Ingo Weinhold 369111e741 * Removed the VMArea::Wire() version that has to allocate a VMAreaWiredRange. ...

Since the requirement is that the area's top cache is locked, allocating
  memory isn't allowed.
* lock_memory_etc(): Create the VMAreaWiredRange object explicitly before
  locking the area's top cache.

Fixes #5680 (deadlocks when using the slab as malloc() backend).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36033 a95241bf-73f2-0310-859d-f6bbb57e9c96

2010-04-05 11:08:01 +00:00

..

arch

x86:

2010-03-26 00:03:53 +00:00

boot

* Added Directory::CreateFile() which can be implemented by file systems to

2010-03-16 17:29:58 +00:00

disk_device_manager

…

fs

…

platform

…

posix

…

slab

Changed object_depot_store() return value to void. It is now always takes

2010-02-24 19:04:41 +00:00

util

Added ring_buffer_get_vecs() that returns iovecs describing the contents of

2010-03-16 17:07:40 +00:00

vm

* Removed the VMArea::Wire() version that has to allocate a VMAreaWiredRange.

2010-04-05 11:08:01 +00:00

block_cache.h

…

boot_device.h

…

boot_item.h

…

boot_splash.h

…

commpage.h

…

condition_variable.h

…

console.h

…

cpu.h

…

debug_heap.h

…

debug_paranoia.h

…

debug.h

* Changed arch_debug_save_registers() to take an arch-specific structure

2010-03-18 17:20:06 +00:00

debugger_keymaps.h

…

elf_priv.h

…

elf.h

…

file_cache.h

…

frame_buffer_console.h

…

generic_syscall.h

…

heap.h

* Introduced {malloc,memalign,free}_etc() which take an additional "flags"

2010-01-27 12:45:53 +00:00

int.h

…

kdevice_manager.h

…

kdriver_settings.h

…

kernel_c++_structs.h

…

kernel_daemon.h

…

kernel.h

…

kimage.h

…

kmodule.h

…

kscheduler.h

…

ksignal.h

…

ksyscalls.h

…

ksystem_info.h

…

listeners.h

…

lock.h

…

low_resource_manager.h

…

messaging.h

…

MessagingServiceDefs.h

…

Notifications.h

…

port.h

…

real_time_clock.h

…

safemode.h

…

scheduling_analysis.h

…

sem.h

…

smp.h

…

syscall_restart.h

…

system_profiler.h

…

team.h

Changed the team shutdown process a bit:

2010-01-20 09:34:53 +00:00

thread_types.h

sigsuspend() would previously wake up correctly when a matching signal was

2010-03-13 13:17:40 +00:00

thread.h

…

timer.h

…

tracing.h

…

user_atomic.h

…

user_debugger.h

Correctly handle cases when a thread single-steps into the kernel as it can

2010-02-25 20:20:16 +00:00

usergroup.h

…

vfs.h

* Introduced {malloc,memalign,free}_etc() which take an additional "flags"

2010-01-27 12:45:53 +00:00

wait_for_objects.h

…