* Added optional parameter "void** oldTable" to Resize(). If given the old
allocation for the table is returned instead of freeing it.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35278 a95241bf-73f2-0310-859d-f6bbb57e9c96
CACHE_DONT_LOCK_KERNEL_SPACE. If the former is given, the slab memory manager
does not wait when reserving memory or pages. The latter prevents area
operations. The new flags add a bit of flexibility. E.g. when allocating page
mapping objects for userland areas CACHE_DONT_WAIT_FOR_MEMORY is sufficient,
i.e. the allocation will succeed as long as pages are available.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35246 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Implemented a more elaborated raw memory allocation backend (MemoryManager).
We allocate 8 MB areas whose pages we allocate and map when needed. An area is
divided into equally-sized chunks which form the basic units of allocation. We
have areas with three possible chunk sizes (small, medium, large), which is
basically what the ObjectCache implementations were using anyway.
* Added "uint32 flags" parameter to several of the slab allocator's object
cache and object depot functions. E.g. object_depot_store() potentially wants
to allocate memory for a magazine. But also in pure freeing functions it
might eventually become useful to have those flags, since they could end up
deleting an area, which might not be allowable in all situations. We should
introduce specific flags to indicate that.
* Reworked the block allocator. Since the MemoryManager allocates block-aligned
areas, maintains a hash table for lookup, and maps chunks to object caches,
we can quickly find out which object cache a to be freed allocation belongs
to and thus don't need the boundary tags anymore.
* Reworked the slab boot strap process. We allocate from the initial area only
when really necessary, i.e. when the object cache for the respective
allocation size has not been created yet. A single page is thus sufficient.
other:
* vm_allocate_early(): Added boolean "blockAlign" parameter. If true, the
semantics is the same as for B_ANY_KERNEL_BLOCK_ADDRESS.
* Use an object cache for page mappings. This significantly reduces the
contention on the heap bin locks.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35232 a95241bf-73f2-0310-859d-f6bbb57e9c96
one for each per CPU store):
* The depot is now protected by a R/W lock combined with a spinlock. It is
required to either hold read lock + spinlock or just the write lock.
* When accessing the per CPU stores we only need to acquire the read lock
and disable interrupts. When switching magazines with the depot we
additionally get the spinlock.
* When allocating a new magazine we do completely unlock.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35200 a95241bf-73f2-0310-859d-f6bbb57e9c96
* The threads beside the main thread are killed earlier now (in the new
team_shutdown_team()), before removing the team from the team hash and from
its process group. This fixes#5296.
* Use a condition variable instead of a semaphore to wait for the non-main
threads to die. We notify the condition right after a thread has left the
team. The semaphore was released by the undertaker.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35196 a95241bf-73f2-0310-859d-f6bbb57e9c96
things a bit.
* Some style cleanup.
* The object depot does now have a cookie that will be passed to the return
hook.
* Fixed object_cache_return_object_wrapper() using the new cookie.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35174 a95241bf-73f2-0310-859d-f6bbb57e9c96
VMCacheRef object which points to the cache. This allows to optimize
VMCache::MoveAllPages(), since it no longer needs to iterate over all pages
to adjust their cache pointer. It can simple swap the cache refs of the two
caches instead.
Reduces the total -j8 Haiku image build time only marginally. The kernel time
drops almost 10%, though.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35155 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Added "bool consumerLocked" parameter to VMCache::Unlock() and
ReleaseRefAndUnlock(). Since Unlock() may cause the cache to be merged with
a consumer cache, the flag is needed to prevent a deadlock in case the
caller still holds a lock to the consumer. Hasn't been a problem yet, since
that situation never occurred.
* VMCacheChainLocker: Reversed unlocking order to bottom-up. The other
direction could cause a deadlock in case caches would be merged, since the
locking order would be reversed. The way VMCacheChainLocker was used this
didn't happen, though.
* fault_get_page(): While copying a page from a lower cache to the top cache,
we do now unlock all caches but the top one, so we don't unnecessarily
kill concurrency.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35153 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Reorganized the code for [un]mapping pages:
- Added new VMTranslationMap::Unmap{Area,Page[s]}() which essentially do what
vm_unmap_page[s]() did before, just in the architecture specific code, which
allows for specific optimizations. UnmapArea() is for the special case that
the complete area is unmapped. Particularly in case the address space is
deleted, some work can be saved. Several TODOs could be slain.
- Since they are only used within vm.cpp vm_map_page() and vm_unmap_page[s]()
are now static and have lost their prefix (and the "preserveModified"
parameter).
* Added VMTranslationMap::Protect{Page,Area}(). They are just inline wrappers
for Protect().
* X86VMTranslationMap::Protect(): Make sure not to accidentally clear the
accessed/dirty flags.
* X86VMTranslationMap::Unmap()/Protect(): Make page table skipping actually
work. It was only skipping to the next page.
* Adjusted the PPC code to at least compile.
No measurable effect for the -j8 Haiku image build time, though the kernel time
drops minimally.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35089 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Pulled the physical page mapping functions out of vm_translation_map into
a new interface VMPhysicalPageMapper.
* Renamed vm_translation_map to VMTranslationMap and made it a proper C++
class. The functions in the operations vector have become methods.
* Added class GenericVMPhysicalPageMapper implementing VMPhysicalPageMapper
as far as possible (without actually writing new code).
* Adjusted the x86 and the PPC specifics accordingly (untested for the
latter). For the other architectures the build is, I'm afraid, seriously
broken.
The next steps will modify and extend the VMTranslationMap interface, so that
it will be possible to fix the bugs in vm_unmap_page[s]() and employ
architecture specific optimizations.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35066 a95241bf-73f2-0310-859d-f6bbb57e9c96
* ioapic_init(): map_physical_memory() was called for already mapped
addresses. This worked fine, but only because the x86 page mapping code
didn't mind.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35059 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Use atomic_{and,or}() instead of atomic_set(), as there are no built-ins
for the latter.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35021 a95241bf-73f2-0310-859d-f6bbb57e9c96
This makes appending the pages to the active queue more efficient and we
don't need the vm_page::is_cleared bit anymore.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35011 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Added Debug{First,Next}() methods to allow easy iteration through the
address spaces in kernel debugger commands.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34978 a95241bf-73f2-0310-859d-f6bbb57e9c96
device path + child partition name. When a "raw" device is unpublished the node
removal notification triggers the partition and child partitions to be
unpublished/removed. Since in that case the "raw" node is already unpublished
trying to resolve it in devfs_unpublish_partition() again to unpublish the child
partitions would fail, leaving the child partition nodes behind. When a new raw
device would then become available publishing its partitions would fail because
of these left behind nodes, causing bug #4587. Seeing that this code is more
compact and straight forward anyway I don't quite see why it was changed in the
first place.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34967 a95241bf-73f2-0310-859d-f6bbb57e9c96
table. It is now inline and uses double-checked locking. This reduces the
contention on the lock to insignificant. Total -j8 Haiku image build speedup
is marginal, but the total kernel time drops 12%.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34934 a95241bf-73f2-0310-859d-f6bbb57e9c96
access to a vm_page. It is basically an atomically accessed thread ID field
in the vm_page structure, which is explicitly set by macros marking the
critical sections. As a first positive effect I had to review quite a bit of
code and found several issues.
* Added several TODOs and comments. Some harmless ones, but also a few
troublesome ones in vm.cpp regarding page unmapping.
* file_cache: PrecacheIO::Prepare()/read_into_cache: Removed superfluous
vm_page_allocate_page() return value checks. It cannot fail anymore.
* Removed the heavily contended "pages" lock. We use different policies now:
- sModifiedTemporaryPages is accessed atomically.
- sPageDeficitLock and sFreePageCondition are protected by a new mutex.
- The page queues have individual locks (mutexes).
- Renamed set_page_state_nolock() to set_page_state(). Unless the caller says
otherwise, it does now lock the affected pages queues itself. Also changed
the return value to void -- we panic() anyway.
* set_page_state(): Add free/clear pages to the beginning of their respective
queues as this is more cache-friendly.
* Pages with the states PAGE_STATE_WIRED or PAGE_STATE_UNUSED are no longer
in any queue. They were in the "active" queue, but there's no good reason
to have them there. In case we decide to let the page daemon work the queues
(like FreeBSD) they would just be in the way.
* Pulled the common part of vm_page_allocate_page_run[_no_base]() into a helper
function. Also fixed a bug I introduced previously: The functions must not
vm_page_unreserve_pages() on success, since they remove the pages from the
free/clear queue without decrementing sUnreservedFreePages.
* vm_page_set_state(): Changed return type to void. The function cannot really
fail and no-one was checking it anyway.
* vm_page_free(), vm_page_set_state(): Added assertion: The page must not be
free/clear before. This is implied by the policy that no-one is allowed to
access free/clear pages without holding the respective queue's lock, which is
not the case at this point. This found the bug fixed in r34912.
* vm_page_requeue(): Added general assertions. panic() when requeuing of
free/clear pages is requested. Same reason as above.
* vm_clone_area(), B_FULL_LOCK case: Don't map busy pages. The implementation is
still not correct, though.
My usual -j8 Haiku build test runs another 10% faster, now. The total kernel
time drops about 18%. As hoped the new locks have only a fraction of the old
"pages" lock contention. Other locks lead the "most wanted list" now.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34933 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Renamed page_queue to VMPageQueue and made it a proper C++ class. Use
DoublyLinkedList instead of own list code.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34874 a95241bf-73f2-0310-859d-f6bbb57e9c96
have one anymore anyway.
* Removed unnecessary setting the list links to NULL after removing a node.
* Replaced "element == NULL" check in Insert() by an assert. This just hid
potential errors.
* Added Insert{Before,After}() methods and declared the Insert() version
with the InsertBefore() semantics obsolete.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34873 a95241bf-73f2-0310-859d-f6bbb57e9c96
sure that the kernel's frame buffer console points to the right data.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34835 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Changed the rw_lock_{read,write}_unlock() return values to void. They
returned a value != B_OK only in case of user error and no-one checked them
anyway.
* Optimized rw_lock_read_[un]lock(). They are inline now and as long as
there's no contending write locker, they will only perform an atomic_add().
* Changed the semantics of nested locking after acquiring a write lock: Read
and write locks are counted separately, so read locks no longer implicitly
become write locks. This does e.g. make degrading a write lock to a read
lock by way of read_lock + write_unlock (as used in the VM) actually work.
These changes speed up the -j8 Haiku image build on my machine by a few
percent, but more interestingly they reduce the total kernel time by 25 %.
Apparently we get more contention on other locks, now.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34830 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Added VMCache::MovePage() and MoveAllPages() to move pages between caches.
* VMAnonymousCache:
- _MergeSwapPages(): Avoid doing anything, if neither cache has swapped out
pages.
- _MergeSwapPages() does now also remove source cache pages that are
shadowed by consumer swap pages. This allows us to call _MergeSwapPages()
before _MergePagesSmallerSource(), save the swap page shadowing check
there and get rid of the vm_page::merge_swap flag. This is an
optimization based on the assumption that usually none or only few pages
are swapped out, so we save a lot of checks.
- Implemented _MergePagesSmallerConsumer() as an alternative to
_MergePagesSmallerSource(). The former is used when the source cache has
more pages than the consumer cache. It iterates over the consumer cache's
pages, moves them to the source and finally moves all pages back to the
consumer. The final move is relatively cheap (though unfortunately we
still have to update all pages' vm_page::cache field), so that overall we
save iterations of the main loop with the more expensive checks.
The optimizations particularly improve the common fork()+exec*() situations.
fork() uses CoW, which is implemented by putting two new empty caches between
the to be copied area and its cache. exec*() destroys one copy of the area,
its cache and thus causes merging of the other new cache with the old cache.
Since this usually happens in a very short time, the old cache does still
contain many pages and the new cache only few. Previously the many pages were
all checked and moved individually. Now we do that for the few pages instead.
A very extreme example of this situation is the Haiku image build. jam has a
huge heap (> 200 MB) and it fork()s+exec*()s for every action to be executed.
Since during the cache merging the cache is locked, any write access to a
heap page causes jam to block until the cache merging is done. Formerly that
took so long that it killed a lot of parallelism in multi-job builds. That
could be observed particularly well when lots of small actions where executed
(like the Link, XRes, Mimeset, SetType, SetVersion combos when building
executables/libraries/add-ons). Those look dramatically better now.
The overall speed improvement for a -j8 image build on my machine is only
about 15%, though.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34784 a95241bf-73f2-0310-859d-f6bbb57e9c96
another. The code originates from vm_copy_on_write_area(). We now generate
the VM cache tracing entries, though.
* count_writable_areas() -> VMCache::CountWritableAreas()
* Added debugger command "cache_stack" which is enabled when VM cache tracing
is enabled. It prints the source caches of a given cache or area at the
time of a specified tracing entry.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34751 a95241bf-73f2-0310-859d-f6bbb57e9c96
- Replaced the "userOnly" parameter by a "flags" parameter, that allows to
specify kernel and userland stack traces individually.
- x86, m68k: Don't always skip the first frame as that prevents the caller
from being able to record its own address.
* capture_tracing_stack_trace(): Replaced the "userOnly" parameter by
"kernelOnly", since one is probably always interested in the kernel stack
trace, but might not want the userland stack trace.
* Added stack trace support for VM cache kernel tracing.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34742 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Added Lock()/Unlock() for explicit locking by a service user.
* Added NotifyLocked() and made Notify() inline.
* Added HasListeners() so one can check whether there is a listener at all
before preparing the event message.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34736 a95241bf-73f2-0310-859d-f6bbb57e9c96
- enqueue_in_run_queue() no longer returns whether rescheduling is supposed
to happen. Instead is sets cpu_ent::invoke_scheduler on the current CPU.
- reschedule() does now handle cpu_ent::invoke_scheduler_if_idle(). No need
to let all callers do that.
* thread_unblock[_locked]() no longer return whether rescheduling is supposed
to happen.
* Got rid of the B_INVOKE_SCHEDULER handling. The interrupt hooks really
can't know, when it makes sense to reschedule or not.
* Introduced scheduler_reschedule_if_necessary[_locked]() functions for
checking+invoking the scheduler.
* Some semaphore functions (e.g. delete_sem()) invoke the scheduler now, if
they wake up anything with greater priority.
I've also tried to add scheduler invocations in the condition variable and
mutex/rw_lock code, but that actually has a negative impact on performance,
probably because it causes too much ping-ponging between threads when
multiple locking primitives are involved.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34657 a95241bf-73f2-0310-859d-f6bbb57e9c96
* simple_smp scheduler: Rewrote the interesting part of
enqueue_in_run_queue(). It always selects a target CPU for the inserted
thread, now. If no CPU is idle, the CPU running the thread with the lowest
priority is chosen. If the thread running on the target CPU has a lower
priority than the inserted one, it will be asked to reschedule. If that's
the current CPU, we'll return the correct value (wasn't done before at
all).
These changes help reducing latencies. On my machine in an idle system
playing music DebugAnalyzer shows maximum latencies of about 1 us. I still
find that a bit much, but it's several orders of magnitude better than
before. The -j8 Haiku image build time dropped about 10%.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34635 a95241bf-73f2-0310-859d-f6bbb57e9c96
implemented by means of an additional member in cpu_ent.
* Removed thread::keep_scheduled and the related functions. The feature
wasn't used yet and wouldn't have worked as implemented anyway.
* Resurrected an older, SMP aware version of our simple scheduler and made it
the default instead of the affine scheduler. The latter is in no state to
be used yet. It causes enormous latencies (I've seen up to 0.1s) even when
six or seven CPUs were idle at the same time, totally killing parallelism.
That's also the reason why a -j8 build was slower than a -j2. This is no
longer the case. On my machine the -j2 build takes about 10% less time now
and the -j8 build saves another 20%. The latter is not particularly
impressive (compared with Linux), but that seems to be due to lock
contention.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34615 a95241bf-73f2-0310-859d-f6bbb57e9c96
system_time_nsecs(), returning the system time in nanoseconds. The function
is only really implemented for x86. For the other architectures
system_time() * 1000 is returned.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34543 a95241bf-73f2-0310-859d-f6bbb57e9c96
waits for certain events on a given page, NotifyPageEvents() wakes up
waiting threads respectively.
* Used the new feature instead of condition variables for waiting on busy
pages. We save publishing and unpublishing of a condition variable whenever
a page is marked busy. There's only something to do, if there's at least
one thread waiting in the list of the respective cache. The general
assumption is that this is only rarely the case and even if it happens,
there should be only very few threads.
* Added an apparently missing notification in cache_io(). At least I didn't
see the reason for it not being there.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34537 a95241bf-73f2-0310-859d-f6bbb57e9c96
to clarify that they never enlarge the area.
* Reimplemented VMKernelAddressSpace. It is somewhat inspired by Bonwick's
vmem resource allocator (though we have different requirements):
- We consider the complete address space to be divided into contiguous
ranges of type free, reserved, or area, each range being represented by
a VMKernelAddressRange object.
- The range objects are managed in an AVL tree and a doubly linked list
(the latter only for faster iteration) sorted by address. This provides
O(log(n)) lookup, insertion and removal.
- For each power of two size we maintain a list of free ranges of at least
that size. Thus for the most common case of B_ANY*_ADDRESS area
allocation, we find a free range in constant time (the rest of the
processing being O(log(n))) with a rather good fit. This should also
help avoiding address space fragmentation.
While the new implementation should be faster, particularly with an
increasing number of areas, I couldn't measure any difference in the -j2
haiku build. From a cursory test the -j8 build hasn't tangibly benefitted
either.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34528 a95241bf-73f2-0310-859d-f6bbb57e9c96
- Renamed to AVLTreeBase and moved it into its own header/source file.
- Renamed FindClose() to FindClosest().
- Added CheckTree() method for debugging purposes. It checks the validity
of the tree.
* Added a templatized class AVLTree which doesn't offer a map-like interface
like AVLTreeMap, but rather one similar to BOpenHashMap and SplayTree. It
is more convenient to use, if one wants to store objects that already
contain the key.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34526 a95241bf-73f2-0310-859d-f6bbb57e9c96
necessary and prevent the structures from being used in a union.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34525 a95241bf-73f2-0310-859d-f6bbb57e9c96
