uma_reclaim(). Reclamation code must not see half-constructed or
destructed zones. Do this by bracing uma_zcreate() and uma_zdestroy()
into a shared-locked sx, and take the sx exclusively in uma_reclaim().
Usually zones are not created/destroyed during the system operation,
but tmpfs mounts do cause zone operations and exposed the bug.
Another solution could be to only expose a new keg on uma_kegs list
after the corresponding zone is fully constructed, and similar
treatment for the destruction. But it probably requires more risky
code rearrangement as well.
Reported and tested by: pho
Discussed with: avg
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
This code has had an extensive rewrite and a good series of reviews, both by the author and other parties. This means a lot of code has been simplified. Pluggable structures for high-rate entropy generators are available, and it is most definitely not the case that /dev/random can be driven by only a hardware souce any more. This has been designed out of the device. Hardware sources are stirred into the CSPRNG (Yarrow, Fortuna) like any other entropy source. Pluggable modules may be written by third parties for additional sources.
The harvesting structures and consequently the locking have been simplified. Entropy harvesting is done in a more general way (the documentation for this will follow). There is some GREAT entropy to be had in the UMA allocator, but it is disabled for now as messing with that is likely to annoy many people.
The venerable (but effective) Yarrow algorithm, which is no longer supported by its authors now has an alternative, Fortuna. For now, Yarrow is retained as the default algorithm, but this may be changed using a kernel option. It is intended to make Fortuna the default algorithm for 11.0. Interested parties are encouraged to read ISBN 978-0-470-47424-2 "Cryptography Engineering" By Ferguson, Schneier and Kohno for Fortuna's gory details. Heck, read it anyway.
Many thanks to Arthur Mesh who did early grunt work, and who got caught in the crossfire rather more than he deserved to.
My thanks also to folks who helped me thresh this out on whiteboards and in the odd "Hallway track", or otherwise.
My Nomex pants are on. Let the feedback commence!
Reviewed by: trasz,des(partial),imp(partial?),rwatson(partial?)
Approved by: so(des)
Acquire the lock in read mode when just needed to ensure the stability
of the keg list. The UMA lock may be held for a long time (relatively
speaking) in uma_reclaim() on machines with lots of zones/kegs. If the
uma_timeout() would fire during that period, subsequent callouts on that
CPU may be significantly delayed.
Reviewed by: jhb
Callers of zone_drain_wait(M_WAITOK) do not need to hold (and were not)
the uma_mtx, but we would attempt to unlock and relock the mutex if we
had to sleep because the zone was already draining. The M_NOWAIT callers
may hold the uma_mtx, but we do not sleep in that case.
Reviewed by: jhb
MFC after: 3 days
These changes prevent sysctl(8) from returning proper output,
such as:
1) no output from sysctl(8)
2) erroneously returning ENOMEM with tools like truss(1)
or uname(1)
truss: can not get etype: Cannot allocate memory
there is an environment variable which shall initialize the SYSCTL
during early boot. This works for all SYSCTL types both statically and
dynamically created ones, except for the SYSCTL NODE type and SYSCTLs
which belong to VNETs. A new flag, CTLFLAG_NOFETCH, has been added to
be used in the case a tunable sysctl has a custom initialisation
function allowing the sysctl to still be marked as a tunable. The
kernel SYSCTL API is mostly the same, with a few exceptions for some
special operations like iterating childrens of a static/extern SYSCTL
node. This operation should probably be made into a factored out
common macro, hence some device drivers use this. The reason for
changing the SYSCTL API was the need for a SYSCTL parent OID pointer
and not only the SYSCTL parent OID list pointer in order to quickly
generate the sysctl path. The motivation behind this patch is to avoid
parameter loading cludges inside the OFED driver subsystem. Instead of
adding special code to the OFED driver subsystem to post-load tunables
into dynamically created sysctls, we generalize this in the kernel.
Other changes:
- Corrected a possibly incorrect sysctl name from "hw.cbb.intr_mask"
to "hw.pcic.intr_mask".
- Removed redundant TUNABLE statements throughout the kernel.
- Some minor code rewrites in connection to removing not needed
TUNABLE statements.
- Added a missing SYSCTL_DECL().
- Wrapped two very long lines.
- Avoid malloc()/free() inside sysctl string handling, in case it is
called to initialize a sysctl from a tunable, hence malloc()/free() is
not ready when sysctls from the sysctl dataset are registered.
- Bumped FreeBSD version to indicate SYSCTL API change.
MFC after: 2 weeks
Sponsored by: Mellanox Technologies
the queue where to enqueue pages that are going to be unwired.
- Add stronger checks to the enqueue/dequeue for the pagequeues when
adding and removing pages to them.
Of course, for unmanaged pages the queue parameter of vm_page_unwire() will
be ignored, just as the active parameter today.
This makes adding new pagequeues quicker.
This change effectively modifies the KPI. __FreeBSD_version will be,
however, bumped just when the full cache of free pages will be
evicted.
Sponsored by: EMC / Isilon storage division
Reviewed by: alc
Tested by: pho
since it will almost certanly fail. Take next bigger zone instead.
This situation should not happen with original bucket zones configuration:
"32 Bucket" zone uses "64 Bucket" and vice versa. But if "64 Bucket" zone
lock is congested, zone may grow its bucket size and start biting itself.
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
(> PAGE_SIZE) zones. If zone is not multiple to PAGE_SIZE, there may
be enough space for the header at the last page, so we may avoid extra
header memory allocation and hash table update/lookup.
ZFS creates bunch of odd-sized UMA zones (5120, 6144, 7168, 10240, 14336).
This change gives good use to at least some of otherwise lost memory there.
Reviewed by: avg
There are good reasons for this to happen, such as recursion prevention, etc.
and they are not fatal since buckets are just an optimization mechanism.
Real bucket allocation failures are any way counted by the bucket zones
themselves, and we don't need double accounting there.
was used without making sure first that it was really passed for us.
On some of my systems this bug made user argument passed by ZFS code to
uma_zalloc_arg() unexpectedly block UMA per-CPU caches for those zones.
This is a last resort for very low memory condition in case other measures
to free memory were ineffective. Sequentially cycle through all CPUs and
extract per-CPU cache buckets into zone cache from where they can be freed.
Lock congestion is the same, whether it happens on alloc or free, so
handle it equally. Now that we have back pressure, there is no problem
to grow buckets a bit faster. Any way growth is much slower then in 9.x.
These new buckets make bucket size self-tuning more soft and precise.
Without them there are buckets for 1, 5, 13, 29, ... items. While at
bigger sizes difference about 2x is fine, at smallest ones it is 5x and
2.6x respectively. New buckets make that line look like 1, 3, 5, 9, 13,
29, reducing jumps between steps, making algorithm work softer, allocating
and freeing memory in better fitting chunks. Otherwise there is quite a
big gap between allocating 128K and 5x128K of RAM at once.
Every time system detects low memory condition decrease bucket sizes for
each zone by one item. As result, higher memory pressure will push to
smaller bucket sizes and so smaller per-CPU caches and so more efficient
memory use.
Before this change there was no force to oppose buckets growth as result
of practically inevitable zone lock conflicts, and after some run time
per-CPU caches could consume enough RAM to kill the system.
The consequence of the bug is that fini calls are not done
when a slab is freed by a call-back from the page daemon.
It went unnoticed for two months because fini is little used.
I spotted the bug while reading the code to learn how it works
so I could write it up for the next edition of the Design and
Implementation of FreeBSD book.
No MFC needed as this code exists only in HEAD.
Reviewed by: kib, jeff
Tested by: pho
additional information, when the page is guaranteed to not belong to a
paging queue. Usually, this results in a lot of type casts which make
reasoning about the code correctness harder.
Sometimes m->object is used instead of pageq, which could cause real
and confusing bugs if non-NULL m->object is leaked. See r141955 and
r253140 for examples.
Change the pageq member into a union containing explicitly-typed
members. Use them instead of type-punning or abusing m->object in x86
pmaps, uma and vm_page_alloc_contig().
Requested and reviewed by: alc
Sponsored by: The FreeBSD Foundation
transparent layering and better fragmentation.
- Normalize functions that allocate memory to use kmem_*
- Those that allocate address space are named kva_*
- Those that operate on maps are named kmap_*
- Implement recursive allocation handling for kmem_arena in vmem.
Reviewed by: alc
Tested by: pho
Sponsored by: EMC / Isilon Storage Division
through bucket_alloc() to uma_zalloc_arg() and uma_zfree_arg().
- Make some smaller buckets for large zones to further reduce memory
waste.
- Implement uma_zone_reserve(). This holds aside a number of items only
for callers who specify M_USE_RESERVE. buckets will never be filled
from reserve allocations.
Sponsored by: EMC / Isilon Storage Division
- Be more explicit about zone vs keg locking. This functionally changes
almost nothing.
- Add a size parameter to uma_zcache_create() so we can size the buckets.
- Pass the zone to bucket_alloc() so it can modify allocation flags
as appropriate.
- Fix a bug in zone_alloc_bucket() where I missed an address of operator
in a failure case. (Found by pho)
Sponsored by: EMC / Isilon Storage Division
performance.
- Always free to the alloc bucket if there is space. This gives LIFO
allocation order to improve hot-cache performance. This also allows
for zones with a single bucket per-cpu rather than a pair if the entire
working set fits in one bucket.
- Enable per-cpu caches of buckets. To prevent recursive bucket
allocation one bucket zone still has per-cpu caches disabled.
- Pick the initial bucket size based on a table driven maximum size
per-bucket rather than the number of items per-page. This gives
more sane initial sizes.
- Only grow the bucket size when we face contention on the zone lock, this
causes bucket sizes to grow more slowly.
- Adjust the number of items per-bucket to account for the header space.
This packs the buckets more efficiently per-page while making them
not quite powers of two.
- Eliminate the per-zone free bucket list. Always return buckets back
to the bucket zone. This ensures that as zones grow into larger
bucket sizes they eventually discard the smaller sizes. It persists
fewer buckets in the system. The locking is slightly trickier.
- Only switch buckets in zalloc, not zfree, this eliminates pathological
cases where we ping-pong between two buckets.
- Ensure that the thread that fills a new bucket gets to allocate from
it to give a better upper bound on allocation time.
Sponsored by: EMC / Isilon Storage Division
backing memory that is only a container for per-cpu caches of arbitrary
pointer items. These zones have no kegs.
- Convert the regular keg based allocator to use the new import/release
functions.
- Move some stats to be atomics since they would require excessive zone
locking/unlocking with the new import/release paradigm. Make
zone_free_item simpler now that callers can manage more stats.
- Check for these cache-only zones in the public APIs and debugging
code by checking zone_first_keg() against NULL.
Sponsored by: EMC / Isilong Storage Division
bitmap using sys/bitset. This is much simpler, has lower space
overhead and is cheaper in most cases.
- Use a second bitmap for invariants asserts and improve the quality of
the asserts as well as the number of erroneous conditions that we will
catch.
- Drastically simplify sizing code. Special case refcnt zones since they
will be going away.
- Update stale comments.
Sponsored by: EMC / Isilon Storage Division
These zones have slab size == sizeof(struct pcpu), but request from VM
enough pages to fit (uk_slabsize * mp_ncpus). An item allocated from such
zone would have a separate twin for each CPU in the system, and these twins
are at a distance of sizeof(struct pcpu) from each other. This magic value
of distance would allow us to make some optimizations later.
To address private item from a CPU simple arithmetics should be used:
item = (type *)((char *)base + sizeof(struct pcpu) * curcpu)
These arithmetics are available as zpcpu_get() macro in pcpu.h.
To introduce non-page size slabs a new field had been added to uma_keg
uk_slabsize. This shifted some frequently used fields of uma_keg to the
fourth cache line on amd64. To mitigate this pessimization, uma_keg fields
were a bit rearranged and least frequently used uk_name and uk_link moved
down to the fourth cache line. All other fields, that are dereferenced
frequently fit into first three cache lines.
Sponsored by: Nginx, Inc.
future further optimizations where the vm_object lock will be held
in read mode most of the time the page cache resident pool of pages
are accessed for reading purposes.
The change is mostly mechanical but few notes are reported:
* The KPI changes as follow:
- VM_OBJECT_LOCK() -> VM_OBJECT_WLOCK()
- VM_OBJECT_TRYLOCK() -> VM_OBJECT_TRYWLOCK()
- VM_OBJECT_UNLOCK() -> VM_OBJECT_WUNLOCK()
- VM_OBJECT_LOCK_ASSERT(MA_OWNED) -> VM_OBJECT_ASSERT_WLOCKED()
(in order to avoid visibility of implementation details)
- The read-mode operations are added:
VM_OBJECT_RLOCK(), VM_OBJECT_TRYRLOCK(), VM_OBJECT_RUNLOCK(),
VM_OBJECT_ASSERT_RLOCKED(), VM_OBJECT_ASSERT_LOCKED()
* The vm/vm_pager.h namespace pollution avoidance (forcing requiring
sys/mutex.h in consumers directly to cater its inlining functions
using VM_OBJECT_LOCK()) imposes that all the vm/vm_pager.h
consumers now must include also sys/rwlock.h.
* zfs requires a quite convoluted fix to include FreeBSD rwlocks into
the compat layer because the name clash between FreeBSD and solaris
versions must be avoided.
At this purpose zfs redefines the vm_object locking functions
directly, isolating the FreeBSD components in specific compat stubs.
The KPI results heavilly broken by this commit. Thirdy part ports must
be updated accordingly (I can think off-hand of VirtualBox, for example).
Sponsored by: EMC / Isilon storage division
Reviewed by: jeff
Reviewed by: pjd (ZFS specific review)
Discussed with: alc
Tested by: pho
Replace the sub-optimal uma_zone_set_obj() primitive with more modern
uma_zone_reserve_kva(). The new primitive reserves before hand
the necessary KVA space to cater the zone allocations and allocates pages
with ALLOC_NOOBJ. More specifically:
- uma_zone_reserve_kva() does not need an object to cater the backend
allocator.
- uma_zone_reserve_kva() can cater M_WAITOK requests, in order to
serve zones which need to do uma_prealloc() too.
- When possible, uma_zone_reserve_kva() uses directly the direct-mapping
by uma_small_alloc() rather than relying on the KVA / offset
combination.
The removal of the object attribute allows 2 further changes:
1) _vm_object_allocate() becomes static within vm_object.c
2) VM_OBJECT_LOCK_INIT() is removed. This function is replaced by
direct calls to mtx_init() as there is no need to export it anymore
and the calls aren't either homogeneous anymore: there are now small
differences between arguments passed to mtx_init().
Sponsored by: EMC / Isilon storage division
Reviewed by: alc (which also offered almost all the comments)
Tested by: pho, jhb, davide
will be printed once the given zone becomes full and cannot allocate an
item. The warning will not be printed more often than every five minutes.
All UMA warnings can be globally turned off by setting sysctl/tunable
vm.zone_warnings to 0.
Discussed on: arch
Obtained from: WHEEL Systems
MFC after: 2 weeks
low memory situation. I've observed a situation where per-CPU
allocations were disabled while there were enough free cached pages.
Basically, cnt.v_free_count was sitting stable at a value lower
than cnt.v_free_min and that caused massive performance drop.
Reviewed by: alc
MFC after: 1 week
excluding other allocations including UMA now entails the addition of
a single flag to kmem_alloc or uma zone create
Reviewed by: alc, avg
MFC after: 2 weeks
