At least in neovim, `│vi:` is not recognized as a modeline because it
has no preceding whitespace. After fixing this, opening a file yields
an error because `net` is not an option. (`noet`, however, is.)
- We now serialize the file descriptor table when spawning / executing
processes on Windows. This means you can now inherit more stuff than
just standard i/o. It's needed by bash, which duplicates the console
to file descriptor #255. We also now do a better job serializing the
environment variables, so you're less likely to encounter E2BIG when
using your bash shell. We also no longer coerce environ to uppercase
- execve() on Windows now remotely controls its parent process to make
them spawn a replacement for itself. Then it'll be able to terminate
immediately once the spawn succeeds, without having to linger around
for the lifetime as a shell process for proxying the exit code. When
process worker thread running in the parent sees the child die, it's
given a handle to the new child, to replace it in the process table.
- execve() and posix_spawn() on Windows will now provide CreateProcess
an explicit handle list. This allows us to remove handle locks which
enables better fork/spawn concurrency, with seriously correct thread
safety. Other codebases like Go use the same technique. On the other
hand fork() still favors the conventional WIN32 inheritence approach
which can be a little bit messy, but is *controlled* by guaranteeing
perfectly clean slates at both the spawning and execution boundaries
- sigset_t is now 64 bits. Having it be 128 bits was a mistake because
there's no reason to use that and it's only supported by FreeBSD. By
using the system word size, signal mask manipulation on Windows goes
very fast. Furthermore @asyncsignalsafe funcs have been rewritten on
Windows to take advantage of signal masking, now that it's much more
pleasant to use.
- All the overlapped i/o code on Windows has been rewritten for pretty
good signal and cancelation safety. We're now able to ensure overlap
data structures are cleaned up so long as you don't longjmp() out of
out of a signal handler that interrupted an i/o operation. Latencies
are also improved thanks to the removal of lots of "busy wait" code.
Waits should be optimal for everything except poll(), which shall be
the last and final demon we slay in the win32 i/o horror show.
- getrusage() on Windows is now able to report RUSAGE_CHILDREN as well
as RUSAGE_SELF, thanks to aggregation in the process manager thread.
- 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 fixes Cosmopolitan so it has fewer opinions about compiler
warnings. The whole repository had to be cleaned up to be buildable in
-Werror -Wall mode. This lets us benefit from things like strict const
checking. Some actual bugs might have been caught too.
- Fix handling of precision in hex float formatting
- Enhance the cocmd interpreter for system() and popen()
- Manually ran the Lua unit tests, which are now passing
- Let stdio i/o operations happen when file is in error state
- We're now saving and restoring xmm in ftrace out of paranoia
- Work towards improving non-optimized build support
- Introduce MODE=zero which is -O0 without ASAN/UBSAN
- Use system GCC when ~/.cosmo.mk has USE_SYSTEM_TOOLCHAIN=1
- Have package.com check .privileged code doesn't call non-privileged
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.
- ASAN memory morgue is now lockless
- Make C11 atomics header more portable
- Rewrote pthread keys support to be lockless
- Simplify Python's unicode table unpacking code
- Make crash report write(2) closer to being atomic
- Make it possible to strace/ftrace a single thread
- ASAN now checks nul-terminated strings fast and properly
- Windows fork() now restores TLS memory of calling thread
The organization of the source files is now much more rational.
Old experiments that didn't work out are now deleted. Naming of
things like files is now more intuitive.
We had previously not enabled TLS in MODE=tiny in order to keep the
smallest example programs (e.g. life.com) just 16kb in size. But it
was error prone doing that, so now we just always enable it because
this change uses hacks to ensure it won't increase life.com's size.
This change also fixes a bug on NetBSD, where signal handlers would
break thread local storage if SA_SIGINFO was being used. This looks
like it might be a bug in NetBSD, but it's got a simple workaround.
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.
- Introduce __assert_disable global
- Improve strsignal() thread safety
- Make system call tracing thread safe
- Fix SO_RCVTIMEO / SO_SNDTIMEO on Windows
- Refactor DescribeFoo() functions into one place
- Fix fork() on Windows when TLS and MAP_STACK exist
- Round upwards in setsockopt(SO_RCVTIMEO) on Windows
- Disable futexes on OpenBSD which seem extremely broken
- Implement a better kludge for monotonic time on Windows
- Finish cleaning up the stdio unlocked APIs
- Make __cxa_finalize() properly thread safe
- Don't log locks if threads aren't being used
- Add some more mutex guards to places using _mmi
- Specific lock names now appear in the --ftrace logs
- Fix mkdeps.com generating invalid Makefiles sometimes
- Simplify and fix bugs in the test runner infrastructure
- Fix issue where sometimes some functions wouldn't be logged
The "no modify self" variant of Actually Portable Executable is now
supported on all platforms. If you use `$(APE_NO_MODIFY_SELF)` then
ld.bfd will embed a 4096 byte ELF binary and a 4096 byte Macho file
which are installed on the fly to ${TMPDIR:-/tmp}, which enables us
launch the executable, without needing to copy the whole executable
To prevent it from copying a tiny executable to your temp directory
you need to install the `ape` command (renamed from ape-loader), to
a system path. For example:
# FreeBSD / NetBSD / OpenBSD
make -j8 o//ape/ape
cp o//ape/ape /usr/bin/ape
# Mac OS
# make -j8 o//ape/ape.macho
curl https://justine.lol/ape.macho >/usr/bin/ape
chmod +x /usr/bin/ape
On Linux you can get even more performance with the new binfmt_misc
support which makes launching non-modifying APE binaries as fast as
launching ELF executables. Running the following command:
# Linux
ape/apeinstall.sh
Will copy APE loader to /usr/bin/ape and register with binfmt_misc
Lastly, this change also fixes a really interesting race condition
with OpenBSD thread joining.
- Document more compiler flags
- Expose new __print_maps() api
- Better overflow checking in mmap()
- Improve the shell example somewhat
- Fix minor runtime bugs regarding stacks
- Make kill() on fork()+execve()'d children work
- Support CLONE_CHILD_CLEARTID for proper joining
- Fix recent possible deadlock regression with --ftrace
- 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 change makes further effort towards improving our poll()
implementation on the New Technology. The stdin worker didn't work out
so well for Python so it's not being used for now. System call tracing
with the --strace flag should now be less noisy now on Windows unless
you modify the strace.internal.h defines to turn on some optional ones
that are most useful for debugging the system call wrappers.
- Introduce fast spinlock API
- Double rand64() perf w/ spinlock
- Improve raise() on New Technology
- Support gettid() across platforms
- Implement SA_NODEFER on New Technology
- Move the lock intrinsics into LIBC_INTRIN
- Make SIGTRAP recoverable on New Technology
- Block SIGCHLD in wait4() on New Technology
- Add threading prototypes for XNU and FreeBSD
- Rewrite abort() fixing its minor bugs on XNU/NT
- Shave down a lot of the content in libc/bits/bits.h
- Let signal handlers modify CPU registers on New Technology
This change fixes minor bugs and adds a feature, which lets us store the
ELF symbol table, inside the ZIP directory. We use the path /zip/.symtab
which can be safely removed using a zip editing tool, to make the binary
smaller after compilation. This supplements the existing method of using
a separate .com.dbg file, which is still supported. The intent is people
don't always know that it's a good idea to download the debug file. It's
not great having someone's first experience be a crash report, that only
has numbers rather than symbols. This will help fix that!
- Fix build flakes
- Polyfill SIGWINCH on Windows
- Fix an execve issue on Windows
- Make strerror show more information
- Improve cmd.exe setup/teardown on Windows
- Support bracketed paste mode in Blinkenlights
- Show keyboard shortcuts in Blinkenlights status bar
- Fixed copy_file_range() and copyfile() w/ zip filesystem
- Size optimize GetDosArgv() to keep life.com 12kb in size
- Improve Blinkenlights ability to load weird ELF executables
- Fix program_executable_name and add GetInterpreterExecutableName
- Make Python in tiny mode fail better if docstrings are requested
- Update Python test exclusions in tiny* modes such as tinylinux
- Add bulletproof unbreakable kprintf() troubleshooting function
- Remove "oldskool" keyword from ape.S for virus scanners
- Fix issue that caused backtraces to not print sometimes
- Improve Blinkenlights serial uart character i/o
- Make clock_gettime() not clobber errno on xnu
- Improve sha256 cpuid check for old computers
- Integrate some bestline linenoise fixes
- Show runit process names better in htop
- Remove SIGPIPE from ShowCrashReports()
- Make realpath() not clobber errno
- Avoid attaching GDB on non-Linux
- Improve img.com example
This commit makes numerous refinements to cosmopolitan memory handling.
The default stack size has been reduced from 2mb to 128kb. A new macro
is now provided so you can easily reconfigure the stack size to be any
value you want. Work around the breaking change by adding to your main:
STATIC_STACK_SIZE(0x00200000); // 2mb stack
If you're not sure how much stack you need, then you can use:
STATIC_YOINK("stack_usage_logging");
After which you can `sort -nr o/$MODE/stack.log`. Based on the unit test
suite, nothing in the Cosmopolitan repository (except for Python) needs
a stack size greater than 30kb. There are also new macros for detecting
the size and address of the stack at runtime, e.g. GetStackAddr(). We
also now support sigaltstack() so if you want to see nice looking crash
reports whenever a stack overflow happens, you can put this in main():
ShowCrashReports();
Under `make MODE=dbg` and `make MODE=asan` the unit testing framework
will now automatically print backtraces of memory allocations when
things like memory leaks happen. Bugs are now fixed in ASAN global
variable overrun detection. The memtrack and asan runtimes also handle
edge cases now. The new tools helped to identify a few memory leaks,
which are fixed by this change.
This change should fix an issue reported in #288 with ARG_MAX limits.
Fixing this doubled the performance of MKDEPS.COM and AR.COM yet again.
PYOBJ.COM was failing when statically analyzing _pyio.py in MODE=dbg
because co_consts contained a big number, which dirtied the interpreter
exception state. We now do comprehensive error checking w/ Python API.
The -DSTACK_FRAME_UNLIMITED CPPFLAG has been removed from DES since its
self test function has been fixed to use heap memory rather than making
aggressive use of the stack.
This change also fixes a regression with function tracing (the --ftrace
flag a.k.a. ftrace_install() a.k.a. cosmo.ftrace) in ASAN build modes.
Lastly, the _tracemalloc module should now always be available for use
in MODE=dbg.
- Python static hello world now 1.8mb
- Python static fully loaded now 10mb
- Python HTTPS client now uses MbedTLS
- Python REPL now completes import stmts
- Increase stack size for Python for now
- Begin synthesizing posixpath and ntpath
- Restore Python \N{UNICODE NAME} support
- Restore Python NFKD symbol normalization
- Add optimized code path for Intel SHA-NI
- Get more Python unit tests passing faster
- Get Python help() pagination working on NT
- Python hashlib now supports MbedTLS PBKDF2
- Make memcpy/memmove/memcmp/bcmp/etc. faster
- Add Mersenne Twister and Vigna to LIBC_RAND
- Provide privileged __printf() for error code
- Fix zipos opendir() so that it reports ENOTDIR
- Add basic chmod() implementation for Windows NT
- Add Cosmo's best functions to Python cosmo module
- Pin function trace indent depth to that of caller
- Show memory diagram on invalid access in MODE=dbg
- Differentiate stack overflow on crash in MODE=dbg
- Add stb_truetype and tools for analyzing font files
- Upgrade to UNICODE 13 and reduce its binary footprint
- COMPILE.COM now logs resource usage of build commands
- Start implementing basic poll() support on bare metal
- Set getauxval(AT_EXECFN) to GetModuleFileName() on NT
- Add descriptions to strerror() in non-TINY build modes
- Add COUNTBRANCH() macro to help with micro-optimizations
- Make error / backtrace / asan / memory code more unbreakable
- Add fast perfect C implementation of μ-Law and a-Law audio codecs
- Make strtol() functions consistent with other libc implementations
- Improve Linenoise implementation (see also github.com/jart/bestline)
- COMPILE.COM now suppresses stdout/stderr of successful build commands
This change reinvents all the GNU Readline features I discovered that I
couldn't live without, e.g. UTF-8, CTRL-R search and CTRL-Y yanking. It
now feels just as good in terms of user interface from the subconscious
workflow perspective. It's real nice to finally have an embeddable line
reader that's actually good with a 30 kb footprint and a bsd-2 license.
This change adds a directory to the examples folder, explaining how the
new Python compiler may be used. Some of the bugs with Python binaries
have been addressed but overall it's still a work in progress.
This change makes SSL virtual hosting possible. You can now load
multiple certificates for multiple domains and redbean will just
figure out which one to use, even if you only have 1 ip address.
You can also use a jumbo certificate that lists all your domains
in the the subject alternative names.
This change also makes performance improvements to MbedTLS. Here
are some benchmarks vs. cc1920749e
BEFORE AFTER (microsecs)
suite_ssl.com 2512881 191738 13.11x faster
suite_pkparse.com 36291 3295 11.01x faster
suite_x509parse.com 854669 120293 7.10x faster
suite_pkwrite.com 6549 1265 5.18x faster
suite_ecdsa.com 53347 18778 2.84x faster
suite_pk.com 49051 18717 2.62x faster
suite_ecdh.com 19535 9502 2.06x faster
suite_shax.com 15848 7965 1.99x faster
suite_rsa.com 353257 184828 1.91x faster
suite_x509write.com 162646 85733 1.90x faster
suite_ecp.com 20503 11050 1.86x faster
suite_hmac_drbg.no_reseed.com 19528 11417 1.71x faster
suite_hmac_drbg.nopr.com 12460 8010 1.56x faster
suite_mpi.com 687124 442661 1.55x faster
suite_hmac_drbg.pr.com 11890 7752 1.53x faster
There aren't any special tricks to the performance imporvements.
It's mostly due to code cleanup, assembly and intel instructions
like mulx, adox, and adcx.
This change boosts SSL handshake performance from 2,627 to ~10,000 per
second which is the same level of performance as NGINX at establishing
secure connections. That's impressive if we consider that redbean is a
forking frontend application server. This was accomplished by:
1. Enabling either SSL session caching or SSL tickets. We choose to
use tickets since they reduce network round trips too and that's
a more important metric than wrk'ing localhost.
2. Fixing mbedtls_mpi_sub_abs() which is the most frequently called
function. It's called about 12,000 times during an SSL handshake
since it's the basis of most arithmetic operations like addition
and for some strange reason it was designed to make two needless
copies in addition to calling malloc and free. That's now fixed.
3. Improving TLS output buffering during the SSL handshake only, so
that only a single is write and read system call is needed until
blocking on the ping pong.
redbean will now do a better job wiping sensitive memory from a child
process as soon as it's not needed. The nice thing about fork is it's
much faster than reverse proxying so the goal is to use the different
address spaces along with setuid() to minimize the risk that a server
key will be compromised in the event that application code is hacked.
