Matthew Parkinson
c02077a06a
Bug fix in umul code for 32-bit Win
2019-02-15 14:36:47 +00:00
Matthew Parkinson
e7d90966f6
Made the malloc tests run on Windows
2019-02-15 14:36:47 +00:00
Matthew Parkinson
ac8e5e4e5b
Fix comment to represent current PAL API.
2019-02-15 09:24:01 +00:00
Matthew Parkinson
bf58015c25
Merge pull request #20 from Theodus/malloc-tests
...
Add malloc tests
2019-02-15 09:12:07 +00:00
Theo Butler
a3fb2b9531
cleanup some tests
2019-02-14 22:31:45 -05:00
Theo Butler
d2909eed0a
Merge branch 'master' into malloc-tests
2019-02-14 22:30:48 -05:00
Theo Butler
90414f82f6
Merge pull request #22 from Microsoft/zero_bug
...
Pal zero bug alignment
2019-02-14 22:26:18 -05:00
Matthew Parkinson
adeecfb0f1
Added bug fix to alignment of using.
2019-02-14 20:08:04 +00:00
Matthew Parkinson
8b9a525500
Added zero test at threshold.
2019-02-14 19:59:36 +00:00
Theo Butler
c555bf7218
fix doc comments for notify_using
2019-02-13 17:15:06 -05:00
Matthew Parkinson
35803305f2
Merge pull request #19 from plietar/visibility
...
Make internal symbols hidden.
2019-02-13 21:50:00 +00:00
Theo Butler
057595a57e
remove -1 check in calloc
2019-02-13 16:03:38 -05:00
Theo Butler
942313fec7
replace asserts with abort
2019-02-13 10:03:41 -05:00
Theo Butler
24ba068dce
remove randomness from malloc tests
2019-02-13 08:42:51 -05:00
Theo Butler
0ff2301083
use malloc shim in malloc tests
2019-02-13 08:12:07 -05:00
Theo Butler
e42551bb15
only build malloc tests on POSIX platforms
2019-02-12 22:05:36 -05:00
theodus
a9615bc6aa
mark variables unused in release build
2019-02-12 20:22:10 -05:00
Theo Butler
f6497e6ee3
Add malloc tests
2019-02-12 19:50:15 -05:00
Paul Liétar
8d8761f248
Make internal symbols hidden.
...
Only the malloc/free/... interface is exported.
This allows all of C++'s weak ODR symbols to be resolved at
compile-time, rather than staying weak and resolved at load-time.
Since internal calls don't need to go through the PLT anymore, we get a
nice speedup, at least on small objects.
2019-02-12 23:27:42 +00:00
Matthew Parkinson
a7f0695776
Merge pull request #17 from plietar/fix-align
...
Fix the condition on when to allocate a new block.
2019-02-12 13:10:53 +00:00
Paul Liétar
79b432ddab
Fix the condition on when to allocate a new block.
...
Previously we would pretty much always allocate a new block for each
allocator, making them 16MiB aligned.
2019-02-12 12:47:22 +00:00
Matthew Parkinson
e004641cec
Added to comment.
2019-02-11 14:47:12 +00:00
Matthew Parkinson
9981bad9b5
clang format
2019-02-11 14:47:12 +00:00
Matthew Parkinson
1272078487
Simplified size class calculation
...
The rounding code can be integrated directly into the sizeclass code,
which means, the code paths can be simplified slightly.
The special case for zero mantissa bits is not required, as the compiler
generates the same code from the more complex path due to
eliminating dead code after folding the constant zero through.
2019-02-11 14:47:12 +00:00
Matthew Parkinson
fdcad7feb4
Add test for sizeclass calculations
2019-02-11 14:47:12 +00:00
David Chisnall
ccd8ec9b4f
Disable rtti / exceptions on the non-Windows builds.
...
RTTI makes the binaries bigger, exceptions make life harder for
optimisers. Neither are actually used.
This doesn't preclude anything #including snmalloc.h with RTTI enabled,
it just disables it in the shim libraries and tests.
The FreeBSD libc builds were already doing this with no ill effect.
2019-02-06 17:46:58 +00:00
Matthew Parkinson
2cd84c80b2
Build shim with 1MiB superslabs.
2019-02-06 17:08:47 +00:00
Matthew Parkinson
731757b29c
Merge pull request #13 from Microsoft/matt-refactor
...
Simplification around remote allocation
2019-02-06 11:25:07 +00:00
Matthew Parkinson
f3897dd3e0
Code review feedback.
2019-02-05 16:57:09 +00:00
Matthew Parkinson
83c55fe5da
Simplify remote allocator.
...
Don't store the sizeclass, as we can find it easily, and already are
going to touch though cache lines in the common case.
2019-02-05 13:27:19 +00:00
Matthew Parkinson
122a5de811
Refactor code to correctly handle low bits of allocator ID.
2019-02-05 13:11:55 +00:00
David Chisnall
fc24b8be3d
Merge pull request #12 from Theodus/override
...
Avoid unnecessary allocation from realloc
2019-02-03 13:30:21 +00:00
theodus
ae3f9f6aec
Get alloc size without wrapper
2019-02-03 08:19:11 -05:00
theodus
2b8b4c884f
Do not realloc if size is in the same sizeclass
2019-02-02 17:28:02 -05:00
theodus
85cf5dd097
Avoid unnecessary allocation from realloc
2019-02-02 14:47:06 -05:00
Matthew Parkinson
c0ceb6e08f
Merge pull request #11 from Theodus/master
...
Fix assertion failure in memalign
2019-01-30 17:36:13 +00:00
theodus
61b310d9d6
Fix assertion failure in memalign
2019-01-30 12:14:16 -05:00
Matthew Parkinson
a838372989
Merge pull request #10 from Microsoft/hardening
...
Hardening allocator
2019-01-25 23:16:30 +00:00
Matthew Parkinson
ec59820970
Use placement new instead of init for metaslab.
2019-01-25 19:46:24 +00:00
Matthew Parkinson
06ff1ffe0a
Renamed zero to init for metaslab.
2019-01-25 16:38:10 +00:00
Matthew Parkinson
e5c40c1231
Remove an auto as not correctly inferred.
2019-01-24 11:48:02 +00:00
Matthew Parkinson
d5fb7150a3
Clang format
2019-01-24 11:39:19 +00:00
Matthew Parkinson
294efe8e03
Hardening access to meta data
...
Introduce a wrapper that ensure various fields of meta data, represent
valid indexes into the the address space, and not beyond the current
slab/superslab.
2019-01-24 11:39:18 +00:00
Matthew Parkinson
2083b29c9b
Remove a raw pointer usage
...
Metaslab can be handled as as ref, rather than a raw pointer.
2019-01-24 11:39:18 +00:00
Matthew Parkinson
e30c94cb2a
Apply ModArray to Metaslab data
...
Prevent bugs accessing outside the Slabs meta data.
2019-01-24 11:39:18 +00:00
Matthew Parkinson
5b81caac87
Merge pull request #9 from Microsoft/two_alloc_types
...
Add example for including the allocator with different settings in two compilation units.
2019-01-24 11:38:20 +00:00
Matthew Parkinson
67fbcba653
Clang format
2019-01-24 11:22:44 +00:00
Matthew Parkinson
28a28bb1ed
Simple example of using two allocators in one application.
...
Example to simulate using two allocators in one application. This will
be useful for scenarios in openenclave.
2019-01-23 13:19:46 +00:00
Matthew Parkinson
5240b3f8c6
Formatting
2019-01-23 11:06:07 +00:00
Matthew Parkinson
ef3a578503
Allow __je_bootstrap methods to not be compiled
...
The __je_bootstrap.. allocator methods can be specified that they
shouldn't be compiled using NO_BOOTSTRAP_ALLOCATION.
2019-01-23 11:05:50 +00:00