Now that the kernel supports startup-time constructors, we were first
doing slab_alloc_init(), and then the constructors ran later on,
zeroing out the freelist pointers.
This meant that all slab allocators thought they were completelty
exhausted and forwarded all requests to kmalloc() instead.
It's now possible to block until another thread in the same process has
exited. We can also retrieve its exit value, which is whatever value it
passed to pthread_exit(). :^)
While executing in the kernel, a thread can acquire various resources
that need cleanup, such as locks and references to RefCounted objects.
This cleanup normally happens on the exit path, such as in destructors
for various RAII guards. But we weren't calling those exit paths when
killing threads that have been executing in the kernel, such as threads
blocked on reading or sleeping, thus causing leaks.
This commit changes how killing threads works. Now, instead of killing
a thread directly, one is supposed to call thread->set_should_die(),
which will unblock it and make it unwind the stack if it is blocked
in the kernel. Then, just before returning to the userspace, the thread
will automatically die.
This patch adds pthread_create() and pthread_exit(), which currently
simply wrap our existing create_thread() and exit_thread() syscalls.
LibThread is also ported to using LibPthread.
Also added an option in the run script to force PIO operation mode with
the IDE controller.
In addition, we're no longer limited to PIIX3 and PIIX4 chipsets for DMA
Now that the SystemMonitor queries which open files can be read and written to,
having can_read()/can_write() unconditionally call ASSERT_NOT_REACHED() leads
to system crashes when inspecting the WindowServer.
Instead, just return true from can_read()/can_write() (indicating that the
read()/write() syscalls should not block) and return -EINVAL when trying to
actually read from or write to these devices.
Added a script to build QEMU from source as part of the Toolchain.
The script content could be in BuildIt.sh but has been put in
a seperate file to make the build optional.
Added PATH=$PATH to sudo calls to hand over the Toolchain's PATH
setup by UseIt.sh. This enabled the script's to use the QEMU
contained in the SerenityOS toolchain.
Deleted old documentation in Meta and replaced it by a new
documentation in the Toolchain folder.
The Launcher's functionality has been replaced by the app shortcuts in
the system menu.
There were various window management hacks to ensure that the launcher
stayed below all other windows while also being movable, etc.
Some syscalls have to pass parameters through a struct, since we can
only fit 3 parameters with our calling convention.
This patch makes use of C++ structured binding to clean up the places
where we expand those parameters structs into local variables.
POSIX's openat() is very similar to open(), except you also provide a
file descriptor referring to a directory from which relative paths
should be resolved.
Passing it the magical fd number AT_FDCWD means "resolve from current
directory" (which is indeed also what open() normally does.)
This fixes libarchive's bsdtar, since it was trying to do something
extremely wrong in the absence of openat() support. The issue has
recently been fixed upstream in libarchive:
https://github.com/libarchive/libarchive/issues/1239
However, we should have openat() support anyway, so I went ahead and
implemented it. :^)
Fixes#748.
Don't wait for someone to wait() on a dead process before releasing its
TTY object. This fixes the child process death detection used by the
Terminal application, which relies on getting an EOF on the master PTY
in order to know it's time to wait() on the child process. :^)
The SysV ABI says that the DF flag should be clear on function entry.
That means we have to clear it when jumping into the kernel from some
random userspace context.
Instead of the big ugly switch statement, build a lookup table using
the syscall enumeration macro.
This greatly simplifies the syscall implementation. :^)
Now the userspace page allocator will search through physical regions,
and stop the search as it finds an available page.
Also remove an "address of" sign since we don't need that when
counting size of physical regions
This was too noisy and important-sounding, when it doesn't really
matter that much. It's not the end of the world if symbolication fails
for one reason or another.
It's not safe to use a raw pointer for Process::m_tty. A pseudoterminal
pair will disappear when file descriptors are closed, and we'd end up
looking dangly. Just use a RefPtr.
Scheduling priority is now set at the thread level instead of at the
process level.
This is a step towards allowing processes to set different priorities
for threads. There's no userspace API for that yet, since only the main
thread's priority is affected by sched_setparam().
Oops, we were creating these and then throwing them away. They will
get instantiated a bit later, when we bring up the mounts in /etc/fstab
from userspace.
This allows us to get rid of all the custom 64-bit division helpers.
I wanted to do this ages ago but couldn't get it working. Turns out it
was unstable due to libgcc using the regular ABI and the kernel being
built with -mregparm=3.
Now that we build the kernel with regular calls, we can just link with
libgcc and get this stuff for free. :^)
Files opened with O_DIRECT will now bypass the disk cache in read/write
operations (though metadata operations will still hit the disk cache.)
This will allow us to test actual disk performance instead of testing
disk *cache* performance, if that's what we want. :^)
There's room for improvment here, we're very aggressively flushing any
dirty cache entries for the specific block before reading/writing that
block. This is done by walking the entire cache, which may be slow.
This helps aid debugging of issues such as #695, where the bridge chip
that controls IDE is NOT a PIIX3/4 compatible controller. Instead of
just hanging when the DMA registers can't be accessed, the system will
inform the user that no valid IDE controller has been found. In this
case, the system will not attempt to initialise the DMA registers and
instead use PIO mode.
Don't keep Inodes around in memory forever after we've interacted with
them once. This is a slight performance pessimization when accessing
the same file repeatedly, but closing it for a while in between.
Longer term we should find a way to keep a limited number of unused
Inodes cached, whichever ones we think are likely to be used again.
Move the kernel image to the 1 MB physical mark. This prevents it from
colliding with stuff like the VGA memory. This was causing us to end
up with the BIOS screen contents sneaking into kernel memory sometimes.
This patch also bumps the kmalloc heap size from 1 MB to 3 MB. It's not
the perfect permanent solution (obviously) but it should get the OOM
monkey off our backs for a while.
dispatch_signal() expected a RegisterDump on the kernel stack. However
in certain cases, like just after a clone, this was not the case and
dispatch_signal() would instead write to an incorrect user stack pointer.
We now use the threads TSS in situations where the RegisterDump may not
be valid, fixing the issue.
