6ad2079 | Alon Zakai | 11 January 2016, 00:42:01 UTC | simplify x != 0 to x when the output is a boolean ; 1.35.20 | 11 January 2016, 01:12:06 UTC |
904f5d7 | Jukka Jylänki | 10 January 2016, 19:47:16 UTC | Properly skip test browser.test_pthread_custom_pthread_main_url on browsers that do not support SharedArrayBuffer. | 10 January 2016, 19:47:16 UTC |
85ec99c | Alon Zakai | 09 January 2016, 22:12:54 UTC | improve atomicrmw_i64 testing | 09 January 2016, 22:12:54 UTC |
0de8794 | juj | 08 January 2016, 22:27:45 UTC | Merge pull request #4028 from kripken/atomicrmw_i64 Add atomicrmw i64 test | 08 January 2016, 22:27:45 UTC |
fdc2a38 | Alon Zakai | 08 January 2016, 20:38:54 UTC | fix atomicrmw i64 and add testing #4025 | 08 January 2016, 21:52:47 UTC |
f2377f8 | Alon Zakai | 08 January 2016, 18:35:41 UTC | Merge pull request #4027 from juj/missing_zlib_include missing_zlib_include | 08 January 2016, 18:35:41 UTC |
d67b514 | Jukka Jylänki | 08 January 2016, 18:30:36 UTC | Add missing include that causes zlib test to fail on missing lseek() and read() declarations. | 08 January 2016, 18:30:36 UTC |
714b546 | Jukka Jylänki | 08 January 2016, 18:12:30 UTC | Add python2/3 shell execution wrappers for emcmake, emconfigure, emmake and emar, which are called via the shell on OSX. | 08 January 2016, 18:20:31 UTC |
3cee0f2 | juj | 08 January 2016, 12:03:27 UTC | Merge pull request #4024 from juj/futex_wait_in_main_thread futex_wait_in_main_thread | 08 January 2016, 12:03:27 UTC |
34322f2 | Jukka Jylänki | 08 January 2016, 02:33:15 UTC | In https://bugzilla.mozilla.org/show_bug.cgi?id=1231333, Atomics.futexWait() has been disallowed from the main browser thread in the SAB specification. Update the futex API to account for that and make main thread futex waits perform a spinloop instead. | 08 January 2016, 02:33:15 UTC |
84d9559 | Alon Zakai | 08 January 2016, 01:21:10 UTC | 1.35.19 | 08 January 2016, 01:21:10 UTC |
b270f7b | Alon Zakai | 08 January 2016, 01:19:22 UTC | 1.35.18 | 08 January 2016, 01:19:22 UTC |
8d9e8c6 | Alon Zakai | 08 January 2016, 01:16:37 UTC | fix other.test_incorrect_static_call | 08 January 2016, 01:16:37 UTC |
9eb7f6f | Alon Zakai | 07 January 2016, 23:13:26 UTC | Merge pull request #4022 from juj/wildcards_in_test_suite Wildcards in test suite | 07 January 2016, 23:13:26 UTC |
1d6c7cd | Jukka Jylänki | 07 January 2016, 23:06:49 UTC | Also update HTML docs on running the tests with wildcards and skipping. | 07 January 2016, 23:06:49 UTC |
84ce40b | Jukka Jylänki | 07 January 2016, 23:00:17 UTC | Simplify population of all test names for wildcard processing. Show a clearly readable error message if selected wildcard set doesn't match anything. | 07 January 2016, 23:00:17 UTC |
67f30bb | Alon Zakai | 07 January 2016, 22:27:26 UTC | Merge branch 'remove_console_log_lz4' of https://github.com/anlambert/emscripten into incoming | 07 January 2016, 22:27:26 UTC |
8e3f1e3 | Alon Zakai | 07 January 2016, 20:45:19 UTC | add note on modularize | 07 January 2016, 22:27:17 UTC |
ca49966 | Antoine Lambert | 07 January 2016, 17:18:42 UTC | remove useless call to console.log | 07 January 2016, 17:18:42 UTC |
61ce920 | Jukka Jylänki | 07 January 2016, 14:40:32 UTC | Clean up the _zzz_ prefixes on the pthread tests in the browser test suite. | 07 January 2016, 14:40:32 UTC |
324114b | Jukka Jylänki | 07 January 2016, 14:39:11 UTC | Allow passing wildcards to test runner "skip:" command, e.g. "python tests/runner.py browser skip:browser.test_pthread_*" will run all tests except the pthread ones. | 07 January 2016, 14:39:11 UTC |
e033398 | Jukka Jylänki | 07 January 2016, 14:32:18 UTC | Add support for passing a wildcard to the test runner to run a specific subset of tests in a suite, e.g. "python tests/runner.py browser.test_pthread_*". | 07 January 2016, 14:32:18 UTC |
d9010c4 | Alon Zakai | 07 January 2016, 01:37:30 UTC | receives calls to __cxa_find_matching_catch that include the argument number, so that we do not need to overload that method | 07 January 2016, 01:37:47 UTC |
c179e73 | Alon Zakai | 07 January 2016, 01:37:27 UTC | add startsWith to compiler.js | 07 January 2016, 01:37:27 UTC |
d2039c6 | Alon Zakai | 06 January 2016, 21:22:33 UTC | Merge branch 'mktime_dst' of https://github.com/TBBle/emscripten into incoming | 06 January 2016, 21:22:33 UTC |
72f178f | Alon Zakai | 06 January 2016, 18:44:49 UTC | Merge branch 'patch-1' of https://github.com/andreasplesch/emscripten into incoming | 06 January 2016, 18:44:49 UTC |
55b1570 | Andreas Plesch | 06 January 2016, 18:39:50 UTC | webidl_test: do not export _main() in order to get rid of warnings during compilation. main() is not needed due to use as library. | 06 January 2016, 18:39:50 UTC |
61b82df | Andreas Plesch | 06 January 2016, 18:36:19 UTC | Update AUTHORS | 06 January 2016, 18:36:19 UTC |
cb3d416 | Alon Zakai | 06 January 2016, 18:14:38 UTC | remove some old non-asm tests | 06 January 2016, 18:14:38 UTC |
a17e0db | juj | 06 January 2016, 11:46:10 UTC | Merge pull request #3976 from juj/llvm_bitreverse_i32 llvm_bitreverse_i32 | 06 January 2016, 11:46:10 UTC |
88fecf4 | Jukka Jylänki | 19 December 2015, 15:03:08 UTC | Implement llvm_bitreverse_i32() which was recently introduced to upstream LLVM at http://reviews.llvm.org/D14569. | 05 January 2016, 22:01:10 UTC |
89fb223 | Alon Zakai | 05 January 2016, 17:46:04 UTC | update test_strcmp_uni for latest llvm optimizer features where memcmp can be computed statically, and the test previously was too specific to musl memcpy results | 05 January 2016, 17:46:36 UTC |
1373838 | juj | 05 January 2016, 15:10:59 UTC | Merge pull request #4006 from juj/optimize_pthread_self optimize_pthread_self | 05 January 2016, 15:10:59 UTC |
7e9c5ed | juj | 05 January 2016, 10:55:29 UTC | Merge pull request #4009 from TBBle/site-emscripten-fork-fix Fix instructions for adding personal Emscripten fork to the SDK | 05 January 2016, 10:55:29 UTC |
4b89ed1 | Paul "TBBle" Hampson | 05 January 2016, 09:48:04 UTC | Fix instructions for adding personal Emscripten fork to the SDK These now match the instructions for a fork of Fastcomp or Fastcomp-Clang, as was probably originally intended. | 05 January 2016, 09:48:04 UTC |
4836551 | Alon Zakai | 05 January 2016, 01:30:39 UTC | Merge branch 'getpeername' of https://github.com/TBBle/emscripten into incoming | 05 January 2016, 01:30:39 UTC |
cac9ec1 | Paul "TBBle" Hampson | 05 January 2016, 01:12:01 UTC | Clarify name of sockets.test_get*name_null tests These tests actually check the behaviour when called with an unconnected socket, so name the tests that way. | 05 January 2016, 01:12:01 UTC |
e3abfbe | Paul "TBBle" Hampson | 05 January 2016, 01:09:51 UTC | Remove extra newline before curly-brace | 05 January 2016, 01:09:51 UTC |
985a4db | Alon Zakai | 05 January 2016, 00:26:36 UTC | don't emit 'use asm' for memory growth, since browsers are removing that feature from asm.js | 05 January 2016, 00:30:27 UTC |
2138e9f | Alon Zakai | 04 January 2016, 23:29:23 UTC | strftime fixes #3993 | 04 January 2016, 23:29:23 UTC |
a8637a8 | Alon Zakai | 04 January 2016, 22:57:24 UTC | 1.35.17 | 04 January 2016, 22:57:24 UTC |
907b627 | Alon Zakai | 04 January 2016, 22:56:39 UTC | 1.35.16 | 04 January 2016, 22:56:39 UTC |
78214c4 | Jukka Jylänki | 04 January 2016, 22:05:10 UTC | Optimize potentially hot functions pthread_self(), emscripten_is_main_runtime_thread() and emscripten_is_main_browser_thread() to run inside the asm.js scope to avoid a FFI transition. Implements #3504. | 04 January 2016, 22:05:10 UTC |
4a2e44c | Jukka Jylänki | 04 January 2016, 21:21:36 UTC | Floating point variants for Atomic.{add, sub, exchange and compareExchange} do not exist, and therefore don't need to be looked for in -s USE_PTHREADS=2 build mode. | 04 January 2016, 21:21:36 UTC |
5eff6a8 | Jukka Jylänki | 04 January 2016, 21:18:44 UTC | Bug https://bugzilla.mozilla.org/show_bug.cgi?id=1141986 has been implemented, so Atomics.exchange() is now available, and therefore importable into the asm.js module. | 04 January 2016, 21:18:44 UTC |
45e7346 | Jukka Jylänki | 04 January 2016, 21:06:46 UTC | The SharedArrayBuffer spec draft no longer has an Atomics.fence() function as asm.js import, so when building with -s USE_PTHREADS=2, we do not need to attempt to patch it out. | 04 January 2016, 21:06:46 UTC |
aae5cad | Jukka Jylänki | 04 January 2016, 15:50:57 UTC | 1.35.15 | 04 January 2016, 15:50:57 UTC |
d3c384d | Jukka Jylanki | 04 January 2016, 14:37:26 UTC | Add support for building native optimizer with GCC 4.3 - 4.6 which did not yet have -std=c++11 but only the experimental -std=c++0x. See https://gcc.gnu.org/projects/cxx0x.html. | 04 January 2016, 14:37:49 UTC |
05f0ba0 | juj | 04 January 2016, 12:00:01 UTC | Merge pull request #3988 from juj/bool_simd_in_js_optimizer Add support for Bool SIMD types in js optimizer. | 04 January 2016, 12:00:01 UTC |
c4418a9 | juj | 04 January 2016, 11:30:44 UTC | Merge pull request #3971 from juj/threadprofiler --threadprofiler | 04 January 2016, 11:30:44 UTC |
1844b19 | Paul "TBBle" Hampson | 03 January 2016, 21:04:37 UTC | Fix mktime handling of tm_isdst #4001 mktime was calculating the dstOffset, but still using summerOffset and winterOffset directly for populating or correcting tm_isdst. | 03 January 2016, 21:07:44 UTC |
dd49b3a | Paul "TBBle" Hampson | 03 January 2016, 21:04:14 UTC | Test 366 days rather than 266 days | 03 January 2016, 21:04:14 UTC |
48fb29b | Paul "TBBle" Hampson | 03 January 2016, 09:36:25 UTC | Implement getpeername and fix getsockname #3997 The existing getsockname was actually returning the address of the other end of the connection, which is getpeername. getpeername can fail, if the destination address is not set ('connect' or 'accept' has never been called). getsockname returns the address of the local end of the connection, and cannot fail; an unbound socket will return '0.0.0.0:0'. However, at this point a socket that was auto-bound (i.e. has not had 'bind' called on it) is also returning '0.0.0.0:0' for getsockname, which is incorrect but at least consistently-behaved now. | 03 January 2016, 10:08:57 UTC |
37aadc4 | Paul "TBBle" Hampson | 03 January 2016, 09:29:34 UTC | Return input name from DNS.lookup_name when given 0.0.0.0 or :: #3997 DNS.lookup_name uses __inet_pton4_raw and __inet_pton6_raw to detect if it has been given a valid address string, but if the result evaluates to false, still attempted to generate a 172.29.0.0/16 IP address. Instead, correctly check the return values of the __inet_pton... methods against precisely null, as is done by the other users. | 03 January 2016, 10:08:08 UTC |
50d52f1 | Jukka Jylänki | 29 December 2015, 16:19:11 UTC | Move emscripten_conditional_set_current_thread_status(), emscripten_set_current_thread_status() and emscripten_set_thread_name() to asm.js side to avoid FFI calls when thread profiling is not enabled. | 29 December 2015, 16:22:57 UTC |
cb2c460 | Jukka Jylänki | 18 December 2015, 23:11:06 UTC | Implement --threadprofiler linker option which incorporates a thread activity profiler to the built page, similar to --cpuprofiler and --memoryprofiler. | 29 December 2015, 16:04:59 UTC |
7da6c29 | Jukka Jylänki | 29 December 2015, 13:33:59 UTC | Revert the removal of the function SIMD_Int32x4_fromBool64x2Bits() in the previous SIMD.js pull request, which is still needed for Float64x2 support. | 29 December 2015, 13:33:59 UTC |
233f7ef | Jukka Jylänki | 29 December 2015, 13:05:48 UTC | Merge remote-tracking branch 'remotes/n8o/incoming' into incoming # Conflicts: # AUTHORS | 29 December 2015, 13:05:48 UTC |
90fcf08 | Jukka Jylänki | 29 December 2015, 12:16:21 UTC | Add support for Bool SIMD types in js optimizer. | 29 December 2015, 12:16:21 UTC |
126812f | juj | 29 December 2015, 11:49:56 UTC | Merge pull request #3986 from juj/simd_js_spec_update Simd js spec update | 29 December 2015, 11:49:56 UTC |
ddf64b2 | juj | 28 December 2015, 20:27:52 UTC | Merge pull request #3987 from eska014/fix-emscripten-result-documentation Fix EMSCRIPTEN_RESULT documentation | 28 December 2015, 20:27:52 UTC |
2aeea43 | eska | 28 December 2015, 19:35:18 UTC | Fix EMSCRIPTEN_RESULT documentation | 28 December 2015, 19:35:18 UTC |
d893bf1 | Jukka Jylänki | 28 December 2015, 17:57:42 UTC | Add support for SIMD.js boolean types to fix asm.js validation on up-to-date spec implementation on FF Nightly. | 28 December 2015, 17:57:42 UTC |
3f8e80e | Jukka Jylänki | 28 December 2015, 17:20:18 UTC | Add boolean SIMD types to native optimizer. | 28 December 2015, 17:20:18 UTC |
a28e81d | Jukka Jylänki | 28 December 2015, 16:04:32 UTC | Implement float32x4 not, and, or and xor with casts under the hood for conveniency. | 28 December 2015, 16:04:32 UTC |
724efd7 | Jukka Jylänki | 28 December 2015, 15:35:27 UTC | SIMD.js operations and, or, xor and not are restricted to integer types. | 28 December 2015, 15:35:27 UTC |
8487502 | Alon Zakai | 27 December 2015, 01:19:44 UTC | fix webidl binder on methods that use the ensureCache and have no strings | 27 December 2015, 01:19:44 UTC |
1f6b09d | Chad Austin | 24 December 2015, 03:31:13 UTC | Fix #610 : add class_property method to class_ for binding static properties of classes and structs. | 24 December 2015, 03:31:13 UTC |
7689b60 | Alon Zakai | 23 December 2015, 20:39:06 UTC | receive all metadata from s2wasm | 23 December 2015, 20:39:06 UTC |
63d0624 | Alon Zakai | 23 December 2015, 20:05:45 UTC | send global base to s2wasm | 23 December 2015, 20:05:45 UTC |
a50b7a9 | Alon Zakai | 23 December 2015, 18:52:43 UTC | move binaryen STATIC_BUMP to right place | 23 December 2015, 18:52:43 UTC |
83dcd5d | Alon Zakai | 23 December 2015, 18:52:19 UTC | support s2wasm --debug | 23 December 2015, 18:52:19 UTC |
7b4069e | Alon Zakai | 22 December 2015, 21:36:11 UTC | support webidl arrays as input parameters in webidl binder | 22 December 2015, 21:36:29 UTC |
84a4848 | Alon Zakai | 22 December 2015, 20:49:55 UTC | refactor webidl code to more clearly special-case array attributes, and prepare for proper full typenames which inclue being an array | 22 December 2015, 21:36:29 UTC |
da50a41 | Jukka Jylänki | 22 December 2015, 16:00:31 UTC | Fix test_pthread_mandelbrot SSE usage on comparisons. | 22 December 2015, 16:33:07 UTC |
9c9f2a7 | Alon Zakai | 21 December 2015, 19:51:19 UTC | move test_link_response_file_does_not_force_absolute_paths to other | 21 December 2015, 19:51:19 UTC |
62fbf06 | Josh Peterson | 21 December 2015, 19:31:45 UTC | Add my name to the AUTHORS file As directed here: http://kripken.github.io/emscripten-site/docs/contributing/index.html I’ve added my name to the AUTHORS file to indicate the I’ve accepted the licenses. | 21 December 2015, 19:31:45 UTC |
a0b439b | Josh Peterson | 21 December 2015, 19:28:32 UTC | Prevent Building.link from forcing absolute paths In some cases, we have paths to objects files that we need to link which contain spaces. Since Building.link cannot create a response file that includes paths with spaces, it will add the object files to the command line. When we have too many object files, the command line gets too long, and we see this error on Windows: WindowsError: [Error 87] The parameter is incorrect We can work around this issue by giving Building.link relative paths to the object files (really just file names), and setting the working directory to by the path with the space. However, Building.link was unnecessarily converting these relative paths to absolute paths. This change prevents Building.link from making that conversion when it calls the linker. Instead it now uses whatever relative paths it was given. | 21 December 2015, 19:28:32 UTC |
71f223f | Alon Zakai | 18 December 2015, 03:29:18 UTC | update s2wasm metadata separator | 19 December 2015, 22:24:14 UTC |
1350873 | Alon Zakai | 19 December 2015, 22:23:37 UTC | Merge pull request #3975 from juj/win_import_emcc_py win_import_emcc_py. | 19 December 2015, 22:23:37 UTC |
5ae6f39 | Jukka Jylänki | 19 December 2015, 21:23:32 UTC | Reintroduce shared->non-shared casts of typed array views in various GL entry points, pending https://bugzilla.mozilla.org/show_bug.cgi?id=1232808. | 19 December 2015, 21:23:32 UTC |
11e0e7e | Jukka Jylänki | 19 December 2015, 14:21:36 UTC | Fix Windows multiprocessing/forking import of emcc and em++. Windows python 2.7 insists on finding em++.py and does not look up em++. Closes #3939. | 19 December 2015, 14:21:36 UTC |
d52939c | Nate Burr | 19 December 2015, 00:43:16 UTC | Fixed issues with the memory profiler * Fxed ui to listen for proper events and to be in proper scope. * Fixed the allocations stats to work properly and be in the correct scope. * Fixed onFree and onMalloc to reference the right functions in the correct scope. | 19 December 2015, 00:43:16 UTC |
7360392 | juj | 18 December 2015, 14:42:34 UTC | Merge pull request #3961 from Daft-Freak/patch-8 Fix glClearBufferfv | 18 December 2015, 14:42:34 UTC |
f9b521c | Alon Zakai | 18 December 2015, 02:00:23 UTC | add more rust testing, of array loads and stores | 18 December 2015, 03:01:18 UTC |
4154faf | Alon Zakai | 17 December 2015, 01:40:09 UTC | use a github ribbon | 17 December 2015, 01:40:09 UTC |
cadfdaf | Alon Zakai | 15 December 2015, 23:40:06 UTC | Merge branch 'incoming' of github.com:kripken/emscripten into incoming | 15 December 2015, 23:40:06 UTC |
48c7375 | Alon Zakai | 15 December 2015, 23:33:34 UTC | 1.35.14 | 15 December 2015, 23:33:34 UTC |
bcb2278 | Alon Zakai | 15 December 2015, 22:32:41 UTC | disable asm1.test_sse2_full, which after llvm update now hits our error on avoiding -O1 | 15 December 2015, 22:32:41 UTC |
92d995a | Alon Zakai | 15 December 2015, 20:33:59 UTC | Merge pull request #3965 from dreamlayers/packager-file-url Accept HTTP status 0 in packager, so file:// URLs work in Chrome | 15 December 2015, 20:33:59 UTC |
6407411 | Alon Zakai | 15 December 2015, 20:23:36 UTC | 1.35.13 | 15 December 2015, 20:23:36 UTC |
6f5ce2e | Alon Zakai | 15 December 2015, 19:40:25 UTC | improve test_dylink_i64 | 15 December 2015, 19:40:25 UTC |
a40ff6b | Alon Zakai | 15 December 2015, 19:31:55 UTC | fix i64Subtract validation in fastLong in side modules | 15 December 2015, 19:31:55 UTC |
eaa0bc6 | Alon Zakai | 15 December 2015, 03:24:20 UTC | parse s2wasm metadata | 15 December 2015, 03:34:07 UTC |
d217814 | Alon Zakai | 15 December 2015, 03:04:44 UTC | fix indentation in emscripten.py wasm code | 15 December 2015, 03:34:07 UTC |
63bddd0 | Alon Zakai | 15 December 2015, 01:43:23 UTC | fall back to staticAlloc if malloc is not present; useful in ONLY_MY_CODE | 15 December 2015, 03:34:07 UTC |
92aa2e9 | Alon Zakai | 15 December 2015, 01:39:09 UTC | fix declares in wasm glue | 15 December 2015, 03:34:07 UTC |
706de1f | Boris Gjenero | 15 December 2015, 02:30:10 UTC | Accept HTTP status 0 in packager, so file:// URLs work in Chrome | 15 December 2015, 02:31:15 UTC |
6f5df54 | Alon Zakai | 14 December 2015, 22:46:27 UTC | Merge pull request #3950 from aidanhs/aphs-useless-sdl2-image-include The header is copied into the sdl2 include path | 14 December 2015, 22:46:27 UTC |
38ddb92 | Alon Zakai | 14 December 2015, 18:47:09 UTC | Merge branch 'incoming' of https://github.com/awtcode/emscripten into incoming | 14 December 2015, 18:47:09 UTC |