Commit Graph

14446 Commits

Author SHA1 Message Date
Elliott Hughes 068ee4d03f am a0f02fa9: Merge "Fix the fs_config call."
* commit 'a0f02fa9e3ca779ed69264fc0977f530328c3f15':
  Fix the fs_config call.
2015-08-25 18:15:53 +00:00
Elliott Hughes 3ecba7f95c am 61ab1618: Merge "Fix 32-bit build."
* commit '61ab1618a8347c503e9185a4d17b957e367bac1b':
  Fix 32-bit build.
2015-08-25 18:15:51 +00:00
Elliott Hughes a0f02fa9e3 Merge "Fix the fs_config call." 2015-08-25 18:11:06 +00:00
Elliott Hughes 7baecbe6a3 Fix the fs_config call.
fs_config ORs in bits to the existing mode value.

Change-Id: I23f0655f9c4c8c8dc43ed979f3b519d7c03a11f8
2015-08-25 11:09:04 -07:00
Elliott Hughes 61ab1618a8 Merge "Fix 32-bit build." 2015-08-25 18:06:19 +00:00
Elliott Hughes 56bf309843 Fix 32-bit build.
fs_config uses "unsigned int" instead of mode_t. Work around this.

Change-Id: I57b97d5bfc1b88089a41b4e8a50f1391b5c5113b
2015-08-25 11:01:39 -07:00
Elliott Hughes ba8fc8293f am d3c72606: Merge "Clean up error handling/reporting in file_sync_service.cpp."
* commit 'd3c726064aa8e676ca4756e8461a5732b16aa7a9':
  Clean up error handling/reporting in file_sync_service.cpp.
2015-08-25 17:50:51 +00:00
Elliott Hughes d3c726064a Merge "Clean up error handling/reporting in file_sync_service.cpp." 2015-08-25 17:46:02 +00:00
Elliott Hughes fdd4e58868 Clean up error handling/reporting in file_sync_service.cpp.
In an earlier code review it was pointed out that there was something
very weird about fail_errno. It didn't seem to make sense that we'd
often try to continue after reporting failure. This patch cleans up
all that and assumes that if we've reported failure to the client,
we should stop what we're doing.

Bug: http://b/23437039
Change-Id: I39c38650ed9f9d5e30adbf68a7545c9e4a6ab812
2015-08-25 08:41:05 -07:00
Elliott Hughes e55e820e66 am e13fd9aa: Merge "lsof fixes."
* commit 'e13fd9aa231b117b03be6b84b7fb0debe6cb620d':
  lsof fixes.
2015-08-25 00:46:24 +00:00
Elliott Hughes e13fd9aa23 Merge "lsof fixes." 2015-08-25 00:37:49 +00:00
Elliott Hughes 703a3f1d07 lsof fixes.
Change-Id: Ie3befc7c8e3540b9ebe93786e711f8445cf6f6c7
2015-08-24 17:21:31 -07:00
Tom Cherry 9b2bc34765 am 77d634f7: Merge "remove TARGET_PROVIDES_INIT_RC"
* commit '77d634f7dddf983605619f83bbacbe8ff27a23bc':
  remove TARGET_PROVIDES_INIT_RC
2015-08-24 23:28:36 +00:00
Tom Cherry 77d634f7dd Merge "remove TARGET_PROVIDES_INIT_RC" 2015-08-24 23:23:36 +00:00
Elliott Hughes 6c04e9a9ec am 42640e52: Merge "Add missing #include <memory> for std::unique_ptr on Windows."
* commit '42640e5224a861690de006952470451d18337256':
  Add missing #include <memory> for std::unique_ptr on Windows.
2015-08-24 21:56:37 +00:00
Elliott Hughes 42640e5224 Merge "Add missing #include <memory> for std::unique_ptr on Windows." 2015-08-24 21:50:26 +00:00
Elliott Hughes a925dba0e0 Add missing #include <memory> for std::unique_ptr on Windows.
Change-Id: Ic02f53003a20a646cbbf2aa08e0e10be9bd8b97f
2015-08-24 14:49:43 -07:00
Nick Kralevich 7ec4c228ee am e80ec0c5: Merge "libutils: Enable LOCAL_SANITIZE := integer"
* commit 'e80ec0c5609d509b17e3e8497a0e0ec5b9caa2e4':
  libutils: Enable LOCAL_SANITIZE := integer
2015-08-24 21:45:36 +00:00
Nick Kralevich e80ec0c560 Merge "libutils: Enable LOCAL_SANITIZE := integer" 2015-08-24 21:39:28 +00:00
Elliott Hughes 5c027334cc am 93b9e865: Merge "There are no big endian hosts, grandpa."
* commit '93b9e8653e95d711b4df163febd662fc022de72a':
  There are no big endian hosts, grandpa.
2015-08-24 21:36:56 +00:00
Elliott Hughes 93b9e8653e Merge "There are no big endian hosts, grandpa." 2015-08-24 21:29:31 +00:00
Elliott Hughes f4465203b6 There are no big endian hosts, grandpa.
Change-Id: I6f64f702f919fe4af10d82c5f395a051571815ed
2015-08-24 14:27:03 -07:00
Tom Cherry 0ea4ffdc06 remove TARGET_PROVIDES_INIT_RC
Change-Id: I15105a8dc11e218d95d4f5a4cee5b2a166d3d639
2015-08-24 13:56:43 -07:00
Elliott Hughes c583c5112b am 186f1dee: Merge "adb sync cleanup."
* commit '186f1dee2afc15c8941a1f78e1aa97e2ec9970ee':
  adb sync cleanup.
2015-08-24 20:51:42 +00:00
Elliott Hughes 186f1dee2a Merge "adb sync cleanup." 2015-08-24 20:48:01 +00:00
Elliott Hughes aa24549190 adb sync cleanup.
We can double the speed of "adb sync" (on N9) if we increase SYNC_DATA_MAX
from 64KiB to 256KiB. This change doesn't do that, because I still haven't
managed to plumb through the information about whether we're a new adb/adbd
to file_sync_client.cpp and file_sync_service.cpp. But this is already a big
change with a lot of cleanup, so let's do the cleanup and worry about the
intended change another day...

This change does improve performance somewhat by halving the number of
lstat(2) calls made on the client side, and ensuring that most packets are
sent with a single write. This has the pleasing result of making the null
sync on an AOSP N9 go from just over 300ms to around 100ms, which means it
now seems instantaneous (https://en.wikipedia.org/wiki/Mental_chronometry).

Change-Id: If9f6d4c1f93ec752b95f71211bbbb1c513045166
2015-08-24 11:15:01 -07:00
Nick Kralevich 4d587f2f9a libutils: Enable LOCAL_SANITIZE := integer
Call abort() on undefined or sketchy integer behavior.
Protects against integer overflow attacks.

Change-Id: I475e5e9fb04160dd2d201e926ab2b897539f8bb6
2015-08-24 10:51:38 -07:00
Badhri Jagan Sridharan 46f24955c8 am e3c02883: Merge "adbd: enable USB SuperSpeed (again)"
* commit 'e3c028836f892e74c737c8b16ff96321cee82d25':
  adbd: enable USB SuperSpeed (again)
2015-08-24 17:41:04 +00:00
Badhri Jagan Sridharan e3c028836f Merge "adbd: enable USB SuperSpeed (again)" 2015-08-24 17:33:20 +00:00
Nick Kralevich 4f0a7cb8e1 am 786e5e26: Merge "libutils: cleanups for -fsanitize=integer"
* commit '786e5e263bdc716f4b2d871191e64726e552acdb':
  libutils: cleanups for -fsanitize=integer
2015-08-24 15:56:42 +00:00
Nick Kralevich 786e5e263b Merge "libutils: cleanups for -fsanitize=integer" 2015-08-24 15:52:57 +00:00
Nick Kralevich 1f28698998 libutils: cleanups for -fsanitize=integer
* Hashing functions rely on integer overflow behavior.
Mark those functions as safe.

* abort() if a passed in size_t value is greater than
UINT32_MAX. This can occur on 64 bit builds where size_t
is larger than uint32_t.

* Special case the index lookup for an empty sorted vector.
Without the special case, size() == 0, and size()-1 underflows.

Change-Id: I343a14b589fc8f0d221c1998ae5d6f0b9e2781f8
2015-08-24 15:52:22 +00:00
Siva Velusamy 5b476736da am 9498fc68: Merge "Use glog for logging in the OSX USB layer"
* commit '9498fc68ec4a6ddf29cf471225eb87dbcb908e3d':
  Use glog for logging in the OSX USB layer
2015-08-24 15:43:16 +00:00
Siva Velusamy 9498fc68ec Merge "Use glog for logging in the OSX USB layer" 2015-08-24 15:35:51 +00:00
Siva Velusamy 743883b570 Use glog for logging in the OSX USB layer
Change-Id: I8a3a4dfc768fdd335f03dc582c85b410c808aeb7
2015-08-24 08:22:15 -07:00
Steve Fung df56400935 am f87f5f90: Merge "crash_reporter: Fix tmp directory location"
* commit 'f87f5f90c56923e88c2a6728d919dabaab0e4177':
  crash_reporter: Fix tmp directory location
2015-08-24 04:21:57 +00:00
Steve Fung 7c6f31293c am ea1e07e9: Merge "crash_reporter: Relicense as Apache 2."
* commit 'ea1e07e94020deca44858108e7a8d0915be36168':
  crash_reporter: Relicense as Apache 2.
2015-08-24 04:21:55 +00:00
Steve Fung f87f5f90c5 Merge "crash_reporter: Fix tmp directory location" 2015-08-24 03:26:21 +00:00
Steve Fung ea1e07e940 Merge "crash_reporter: Relicense as Apache 2." 2015-08-24 01:03:14 +00:00
Steve Fung da98133cc0 crash_reporter: Fix tmp directory location
It is unsafe to use /data/local/tmp.  Use our own temporary
directory under /data/misc/crash_reporter.

Bug: 23430484
Change-Id: I186ba6b146bd884064177a3bb29b18b8edcc0b62
2015-08-23 17:57:19 -07:00
Yusuke Sato eaf04aa098 am 3b94cef1: Merge "Add FORK_EXECVP_OPTION_CAPTURE_OUTPUT to logwrap.h"
* commit '3b94cef1b75893f98ad3011a09582405bb32baf4':
  Add FORK_EXECVP_OPTION_CAPTURE_OUTPUT to logwrap.h
2015-08-22 02:35:08 +00:00
Yusuke Sato 3b94cef1b7 Merge "Add FORK_EXECVP_OPTION_CAPTURE_OUTPUT to logwrap.h" 2015-08-22 02:27:46 +00:00
Yusuke Sato 7980426a6f Add FORK_EXECVP_OPTION_CAPTURE_OUTPUT to logwrap.h
This allows raw popen calls in e.g. system/netd/ to be replaced
with android_fork_execvp_ext().

Change-Id: I159ece7369fa38ff8782024bef0d7cfafe74ecee
2015-08-21 17:14:59 -07:00
Chih-hung Hsieh f91d63e2b8 am 34d0e782: Merge "To compile with llvm integrated assembler."
* commit '34d0e782aff894a649f2672e7e9df7c7ccab8377':
  To compile with llvm integrated assembler.
2015-08-21 23:48:49 +00:00
Chih-hung Hsieh 34d0e782af Merge "To compile with llvm integrated assembler." 2015-08-21 23:42:31 +00:00
Steve Fung 975f85110b am 2e2dc003: Merge "crash_reporter: Fix path to dbus-send"
* commit '2e2dc0039950014d0948589899517e065a82a8b4':
  crash_reporter: Fix path to dbus-send
2015-08-21 23:29:35 +00:00
Steve Fung 2e2dc00399 Merge "crash_reporter: Fix path to dbus-send" 2015-08-21 23:23:34 +00:00
Chih-Hung Hsieh 7244cf2aa5 To compile with llvm integrated assembler.
* Explicitly specify default .align 0.
* Use standard ldmfdlo instruction.
* Before and after gas outputs are identical,
  with align 0 sections.
* Objdump showed .text/.data/.bss section alignment
  attributes are 2^0 from gas and 2^2 from llvm
  assembler. These .S files might be working when
  compiled by gas, but llvm assembler's output should
  be more correct or conservative.

Change-Id: I4e578dbc8155c0d06d1bbc1c33ec4cc851a18479
2015-08-21 15:58:26 -07:00
Steve Fung 916a5c8624 am 0f6c07b7: Merge "crash_reporter: Add Android.mk"
* commit '0f6c07b7b89f77927e94d6f87228eb5b15c607eb':
  crash_reporter: Add Android.mk
2015-08-21 22:14:55 +00:00
Steve Fung 0f6c07b7b8 Merge "crash_reporter: Add Android.mk" 2015-08-21 22:08:32 +00:00