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
- This change fixes a bug that allowed unbuffered printf() output (to
streams like stderr) to be truncated. This regression was introduced
some time between now and the last release.
- POSIX specifies all functions as thread safe by default. This change
works towards cleaning up our use of the @threadsafe / @threadunsafe
documentation annotations to reflect that. The goal is (1) to use
@threadunsafe to document functions which POSIX say needn't be thread
safe, and (2) use @threadsafe to document functions that we chose to
implement as thread safe even though POSIX didn't mandate it.
- Tidy up the clock_gettime() implementation. We're now trying out a
cleaner approach to system call support that aims to maintain the
Linux errno convention as long as possible. This also fixes bugs that
existed previously, where the vDSO errno wasn't being translated
properly. The gettimeofday() system call is now a wrapper for
clock_gettime(), which reduces bloat in apps that use both.
- The recently-introduced improvements to the execute bit on Windows has
had bugs fixed. access(X_OK) on a directory on Windows now succeeds.
fstat() will now perform the MZ/#! ReadFile() operation correctly.
- Windows.h is no longer included in libc/isystem/, because it confused
PCRE's build system into thinking Cosmopolitan is a WIN32 platform.
Cosmo's Windows.h polyfill was never even really that good, since it
only defines a subset of the subset of WIN32 APIs that Cosmo defines.
- The setlongerjmp() / longerjmp() APIs are removed. While they're nice
APIs that are superior to the standardized setjmp / longjmp functions,
they weren't superior enough to not be dead code in the monorepo. If
you use these APIs, please file an issue and they'll be restored.
- The .com appending magic has now been removed from APE Loader.
- Every unit test now passes on Apple Silicon. The final piece of this
puzzle was porting our POSIX threads cancelation support, since that
works differently on ARM64 XNU vs. AMD64. Our semaphore support on
Apple Silicon is also superior now compared to AMD64, thanks to the
grand central dispatch library which lets *NSYNC locks go faster.
- The Cosmopolitan runtime is now more stable, particularly on Windows.
To do this, thread local storage is mandatory at all runtime levels,
and the innermost packages of the C library is no longer being built
using ASAN. TLS is being bootstrapped with a 128-byte TIB during the
process startup phase, and then later on the runtime re-allocates it
either statically or dynamically to support code using _Thread_local.
fork() and execve() now do a better job cooperating with threads. We
can now check how much stack memory is left in the process or thread
when functions like kprintf() / execve() etc. call alloca(), so that
ENOMEM can be raised, reduce a buffer size, or just print a warning.
- POSIX signal emulation is now implemented the same way kernels do it
with pthread_kill() and raise(). Any thread can interrupt any other
thread, regardless of what it's doing. If it's blocked on read/write
then the killer thread will cancel its i/o operation so that EINTR can
be returned in the mark thread immediately. If it's doing a tight CPU
bound operation, then that's also interrupted by the signal delivery.
Signal delivery works now by suspending a thread and pushing context
data structures onto its stack, and redirecting its execution to a
trampoline function, which calls SetThreadContext(GetCurrentThread())
when it's done.
- We're now doing a better job managing locks and handles. On NetBSD we
now close semaphore file descriptors in forked children. Semaphores on
Windows can now be canceled immediately, which means mutexes/condition
variables will now go faster. Apple Silicon semaphores can be canceled
too. We're now using Apple's pthread_yield() funciton. Apple _nocancel
syscalls are now used on XNU when appropriate to ensure pthread_cancel
requests aren't lost. The MbedTLS library has been updated to support
POSIX thread cancelations. See tool/build/runitd.c for an example of
how it can be used for production multi-threaded tls servers. Handles
on Windows now leak less often across processes. All i/o operations on
Windows are now overlapped, which means file pointers can no longer be
inherited across dup() and fork() for the time being.
- We now spawn a thread on Windows to deliver SIGCHLD and wakeup wait4()
which means, for example, that posix_spawn() now goes 3x faster. POSIX
spawn is also now more correct. Like Musl, it's now able to report the
failure code of execve() via a pipe although our approach favors using
shared memory to do that on systems that have a true vfork() function.
- We now spawn a thread to deliver SIGALRM to threads when setitimer()
is used. This enables the most precise wakeups the OS makes possible.
- The Cosmopolitan runtime now uses less memory. On NetBSD for example,
it turned out the kernel would actually commit the PT_GNU_STACK size
which caused RSS to be 6mb for every process. Now it's down to ~4kb.
On Apple Silicon, we reduce the mandatory upstream thread size to the
smallest possible size to reduce the memory overhead of Cosmo threads.
The examples directory has a program called greenbean which can spawn
a web server on Linux with 10,000 worker threads and have the memory
usage of the process be ~77mb. The 1024 byte overhead of POSIX-style
thread-local storage is now optional; it won't be allocated until the
pthread_setspecific/getspecific functions are called. On Windows, the
threads that get spawned which are internal to the libc implementation
use reserve rather than commit memory, which shaves a few hundred kb.
- sigaltstack() is now supported on Windows, however it's currently not
able to be used to handle stack overflows, since crash signals are
still generated by WIN32. However the crash handler will still switch
to the alt stack, which is helpful in environments with tiny threads.
- Test binaries are now smaller. Many of the mandatory dependencies of
the test runner have been removed. This ensures many programs can do a
better job only linking the the thing they're testing. This caused the
test binaries for LIBC_FMT for example, to decrease from 200kb to 50kb
- long double is no longer used in the implementation details of libc,
except in the APIs that define it. The old code that used long double
for time (instead of struct timespec) has now been thoroughly removed.
- ShowCrashReports() is now much tinier in MODE=tiny. Instead of doing
backtraces itself, it'll just print a command you can run on the shell
using our new `cosmoaddr2line` program to view the backtrace.
- Crash report signal handling now works in a much better way. Instead
of terminating the process, it now relies on SA_RESETHAND so that the
default SIG_IGN behavior can terminate the process if necessary.
- Our pledge() functionality has now been fully ported to AARCH64 Linux.
This change implements a new approach to function call logging, that's
based on the GCC flag: -fpatchable-function-entry. Read the commentary
in build/config.mk to learn how it works.
There's a new program named ape/ape-m1.c which will be used to build an
embeddable binary that can load ape and elf executables. The support is
mostly working so far, but still chasing down ABI issues.
- Utilities like pledge.com now build
- kprintf() will no longer balk at 48-bit addresses
- There's a new aarch64-dbg build mode that should work
- gc() and defer() are mostly pacified; avoid using them on aarch64
- THIRD_PART_STB now has Arm Neon intrinsics for fast image handling
640 bytes for old kDos2Errno table
182 bytes for new kDos2Errno under hello2.com (MODE=fastbuild)
122 bytes for new kDos2Errno under hello2.com (MODE=tiny)
The whole repository is now buildable with GNU Make Landlock sandboxing.
This proves that no Makefile targets exist which touch files other than
their declared prerequisites. In order to do this, we had to:
1. Stop code morphing GCC output in package.com and instead run a
newly introduced FIXUPOBJ.COM command after GCC invocations.
2. Disable all the crumby Python unit tests that do things like create
files in the current directory, or rename() files between folders.
This ended up being a lot of tests, but most of them are still ok.
3. Introduce an .UNSANDBOXED variable to GNU Make to disable Landlock.
We currently only do this for things like `make tags`.
4. This change deletes some GNU Make code that was preventing the
execve() optimization from working. This means it should no longer
be necessary in most cases for command invocations to be indirected
through the cocmd interpreter.
5. Missing dependencies had to be declared in certain places, in cases
where they couldn't be automatically determined by MKDEPS.COM
6. The libcxx header situation has finally been tamed. One of the
things that makes this difficult is MKDEPS.COM only wants to
consider the first 64kb of a file, in order to go fast. But libcxx
likes to have #include lines buried after huge documentation.
7. An .UNVEIL variable has been introduced to GNU Make just in case
we ever wish to explicitly specify additional things that need to
be whitelisted which aren't strictly prerequisites. This works in
a manner similar to the recently introduced .EXTRA_PREREQS feature.
There's now a new build/bootstrap/make.com prebuilt binary available. It
should no longer be possible to write invalid Makefile code.
This change simplifies the thread-local storage support code. On Windows
and Mac OS X the startup latency of __enable_tls() has been reduced from
30ms to 1ms. On Windows, TLS memory accesses will now go much faster due
to better self-modifying code that prevents a function call and acquires
our thread information block pointer in a single instruction.
We now rewrite the binary image at runtime on Windows and XNU to change
mov %fs:0,%reg instructions to use %gs instead. There's also simpler
threading API introduced by this change and it's called _spawn() and
_join(), which has replaced most clone() usage.
- Write tests for cthreads
- Fix bugs in pe2.com tool
- Fix ASAN issue with GetDosEnviron()
- Consolidate the cthread header files
- Some code size optimizations for MODE=
- Attempted to squash a tls linker warning
- Attempted to get futexes working on FreeBSD
- Document redbean's argon2 module
- Fix regressions in cthreads library
- Make testlib work better with threads
- Give the cthreads library lots of love
- Remove some of the stdio assembly code
- Implement getloadavg() across platforms
- Code size optimizations for errnos, etc.
- Only check for signals in main thread on Windows
- Make errnos for dup2 / dup3 consistent with posix
This change also fixes a bug in the argon2 module, where the NUL
terminator was being included in the hash encoded ascii string. This
shouldn't require any database migrations to folks who found this module
and productionized it, since the argon2 library treats it as a c string.
This change fixes a nasty regression caused by
80b211e314 which deadlocked.
This change also causes MbedTLS to prefer the ChaCha ciphersuite on
older CPUs that don't have AES hardware instructions.
- add vdso dump utility
- tests now log stack usage
- rename g_ftrace to __ftrace
- make internal spinlocks go faster
- add conformant c11 atomics library
- function tracing now logs stack usage
- make function call tracing thread safe
- add -X unsecure (no ssl) mode to redbean
- munmap() has more consistent behavior now
- pacify fsync() calls on python unit tests
- make --strace flag work better in redbean
- start minimizing and documenting compiler flags
This is similar to the --ftrace (c function call trace) flag, except
it's less noisy since it only logs system calls to stderr. Having this
flag is valuable because (1) system call tracing tells us a lot about
the behavior of complex programs and (2) it's usually very hard to get
system call tracing on various operating systems, e.g. strace, ktrace,
dtruss, truss, nttrace, etc. Especially on Apple platforms where even
with the special boot trick, debuggers still aren't guaranteed to work.
make -j8 o//examples
o//examples/hello.com --strace
This is enabled by default in MODE=, MODE=opt, and MODE=dbg. In MODE=dbg
extra information will be printed.
make -j8 MODE=dbg o/dbg/examples
o/dbg/examples/hello.com --strace |& less
This change also changes:
- Rename IsText() → _istext()
- Rename IsUtf8() → _isutf8()
- Fix madvise() on Windows NT
- Fix empty string case of inet_ntop()
- vfork() wrapper now saves and restores errno
- Update xsigaction() to yoink syscall support
It turned out that the linker was doing the wrong with the amalgamation
library concerning weak stubs. A regression test has been added and new
binaries have been uploaded to https://justine.lol/cosmopolitan/
Ideally this should be fixed by building a tool that turns multiple .a
files into a single .a file with deduplication. As a workaround for now
the cosmopolitan.a build is restructured to not include LIBC_STUBS which
meant technical debt needed to be paid off where non-stub interfaces
were moved to LIBC_INTRIN and LIBC_NEXGEN32E.
Thank @PerfectProductions in #31 for the report!
This is done without using Microsoft's internal APIs. MAP_PRIVATE
mappings are copied to the subprocess via a pipe, since Microsoft
doesn't want us to have proper COW pages. MAP_SHARED mappings are
remapped without needing to do any copying. Global variables need
copying along with the stack and the whole heap of anonymous mem.
This actually improves the reliability of the redbean http server
although one shouldn't expect 10k+ connections on a home computer
that isn't running software built to serve like Linux or FreeBSD.
I wanted a tiny scriptable meltdown proof way to run userspace programs
and visualize how program execution impacts memory. It helps to explain
how things like Actually Portable Executable works. It can show you how
the GCC generated code is going about manipulating matrices and more. I
didn't feel fully comfortable with Qemu and Bochs because I'm not smart
enough to understand them. I wanted something like gVisor but with much
stronger levels of assurances. I wanted a single binary that'll run, on
all major operating systems with an embedded GPL barrier ZIP filesystem
that is tiny enough to transpile to JavaScript and run in browsers too.
https://justine.storage.googleapis.com/emulator625.mp4
