If we pass a null path to these syscall wrappers, just return EFAULT
directly from the wrapper instead of segfaulting by calling strlen().
This is a compromise, since we now have to pass the path length to the
kernel, so we can't rely on the kernel to tell us that the path is at
a bad memory address.
I though it would be nice to also show the style that the browser uses
to display an element.
In order to do that, in place of the styles table I've put a tab widget,
with tabs for both element and computed element styles.
LibHTML will now use the palette colors for the default document background and
the text. As always, a page can override this default styling with CSS if it
really wants a specific color or style.
Fixes https://github.com/SerenityOS/serenity/issues/963
Previously we would consider anything in the large padded area around
each item to also be part of the item for mouse event purposes.
This didn't feel right when rubberbanding, so this patch factors out
the per-item rect computation into a get_item_rects() helper which can
then be used by the various functions that need it.
Separate some responsibilities:
ELFDynamicLoader is responsible for loading elf binaries from disk and
performing relocations, calling init functions, and eventually calling
finalizer functions.
ELFDynamicObject is a helper class to parse the .dynamic section of an
elf binary, or the table of Elf32_Dyn entries at the _DYNAMIC symbol.
ELFDynamicObject now owns the helper classes for Relocations, Symbols,
Sections and the like that ELFDynamicLoader will use to perform
relocations and symbol lookup.
Because these new helpers are constructed from offsets into the .dynamic
section within the loaded .data section of the binary, we don't need the
ELFImage for nearly as much of the loading processes as we did before.
Therefore we can remove most of the extra DynamicXXX classes and just
keep the one that lets us find the location of _DYNAMIC in the new ELF.
And finally, since we changed the name of the class that dlopen/dlsym
care about, we need to compile/link and use the new ELFDynamicLoader
class in LibC.
This fixes an issue in SystemMonitor where old data would linger in the
table views after selecting a process owned by another user.
Since we can no longer read /proc/PID/* unless PID belongs to us,
we will now present empty views for these processes. :^)
When selecting an element in the browser's DOM inspector, we now also
show the resolved CSS properties (and their values) for that element.
Since the inspector was growing a bit more complex, I moved it out of
the "show inspector" action callback and into its own class.
In the future, we will probably want to migrate the inspector down to
LibHTML to make it accessible to other clients of the library, but for
now we can keep working on it inside Browser. :^)
This code never worked, as was never used for anything. We can build
a much better SHM implementation on top of TmpFS or similar when we
get to the point when we need one.
For dynamic loading, the symbol bind of a symbol actually doesn't
matter. We could do what old glibc did and try to find a strong
symbol for any weak definitions, but the ELF spec doesn't require
it and they changed that a few years ago anyway. So, moot point. :)
We were not sending the ID of the window that was listening for window
management (WM) events along with the WM messages. They only included
the "target" window's ID.
Since the taskbar's single window had the first window ID for its own
connection to the WindowServer, it meant that it would only receive
WM events for the first window ID in other processes as well.
This broke when I ported WindowServer to LibIPC.
Fix this by including the WM listener ID in all WM messages, and since
we're here anyway, get rid of a bunch of unnecessary indirection where
we were passing WM events through the WindowServer event loop before
sending them to the listener.
ELFDynamicObject::load looks a lot better with all the steps
re-organized into helpers.
Add plt_trampoline.S to handle PLT fixups for lazy loading.
Add the needed trampoline-trampolines in ELFDynamicObject to get to
the proper relocations and to return the symbol back to the assembly
method to call into from the PLT once we return back to user code.
We weren't calling the method here before because it was ill-formed.
No start files meant that we got the front half of the init section but
not the back half (no 'ret' in _init!). Now that we have the proper
crtbeginS and crtendS files from libgcc to help us out, we can assume
that DSOs will have the proper _init method defined.
The scrollbar width must be factored in, and one too many
m_line_spacing were being factored into the height. These caused an
initial terminal opening in 80x25 to get resized right away and
shrunk down to 77x24.
This patch also adds some missing relocation defines to exec_elf.h,
and a few helper classes/methods to ELFImage so that we can use it
for our dynamically loaded libs and not just main program images from
the kernel :)
This patch introduces a syscall:
int set_thread_boost(int tid, int amount)
You can use this to add a permanent boost value to the effective thread
priority of any thread with your UID (or any thread in the system if
you are the superuser.)
This is quite crude, but opens up some interesting opportunities. :^)
Threads now have numeric priorities with a base priority in the 1-99
range.
Whenever a runnable thread is *not* scheduled, its effective priority
is incremented by 1. This is tracked in Thread::m_extra_priority.
The effective priority of a thread is m_priority + m_extra_priority.
When a runnable thread *is* scheduled, its m_extra_priority is reset to
zero and the effective priority returns to base.
This means that lower-priority threads will always eventually get
scheduled to run, once its effective priority becomes high enough to
exceed the base priority of threads "above" it.
The previous values for ThreadPriority (Low, Normal and High) are now
replaced as follows:
Low -> 10
Normal -> 30
High -> 50
In other words, it will take 20 ticks for a "Low" priority thread to
get to "Normal" effective priority, and another 20 to reach "High".
This is not perfect, and I've used some quite naive data structures,
but I think the mechanism will allow us to build various new and
interesting optimizations, and we can figure out better data structures
later on. :^)
This removes a bunch of JsonValue copying from the hot path in thread
statistics fetching.
Also pre-size the thread statistics vector since we know the final size
up front. :^)
Instead of using ByteBuffer (which always malloc() their storage) for
IPC message encoding, we now use a Vector<u8, 1024>, which means that
messages smaller than 1 KB avoid heap allocation entirely.
The widget layout system currently works by having layouts size the
children of a widgets. The children then get resize events, giving them
a chance to lay out their own children, etc.
In keeping with this, we need to handle the resize event before calling
do_layout() in a widget, since the resize event handler may do things
that end up affecting the layout, but layout should not affect the
resize event since the event comes from the widget parent, not itself.
LibCore timers now have a TimerShouldFireWhenNotVisible flag which is
set to "No" by default.
If "No", the timer will not be fired by the event loop if it's within
a CObject tree whose nearest GWindow ancestor is currently not visible
for timer purposes. (Specificially, this means that the window is
either minimized or fully occluded, and so does not want to fire timers
just to update the UI.)
This is another nice step towards a calm and serene operating system.
This is memory that's loaded from an inode (file) but not modified in
memory, so still identical to what's on disk. This kind of memory can
be freed and reloaded transparently from disk if needed.
Dirty private memory is all memory in non-inode-backed mappings that's
process-private, meaning it's not shared with any other process.
This patch exposes that number via SystemMonitor, giving us an idea of
how much memory each process is responsible for all on its own.
Palette is now a value wrapper around a NonnullRefPtr<PaletteImpl>.
A new function, set_color(ColorRole, Color) implements a simple
copy-on-write mechanism so that we're sharing the PaletteImpl in the
common case, but allowing you to create custom palettes if you like,
by getting a GWidget's palette, modifying it, and then assigning the
modified palette to the widget via GWidget::set_palette().
Use this to make PaintBrush show its palette colors once again.
Fixes#943.
Lock each directory before entering it so when using -j, the same
dependency isn't built more than once at a time.
This doesn't get full -j parallelism though, since one make child
will be sitting idle waiting for flock to receive its lock and
continue making (which should then do nothing since it will have
been built already). Unfortunately there's not much that can be
done to fix that since it can't proceed until its dependency is
built by another make process.
When filling in some missing part of a window (typically happens during
interactive window resize) we now use the ColorRole::Background from
the system theme palette instead of expecting the clients to send us
the same information when creating windows.
WindowServer now tracks whether windows are occluded (meaning that
they are completely covered by one or more opaque windows sitting above
them.) This state is communicated to the windows via WindowStateChanged
messages, which then allow GWindow to mark its backing store volatile.
This reduces the effective memory impact of windows that are not at all
visible to the user. Very cool. :^)
Compiling LibCore on macOS is needed if one wants to compile host tools (like IPCCompiler) on a non Linux host. These changes could be possibly reverted once "event loop" functionality and "base library" (Vector, String etc.) will be split in two separate libraries,
updating all relevant projects.
This functon will draw an ellipse which is intersecting the corners of
the rect given. It is a very naive implementation, taking 200 samples of
points around the ellipse, and drawing straight lines between each of
these points.
The ellipses look good enough to me though!
WindowServer will now send out a WindowStateChanged message to clients
when one of their windows is minimized.
This is then forwarded to the GWindow, which will try to mark its
underlying window backing store as volatile.
This allows the kernel to steal the memory used by minimized windows
in case it starts running low. Very cool! :^)
This patch implements a simple version of the futex (fast userspace
mutex) API in the kernel and uses it to make the pthread_cond_t API's
block instead of busily sched_yield().
An arbitrary userspace address is passed to the kernel as a "token"
that identifies the futex and you can then FUTEX_WAIT and FUTEX_WAKE
that specific userspace address.
FUTEX_WAIT corresponds to pthread_cond_wait() and FUTEX_WAKE is used
for pthread_cond_signal() and pthread_cond_broadcast().
I'm pretty sure I'm missing something in this implementation, but it's
hopefully okay for a start. :^)
Make sure we always line up the "hole" in the progress bar with the
right side of the paint rect. This fixes a 1px wide glitch seen when
using a darker system theme.
Instead of directly manipulating LDFLAGS, set LIB_DEPS in each
subdirectory Makefile listing the libraries needed for
building/linking such as "LIB_DEPS = Core GUI Draw IPC Core".
This adds each library as an -L and -l argument in LDFLAGS, but
also adds the library.a file as a link dependency on the current
$(PROGRAM). This causes the given library to be (re)built before
linking the current $(PROGRAM), but will also re-link any binaries
depending on that library when it is modified, when running make
from the root directory.
Also turn generator tools like IPCCompiler into dependencies on the
files they generate, so they are built on-demand when a particular
directory needs them.
This all allows the root Makefile to just list directories and not
care about the order, as all of the dependency tracking will figure
it out.
GApplication now has a palette. This palette contains all the system
theme colors by default, and is inherited by a new top-level GWidget.
New child widgets inherit their parents palette.
It is possible to override the GApplication palette, and the palette
of any GWidget.
The Palette object contains a bunch of colors, each corresponding to
a ColorRole. Each role has a convenience getter as well.
Each GWidget now has a background_role() and foreground_role(), which
are then looked up in their current palette when painting. This means
that you no longer alter the background color of a widget by setting
it directly, rather you alter either its background role, or the
widget's palette.
Color themes are loaded from .ini files in /res/themes/
The theme can be switched from the "Themes" section in the system menu.
The basic mechanism is that WindowServer broadcasts a SharedBuffer with
all of the color values of the current theme. Clients receive this with
the response to their initial WindowServer::Greet handshake.
When the theme is changed, WindowServer tells everyone by sending out
an UpdateSystemTheme message with a new SharedBuffer to use.
This does feel somewhat bloated somehow, but I'm sure we can iterate on
it over time and improve things.
To get one of the theme colors, use the Color(SystemColor) constructor:
painter.fill_rect(rect, SystemColor::HoverHighlight);
Some things don't work 100% right without a reboot. Specifically, when
constructing a GWidget, it will set its own background and foreground
colors based on the current SystemColor::Window and SystemColor::Text.
The widget is then stuck with these values, and they don't update on
system theme change, only on app restart.
All in all though, this is pretty cool. Merry Christmas! :^)
We were casting the pthread_mutex_t* instead of pthread_mutex_t::lock
to an Atomic<u32>. This still worked fine, since "lock" is the first
member of pthread_mutex_t.
This is a bit sad, but, with the Allocators as static globals their
destructors were running before some user code. Which doesn't really
make much sense, as none of the members of (at least the basic one) do
any real heavy lifting or have many resources to RAII.
To avoid the problem, just mmap the memory for the global arrays of
Allocators in __malloc_init and let the Kernel collect the memory when
we're done with the process.
These guys are all declared as globals, and their ASSERT_NOT_REACHED
in the destructor doesn't play nice with __cxa_atexit. As in, every
application will assert in __cxa_finalize if this assert isn't removed.
Implement __cxa_atexit and __cxa_finalize per the Itanium spec,
and convert stdlib's atexit and exit() to to call them instead of
a custom 'C-only' atexit implementation.
We always want to put crt0.o in the location where it can get picked
up by the i686-pc-serenity toolchain.
This feels a bit hackish but should get the build working again. :^)
Use simple stack cookies to try to provoke an assertion failure on
stack overflow.
This is far from perfect, since we use a constant cookie instead of
generating a random one on startup, but it can still help us catch
bugs, which is the primary concern right now. :^)
Allow everything to be built from the top level directory with just
'make', cleaned with 'make clean', and installed with 'make
install'. Also support these in any particular subdirectory.
Specifying 'make VERBOSE=1' will print each ld/g++/etc. command as
it runs.
Kernel and early host tools (IPCCompiler, etc.) are built as
object.host.o so that they don't conflict with other things built
with the cross-compiler.
These fields are intended to carry the real meat of a drag operation,
and the "text" is just for what we show on screen (alongside the cursor
during the actual drag.)
The data field is just a String for now, but in the future we should
make it something more flexible.
Now that Frame knows the visible viewport rect, it can easily ignore
repaint requests from e.g <blink> elements that are not currently
scrolled into view. :^)
When the visible viewport rect changes, we walk the layout tree and
check where each LayoutImage is in relation to the viewport rect.
Images outside have their bitmaps marked as volatile.
Note that the bitmaps are managed by ImageDecoder objects. If a bitmap
is purged by the kernel while volatile, we construct a new ImageDecoder
next time we need pixels for the image.
A client that only ever does synchronous IPC calls from its side would
never actually process incoming asynchronous messages since they would
arrive while waiting for a synchronous response and then end up sitting
forever in the "unhandled messages" queue.
We now always handle unhandled messages using a deferred invocation.
This fixes the bug where Audio.MenuApplet didn't learn that the muted
state changed in response to its own request to change it. :^)
Instead of implementing menu applets as their own thing, they are now
WSWindows of WSWindowType::MenuApplet.
This makes it much easier to work with them on the client side, since
you can just create a GWindow with the right type and you're in the
menubar doing applet stuff :^)
This patch adds a single "kernel info page" that is mappable read-only
by any process and contains the current time of day.
This is then used to implement a version of gettimeofday() that doesn't
have to make a syscall.
To protect against race condition issues, the info page also has a
serial number which is incremented whenever the kernel updates the
contents of the page. Make sure to verify that the serial number is the
same before and after reading the information you want from the page.
Currently only Ext2FS and TmpFS supports InodeWatchers. We now fail
with ENOTSUPP if watch_file() is called on e.g ProcFS.
This fixes an issue with FileManager chewing up all the CPU when /proc
was opened. Watchers don't keep the watched Inode open, and when they
close, the watcher FD will EOF.
Since nothing else kept /proc open in FileManager, the watchers created
for it would EOF immediately, causing a refresh over and over.
Fixes#879.
This logic is all taken care of by GAbstractColumnView now, so we can
simply delete GTreeView::context_menu_event(). :^)
Fixes an issue mentioned in #826
4KB gets pretty mmap/munmap heavy when downloading larger files,
so bump this a bit to reduce time spent in memory allocation.
This can be improved in various ways, but I'm not sure what the
best way forward is at the moment.
This makes GTreeView able to support multi-column models!
Only one column can be the "tree column", this is column 0 by default
but can be changed by overriding GModel::tree_column().
When the filesystem model is updated, it is rebuilt. This means dangling
indexes inside the TreeView metadata table will have old information and random
directories will toggle open. Clearing the table alleviates this issue.
The kernel now supports basic profiling of all the threads in a process
by calling profiling_enable(pid_t). You finish the profiling by calling
profiling_disable(pid_t).
This all works by recording thread stacks when the timer interrupt
fires and the current thread is in a process being profiled.
Note that symbolication is deferred until profiling_disable() to avoid
adding more noise than necessary to the profile.
A simple "/bin/profile" command is included here that can be used to
start/stop profiling like so:
$ profile 10 on
... wait ...
$ profile 10 off
After a profile has been recorded, it can be fetched in /proc/profile
There are various limits (or "bugs") on this mechanism at the moment:
- Only one process can be profiled at a time.
- We allocate 8MB for the samples, if you use more space, things will
not work, and probably break a bit.
- Things will probably fall apart if the profiled process dies during
profiling, or while extracing /proc/profile
When a GAction is activated by a menu, or by a toolbar button, you can
now use GAction::activator() to get a pointer to whomever activated it.
This can be used to implement context-specific behaviors in situations
where the same action is exposed through multiple paths.
This addresses an issue that was brought up in #826.
We now take advantage of SharedBuffers being purgeable memory by
setting the volatile flag on window back buffers while not painting
into them.
This means that one of the two backing stores used by each window
is purgeable+volatile most of the time, allowing the kernel to purge
it to recover memory if needed.
Note that this is only relevant when double-buffering is turned on,
but since that is the default, this does affect most apps. :^)
This patch makes SharedBuffer use a PurgeableVMObject as its underlying
memory object.
A new syscall is added to control the volatile flag of a SharedBuffer.
This patch exposes some fields about purgeable memory regions.
We now also show total purgeable volatile and non-volatile memory in
the big process table.
It's now possible to get purgeable memory by using mmap(MAP_PURGEABLE).
Purgeable memory has a "volatile" flag that can be set using madvise():
- madvise(..., MADV_SET_VOLATILE)
- madvise(..., MADV_SET_NONVOLATILE)
When in the "volatile" state, the kernel may take away the underlying
physical memory pages at any time, without notifying the owner.
This gives you a guilt discount when caching very large things. :^)
Setting a purgeable region to non-volatile will return whether or not
the memory has been taken away by the kernel while being volatile.
Basically, if madvise(..., MADV_SET_NONVOLATILE) returns 1, that means
the memory was purged while volatile, and whatever was in that piece
of memory needs to be reconstructed before use.
Using int was a mistake. This patch changes String, StringImpl,
StringView and StringBuilder to use size_t instead of int for lengths.
Obviously a lot of code needs to change as a result of this.
This bitmap is displayed alongside the dragged text underneath the
mouse cursor while dragging.
This will be a perfect fit for dragging e.g files around. :^)
This function returns the bitmap itself if it's already backed by a
SharedBuffer object, otherwise it creates a shareable copy of itself
and returns that.
This patch enables basic drag&drop between applications.
You initiate a drag by creating a GDragOperation object and calling
exec() on it. This creates a nested event loop in the calling program
that only returns once the drag operation has ended.
On the receiving side, you get a call to GWidget::drop_event() with
a GDropEvent containing information about the dropped data.
The only data passed right now is a piece of text that's also used
to visually indicate that a drag is happening (by showing the text in
a little box that follows the mouse cursor around.)
There are things to fix here, but we're off to a nice start. :^)
These should be the last thing needed to make SDL build with threads
support. I think we can survive just fine with stubs of these for now,
especially given that the kernel doesn't care super much about thread
priorities anyway.
This patch adds pthread_key_create() and pthread_{get,set}specific().
There's a maximum of 64 thread-specific keys for simplicity.
Key destructors are not invoked on thread exit.
This feels like a pretty naive implementation, but I think it can work.
Basically each waiter creates an object on its stack that is then
added to a linked list inside by the pthread_cond_t.
Signalling is then done by walking the list and unsetting the "waiting"
flag on as many of the waiters as you like.
