1 ms to 1 us. The long minimum timeout seriously screwed profiling,
since by default it works with 1 ms ticks, stopping the timer when the
thread is unscheduled and restarting it with the remaining time when it
is scheduled again.
I could also imagine that this had a negative effect on latencies and
the precision of thread wakeup times.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27672 a95241bf-73f2-0310-859d-f6bbb57e9c96
flags on the wrong thread (the current one instead of the newly created
one). This would cause the thread not to be debugged as it should, e.g.
profiling wouldn't work correctly.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27671 a95241bf-73f2-0310-859d-f6bbb57e9c96
* The class calculates a minimum width now, which is based on the line height,
this may also fix the bug with the small text inputs in the Pe Find window.
* Added TODOs about implementing GetHeightForWidth(), which may be a good idea
when a BTextView is used as non-editable informative text in an interface,
and one wants to make sure that the entire text is shown.
* Replaced the call to _Refresh() in Draw(), which recalculates all the line
breaks for no reason with _DrawLines() again. The TODO mentioned that text
will be drawn without drawing the background first, but maybe this is a
relict from times where Draw() was invoked directly? At least I cannot see
any negative consequences, and this should be much more efficient.
(Other than that, this patch should hopefully have no potential negative
side effects...<crosses fingers>)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27670 a95241bf-73f2-0310-859d-f6bbb57e9c96
usually they won't change during the runtime of the profiled program
anyway.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27668 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Adjusted used files systems accordingly.
* BFS::Stream::GetName() was broken. It accessed the small data region
which wasn't loaded, since BFS::Stream derived from bfs_inode, which
is a variably-sized structure with the small data region at the end.
Changed that to a ref-counted, shared member instead.
* Implemented RootFileSystem::GetName().
* Added Directory::GetPath() to get a full path of the directory or an
entry.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27664 a95241bf-73f2-0310-859d-f6bbb57e9c96
into BTextView class as well, hopefully fixing the GCC4 build.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27661 a95241bf-73f2-0310-859d-f6bbb57e9c96
buffer during a timer event that interrupted a kernel function: We do
now flush the buffer as soon as it is 70% full, *if* we didn't interrupt
a kernel function. When the buffer runs full and we still haven't hit a
user function, we drop the tick. The number of dropped ticks is recorded
and sent to the debugger with the next update message.
Reverted the previous partial solution (the temporary disabling of
profiling while in debugger support code).
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27658 a95241bf-73f2-0310-859d-f6bbb57e9c96
various debugger messages. We do now correctly match all sampled
addresses.
* Update the saved info of a team's main thread after exec*(), so we
can print a more useful name.
* Improved the output: The images are listed in a table, now. The
function table also includes the image IDs for the functions.
* Cleaned up debug output.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27657 a95241bf-73f2-0310-859d-f6bbb57e9c96
is created or deleted (or exec*() has been invoked). The counter is
sent with several debugger messages.
* Track the image event counter that is used when samples are added to
the profiling buffer. If the current team counter differs, we flush
the buffer first (sending an update message to the debugger), so that
the debugger has a chance to match the addresses to the correct images.
* Disable profiling for a thread while it runs in the debugger support
code. This fixes potential deadlocks which could occur when a
profiling timer event occurred that would require the buffer to be
flushed while the thread was just sending something to the debugger or
waiting for a command. As it turns out, this is not sufficient either,
since we should never try to flush the buffer when the timer event
occurred in the kernel, since the thread might hold a lock that the
debugger thread could try to acquire. Will implement a more general
solution later.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27656 a95241bf-73f2-0310-859d-f6bbb57e9c96
BTextView classes:
* Declared the directly used BTextView helper classes as private BTextView
classes and changed all affected files.
* Realized that Tracker's BPoseView was (accidentally?) using what used to
be _BWidthBuffer_. It had declared it's own class with the same name and
same members/size in headers/private/tracker/TextViewSupport.h, but the
implementation was nowhere to be found. I can only explain this that
the BTextView implementation was then actually linked and used. But the big
problem was that it was used without locking (unlike in BTextView)! When
many Tracker windows opened during system startup or later and they happened
to each request characters not yet in the cache, I imagine things could have
gone bad and corrupted memory. Anyways, since I can see the usefulness of
the cache, BPoseView uses BTextView::WidthBuffer on purpose now. And I moved
the locking inside BTextView::WidthBuffer::StringWidth().
* Adjusted InterfaceDefs.cpp accordingly.
* TODO: Move subsequent classes into BTextView namespace as well, ie derived
classes that BTextView doesn't directly know about. All stuff in src/kits/
inteface/textview_support/
* Added preliminary and not yet implemented layout friendly BTextView
constructors.
* I will try to handle the insets imposed by BTextView::fTextRect a bit
differently when used inside the new layout management framework. For this,
I added BTextView::SetInsets() and GetInsets(). SetInsets() doesn't do
anything yet.
So far, everything seems to work still... ;-)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27654 a95241bf-73f2-0310-859d-f6bbb57e9c96
available for hinted font rendering. Instead, display the information that
only this combination has no effect on Haiku builds without subpixel based
freetype rendering. Better fix for 2753.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27653 a95241bf-73f2-0310-859d-f6bbb57e9c96
yet, since there's no synchronization of the sample evalution with the
image creation/deletion. That is samples taken before an exec*() could
be evaluated after the new images have already been loaded and could
thus be translated to the wrong functions.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27651 a95241bf-73f2-0310-859d-f6bbb57e9c96
* We can't enable interrupts in an interrupt handler. Instead we use the
newly introduced callback feature, which notifies the debugger right
before returning from the interrupt.
* We didn't indicate that the profiling buffer was full and that the
thread shouldn't be profiled ATM. Therefore it could happen that it
was profiled while trying to notify the debugger that the profiling
buffer was full, resulting in a deadlock. Introduce a respective flag
in the thread debug structure.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27649 a95241bf-73f2-0310-859d-f6bbb57e9c96
interrupt handler and will be executed right before returning from the
interrupt.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27648 a95241bf-73f2-0310-859d-f6bbb57e9c96
* truncate the path label, since a long path would automatically
resize the window to something unusable in that case...
Note: somehow the PopUpMenu does ignore a set max column size
expanding from small to large leaves some drawing artefacts
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27647 a95241bf-73f2-0310-859d-f6bbb57e9c96
label, a TextControl maintains it's current width in ResizeToPreferred() if
it is wide enough. Should fix#2751.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27646 a95241bf-73f2-0310-859d-f6bbb57e9c96
i did exactly that way in my code, which messed up the location popup
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27645 a95241bf-73f2-0310-859d-f6bbb57e9c96
* have png and home as default
* still misses some kind of settings...
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27642 a95241bf-73f2-0310-859d-f6bbb57e9c96
be tracked to the kernel, which then counts the hits, an area is
passed to kernel in which the hits are recorded. When the area is
full, the debugger is notified. For some reason that part doesn't work
yet -- the whole system freezes when waiting for a reply.
* Reorganized the profile tool code a bit. For one with respect to the
changed API, but also to prepare tracking of image creation/deletion.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27640 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Remove unused code.
* Remove INSTANT_UPDATE define, I guess instant update is always wanted.
* Replaced text control for setting the "averaging strength" with a slider
and hopefully improved label.
* Simplified other labels a bit as well.
* Added detection of subpixel rendering availability by peeking into the
Freetype option header. If it's not turned on, then the controls are not
displayed and instead there is a text explaining the problem and hinting
at what needs to be done in order to enable the feature.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27634 a95241bf-73f2-0310-859d-f6bbb57e9c96
that makes drivers using it more independent from internal net_timer changes.
* Correctly implemented callout_active() using the new is_timer_running()
stack function.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27632 a95241bf-73f2-0310-859d-f6bbb57e9c96
object files. Thus static functions will be found, too.
* debug_lookup_symbol_address() and debug_next_image_symbol() no longer
need to read the symbol name via the debugger API, since the
respective SymbolLookup methods compute the length of the symbol name
that can safely be accessed locally, now.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27628 a95241bf-73f2-0310-859d-f6bbb57e9c96
