OSDN Git Service
Ian Rogers [Wed, 14 Aug 2013 00:07:05 +0000 (17:07 -0700)]
If CheckJNI is enabled don't allow app workarounds
(cherry picked from commit
46963dff1d1592729ba1e8b274835bf6c80a50bd)
Change-Id: I2d054bf0824ec8378291d2035fc0586c6ec9aa2e
Mathieu Chartier [Wed, 7 Aug 2013 18:33:19 +0000 (11:33 -0700)]
Remove updateProcessState.
Change-Id: I46588f15f92c8298130c49773951c62b05b44a4f
(cherry picked from commit
6d6dfeaa7f9d22fcf323f35712ab82d541ca3030)
Jun Tian [Tue, 6 Aug 2013 07:03:15 +0000 (15:03 +0800)]
Fix the protection of code cache for x86 JIT
During generation of code into code cache
an unprotected region of memory does not correspond to
protected one, The patch fixes that.
Author: Katkov Serguei <serguei.i.katkov@intel.com>
(cherry picked from commit
74a62214ef262380371bc21be2a1c42295046fb2)
Change-Id: I362a10897564b987c8a3b2dfc9ded8f0a9efd56a
Ian Rogers [Thu, 8 Aug 2013 23:21:57 +0000 (16:21 -0700)]
am
13aeac96: Merge "Fix broken card table asserts."
* commit '
13aeac968f2473fcb4f31ccc6f43c6afc62db4bf':
Fix broken card table asserts.
Ian Rogers [Thu, 8 Aug 2013 21:15:59 +0000 (21:15 +0000)]
Merge "Fix broken card table asserts."
Mathieu Chartier [Thu, 8 Aug 2013 18:43:30 +0000 (11:43 -0700)]
Fix broken card table asserts.
We had an off by one error due to getting the card for the heap
source limit. This is not a valid card when the heap is at
maximum size, but doesn't need to be since we do less than
comparison on it. Fixed another error where we didn't take into
account the biased begin in an assert.
Did a bit of refactoring by removing useless if statement.
Fixes 061-out-of-memory, 080-oom-throw.
Bug: https://code.google.com/p/android/issues/detail?id=42868
(cherry picked from commit
ee903e174872edd0ecc6f1940c7412892cd49123)
Change-Id: If677c93e46fe1ee5729eec1820b1bd9682b4f924
Mathieu Chartier [Thu, 8 Aug 2013 18:43:30 +0000 (11:43 -0700)]
Fix broken card table asserts.
We had an off by one error due to getting the card for the heap
source limit. This is not a valid card when the heap is at
maximum size, but doesn't need to be since we do less than
comparison on it. Fixed another error where we didn't take into
account the biased begin in an assert.
Did a bit of refactoring by removing useless if statement.
Fixes 061-out-of-memory.
Bug:
8282011
Change-Id: I66e3026e7b98b738ecfbced101846ec9f184ab70
(cherry picked from commit
ee903e174872edd0ecc6f1940c7412892cd49123)
Elliott Hughes [Mon, 5 Aug 2013 22:51:57 +0000 (15:51 -0700)]
am
cec783df: Merge "More consistent JNI error reporting."
* commit '
cec783df5a51233dab94f2fa4506f677d45132e8':
More consistent JNI error reporting.
Elliott Hughes [Mon, 5 Aug 2013 22:15:28 +0000 (22:15 +0000)]
Merge "More consistent JNI error reporting."
Elliott Hughes [Mon, 5 Aug 2013 22:07:43 +0000 (15:07 -0700)]
More consistent JNI error reporting.
Always include the JNI function's name in the log message.
Bug:
10162113
Change-Id: I4ebd83c48f094f753190f70f644f5a59359b6327
Elliott Hughes [Sat, 3 Aug 2013 00:59:04 +0000 (17:59 -0700)]
am
d76fe5d0: Merge "Not all shared libraries have a JNI_OnLoad."
* commit '
d76fe5d00d0df2ff09dd311688112a496f782b9b':
Not all shared libraries have a JNI_OnLoad.
Elliott Hughes [Sat, 3 Aug 2013 00:59:03 +0000 (17:59 -0700)]
am
f20307a9: Merge "Returning JNI_ERR from JNI_OnLoad is still an error!"
* commit '
f20307a9e7ba4216790a36caa971e12183544ef8':
Returning JNI_ERR from JNI_OnLoad is still an error!
Elliott Hughes [Sat, 3 Aug 2013 00:59:02 +0000 (17:59 -0700)]
am
8c82bccc: Merge "Check JNI versions handed to JNI and JII functions."
* commit '
8c82bcccdf3f179a8e5d7fe2753e212a88e20dd6':
Check JNI versions handed to JNI and JII functions.
Elliott Hughes [Sat, 3 Aug 2013 00:42:00 +0000 (00:42 +0000)]
Merge "Not all shared libraries have a JNI_OnLoad."
Elliott Hughes [Sat, 3 Aug 2013 00:41:25 +0000 (17:41 -0700)]
Not all shared libraries have a JNI_OnLoad.
Change-Id: Iace5d33cbeafbf8b840bebb4bb709e32053f142d
Elliott Hughes [Fri, 2 Aug 2013 23:53:10 +0000 (23:53 +0000)]
Merge "Returning JNI_ERR from JNI_OnLoad is still an error!"
Elliott Hughes [Fri, 2 Aug 2013 23:49:57 +0000 (16:49 -0700)]
Returning JNI_ERR from JNI_OnLoad is still an error!
Fixes a bug introduced in the previous patch.
Change-Id: I34ea63d323a0125c3c4ae140a2ae9b8e9ed3e1df
Elliott Hughes [Fri, 2 Aug 2013 23:35:59 +0000 (23:35 +0000)]
Merge "Check JNI versions handed to JNI and JII functions."
Elliott Hughes [Fri, 2 Aug 2013 23:02:34 +0000 (16:02 -0700)]
Check JNI versions handed to JNI and JII functions.
Bug: https://code.google.com/p/android/issues/detail?id=58012
Change-Id: I7b062564573107df54f9e8d71df0e1583d83d421
Elliott Hughes [Fri, 2 Aug 2013 19:41:40 +0000 (12:41 -0700)]
am
813a3a2d: Merge "If dalvik wants ASCII casing, it needs to ask for it."
* commit '
813a3a2d1c295266299425cac35a817b06f026e3':
If dalvik wants ASCII casing, it needs to ask for it.
Brian Carlstrom [Fri, 2 Aug 2013 19:28:22 +0000 (12:28 -0700)]
resolved conflicts for merge of
ae77d6b6 to master
Change-Id: If7712cbddd6786c91648c4fc31f04e96937d4670
Elliott Hughes [Fri, 2 Aug 2013 18:51:50 +0000 (18:51 +0000)]
Merge "If dalvik wants ASCII casing, it needs to ask for it."
Brian Carlstrom [Fri, 2 Aug 2013 18:27:16 +0000 (18:27 +0000)]
Merge "Remove dependency on frameworks/native"
Elliott Hughes [Fri, 2 Aug 2013 17:34:29 +0000 (10:34 -0700)]
If dalvik wants ASCII casing, it needs to ask for it.
http://elliotth.blogspot.com/2012/01/beware-convenience-methods.html
Bug: https://code.google.com/p/android/issues/detail?id=58359
Change-Id: I1601cda215ebc5995117323e40a684ad352fe733
Elliott Hughes [Fri, 2 Aug 2013 15:52:03 +0000 (08:52 -0700)]
am
f65d478a: Merge "Throw NoSuchMethodError if RegisterNatives fails."
* commit '
f65d478a7005230264596f4814c94a0a9571259f':
Throw NoSuchMethodError if RegisterNatives fails.
Elliott Hughes [Fri, 2 Aug 2013 15:44:29 +0000 (15:44 +0000)]
Merge "Throw NoSuchMethodError if RegisterNatives fails."
Brian Carlstrom [Fri, 2 Aug 2013 06:30:07 +0000 (23:30 -0700)]
Remove dependency on frameworks/native
Change-Id: I66e226f8390bd499e956b00e4088bc0e1e150cb1
Elliott Hughes [Fri, 2 Aug 2013 00:29:14 +0000 (17:29 -0700)]
Throw NoSuchMethodError if RegisterNatives fails.
Bug:
9963858
Change-Id: I5b3ce3d3dbdc949e69f3389e2df19bcca5648d7e
Brian Carlstrom [Wed, 31 Jul 2013 18:43:36 +0000 (18:43 +0000)]
Merge "Show size/alloc/free per Dalvik heap space in dumpsys"
Elliott Hughes [Wed, 31 Jul 2013 18:04:21 +0000 (11:04 -0700)]
am
58de514b: Merge "Fix Class.getDex."
* commit '
58de514b160674a483bcb912d367072c7354cc1e':
Fix Class.getDex.
Elliott Hughes [Wed, 31 Jul 2013 17:47:22 +0000 (17:47 +0000)]
Merge "Fix Class.getDex."
Elliott Hughes [Wed, 31 Jul 2013 16:41:16 +0000 (09:41 -0700)]
Fix Class.getDex.
Dalvik doesn't check its own .so when looking for JNI methods, so
manually register Class.getDex. Also fix a bug in that native method's
implementation.
Change-Id: Ic87484fac888e9eea0e5d51a2d2447eee860ece5
Hiroshi Yamauchi [Mon, 15 Jul 2013 21:20:41 +0000 (14:20 -0700)]
Show size/alloc/free per Dalvik heap space in dumpsys
Add the heap size/alloc/free stats to the Dalvik heap space breakdown section in dumpsys meminfo.
Also, now the zygote heap has a distict ashmem region name.
Bug:
9532137
Bug:
8266259
(cherry picked from commit
3edfe0b32753309ad7dcccd894239cb0a8aefb85)
Change-Id: I9b32727e4211f571a5205b6b7281958514d245d4
Ben Cheng [Mon, 29 Jul 2013 20:40:51 +0000 (13:40 -0700)]
am
73a09433: Merge "Minor code cleanup to address warnings found by Clang."
* commit '
73a0943392a2927dae96195f5293940877f11c05':
Minor code cleanup to address warnings found by Clang.
Ben Cheng [Mon, 29 Jul 2013 20:29:27 +0000 (20:29 +0000)]
Merge "Minor code cleanup to address warnings found by Clang."
Ben Cheng [Fri, 26 Jul 2013 16:46:06 +0000 (09:46 -0700)]
Minor code cleanup to address warnings found by Clang.
cherry-picked from internal Android branch
55617c82a73d84ff3695bdd5526159448262d009
Change-Id: I0f78ca6b8293c13d7dbb535556543f6ea9f4dd45
Elliott Hughes [Mon, 29 Jul 2013 20:17:48 +0000 (13:17 -0700)]
am
4550b42e: Merge "Implement the dalvik side of libcore.reflect."
* commit '
4550b42e776ecfe26b0b5538d14a4d57342b5ce0':
Implement the dalvik side of libcore.reflect.
Elliott Hughes [Mon, 29 Jul 2013 20:06:13 +0000 (20:06 +0000)]
Merge "Implement the dalvik side of libcore.reflect."
Elliott Hughes [Mon, 29 Jul 2013 19:57:23 +0000 (12:57 -0700)]
am
fb1d9461: Merge "dx.bat: Don\'t change directory."
* commit '
fb1d9461a57c67a7fcf1f8667d9fa65b609d2f61':
dx.bat: Don't change directory.
Elliott Hughes [Mon, 29 Jul 2013 19:49:36 +0000 (19:49 +0000)]
Merge "dx.bat: Don't change directory."
Elliott Hughes [Fri, 26 Jul 2013 00:43:39 +0000 (17:43 -0700)]
Implement the dalvik side of libcore.reflect.
(Cherry-pick of
b4d5cf78d4731ffc6211d07a0d69c92c559ba792.)
Change-Id: I5d1b41f029c9c49a2d60a066a5fb0ed0f52f4b52
Ray Donnelly [Fri, 3 May 2013 00:41:36 +0000 (01:41 +0100)]
dx.bat: Don't change directory.
Changing directory to that of dx.bat prevents dx.bat from being
used with relative directories. This leads to all sorts of
issues integrating it with build systems that expect this to
work.
I've tested that this change works under the following conditions:
1. Android SDK in a path containing spaces.
2. java.exe in a path containing spaces.
3. Java object files in a path containing spaces.
4. Output jar file in a path containing spaces.
5. Java object files specified with relative paths.
6. Java object files specified with absolute paths.
7. Output jar file specified with relative paths.
8. Output jar file specified with absolute paths.
Change-Id: I1b47a6cc73169607cc6238acf197e3b467c9bcbc
Elliott Hughes [Fri, 26 Jul 2013 00:43:39 +0000 (17:43 -0700)]
Implement the dalvik side of libcore.reflect.
Change-Id: I9ef0da6f715e2727d8530aa7a8edee97b5bfa48d
Alex Ray [Wed, 24 Jul 2013 05:13:45 +0000 (05:13 +0000)]
Merge "libenc: remove libutils dependency"
Colin Cross [Wed, 24 Jul 2013 01:05:04 +0000 (01:05 +0000)]
Merge "add static dependency on liblog"
Alex Ray [Wed, 24 Jul 2013 00:48:05 +0000 (17:48 -0700)]
libenc: remove libutils dependency
Change-Id: I679fd6f06e007921251d15d7003615d7b0d91c52
Colin Cross [Wed, 24 Jul 2013 00:45:23 +0000 (17:45 -0700)]
add static dependency on liblog
Change-Id: Id431382f0fc71ef126dd35e5949d5de2abe2b785
Alex Ray [Wed, 24 Jul 2013 00:38:25 +0000 (00:38 +0000)]
Merge "libdvm: move thread definitions dependency"
Alex Ray [Wed, 24 Jul 2013 00:17:50 +0000 (17:17 -0700)]
libdvm: move thread definitions dependency
Change-Id: I88351f5b1a8bd2dccab8ab86fd400ec707f5809a
Elliott Hughes [Tue, 23 Jul 2013 23:43:14 +0000 (16:43 -0700)]
am
beeeb788: am
05fa5fd5: Merge "Simplify merges of the annotation code."
* commit '
beeeb78893dded91597a82b6dd9a7e4960f7549b':
Simplify merges of the annotation code.
Elliott Hughes [Tue, 23 Jul 2013 23:37:57 +0000 (16:37 -0700)]
am
05fa5fd5: Merge "Simplify merges of the annotation code."
* commit '
05fa5fd5ce0921e8e2a8da5114c5b557434a28db':
Simplify merges of the annotation code.
Elliott Hughes [Tue, 23 Jul 2013 23:35:35 +0000 (23:35 +0000)]
Merge "Simplify merges of the annotation code."
Elliott Hughes [Tue, 23 Jul 2013 23:17:22 +0000 (16:17 -0700)]
Simplify merges of the annotation code.
Change-Id: I9fd3b53032b2e6053094500571775ed2fb5edf78
buzbee [Sun, 21 Jul 2013 18:45:37 +0000 (11:45 -0700)]
Suppress warning if JIT disabled
Setting gDvmJit.code_cache_size=0 disables the JIT, but also
causes a warning. This change disables the JIT earlier in the
startup process, avoiding the warning.
Change-Id: Id6acd32dc0067218d503add90119008c953c4ea5
Mathieu Chartier [Thu, 18 Jul 2013 01:11:16 +0000 (18:11 -0700)]
Add updateProcessState to VMRuntime
updateProcessState is called from the ActivityManager. It is used in ART
to help determine GC behavior.
Change-Id: I65662708ffd8443fff0c3d0e320bf8922e22d7cf
Mathieu Chartier [Tue, 16 Jul 2013 22:46:42 +0000 (22:46 +0000)]
Merge "Backport native allocation tracking."
Elliott Hughes [Tue, 16 Jul 2013 22:36:11 +0000 (15:36 -0700)]
am
6986ad03: am
bc03380a: Merge "Remove an assertion that JDWP packets are small."
* commit '
6986ad0343ebe4a7379a6d726005fa550b186b5d':
Remove an assertion that JDWP packets are small.
Elliott Hughes [Tue, 16 Jul 2013 22:31:38 +0000 (15:31 -0700)]
am
bc03380a: Merge "Remove an assertion that JDWP packets are small."
* commit '
bc03380a9d5565cdef0c4b4c41fdd19813b05799':
Remove an assertion that JDWP packets are small.
Elliott Hughes [Tue, 16 Jul 2013 20:45:54 +0000 (20:45 +0000)]
Merge "Remove an assertion that JDWP packets are small."
Elliott Hughes [Tue, 16 Jul 2013 20:41:03 +0000 (13:41 -0700)]
Remove an assertion that JDWP packets are small.
Now we support longer allocation tracker runs, it's possible
to collect more than 1MiB of data.
Bug:
9871848
Change-Id: Ifb575538b97563ce909f7e0aa6840ccbd6e93569
Mathieu Chartier [Mon, 15 Jul 2013 23:58:59 +0000 (16:58 -0700)]
Backport native allocation tracking.
From:
https://googleplex-android-review.googlesource.com/#/c/328818/
Change-Id: I5a930e629c5a2062f01f870b4a44dcd98cb4ffe9
Mathieu Chartier [Wed, 10 Jul 2013 00:30:51 +0000 (17:30 -0700)]
Add low memory mode option.
Added a low memory mode option which can be enabled by passing the
flag "-XX:LowMemoryMode" to the VM.
Changed dvmHeapSourceAlloc to use mspace_malloc and madvise page
aligned regions of allocations back to the kernel if low
memory mode is enabled. No measurable performance penalty on a
benchmark which does 4 million object allocations. This seems to
reduce the Dalvik total PSS by 10 - 15% on devices with
large bitmaps (such as manta or flo).
Another change is that cards are madvised instead of memset when
cleared if low memory mode is enabled. This saves a minor amount
of memory at the cost of having slightly more page faults.
Change-Id: I0c850bc370ebabf4393bff1c47d3257b2f1518d2
Elliott Hughes [Fri, 12 Jul 2013 01:57:44 +0000 (18:57 -0700)]
am
d9369dc7: am
57914adc: am
93c3b11e: Merge "Increase the default alloc tracker maximum."
* commit '
d9369dc790004aedbd772737926329485f794548':
Increase the default alloc tracker maximum.
Elliott Hughes [Fri, 12 Jul 2013 01:55:50 +0000 (18:55 -0700)]
am
57914adc: am
93c3b11e: Merge "Increase the default alloc tracker maximum."
* commit '
57914adc7e0e09a74015c2dd0d3e98f1cc57e3ec':
Increase the default alloc tracker maximum.
Elliott Hughes [Fri, 12 Jul 2013 01:52:23 +0000 (18:52 -0700)]
am
93c3b11e: Merge "Increase the default alloc tracker maximum."
* commit '
93c3b11ea6bcc25e8f0bb6ba701c71ed05906b52':
Increase the default alloc tracker maximum.
Elliott Hughes [Fri, 12 Jul 2013 00:36:54 +0000 (00:36 +0000)]
Merge "Increase the default alloc tracker maximum."
Elliott Hughes [Fri, 12 Jul 2013 00:23:15 +0000 (17:23 -0700)]
Increase the default alloc tracker maximum.
Also add a new system property "dalvik.vm.allocTrackerMax" that allows
developers to choose an arbitrary limit.
Change-Id: I246a31d0ee46557544b1babae412246914495f80
Brian Carlstrom [Fri, 28 Jun 2013 21:38:35 +0000 (14:38 -0700)]
add dalvik.system.VMRuntime.vmLibrary implementation for Dalvik
(cherry picked from commit
70c799fc91564b497dedc408943065ed58a275bb)
Change-Id: I8392e9f9dd92420b5d5c61f83a32fae6e23ebee3
Elliott Hughes [Tue, 2 Jul 2013 20:14:29 +0000 (13:14 -0700)]
am
f8696fa7: am
2115c7ed: am
f8fe8731: Merge "Disable -Xjnigreflimit."
* commit '
f8696fa79b0bb1ecc9eea0229eb989f25feaa1c0':
Disable -Xjnigreflimit.
Elliott Hughes [Tue, 2 Jul 2013 20:13:17 +0000 (13:13 -0700)]
am
2115c7ed: am
f8fe8731: Merge "Disable -Xjnigreflimit."
* commit '
2115c7ed027f20edca0b8e0ab819c63ff16a389c':
Disable -Xjnigreflimit.
Elliott Hughes [Tue, 2 Jul 2013 20:11:33 +0000 (13:11 -0700)]
am
f8fe8731: Merge "Disable -Xjnigreflimit."
* commit '
f8fe8731a5302105b285d549e4a0265a35db3e2e':
Disable -Xjnigreflimit.
Elliott Hughes [Tue, 2 Jul 2013 20:09:00 +0000 (20:09 +0000)]
Merge "Disable -Xjnigreflimit."
Elliott Hughes [Tue, 2 Jul 2013 19:50:29 +0000 (12:50 -0700)]
Disable -Xjnigreflimit.
This was supposed to help find bugs, but in practice (unlike the
locale reference limit and the pinned array limit) doesn't. What it
does do is cause well-behaved code to fail if its allocation rate
is high. This has caused trouble for both audio and graphics code
running on eng builds.
Bug:
7903975
Change-Id: If475cb51c9cab13270a83a60d6d0aecfab758e88
Brian Carlstrom [Tue, 18 Jun 2013 20:12:05 +0000 (20:12 +0000)]
Merge "Move to new art dalvikvm"
Brian Carlstrom [Mon, 17 Jun 2013 19:52:41 +0000 (12:52 -0700)]
Move to new art dalvikvm
Change-Id: I93f66d294f6c02069dc8a539c54966ba41417ca2
Elliott Hughes [Mon, 17 Jun 2013 17:21:26 +0000 (10:21 -0700)]
am
bbd903de: am
28702c58: am
ae2fb913: Merge "JIT: Use rsb and shift in easy multiply."
* commit '
bbd903dec7c25859849fa447aaeef365f20440d4':
JIT: Use rsb and shift in easy multiply.
Elliott Hughes [Mon, 17 Jun 2013 17:18:54 +0000 (10:18 -0700)]
am
28702c58: am
ae2fb913: Merge "JIT: Use rsb and shift in easy multiply."
* commit '
28702c58c9af24f735091ea83243af012956099e':
JIT: Use rsb and shift in easy multiply.
Elliott Hughes [Mon, 17 Jun 2013 17:17:03 +0000 (10:17 -0700)]
am
ae2fb913: Merge "JIT: Use rsb and shift in easy multiply."
* commit '
ae2fb91351509897699a29ae54a3d120a90b62ec':
JIT: Use rsb and shift in easy multiply.
Elliott Hughes [Mon, 17 Jun 2013 17:14:21 +0000 (17:14 +0000)]
Merge "JIT: Use rsb and shift in easy multiply."
Elliott Hughes [Fri, 14 Jun 2013 17:47:25 +0000 (10:47 -0700)]
am
d189d8ca: am
6fb2ea56: am
2f0ce6ab: Merge "Fix inner class access flags and array modifiers."
* commit '
d189d8ca6ef2be53e80247c735ffb38279ae3f0f':
Fix inner class access flags and array modifiers.
Elliott Hughes [Fri, 14 Jun 2013 17:45:27 +0000 (10:45 -0700)]
am
6fb2ea56: am
2f0ce6ab: Merge "Fix inner class access flags and array modifiers."
* commit '
6fb2ea567970f155386c6c0301a94142a954cc27':
Fix inner class access flags and array modifiers.
Elliott Hughes [Fri, 14 Jun 2013 17:44:06 +0000 (10:44 -0700)]
am
2f0ce6ab: Merge "Fix inner class access flags and array modifiers."
* commit '
2f0ce6ab114b1036b06c4bd425cd01cb509260bd':
Fix inner class access flags and array modifiers.
Elliott Hughes [Fri, 14 Jun 2013 17:41:43 +0000 (17:41 +0000)]
Merge "Fix inner class access flags and array modifiers."
Elliott Hughes [Fri, 14 Jun 2013 01:57:33 +0000 (18:57 -0700)]
Fix inner class access flags and array modifiers.
Inner class access flags should come from the InnerClass attribute.
Array classes should all have the ACC_ABSTRACT modifier set.
Bug: https://code.google.com/p/android/issues/detail?id=56267
Change-Id: If4793ee6c7a6b24bf4d32ef05167d81b58011188
Jean-Baptiste Queru [Fri, 14 Jun 2013 16:11:23 +0000 (09:11 -0700)]
am
11c768b7: (-s ours) am
60d1c05b: am
eb82cca7: (-s ours) am
381868f3: (-s ours) am
840d233c: am
800e4320: am
63c34b5e: am
68719a17: (-s ours) Reconcile with jb-mr1-release - do not merge
* commit '
11c768b7ad45e20d9182e8fe2063b3643219b400':
Jean-Baptiste Queru [Fri, 14 Jun 2013 15:35:51 +0000 (08:35 -0700)]
am
60d1c05b: am
eb82cca7: (-s ours) am
381868f3: (-s ours) am
840d233c: am
800e4320: am
63c34b5e: am
68719a17: (-s ours) Reconcile with jb-mr1-release - do not merge
* commit '
60d1c05bc70c5a3acf2f345cf434524aecdeab43':
Jean-Baptiste Queru [Fri, 14 Jun 2013 15:34:02 +0000 (08:34 -0700)]
am
eb82cca7: (-s ours) am
381868f3: (-s ours) am
840d233c: am
800e4320: am
63c34b5e: am
68719a17: (-s ours) Reconcile with jb-mr1-release - do not merge
* commit '
eb82cca7f4b40bfb9aada2dfa56595d9e3ff1eda':
Anders O Nilsson [Wed, 10 Oct 2012 13:26:32 +0000 (15:26 +0200)]
JIT: Use rsb and shift in easy multiply.
For easy multiplication using reverse subtract (when
lit is 2^n-1) use the barrel shifter for rsb.
This improves arithmetic performance for code executing
in Dalvik. E.g String.hashCode.
Change-Id: Ifb086dcec344b30fd3e392ac21d508b43e820cdc
Signed-off-by: Patrik Ryd <patrik.ryd@stericsson.com>
Jean-Baptiste Queru [Fri, 14 Jun 2013 05:33:07 +0000 (22:33 -0700)]
am
381868f3: (-s ours) am
840d233c: am
800e4320: am
63c34b5e: am
68719a17: (-s ours) Reconcile with jb-mr1-release - do not merge
* commit '
381868f347d413a671a6ad9c0af64d258a060b99':
Jean-Baptiste Queru [Fri, 14 Jun 2013 05:31:29 +0000 (22:31 -0700)]
am
840d233c: am
800e4320: am
63c34b5e: am
68719a17: (-s ours) Reconcile with jb-mr1-release - do not merge
* commit '
840d233c12200de9260c85d6eadbc73b41cef00b':
Jean-Baptiste Queru [Fri, 14 Jun 2013 04:07:35 +0000 (21:07 -0700)]
am
800e4320: am
63c34b5e: am
68719a17: (-s ours) Reconcile with jb-mr1-release - do not merge
* commit '
800e4320ba0815bfe7831deab99f5d600800663f':
Jean-Baptiste Queru [Fri, 14 Jun 2013 01:07:45 +0000 (18:07 -0700)]
am
63c34b5e: am
68719a17: (-s ours) Reconcile with jb-mr1-release - do not merge
* commit '
63c34b5e2aae446b76fdf34f0ff895e35cd7376a':
Elliott Hughes [Thu, 13 Jun 2013 21:49:44 +0000 (14:49 -0700)]
am
15139109: am
6529809b: Merge "JIT: Allow use of cbz/cbnz for conditional branch gen."
* commit '
151391097e94585d4dd062098438e5c894b98d6d':
JIT: Allow use of cbz/cbnz for conditional branch gen.
Elliott Hughes [Thu, 13 Jun 2013 21:46:07 +0000 (14:46 -0700)]
am
6529809b: Merge "JIT: Allow use of cbz/cbnz for conditional branch gen."
* commit '
6529809bd4448a76622a08cf63829cab73d91b9a':
JIT: Allow use of cbz/cbnz for conditional branch gen.
Elliott Hughes [Thu, 13 Jun 2013 21:43:55 +0000 (21:43 +0000)]
Merge "JIT: Allow use of cbz/cbnz for conditional branch gen."
Elliott Hughes [Tue, 11 Jun 2013 18:07:24 +0000 (11:07 -0700)]
am
70605c75: am
a163f4a9: Merge "Now that abort(3) is fixed, dvmAbort can be simplified."
* commit '
70605c75b9418a3ef0450ba37bcf29a0c61609b4':
Now that abort(3) is fixed, dvmAbort can be simplified.
Elliott Hughes [Tue, 11 Jun 2013 18:03:21 +0000 (11:03 -0700)]
am
a163f4a9: Merge "Now that abort(3) is fixed, dvmAbort can be simplified."
* commit '
a163f4a944cc040aaf25f92ae9c28c1e9cf981f8':
Now that abort(3) is fixed, dvmAbort can be simplified.
Elliott Hughes [Tue, 11 Jun 2013 18:01:49 +0000 (18:01 +0000)]
Merge "Now that abort(3) is fixed, dvmAbort can be simplified."
Elliott Hughes [Tue, 11 Jun 2013 17:52:34 +0000 (10:52 -0700)]
Now that abort(3) is fixed, dvmAbort can be simplified.
The fix to abort(3) has SHA
7e6ce1a3c52d8533fed92c143419fedb0c93988a.
(https://android-review.googlesource.com/#/c/60341/)
Change-Id: I89e7ed174cbd08ad9046a98e2fdd73af8ae3ee5c
Brian Carlstrom [Sat, 8 Jun 2013 00:36:51 +0000 (17:36 -0700)]
am
520f6ab1: am
6531a977: Merge "Lose System.currentTimeMillis, nanoTime, and mapLibraryName to the library."
* commit '
520f6ab13b56bdaeb21ac99d105ac59a052539a7':
Lose System.currentTimeMillis, nanoTime, and mapLibraryName to the library.