Commit Graph

48 Commits

Author SHA1 Message Date
Qi Wang
83f3294027 Small refactors around 7bb05e0. 2021-09-27 16:05:13 -07:00
Piotr Balcer
7bb05e04be add experimental.arenas_create_ext mallctl
This mallctl accepts an arena_config_t structure which
can be used to customize the behavior of the arena.
Right now it contains extent_hooks and a new option,
metadata_use_hooks, which controls whether the extent
hooks are also used for metadata allocation.

The medata_use_hooks option has two main use cases:

1. In heterogeneous memory systems, to avoid metadata
being placed on potentially slower memory.

2. Avoiding virtual memory from being leaked as a result
of metadata allocation failure originating in an extent hook.
2021-09-24 13:43:18 -07:00
Qi Wang
7c964b0352 Add rtree_write_range(): writing the same content to multiple leaf elements.
Apply to emap_(de)register_interior which became noticeable in perf profiles.
2021-03-29 17:19:53 -07:00
Qi Wang
4d8c22f9a5 Store edata->state in rtree leaf and make edata_t 128B aligned.
Verified that this doesn't result in any real increase of edata_t bytes
allocated.
2021-03-29 17:19:53 -07:00
Qi Wang
70d1541c5b Track extent is_head state in rtree leaf. 2021-03-29 17:19:53 -07:00
David Goldblatt
e034500698 Edata: rename "ranged" bit to "pai".
This better represents its intended purpose; the hugepage allocator design
evolved away from needing contiguity of hugepage virtual address space.
2020-09-18 12:39:25 -07:00
David Goldblatt
eda9c2858f Edata: zero stack edatas before initializing.
This avoids some UB. No compilers take advantage of it for now, but no sense in
tempting fate.
2020-05-14 10:30:20 -07:00
David Goldblatt
79ae7f9211 Rtree: Remove the per-field accessors.
We instead split things into "edata" and "metadata".
2020-04-10 13:12:47 -07:00
David Goldblatt
dc26b30094 Rtree: Clean up compact/non-compact split. 2020-04-10 13:12:47 -07:00
David Goldblatt
12eb888e54 Edata: Add a ranged bit.
We steal the dumpable bit, which we ended up not needing.
2020-04-10 13:12:47 -07:00
David Goldblatt
bd4fdf295e Rtree: Pull leaf contents into their own struct. 2020-04-10 13:12:47 -07:00
Yinan Zhang
21dfa4300d Change assert_* to expect_* in tests
```
grep -Irl assert_ test/ | xargs sed -i \
    's/witness_assert/witness_do_not_replace/g';
grep -Irl assert_ test/ | xargs sed -i \
    's/malloc_mutex_assert_owner/malloc_mutex_do_not_replace_owner/g';

grep -Ir assert_ test/ | grep -o "[_a-zA-Z]*assert_[_a-zA-Z]*" | \
    grep -v "^assert_"; # confirm no output
grep -Irl assert_ test/ | xargs sed -i 's/assert_/expect_/g';

grep -Irl witness_do_not_replace test/ | xargs sed -i \
    's/witness_do_not_replace/witness_assert/g';
grep -Irl malloc_mutex_do_not_replace_owner test/ | xargs sed -i \
    's/malloc_mutex_do_not_replace_owner/malloc_mutex_assert_owner/g';
```
2020-02-19 16:03:16 -08:00
David T. Goldblatt
a0c1f4ac57 Rtree: take the base allocator as a parameter.
This facilitates better testing by avoiding mixing of the "real" base with the
base used by the rtree under test.
2020-02-18 11:22:09 -08:00
David Goldblatt
a7862df616 Rename extent_t to edata_t.
This frees us up from the unfortunate extent/extent2 naming collision.
2019-12-20 10:18:40 -08:00
David T. Goldblatt
41187bdfb0 Extents: Break extent-struct/arena interactions
Specifically, the extent_arena_[g|s]et functions and the address randomization.

These are the only things that tie the extent struct itself to the arena code.
2019-09-23 23:06:27 -07:00
Qi Wang
9a86c65abc Implement retain on Windows.
The VirtualAlloc and VirtualFree APIs are different because MEM_DECOMMIT cannot
be used across multiple VirtualAlloc regions.  To properly support decommit,
only allow merge / split within the same region -- this is done by tracking the
"is_head" state of extents and not merging cross-region.

Add a new state is_head (only relevant for retain && !maps_coalesce), which is
true for the first extent in each VirtualAlloc region.  Determine if two extents
can be merged based on the head state, and use serial numbers for sanity checks.
2019-07-23 22:18:55 -07:00
David Goldblatt
55e5cc1341 SC: Make some key size classes static.
The largest small class, smallest large class, and largest large class may all
be needed down fast paths; to avoid the risk of touching another cache line, we
can make them available as constants.
2018-07-12 20:53:06 -07:00
David Goldblatt
e904f813b4 Hide size class computation behind a layer of indirection.
This class removes almost all the dependencies on size_classes.h, accessing the
data there only via the new module sc.h, which does not depend on any
configuration options.

In a subsequent commit, we'll remove the configure-time size class computations,
doing them at boot time, instead.
2018-07-12 20:53:06 -07:00
David Goldblatt
d14bbf8d81 Add a "dumpable" bit to the extent state.
Currently, this is unused (i.e. all extents are always marked dumpable).  In the
future, we'll begin using this functionality.
2017-10-16 15:35:49 -07:00
David Goldblatt
8261e581be Header refactoring: Pull size helpers out of jemalloc module. 2017-05-31 13:08:45 -07:00
David Goldblatt
44f9bd147a Header refactoring: unify and de-catchall rtree module. 2017-05-31 13:08:45 -07:00
David Goldblatt
3f685e8824 Protect the rtree/extent interactions with a mutex pool.
Instead of embedding a lock bit in rtree leaf elements, we associate extents
with a small set of mutexes.  This gets us two things:

- We can use the system mutexes.  This (hypothetically) protects us from
  priority inversion, and lets us stop doing a backoff/sleep loop, instead
  opting for precise wakeups from the mutex.
- Cuts down on the number of mutex acquisitions we have to do (from 4 in the
  worst case to two).

We end up simplifying most of the rtree code (which no longer has to deal with
locking or concurrency at all), at the cost of additional complexity in the
extent code: since the mutex protecting the rtree leaf elements is determined by
reading the extent out of those elements, the initial read is racy, so that we
may acquire an out of date mutex.  We re-check the extent in the leaf after
acquiring the mutex to protect us from this race.
2017-05-19 14:21:27 -07:00
Qi Wang
9ed84b0d45 Add init function support to tsd members.
This will facilitate embedding tcache into tsd, which will require proper
initialization cannot be done via the static initializer.  Make tsd->rtree_ctx
to be initialized via rtree_ctx_data_init().
2017-04-04 00:19:21 -07:00
Jason Evans
ce41ab0c57 Embed root node into rtree_t.
This avoids one atomic operation per tree access.
2017-03-22 18:33:32 -07:00
Jason Evans
99d68445ef Incorporate szind/slab into rtree leaves.
Expand and restructure the rtree API such that all common operations can
be achieved with minimal work, regardless of whether the rtree leaf
fields are independent versus packed into a single atomic pointer.
2017-03-22 18:33:32 -07:00
Jason Evans
944c8a3383 Split rtree_elm_t into rtree_{node,leaf}_elm_t.
This allows leaf elements to differ in size from internal node elements.

In principle it would be more correct to use a different type for each
level of the tree, but due to implementation details related to atomic
operations, we use casts anyway, thus counteracting the value of
additional type correctness.  Furthermore, such a scheme would require
function code generation (via cpp macros), as well as either unwieldy
type names for leaves or type aliases, e.g.

  typedef struct rtree_elm_d2_s rtree_leaf_elm_t;

This alternate strategy would be more correct, and with less code
duplication, but probably not worth the complexity.
2017-03-22 18:33:32 -07:00
Jason Evans
650c070e10 Remove rtree support for 0 (NULL) keys.
NULL can never actually be inserted in practice, and removing support
allows a branch to be removed from the fast path.
2017-02-08 18:50:03 -08:00
Jason Evans
f5cf9b19c8 Determine rtree levels at compile time.
Rather than dynamically building a table to aid per level computations,
define a constant table at compile time.  Omit both high and low
insignificant bits.  Use one to three tree levels, depending on the
number of significant bits.
2017-02-08 18:50:03 -08:00
Jason Evans
c0cc5db871 Replace tabs following #define with spaces.
This resolves #564.
2017-01-20 21:45:53 -08:00
Jason Evans
f408643a4c Remove extraneous parens around return arguments.
This resolves #540.
2017-01-20 21:43:07 -08:00
Jason Evans
c4c2592c83 Update brace style.
Add braces around single-line blocks, and remove line breaks before
function-opening braces.

This resolves #537.
2017-01-20 21:43:07 -08:00
Jason Evans
ffbb7dac3d Remove leading blank lines from function bodies.
This resolves #535.
2017-01-13 14:49:24 -08:00
Dave Watson
25f7bbcf28 Fix long spinning in rtree_node_init
rtree_node_init spinlocks the node, allocates, and then sets the node.
This is under heavy contention at the top of the tree if many threads
start to allocate at the same time.

Instead, take a per-rtree sleeping mutex to reduce spinning.  Tested
both pthreads and osx OSSpinLock, and both reduce spinning adequately

Previous benchmark time:
./ttest1 500 100
~15s

New benchmark time:
./ttest1 500 100
.57s
2016-11-02 20:30:53 -07:00
Jason Evans
6f29a83924 Add rtree lookup path caching.
rtree-based extent lookups remain more expensive than chunk-based run
lookups, but with this optimization the fast path slowdown is ~3 CPU
cycles per metadata lookup (on Intel Core i7-4980HQ), versus ~11 cycles
prior.  The path caching speedup tends to degrade gracefully unless
allocated memory is spread far apart (as is the case when using a
mixture of sbrk() and mmap()).
2016-06-05 20:59:57 -07:00
Jason Evans
e75e9be130 Add rtree element witnesses. 2016-06-03 12:27:41 -07:00
Jason Evans
8c9be3e837 Refactor rtree to always use base_alloc() for node allocation. 2016-06-03 12:27:41 -07:00
Jason Evans
2d2b4e98c9 Add element acquire/release capabilities to rtree.
This makes it possible to acquire short-term "ownership" of rtree
elements so that it is possible to read an extent pointer *and* read the
extent's contents with a guarantee that the element will not be modified
until the ownership is released.  This is intended as a mechanism for
resolving rtree read/write races rather than as a way to lock extents.
2016-06-03 12:27:33 -07:00
Jason Evans
a7a6f5bc96 Rename extent_node_t to extent_t. 2016-05-16 12:21:28 -07:00
Jason Evans
4be0c3ca42 Add no-OOM assertions to test. 2015-08-07 00:51:11 -07:00
Jason Evans
5fae7dc1b3 Fix MinGW-related portability issues.
Create and use FMT* macros that are equivalent to the PRI* macros that
inttypes.h defines.  This allows uniform use of the Unix-specific format
specifiers, e.g. "%zu", as well as avoiding Windows-specific definitions
of e.g. PRIu64.

Add ffs()/ffsl() support for compiling with gcc.

Extract compatibility definitions of ENOENT, EINVAL, EAGAIN, EPERM,
ENOMEM, and ENORANGE into include/msvc_compat/windows_extra.h and
use the file for tests as well as for core jemalloc code.
2015-07-23 13:56:25 -07:00
Jason Evans
ae93d6bf36 Avoid function prototype incompatibilities.
Add various function attributes to the exported functions to give the
compiler more information to work with during optimization, and also
specify throw() when compiling with C++ on Linux, in order to adequately
match what __THROW does in glibc.

This resolves #237.
2015-07-10 16:09:40 -07:00
Jason Evans
fd5f9e43c3 Avoid atomic operations for dependent rtree reads. 2015-05-15 17:02:30 -07:00
Jason Evans
8d6a3e8321 Implement dynamic per arena control over dirty page purging.
Add mallctls:
- arenas.lg_dirty_mult is initialized via opt.lg_dirty_mult, and can be
  modified to change the initial lg_dirty_mult setting for newly created
  arenas.
- arena.<i>.lg_dirty_mult controls an individual arena's dirty page
  purging threshold, and synchronously triggers any purging that may be
  necessary to maintain the constraint.
- arena.<i>.chunk.purge allows the per arena dirty page purging function
  to be replaced.

This resolves #93.
2015-03-18 18:55:33 -07:00
Jason Evans
8d0e04d42f Refactor rtree to be lock-free.
Recent huge allocation refactoring associates huge allocations with
arenas, but it remains necessary to quickly look up huge allocation
metadata during reallocation/deallocation.  A global radix tree remains
a good solution to this problem, but locking would have become the
primary bottleneck after (upcoming) migration of chunk management from
global to per arena data structures.

This lock-free implementation uses double-checked reads to traverse the
tree, so that in the steady state, each read or write requires only a
single atomic operation.

This implementation also assures that no more than two tree levels
actually exist, through a combination of careful virtual memory
allocation which makes large sparse nodes cheap, and skipping the root
node on x64 (possible because the top 16 bits are all 0 in practice).
2015-02-04 16:51:53 -08:00
Jason Evans
5460aa6f66 Convert all tsd variables to reside in a single tsd structure. 2014-09-23 02:36:08 -07:00
Jason Evans
ab8c79fdaf Fix message formatting errors uncovered by p_test_fail() refactoring. 2014-03-30 11:21:09 -07:00
Jason Evans
b954bc5d3a Convert rtree from (void *) to (uint8_t) storage.
Reduce rtree memory usage by storing booleans (1 byte each) rather than
pointers.  The rtree code is only used to record whether jemalloc manages
a chunk of memory, so there's no need to store pointers in the rtree.

Increase rtree node size to 64 KiB in order to reduce tree depth from 13
to 3 on 64-bit systems.  The conversion to more compact leaf nodes was
enough by itself to make the rtree depth 1 on 32-bit systems; due to the
fact that root nodes are smaller than the specified node size if
possible, the node size change has no impact on 32-bit systems (assuming
default chunk size).
2014-01-02 17:36:38 -08:00
Jason Evans
b980cc774a Add rtree unit tests. 2014-01-02 16:17:15 -08:00