned one sat ready.bug fixedthe project
Oct 6Skipper #4324The steady-state fade-in path copied every endpoint on each load-balanced request.1.5× lowerMustafa Abdelrahman
Sep 30LiveKit #4681Forwarding packets with common 4-byte dependency descriptors allocated memory for their headers.to zerocnderrauber
Sep 25Pipecat #5906RTVIObserver kept the ID of every frame it saw, audio frames included, for the life of a session.Reported by Perfloop in pipecat-ai/pipecat#5882.bug fixedAleix Conchillo Flaqué (their fix, Perfloop's report)
Sep 21Prometheus® #19717OTLP explicit histogram conversion repeatedly reconstructed metric suffix strings inside data-point and bucket loops.−18%Arve Knudsen
Sep 17Pipecat #5811AudioVolumeTracker rebuilt the entire rolling window buffer on every audio chunk.1.9× lowerAleix Conchillo Flaqué
Sep 16Pipecat #5805VADController broadcast redundant UserSpeakingFrame events downstream on every audio frame while speaking.−16%Aleix Conchillo Flaqué
Sep 16Axiom FSST #7Table deserialization rebuilt the default lookup state right before overwriting it from the stream.3.0× lowerSeif Lotfy
Sep 9Tigris ocache #253YCSB key preloading ran serially, ignoring the configured concurrency during the load phase.5.1× lowerOvais Tariq
Sep 9Tigris ocache #252Sub-second ring heartbeats triggered redundant CAS retries because ring timestamps use one-second precision.2.2× lowerOvais Tariq
Sep 9Tigris ocache #245Segment writes allocated an extra copy buffer for SectionReader inputs during recompaction.−31%Ovais Tariq
Sep 9Tigris ocache #220File lock lookups allocated a throwaway mutex on every lookup for an existing path.−10%Ovais Tariq
Sep 9Tigris ocache #273YCSB latency reporting copied every per-operation latency slice before sorting percentile values.1,015× lowerOvais Tariq
Sep 3Tigris TAG #198Background full-object cache warms routed through broadcaster listeners instead of streaming directly to cache writers.1.6× lowerOvais Tariq
Sep 3Tigris TAG #197Large cache hits staged data through an io.Pipe goroutine and extra copy.1.8× lowerOvais Tariq
Sep 3Tigris TAG #199Transparent forwarding deep-cloned every inbound header map and value slice on each uncached S3 operation.−15%Ovais Tariq
Aug 28Tigris ocache #241Local streamed Puts crossing the inline threshold bypassed FileManager's pooled copy buffer.8.9× lowerOvais Tariq
Aug 28Tigris ocache #244Inline key-value listings allocated and copied protobuf-decoded values a second time.2.0× lowerOvais Tariq
Aug 27RoaringBitmap #557Bitmap.ToArray decoded dense bitmap containers one set bit at a time.Reported by Perfloop in RoaringBitmap/roaring#556.1.8× lowerDaniel Lemire (their fix, Perfloop's report)
Aug 24RoaringBitmap #555Non-inplace array subtraction converted array containers into bitmap containers and scanned them.4.2× lowerDaniel Lemire
Aug 24Prometheus® #19503Sampled remote reads cloned and merged response labels even when no external labels were configured.−16%Arve Knudsen
Aug 23Tigris ocache #240Delete and overwrite cleanup decoded unused inline payload data when reading value messages.1.7× lowerOvais Tariq
Aug 22Tigris ocache #237Every topology response allocated and sorted a copy of normalized token slices.1.9× lowerOvais Tariq
Aug 22Tigris ocache #238Cleaner scans decoded full inline payloads and control messages for ordinary non-expiring rows.110× lowerOvais Tariq
Aug 20Tigris TAG #175Normal PUT forwarding cloned upstream response headers even when it did not retain them.to zeroOvais Tariq
Aug 20Tigris ocache #234Discarded YCSB reads materialized unused result slices in memory.1.6× lowerOvais Tariq
Aug 20Tigris TAG #173Cache-rejected PUT requests still ran the full write-through tee admission path and upstream metadata HEAD.25× lowerOvais Tariq
Aug 20Tigris TAG #172Canonical query construction formatted an intermediate string for each escaped byte.−17%Ovais Tariq
Aug 19RoaringBitmap #551Two-pointer array intersection loops executed repeated input bounds checks on every element read.−21%Daniel Lemire
Aug 18Tigris ocache #219File compaction and segment recompaction copied payloads with no shared throughput bound.bug fixedOvais Tariq
Aug 18Tigris TAG #164Signing endpoint URLs were serialized and reparsed on every request.−9%Ovais Tariq
Aug 17Tigris ocache #217Remote GetRangeStream copied every gRPC response chunk into an intermediate buffer before delivery.−19%Ovais Tariq
Aug 15Tigris ocache #210Gated access-time reads used Peek, evicting hot keys from the gate cache under capacity pressure.1.5× lowerOvais Tariq
Aug 14Zoekt #1127Code search walked every branch bitmap to reject shards it could rule out up front.24× lowerKeegan Carruthers-Smith
Aug 14kway-go #7Two-input MergeSliceFunc discarded comparator signs from failed lookaheads and gallop boundaries.1.5× lowerAchille
Aug 13kway-go #6Buffers allocated backing arrays that empty sources and sources ending at growth boundaries never used.−29%Achille
Aug 13ClickHouse #113899ClickHouse checked whether constant columns were sorted. Every insert, every row.gain, not sizedAlexey Milovidov
Aug 12kway-go #5Zero-copy passthrough batches repeatedly grew the merge aggregation buffer and allocated unused capacity.2.2× lowerAchille
Aug 12Prometheus® #19388The OTLP write handler built fresh scratch state for every request.−27%Arve Knudsen
Aug 11RoaringBitmap #542Lazy bitmap unions cloned and overwrote the receiver container while repeating slice bounds checks.−15%Daniel Lemire
Aug 11RoaringBitmap #539Parallel BSI query paths called run-optimization on empty worker-local result bitmaps across workers.14× lowerDaniel Lemire
Aug 11Tigris TAG #148Cacheable full-GET requests blocked handler completion until detached cache population finished.177× lowerOvais Tariq
Aug 7Zoekt #1110Unlimited sharded searches repeatedly sorted and truncated matches across every intermediate result chunk.26× lowerKeegan Carruthers-Smith
Jul 27Axiom FSST #5Candidate selection during training maintained a min-heap and a separate output list.−5.7%Seif Lotfy
Jul 24parquet-go #581Reconstruction of wide maps with repeated descendants allocated 347 times per operation.−19%Achille
Jul 24Ruff #27159Cached auto-import completions spent extra time in scope-member lookups.−14%Micha Reiser
Jul 23parquet-go #579Default-page INT64 bounds calculations lacked AVX-512VL acceleration.−27%Achille
Jul 22RoaringBitmap #535Interleaved source keys during bitmap union shifted metadata slices once for every inserted key.6.9× lowerDaniel Lemire
Jul 13RoaringBitmap #534Dense bitmap container materialization decoded set bits using popcount and isolate-lowest-bit operations.−30%Daniel Lemire
Jul 10RoaringBitmap #533Scattered IN-list queries in BSI BatchEqual materialized a large number of intermediate bitmaps.1.8× lowerDaniel Lemire
Jul 9Zoekt #1083Case-insensitive match verification decoded UTF-8 and performed Unicode lowercasing lookups for every byte.2.3× lowerKeegan Carruthers-Smith
Jul 7parquet-go #550variant.Marshal built an intermediate variant.Value tree before encoding it into bytes.4.2× lowerAchille
Jul 2RoaringBitmap #532BSI BatchEqual scanned columns with individual binary searches across roaring bitmap bitplanes.31× lowerGuy A Molinari
Jun 25parquet-go #547Single-column row comparisons incurred interface dispatch and copied two 24-byte Value structures per comparison.1.7× lowerAchille
Jun 21parquet-go #542Writing time columns boxed each value individually, allocating on the heap for every row.3.8× lowerAchille

Merge dates and the people who merged were read from GitHub on November 18, 2026.

See what it finds in your code.

sign up →

Or write to hello@perfloop.ai