It is still somewhat broken and somewhat of a mess, but it is much
lesser mess now.
As part of this change I also amputated a number of unnecessary or too
complicated bits. Like attempt to build both static and shared
versions of tcmalloc libraries at the same time. We stick to cmake's
"common" (seemingly) behavior of defaulting to something (in our case
shared library) and letting users override BUILD_SHARED_LIBS.
All the "install" bits are amputated as well, they're not ready.
We now wrap StackTraceScope thingy in tcmalloc-specific parts, instead
of automagically inside every stacktrace.cc function. TCMalloc bits
all need to grab stacktraces via newly introduced
tcmalloc::GrabBacktrace (which handles emergency malloc wrapping).
New approach eliminates the need for doing fake stacktrace scope. CPU
profiler, being distinct .so library couldn't take advantage of
emergency malloc anyways.
This simplifies the build further and eliminates another potential
point of runtime divergence when stacktrace is linked to both
libprofiler and libtcmalloc.
Logic was removed from thread_cache.{h,cc} into
thread_cache_ptr.{h,cc}.
Separation will help possible future evolution, and we already changed
the logic quite a bit:
* early access (when TLS isn't initialized yet) now uses global
ThreadCache instance. We therefore have ThreadCachePtr instances
managing required locking. This eliminates unnecessary complication of
PTHREADS_CRASHES_IF_RUN_TOO_EARLY logic, and any other danger of
touching TLS too early. BTW previous implementation actually leaked
initial early-initialized ThreadCache instance(!)
* old configure-time HAVE_TLS logic is amputated. Config-time part of
it made little sense as C++ 17 guarantees availability of
thread_local, but we have manually curated deny-list of "bad" OSes,
that we tested (via compile checks!) at configure time. Now this
is all compile time. There is now compile-time kHaveGoodTLS variable
and we're using it mostly via if constexpr.
* kHaveGoodTLS case of creating thread cache is simplified and made
more straightforward (no need to have in_setspecific logic).
* !kHaveGoodTLS case if fixed and improved too. We avoid
std:🧵:get_id, as it deadlocks on mingw. We use errno address as
a portable and (usually) async-signal safe 'my thread' identifier. We
also eliminate linear searching of thread's cache and replace it with
straightforward hash table lookup.
This allows us, later, to avoid building this stuff in configurations
that don't use it. I have also reduced API and ABI surface to enable
further refactorings.
* Remove build dependency on HAVE_PTHREAD
* Remove build dependency on HAVE_STD_ALIGNED_VAL_T and ENABLE_ALIGNED_NEW_DELETE
* Remove redundant tcmalloc.h files & ensure there are no cross-build-tool references
* Adopt automake commit 26927d1 in the CMake build
- Fix CMake builds for MinGW and MSVC
- Ensure the Autotools, CMake and VSProj builds do not reference each others' config.h
- Use std:🧵:id instead of our own thread ID wrappers
- Moved explicit TLS wrapper functions into the tcmalloc:: namespace and change their visibility to hidden
Resolves#1486
Pretty much everything has fork, except windows. Which is much simpler
to test at compile time, then duplicating and complicating the test at
configure time.
As part of cpu profiler we're extracting current PC (program counter)
of out signal's ucontext. Different OS and hardware combinations have
different ways for that. We had a list of variants that we tested at
compile time and populated PC_FROM_UCONTEXT macro into config.h. It
caused duplication and occasional mismatches between our autoconf and
cmake bits.
So this commit changes testing to be compile-time. We remove
complexity from build system and add some to C++ source.
We use SFINAE to find which of those variants compile (and we silently
assume that 'compiles' implies 'works'; this is what config-time
testing did too). Occasionally we'll face situations where several
variants compile. And we couldn't handle this case in pure C++. So we
have a small Ruby program that generates chain of inheritance among
SFINAE-specialized class templates. This handles prioritization among
variants.
List of ucontext->pc extraction variants is mostly same. We dropped
super-obsolete (circa Linux kernel 2.0) arm variant. And NetBSD case
is now improved. We now use their nice architecture-independent macro
instead of x86-specific access.
We used msync to verify that address is readable. But msync gives
false positives for PROT_NONE mappings. And we recently got bug report
from user hitting this exact condition.
For correct access check, we steal idea from Abseil and do sigprocmask
with address used as new signal mask and with invalid HOW
argument. This works in today's Linux kernels and is among fastest
methods available. But is brittle w.r.t. possible kernel changes. So
we supply fallback method that does 2 syscalls.
For non-Linux systems we implement usual "write to pipe" trick. Which
also has decent performance, but requires occasional pipe draining and
uses fds which could occasionally be damaged by some forking codes.
We also finally cover all new code with unit test.
Fixes github issue #1426