cosmopolitan/libc
Justine Tunney 85f64f3851
Make futexes 100x better on x86 MacOS
Thanks to @autumnjolitz (in #876) the Cosmopolitan codebase is now
acquainted with Apple's outstanding ulock system calls which offer
something much closer to futexes than Grand Central Dispatch which
wasn't quite as good, since its wait function can't be interrupted
by signals (therefore necessitating a busy loop) and it also needs
semaphore objects to be created and freed. Even though ulock is an
internal Apple API, strictly speaking, the benefits of futexes are
so great that it's worth the risk for now especially since we have
the GCD implementation still as a quick escape hatch if it changes

Here's why this change is important for x86 XNU users. Cosmo has a
suboptimal polyfill when the operating system doesn't offer an API
that let's us implement futexes properly. Sadly we had to use that
on X86 XNU until now. The polyfill works using clock_nanosleep, to
poll the futex in a busy loop with exponential backoff. On XNU x86
clock_nanosleep suffers from us not being able to use a fast clock
gettime implementation, which had a compounding effect that's made
the polyfill function even more poorly. On X86 XNU we also need to
polyfill sched_yield() using select(), which made things even more
troublesome. Now that we have futexes we don't have any busy loops
anymore for both condition variables and thread joining so optimal
performance is attained. To demonstrate, consider these benchmarks

Before:

    $ ./lockscale_test.com -b
    consumed 38.8377   seconds real time and
              0.087131 seconds cpu time

After:

    $ ./lockscale_test.com -b
    consumed 0.007955 seconds real time and
             0.011515 seconds cpu time

Fixes #876
2023-10-03 15:15:43 -07:00
..
calls Make futexes 100x better on x86 MacOS 2023-10-03 15:15:43 -07:00
crt Implement thread cancellation for aarch64 2023-09-07 08:48:38 -07:00
dns Make improvements 2023-10-03 06:17:16 -07:00
elf Make improvements 2023-09-18 21:04:47 -07:00
fmt Make improvements 2023-10-03 06:17:16 -07:00
integral Make improvements 2023-09-06 12:34:59 -07:00
intrin Make futexes 100x better on x86 MacOS 2023-10-03 15:15:43 -07:00
isystem Make improvements 2023-10-03 06:17:16 -07:00
log Make futexes 100x better on x86 MacOS 2023-10-03 15:15:43 -07:00
mem Make improvements 2023-10-03 06:17:16 -07:00
nexgen32e Make improvements 2023-10-03 06:17:16 -07:00
nt Make improvements 2023-10-03 06:17:16 -07:00
proc Make improvements 2023-10-03 06:17:16 -07:00
runtime Make futexes 100x better on x86 MacOS 2023-10-03 15:15:43 -07:00
sock Make improvements 2023-10-03 06:17:16 -07:00
stdio Revert "Rewrite ZipOS" 2023-10-03 14:40:03 -07:00
str Make improvements 2023-10-03 06:17:16 -07:00
sysv Make futexes 100x better on x86 MacOS 2023-10-03 15:15:43 -07:00
testlib Make improvements 2023-10-03 06:17:16 -07:00
thread Make futexes 100x better on x86 MacOS 2023-10-03 15:15:43 -07:00
time Make improvements 2023-10-03 06:17:16 -07:00
tinymath Reduce mandatory stack rss by 256kb 2023-09-07 04:33:01 -07:00
vga [metal] Fix video mode filtering & frame buffer ref-counting (#889) 2023-09-06 03:41:07 -07:00
x Make improvements 2023-10-03 06:17:16 -07:00
ar.h Rewrite Cosmopolitan Ar 2023-07-02 10:19:16 -07:00
assert.h Make greenbean web server better 2023-09-07 03:44:50 -07:00
atomic.h
complex.h Port a lot more code to AARCH64 2023-05-14 09:37:26 -07:00
cosmo.h Invent systemvpe() function 2023-08-09 00:27:55 -07:00
dce.h Replace COSMO define with _COSMO_SOURCE 2023-08-13 20:55:04 -07:00
disclaimer.inc
dos.internal.h Introduce #include <cosmo.h> to toolchain users 2023-06-09 18:03:05 -07:00
empty.s Do some basic build tuning 2023-05-10 04:20:46 -07:00
errno.h Make improvements 2023-09-18 21:04:47 -07:00
imag.internal.h Remove evil constants from cosmopolitan.h 2022-12-17 00:42:45 -08:00
inttypes.h Fix inttypes.h FAST16 macros to have a correct definition (#791) 2023-03-29 00:19:40 -07:00
iso646.internal.h
libc.mk Make improvements 2023-09-18 21:04:47 -07:00
limits.h Make improvements 2023-09-21 07:30:39 -07:00
literal.h Get GNU MPFR and MPC tests to pass 2023-08-21 15:05:10 -07:00
mach.internal.h Introduce #include <cosmo.h> to toolchain users 2023-06-09 18:03:05 -07:00
macho.internal.h Fiddle around with Mach-O 2023-05-20 04:13:49 -07:00
macros.internal.h Make improvements 2023-10-03 06:17:16 -07:00
math.h Get GNU MPFR and MPC tests to pass 2023-08-21 15:05:10 -07:00
notice.inc
notice.internal.h
paths.h Embed cocmd.com interpreter for system() / open() 2022-10-02 15:29:57 -07:00
README.md
stdalign.internal.h
stdbool.h Make fatcosmocc good enough to build ncurses 6.4 2023-08-12 22:30:05 -07:00
stdckdint.h Perform inconsequential code cleanup 2023-08-07 20:24:50 -07:00
temp.h Overhaul process spawning 2023-09-10 08:17:44 -07:00
testlib-test.txt Fix some zipos directory related bugs 2023-09-19 02:30:42 -07:00
type2str.h Improve new C23 checked arithmetic feature 2023-06-16 15:32:18 -07:00
zip.internal.h Revert "Rewrite ZipOS" 2023-10-03 14:40:03 -07:00

Cosmopolitan Standard Library

This directory defines static archives defining functions, like printf(), mmap(), win32, etc. Please note that the Cosmopolitan build configuration doesn't link any C/C++ library dependencies by default, so you still have the flexibility to choose the one provided by your system. If you'd prefer Cosmopolitan, just add $(LIBC) and $(CRT) to your linker arguments.

Your library is compromised of many bite-sized static archives. We use the checkdeps tool to guarantee that the contents of the archives are organized in a logical way that's easy to use with or without our makefile infrastructure, since there's no cyclic dependencies.

The Cosmopolitan Library exports only the most stable canonical system calls for all supported operating systems, regardless of which platform is used for compilation. We polyfill many of the APIs, e.g. read(), write() so they work consistently everywhere while other apis, e.g. CreateWindowEx(), might only work on one platform, in which case they become no-op functions on others.

Cosmopolitan polyfill wrappers will usually use the dollar sign naming convention, so they may be bypassed when necessary. This same convention is used when multiple implementations of string library and other performance-critical function are provided to allow Cosmopolitan to go fast on both old and newer computers.

We take an approach to configuration that relies heavily on the compiler's dead code elimination pass (libc/dce.h). Most of the code is written so that, for example, folks not wanting support for OpenBSD can flip a bit in SUPPORT_VECTOR and that code will be omitted from the build. The same is true for builds that are tuned using -march=native which effectively asks the library to not include runtime support hooks for x86 processors older than what you use.

Please note that, unlike Cygwin or MinGW, Cosmopolitan does not achieve broad support by bolting on a POSIX emulation layer. We do nothing more than (in most cases) stateless API translations that get you 90% of the way there in a fast lightweight manner. We therefore can't address some of the subtle differences, such as the nuances of absolute paths on Windows. Our approach could be compared to something more along the lines of, "the Russians just used a pencil to write in space", versus spending millions researching a pen like NASA.