maria:bb-10.5-all-builders

Last commit made on 2024-02-16
Get this branch:
git clone -b bb-10.5-all-builders https://git.launchpad.net/maria

Branch merges

Branch information

Name:
bb-10.5-all-builders
Repository:
lp:maria

Recent commits

318a7e2... by VladislavVaintroub

speed up SSL connection time by approx factor of 2, for WolfSSL.

Optimize common RSA math operations (WOLFSSL_HAVE_SP_RSA),
Use intel assembly optimizations for it.

"slow" ssl tests, such as main.ssl-big and main.ssl_connect now execute
in half the time, with this patch.

6a2461b... by VladislavVaintroub

WolfSSL - Use MSVC_INTEL rather than MSVC, to indicate optimizations

Take into account that server can be compiler on ARM64

2ae2f88... by VladislavVaintroub

WolfSSL - restore workaround for aes_asm.S

70ba46c... by VladislavVaintroub

Refactor WolfSSL build

Remove old workarounds.

73709b4... by VladislavVaintroub

WolfSSL - use "SP Math All" library for multi-precision math, rather than
fastmath

"Sp Math All" marketed as faster more portable library. And indeed,
ssl_connect and ssl-big tests run somewhat faster with it.

Also remove some older workarounds for building wolfssl

8a1904d... by Daniel Black

MDEV-33301 memlock with systemd still not working

CapabilityBoundingSet included CAP_IPC_LOCK in MDEV-9095, however
it requires that the executable has the capability marked in extended
attributes also.

The alternate to this is raising the RLIMIT_MEMLOCK for the service/
process to be able to complete the mlockall system call. This needs to
be adjusted to whatever the MariaDB server was going to allocate.
Rather than leave the non-obvious mapping of settings and tuning,
add the capability so its easier for the user.

We set the capability, if possible, but may never be used depending
on user settings. As such in the Debian postinst script, don't
complain if this fails.

The CAP_IPC_LOCK also facilitates the mmaping of huge memory pages.
(see man mmap), like mariadb uses with --large-pages.

0185ac6... by Dave Gosselin <email address hidden>

MDEV-30975 Wrong result with cross Join given join order

For queries with derived tables populated having some side-effect, we
will fill such a derived table more than once, but without clearing
its rows. Consequently it will have duplicate rows.
An example query exhibiting the problem is
  SELECT STRAIGHT_JOIN c1 FROM t1 JOIN (SELECT @a := 0) x;
Since mysql_derived_fill will, for UNCACHEABLE_DEPENDENT tables, drop
all rows and repopulate, we relax the condition at line 1204: rather
than assume all uncacheable values prevent early return, we now
allow an early return for uncacheable values other than
UNCACHEABLE_DEPENDENT. In general, we only populate derived tables
once unless they're dependent tables.

85517f6... by Alexey Botchkov

MDEV-33393 audit plugin do not report user did the action..

The '<replication_slave>' user is assigned to the slave replication
thread so this name appears in the auditing logs.

b770633... by Marko Mäkelä

Merge 10.4 into 10.5

68d9deb... by Marko Mäkelä

MDEV-33332 SIGSEGV in buf_read_ahead_linear() when bpage is in buf_pool.watch

buf_read_ahead_linear(): If buf_pool.watch_is_sentinel(*bpage),
do not attempt to read the page frame because the pointer would be null
for the elements of buf_pool.watch[].

Hitting this bug requires the use of a non-default value of
innodb_change_buffering.