Closing usb_midi now wake up midi_server port reader, as expected.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33782 a95241bf-73f2-0310-859d-f6bbb57e9c96
Global lock/unlock assembly code was broken, reworked based on
what FreeBSD does.
Thanks!
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33781 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Now, killing a team shut properly cause the app_server to close the windows
of that team, too.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33777 a95241bf-73f2-0310-859d-f6bbb57e9c96
posixtestsuite into our repository.
* A few of the latter ones actually fail on Haiku.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33776 a95241bf-73f2-0310-859d-f6bbb57e9c96
pthread_create 2-1 interface conformance test to pass.
* Also fixed return values to be in the POSIX error range in case we ever switch
them by default.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33775 a95241bf-73f2-0310-859d-f6bbb57e9c96
ID.
* Accidently, this should also fix bug #4839.
* Optimized _get_next_sem_info() a whole lot by iterating over the team's
semaphore list.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33773 a95241bf-73f2-0310-859d-f6bbb57e9c96
to the owning team.
* Instead, the team now maintains a list containing the ports it owns.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33771 a95241bf-73f2-0310-859d-f6bbb57e9c96
renamed files. After more investigation i found out that the problem was
caused by a non usual use of the zombie list in r31307. In the conditions
to reproduce #4130 it would call AttrChanged() or StatChanged() on zombie
poses which by convention should always be invalid models, ie: null node.
The logic would potentially create other problems too as it wasn't only
adding refFiltered models to the zombies.
I reverted part of r31307, clarified the logic and updated the comments.
I'd prefer another implementation for this as reusing the zombie list
here is only calling for future mysterious problems. #4131, #4271, #4610
and #4831 are still closed, but i'm reopening #1717 which is more of an
enhancement btw.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33769 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Brecht, is it possible your line break column is one character too short? At
least you broke some lines without need.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33766 a95241bf-73f2-0310-859d-f6bbb57e9c96
This changeset correct usage of this language abbreviation and restore normal function of the localized preflets.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33765 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Instead, they should call the protected virtual ExpandOrCollapse(), this makes
one of my apps work again under Haiku (where the list view contents are
filled on demand).
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33760 a95241bf-73f2-0310-859d-f6bbb57e9c96
I thought there was a ticket about this bug, but couldn't find it.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33750 a95241bf-73f2-0310-859d-f6bbb57e9c96
* added header for dealing with binary numbers and bitmasks (C++ templates)
these "macro's" might not work well for long words, though
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33749 a95241bf-73f2-0310-859d-f6bbb57e9c96
returns valid data. Since the data is still valid in the copy, we don't need
to clear the corresponding flag when copying a message. This is most probably
the same behaviour as in R5 as Tracker seems to expect that in the case where
a drop is done after a menu was displayed as in #4757. Thanks to aldeck for
tracking that one down inside Tracker, should fix#4757.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33746 a95241bf-73f2-0310-859d-f6bbb57e9c96
Using macro nonnull which actually resolves to an attribute doesn't work
on inline functions.
Thx to mmlr for helping out.
Eitherway it should have worked in the first place, because the compiler
has all the information it needs.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33745 a95241bf-73f2-0310-859d-f6bbb57e9c96
busy processing other things, the file could be gone by the time we processed
the fs notification message of the entry creation. This case wasn't handled
properly and would cause #4130 and most certainly duplicate #4831, #4610 and
also possibly #4271.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33744 a95241bf-73f2-0310-859d-f6bbb57e9c96
* This makes sem_delete_owned_sems() a lot more efficient; before it would need
to scan the entire semaphore table.
* This speeds up the test build of the kernel by another 2 seconds (with
KDEBUG=2) on my laptop.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33743 a95241bf-73f2-0310-859d-f6bbb57e9c96
the network drivers needed to be adjusted to the header reordering, too
sorry
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33740 a95241bf-73f2-0310-859d-f6bbb57e9c96
ordering.
this makes porting driver (especially wifi ones) more
convinient
* based on FreeBSD svn revision 196691
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33739 a95241bf-73f2-0310-859d-f6bbb57e9c96
iterated over all known thread *IDs* with interrupts disabled.
Now it iterates over the team's thread list (going from back to front, since
new threads are added at the front of the singly linked queue).
* Alexandre restarted Tracker quite a lot, and let the shell script run to
reproduce a certain bug - and then wondered why ProcessController would
take several seconds to open its windows until it passed through more than
8 million IDs... :-)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33737 a95241bf-73f2-0310-859d-f6bbb57e9c96
(hopefully) correct place.
* It seems to be even harder to understand basic locking primitives: when you
think about it, it shouldn't surprise you that conditional variables never
return B_WOULD_BLOCK. This fixes gdb again.
* Added tracing support to the ports subsystem.
* get_port_message() will now resize the port heap if needed (but will also
take timeouts into account while doing so, more or less). The initial port
space is 4MB (as before), the growth rate is the same, and the system wide
limit is arbitrarily set to 64 MB (all swappable). A team limit has been set
to 8 MB, but is not enforced yet. Since ports are using up address space in
the kernel, those seems to be proper limits.
* This also fixes a strange, and rare lockup where the mouse cursor would still
move, but everything else would basically hang, but look perfectly normal from
KDL on the first look. As recently happened on Brecht's laptop, and debugged
by mmlr and me: the cbuf space got used up when lots of windows wanted to
redraw after a workspace switch. The app_server wouldn't answer anymore to
client requests, but thought it would have done so, as LinkSender::Flush()
doesn't care if it got a B_NO_MEMORY (the ports will now block until memory
is available if possible, so that should not be a problem anymore).
* Improved "port" KDL command, it now also prints the messages in the port.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33735 a95241bf-73f2-0310-859d-f6bbb57e9c96
