diff options
author | Monty <monty@mariadb.org> | 2021-04-18 15:29:13 +0300 |
---|---|---|
committer | Monty <monty@mariadb.org> | 2021-04-20 12:30:09 +0300 |
commit | 031f11717d9f351dfb12cd27c225c533e289261a (patch) | |
tree | 66da9cdbd1e8acb82da2e75e36a08b3461504716 /sql/sql_select.h | |
parent | eb4123eefc8e0ee4751d3cd3cc49ebf256aa9486 (diff) | |
download | mariadb-git-031f11717d9f351dfb12cd27c225c533e289261a.tar.gz |
Fix all warnings given by UBSAN
The easiest way to compile and test the server with UBSAN is to run:
./BUILD/compile-pentium64-ubsan
and then run mysql-test-run.
After this commit, one should be able to run this without any UBSAN
warnings. There is still a few compiler warnings that should be fixed
at some point, but these do not expose any real bugs.
The 'special' cases where we disable, suppress or circumvent UBSAN are:
- ref10 source (as here we intentionally do some shifts that UBSAN
complains about.
- x86 version of optimized int#korr() methods. UBSAN do not like unaligned
memory access of integers. Fixed by using byte_order_generic.h when
compiling with UBSAN
- We use smaller thread stack with ASAN and UBSAN, which forced me to
disable a few tests that prints the thread stack size.
- Verifying class types does not work for shared libraries. I added
suppression in mysql-test-run.pl for this case.
- Added '#ifdef WITH_UBSAN' when using integer arithmetic where it is
safe to have overflows (two cases, in item_func.cc).
Things fixed:
- Don't left shift signed values
(byte_order_generic.h, mysqltest.c, item_sum.cc and many more)
- Don't assign not non existing values to enum variables.
- Ensure that bool and enum values are properly initialized in
constructors. This was needed as UBSAN checks that these types has
correct values when one copies an object.
(gcalc_tools.h, ha_partition.cc, item_sum.cc, partition_element.h ...)
- Ensure we do not called handler functions on unallocated objects or
deleted objects.
(events.cc, sql_acl.cc).
- Fixed bugs in Item_sp::Item_sp() where we did not call constructor
on Query_arena object.
- Fixed several cast of objects to an incompatible class!
(Item.cc, Item_buff.cc, item_timefunc.cc, opt_subselect.cc, sql_acl.cc,
sql_select.cc ...)
- Ensure we do not do integer arithmetic that causes over or underflows.
This includes also ++ and -- of integers.
(Item_func.cc, Item_strfunc.cc, item_timefunc.cc, sql_base.cc ...)
- Added JSON_VALUE_UNITIALIZED to json_value_types and ensure that
value_type is initialized to this instead of to -1, which is not a valid
enum value for json_value_types.
- Ensure we do not call memcpy() when second argument could be null.
- Fixed that Item_func_str::make_empty_result() creates an empty string
instead of a null string (safer as it ensures we do not do arithmetic
on null strings).
Other things:
- Changed struct st_position to an OBJECT and added an initialization
function to it to ensure that we do not copy or use uninitialized
members. The change to a class was also motived that we used "struct
st_position" and POSITION randomly trough the code which was
confusing.
- Notably big rewrite in sql_acl.cc to avoid using deleted objects.
- Changed in sql_partition to use '^' instead of '-'. This is safe as
the operator is either 0 or 0x8000000000000000ULL.
- Added check for select_nr < INT_MAX in JOIN::build_explain() to
avoid bug when get_select() could return NULL.
- Reordered elements in POSITION for better alignment.
- Changed sql_test.cc::print_plan() to use pointers instead of objects.
- Fixed bug in find_set() where could could execute '1 << -1'.
- Added variable have_sanitizer, used by mtr. (This variable was before
only in 10.5 and up). It can now have one of two values:
ASAN or UBSAN.
- Moved ~Archive_share() from ha_archive.cc to ha_archive.h and marked
it virtual. This was an effort to get UBSAN to work with loaded storage
engines. I kept the change as the new place is better.
- Added in CONNECT engine COLBLK::SetName(), to get around a wrong cast
in tabutil.cpp.
- Added HAVE_REPLICATION around usage of rgi_slave, to get embedded
server to compile with UBSAN. (Patch from Marko).
- Added #ifdef for powerpc64 to avoid a bug in old gcc versions related
to integer arithmetic.
Changes that should not be needed but had to be done to suppress warnings
from UBSAN:
- Added static_cast<<uint16_t>> around shift to get rid of a LOT of
compiler warnings when using UBSAN.
- Had to change some '/' of 2 base integers to shift to get rid of
some compile time warnings.
Reviewed by:
- Json changes: Alexey Botchkov
- Charset changes in ctype-uca.c: Alexander Barkov
- InnoDB changes & Embedded server: Marko Mäkelä
- sql_acl.cc changes: Vicențiu Ciorbaru
- build_explain() changes: Sergey Petrunia
Diffstat (limited to 'sql/sql_select.h')
-rw-r--r-- | sql/sql_select.h | 97 |
1 files changed, 49 insertions, 48 deletions
diff --git a/sql/sql_select.h b/sql/sql_select.h index 1d928334bd8..d21d1bcc305 100644 --- a/sql/sql_select.h +++ b/sql/sql_select.h @@ -698,8 +698,6 @@ end_write_group(JOIN *join, JOIN_TAB *join_tab __attribute__((unused)), bool end_of_records); -struct st_position; - class Semi_join_strategy_picker { public: @@ -710,7 +708,7 @@ public: Update internal state after another table has been added to the join prefix */ - virtual void set_from_prev(struct st_position *prev) = 0; + virtual void set_from_prev(POSITION *prev) = 0; virtual bool check_qep(JOIN *join, uint idx, @@ -720,7 +718,7 @@ public: double *read_time, table_map *handled_fanout, sj_strategy_enum *strategy, - struct st_position *loose_scan_pos) = 0; + POSITION *loose_scan_pos) = 0; virtual void mark_used() = 0; @@ -751,7 +749,7 @@ public: first_dupsweedout_table= MAX_TABLES; is_used= FALSE; } - void set_from_prev(struct st_position *prev); + void set_from_prev(POSITION *prev); bool check_qep(JOIN *join, uint idx, @@ -761,7 +759,7 @@ public: double *read_time, table_map *handled_fanout, sj_strategy_enum *stratey, - struct st_position *loose_scan_pos); + POSITION *loose_scan_pos); void mark_used() { is_used= TRUE; } friend void fix_semijoin_strategies_for_picked_join_order(JOIN *join); @@ -797,7 +795,7 @@ public: is_used= FALSE; } - void set_from_prev(struct st_position *prev); + void set_from_prev(POSITION *prev); bool check_qep(JOIN *join, uint idx, table_map remaining_tables, @@ -806,7 +804,7 @@ public: double *read_time, table_map *handled_fanout, sj_strategy_enum *strategy, - struct st_position *loose_scan_pos); + POSITION *loose_scan_pos); void mark_used() { is_used= TRUE; } friend void fix_semijoin_strategies_for_picked_join_order(JOIN *join); @@ -815,6 +813,7 @@ public: class LooseScan_picker : public Semi_join_strategy_picker { +public: /* The first (i.e. driving) table we're doing loose scan for */ uint first_loosescan_table; /* @@ -833,14 +832,13 @@ class LooseScan_picker : public Semi_join_strategy_picker uint loosescan_parts; /* Number of keyparts to be kept distinct */ bool is_used; -public: void set_empty() { first_loosescan_table= MAX_TABLES; is_used= FALSE; } - void set_from_prev(struct st_position *prev); + void set_from_prev(POSITION *prev); bool check_qep(JOIN *join, uint idx, table_map remaining_tables, @@ -849,19 +847,19 @@ public: double *read_time, table_map *handled_fanout, sj_strategy_enum *strategy, - struct st_position *loose_scan_pos); + POSITION *loose_scan_pos); void mark_used() { is_used= TRUE; } friend class Loose_scan_opt; friend void best_access_path(JOIN *join, JOIN_TAB *s, table_map remaining_tables, - const struct st_position *join_positions, + const POSITION *join_positions, uint idx, bool disable_jbuf, double record_count, - struct st_position *pos, - struct st_position *loose_scan_pos); + POSITION *pos, + POSITION *loose_scan_pos); friend bool get_best_combination(JOIN *join); friend int setup_semijoin_loosescan(JOIN *join); friend void fix_semijoin_strategies_for_picked_join_order(JOIN *join); @@ -888,7 +886,7 @@ public: sjm_scan_last_inner= 0; is_used= FALSE; } - void set_from_prev(struct st_position *prev); + void set_from_prev(POSITION *prev); bool check_qep(JOIN *join, uint idx, table_map remaining_tables, @@ -897,7 +895,7 @@ public: double *read_time, table_map *handled_fanout, sj_strategy_enum *strategy, - struct st_position *loose_scan_pos); + POSITION *loose_scan_pos); void mark_used() { is_used= TRUE; } friend void fix_semijoin_strategies_for_picked_join_order(JOIN *join); @@ -912,8 +910,9 @@ class Rowid_filter; Information about a position of table within a join order. Used in join optimization. */ -typedef struct st_position +class POSITION { +public: /* The table that's put into join order */ JOIN_TAB *table; @@ -925,7 +924,7 @@ typedef struct st_position double records_read; /* The selectivity of the pushed down conditions */ - double cond_selectivity; + double cond_selectivity; /* Cost accessing the table in course of the entire complete join execution, @@ -934,8 +933,6 @@ typedef struct st_position */ double read_time; - /* Cumulative cost and record count for the join prefix */ - Cost_estimate prefix_cost; double prefix_record_count; /* @@ -944,29 +941,46 @@ typedef struct st_position */ KEYUSE *key; + /* Info on splitting plan used at this position */ + SplM_plan_info *spl_plan; + + /* Cost info for the range filter used at this position */ + Range_rowid_filter_cost_info *range_rowid_filter_info; + /* If ref-based access is used: bitmap of tables this table depends on */ table_map ref_depend_map; - + /* - TRUE <=> join buffering will be used. At the moment this is based on - *very* imprecise guesses made in best_access_path(). + Bitmap of semi-join inner tables that are in the join prefix and for + which there's no provision for how to eliminate semi-join duplicates + they produce. */ - bool use_join_buffer; - + table_map dups_producing_tables; + + table_map inner_tables_handled_with_other_sjs; + + Duplicate_weedout_picker dups_weedout_picker; + Firstmatch_picker firstmatch_picker; + LooseScan_picker loosescan_picker; + Sj_materialization_picker sjmat_picker; + + /* Cumulative cost and record count for the join prefix */ + Cost_estimate prefix_cost; + /* Current optimization state: Semi-join strategy to be used for this and preceding join tables. - + Join optimizer sets this for the *last* join_tab in the - duplicate-generating range. That is, in order to interpret this field, + duplicate-generating range. That is, in order to interpret this field, one needs to traverse join->[best_]positions array from right to left. When you see a join table with sj_strategy!= SJ_OPT_NONE, some other - field (depending on the strategy) tells how many preceding positions + field (depending on the strategy) tells how many preceding positions this applies to. The values of covered_preceding_positions->sj_strategy must be ignored. */ enum sj_strategy_enum sj_strategy; - + /* Valid only after fix_semijoin_strategies_for_picked_join_order() call: if sj_strategy!=SJ_OPT_NONE, this is the number of subsequent tables that @@ -975,26 +989,12 @@ typedef struct st_position uint n_sj_tables; /* - Bitmap of semi-join inner tables that are in the join prefix and for - which there's no provision for how to eliminate semi-join duplicates - they produce. + TRUE <=> join buffering will be used. At the moment this is based on + *very* imprecise guesses made in best_access_path(). */ - table_map dups_producing_tables; - - table_map inner_tables_handled_with_other_sjs; - - Duplicate_weedout_picker dups_weedout_picker; - Firstmatch_picker firstmatch_picker; - LooseScan_picker loosescan_picker; - Sj_materialization_picker sjmat_picker; - - /* Info on splitting plan used at this position */ - SplM_plan_info *spl_plan; - - /* Cost info for the range filter used at this position */ - Range_rowid_filter_cost_info *range_rowid_filter_info; - -} POSITION; + bool use_join_buffer; + POSITION(); +}; typedef Bounds_checked_array<Item_null_result*> Item_null_array; @@ -1590,6 +1590,7 @@ public: fields_list= fields_arg; non_agg_fields.empty(); bzero((char*) &keyuse,sizeof(keyuse)); + having_value= Item::COND_UNDEF; tmp_table_param.init(); tmp_table_param.end_write_records= HA_POS_ERROR; rollup.state= ROLLUP::STATE_NONE; |