db/pgbackrest
1
0
Fork 0
mirror of https://github.com/pgbackrest/pgbackrest.git synced 2025-07-05 00:28:52 +02:00
Code Issues Releases Activity

Add support for alternate compile-time page sizes.

Browse Source 
Alternate pages sizes can be selected at compile-time, .e.g. 4096. While compile-time settings are generally not well tested by core, some established forks such as Greenplum use them.
This commit is contained in:
Viktor Kurilko
2023-12-14 19:28:52 +03:00
committed by GitHub
parent d205a61949
commit 89d5278b74
17 changed files with 449 additions and 353 deletions
Download Patch File Download Diff File Expand all files Collapse all files

143
src/postgres/interface/page.c
View File

@ -1,5 +1,7 @@
/***********************************************************************************************************************************
PostgreSQL Page Interface
PostgreSQL Page Checksum
Adapted from PostgreSQL src/include/storage/checksum_impl.h.
***********************************************************************************************************************************/
#include "build.auto.h"
@ -8,13 +10,144 @@ PostgreSQL Page Interface
#include "postgres/interface/static.vendor.h"
/***********************************************************************************************************************************
Include the page checksum code
Page checksum calculation
***********************************************************************************************************************************/
#include "postgres/interface/pageChecksum.vendor.c.inc"
// Number of checksums to calculate in parallel
#define PARALLEL_SUM 32
// Prime multiplier of FNV-1a hash
#define FNV_PRIME 16777619
// Calculate one round of the checksum
#define CHECKSUM_ROUND(checksum, value) \
do \
{ \
const uint32_t tmp = (checksum) ^ (value); \
checksum = tmp * FNV_PRIME ^ (tmp >> 17); \
} while (0)
// Main calculation loop
#define CHECKSUM_CASE(pageSize) \
case pageSize: \
for (uint32_t i = 0; i < (uint32) (pageSize / (sizeof(uint32) * PARALLEL_SUM)); i++) \
for (uint32_t j = 0; j < PARALLEL_SUM; j++) \
CHECKSUM_ROUND(sums[j], ((PgPageChecksum##pageSize *)page)->data[i][j]); \
\
break;
/***********************************************************************************************************************************
Define unions that will make the code valid under strict aliasing for each page size
***********************************************************************************************************************************/
#define CHECKSUM_UNION(pageSize) \
typedef union \
{ \
PageHeaderData phdr; \
uint32_t data[pageSize / (sizeof(uint32_t) * PARALLEL_SUM)][PARALLEL_SUM]; \
} PgPageChecksum##pageSize;
CHECKSUM_UNION(pgPageSize1);
CHECKSUM_UNION(pgPageSize2);
CHECKSUM_UNION(pgPageSize4);
CHECKSUM_UNION(pgPageSize8);
CHECKSUM_UNION(pgPageSize16);
CHECKSUM_UNION(pgPageSize32);
/**********************************************************************************************************************************/
FN_EXTERN uint16_t
pgPageChecksum(unsigned char *page, uint32_t blockNo)
pgPageChecksum(unsigned char *const page, const uint32_t blockNo, const PgPageSize pageSize)
{
return pg_checksum_page((char *)page, blockNo);
FUNCTION_TEST_BEGIN();
FUNCTION_TEST_PARAM_P(UCHARDATA, page);
FUNCTION_TEST_PARAM(UINT, blockNo);
FUNCTION_TEST_PARAM(ENUM, pageSize);
FUNCTION_TEST_END();
// Save pd_checksum and temporarily set it to zero, so that the checksum calculation isn't affected by the old checksum stored
// on the page. Restore it after, because actually updating the checksum is NOT part of the API of this function.
const uint16_t checksumPrior = ((PageHeaderData *)page)->pd_checksum;
((PageHeaderData *)page)->pd_checksum = 0;
// Initialize partial checksums to their corresponding offsets
uint32_t sums[PARALLEL_SUM] =
{
0x5b1f36e9, 0xb8525960, 0x02ab50aa, 0x1de66d2a, 0x79ff467a, 0x9bb9f8a3, 0x217e7cd2, 0x83e13d2c,
0xf8d4474f, 0xe39eb970, 0x42c6ae16, 0x993216fa, 0x7b093b5d, 0x98daff3c, 0xf718902a, 0x0b1c9cdb,
0xe58f764b, 0x187636bc, 0x5d7b3bb1, 0xe73de7de, 0x92bec979, 0xcca6c0b2, 0x304a0979, 0x85aa43d4,
0x783125bb, 0x6ca8eaa2, 0xe407eac6, 0x4b5cfc3e, 0x9fbf8c76, 0x15ca20be, 0xf2ca9fd3, 0x959bd756,
};
// Main checksum calculation
switch (pageSize)
{
CHECKSUM_CASE(pgPageSize8); // Default page size should be checked first
CHECKSUM_CASE(pgPageSize1);
CHECKSUM_CASE(pgPageSize2);
CHECKSUM_CASE(pgPageSize4);
CHECKSUM_CASE(pgPageSize16);
CHECKSUM_CASE(pgPageSize32);
default:
pgPageSizeCheck(pageSize);
}
// Add in two rounds of zeroes for additional mixing
for (uint32_t i = 0; i < 2; i++)
for (uint32_t j = 0; j < PARALLEL_SUM; j++)
CHECKSUM_ROUND(sums[j], 0);
// Xor fold partial checksums together
uint32 result = 0;
for (uint32_t i = 0; i < PARALLEL_SUM; i++)
result ^= sums[i];
// Restore prior checksum
((PageHeaderData *)page)->pd_checksum = checksumPrior;
// Mix in the block number to detect transposed pages
result ^= blockNo;
// Reduce to a uint16 (to fit in the pd_checksum field) with an offset of one. That avoids checksums of zero, which seems like a
// good idea.
FUNCTION_TEST_RETURN(UINT16, (uint16_t)((result % 65535) + 1));
}
/**********************************************************************************************************************************/
FN_EXTERN bool
pgPageSizeValid(const PgPageSize pageSize)
{
FUNCTION_TEST_BEGIN();
FUNCTION_TEST_PARAM(ENUM, pageSize);
FUNCTION_TEST_END();
switch (pageSize)
{
case pgPageSize1:
case pgPageSize2:
case pgPageSize4:
case pgPageSize8:
case pgPageSize16:
case pgPageSize32:
FUNCTION_TEST_RETURN(BOOL, true);
}
FUNCTION_TEST_RETURN(BOOL, false);
}
/**********************************************************************************************************************************/
FN_EXTERN void
pgPageSizeCheck(const PgPageSize pageSize)
{
FUNCTION_TEST_BEGIN();
FUNCTION_TEST_PARAM(ENUM, pageSize);
FUNCTION_TEST_END();
if (!pgPageSizeValid(pageSize))
{
THROW_FMT(
FormatError, "page size is %u but only %i, %i, %i, %i, %i, and %i are supported", pageSize, pgPageSize1, pgPageSize2,
pgPageSize4, pgPageSize8, pgPageSize16, pgPageSize32);
}
FUNCTION_TEST_RETURN_VOID();
}

227
src/postgres/interface/pageChecksum.vendor.c.inc
View File

@ -1,227 +0,0 @@
/***********************************************************************************************************************************
PostgreSQL Page Checksum Algorithm
For each supported release of PostgreSQL check the code in this file to see if it has changed. The easiest way to do this is to
copy and paste in place and check git to see if there are any diffs. Tabs should be copied as is to make this process easy even
though the pgBackRest project does not use tabs elsewhere.
Since the checksum implementation and page format do not (yet) change between versions this code should be copied verbatim from
src/include/storage/checksum_impl.h for each new release. Only the newest released version of the code should be used.
Modifications need to be made after copying:
1) Remove `#include "storage/bufpage.h"`.
2) Make pg_checksum_page() static.
3) Remove Assert(!PageIsNew(&cpage->phdr)).
***********************************************************************************************************************************/
/*-------------------------------------------------------------------------
*
* checksum_impl.h
* Checksum implementation for data pages.
*
* This file exists for the benefit of external programs that may wish to
* check Postgres page checksums. They can #include this to get the code
* referenced by storage/checksum.h. (Note: you may need to redefine
* Assert() as empty to compile this successfully externally.)
*
* Portions Copyright (c) 1996-2023, PostgreSQL Global Development Group
* Portions Copyright (c) 1994, Regents of the University of California
*
* src/include/storage/checksum_impl.h
*
*-------------------------------------------------------------------------
*/
/*
* The algorithm used to checksum pages is chosen for very fast calculation.
* Workloads where the database working set fits into OS file cache but not
* into shared buffers can read in pages at a very fast pace and the checksum
* algorithm itself can become the largest bottleneck.
*
* The checksum algorithm itself is based on the FNV-1a hash (FNV is shorthand
* for Fowler/Noll/Vo). The primitive of a plain FNV-1a hash folds in data 1
* byte at a time according to the formula:
*
* hash = (hash ^ value) * FNV_PRIME
*
* FNV-1a algorithm is described at http://www.isthe.com/chongo/tech/comp/fnv/
*
* PostgreSQL doesn't use FNV-1a hash directly because it has bad mixing of
* high bits - high order bits in input data only affect high order bits in
* output data. To resolve this we xor in the value prior to multiplication
* shifted right by 17 bits. The number 17 was chosen because it doesn't
* have common denominator with set bit positions in FNV_PRIME and empirically
* provides the fastest mixing for high order bits of final iterations quickly
* avalanche into lower positions. For performance reasons we choose to combine
* 4 bytes at a time. The actual hash formula used as the basis is:
*
* hash = (hash ^ value) * FNV_PRIME ^ ((hash ^ value) >> 17)
*
* The main bottleneck in this calculation is the multiplication latency. To
* hide the latency and to make use of SIMD parallelism multiple hash values
* are calculated in parallel. The page is treated as a 32 column two
* dimensional array of 32 bit values. Each column is aggregated separately
* into a partial checksum. Each partial checksum uses a different initial
* value (offset basis in FNV terminology). The initial values actually used
* were chosen randomly, as the values themselves don't matter as much as that
* they are different and don't match anything in real data. After initializing
* partial checksums each value in the column is aggregated according to the
* above formula. Finally two more iterations of the formula are performed with
* value 0 to mix the bits of the last value added.
*
* The partial checksums are then folded together using xor to form a single
* 32-bit checksum. The caller can safely reduce the value to 16 bits
* using modulo 2^16-1. That will cause a very slight bias towards lower
* values but this is not significant for the performance of the
* checksum.
*
* The algorithm choice was based on what instructions are available in SIMD
* instruction sets. This meant that a fast and good algorithm needed to use
* multiplication as the main mixing operator. The simplest multiplication
* based checksum primitive is the one used by FNV. The prime used is chosen
* for good dispersion of values. It has no known simple patterns that result
* in collisions. Test of 5-bit differentials of the primitive over 64bit keys
* reveals no differentials with 3 or more values out of 100000 random keys
* colliding. Avalanche test shows that only high order bits of the last word
* have a bias. Tests of 1-4 uncorrelated bit errors, stray 0 and 0xFF bytes,
* overwriting page from random position to end with 0 bytes, and overwriting
* random segments of page with 0x00, 0xFF and random data all show optimal
* 2e-16 false positive rate within margin of error.
*
* Vectorization of the algorithm requires 32bit x 32bit -> 32bit integer
* multiplication instruction. As of 2013 the corresponding instruction is
* available on x86 SSE4.1 extensions (pmulld) and ARM NEON (vmul.i32).
* Vectorization requires a compiler to do the vectorization for us. For recent
* GCC versions the flags -msse4.1 -funroll-loops -ftree-vectorize are enough
* to achieve vectorization.
*
* The optimal amount of parallelism to use depends on CPU specific instruction
* latency, SIMD instruction width, throughput and the amount of registers
* available to hold intermediate state. Generally, more parallelism is better
* up to the point that state doesn't fit in registers and extra load-store
* instructions are needed to swap values in/out. The number chosen is a fixed
* part of the algorithm because changing the parallelism changes the checksum
* result.
*
* The parallelism number 32 was chosen based on the fact that it is the
* largest state that fits into architecturally visible x86 SSE registers while
* leaving some free registers for intermediate values. For future processors
* with 256bit vector registers this will leave some performance on the table.
* When vectorization is not available it might be beneficial to restructure
* the computation to calculate a subset of the columns at a time and perform
* multiple passes to avoid register spilling. This optimization opportunity
* is not used. Current coding also assumes that the compiler has the ability
* to unroll the inner loop to avoid loop overhead and minimize register
* spilling. For less sophisticated compilers it might be beneficial to
* manually unroll the inner loop.
*/
/* number of checksums to calculate in parallel */
#define N_SUMS 32
/* prime multiplier of FNV-1a hash */
#define FNV_PRIME 16777619
/* Use a union so that this code is valid under strict aliasing */
typedef union
{
PageHeaderData phdr;
uint32 data[BLCKSZ / (sizeof(uint32) * N_SUMS)][N_SUMS];
} PGChecksummablePage;
/*
* Base offsets to initialize each of the parallel FNV hashes into a
* different initial state.
*/
static const uint32 checksumBaseOffsets[N_SUMS] = {
0x5B1F36E9, 0xB8525960, 0x02AB50AA, 0x1DE66D2A,
0x79FF467A, 0x9BB9F8A3, 0x217E7CD2, 0x83E13D2C,
0xF8D4474F, 0xE39EB970, 0x42C6AE16, 0x993216FA,
0x7B093B5D, 0x98DAFF3C, 0xF718902A, 0x0B1C9CDB,
0xE58F764B, 0x187636BC, 0x5D7B3BB1, 0xE73DE7DE,
0x92BEC979, 0xCCA6C0B2, 0x304A0979, 0x85AA43D4,
0x783125BB, 0x6CA8EAA2, 0xE407EAC6, 0x4B5CFC3E,
0x9FBF8C76, 0x15CA20BE, 0xF2CA9FD3, 0x959BD756
};
/*
* Calculate one round of the checksum.
*/
#define CHECKSUM_COMP(checksum, value) \
do { \
uint32 __tmp = (checksum) ^ (value); \
(checksum) = __tmp * FNV_PRIME ^ (__tmp >> 17); \
} while (0)
/*
* Block checksum algorithm. The page must be adequately aligned
* (at least on 4-byte boundary).
*/
static uint32
pg_checksum_block(const PGChecksummablePage *page)
{
uint32 sums[N_SUMS];
uint32 result = 0;
uint32 i,
j;
/* ensure that the size is compatible with the algorithm */
Assert(sizeof(PGChecksummablePage) == BLCKSZ);
/* initialize partial checksums to their corresponding offsets */
memcpy(sums, checksumBaseOffsets, sizeof(checksumBaseOffsets));
/* main checksum calculation */
for (i = 0; i < (uint32) (BLCKSZ / (sizeof(uint32) * N_SUMS)); i++)
for (j = 0; j < N_SUMS; j++)
CHECKSUM_COMP(sums[j], page->data[i][j]);
/* finally add in two rounds of zeroes for additional mixing */
for (i = 0; i < 2; i++)
for (j = 0; j < N_SUMS; j++)
CHECKSUM_COMP(sums[j], 0);
/* xor fold partial checksums together */
for (i = 0; i < N_SUMS; i++)
result ^= sums[i];
return result;
}
/*
* Compute the checksum for a Postgres page.
*
* The page must be adequately aligned (at least on a 4-byte boundary).
* Beware also that the checksum field of the page is transiently zeroed.
*
* The checksum includes the block number (to detect the case where a page is
* somehow moved to a different location), the page header (excluding the
* checksum itself), and the page data.
*/
static uint16
pg_checksum_page(char *page, BlockNumber blkno)
{
PGChecksummablePage *cpage = (PGChecksummablePage *) page;
uint16 save_checksum;
uint32 checksum;
/*
* Save pd_checksum and temporarily set it to zero, so that the checksum
* calculation isn't affected by the old checksum stored on the page.
* Restore it after, because actually updating the checksum is NOT part of
* the API of this function.
*/
save_checksum = cpage->phdr.pd_checksum;
cpage->phdr.pd_checksum = 0;
checksum = pg_checksum_block(cpage);
cpage->phdr.pd_checksum = save_checksum;
/* Mix in the block number to detect transposed pages */
checksum ^= blkno;
/*
* Reduce to a uint16 (to fit in the pd_checksum field) with an offset of
* one. That avoids checksums of zero, which seems like a good idea.
*/
return (uint16) ((checksum % 65535) + 1);
}

10
src/postgres/interface/static.vendor.h
View File

@ -23,16 +23,6 @@ all versions of PostgreSQL supported by pgBackRest.
#include "common/assert.h"
#include "postgres/interface.h"
/***********************************************************************************************************************************
Define Assert() as ASSERT()
***********************************************************************************************************************************/
#define Assert(condition) ASSERT(condition)
/***********************************************************************************************************************************
Define BLCKSZ as PG_PAGE_SIZE_DEFAULT
***********************************************************************************************************************************/
#define BLCKSZ PG_PAGE_SIZE_DEFAULT
/***********************************************************************************************************************************
Types from src/include/c.h
***********************************************************************************************************************************/