- 22 Apr, 2015 2 commits
-
-
Andrew Hsieh authored
-
Duane Sand authored
Change-Id: Idb5956dad7f540ae2a7d90405d91de9e62e46cf5
-
- 07 Apr, 2015 1 commit
-
-
Andrew Hsieh authored
-
- 06 Apr, 2015 1 commit
-
-
Lai Wei-Chih authored
Change-Id: I7952d4c2b9655bca605c8a25027cd35d730dd540
-
- 01 Apr, 2015 2 commits
-
-
Andrew Hsieh authored
-
Andrew Hsieh authored
See b.android.com/161622 Change-Id: I2cbd1cc0dbee1337d40be1f28daafc1b2b392c91
-
- 31 Mar, 2015 1 commit
-
-
Andrew Hsieh authored
-
- 30 Mar, 2015 9 commits
-
-
Andrew Hsieh authored
To determin project path, ndk-build searches parent directory (recursively) if current doesn't have jni/Android.mk or AndroidManifest.xml. In that case, path is expanded to absolute, eg. C:\android-ndk\samples\hello-jni\\AndroidManifest.xml, and the \\ may fail subsequent $(wildcard ...). Remove trailing / in project path. See b.android.com/160584 Change-Id: I234b065df9bb10a10e183234dffd92299699566b
-
Andrew Hsieh authored
-
Andrew Hsieh authored
-
Andrew Hsieh authored
Change-Id: Ia05e441e5d4876dccb2658816c80b394ae81c834
-
Andrew Hsieh authored
3 more files: COPYING3, COPYING3.LIB and COPYING.RUNTIME Change-Id: I207c3d24ced01ab8f5b28fea07c1b0c49bf95656
-
Andrew Hsieh authored
-
Andrew Hsieh authored
-
Andrew Hsieh authored
-
Andrew Hsieh authored
-
- 26 Mar, 2015 5 commits
-
-
Yabin Cui authored
-
Andrew Hsieh authored
-
Yabin Cui authored
-
Yabin Cui authored
Change-Id: I2f7379326bb56af4a874d6e2e67b2ac0fa669e0d
-
- 24 Mar, 2015 2 commits
-
-
Elliott Hughes authored
-
Elliott Hughes authored
Change-Id: I5b7ebaa57defe5ce706a55d7a48b2406ba5bd224
-
- 23 Mar, 2015 1 commit
-
-
Andrew Hsieh authored
-
- 20 Mar, 2015 1 commit
-
-
Daniel Micay authored
This results in nearly all functions with the possibility of stack corruption getting stack canaries, because it applies to any function taking a reference to the frame or with a local array rather than just the functions with arrays larger than 8 bytes. It was developed for use in Chrome (and Chrome OS) and has also been adopted by various other distributions (Arch, Fedora, Ubuntu, etc). The code size increase ranges from ~1.5% to ~2.5%, compared to ~0.3% to ~0.7% with the more conservative switch. The increase in the performance loss is usually minimal. The overall size increase once everything other than C and C++ code is taken into account is minimal, and it greatly improves the mitigation of stack buffer overflow vulnerabilities. https://lwn.net/Articles/584225/ Change-Id: I646c511e4bc2f8143df4dac2408dbe50cca45fa1
-
- 19 Mar, 2015 1 commit
-
-
Steve Ellcey authored
are no longer needed. Change-Id: I2213255d80a223de33dd12b0be0179c59c27c739
-
- 04 Mar, 2015 2 commits
-
-
Dan Albert authored
-
Dan Albert authored
Change-Id: I03e7146e445cec3f99781c19c8a204f6bf4805cf
-
- 03 Mar, 2015 2 commits
-
-
Andrew Hsieh authored
-
Ilya Konstantinov authored
Fix the patch to apply on gcc revision 7e8c3dad070b469494f5e8e09ee73cd99bb84c12.
-
- 24 Feb, 2015 5 commits
-
-
Andrew Hsieh authored
Also compile all ld.gold with --enable-threads which adds two options --threads Run the linker multi-threaded --no-threads Do not run the linker multi-threaded [default] Change-Id: I6fd6fc9715e87bfa4984d9fad45b17dfd3198e4f
-
Andrew Hsieh authored
x86/x86_64 compilers since r10d stop adding -mstackrealign implicitly for performance reason: compilers assume stack is properly aligned BEFORE call, and manipulate stack pointer to observe this invariance already. This does assume stack pointer is properly aligned to begin with, and there is no rogue hand-coded assembly in the call chains... Unfortunately not all x86-based device call JNI with stack 16-byte aligned (eg. ASUS_T00F). This CL compiles x86 prebuilt STL libraries with -mstackrealign regardless. Note that x86 STL libraries in r10d already compiled with -mstackrealign Change-Id: I7a74799b5a5d845672b0206954829eb46e81fd78
-
Andrew Hsieh authored
Change-Id: I78abd8e818fbf212e8408275f7ebec6e975ea075
-
Andrew Hsieh authored
Fixes for mips64 and/or r6 are only backported to 2.25 Change-Id: I9f5ddb48fa880489144a7a15e04810dff9e53b90
-
Andrew Hsieh authored
-
- 13 Feb, 2015 4 commits
-
-
Andrew Hsieh authored
"adb shell cmd" always return 0 even though when cmd returns non-zero. Fix those instance of adb whose return values are checked to call adb_var_shell* instead. Without this fix, "adb shell test -e /system/bin/app_process32" always return 0 on pre-L devices, and causes ndk-gdb to use the non-existance app_process32 and fail Change-Id: I1e8925dc989009859cf5c02a2fb92655864f1d7b
-
Andrew Hsieh authored
-
Andrew Hsieh authored
-
Andrew Hsieh authored
-
- 02 Feb, 2015 1 commit
-
-
Fabrice de Gans-Riberi authored
Cherry-picked from chromium https://chromium.googlesource.com/android_tools/+/ea176cccda8edad9ca2cea99f27b1a28a5c222fe Change-Id: Ifdf27ef129ca0fdb47946f13c7749fca35ccfdb4
-