| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
* rickard/gcc-atomics/OTP-12383:
Improve ethread atomics based on GCC builtins
Conflicts:
erts/aclocal.m4
|
|
|
|
The old time API is based on erlang:now/0. The major issue with
erlang:now/0 is that it was intended to be used for so many
unrelated things. This tied these unrelated operations together
and unnecessarily caused performance, scalability as well as
accuracy, and precision issues for operations that do not need
to have such issues. The new API spreads different functionality
over multiple functions in order to improve on this.
The new API consists of a number of new BIFs:
- erlang:convert_time_unit/3
- erlang:monotonic_time/0
- erlang:monotonic_time/1
- erlang:system_time/0
- erlang:system_time/1
- erlang:time_offset/0
- erlang:time_offset/1
- erlang:timestamp/0
- erlang:unique_integer/0
- erlang:unique_integer/1
- os:system_time/0
- os:system_time/1
and a number of extensions of existing BIFs:
- erlang:monitor(time_offset, clock_service)
- erlang:system_flag(time_offset, finalize)
- erlang:system_info(os_monotonic_time_source)
- erlang:system_info(time_offset)
- erlang:system_info(time_warp_mode)
- erlang:system_info(time_correction)
- erlang:system_info(start_time)
See the "Time and Time Correction in Erlang" chapter of the
ERTS User's Guide for more information.
|
|
* Use of __atomic builtins when available.
* Improved configure test that checks for missing memory
barrier in __sync_synchronize(). The old approach was to
verify known working gcc versions and check gcc version at
compile time. Besides not being very safe, the old approach
often unnecessarily caused usage of the very expensive
workaround.
* Introduced (no overhead) workaround for missing clobber in
__sync_synchronize() when using buggy LLVM implementation of
__sync_synchronize().
* Implement native memory barriers for ARM processors supporting
the DMB instruction.
* Use of volatile store on Alpha as atomic set operation if no
__atomic_store_n() is available (already used on x86/x86_64
Sparc V9, PowerPC, and MIPS). Fallback used when not using
volatile store is typically very expensive.
* Use volatile load on Alpha and ARM as atomic read operation
if no __atomic_load_n() is available (already used on
x86/x86_64 Sparc V9, PowerPC, and MIPS). Fallback when not
using volatile load is typically very expensive.
|
|
|
|
The ethread atomics API now also provide double word size atomics.
Double word size atomics are implemented using native atomic
instructions on x86 (when the cmpxchg8b instruction is available)
and on x86_64 (when the cmpxchg16b instruction is available). On
other hardware where 32-bit atomics or word size atomics are
available, an optimized fallback is used; otherwise, a spinlock,
or a mutex based fallback is used.
The ethread library now performs runtime tests for presence of
hardware features, such as for example SSE2 instructions, instead
of requiring this to be determined at compile time.
There are now functions implementing each atomic operation with the
following implied memory barrier semantics: none, read, write,
acquire, release, and full. Some of the operation-barrier
combinations aren't especially useful. But instead of filtering
useful ones out, and potentially miss a useful one, we implement
them all.
A much smaller set of functionality for native atomics are required
to be implemented than before. More or less only cmpxchg and a
membar macro are required to be implemented for each atomic size.
Other functions will automatically be constructed from these. It is,
of course, often wise to implement more that this if possible from a
performance perspective.
|
|
|
|
|
|
The atomic memory operations interface used the 'long' type and assumed that
it was of the same size as 'void *'. This is true on most platforms, however,
not on Windows 64.
|
|
Large parts of the ethread library have been rewritten. The
ethread library is an Erlang runtime system internal, portable
thread library used by the runtime system itself.
Most notable improvement is a reader optimized rwlock
implementation which dramatically improve the performance of
read-lock/read-unlock operations on multi processor systems by
avoiding ping-ponging of the rwlock cache lines. The reader
optimized rwlock implementation is used by miscellaneous
rwlocks in the runtime system that are known to be read-locked
frequently, and can be enabled on ETS tables by passing the
`{read_concurrency, true}' option upon table creation. See the
documentation of `ets:new/2' for more information.
The ethread library can now also use the libatomic_ops library
for atomic memory accesses. This makes it possible for the
Erlang runtime system to utilize optimized atomic operations
on more platforms than before. Use the
`--with-libatomic_ops=PATH' configure command line argument
when specifying where the libatomic_ops installation is
located. The libatomic_ops library can be downloaded from:
http://www.hpl.hp.com/research/linux/atomic_ops/
The changed API of the ethread library has also caused
modifications in the Erlang runtime system. Preparations for
the to come "delayed deallocation" feature has also been done
since it depends on the ethread library.
Note: When building for x86, the ethread library will now use
instructions that first appeared on the pentium 4 processor. If
you want the runtime system to be compatible with older
processors (back to 486) you need to pass the
`--enable-ethread-pre-pentium4-compatibility' configure command
line argument when configuring the system.
|
|
Writer preferred pthread read/write locks has been enabled on Linux.
|
|
Support for using gcc's built-in functions for atomic memory access has
been added. This functionallity will be used if available and no other
native atomic implementation in ERTS is available.
|
|
|
