|
326b5ebf
|
2015-08-30T17:29:21
|
|
Poison freed objects such that double-free is detected
Previously we were setting refcount of freed objects to the inert value, which
was harmful because it caused further destroy()s of the freed object to NOT
call free() and hence hide the bug. Indeed, after eb0bf3ae6688b7 test-object
was double-free'ing objects and this was never caught on Linux. It only was
caught as crashing on Mac.
Now we poison refcount upon freeing and check that it's valid whenever reading
it. Makes test-object fail now.
|
|
3f174cd0
|
2015-03-28T00:49:33
|
|
Minor refactoring to the atomics implementation
s/atomic_int/atomic_int_impl/ and s/atomic_ptr/atomic_ptr_impl/
to bring it in par with hb_mutex_impl_t, then re-introduce
hb_atomic_int_t as a wrapper around hb_atomic_int_impl_t.
In hb_reference_count_t, make it clear the non-atomic get and set
are intentional due to nature of the cases they are used in
(comparison to -1 and the debug output/tracing).
|
|
|
0d2c2f23
|
2014-08-14T13:25:55
|
|
Simplify hb-object
|
|
|
9d861b81
|
2014-08-14T13:15:21
|
|
Really fix clang build this time
|
|
|
272226f2
|
2014-08-14T13:11:33
|
|
Revert "Fix build on clang after previous commit"
This reverts commit 9c80cbc87f23040a334e30c750f64523d9ef10c8.
|
|
|
9c80cbc8
|
2014-08-14T13:09:52
|
|
Fix build on clang after previous commit
|
|
|
a5a27073
|
2014-08-14T13:05:36
|
|
Rewrite this==NULL checks to avoid undefined behavior
Fixes https://code.google.com/p/chromium/issues/detail?id=403594
|
|
|
cd7ea4f7
|
2014-08-14T12:57:02
|
|
Make hb_object_t members private
In preparation for fixing:
https://code.google.com/p/chromium/issues/detail?id=403594
|
|
|
7babfe5a
|
2012-12-04T00:35:54
|
|
Move object mutext into the user-data array
We are not using it for anything lse it seems.
|
|
|
a1900114
|
2012-12-04T00:29:35
|
|
Remove unused functions
|
|
|
2cf30196
|
2012-06-09T14:58:01
|
|
Add hb_object_lock/unlock()
|
|
|
bd300df9
|
2012-06-08T20:35:18
|
|
Minor
|
|
|
6220e5fc
|
2012-06-06T03:30:09
|
|
Add ASSERT_POD for most objects
|
|
|
cdafe3a7
|
2012-06-05T16:34:49
|
|
Add gcc intrinsics implementations for atomic and mutex
|
|
|
0e253e97
|
2012-06-05T15:37:19
|
|
Add a mutex to object header
Removes one more static-initialization. A few more to go.
|
|
|
bf93b636
|
2012-06-05T14:17:32
|
|
Remove constructor from hb_prealloced_array_t
This was causing all object types to be non-POD and have static
initializers. We don't need that!
Now, most nil objects just moved from .bss to .data. Fixing for that
coming soon.
|
|
|
3b8fd9c4
|
2012-06-03T15:54:19
|
|
Remove const from ref_count.ref_count
According to Tom Hacohen this was breaking build with some compilers.
In file included from hb-buffer-private.hh:35:0,
from hb-ot-map-private.hh:32,
from hb-ot-shape-private.hh:32,
from hb-ot-shape.cc:29:
hb-object-private.hh: In constructor '_hb_object_header_t::_hb_object_header_t()':
hb-object-private.hh:97:8: error: uninitialized const member in 'struct hb_reference_count_t'
hb-object-private.hh:51:25: note: 'hb_reference_count_t::ref_count' should be initialized
In file included from hb-ot-shape.cc:33:0:
hb-set-private.hh: In constructor '_hb_set_t::_hb_set_t()':
hb-set-private.hh:37:8: note: synthesized method '_hb_object_header_t::_hb_object_header_t()' first required here
hb-ot-shape.cc: In function 'void hb_ot_shape_glyphs_closure(hb_font_t*, hb_buffer_t*, const hb_feature_t*, unsigned int, hb_set_t*)':
hb-ot-shape.cc:521:12: note: synthesized method '_hb_set_t::_hb_set_t()' first required here
|
|
|
70600dbf
|
2012-06-03T15:52:51
|
|
Minor
|
|
|
0558d55b
|
2012-05-28T10:46:47
|
|
Remove hb_atomic_int_set/get()
We never use them in fact...
I'm just adjusting these as I better understand the requirements of
the code and the guarantees of each operation.
|
|
|
ec3ba4b9
|
2012-05-17T20:30:46
|
|
Move atomic ops into their own header
|
|
|
7bd2b04f
|
2012-05-11T03:40:58
|
|
Minor
|
|
|
cf26510d
|
2012-05-11T03:35:08
|
|
Some more...
Done. I promise.
|
|
|
bc71ad49
|
2012-03-01T17:30:29
|
|
Fix atomic-int op on Apple
The OSAtomicAdd32Barrier operator returns the new value, we want the
old value.
|
|
|
a1970d9a
|
2012-02-24T13:51:09
|
|
Add support for atomic int and mutex on Apple systems
So, apparently there's no atomic int 'get' method on Apple. You have to
add(0) to get. And that's not const-friendly. So switch inert-object
checking to a non-atomic get. This, however, is safe, and a negligible
performance boost too.
|
|
|
80044291
|
2012-02-23T18:47:17
|
|
Remove unused hb_atomic_int_set()
Apparently it can't be implemented on OS X. We weren't using it anyway.
|
|
|
bd7ff1de
|
2012-02-23T15:06:16
|
|
Allow disabling multi-threaded support
By defining HB_NO_MT.
Also, only warn once per missing MT feature support.
Mozilla Bug 666661 - gfx/harfbuzz/src/hb-prive.h - compiler warnings on mac
|
|
|
af921354
|
2011-10-21T09:18:43
|
|
Minor
|
|
|
89d89646
|
2011-10-17T11:50:54
|
|
Fix intrin.h b0rkage with older MSVC
Reported by Jonathan Kew.
|
|
|
d5476a30
|
2011-09-16T12:30:50
|
|
Minor
|
|
|
33ccc779
|
2011-08-09T00:43:24
|
|
[API] Make set_user_data() functions take a replace parameter
We need this to set data on objects safely without worrying that some
other thread unsets it by setting it at the same time.
|
|
|
c605bbbb
|
2011-08-04T20:00:53
|
|
Remove C++ guards from source files
Where causing issues for people with MSVC.
|
|
|
cc06c243
|
2011-07-25T20:25:44
|
|
Streamline debugging infrastructure even more
|
|
|
43ff203d
|
2011-07-25T17:35:24
|
|
Use variadic macros for debugging
Looks *so* much nicer!
|
|
|
decd4e3e
|
2011-07-25T16:47:02
|
|
Add sugar syntax for debug messages
Buffer debugging coming soon.
|
|
|
31f18abe
|
2011-06-15T09:49:58
|
|
Minor compiler warning fixes
|
|
|
b9452bfc
|
2011-06-14T14:47:07
|
|
Fix compiler warnings with -pedantic
|
|
|
df077fad
|
2011-05-12T01:19:39
|
|
[object] Make object inert during destruction
Such that user_data and other finalizers cannot resurrect object
|
|
|
45bfa990
|
2011-05-10T19:12:49
|
|
Fix set implementation to be truly threadsafe even with destroy() callbacks
The test/object test is passing again, instead of deadlocking.
|
|
|
46df6828
|
2011-05-05T15:33:19
|
|
Make user_data access threadsafe
For now, by taking a global user_data mutex.
|
|
|
218e67b9
|
2011-05-05T15:28:37
|
|
Shrink code
|
|
|
b45f32ee
|
2011-05-05T15:00:43
|
|
Use hb_array_t for hb_language_t mapping
|
|
|
21d2c92f
|
2011-05-05T14:47:53
|
|
Move code around
|
|
|
811482bd
|
2011-05-05T13:21:04
|
|
Replace hb_map_t with hb_set_t which is more intuitive and flexible
|
|
|
478a4253
|
2011-05-05T12:39:51
|
|
Make array/map implementation more generic
|
|
|
56eb5ad6
|
2011-05-04T19:27:37
|
|
Move code around
Mutex (and Windows.h by extension) are fairly isolated now.
|
|
|
20001794
|
2011-05-03T00:49:06
|
|
Move Win32 thread-safety stuff to hb-object-private.h
The Win32 definitions for LONG, ULONG, etc conflicts with
hb-open-type.h. Avoid that by making sure hb-object-private.h
and hb-open-type.h are not included in the same compilation unit.
|
|
|
1cd5969f
|
2011-05-02T19:53:39
|
|
[object] Fix bug in get_user_data() implementation
|
|
|
852e08ec
|
2011-04-27T21:45:51
|
|
Move code around
|
|
|
29c67d3f
|
2011-04-27T21:22:32
|
|
Add initial implementation of user_data to objects
|
|
|
47e71d96
|
2011-04-27T16:38:03
|
|
[object] Remove unnecessary use of macros
|
|
|
ae008b90
|
2011-04-27T16:12:12
|
|
[object] Add tracing support back in
|
|
|
40a9b815
|
2011-04-27T01:48:56
|
|
Add TODO item
|
|
|
ec6f9c2f
|
2011-04-21T18:35:58
|
|
Further simplify object handling
|
|
|
fca368c4
|
2011-04-21T18:24:02
|
|
Add hb_object_header_t which is the common part of all objects
Makes way for adding arbitrary user_data support.
|
|
|
a9f24c80
|
2011-04-21T17:18:22
|
|
Move hb_reference_count_t to hb-private.h
|
|
|
2409d5f8
|
2011-04-21T17:14:28
|
|
Update Copyright headers
|
|
|
24229eb1
|
2011-04-21T16:55:17
|
|
Remove obsolete comment
Talking to Ryan Lortie, he thinks my comment doesn't make sense.
So I'm making the getter const. Note that g_atomic_int_get()
casts that away itself, so we don't need to worry about that
(which kinda makes me uncomfortable actually).
|
|
|
3e8bdbf9
|
2011-04-21T16:16:21
|
|
Cleanup hb_refrence_count_t
|
|
|
d18431b4
|
2011-04-20T18:59:10
|
|
Move hb_reference_count_t from macros to inline methods
|
|
|
c57d454a
|
2011-04-20T18:50:27
|
|
Rename all private sources and headers to C++ files
So we can liberally use the simple features of C++ that parts of the
codebase is already using.
|