v6-0001-Simplify-abbreviated-key-hashing-using-murmurhash.patch
text/x-patch
Filename: v6-0001-Simplify-abbreviated-key-hashing-using-murmurhash.patch
Type: text/x-patch
Part: 0
Patch
Same data as JSON:
GET /api/v1/attachments/:id/patch
the parsed metadata as JSON — format, series position, per-file stats; never the diff bytes.
API reference →
Format: format-patch
Series: patch v6-0001
Subject: Simplify abbreviated key hashing using murmurhash64()
| File | + | − |
|---|---|---|
| src/backend/utils/adt/bytea.c | 2 | 7 |
| src/backend/utils/adt/network.c | 1 | 7 |
| src/backend/utils/adt/numeric.c | 1 | 6 |
| src/backend/utils/adt/uuid.c | 1 | 7 |
| src/backend/utils/adt/varlena.c | 2 | 7 |
From 3312663274d2278a58fa3e24d8c157ff8d47b620 Mon Sep 17 00:00:00 2001
From: Aleksander Alekseev <aleksander@tigerdata.com>
Date: Tue, 13 Jan 2026 14:51:21 +0300
Subject: [PATCH v6 1/2] Simplify abbreviated key hashing using murmurhash64()
Now when all Datums are 64-bit values we can simplify the code by using
murmurhash64() in *_abbrev_convert() functions. Also replace hash_uint32()
with murmurhash32() in a few other places for consistency.
Author: Aleksander Alekseev <aleksander@tigerdata.com>
Suggested-by: John Naylor <johncnaylorls@gmail.com>
Reviewed-by: John Naylor <johncnaylorls@gmail.com>
Discussion: https://postgr.es/m/CAJ7c6TMPhDRQMmkUHPv8oOK97B1mR8NRS61DgjpdaZUPAwaeZQ%40mail.gmail.com
---
src/backend/utils/adt/bytea.c | 9 ++-------
src/backend/utils/adt/network.c | 8 +-------
src/backend/utils/adt/numeric.c | 7 +------
src/backend/utils/adt/uuid.c | 8 +-------
src/backend/utils/adt/varlena.c | 9 ++-------
5 files changed, 7 insertions(+), 34 deletions(-)
diff --git a/src/backend/utils/adt/bytea.c b/src/backend/utils/adt/bytea.c
index a1f5d4e572b..71dc591438d 100644
--- a/src/backend/utils/adt/bytea.c
+++ b/src/backend/utils/adt/bytea.c
@@ -1110,17 +1110,12 @@ bytea_abbrev_convert(Datum original, SortSupport ssup)
Min(len, PG_CACHE_LINE_SIZE)));
if (len > PG_CACHE_LINE_SIZE)
- hash ^= DatumGetUInt32(hash_uint32((uint32) len));
+ hash ^= murmurhash32((uint32) len);
addHyperLogLog(&bss->full_card, hash);
/* Hash abbreviated key */
- {
- uint32 tmp;
-
- tmp = DatumGetUInt32(res) ^ (uint32) (DatumGetUInt64(res) >> 32);
- hash = DatumGetUInt32(hash_uint32(tmp));
- }
+ hash = (uint32) murmurhash64(DatumGetUInt64(res));
addHyperLogLog(&bss->abbr_card, hash);
diff --git a/src/backend/utils/adt/network.c b/src/backend/utils/adt/network.c
index f4bf9c3b532..cc2bd3c7344 100644
--- a/src/backend/utils/adt/network.c
+++ b/src/backend/utils/adt/network.c
@@ -738,13 +738,7 @@ network_abbrev_convert(Datum original, SortSupport ssup)
/* Hash abbreviated key */
if (uss->estimating)
- {
- uint32 tmp;
-
- tmp = DatumGetUInt32(res) ^ (uint32) (DatumGetUInt64(res) >> 32);
-
- addHyperLogLog(&uss->abbr_card, DatumGetUInt32(hash_uint32(tmp)));
- }
+ addHyperLogLog(&uss->abbr_card, (uint32) murmurhash64(DatumGetUInt64(res)));
return res;
}
diff --git a/src/backend/utils/adt/numeric.c b/src/backend/utils/adt/numeric.c
index c9717faea26..3712cd0e9df 100644
--- a/src/backend/utils/adt/numeric.c
+++ b/src/backend/utils/adt/numeric.c
@@ -2404,12 +2404,7 @@ numeric_abbrev_convert_var(const NumericVar *var, NumericSortSupport *nss)
result = -result;
if (nss->estimating)
- {
- uint32 tmp = ((uint32) result
- ^ (uint32) ((uint64) result >> 32));
-
- addHyperLogLog(&nss->abbr_card, DatumGetUInt32(hash_uint32(tmp)));
- }
+ addHyperLogLog(&nss->abbr_card, (uint32) murmurhash64(result));
return NumericAbbrevGetDatum(result);
}
diff --git a/src/backend/utils/adt/uuid.c b/src/backend/utils/adt/uuid.c
index 2d33ba2640b..7b55528cbb7 100644
--- a/src/backend/utils/adt/uuid.c
+++ b/src/backend/utils/adt/uuid.c
@@ -413,13 +413,7 @@ uuid_abbrev_convert(Datum original, SortSupport ssup)
uss->input_count += 1;
if (uss->estimating)
- {
- uint32 tmp;
-
- tmp = DatumGetUInt32(res) ^ (uint32) (DatumGetUInt64(res) >> 32);
-
- addHyperLogLog(&uss->abbr_card, DatumGetUInt32(hash_uint32(tmp)));
- }
+ addHyperLogLog(&uss->abbr_card, (uint32) murmurhash64(DatumGetUInt64(res)));
/*
* Byteswap on little-endian machines.
diff --git a/src/backend/utils/adt/varlena.c b/src/backend/utils/adt/varlena.c
index 0c6d3ba4d22..18ef69f7e0c 100644
--- a/src/backend/utils/adt/varlena.c
+++ b/src/backend/utils/adt/varlena.c
@@ -2158,17 +2158,12 @@ varstr_abbrev_convert(Datum original, SortSupport ssup)
Min(len, PG_CACHE_LINE_SIZE)));
if (len > PG_CACHE_LINE_SIZE)
- hash ^= DatumGetUInt32(hash_uint32((uint32) len));
+ hash ^= murmurhash32((uint32) len);
addHyperLogLog(&sss->full_card, hash);
/* Hash abbreviated key */
- {
- uint32 tmp;
-
- tmp = DatumGetUInt32(res) ^ (uint32) (DatumGetUInt64(res) >> 32);
- hash = DatumGetUInt32(hash_uint32(tmp));
- }
+ hash = (uint32) murmurhash64(DatumGetUInt64(res));
addHyperLogLog(&sss->abbr_card, hash);
--
2.43.0