This was a premature optimization from the early days of SerenityOS.
The eternal heap was a simple bump pointer allocator over a static
byte array. My original idea was to avoid heap fragmentation and improve
data locality, but both ideas were rooted in cargo culting, not data.
We would reserve 4 MiB at boot and only ended up using ~256 KiB, wasting
the rest.
This patch replaces all kmalloc_eternal() usage by regular kmalloc().
This is an interface to downcast(), which degrades errors into runtime
errors, and allows seemingly-correct-but-not-quite constructs like the
following to appear to compile, but fail at runtime:
Variant<NonnullRefPtr<T>, U> foo = ...;
Variant<RefPtr<T>, U> bar = foo;
The expectation here is that `foo` is converted to a RefPtr<T> if it
contains one, and remains a U otherwise, but in reality, the
NonnullRefPtr<T> variant is simply dropped on the floor, and the
resulting variant becomes invalid, failing the assertion in downcast().
This commit adds a Variant<Ts...>(Variant<NewTs...>) constructor that
ensures that no alternative can be left out at compiletime, for the
users that were using this interface for merely increasing the number of
alternatives (for instance, LibSQL's Value class).
This is a raffinement of 49cbd4dcca.
Previously, the container was scanned to compute the size in the unhappy
path. Now, using `all_of` happy and unhappy path should be fast.
In order to reduce our reliance on __builtin_{ffs, clz, ctz, popcount},
this commit removes all calls to these functions and replaces them with
the equivalent functions in AK/BuiltinWrappers.h.
While watching Andreas' most recent video, I noticed that this function
only worked with 32 bit values, but was a serious performance
bottleneck for the kernel. As such, I reworked it to use `size_t`, so
it now can switch to 64-bit sweeps on 64-bit platforms. This caused
test-js to go from 12.5 seconds hot to 11.5 seconds hot on my machine
when running on KVM x86_64.
The goal of this file is to enable C++ overloaded functions for
standard builtin functions that we use. It contains fallback
implementations for systems that do not have the builtins available.
Before, if we couldn't read enough data out of the buffer, we would re-
fill the buffer and recursively call read(), which in turn reads data
from the buffer into the resliced target span. This incurs very
intensive superflous memmove's when large chunks of data are read from
a buffered stream.
This commit changes the behavior so that when we exhaust the buffer, we
first read any necessary additional data directly into the target, then
fill up the buffer again. Effectively, this results in drastically
reduced overhead from Buffered when reading large contiguous chunks.
Of course, Buffered is designed to speed up data access patterns with
small frequent reads, but it's nice to be able to combine both access
patterns on one stream without penalties either way.
The final performance gain is about an additional 80% of abench decoding
speed.
This unbreaks the /var/run/utmp system which starts out as an empty
string, and is then turned into an object by the first update.
This isn't necessarily the best way for this to work, but it's how
it used to work, so this just fixes the regression for now.
This fixes at least half of our LibC includes in the kernel. The source
of truth for errno codes and their description strings now lives in
Kernel/API/POSIX/errno.h as an enumeration, which LibC includes.
This is the same strategy that LLVM's compiler-rt uses to make sure that
each UBSAN error is only reported once, when UBSAN is *not* deadly.
Otherwise, each time we head through a UB codepath, we will log the same
error over and over. That behavior just adds noise to the logs and makes
it nearly impossible to run binaires that have some common code path
with flagged UB in them.
compiler-rt goes the extra step to make sure the "clear" action is
atomic, but we don't really have that many multi-threaded apps gettting
tested with UBSAN yet, so we can add that later.
This will allow us to avoid some potentially expensive type conversion
during lookup, like form String to StringView, which would allocate
memory otherwise.
Instead of checking __linux__ macro directly, the code should check if
this macro is defined. This is already done correctly a couple of lines
above.
I ran into this when trying to build libjs-test262 on MacOS where I got
the following error message
error: "__linux__" is not defined, evaluates to 0 [-Werror=undef]
This is a convenience template that implements reference count
forwarding. This means that an object forwards ref() and unref() to
another object.
We can use this when two ref-counted objects need to keep each other
alive. This situation poses two problems:
- Using 2x RefPtr would cause a ref cycle and leak both objects.
- Using 2x WeakPtr would allow one of them to be destroyed early.
With RefCountForwarder, only one of the objects has a ref count. The
object with the ref count points to the forwarding object by using a
non-counting smart pointer (OwnPtr or NonnullOwnPtr). Thus, both objects
are kept alive by the same ref count, and they can safely point to each
other without worrying about disjoint lifetimes.
Note that the return type for the non-const method error() changed. This
is most likely an accident, hidden by the fact that ErrorType typically
is Error.
This makes it an error to not do something with a returned smart
pointer, which should help prevent mistakes. In cases where you do need
to ignore the value, casting to void will placate the compiler.
I did have to add comments to disable clang-format on a couple of lines,
where it wanted to format the code like this:
```c++
private : NonnullRefPtr() = delete;
```
Previously, Vector::extend for a moved vector would move the other
vector into this vector if this vector was empty, thereby throwing away
existing allocated capacity. Therefore, this commit allows the move to
only happen if this vector's capacity is too small to fit the other
vector. This will also alleviate bugs where callers relied on the
capacity to never shrink with calls to unchecked_append, extend and the
like.
This mirrors the existence of append() for data pointers and is very
useful when the program needs to have a guarantee of no allocations,
as is necessary for real-time audio.
During the build process on macOS, multiple versions of <unistd.h> were
being included (Apple's version and GCC's version). It appears that
all other places #include the version from GCC, but in Platform.h the
Apple header was being used. GCC's <unistd.h> is wrapped in
`extern "C"`, while Apple's is not. This causes a conflicting
declaration, so we need to wrap the #include with extern "C".
Issue has been observed on macOS Mojave.
See https://github.com/microsoft/vcpkg/issues/11320 for a similar issue.
This creates an error that contains the name of the syscall that failed.
This allows error handlers to print out the name of the call if they
want to. :^)
