Bypassing background thread creation for the oversize_arena used to be an
optimization since that arena had eager purging. However #2466 changed the
purging policy for the oversize_arena -- specifically it switched to the default
decay time when background_thread is enabled.
This issue is noticable when the number of arenas is low: whenever the total #
of arenas is <= 4 (which is the default max # of background threads), in which
case the purging will be stalled since no background thread is created for the
oversize_arena.
One of the condition to start purging is `hpa_hugify_blocked_by_ndirty`
function call returns true. This can happen in cases where we have no
dirty memory for this shard at all. In this case purging loop will be an
infinite loop.
`hpa_hugify_blocked_by_ndirty` was introduced at 0f6c420, but at that
time purging loop has different form and additional `break` was not
required. Purging loop form was re-written at 6630c5989, but additional
exit condition wasn't added there at the time.
Repo code was shared by Patrik Dokoupil at [1], I stripped it down to
minimum to reproduce issue in jemalloc unit tests.
[1]: https://github.com/jemalloc/jemalloc/pull/2533
When there were many items stashed, it's possible that after flushing stashed,
ncached is already lower than the remain, in which case the flush can simply
return at that point.
This change adds support for writing pid namespaces to the filename of a
heap profile. When running with namespaces pids may reused across
namespaces and if mounts are shared where profiles are written there is
not a great way to differentiate profiles between pids.
Signed-off-by: Daniel Hodges <hodges.daniel.scott@gmail.com>
Signed-off-by: Daniel Hodges <hodgesd@fb.com>
realloc() and rallocx() shares path, and realloc() should set errno to
ENOMEM upon OOM failures.
Fixes: ee961c2310 ("Merge realloc and rallocx pathways.")
Signed-off-by: Juhyung Park <qkrwngud825@gmail.com>
Access nactive, ndirty and nmuzzy throught getters and not directly.
There are no functional change, but getters are required to propagate
HPA's statistics up to Page Allocator's statitics.
Instead of failing on assertions. Previously the same change was made for
posix_memalign and aligned_alloc (#1554). Make memalign behave the same way
even though it's obsolete.
it's within the huge page size. These requests do not concern internal
fragmentation with huge pages, since the entire range is expected to be
accessed.
```
sub cleanup {
unlink($main::tmpfile_sym);
unlink(keys %main::tempnames);
# We leave any collected profiles in $HOME/jeprof in case the user wants
# to look at them later. We print a message informing them of this.
if ((scalar(@main::profile_files) > 0) &&
defined($main::collected_profile)) {
if (scalar(@main::profile_files) == 1) {
print STDERR "Dynamically gathered profile is in $main::collected_profile\n";
}
print STDERR "If you want to investigate this profile further, you can do:\n";
print STDERR "\n";
print STDERR " jeprof \\\n";
print STDERR " $main::prog \\\n";
print STDERR " $main::collected_profile\n";
print STDERR "\n";
}
}
```
On cleanup, it would print out a message for the collected profile.
If there is only one collected profile, it would pop by L691, then `scalar(@main::profile_files)` would be 0, and no message would be printed.
The current autogen.sh script doesn't allow receiving quoted extra
options.
If someone wants to pass extra CFLAGS that is split into multiple
options with a whitespace, then a quote is required.
However, the configure inside autogen.sh fails in this case as follows.
$ ./autogen.sh CFLAGS="-Dmmap=cxl_mmap -Dmunmap=cxl_munmap"
autoconf
./configure --enable-autogen CFLAGS=-Dmmap=cxl_mmap -Dmunmap=cxl_munmap
configure: error: unrecognized option: `-Dmunmap=cxl_munmap'
Try `./configure --help' for more information
Error 0 in ./configure
It's because the quote discarded unexpectedly when calling configure.
This patch is to fix this problem.
Signed-off-by: Honggyu Kim <honggyu.kim@sk.com>
1. Pre-generate all default tcache ncached_max in tcache_boot;
2. Add getters returning default ncached_max and ncached_max_set;
3. Refactor tcache init so that it is always init with a given setting.
An allocation small enough will be promoted so that it does not
share an extent with others. However, when dalloc, such allocations
may not be dalloc as a promoted one if nbins < SC_NBINS. This
commit fixes the bug.
1. `thread_tcache_ncached_max_read_sizeclass` allows users to get the
ncached_max of the bin with the input sizeclass, passed in through
oldp (will be upper casted if not an exact bin size is given).
2. `thread_tcache_ncached_max_write` takes in a char array
representing the settings for bins in the tcache.
When using metadata_thp, allocate tcache bin stacks from base0, which means they
will be placed on huge pages along with other metadata, instead of mixed with
other regular allocations.
In order to do so, modified the base allocator to support limited reuse: freed
tcached stacks (from thread termination) will be returned to base0 and made
available for reuse, but no merging will be attempted since they were bump
allocated out of base blocks. These reused base extents are managed using
separately allocated base edata_t -- they are cached in base->edata_avail when
the extent is all allocated.
One tricky part is, stats updating must be skipped for such reused extents
(since they were accounted for already, and there is no purging for base). This
requires tracking the "if is reused" state explicitly and bypass the stats
updates when allocating from them.
1. add tcache_max and nhbins into tcache_t so that they are per-tcache,
with one auto tcache per thread, it's also per-thread;
2. add mallctl for each thread to set its own tcache_max (of its auto tcache);
3. store the maximum number of items in each bin instead of using a global storage;
4. add tests for the modifications above.
5. Rename `nhbins` and `tcache_maxclass` to `global_do_not_change_nhbins` and `global_do_not_change_tcache_maxclass`.
