7a89a7b8c5
support for slabs with 1 slot for largest sizes
2019-04-10 16:26:49 -04:00
6c31f6710a
support extended range of small size classes
2019-04-10 08:31:51 -04:00
d5f18c47b3
micro-optimize initialization with arenas
2019-04-10 08:07:24 -04:00
62c73d8b41
harden thread_arena check
2019-04-10 07:40:29 -04:00
d5c00b4d0d
disable current in-place growth code path for now
2019-04-09 19:20:34 -04:00
d5c1bca915
use round-robin assignment to arenas
...
The initial implementation was a temporary hack rather than a serious
implementation of random arena selection. It may still make sense to
offer it but it should be implemented via the CSPRNG instead of this
silly hack. It would also make sense to offer dynamic load balancing,
particularly with sched_getcpu().
This results in a much more predictable spread across arenas. This is
one place where randomization probably isn't a great idea because it
makes the benefits of arenas unpredictable in programs not creating a
massive number of threads. The security benefits of randomization for
this are also quite small. It's not certain that randomization is even a
net win for security since it's not random enough and can result in a
more interesting mix of threads in the same arena for an attacker if
they're able to attempt multiple attacks.
2019-04-09 16:54:14 -04:00
9a0de626fc
move stats accounting to utility functions
2019-04-09 03:57:44 -04:00
9453332e57
remove redundant else block
2019-04-09 00:06:17 -04:00
a4cff7a960
factor out slab memory_set_name into label_slab
2019-04-07 18:02:56 -04:00
ef90f404a6
add sanity check for stats option
2019-04-07 09:06:03 -04:00
e0891c8cfc
implement the option of large size classes
...
This extends the size class scheme used for slab allocations to large
allocations. This drastically improves performance for many real world
programs using incremental realloc growth instead of using proper growth
factors. There are 4 size classes for every doubling in size, resulting
in a worst case of ~20% extra virtual memory being reserved and a huge
increase in performance for pathological cases. For example, growing
from 4MiB to 8MiB by calling realloc in increments of 32 bytes will only
need to do work beyond looking up the size 4 times instead of 1024 times
with 4096 byte granularity.
2019-04-07 08:52:17 -04:00
c68de6141d
factor out duplicated code in malloc/realloc
2019-04-07 05:48:10 -04:00
ce36d0c826
split out allocate_large function
2019-04-07 05:44:09 -04:00
3d18fb8074
implement Android M_PURGE mallopt via malloc_trim
2019-04-07 03:35:26 -04:00
4f08e40fe5
move thread sealing implementation
2019-04-07 00:50:26 -04:00
55891357ff
clean up the exported API section of the code
2019-04-07 00:36:53 -04:00
491ce6b0b1
no need to provide valloc and pvalloc on Android
2019-04-07 00:31:09 -04:00
1eed432b9a
limit more glibc cruft to that environment
2019-04-07 00:30:05 -04:00
27a4c883ce
extend stats with nmalloc and ndalloc
2019-04-06 23:19:03 -04:00
e94fe50a0d
include zero byte size class in stats
...
The allocations don't consume any actual memory, but it does still use
up the virtual memory assigned to the size class and requires metadata.
2019-04-06 22:43:56 -04:00
712748aaa8
add implementation of Android mallinfo extensions
...
These are used internally by Bionic to implement malloc_info.
2019-04-06 22:39:01 -04:00
0f107cd2a3
only provide malloc_info stub for glibc
...
This has a proper implementation in Bionic outside of the malloc
implementation via the extended mallinfo API.
2019-04-06 22:01:12 -04:00
350d0e5fd2
add real mallinfo implementation for Android
...
Android Q uses the mallinfo implementation in the ART GC:
c220f981801575267302
2019-04-06 20:54:26 -04:00
df9650fe64
conditionally include threads.h
2019-03-26 01:28:27 -04:00
98deb9de52
relabel malloc read-only after init data
2019-03-25 20:34:10 -04:00
fc8f2c3b60
move pthread_atfork wrapper to util header
2019-03-25 17:16:52 -04:00
b5187a0aff
only use __register_atfork hack for old glibc
2019-03-25 17:16:22 -04:00
c5e911419d
add initial implementation of arenas
2019-03-25 14:59:50 -04:00
55769496dc
move hash_page to pages.h
2019-03-25 14:54:22 -04:00
13de480bde
rename quarantine bitmap field for clarity
2019-03-24 20:24:40 -04:00
3d142eb4c2
relabel large allocation guards when shrinking
2019-03-23 23:01:12 -04:00
64dfd23f7b
relabel purged slabs
2019-03-23 22:59:59 -04:00
178ec6e3f9
relabel quarantined large allocation regions
2019-03-23 22:57:19 -04:00
6e67106882
label malloc slab region gaps
2019-03-23 22:54:56 -04:00
1d62075291
label allocate_aligned_pages mappings
2019-03-23 22:29:04 -04:00
45337ebe07
label allocate_pages mappings
2019-03-22 23:17:38 -04:00
65311a5df2
relabel region table mapping
2019-03-22 21:59:44 -04:00
4a000d96e2
pkey state is now preserved on fork for Linux 5.0+
...
This patch is going to be backported to stable kernels, so the check
could be expanded to allow recent enough stable kernel branches.
2019-03-20 11:05:31 -04:00
c9df70d934
add support for labelling memory regions
2019-02-13 13:34:33 -05:00
2d7882ec0e
remove redundant unseal / seal metadata
2019-01-08 17:01:56 -05:00
fa17f70a73
add more configuration sanity checks
2019-01-06 00:52:25 -05:00
57f115b33c
scale slab quarantine based on size
2019-01-02 14:52:13 -05:00
ccc2a86501
rename quarantine size -> length for clarity
2019-01-02 14:17:02 -05:00
bc2cb5c828
fix builds with both random and queue quarantine
2019-01-02 13:23:49 -05:00
5d3b299209
support random quarantine sizes other than 0 or 16
2019-01-02 13:04:40 -05:00
4dc3385bc0
fix calloc for nmemb == 0 and !ZERO_ON_FREE
2018-12-09 10:59:21 -05:00
830a7d338d
make class_region_size configurable
2018-12-05 09:38:13 -05:00
ced3b0e070
support Android malloc_usable_size
2018-12-05 07:59:35 -05:00
7917966eca
avoid unnecessary GNU pointer arithmetic extension
2018-12-05 07:59:35 -05:00
7f781af929
explicitly include system malloc.h
2018-12-05 07:59:35 -05:00
Daniel Micay