Followup from 5.5 patch. Removing memory barriers on intel is wrong as
this doesn't prevent the compiler and/or processor from reorganizing reads
before the mutex release. Forcing a memory barrier before reading the waiters will
guarantee that no speculative reading takes place.
Fix memory barrier issues on releasing mutexes. We must have a full
memory barrier between releasing a mutex lock and reading its waiters.
This prevents us from missing to release waiters due to reading the
number of waiters speculatively before releasing the lock. If threads
try and wait between us reading the waiters count and releasing the
lock, those threads might stall indefinitely.
Also, we must use proper ACQUIRE/RELEASE semantics for atomic
operations, not ACQUIRE/ACQUIRE.
commit ef92aaf9ec
Author: Jan Lindström <jan.lindstrom@mariadb.com>
Date: Wed Jun 22 22:37:28 2016 +0300
MDEV-10083: Orphan ibd file when playing with foreign keys
Analysis: row_drop_table_for_mysql did not allow dropping
referenced table even in case when actual creating of the
referenced table was not successfull if foreign_key_checks=1.
Fix: Allow dropping referenced table even if foreign_key_checks=1
if actual table create returned error.
Analysis: row_drop_table_for_mysql did not allow dropping
referenced table even in case when actual creating of the
referenced table was not successfull if foreign_key_checks=1.
Fix: Allow dropping referenced table even if foreign_key_checks=1
if actual table create returned error.
MySQL 5.6 do not work with MariaDB 10.1
Analysis: Problem is that tablespace flags bit DATA_DIR
is on different position on MySQL 5.6 compared to
MariaDB 10.1.
Fix: If we detect that there is difference between dictionary
flags and tablespace flags we remove DATA_DIR flag and compare
again. Remote tablespace is tried to locate even in case
when DATA_DIR flag is not set.
Analysis: When pages in doublewrite buffer are analyzed compressed
pages do not have correct checksum.
Fix: Decompress page before checksum is compared. If decompression
fails we still check checksum and corrupted pages are found.
If decompression succeeds, page now contains the original
checksum.
Problem was that in-place online alter table was used on a table
that had mismatch between MySQL frm file and InnoDB data dictionary.
Fixed so that traditional "Copy" method is used if the MySQL frm
and InnoDB data dictionary is not consistent.
Using __ppc_get_timebase will translate to mfspr instruction
The mfspr instruction will block FXU1 until complete but the other
Pipelines are available for execution of instructions from other
SMT threads on the same core.
The latency time to read the timebase SPR is ~10 cycles.
So any impact on other threads is limited other FXU1 only instructions
(basically other mfspr/mtspr ops).
Suggested by Steven J. Munroe, Linux on Power Toolchain Architect,
Linux Technology Center
IBM Corporation
Bug#18842925 : SET THREAD PRIORITY IN INNODB MUTEX SPINLOOP
Like "pause" instruction for hyper-threading at Intel CPUs,
POWER has special instructions only for hinting priority of hardware-threads.
Approved by Sunny in rb#6256
Backport of the 5.7 fix - c92102a6ef
(excluded cache line size patch)
Suggestion by Stewart Smith
UT_RELAX_CPU(): Use a compiler barrier.
ut_delay(): Remove the dummy global variable ut_always_false.
RB: 11399
Reviewed-by: Jimmy Yang <jimmy.yang@oracle.com>
Backported from MySQL-5.7 - patch 5e3efb0396
Suggestion by Stewart Smith
Make sure that we read all possible encryption keys from checkpoint
and if log block checksum does not match, print all found
checkpoint encryption keys.
Analysis:
-- InnoDB has n (>0) redo-log files.
-- In the first page of redo-log there is 2 checkpoint records on fixed location (checkpoint is not encrypted)
-- On every checkpoint record there is up to 5 crypt_keys containing the keys used for encryption/decryption
-- On crash recovery we read all checkpoints on every file
-- Recovery starts by reading from the latest checkpoint forward
-- Problem is that latest checkpoint might not always contain the key we need to decrypt all the
redo-log blocks (see MDEV-9422 for one example)
-- Furthermore, there is no way to identify is the log block corrupted or encrypted
For example checkpoint can contain following keys :
write chk: 4 [ chk key ]: [ 5 1 ] [ 4 1 ] [ 3 1 ] [ 2 1 ] [ 1 1 ]
so over time we could have a checkpoint
write chk: 13 [ chk key ]: [ 14 1 ] [ 13 1 ] [ 12 1 ] [ 11 1 ] [ 10 1 ]
killall -9 mysqld causes crash recovery and on crash recovery we read as
many checkpoints as there is log files, e.g.
read [ chk key ]: [ 13 1 ] [ 12 1 ] [ 11 1 ] [ 10 1 ] [ 9 1 ]
read [ chk key ]: [ 14 1 ] [ 13 1 ] [ 12 1 ] [ 11 1 ] [ 10 1 ] [ 9 1 ]
This is problematic, as we could still scan log blocks e.g. from checkpoint 4 and we do
not know anymore the correct key.
CRYPT INFO: for checkpoint 14 search 4
CRYPT INFO: for checkpoint 13 search 4
CRYPT INFO: for checkpoint 12 search 4
CRYPT INFO: for checkpoint 11 search 4
CRYPT INFO: for checkpoint 10 search 4
CRYPT INFO: for checkpoint 9 search 4 (NOTE: NOT FOUND)
For every checkpoint, code generated a new encrypted key based on key
from encryption plugin and random numbers. Only random numbers are
stored on checkpoint.
Fix: Generate only one key for every log file. If checkpoint contains only
one key, use that key to encrypt/decrypt all log blocks. If checkpoint
contains more than one key (this is case for databases created
using MariaDB server version 10.1.0 - 10.1.12 if log encryption was
used). If looked checkpoint_no is found from keys on checkpoint we use
that key to decrypt the log block. For encryption we use always the
first key. If the looked checkpoint_no is not found from keys on checkpoint
we use the first key.
Modified code also so that if log is not encrypted, we do not generate
any empty keys. If we have a log block and no keys is found from
checkpoint we assume that log block is unencrypted. Log corruption or
missing keys is found by comparing log block checksums. If we have
a keys but current log block checksum is correct we again assume
log block to be unencrypted. This is because current implementation
stores checksum only before encryption and new checksum after
encryption but before disk write is not stored anywhere.
There was two problems. Firstly, if page in ibuf is encrypted but
decrypt failed we should not allow InnoDB to start because
this means that system tablespace is encrypted and not usable.
Secondly, if page decrypt is detected we should return false
from buf_page_decrypt_after_read.
