Partial commit of the greater MDEV-34348 scope.
MDEV-34348: MariaDB is violating clang-16 -Wcast-function-type-strict
Change the type of my_hash_get_key to:
1) Return const
2) Change the context parameter to be const void*
Also fix casting in hash adjacent areas.
Reviewed By:
============
Marko Mäkelä <marko.makela@mariadb.com>
PFS_atomic class contains wrappers around my_atomic_* operations, which
are macros to GNU atomic operations (__atomic_*). Due to different
implementations of compilers, clang may encounter errors when compiling
on x86_32 architecture.
The following functions are replaced with C++ std::atomic type in
performance schema code base:
- PFS_atomic::store_*()
-> my_atomic_store*
-> __atomic_store_n()
=> std::atomic<T>::store()
- PFS_atomic::load_*()
-> my_atomic_load*
-> __atomic_load_n()
=> std::atomic<T>::load()
- PFS_atomic::add_*()
-> my_atomic_add*
-> __atomic_fetch_add()
=> std::atomic<T>::fetch_add()
- PFS_atomic::cas_*()
-> my_atomic_cas*
-> __atomic_compare_exchange_n()
=> std::atomic<T>::compare_exchange_strong()
and PFS_atomic class could be dropped completely.
Note that in the wrapper memory order passed to original GNU atomic
extensions are hard-coded as `__ATOMIC_SEQ_CST`, which is equivalent to
`std::memory_order_seq_cst` in C++, and is the default parameter for
std::atomic_* functions.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services.
A few different incorrect function type UBSAN issues have been
grouped into this patch.
The only real potentially undefined behavior is an error about
show_func_mutex_instances_lost, which when invoked in
sql_show.cc::show_status_array(), puts 5 arguments onto the stack;
however, the implementing function only actually has 3 parameters (so
only 3 would be popped). This was fixed by adding in the remaining
parameters to satisfy the type mysql_show_var_func.
The rest of the findings are pointer type mismatches that wouldn't
lead to actual undefined behavior. The lf_hash_initializer function
type definition is
typedef void (*lf_hash_initializer)(LF_HASH *hash, void *dst, const void *src);
but the MDL_lock and table cache's implementations of this function
do not have that signature. The MDL_lock has specific MDL object
parameters:
static void lf_hash_initializer(LF_HASH *hash __attribute__((unused)),
MDL_lock *lock, MDL_key *key_arg)
and the table cache has specific TDC parameters:
static void tdc_hash_initializer(LF_HASH *,
TDC_element *element, LEX_STRING *key)
leading to UBSAN runtime errors when invoking these functions.
This patch fixes these type mis-matches by changing the
implementing functions to use void * and const void * for their
respective parameters, and later casting them to their expected
type in the function body.
Note too the functions tdc_hash_key and tc_purge_callback had
a similar problem to tdc_hash_initializer and was fixed
similarly.
Reviewed By:
============
Sergei Golubchik <serg@mariadb.com>
I checked all stack overflow potential problems found with
gcc -Wstack-usage=16384
and
clang -Wframe-larger-than=16384 -no-inline
Fixes:
Added '#pragma clang diagnostic ignored "-Wframe-larger-than="'
to a lot of function to where stack usage large but resonable.
- Added stack check warnings to BUILD scrips when using clang and debug.
Function changed to use malloc instead allocating things on stack:
- read_bootstrap_query() now allocates line_buffer (20000 bytes) with
malloc() instead of using stack. This has a small performance impact
but this is not releant for bootstrap.
- mroonga grn_select() used 65856 bytes on stack. Changed it to use
malloc().
- Wsrep_schema::replay_transaction() and
Wsrep_schema::recover_sr_transactions().
- Connect zipOpen3()
Not fixed:
- mroonga/vendor/groonga/lib/expr.c grn_proc_call() uses
43712 byte on stack. However this is not easy to fix as the stack
used is caused by a lot of code generated by defines.
- Most changes in mroonga/groonga where only adding of pragmas to disable
stack warnings.
- rocksdb/options/options_helper.cc uses 20288 of stack space.
(no reason to fix except to get rid of the compiler warning)
- Causes using alloca() where the allocation size is resonable.
- An issue in libmariadb (reported to connectors).
to iterate over all status variables one should use
LOCK_all_status_vars not LOCK_status
this fixes sporadic mutex lock inversion in plugins.password_reuse_check:
* acl_cache->lock is taken over complex operations that might increment
status counters (under LOCK_status).
* acl_cache->lock is needed to get the values of Acl% status variables
when iterating over status variables
This column (`COUNT_TRANSACTIONS_RETRIES`) is defined as `BIGINT UNSIGNED`
(64-bit unsigned integer) in the user-visible SQL definition:
182ff21ace/storage/perfschema/table_replication_applier_status.cc (L66)
"COUNT_TRANSACTIONS_RETRIES BIGINT unsigned not null comment 'The number of retries that were made because the replication SQL thread failed to apply a transaction.',"
And its value is internally set/updated using the `set_field_ulonglong`
function:
182ff21ace/storage/perfschema/table_replication_applier_status.cc (L231-L233)
case 3: /* total number of times transactions were retried */
set_field_ulonglong(f, m_row.count_transactions_retries);
break;
… but the structure where it is stored allocates only `ulong` for it:
182ff21ace/storage/perfschema/table_replication_applier_status.h (L62)
ulong count_transactions_retries;
As a result of this inconsistency:
1. On any platform where `ulong` is `uint32_t` and `ulonglong` is `uint64_t`,
setting this value would corrupt the 4 bytes of memory *following* the 4
bytes actually allocated for it.
Likely this problem was never noticed because this is the final element in
the structure, and the structure is padded by the compiler to prevent
memory corruption errors.
2. On any BIG-ENDIAN platform where `ulong` is `uint32_t` and `ulonglong`
is `uint64_t`, reading back the value of this column will result in
total garbage.
Likely this problem was never noticed because MariaDB has not been
tested on 32-bit big-endian platforms.
In order not to affect the user-visible/SQL definition of this column, the
correct way to fix this issue is to change it to `ulonglong` in the
structure definition. See
https://github.com/MariaDB/server/pull/2763/files#r1329110832 for the
original identification and discussion of this issue.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the BSD-new
license. I am contributing on behalf of my employer Amazon Web Services
perfschema thread walker needs to take thread's LOCK_thd_kill to prevent
the thread from disappearing why it's being looked at.
But there's no need to lock it for the current thread.
In fact, it was harmful as some code down the stack might take
LOCK_thd_kill (e.g. set_killed() does it, and my_malloc_size_cb_func()
calls set_killed()). And it caused a bunch of mutexes being locked under
LOCK_thd_kill, which created problems later when my_malloc_size_cb_func()
called set_killed() at some unspecified point under some
random mutexes.
fixes the failure of
./mtr --ps sys_vars.gtid_slave_pos_grant sysschema.v_session_ssl_status
safe_mutex: Found wrong usage of mutex 'LOCK_thd_data' and 'LOCK_active_mi'
* treat FUNC/ARRAY variables as SESSION (otherwise they won't be shown)
* allow SHOW_SIMPLE_FUNC everywhere where SHOW_FUNC is
* increase row buffer size to avoid "too short" assert
This reverts commit 03c9a4ef4a.
The fix is wrong. It was doing this: if the uninitialized
wait->m_class has some specific value, then don't initialize it.
…with: Test assertion failed
Problem:
=======
Assertion text: 'Value returned by SSS and PS table for Last_Error_Number
should be same.'
Assertion condition: '"1146" = "0"'
Assertion condition, interpolated: '"1146" = "0"'
Assertion result: '0'
Analysis:
========
In parallel replication when slave is started the worker pool gets
activated and it gets cleared when slave stops. Each time the worker pool
gets activated a backup worker pool also gets created to store worker
specific perforance schema information in case of errors. On error, all
relevant information is copied from rpl_parallel_thread to rli and it gets
cleared from thread. Then server waits for all workers to complete their
work, during this stage performance schema table specific worker info is
stored into the backup pool and finally the actual pool gets cleared. If
users query the performance schema table to know the status of workers the
information from backup pool will be used. The test simulates
ER_NO_SUCH_TABLE error and verifies the worker information in pfs table.
Test works fine if execution occurs in following order.
Step 1. Error occurred 'worker information is copied to backup pool'.
Step 2. handle_slave_sql invokes 'rpl_parallel_resize_pool_if_no_slaves' to
deactivate worker pool, it marks the pool->count=0
Step 3. PFS table is queried, since actual pool is deactivated backup pool
information is read.
If the Step 3 happens prior to Step2 the pool is yet to be deactivated and
the actual pool is read, which doesn't have any error details as they were
cleared. Hence test ocasionally fails.
Fix:
===
Upon error mark the back pool as being active so that if PFS table is
quried since the backup pool is flagged as valid its information will be
read, in case it is not flagged regular pool will be read.
This work is one of the last pieces created by the late Sujatha Sivakumar.
When one session SELECT ... FOR UPDATE and holds the lock, subsequent
sessions that SELECT ... FOR UPDATE will wait to get the lock.
Currently, that event is labeled as `wait/io/table/sql/handler`, which
is incorrect. Instead, it should have been
`wait/lock/table/sql/handler`.
Two factors contribute to this bug:
1. Instrumentation interface and the heavy usage of `TABLE_IO_WAIT` in
`sql/handler.cc` file. See interface [^1] for better understanding;
2. The balancing act [^2] of doing instrumentation aggregration _AND_
having good performance. For example, EVENTS_WAITS_SUMMARY... is
aggregated using EVENTS_WAITS_CURRENT. Aggregration needs to be based
on the same wait class, and the code was overly aggressive in label a
LOCK operation as an IO operation in this case.
The proposed fix is pretty simple, but understanding the bug took a
while. Hence the footnotes below. For future improvement and
refactoring, we may want to consider renaming `TABLE_IO_WAIT` and making
it less coarse and more targeted.
Note that newly added test case, events_waits_current_MDEV-29091,
initially didn't pass Buildbot CI for embedded build tests. Further
research showed that other impacted tests all included not_embedded.inc.
This oversight was fixed later.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
[^1]: To understand `performance_schema` instrumentation interface, I
found this URL is the most helpful:
https://dev.mysql.com/doc/dev/mysql-server/latest/PAGE_PFS_PSI.html
[^2]: The best place to understand instrumentation projection,
composition, and aggregration is through the source file. Although I
prefer reading Doxygen produced html file, but for whatever reason, the
rendering is not ideal. Here is link to 10.6's pfs.cc:
https://github.com/MariaDB/server/blob/10.6/storage/perfschema/pfs.cc
This patch is the result of running
run-clang-tidy -fix -header-filter=.* -checks='-*,modernize-use-equals-default' .
Code style changes have been done on top. The result of this change
leads to the following improvements:
1. Binary size reduction.
* For a -DBUILD_CONFIG=mysql_release build, the binary size is reduced by
~400kb.
* A raw -DCMAKE_BUILD_TYPE=Release reduces the binary size by ~1.4kb.
2. Compiler can better understand the intent of the code, thus it leads
to more optimization possibilities. Additionally it enabled detecting
unused variables that had an empty default constructor but not marked
so explicitly.
Particular change required following this patch in sql/opt_range.cc
result_keys, an unused template class Bitmap now correctly issues
unused variable warnings.
Setting Bitmap template class constructor to default allows the compiler
to identify that there are no side-effects when instantiating the class.
Previously the compiler could not issue the warning as it assumed Bitmap
class (being a template) would not be performing a NO-OP for its default
constructor. This prevented the "unused variable warning".
On GNU/Linux, even though the C11 aligned_alloc() appeared in
GNU libc early on, some custom memory allocators did not
implement it until recently. For example, before
gperftools/gperftools@d406f22853
the free() in tcmalloc would fail to free memory that was
returned by aligned_alloc(), because the latter would map to the
built-in allocator of libc. The Linux specific memalign() has a
similar interface and is safer to use, because it has been
available for a longer time. For AddressSanitizer, we will use
aligned_alloc() so that the constraint on size can be enforced.
buf_tmp_reserve_compression_buf(): When HAVE_ALIGNED_ALLOC holds,
round up the size to be an integer multiple of the alignment.
pfs_malloc(): In the unit test stub, round up the size to be an
integer multiple of the alignment.
Table_cache_instance: Define the structure aligned at
the CPU cache line, and remove a pad[] data member.
Krunal Bauskar reported this to improve performance on ARMv8.
aligned_malloc(): Wrapper for the Microsoft _aligned_malloc()
and the ISO/IEC 9899:2011 <stdlib.h> aligned_alloc().
Note: The parameters are in the Microsoft order (size, alignment),
opposite of aligned_alloc(alignment, size).
Note: The standard defines that size must be an integer multiple
of alignment. It is enforced by AddressSanitizer but not by GNU libc
on Linux.
aligned_free(): Wrapper for the Microsoft _aligned_free() and
the standard free().
HAVE_ALIGNED_ALLOC: A new test. Unfortunately, support for
aligned_alloc() may still be missing on some platforms.
We will fall back to posix_memalign() for those cases.
HAVE_MEMALIGN: Remove, along with any use of the nonstandard memalign().
PFS_ALIGNEMENT (sic): Removed; we will use CPU_LEVEL1_DCACHE_LINESIZE.
PFS_ALIGNED: Defined using the C++11 keyword alignas.
buf_pool_t::page_hash_table::create(),
lock_sys_t::hash_table::create():
lock_sys_t::hash_table::resize(): Pad the allocation size to an
integer multiple of the alignment.
Reviewed by: Vladislav Vaintroub
