- TSA is a static analyzer build by Google which finds race conditions
and deadlocks at compile time.
- It works by associating a shared member variable with a
synchronization primitive that protects it. The compiler can then
check at each access if proper locking happened before. A good
introduction are [0] and [1].
- TSA requires some help by the programmer via annotations. Luckily,
LLVM's libcxx already has annotations for std::mutex, std::lock_guard,
std::shared_mutex and std::scoped_lock. This commit enables them
(--> contrib/libcxx-cmake/CMakeLists.txt).
- Further, this commit adds convenience macros for the low-level
annotations for use in ClickHouse (--> base/defines.h). For
demonstration, they are leveraged in a few places.
- As we compile with "-Wall -Wextra -Weverything", the required compiler
flag "-Wthread-safety-analysis" was already enabled. Negative checks
are an experimental feature of TSA and disabled
(--> cmake/warnings.cmake). Compile times did not increase noticeably.
- TSA is used in a few places with simple locking. I tried TSA also
where locking is more complex. The problem was usually that it is
unclear which data is protected by which lock :-(. But there was
definitely some weird code where locking looked broken. So there is
some potential to find bugs.
*** Limitations of TSA besides the ones listed in [1]:
- The programmer needs to know which lock protects which piece of shared
data. This is not always easy for large classes.
- Two synchronization primitives used in ClickHouse are not annotated in
libcxx:
(1) std::unique_lock: A releaseable lock handle often together with
std::condition_variable, e.g. in solve producer-consumer problems.
(2) std::recursive_mutex: A re-entrant mutex variant. Its usage can be
considered a design flaw + typically it is slower than a standard
mutex. In this commit, one std::recursive_mutex was converted to
std::mutex and annotated with TSA.
- For free-standing functions (e.g. helper functions) which are passed
shared data members, it can be tricky to specify the associated lock.
This is because the annotations use the normal C++ rules for symbol
resolution.
[0] https://clang.llvm.org/docs/ThreadSafetyAnalysis.html
[1] https://static.googleusercontent.com/media/research.google.com/en//pubs/archive/42958.pdf
- the CI builds were recently upgraded to Clang 14
- this commit bumps versions of other LLVM tools needed for the build
- this is important for people who have multiple LLVM versions installed
via their package manager
First part, updated most UTF8, hashing, memory and codecs. Except
utf8lower and upper, maybe a little later.
That includes huge amount of research with movemask dealing. Exact
details and blog post TBD.
- It was noticed that in (*), the crashstack says "There is no
information about the reference checksum."
- The binaries are pulled via docker hub and upon inspection they indeed
lack the hash embedded as ELF section ".note.ClickHouse.hash" in the
clickhouse binary. This is weird because docker hub images are
"official" builds which should trigger the hash embedding.
- Turns out that the docker hub binaries are also stripped which was too
aggressive. We now no longer remove sections ".comment" and ".note"
which are anyways only 140 bytes in size, i.e. binary size still goes
down (on my stystem) from 2.1 GB to 0.47 + 0.40 GB binary + dbg info.
(*) https://playground.lodthe.me/ba75d494-95d1-4ff6-a0ad-60c138636c9b
- before: usr/lib/debug/usr/bin/clickhouse.debug/clickhouse.debug
- after : usr/lib/debug/usr/bin/clickhouse.debug
Note, clickhouse_make_empty_debug_info_for_nfpm() is fine.
Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
Globbing generally misses to pick up files which were added/deleted
after CMake's configure. This is a nuissance but can be alleviated using
CONFIGURE_DEPENDS (available since CMake 3.12) which adds a check for
new/deleted files before each compile and - if necessary - restarts the
configuration. On my system, the check takes < 0.1 sec.
(Side note: CONFIGURE_DEPENDS is not guaranteed to work accross all
generators, but at least it works for Ninja which everyone @CH seems to
use.)
The check activated ccache unconditionally for all non-Clang compilers
(= GCC) while allowing ancient ccache versions for these. Perhaps there
was a reason for that in the past but it's simpler to only require a
minimum ccache version.
To simplify further, also require at least ccache 3.3 (released in 2016)
instead of 3.2.1 (released in 2014).
The compiler launcher (ccache, distcc) can be set externally via
-DCMAKE_CXX_COMPILER_LAUNCHER=<tool>. We previously silently ignored
this setting and continued without any launcher (e.g. ccache). Changed
this to now respect the externally specified launcher.
WEVERYTHING enables on Clang literally every warning. People on the
internet are divided if this is a good thing or not but ClickHouse
compiles with -Weverything + some exceptions for noisy warnings since at
least a year.
I tried to build with WEVERYTHING = OFF and the build was badly broken.
It seems nobody actually turns WEVERYTHING off. Actually, why would one
if the CI builds (configured with WEVERYTHING = ON) potentially generate
errors not generated in local development.
To simplify the build scripts and to remove the need to maintain two
sets of compiler warnings, I made WEVERYTHING the default and threw
WEVERYTHING = OFF out.
