From 2389ee8dd800b7e95eb6ee8238e67db88c60d996 Mon Sep 17 00:00:00 2001 From: Michael Paquier Date: Wed, 16 Mar 2022 11:20:57 +0900 Subject: [PATCH] pageinspect: Fix handling of page sizes and AM types This commit fixes a set of issues related to the use of the SQL functions in this module when the caller is able to pass down raw page data as input argument: - The page size check was fuzzy in a couple of places, sometimes looking after only a sub-range, but what we are looking for is an exact match on BLCKSZ. After considering a few options here, I have settled down to do a generalization of get_page_from_raw(). Most of the SQL functions already used that, and this is not strictly required if not accessing an 8-byte-wide value from a raw page, but this feels safer in the long run for alignment-picky environment, particularly if a code path begins to access such values. This also reduces the number of strings that need to be translated. - The BRIN function brin_page_items() uses a Relation but it did not check the access method of the opened index, potentially leading to crashes. All the other functions in need of a Relation already did that. - Some code paths could fail on elog(), but we should to use ereport() for failures that can be triggered by the user. Tests are added to stress all the cases that are fixed as of this commit, with some junk raw pages (\set VERBOSITY ensures that this works across all page sizes) and unexpected index types when functions open relations. Author: Michael Paquier, Justin Prysby Discussion: https://postgr.es/m/20220218030020.GA1137@telsasoft.com Backpatch-through: 10 --- contrib/pageinspect/brinfuncs.c | 36 ++++++++++---------------- contrib/pageinspect/btreefuncs.c | 28 ++++++++++---------- contrib/pageinspect/expected/brin.out | 4 +++ contrib/pageinspect/expected/btree.out | 15 +++++++++++ contrib/pageinspect/expected/gin.out | 11 ++++++++ contrib/pageinspect/expected/hash.out | 17 ++++++++++++ contrib/pageinspect/expected/page.out | 11 ++++++++ contrib/pageinspect/fsmfuncs.c | 4 ++- contrib/pageinspect/hashfuncs.c | 6 +++-- contrib/pageinspect/rawpage.c | 29 +++------------------ contrib/pageinspect/sql/brin.sql | 4 +++ contrib/pageinspect/sql/btree.sql | 13 ++++++++++ contrib/pageinspect/sql/gin.sql | 9 +++++++ contrib/pageinspect/sql/hash.sql | 13 ++++++++++ contrib/pageinspect/sql/page.sql | 9 +++++++ 15 files changed, 142 insertions(+), 67 deletions(-) diff --git a/contrib/pageinspect/brinfuncs.c b/contrib/pageinspect/brinfuncs.c index f4f0dea860f..fa512aec936 100644 --- a/contrib/pageinspect/brinfuncs.c +++ b/contrib/pageinspect/brinfuncs.c @@ -18,6 +18,7 @@ #include "access/brin_revmap.h" #include "access/brin_tuple.h" #include "catalog/index.h" +#include "catalog/pg_am_d.h" #include "catalog/pg_type.h" #include "funcapi.h" #include "lib/stringinfo.h" @@ -33,6 +34,8 @@ PG_FUNCTION_INFO_V1(brin_page_items); PG_FUNCTION_INFO_V1(brin_metapage_info); PG_FUNCTION_INFO_V1(brin_revmap_data); +#define IS_BRIN(r) ((r)->rd_rel->relam == BRIN_AM_OID) + typedef struct brin_column_state { int nstored; @@ -47,8 +50,7 @@ Datum brin_page_type(PG_FUNCTION_ARGS) { bytea *raw_page = PG_GETARG_BYTEA_P(0); - Page page = VARDATA(raw_page); - int raw_page_size; + Page page; char *type; if (!superuser()) @@ -56,14 +58,7 @@ brin_page_type(PG_FUNCTION_ARGS) (errcode(ERRCODE_INSUFFICIENT_PRIVILEGE), (errmsg("must be superuser to use raw page functions")))); - raw_page_size = VARSIZE(raw_page) - VARHDRSZ; - - if (raw_page_size != BLCKSZ) - ereport(ERROR, - (errcode(ERRCODE_INVALID_PARAMETER_VALUE), - errmsg("input page too small"), - errdetail("Expected size %d, got %d", - BLCKSZ, raw_page_size))); + page = get_page_from_raw(raw_page); switch (BrinPageType(page)) { @@ -91,19 +86,7 @@ brin_page_type(PG_FUNCTION_ARGS) static Page verify_brin_page(bytea *raw_page, uint16 type, const char *strtype) { - Page page; - int raw_page_size; - - raw_page_size = VARSIZE(raw_page) - VARHDRSZ; - - if (raw_page_size != BLCKSZ) - ereport(ERROR, - (errcode(ERRCODE_INVALID_PARAMETER_VALUE), - errmsg("input page too small"), - errdetail("Expected size %d, got %d", - BLCKSZ, raw_page_size))); - - page = VARDATA(raw_page); + Page page = get_page_from_raw(raw_page); /* verify the special space says this page is what we want */ if (BrinPageType(page) != type) @@ -171,6 +154,13 @@ brin_page_items(PG_FUNCTION_ARGS) MemoryContextSwitchTo(oldcontext); indexRel = index_open(indexRelid, AccessShareLock); + + if (!IS_BRIN(indexRel)) + ereport(ERROR, + (errcode(ERRCODE_WRONG_OBJECT_TYPE), + errmsg("\"%s\" is not a %s index", + RelationGetRelationName(indexRel), "BRIN"))); + bdesc = brin_build_desc(indexRel); /* minimally verify the page we got */ diff --git a/contrib/pageinspect/btreefuncs.c b/contrib/pageinspect/btreefuncs.c index 439d706edee..33d809f4f69 100644 --- a/contrib/pageinspect/btreefuncs.c +++ b/contrib/pageinspect/btreefuncs.c @@ -181,8 +181,10 @@ bt_page_stats(PG_FUNCTION_ARGS) rel = relation_openrv(relrv, AccessShareLock); if (!IS_INDEX(rel) || !IS_BTREE(rel)) - elog(ERROR, "relation \"%s\" is not a btree index", - RelationGetRelationName(rel)); + ereport(ERROR, + (errcode(ERRCODE_WRONG_OBJECT_TYPE), + errmsg("\"%s\" is not a %s index", + RelationGetRelationName(rel), "btree"))); /* * Reject attempts to read non-local temporary relations; we would be @@ -335,8 +337,10 @@ bt_page_items(PG_FUNCTION_ARGS) rel = relation_openrv(relrv, AccessShareLock); if (!IS_INDEX(rel) || !IS_BTREE(rel)) - elog(ERROR, "relation \"%s\" is not a btree index", - RelationGetRelationName(rel)); + ereport(ERROR, + (errcode(ERRCODE_WRONG_OBJECT_TYPE), + errmsg("\"%s\" is not a %s index", + RelationGetRelationName(rel), "btree"))); /* * Reject attempts to read non-local temporary relations; we would be @@ -424,7 +428,6 @@ bt_page_items_bytea(PG_FUNCTION_ARGS) Datum result; FuncCallContext *fctx; struct user_args *uargs; - int raw_page_size; if (!superuser()) ereport(ERROR, @@ -437,19 +440,12 @@ bt_page_items_bytea(PG_FUNCTION_ARGS) MemoryContext mctx; TupleDesc tupleDesc; - raw_page_size = VARSIZE(raw_page) - VARHDRSZ; - - if (raw_page_size < SizeOfPageHeaderData) - ereport(ERROR, - (errcode(ERRCODE_INVALID_PARAMETER_VALUE), - errmsg("input page too small (%d bytes)", raw_page_size))); - fctx = SRF_FIRSTCALL_INIT(); mctx = MemoryContextSwitchTo(fctx->multi_call_memory_ctx); uargs = palloc(sizeof(struct user_args)); - uargs->page = VARDATA(raw_page); + uargs->page = get_page_from_raw(raw_page); uargs->offset = FirstOffsetNumber; @@ -525,8 +521,10 @@ bt_metap(PG_FUNCTION_ARGS) rel = relation_openrv(relrv, AccessShareLock); if (!IS_INDEX(rel) || !IS_BTREE(rel)) - elog(ERROR, "relation \"%s\" is not a btree index", - RelationGetRelationName(rel)); + ereport(ERROR, + (errcode(ERRCODE_WRONG_OBJECT_TYPE), + errmsg("\"%s\" is not a %s index", + RelationGetRelationName(rel), "btree"))); /* * Reject attempts to read non-local temporary relations; we would be diff --git a/contrib/pageinspect/expected/brin.out b/contrib/pageinspect/expected/brin.out index 71eb190380c..10cd36c1778 100644 --- a/contrib/pageinspect/expected/brin.out +++ b/contrib/pageinspect/expected/brin.out @@ -48,4 +48,8 @@ SELECT * FROM brin_page_items(get_raw_page('test1_a_idx', 2), 'test1_a_idx') 1 | 0 | 1 | f | f | f | {1 .. 1} (1 row) +-- Failure for non-BRIN index. +CREATE INDEX test1_a_btree ON test1 (a); +SELECT brin_page_items(get_raw_page('test1_a_btree', 0), 'test1_a_btree'); +ERROR: "test1_a_btree" is not a BRIN index DROP TABLE test1; diff --git a/contrib/pageinspect/expected/btree.out b/contrib/pageinspect/expected/btree.out index 2aaa4df53b1..da86c12846b 100644 --- a/contrib/pageinspect/expected/btree.out +++ b/contrib/pageinspect/expected/btree.out @@ -57,4 +57,19 @@ data | 01 00 00 00 00 00 00 01 SELECT * FROM bt_page_items(get_raw_page('test1_a_idx', 2)); ERROR: block number 2 is out of range for relation "test1_a_idx" +-- Failure when using a non-btree index. +CREATE INDEX test1_a_hash ON test1 USING hash(a); +SELECT bt_metap('test1_a_hash'); +ERROR: "test1_a_hash" is not a btree index +SELECT bt_page_stats('test1_a_hash', 0); +ERROR: "test1_a_hash" is not a btree index +SELECT bt_page_items('test1_a_hash', 0); +ERROR: "test1_a_hash" is not a btree index +-- Failure with incorrect page size +-- Suppress the DETAIL message, to allow the tests to work across various +-- page sizes. +\set VERBOSITY terse +SELECT bt_page_items('aaa'::bytea); +ERROR: invalid page size +\set VERBOSITY default DROP TABLE test1; diff --git a/contrib/pageinspect/expected/gin.out b/contrib/pageinspect/expected/gin.out index 82f63b23b19..c0f75950987 100644 --- a/contrib/pageinspect/expected/gin.out +++ b/contrib/pageinspect/expected/gin.out @@ -35,3 +35,14 @@ FROM gin_leafpage_items(get_raw_page('test1_y_idx', -[ RECORD 1 ] ?column? | t +-- Failure with incorrect page size +-- Suppress the DETAIL message, to allow the tests to work across various +-- page sizes. +\set VERBOSITY terse +SELECT gin_leafpage_items('aaa'::bytea); +ERROR: invalid page size +SELECT gin_metapage_info('bbb'::bytea); +ERROR: invalid page size +SELECT gin_page_opaque_info('ccc'::bytea); +ERROR: invalid page size +\set VERBOSITY default diff --git a/contrib/pageinspect/expected/hash.out b/contrib/pageinspect/expected/hash.out index 75d7bcfad5f..12369ddfa29 100644 --- a/contrib/pageinspect/expected/hash.out +++ b/contrib/pageinspect/expected/hash.out @@ -159,4 +159,21 @@ SELECT * FROM hash_page_items(get_raw_page('test_hash_a_idx', 4)); SELECT * FROM hash_page_items(get_raw_page('test_hash_a_idx', 5)); ERROR: page is not a hash bucket or overflow page +-- Failure with non-hash index +CREATE INDEX test_hash_a_btree ON test_hash USING btree (a); +SELECT hash_bitmap_info('test_hash_a_btree', 0); +ERROR: "test_hash_a_btree" is not a hash index +-- Failure with incorrect page size +-- Suppress the DETAIL message, to allow the tests to work across various +-- page sizes. +\set VERBOSITY terse +SELECT hash_metapage_info('aaa'::bytea); +ERROR: invalid page size +SELECT hash_page_items('bbb'::bytea); +ERROR: invalid page size +SELECT hash_page_stats('ccc'::bytea); +ERROR: invalid page size +SELECT hash_page_type('ddd'::bytea); +ERROR: invalid page size +\set VERBOSITY default DROP TABLE test_hash; diff --git a/contrib/pageinspect/expected/page.out b/contrib/pageinspect/expected/page.out index e4e0c221dba..75d7e8232b4 100644 --- a/contrib/pageinspect/expected/page.out +++ b/contrib/pageinspect/expected/page.out @@ -113,3 +113,14 @@ select tuple_data_split('test8'::regclass, t_data, t_infomask, t_infomask2, t_bi (1 row) drop table test8; +-- Failure with incorrect page size +-- Suppress the DETAIL message, to allow the tests to work across various +-- page sizes. +\set VERBOSITY terse +SELECT fsm_page_contents('aaa'::bytea); +ERROR: invalid page size +SELECT page_checksum('bbb'::bytea, 0); +ERROR: invalid page size +SELECT page_header('ccc'::bytea); +ERROR: invalid page size +\set VERBOSITY default diff --git a/contrib/pageinspect/fsmfuncs.c b/contrib/pageinspect/fsmfuncs.c index 86e8075845e..788c6fb397c 100644 --- a/contrib/pageinspect/fsmfuncs.c +++ b/contrib/pageinspect/fsmfuncs.c @@ -37,6 +37,7 @@ fsm_page_contents(PG_FUNCTION_ARGS) { bytea *raw_page = PG_GETARG_BYTEA_P(0); StringInfoData sinfo; + Page page; FSMPage fsmpage; int i; @@ -45,7 +46,8 @@ fsm_page_contents(PG_FUNCTION_ARGS) (errcode(ERRCODE_INSUFFICIENT_PRIVILEGE), (errmsg("must be superuser to use raw page functions")))); - fsmpage = (FSMPage) PageGetContents(VARDATA(raw_page)); + page = get_page_from_raw(raw_page); + fsmpage = (FSMPage) PageGetContents(page); initStringInfo(&sinfo); diff --git a/contrib/pageinspect/hashfuncs.c b/contrib/pageinspect/hashfuncs.c index c49adf207cb..c188a68bc29 100644 --- a/contrib/pageinspect/hashfuncs.c +++ b/contrib/pageinspect/hashfuncs.c @@ -420,8 +420,10 @@ hash_bitmap_info(PG_FUNCTION_ARGS) indexRel = index_open(indexRelid, AccessShareLock); if (!IS_HASH(indexRel)) - elog(ERROR, "relation \"%s\" is not a hash index", - RelationGetRelationName(indexRel)); + ereport(ERROR, + (errcode(ERRCODE_WRONG_OBJECT_TYPE), + errmsg("\"%s\" is not a %s index", + RelationGetRelationName(indexRel), "hash"))); if (RELATION_IS_OTHER_TEMP(indexRel)) ereport(ERROR, diff --git a/contrib/pageinspect/rawpage.c b/contrib/pageinspect/rawpage.c index d7bf782ccd5..072f1efe221 100644 --- a/contrib/pageinspect/rawpage.c +++ b/contrib/pageinspect/rawpage.c @@ -218,7 +218,6 @@ Datum page_header(PG_FUNCTION_ARGS) { bytea *raw_page = PG_GETARG_BYTEA_P(0); - int raw_page_size; TupleDesc tupdesc; @@ -235,18 +234,7 @@ page_header(PG_FUNCTION_ARGS) (errcode(ERRCODE_INSUFFICIENT_PRIVILEGE), (errmsg("must be superuser to use raw page functions")))); - raw_page_size = VARSIZE(raw_page) - VARHDRSZ; - - /* - * Check that enough data was supplied, so that we don't try to access - * fields outside the supplied buffer. - */ - if (raw_page_size < SizeOfPageHeaderData) - ereport(ERROR, - (errcode(ERRCODE_INVALID_PARAMETER_VALUE), - errmsg("input page too small (%d bytes)", raw_page_size))); - - page = (PageHeader) VARDATA(raw_page); + page = (PageHeader) get_page_from_raw(raw_page); /* Build a tuple descriptor for our result type */ if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE) @@ -299,25 +287,14 @@ page_checksum(PG_FUNCTION_ARGS) { bytea *raw_page = PG_GETARG_BYTEA_P(0); uint32 blkno = PG_GETARG_INT32(1); - int raw_page_size; - PageHeader page; + Page page; if (!superuser()) ereport(ERROR, (errcode(ERRCODE_INSUFFICIENT_PRIVILEGE), (errmsg("must be superuser to use raw page functions")))); - raw_page_size = VARSIZE(raw_page) - VARHDRSZ; - - /* - * Check that the supplied page is of the right size. - */ - if (raw_page_size != BLCKSZ) - ereport(ERROR, - (errcode(ERRCODE_INVALID_PARAMETER_VALUE), - errmsg("incorrect size of input page (%d bytes)", raw_page_size))); - - page = (PageHeader) VARDATA(raw_page); + page = get_page_from_raw(raw_page); PG_RETURN_INT16(pg_checksum_page((char *) page, blkno)); } diff --git a/contrib/pageinspect/sql/brin.sql b/contrib/pageinspect/sql/brin.sql index 735bc3b6733..8717229c5d2 100644 --- a/contrib/pageinspect/sql/brin.sql +++ b/contrib/pageinspect/sql/brin.sql @@ -15,4 +15,8 @@ SELECT * FROM brin_revmap_data(get_raw_page('test1_a_idx', 1)) LIMIT 5; SELECT * FROM brin_page_items(get_raw_page('test1_a_idx', 2), 'test1_a_idx') ORDER BY blknum, attnum LIMIT 5; +-- Failure for non-BRIN index. +CREATE INDEX test1_a_btree ON test1 (a); +SELECT brin_page_items(get_raw_page('test1_a_btree', 0), 'test1_a_btree'); + DROP TABLE test1; diff --git a/contrib/pageinspect/sql/btree.sql b/contrib/pageinspect/sql/btree.sql index 8eac64c7b3c..174814fb7a4 100644 --- a/contrib/pageinspect/sql/btree.sql +++ b/contrib/pageinspect/sql/btree.sql @@ -18,4 +18,17 @@ SELECT * FROM bt_page_items(get_raw_page('test1_a_idx', 0)); SELECT * FROM bt_page_items(get_raw_page('test1_a_idx', 1)); SELECT * FROM bt_page_items(get_raw_page('test1_a_idx', 2)); +-- Failure when using a non-btree index. +CREATE INDEX test1_a_hash ON test1 USING hash(a); +SELECT bt_metap('test1_a_hash'); +SELECT bt_page_stats('test1_a_hash', 0); +SELECT bt_page_items('test1_a_hash', 0); + +-- Failure with incorrect page size +-- Suppress the DETAIL message, to allow the tests to work across various +-- page sizes. +\set VERBOSITY terse +SELECT bt_page_items('aaa'::bytea); +\set VERBOSITY default + DROP TABLE test1; diff --git a/contrib/pageinspect/sql/gin.sql b/contrib/pageinspect/sql/gin.sql index d516ed3cbd4..2a653609750 100644 --- a/contrib/pageinspect/sql/gin.sql +++ b/contrib/pageinspect/sql/gin.sql @@ -17,3 +17,12 @@ SELECT COUNT(*) > 0 FROM gin_leafpage_items(get_raw_page('test1_y_idx', (pg_relation_size('test1_y_idx') / current_setting('block_size')::bigint)::int - 1)); + +-- Failure with incorrect page size +-- Suppress the DETAIL message, to allow the tests to work across various +-- page sizes. +\set VERBOSITY terse +SELECT gin_leafpage_items('aaa'::bytea); +SELECT gin_metapage_info('bbb'::bytea); +SELECT gin_page_opaque_info('ccc'::bytea); +\set VERBOSITY default diff --git a/contrib/pageinspect/sql/hash.sql b/contrib/pageinspect/sql/hash.sql index 87ee549a7b4..546c780e0e4 100644 --- a/contrib/pageinspect/sql/hash.sql +++ b/contrib/pageinspect/sql/hash.sql @@ -76,5 +76,18 @@ SELECT * FROM hash_page_items(get_raw_page('test_hash_a_idx', 3)); SELECT * FROM hash_page_items(get_raw_page('test_hash_a_idx', 4)); SELECT * FROM hash_page_items(get_raw_page('test_hash_a_idx', 5)); +-- Failure with non-hash index +CREATE INDEX test_hash_a_btree ON test_hash USING btree (a); +SELECT hash_bitmap_info('test_hash_a_btree', 0); + +-- Failure with incorrect page size +-- Suppress the DETAIL message, to allow the tests to work across various +-- page sizes. +\set VERBOSITY terse +SELECT hash_metapage_info('aaa'::bytea); +SELECT hash_page_items('bbb'::bytea); +SELECT hash_page_stats('ccc'::bytea); +SELECT hash_page_type('ddd'::bytea); +\set VERBOSITY default DROP TABLE test_hash; diff --git a/contrib/pageinspect/sql/page.sql b/contrib/pageinspect/sql/page.sql index 42b69e62f52..dddb80b457c 100644 --- a/contrib/pageinspect/sql/page.sql +++ b/contrib/pageinspect/sql/page.sql @@ -52,3 +52,12 @@ select t_bits, t_data from heap_page_items(get_raw_page('test8', 0)); select tuple_data_split('test8'::regclass, t_data, t_infomask, t_infomask2, t_bits) from heap_page_items(get_raw_page('test8', 0)); drop table test8; + +-- Failure with incorrect page size +-- Suppress the DETAIL message, to allow the tests to work across various +-- page sizes. +\set VERBOSITY terse +SELECT fsm_page_contents('aaa'::bytea); +SELECT page_checksum('bbb'::bytea, 0); +SELECT page_header('ccc'::bytea); +\set VERBOSITY default