all teams in serveral server areas, and instead of having to eventually clone
them all several times in BBitmap, we now have one or more areas per team,
and BBitmap will only clone areas once if needed. As a side effect, this
method should be magnitudes faster than the previous version.
* This method is also much more secure: instead of putting the allocation
maintenance structures into those everyone-read-write areas, they are now
separated, so that faulty applications cannot crash the app_server this
way anymore. This should fix bug #172.
* Freeing memory is not yet implemented though! (although all memory will
be freed upon app exit)
* There are now 3 different bitmap allocation strategies: per ClientMemoryAllocator
(ie. via ServerApp), per area (for overlays, not yet implemented), and using
malloc()/free() for server-only bitmaps.
* ServerBitmap now deletes its buffers itself.
* Cleaned up BBitmap and BApplication a bit.
* The test environment currently doesn't build anymore, will fix it next.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16826 a95241bf-73f2-0310-859d-f6bbb57e9c96
was a bit strange anyway - this closed bug #317.
* The selected window now gets a selection frame in case it can be moved.
* If the window cannot be moved, a new workspace isn't selected either.
* When choosing a workspace, the one currently selected also gets a selection
frame.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16823 a95241bf-73f2-0310-859d-f6bbb57e9c96
file (which does no longer exist) - it now uses the one from the kernel as a
quick fix.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16821 a95241bf-73f2-0310-859d-f6bbb57e9c96
into account when copying to the same color space (new code isn't that fast,
but it works).
* BBitmap::ImportBits() did not take the specified offset into account at all.
This fixes bug #313.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16812 a95241bf-73f2-0310-859d-f6bbb57e9c96
chose the front window as our next focus window - but this proved
to be problematic with B_AVOID_FRONT windows. Therefore, we now
simply chose the top-most window as the next focus window.
This fixes bug #281, and potentially also fixes bug #181.
* This also revealed another bug in SetFocusWindow(): when the window
to have focus already had focus, but were hidden before, the focus
did not change; if that window was subsequently removed, the app_server
would have crashed.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16811 a95241bf-73f2-0310-859d-f6bbb57e9c96
in HWInterface::_DrawCursor(). Axel, even though we
understood the problem, we didn't really fix it back
then: When the sync flag was set to true in
BView::SetViewCursor(), the link was only flushed,
which means that the function still returned before the
ServerWindow thread processed the message. This means
that the race condition (the cursor being immediately
deleted after SetViewCursor returns, which might be
processed in ServerApp thread before the SetViewCursor
request in ServerWindow thread) still existed. I changed
SetViewCursor now to do a real sync (wait for the
ServerWindow reply) before returning. The alternative
would be to set the fPendingViewCursor flag in either case.
Anyhow, I could reproduce the error quite reliably before
this change, and now it is gone... here is to hoping!
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16809 a95241bf-73f2-0310-859d-f6bbb57e9c96
Unfortunately, Tracker resolves the link, and doesn't pick up its name or shortcut
(should be fixed in Tracker, though).
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16806 a95241bf-73f2-0310-859d-f6bbb57e9c96
these should be public (they don't match any basic Be naming style
anyway :-).
* Put the code that's used by the app_server where it's needed.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16804 a95241bf-73f2-0310-859d-f6bbb57e9c96
this one, maybe we should find out what it does, make it public and
document it.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16801 a95241bf-73f2-0310-859d-f6bbb57e9c96
for the kernel as well, and thus we need to export _kern_get_timezone(),
too.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16792 a95241bf-73f2-0310-859d-f6bbb57e9c96
(realized as an option enabled in the menu).
* Minor bug fixes and changes.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16791 a95241bf-73f2-0310-859d-f6bbb57e9c96
using the new _kern_get_timezone() syscall.
* Added an implementation of ftime() based on the above function, this
may fix bug #308.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16786 a95241bf-73f2-0310-859d-f6bbb57e9c96
dunno if we want to keep this, though.
* Improved some methods by no longer calling strlen() more than once.
* We're now using snprintf() instead of sprintf() in the << operators to
make them more secure (even though the string lengths should be long
enough).
* Improved << operators by taking the return (the resulting string length) of
snprintf() into account.
* Replaced calls to _GrowBy() with a negative argument with calls to
_Alloc() which actually safes some computation.
* Cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16783 a95241bf-73f2-0310-859d-f6bbb57e9c96
audio/multi changed to audio/hmulti (avoid confusion with Be's audio/multi)
hmultiaudio.addon won't load ich_ac97 any more (reason is it's not compatible)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16782 a95241bf-73f2-0310-859d-f6bbb57e9c96
to be killed. Reverted ShutdownProcess.cpp to continue to use kill_team()
instead of sending a signal.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16780 a95241bf-73f2-0310-859d-f6bbb57e9c96
