OSDN Git Service
Conley Owens [Thu, 5 May 2011 20:22:18 +0000 (13:22 -0700)]
Merge "Add missing clean-up of idmap file descriptors."
Mårten Kongstad [Thu, 5 May 2011 08:40:42 +0000 (10:40 +0200)]
Add missing clean-up of idmap file descriptors.
Change-Id: I9bdc9a4b7962f1a8dce77f4b213c8b9dc26e4b0f
Conley Owens [Mon, 2 May 2011 16:53:25 +0000 (09:53 -0700)]
Merge "Removing dead store; it was likely a relict of debuging code."
Conley Owens [Fri, 29 Apr 2011 17:01:23 +0000 (10:01 -0700)]
Merge "Adjust mBiggerTouchSlopSquare to the suitable value"
Conley Owens [Fri, 29 Apr 2011 17:00:26 +0000 (10:00 -0700)]
Merge "Scolling using arrow keys with padding"
Conley Owens [Fri, 29 Apr 2011 17:00:20 +0000 (10:00 -0700)]
Merge "CHAR_SEQUENCE_CREATOR cannot handle null string"
Conley Owens [Fri, 29 Apr 2011 17:00:12 +0000 (10:00 -0700)]
Merge "Fixed to draw texture when loadTexture() is called at first time"
Conley Owens [Fri, 29 Apr 2011 17:00:06 +0000 (10:00 -0700)]
Merge "Fixing wrong Japanese phone formatter."
Conley Owens [Fri, 29 Apr 2011 17:00:00 +0000 (10:00 -0700)]
Merge "Fix memory leak of SkMovie class"
Conley Owens [Fri, 29 Apr 2011 16:59:55 +0000 (09:59 -0700)]
Merge "Implement support for ALT and SHIFT modifiers"
Conley Owens [Fri, 29 Apr 2011 16:59:48 +0000 (09:59 -0700)]
Merge "Fix startSearch condition on Dialog, because pressing search key causes ANR popup. If search key is pressed on Dialog and there is no search item then do nothing. How to reproduce the issue: Settings > About phone > Legal information > Google legal > Press search key > Touch list item or press back key > ANR popup is shown"
Conley Owens [Fri, 29 Apr 2011 16:59:42 +0000 (09:59 -0700)]
Merge "modify the device to enable to get the state as slide-open/slide-close."
Conley Owens [Fri, 29 Apr 2011 16:59:36 +0000 (09:59 -0700)]
Merge "Adding missing breaks in BnSurfaceComposer::onTransact"
Conley Owens [Fri, 29 Apr 2011 16:59:30 +0000 (09:59 -0700)]
Merge "Fixed to call computeVisibleRegions() when transparent value changed"
Conley Owens [Fri, 29 Apr 2011 16:59:22 +0000 (09:59 -0700)]
Merge "frameworks/base: acquire lock on am only when needed"
Conley Owens [Fri, 29 Apr 2011 16:56:58 +0000 (09:56 -0700)]
Merge "Adding a sanity test in the while loop in readHistory()"
Conley Owens [Fri, 29 Apr 2011 16:48:10 +0000 (09:48 -0700)]
Merge "Improved conversion accuracy of exif geotag data"
Conley Owens [Fri, 29 Apr 2011 16:48:04 +0000 (09:48 -0700)]
Merge "Fixes broken image link in "Tools -> Traceview" documentation page."
Jozef BABJAK [Fri, 29 Apr 2011 11:00:04 +0000 (13:00 +0200)]
Removing dead store; it was likely a relict of debuging code.
Change-Id: I685c6df119212f48f90dd40d01a3638a33ce2800
Conley Owens [Thu, 28 Apr 2011 22:07:00 +0000 (15:07 -0700)]
Merge "Added null check on return value for getKeyDispatcherState()"
Conley Owens [Thu, 28 Apr 2011 21:57:14 +0000 (14:57 -0700)]
Merge "Eliminating dead logic - bitmap state simply cannot be null here."
Conley Owens [Thu, 28 Apr 2011 21:57:11 +0000 (14:57 -0700)]
Merge "Removing unnecessary computation."
Conley Owens [Thu, 28 Apr 2011 21:57:08 +0000 (14:57 -0700)]
Merge "Using robust equality check."
Conley Owens [Thu, 28 Apr 2011 21:56:59 +0000 (14:56 -0700)]
Merge "We don't need these local references."
Conley Owens [Thu, 28 Apr 2011 21:56:55 +0000 (14:56 -0700)]
Merge "Eliminating dead code with no other effect than consuming battery life."
Conley Owens [Thu, 28 Apr 2011 18:57:48 +0000 (11:57 -0700)]
am
8fff7805: Merge "Remove duplicate \'=\' in exception message"
* commit '
8fff7805ab1ccde8321697116ed3dd2fae5340ed':
Remove duplicate '=' in exception message
Conley Owens [Thu, 28 Apr 2011 18:46:39 +0000 (11:46 -0700)]
Merge "Remove duplicate '=' in exception message"
Bernhard Reutner-Fischer [Thu, 28 Apr 2011 18:06:44 +0000 (20:06 +0200)]
Remove duplicate '=' in exception message
Remove duplicate '=' in IllegalArgumentException of MyAdapter::getItem()
Change-Id: I151a1432a9f54c8d42bae295f3967b0be2223026
Per Andersson [Mon, 17 Jan 2011 10:16:23 +0000 (11:16 +0100)]
Added null check on return value for getKeyDispatcherState()
The result from getKeyDispatcherState() was used without
checking if it returned null, which resulted in a NullPointerException.
Change-Id: I4b55ad44d5c08b7f729dbbdbcaed0e978a430258
Conley Owens [Wed, 27 Apr 2011 20:53:46 +0000 (13:53 -0700)]
am
13b21e60: Merge "Auto hide virtual keyboard"
* commit '
13b21e608e47382cb94a0aeac13b24ac97f355d2':
Auto hide virtual keyboard
Conley Owens [Wed, 27 Apr 2011 20:53:37 +0000 (13:53 -0700)]
am
a538b4a2: Merge "onDetachedFromWindow is called before onAttachedToWindow"
* commit '
a538b4a206c54ef763dab4da00990a8991425e77':
onDetachedFromWindow is called before onAttachedToWindow
Conley Owens [Wed, 27 Apr 2011 20:48:45 +0000 (13:48 -0700)]
am
4a4d96e7: Merge changes I0ab47e2f,Ib8016996
* commit '
4a4d96e7ee682167861434cba05531f04642bf97':
Remove errant tag terminator
Normalize output of XMLtree content
Conley Owens [Wed, 27 Apr 2011 20:20:16 +0000 (13:20 -0700)]
Merge "Auto hide virtual keyboard"
Conley Owens [Wed, 27 Apr 2011 20:14:57 +0000 (13:14 -0700)]
Merge "onDetachedFromWindow is called before onAttachedToWindow"
Conley Owens [Wed, 27 Apr 2011 20:06:46 +0000 (13:06 -0700)]
Merge changes I0ab47e2f,Ib8016996
* changes:
Remove errant tag terminator
Normalize output of XMLtree content
Conley Owens [Wed, 27 Apr 2011 19:43:19 +0000 (12:43 -0700)]
am
86d1d747: Merge "Add lock before calling initEglImage"
* commit '
86d1d74762e65b6f64c2d4758aa5fd4af6275696':
Add lock before calling initEglImage
Conley Owens [Wed, 27 Apr 2011 19:28:05 +0000 (12:28 -0700)]
Merge "Add lock before calling initEglImage"
Conley Owens [Wed, 27 Apr 2011 19:09:31 +0000 (12:09 -0700)]
Ignore fix for startup NPE as file is deleted.
Change-Id: Ifcedb4cc14b7079b03f585218c8c29b8f5d96d20
Conley Owens [Wed, 27 Apr 2011 18:40:46 +0000 (11:40 -0700)]
Merge "Fix for startup NPE."
Conley Owens [Tue, 26 Apr 2011 22:46:32 +0000 (15:46 -0700)]
am
17ba30b5: Merge "Load mnc/mcc configuration value from SIM (KDDI-CDMA)"
* commit '
17ba30b549a847f3aab854173a9484851f694453':
Load mnc/mcc configuration value from SIM (KDDI-CDMA)
Conley Owens [Tue, 26 Apr 2011 21:36:48 +0000 (14:36 -0700)]
Merge "Load mnc/mcc configuration value from SIM (KDDI-CDMA)"
Kenny Root [Tue, 26 Apr 2011 18:32:22 +0000 (11:32 -0700)]
am
13ce221e: Merge "libutils: Fix an improper const-cast in RefBase"
* commit '
13ce221e4316f7956ba072e774a7b97646e5d99b':
libutils: Fix an improper const-cast in RefBase
Kenny Root [Tue, 26 Apr 2011 18:26:33 +0000 (11:26 -0700)]
Merge "libutils: Fix an improper const-cast in RefBase"
Jean-Baptiste Queru [Tue, 26 Apr 2011 07:39:44 +0000 (00:39 -0700)]
am
4b72eec3: Merge "Re-assingning local variable if mStringBlocks was initialized here. NPE is thrown otherwise."
* commit '
4b72eec34c983abb7a3afcbadd74d3bd807c4765':
Re-assingning local variable if mStringBlocks was initialized here. NPE is thrown otherwise.
Jean-Baptiste Queru [Mon, 25 Apr 2011 20:33:37 +0000 (13:33 -0700)]
Merge "Re-assingning local variable if mStringBlocks was initialized here. NPE is thrown otherwise."
Josh Stone [Fri, 22 Apr 2011 18:13:35 +0000 (11:13 -0700)]
libutils: Fix an improper const-cast in RefBase
Under Fedora 15 Beta, gcc 4.6.0 warns:
frameworks/base/libs/utils/RefBase.cpp: In member function
‘void android::RefBase::weakref_type::trackMe(bool, bool)’:
frameworks/base/libs/utils/RefBase.cpp:483:67: error: passing
‘const android::RefBase::weakref_impl’ as ‘this’ argument of
‘void android::RefBase::weakref_impl::trackMe(bool, bool)’
discards qualifiers [-fpermissive]
trackMe is not a const function, so don't use const in the static_cast
to a weakref_impl pointer.
Change-Id: I3c9ba73eb127985f5f54197ffecf2939c50f632c
Jaikumar Ganesh [Tue, 19 Apr 2011 22:57:24 +0000 (15:57 -0700)]
am
c34c88b9: am
ec1a8f7a: Merge "Unable to push files to Windows 7"
* commit '
c34c88b948cb60c057bd2145df9f1b2091f3c894':
Unable to push files to Windows 7
Jaikumar Ganesh [Tue, 19 Apr 2011 22:28:26 +0000 (15:28 -0700)]
am
ec1a8f7a: Merge "Unable to push files to Windows 7"
* commit '
ec1a8f7aa7e71e292ef813b1e886a9e26e3635af':
Unable to push files to Windows 7
Jaikumar Ganesh [Tue, 19 Apr 2011 21:48:08 +0000 (14:48 -0700)]
Merge "Unable to push files to Windows 7"
Jozef BABJAK [Tue, 19 Apr 2011 11:24:37 +0000 (13:24 +0200)]
Eliminating dead logic - bitmap state simply cannot be null here.
Change-Id: I33fdf650b06242efa22ca30b3f7252f6854b42f6
Jozef BABJAK [Tue, 19 Apr 2011 10:28:22 +0000 (12:28 +0200)]
Re-assingning local variable if mStringBlocks was initialized here. NPE is thrown otherwise.
Change-Id: Ic8648aa5ae6a61348803171243c1c217feb058ec
Jozef BABJAK [Tue, 19 Apr 2011 09:27:30 +0000 (11:27 +0200)]
Fix for startup NPE.
Change-Id: I6928939d235faf544a85bb84bba2fa135ddc8e45
Andreas Huber [Mon, 18 Apr 2011 22:39:09 +0000 (15:39 -0700)]
am
f0f1ceeb: DO NOT MERGE: Make sure we restart the prefetcher immediately when trying to satisfy a read.
* commit '
f0f1ceeb517ff226532a407da6d55602f195e5b5':
DO NOT MERGE: Make sure we restart the prefetcher immediately when trying to satisfy a read.
Kobi Cohen Arazi [Fri, 15 Apr 2011 17:38:33 +0000 (10:38 -0700)]
Add lock before calling initEglImage
Without that lock, there is a chance of race condition
where while composing a specific index, requestBuf with
the same index can be executed and touch the
same data that is being used in initEglImage.
(e.g. dirty flag in texture)
Andreas Huber [Thu, 14 Apr 2011 17:01:41 +0000 (10:01 -0700)]
DO NOT MERGE: Make sure we restart the prefetcher immediately when trying to satisfy a read.
Change-Id: I27c2b5d20ae577e71936d32522f70f6ba08cc247
related-to-bug:
4286618
Mattias Petersson [Wed, 13 Apr 2011 13:09:16 +0000 (15:09 +0200)]
Auto hide virtual keyboard
Sometimes the virtual keyboard was not hidden when switching between
applications. An example of this was when launching the browser from
the Google Search widget:
1) Tap the Google Search widget and enter some text, e.g. "google"
2) Select one search items, e.g. "google maps"
3) Browser opens. Press back button.
4) Select an item again, e.g. "google maps" - Keyboard does not
close.
When switching application, the virtual keyboard needs to find a new
Z position (window index) among the other windows. Normally it is
placed on top of the first window that is visible and can get focus
(canBeImeTarget()).
With a new application being launched, there is
an exception: a special "starting window" is placed on top of the
Activity window while the application is starting up. Since this
window should not get input, we need to look below that window.
When doing this, the previous implementation assumed that the
first window below always was focusable. If it wasn't, the
input method was placed above the "starting window", which
caused confusion that led to the keyboard not being closed
automatically.
In the case of the Browser, it sometimes has a "fake TitleBar"
window that can not get focus and that is placed above the
Activity window.
With this fix, we now keep looking through the windows below
the "starting window" until we find a window that can receive
input.
Change-Id: I1117846eb0f57603e64329bd955e28182f98f226
Martin Wallgren [Wed, 13 Apr 2011 13:03:35 +0000 (15:03 +0200)]
onDetachedFromWindow is called before onAttachedToWindow
Multiple threads are adding messages about the current
state of the views to the main looper. This can cause
onDetachedFromWindow to be posted on the looper before
onAttachedToWindow. This change will make sure to only
dispatch onDetachedFromWindow if we have previously
dispatched onAttachToWindow.
Change-Id: Ibc7cbcafb098bc000d2ef5480d2110d3fff4d55a
Eric Fischer [Tue, 12 Apr 2011 20:02:15 +0000 (13:02 -0700)]
am
71a556f2: Import revised translations. DO NOT MERGE
* commit '
71a556f24e2b8d489bf81b22fab6fe5e99a9eae0':
Import revised translations. DO NOT MERGE
Eric Fischer [Tue, 12 Apr 2011 19:08:59 +0000 (12:08 -0700)]
Import revised translations. DO NOT MERGE
Change-Id: Idd866750d0c53ae0665ef2dd6b64230182d185ae
Masanori Ogino [Tue, 12 Apr 2011 11:42:49 +0000 (20:42 +0900)]
Load mnc/mcc configuration value from SIM (KDDI-CDMA)
Now support to load mnc/mcc value from IMSI on KDDI-CDMA.
Change-Id: I6f1d137ecf2e3130ca8f1a782dccec13b51f9933
Kenny Root [Mon, 11 Apr 2011 22:33:19 +0000 (15:33 -0700)]
am
08d9d9a4: Merge "Runtime resource overlay, iteration 1."
* commit '
08d9d9a46250c4fad66e9b637e8898a3524c4286':
Runtime resource overlay, iteration 1.
Kenny Root [Mon, 11 Apr 2011 17:07:15 +0000 (10:07 -0700)]
Merge "Runtime resource overlay, iteration 1."
Shachar Shemesh [Mon, 11 Apr 2011 10:48:35 +0000 (13:48 +0300)]
Remove errant tag terminator
Element android.widget.QuickContactBadge had two closing markers.
Change-Id: I0ab47e2fc999b4cf09fa10a7f5df58090816e315
Shachar Shemesh [Mon, 11 Apr 2011 10:32:19 +0000 (13:32 +0300)]
Normalize output of XMLtree content
Make the output from aapt dump xmltree normalized, so that it is unambigously
displayed regardless of the content of the strings. Previous patch left out
handling of XML element content.
Change-Id: Ib8016996c769f3dde7a87f7ecbdf850333f2426a
Brad Fitzpatrick [Thu, 7 Apr 2011 16:08:58 +0000 (09:08 -0700)]
am
ca7ad44a: Merge "audioflinger: Enable ARMv5TE optimized resampler"
* commit '
ca7ad44a8b6764b5935143bfe55700cbddb0d990':
audioflinger: Enable ARMv5TE optimized resampler
Brad Fitzpatrick [Thu, 7 Apr 2011 15:57:00 +0000 (08:57 -0700)]
Merge "audioflinger: Enable ARMv5TE optimized resampler"
Mattias Petersson [Thu, 7 Apr 2011 13:46:56 +0000 (15:46 +0200)]
Scolling using arrow keys with padding
It was impossible to reach the end of a large test when scrolling down
using the arrowkeys when the ScrollView had padding. A common example
of this would be an AlertDialog with a text that is too long to fit on
the screen.
Change-Id: I55464290a0cdeabde83ccccc76fe8d015ae57a8d
Jim Huang [Wed, 6 Apr 2011 06:19:29 +0000 (14:19 +0800)]
audioflinger: Enable ARMv5TE optimized resampler
Previously, the optimized asm option is only enabled when
__ARM_ARCH_5E__ is defined, which is assigned in armv5te.mk
rather than armv7-a series targets. This patch checks the ARM CPU
feature about half-word multiply instructions to enable ARMv5TE
resampler optimization routines properly.
Change-Id: I4c5a5d8c932416f23bedb0b389db958349f21ea4
Martin Wallgren [Thu, 7 Apr 2011 12:45:43 +0000 (14:45 +0200)]
CHAR_SEQUENCE_CREATOR cannot handle null string
TextUtils.writeToParcel can handle null as input CharSequence
but createFromParcel will throw NullPointerException. Transforming
to and from parcel should handle null in the same way.
Change-Id: I707ea9be2091d2655c5f63eaa57c5674befa5ad3
Xavier Ducrohet [Thu, 7 Apr 2011 02:08:11 +0000 (19:08 -0700)]
am
59aad78e: LayoutLib: don\'t render when shader\'s local matrix is set to 0 scale.
* commit '
59aad78eae14939c169876b2f632d005f382cf7d':
LayoutLib: don't render when shader's local matrix is set to 0 scale.
Xavier Ducrohet [Thu, 7 Apr 2011 02:08:02 +0000 (19:08 -0700)]
Xavier Ducrohet [Thu, 7 Apr 2011 00:03:31 +0000 (17:03 -0700)]
LayoutLib: don't render when shader's local matrix is set to 0 scale.
Change-Id: I7726d87f3dd0475ac662f535a08c6435b8b9ed1f
Xavier Ducrohet [Wed, 6 Apr 2011 03:19:17 +0000 (20:19 -0700)]
Merge
c5aeac7f from honeycomb.
LayoutLib: Fix issue with rendering mode.
When the layout content is embedded inside a decor layout
to emulate system bar (top or bottom) and title bar (or action bar)
then the code computing the full required size based on the RenderingMode
would fail because the decor layout would prevents the content layout to
take as much room as possible.
There is also an issue with the way we know render dialogs as the
dialogs usually as a frame with some padding and the previous measurements
would not take into account the dialog padding when increasing the
screen size.
This fix makes the code measure the size of the content layout in the
normal rendering, and then separately from the root layout with the proper
MeasureSpec to let the content grown. The difference in size is then added
to the screen size.
Bug: http://code.google.com/p/android/issues/detail?id=15892
Change-Id: Ie9a6c5e254b16785f817dcb9fae755d4936880aa
Ichitaro Kohara [Fri, 18 Feb 2011 12:15:04 +0000 (13:15 +0100)]
Fixed to draw texture when loadTexture() is called at first time
SW renderer of Stagefright reaches loadTexture() to draw image.
The first time loadTexture() is called, it just initializes OGL
texture, then returns. Thus, the first time call doesn't draw.
This patch fixes to move on to draw stage after the initialization.
Change-Id: I3ec1ad68fb8d376a4ad7aefded1c18a002d175c4
Mike Lockwood [Mon, 4 Apr 2011 21:43:06 +0000 (14:43 -0700)]
am
43707a83: UsbManager: minor Javadoc tweak
* commit '
43707a83ec12957fb84a406276a38fbaf4e23fd3':
UsbManager: minor Javadoc tweak
Jason parks [Mon, 4 Apr 2011 21:42:54 +0000 (14:42 -0700)]
am
78a9766b: Merge "Implement teardown script." into gingerbread
* commit '
78a9766be2a08e1f6a448bdb336f2afeff71a52b':
Implement teardown script.
Jeff Hamilton [Mon, 4 Apr 2011 21:42:44 +0000 (14:42 -0700)]
am
84d34078: Make the NFC UID forward compatible.
* commit '
84d340781b1e148ead9fb4790755737c9dcc1cde':
Make the NFC UID forward compatible.
James Dong [Mon, 4 Apr 2011 21:42:05 +0000 (14:42 -0700)]
am
ea5d2c58: Merge "When read() from AudioRecord returns 0 or negative value, report an error to application. DO NOT MERGE." into gingerbread
* commit '
ea5d2c584490424be6bbc27dec58b4f9868d631e':
When read() from AudioRecord returns 0 or negative value, report an error to application. DO NOT MERGE.
Eric Laurent [Mon, 4 Apr 2011 19:25:37 +0000 (12:25 -0700)]
am
a482d83c: Merge "Issue
4157048: mic gain for VoIP/SIP calls." into gingerbread
* commit '
a482d83ccf35ccd6cc29a9e1ace3d77b5f28d013':
Issue
4157048: mic gain for VoIP/SIP calls.
Mike Lockwood [Mon, 4 Apr 2011 19:25:33 +0000 (12:25 -0700)]
am
a8731e1e: Merge "DO NOT MERGE: UsbService: Make USB disconnect handling for accessory mode more robust" into gingerbread
* commit '
a8731e1e922512e5c22485e48059f33b61ab93ea':
DO NOT MERGE: UsbService: Make USB disconnect handling for accessory mode more robust
Trevor Johns [Mon, 4 Apr 2011 19:25:29 +0000 (12:25 -0700)]
am
cafbe87d: Merge "Fixing line in Market IAB docs that explains how to bind to the MarketService." into gingerbread
* commit '
cafbe87d5b6108d71e85950796b587f95d00cfac':
Fixing line in Market IAB docs that explains how to bind to the MarketService.
Simon Wilson [Mon, 4 Apr 2011 19:25:25 +0000 (12:25 -0700)]
am
ec944728: Merge "DO NOT MERGE Disable WiMAX when "Data Enabled" is off" into gingerbread
* commit '
ec94472835787b66e1799110411d5077d0ef37cd':
DO NOT MERGE Disable WiMAX when "Data Enabled" is off
Eric Fischer [Mon, 4 Apr 2011 19:20:38 +0000 (12:20 -0700)]
am
8b312c6c: Merge "Import revised translations. DO NOT MERGE" into gingerbread
* commit '
8b312c6ca1c3f6dc4bec0c7d45bc9509afef4984':
Import revised translations. DO NOT MERGE
Jason parks [Mon, 4 Apr 2011 19:20:33 +0000 (12:20 -0700)]
am
cca2f2f9: Merge "Send out a broadcast that the system is being factory reset." into gingerbread
* commit '
cca2f2f987ab772a4bae3176df7259f59d6502ce':
Send out a broadcast that the system is being factory reset.
Justin Ho [Mon, 4 Apr 2011 19:19:39 +0000 (12:19 -0700)]
am
87805ca3: Updated CDMA roaming icons
* commit '
87805ca3a477e165891baa18efe5239a70d11bc9':
Updated CDMA roaming icons
Jean-Baptiste Queru [Mon, 4 Apr 2011 19:19:26 +0000 (12:19 -0700)]
am
bef31431: No-op - reconcile with open-source gingerbread
* commit '
bef31431cafe100e382a3caf7ea7dce343832d68':
Robert Greenwalt [Mon, 4 Apr 2011 19:18:58 +0000 (12:18 -0700)]
am
27725e8c: Only leave one default network up at a time.
* commit '
27725e8cd1e2eb3b88186ae8d43e437f130b1f84':
Only leave one default network up at a time.
Robert Greenwalt [Mon, 4 Apr 2011 18:23:51 +0000 (11:23 -0700)]
Merge "resolved conflicts for merge of
7544b23e to gingerbread-plus-aosp" into gingerbread-plus-aosp
Robert Greenwalt [Mon, 4 Apr 2011 18:21:23 +0000 (11:21 -0700)]
resolved conflicts for merge of
7544b23e to gingerbread-plus-aosp
Change-Id: I28cf91f473eb9e3e758129b2f4d89827e73049dd
Brad Fitzpatrick [Mon, 4 Apr 2011 18:03:03 +0000 (11:03 -0700)]
am
9907d161: Merge "Fix data corruption when writing to Bluetooth socket"
* commit '
9907d161584415c81de1099678f160da172fd1a6':
Fix data corruption when writing to Bluetooth socket
Brad Fitzpatrick [Mon, 4 Apr 2011 17:49:05 +0000 (10:49 -0700)]
Merge "Fix data corruption when writing to Bluetooth socket"
Brad Fitzpatrick [Mon, 4 Apr 2011 17:17:43 +0000 (10:17 -0700)]
am
2ed72048: Merge "Always return a valid index from Rfc822Tokenizer.findTokenEnd()"
* commit '
2ed72048828cf9f617da971e5d2b8a062de08e5b':
Always return a valid index from Rfc822Tokenizer.findTokenEnd()
Brad Fitzpatrick [Mon, 4 Apr 2011 17:17:35 +0000 (10:17 -0700)]
am
e558a000: Merge "Avoid crash in system server when mounting container"
* commit '
e558a0000389a2523b8d3d3a787db6b4579d2bdc':
Avoid crash in system server when mounting container
Brad Fitzpatrick [Mon, 4 Apr 2011 16:41:10 +0000 (09:41 -0700)]
Merge "Always return a valid index from Rfc822Tokenizer.findTokenEnd()"
Brad Fitzpatrick [Mon, 4 Apr 2011 16:38:16 +0000 (09:38 -0700)]
Merge "Avoid crash in system server when mounting container"
Mattias Niklewski [Tue, 18 Jan 2011 13:27:23 +0000 (14:27 +0100)]
Always return a valid index from Rfc822Tokenizer.findTokenEnd()
If an invalid input string ends with a backslash inside a comment
or quoted string, the returned index would be past the end of the
string. In one case this would lead to a runtime exception being
thrown from MultiAutoCompleteTextView.performValidation.
Change-Id: If629372b429716c25cdc25764f088e95d4812d57
Fredrik Helmer [Wed, 9 Feb 2011 15:16:10 +0000 (16:16 +0100)]
Avoid crash in system server when mounting container
A race condition when mounting a container in PackageHelper may cause
the system_server to crash (uncaught exception). Calling methods are
prepared to handle null, so return null instead.
Change-Id: I852ee21a2d847e37d81c1b900c27ddf94ef24fcb
Mike Playle [Fri, 22 Oct 2010 12:58:17 +0000 (13:58 +0100)]
Fix data corruption when writing to Bluetooth socket
Writes to Bluetooth sockets are handled by writeNative() in
android_bluetooth_BluetoothSocket.cpp. This calls asocket_write()
which is implemented in abort_socket.c.
This latter function sleeps until poll() indicates that the socket is
writeable, then calls write() once, returning the number of bytes
written. However writeNative() just returns this byte count to the
caller; it's eventually ignored in BluetoothOutputStream.java. This
doesn't match the semantics of a Java OutputStream, which is required
to block until all bytes have been written.
This fix adds a loop to writeNative() that repeatedly calls the lower
level write function until all the data has been written (or an error
occurred in which case we should exit the loop early). With this
change it is possible to write large amounts of data to a Bluetooth
socket without experiencing data loss.
Change-Id: I0b464382817e15adec32ba0e3cb37e7d1cccc730
Mike Lockwood [Fri, 1 Apr 2011 21:28:45 +0000 (17:28 -0400)]
UsbManager: minor Javadoc tweak
Change-Id: I44efc54cd2a0991465aabe0ecea1b8a3291014c2
Signed-off-by: Mike Lockwood <lockwood@android.com>
Robert Greenwalt [Fri, 1 Apr 2011 21:11:30 +0000 (14:11 -0700)]
Fix bad merge.
Add host route for a default gateway before adding the default route.
Also fixing bug were "already active" condtion of APN failed to
propagate the default gateway info, resulting in a loss of that address.
bug:
4175328
Change-Id: I38f9d108693382a43e6605d56ca3963adf4eb6f4