The extents and hpa-nonfull "ind" columns are emitter_type_unsigned but were
set via col_ind.size_val -- correct only by little-endian union overlap.
Output is byte-identical. Pre-existing (predates the refactor).
mutex_stats_read_global and mutex_stats_read_arena had byte-identical
bodies; merge them into a single mutex_stats_read used by both the global
and per-arena mutex tables. Pure cleanup; no functional or output change.
Table-output change: label size classes above 8*PAGE as a range
"prev_size+1..size" (e.g. "32769..40960") in the extents / lextents /
hpa-nonfull tables, instead of a single size. Text only; JSON is unchanged
(the ordered size-class metadata already implies the range).
Table-output change: fold the HPA "In full slabs:" / "In empty slabs:" prose
blocks into the per-size-class table as two symbolic rows (size "full"/"empty")
and retitle it "pageslabs:". JSON is byte-identical; only the human-readable
table changes.
Table-output change: emit the HPA shard counters as structured key:value rows
keyed by the JSON field names, instead of the previous free-form prose. JSON
output is byte-identical; only the human-readable table changes.
src/stats.c fused mallctl reads with dual-format (text/JSON) emission inside a
few very large, bottom-up functions, behind dense macros, with the public entry
point at the very bottom. Rework it to be output-preserving but much easier to
read and extend:
- Separate gathering from emission where it is worth it: most sections become a
small gather (stats_gather_*, the mallctl reads into a typed struct) plus an
emit that renders it, with the plumbing macros (CTL_*, COL*) and gather
structs moved to internal header (stats_internal.h). This is a balance,
not a rule: the per-size-class tables stay O(1) streaming (per-row gather+emit,
no large stack buffers, avoiding heap buffering), the mutex rows keep reading
straight into emitter columns, and the already-readable config/opt probes are
left as they are.
- Read like the output: open with a table-of-contents comment and the public
stats_print(), then the section functions top-down, keeping the
interval/boot/fork plumbing at the bottom. The ToC names the function that
prints each section, and its JSON sub-path.
- Keep one emission engine: the emitter remains the single dual-format renderer,
so each section has one rendering path rather than parallel renderers (the two
formats are not field-identical). Move some table-only concerns into it --
gap-collapsing ("---") and a dual columnar row op (emitter_row) -- so section
code carries less table-format detail.
Output-preserving: text is byte-identical and JSON keys/values are unchanged.
It is unused and no longer needed: hpa_dirty_mult, hpa_purge_threshold,
hpa_min_purge_interval_ms, and hpa_min_purge_delay_ms now provide finer
control over the purge rate.
Add a small extent cache in front of the PAC ecaches. Allocs and dallocs
that fit are served from per-shard SEC bins without taking the ecache
mutex; overflow falls through to the backing ecaches, including
ecache_pinned for pinned extents.
The feature is gated behind experimental_pac_sec_nshards (default 0,
disabled). To support independent HPA and PAC SEC instances,
sec_alloc/sec_dalloc/sec_fill take an explicit shard argument, with HPA
and PAC using separate TSD shard slots.
The runtime option aborted on every OOM, breaking new(std::nothrow)
semantics. Replace with configure-time --enable-cxx-infallible-new
(default off): when on, throwing new aborts (size logged) and
nothrow returns null; when off, standard new_handler + bad_alloc /
null behavior is preserved. Under LTO the on-path lets the compiler
prove operator new is no-throw.
The split managed one ordering constraint: arena_choose() had to be
defined before arena_choose_maybe_huge() but after the tsd/tcache
inlines it depends on. After the malloc_dispatch refactor moved the
heaviest tcache-pulling inlines out of arena_inlines_b.h, the
remaining arena-side inlines all belong together. The merged
arena_inlines.h explicitly includes jemalloc_internal_inlines_a.h
and tcache.h (previously transitively pulled).
Convert the production source files in src/ (69 .c/.cpp) and
test/jemalloc_test.h.in to list the headers they actually use, then
delete the umbrella. Three consolidated headers (peak_event.h,
prof_sys.h, sz.h) also gain explicit transitive includes.
Every translation unit now declares what it uses. A missing include
now fails at the failing file rather than silently working because
something upstream pulled in the world.
This change includes the following improvements:
- Remove the hpa_sec_batch_fill_extra parameter.
- Refactor the hpa_alloc() code and helper functions to be able to
allocate more than one extent out of a single pageslab. This way
we can amortize the per-pageslab costs (active bitmap iteration,
pageslab metadata updates) across multiple extents.
- Decide on a min and max number of extents that will be allocated
in hpa_alloc(). The code will try to allocate at least the min
and allocate up to the max as long as we can allocate additional
ones from the pageslab we already have, as additional allocations
are relatively cheap.
- Add extent allocation distribution stats.
- Amend hpa_sec_integration.c unit test.
Some pages (e.g., hugetlb pages) cannot be purged, and should be
prioritized for reuse. A custom extent_alloc hook signals this by
OR'ing EXTENT_ALLOC_FLAG_PINNED into the low bits of the returned
pointer; jemalloc strips the flag bits and caches pinned extents in
a dedicated ecache_pinned, separate from the dirty/muzzy decay
pipeline.
Pinned extents do not coalesce eagerly, except for ones larger than
SC_LARGE_MINCLASS. A prefer-small policy reuses the smallest fitting
pinned extent, to avoid unnecessary split/fragmentation.
In both the full_slabs and empty_slabs JSON sections of HPA shard
stats, "nactive_huge" was emitted twice instead of emitting
"ndirty_huge" as the second entry. This caused ndirty_huge to be
missing from the JSON output entirely.
Add a unit test that verifies both sections contain "ndirty_huge".
During mutex stats emit, derived counters are not emitted for json.
Yet the array indexing counter should still be increased to skip
derived elements in the output, which was not. This commit fixes it.
Implementation inspired by idea described in "Beyond malloc efficiency
to fleet efficiency: a hugepage-aware memory allocator" paper [1].
Primary idea is to track maximum number (peak) of active pages in use
with sliding window and then use this number to decide how many dirty
pages we would like to keep.
We are trying to estimate maximum amount of active memory we'll need in
the near future. We do so by projecting future active memory demand
(based on peak active memory usage we observed in the past within
sliding window) and adding slack on top of it (an overhead is reasonable
to have in exchange of higher hugepages coverage). When peak demand
tracking is off, projection of future active memory is active memory we
are having right now.
Estimation is essentially the same as `nactive_max * (1 + dirty_mult)`.
Peak demand purging algorithm controlled by two config options. Option
`hpa_peak_demand_window_ms` controls duration of sliding window we track
maximum active memory usage in and option `hpa_dirty_mult` controls
amount of slack we are allowed to have as a percent from maximum active
memory usage. By default `hpa_peak_demand_window_ms == 0` now and we
have same behaviour (ratio based purging) that we had before this
commit.
[1]: https://storage.googleapis.com/gweb-research2023-media/pubtools/6170.pdf
Add opt.process_madvise_max_batch which determines if process_madvise is enabled
(non-zero) and the max # of regions in each batch. Added another limiting
factor which is the space to reserve on stack, which results in the max batch of
128.
When evaluating changes in HPA logic, it is useful to know internal
`hpa_shard` state. Great deal of this state is `psset`. Some of the
`psset` stats was available, but in disaggregated form, which is not
very convenient. This commit exposed `psset` counters to `mallctl`
and malloc stats dumps.
Example of how malloc stats dump will look like after the change.
HPA shard stats:
Pageslabs: 14899 (4354 huge, 10545 nonhuge)
Active pages: 6708166 (2228917 huge, 4479249 nonhuge)
Dirty pages: 233816 (331 huge, 233485 nonhuge)
Retained pages: 686306
Purge passes: 8730 (10 / sec)
Purges: 127501 (146 / sec)
Hugeifies: 4358 (5 / sec)
Dehugifies: 4 (0 / sec)
Pageslabs, active pages, dirty pages and retained pages are rows added
by this change.
Linux 6.1 introduced `MADV_COLLAPSE` flag to perform a best-effort
synchronous collapse of the native pages mapped by the memory range into
transparent huge pages.
Synchronous hugification might be beneficial for at least two reasons:
we are not relying on khugepaged anymore and get an instant feedback if
range wasn't hugified.
If `hpa_hugify_sync` option is on, we'll try to perform synchronously
collapse and if it wasn't successful, we'll fallback to asynchronous
behaviour.
Option `experimental_hpa_strict_min_purge_interval` was expected to be
temporary to simplify rollout of a bugfix. Now, when bugfix rollout is
complete it is safe to remove this option.
Option `experimental_hpa_max_purge_nhp` introduced for backward
compatibility reasons: to make it possible to have behaviour similar
to buggy `hpa_strict_min_purge_interval` implementation.
When `experimental_hpa_max_purge_nhp` is set to -1, there is no limit
to number of slabs we'll purge on each iteration. Otherwise, we'll purge
no more than `experimental_hpa_max_purge_nhp` hugepages (slabs). This in
turn means we might not purge enough dirty pages to satisfy
`hpa_dirty_mult` requirement.
Combination of `hpa_dirty_mult`, `experimental_hpa_max_purge_nhp` and
`hpa_strict_min_purge_interval` options allows us to have steady rate of
pages returned back to the system. This provides a strickier latency
guarantees as number of `madvise` calls is bounded (and hence number of
TLB shootdowns is limited) in exchange to weaker memory usage
guarantees.
Change in `hpa_min_purge_interval_ms` handling logic is not backward
compatible as it might increase memory usage. Now this logic guarded by
`hpa_strict_min_purge_interval` option.
When `hpa_strict_min_purge_interval` is true, we will purge no more than
`hpa_min_purge_interval_ms`. When `hpa_strict_min_purge_interval` is
false, old purging logic behaviour is preserved.
Long term strategy migrate all users of hpa to new logic and then delete
`hpa_strict_min_purge_interval` option.
This lets us easily see what fraction of flush load is being taken up by the
bins, and helps guide future optimization approaches (for example: should we
prefetch during cache bin fills? It depends on how many objects the average fill
pops out of the batch).
