* fixed ticket #1500
* big cleanup and removal of unused code
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22411 a95241bf-73f2-0310-859d-f6bbb57e9c96
of locking before processing the message (single/all window lock)
-> in most message cases, I could comment out the unlocking/locking which
switched to the different lock, because the required lock is now already held,
this removes some race conditions which were commented in the code already
* EventDispatcher::SetDragMessage() didn't lock the object, this would have
been bad if multiple windows tried to set a drag bitmap at once
* the Desktop object keeps track of mouse position and pressed buttons, so
that it doesn't need to lock the EventDispatcher for sending fake mouse
moves to windows on show/hide of windows (solves some cases of possible
dead locks with the new locking strategy)
* the keyboard EventFilter switches the current workspace asynchrnously from
the Desktop thread (another source of possible deadlocks)
* the "reader is trying to become writer" check in MultiLocker is only used
in DEBUG mode now
As a summary: It would be nice if ServerWindow used a readlock for all messages
it processes itself, and forwards all messages for which it needs a write lock
to the Desktop thread. All cases where either the Desktop or the ServerWindow
mess with the EventDispatcher are possible sources of deadlocks. This is solved
right now by making sure that the lock is released before using the
EventDispatcher.
I have not observed any deadlocks while switching workspaces and launching
many apps anymore, neither crashes. But I have not tested extensively except
for in the test environment. That being said, I could reproduce the problems
on first try before in Haiku.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22410 a95241bf-73f2-0310-859d-f6bbb57e9c96
pointer - which it now handles gracefully.
* This also fixes starting the runtime loader directly: it no longer crashes
but will just return an error.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22409 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Check safemode settings only when it's not already disabled (doesn't make
sense to check those then).
* Minor cleanup
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22408 a95241bf-73f2-0310-859d-f6bbb57e9c96
snprintf instead of sprintf and reduced the size of the char array,
though. Hope you don't mind, Rene.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22406 a95241bf-73f2-0310-859d-f6bbb57e9c96
a second time would magically work, as it skipped its initialization then...
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22405 a95241bf-73f2-0310-859d-f6bbb57e9c96
could be written back. This should stop the page thief from stealing active
pages that cannot be recreated easily :-)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22404 a95241bf-73f2-0310-859d-f6bbb57e9c96
the death stack and reacquiring the thread lock in which another thread
could snatch our stack that we were still going to use for the
scheduler. Now we've got a second spinlock that we can hold while
releasing a semaphore.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22403 a95241bf-73f2-0310-859d-f6bbb57e9c96
and thread spinlock was reverse in Wait() and Notify(). The thread lock
is now the outer lock -- this way it is still possible to call Notify()
with the thread lock being held.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22401 a95241bf-73f2-0310-859d-f6bbb57e9c96
* In select_fd(): First get the file descriptor, then check whether any
events have to be selected at all. This has the advantage that the
caller can interpret an error return code as invalid FD. Consequently
common_poll() no longer checks FD validity separately -- this was a
race condition.
* common_poll() always selects POLLERR and POLLHUP now, which it has to
do according to the specs.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22400 a95241bf-73f2-0310-859d-f6bbb57e9c96
already exist, it won't be zeroed out. The image will still be
initialized with BFS, though. Handy for installing Haiku on a partition.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22397 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Reworked the select support:
- The io_context additionally stores a table of lists of select_infos,
which enables it to deselect events of a pending select() when
closing a FD. This prevents a race condition potentially causing a
write to stale memory.
- The opaque selectsync* passed to FSs is now actually a select_info*.
This was necessary, since the FDs deselect() hook (unlike the
select() hook) doesn't take a "ref" argument and deselecting a
single info (e.g. caused by a premature close()) was not possible.
The select() hook's "ref" argument has become superfluous.
- It should now be relatively easy to implement a poll_on_steroids()
that can also wait for objects other than FDs (e.g. semaphores,
ports, threads etc.).
* Set/reset the signal mask in common_select(). This makes pselect()
work as required.
* Reorganized vfs_resize_fd_table().
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22391 a95241bf-73f2-0310-859d-f6bbb57e9c96
stdin/out/err according the what was specified via arguments.
select_close_test select()s a file descriptor that is closed a little
later by another thread.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22390 a95241bf-73f2-0310-859d-f6bbb57e9c96
client and the server. This should fix bug #1490, but I haven't tested it yet.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22388 a95241bf-73f2-0310-859d-f6bbb57e9c96
characters first, and then falls back to take everything. fTrigger is now
uint32 and works with all unicode characters; unlike in BeOS, the
MenuTriggerTest application now works correctly in Haiku.
* fTriggerIndex is now a character position, not a byte position of the label;
this allows BMenuItem::DrawContent() to draw the trigger at the correct
position, even if there are multi-byte UTF-8 characters.
* The above fixed bug #1506; triggers are still not working, though.
* Rewrote Menu.h header.
* Renamed all private methods (that are not called by BWindow) to have the
underscore prefix.
* Removed unused methods.
* Some minor cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22384 a95241bf-73f2-0310-859d-f6bbb57e9c96
necessary because low memory handlers might need to wait for other services
while we need to make sure the page thief can always steal pages.
* Made the page thief more aggressive when memory is critical: added a third
run where it will also steal a small amount of active pages to keep things
going.
* The page thief did not maintain the cache's reference correctly in all
situations.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22382 a95241bf-73f2-0310-859d-f6bbb57e9c96
shapes) which have so far been done for Haiku, the website and other apps
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22381 a95241bf-73f2-0310-859d-f6bbb57e9c96
account when checking for the number of readable bytes. For readers
it doesn't make a difference, since they were looping anyway, but
select() would report read-availability incorrectly.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22379 a95241bf-73f2-0310-859d-f6bbb57e9c96
writers.
* Removed the optimization for piping data between two threads of the
same team. This greatly simplifies the code. It seems to me the case
is very uncommon; it could be added back later, though.
* Basically rewrote reading from and writing to the pipe:
- A blockable writer can now write more than what is currently
available in the ring buffer.
- Writing respects the PIPE_BUF non-interleaving limit, though our
headers don't seem to define PIPE_BUF anywhere.
- Unblock writers, when the last reader is gone and send those that
haven't written anything yet a SIGPIPE. Fixes bug #1476.
* Correctly implemented select() support. We were only notifying
writers. We manage two separate select sync pools per pipe now: one
for the reader end and one for the writer end.
* Reading/writing from the root dir does no longer end in KDL.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22378 a95241bf-73f2-0310-859d-f6bbb57e9c96
Added support for non-48-bit LBA, added write support.
Proper error handling still missing, write support disabled for your protection.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22376 a95241bf-73f2-0310-859d-f6bbb57e9c96
components - now, we divide the initialization of that service into two parts
which allows this (before, most handlers were gone after boot).
* Added debugger command that dumps the low memory handlers.
* The slab allocator now registers its low memory handler with a higher
priority, so that it'll run before the potentially heavier ones.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22374 a95241bf-73f2-0310-859d-f6bbb57e9c96
on another machine, don't know if this was really the cause, but it should
be checked nevertheless I guess)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22373 a95241bf-73f2-0310-859d-f6bbb57e9c96
deadlock whenever someone without a lock (like the page_writer()) is calling
this function.
* Added a new ReadWriteLock::TryLock() method.
* Renamed bfs_read_vnode() and bfs_release_vnode() to bfs_get_vnode() and
bfs_put_vnode() just like those functions are called in fs_interface.h.
* Fixed a warning in BPlusTree::_SplitNode().
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22372 a95241bf-73f2-0310-859d-f6bbb57e9c96
takes menu margins into account, drawing is now fully scalable)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22371 a95241bf-73f2-0310-859d-f6bbb57e9c96
the BeBook mentions it does
* make sure the view is invalidated on resize
* needs more fixes (for example GetPreferredSize() does not take everything
into account)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22370 a95241bf-73f2-0310-859d-f6bbb57e9c96
doesn't exist, so I just used a BFile instead... this should fix part of
ticket #1423, if someone wants to look into the gateway retrieval
(settings.cpp) to fix the other half of the bug, that would be great
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22366 a95241bf-73f2-0310-859d-f6bbb57e9c96
