mirror of
https://github.com/boostorg/context.git
synced 2026-01-26 06:22:42 +00:00
aadb5968f4
- tracking parent inside of execution_context: - costs performance - execution_context doesn't realy know which execution_context should be executed next (if it's terminating) - fibers do not care of the parent context -> scheduler decides which is the next context
38 lines
1.1 KiB
Plaintext
38 lines
1.1 KiB
Plaintext
[/
|
|
Copyright Oliver Kowalke 2014.
|
|
Distributed under the Boost Software License, Version 1.0.
|
|
(See accompanying file LICENSE_1_0.txt or copy at
|
|
http://www.boost.org/LICENSE_1_0.txt
|
|
]
|
|
|
|
[section:performance Performance]
|
|
|
|
Performance of __boost_context__ was measured on the platforms shown in the
|
|
following table. Performance measurements were taken using `rdtsc` and
|
|
`boost::chrono::high_resolution_clock`, with
|
|
overhead corrections, on x86 platforms. In each case, cache warm-up was
|
|
accounted for, and the one running thread was pinned
|
|
to a single CPU. The code was compiled using the build options,
|
|
'variant = release cxxflags = -DBOOST_DISABLE_ASSERTS'.
|
|
|
|
[table Performance of context switch
|
|
[[Platform] [ucontext_t] [fcontext_t] [execution_context] [windows fibers]]
|
|
[
|
|
[i386 [footnote AMD Athlon 64 DualCore 4400+]]
|
|
[708 ns / 754 cycles]
|
|
[37 ns / 37 cycles]
|
|
[ ns / cycles]
|
|
[ ns / cycles]
|
|
]
|
|
[
|
|
[x86_64 [footnote Intel Core2 Q6700]]
|
|
[547 ns / 1433 cycles]
|
|
[8 ns / 23 cycles]
|
|
[16 ns / 46 cycles]
|
|
[ ns / cycles]
|
|
]
|
|
]
|
|
|
|
|
|
[endsect]
|