* Better refcounting
Cribbed from [Rust Arc][1] and the [Boost docs][2]:
"""
Increasing the reference counter can always be done with
memory_order_relaxed: New references to an object can only be formed
from an existing reference, and passing an existing reference from one
thread to another must already provide any required synchronization.
It is important to enforce any possible access to the object in one
thread (through an existing reference) to happen before deleting the
object in a different thread. This is achieved by a "release" operation
after dropping a reference (any access to the object through this
reference must obviously happened before), and an "acquire" operation
before deleting the object.
It would be possible to use memory_order_acq_rel for the fetch_sub
operation, but this results in unneeded "acquire" operations when the
reference counter does not yet reach zero and may impose a performance
penalty.
"""
[1] https://moshg.github.io/rust-std-ja/src/alloc/arc.rs.html
[2] https://www.boost.org/doc/libs/1_55_0/doc/html/atomic/usage_examples.html
* Make ZiposHandle's pos atomic
Implements a somewhat stronger guarantee than POSIX specifies: reads and
seeks are atomic. They may be arbitrarily reordered between threads, but
each one happens all the way and leaves the fd in a consistent state.
This is achieved by "locking" pos in __zipos_read by storing SIZE_MAX to
pos during the operation, so only one can be in-flight at a time. Seeks,
on the other hand, just update pos in one go, and rerun if it changed in
the meantime.
I used `LIKELY` / `UNLIKELY` to pessimize the concurrent case; hopefully
that buys back some performance.
On aarch64 hosts, MODE= is changed to MODE=aarch64, so o// targets don't
work. So On aarch64, get apelink.com out of o/aarch64/. Also prepend ape
when calling it. And finally, fetch with curl when wget isn't installed.
* Implement __zipos_dup
Makes ZiposHandle reference-counted by an `rc` field in a union with its
freelist `next` pointer. The functions `__zipos_free` and `__zipos_keep`
function as incref/decref for it. Adds `__zipos_postdup` to fix metadata
on file descriptors after dup-like operations, and adds zipos support to
`sys_dup_nt` + `sys_close_nt`.
* Remove noop __zipos_postdup
rc is never a zipos file because it is always a previously unused file
descriptor. fd is never a zipos file because that case has been handled
above by __zipos_fcntl.
Includes additional fixes from main repo's unmerged PRs:
- quickjs#132: Fix undefined behavior: shift 32 bits for uint32_t in bf_set_ui
- quickjs#171: Fix locale-aware representation of hours in Date class
- quickjs#182: Fix stack overflow in CVE-2023-31922
On UNIX if dup2(newfd) was a ZipOS file descriptor, then its resources
weren't being released, and the newly created file descriptor would be
mistaken for ZipOS due to its memory not being cleared. On Windows, an
issue also existed relating to newfd resources not being released.
I'm no longer able to reproduce the PE import table corruption that was
previously observed. Since this optimization shaves up to 64kb off each
fat APE binary we build, it's worth turning this back on again, to wait
and see if something breaks, and if so, fix it. At least until the next
release is shipped.
See #965
This change upgrades to superconfigure z0.0.23 which fixes an issue
where the compiler had harmless /home/... paths baked-in, which are
normally only present in the build environment, and usually skipped
over. Sadly on MacOS calling fstatat() on these paths would lead to
cloud file system ops that caused system calls to take a long time.
That's problematic, since cosmocc needs to be a 100% local command.
This change fixes a regression that happened some time ago when building
for AARCH64 using the vendored toolchain rather than cosmocc. The errors
that would show up `Relocations in generic ELF (EM: 62)` have been fixed
Munging of paths passed inside the system() interpreter command is no
longer supported. You have to pass your paths to posix_spawn() or the
execve() family of functions if you want them to be munged. The first
three characters must match `^/[a-z]/` in which case, it'll be turned
into a DOS-style drive path with backslashes.
- Introduce MAP_JIT which is zero on other platforms
- Invent __jit_begin() and __jit_end() which wrap Apple's APIs
- Runtime dispatch to sys_icache_invalidate() in __clear_cache()
- Use good ELF technique in cosmo_dlopen()
- Make strerror() conform more to other libc impls
- Introduce __clear_cache() and use it in cosmo_dlopen()
- Remove libc/fmt/fmt.h header (trying to kill off LIBC_FMT)
It turns out my earlier commit ddc08dc974 caused a build with
MODE=aarch64 to fail. The commit changed deathstar.c to link
in code to support a VGA console, but this is not implemented
yet for AArch64. Thanks to @ahgamut for spotting this issue.
The test was failing if the process's umask happened to be
0077, for example. The file `foo` was then created with a
file mode of 0100600, rather than the expected 0100644.
After hearing horror stories from a trusted colleague, I don't think
this is the kind of API we want to be supporting. Also SQLite wisdom
regarding fdatasync() has been added to the documentation.
