OSDN Git Service
Jesse Wilson [Thu, 11 Mar 2010 23:03:28 +0000 (15:03 -0800)]
Merge "Implementing getBaseUri() for DOM."
Jesse Wilson [Thu, 11 Mar 2010 02:42:32 +0000 (18:42 -0800)]
Implementing getBaseUri() for DOM.
The tests for this uncover problems with the RI's implementation.
It does bizarre things with hashes, trashing the absolute URI and
returning a page-relative one (like "#foo").
Change-Id: Ib8af163a7b359e3f72a9c94eb3dd7e81e3a9a95c
Bill Buzbee [Thu, 11 Mar 2010 19:12:15 +0000 (11:12 -0800)]
Jit: Make debugging mode aware of inlineExecute/moveResult optimization
The Jit has a mode in which selected opcodes can be handled normally
or single-stepped in the interpter. This was broken for cases in
which the Jit applied an optimization to fold inlineExecute/moveResult
intruction pairs into a single operation and the debug mode was set
to handle the two opcodes differently.
Change-Id: Ifa436d4ba66ba0c13ea366c0956e6cf92ce9cdfd
Colin Cross [Thu, 11 Mar 2010 02:14:13 +0000 (18:14 -0800)]
Merge "Add armv7-a-neon build target"
Xavier Ducrohet [Wed, 10 Mar 2010 23:33:45 +0000 (15:33 -0800)]
Merge "remove System.*.println when ddm dispatch fails."
Xavier Ducrohet [Wed, 10 Mar 2010 22:11:40 +0000 (14:11 -0800)]
remove System.*.println when ddm dispatch fails.
Those messages are sometimes displayed to the user executing
the am command and can be really confusing (they are totally
benign)
Change-Id: I09c194a02ddc956ec52f19faa03ca9aa1eba604b
Andy McFadden [Wed, 10 Mar 2010 22:05:54 +0000 (14:05 -0800)]
Merge "Rearrange the way SignalCatcher writes logs."
Andy McFadden [Wed, 10 Mar 2010 21:42:22 +0000 (13:42 -0800)]
Rearrange the way SignalCatcher writes logs.
This is an attempt to reduce the amount of time that threads are suspended
while dumping stack traces in response to a SIGQUIT. This is primarily
for ANRs and bugreport generation, where many different apps may be trying
to write their stacks out concurrently.
Instead of grabbing the file with flock() and then printing the traces
to the file, we render the traces to memory and write them with a single
write() call to an O_APPEND fd. Also, the file access now occurs after
the VM threads have been resumed to avoid stalling on I/O.
No change was made to the format of the output.
For bug
2504471.
Also: did some rearranging in the file.
Change-Id: If505ec014237289a6cc86aa09bc66b3a126b345c
Ben Cheng [Wed, 10 Mar 2010 21:12:55 +0000 (13:12 -0800)]
Pad the page containing ARM code in the JIT code cache to work around a CPU bug.
Bug:
2501147
Verified with seeing the following line in the log. The 4096 number confirms
that the ARM handler code is using a full page.
D/dalvikvm( 369): 21 compilations using 4096 + 1288 bytes
Change-Id: I89e1473ce9a4353f58f740e1c86d358d3fec33b2
Jesse Wilson [Wed, 10 Mar 2010 18:11:59 +0000 (10:11 -0800)]
Merge "New tests for JSONObject and for JSON's self-use."
Jesse Wilson [Wed, 10 Mar 2010 01:54:35 +0000 (17:54 -0800)]
New tests for JSONObject and for JSON's self-use.
Alongside development of these tests, I'm working on a new
cleanroom implementation. The self use test was written to
prevent me from self-using in a way that the original
implementation does not.
Change-Id: Ie617aca1978bd39d85b05e5c2c7bd657ed159dd6
Andy McFadden [Wed, 10 Mar 2010 15:25:33 +0000 (07:25 -0800)]
Merge "If a finalizer wedges, raise prio and retry."
Elliott Hughes [Wed, 10 Mar 2010 06:59:36 +0000 (22:59 -0800)]
Merge "Work around droiddoc bug http://b/
2022288."
Elliott Hughes [Wed, 10 Mar 2010 06:58:34 +0000 (22:58 -0800)]
Work around droiddoc bug http://b/
2022288.
Change-Id: Ib46a260916dee99f190aa8b9465f4f2d3b04aa67
Ben Cheng [Wed, 10 Mar 2010 06:38:34 +0000 (22:38 -0800)]
Merge "Align fake data in the same page offsets as those in the bugreport."
Ben Cheng [Wed, 10 Mar 2010 06:32:03 +0000 (22:32 -0800)]
Align fake data in the same page offsets as those in the bugreport.
Sometimes the crash in the JIT'ed code is due to CPU bugs which are sensitive
to placement of the code.
Change-Id: I017ec3620f8172e2fac9e7abfa07f76b65db2306
Elliott Hughes [Wed, 10 Mar 2010 06:19:34 +0000 (22:19 -0800)]
Another go at making droiddoc as happy as javadoc...
Change-Id: I70dd8d57053a6e60b9be0cbe8a95d9937d9b00ef
Elliott Hughes [Wed, 10 Mar 2010 05:49:16 +0000 (21:49 -0800)]
Fix javadoc errors.
Change-Id: Ib3eb500006f5b4b1dadf959397fce7737fb53fe7
Jesse Wilson [Wed, 10 Mar 2010 02:10:19 +0000 (18:10 -0800)]
Merge "Print errors encountered while writing XML reports"
Jesse Wilson [Wed, 10 Mar 2010 02:08:04 +0000 (18:08 -0800)]
Merge "Implement adoptNode() and importNode()."
Jesse Wilson [Wed, 10 Mar 2010 00:33:18 +0000 (16:33 -0800)]
Implement adoptNode() and importNode().
importNode() shares all of its implementation with Node.clone();
the only areas they differ are the ones we don't support. The
shared code is in 2 new methods, shallowCopy() and cloneOrImportNode(),
both based on the old clone() method.
Also removing some unnecessary "throws DOMException" clauses.
The tests for the new methods are slightly cumbersome. Most move/copy
nodes to another document, serialize that, and compare.
Change-Id: Id9fb076e020d8327a8f70da401af9bd95d7a3d1b
Jesse Wilson [Wed, 10 Mar 2010 01:59:43 +0000 (17:59 -0800)]
Print errors encountered while writing XML reports
Change-Id: I7fdf9f6989c94a3f0fd3ff93db3426a4d080120f
Andy McFadden [Wed, 10 Mar 2010 00:38:36 +0000 (16:38 -0800)]
If a finalizer wedges, raise prio and retry.
The VM uses a watchdog mechanism to detect stuck finalizers. It appears
that, in some cases, the watchdog may be firing because the HeapWorker
thread in a background process doesn't get any CPU time after a
remotely-induced GC finishes. With this change, if the HeapWorker is
running at a reduced priority, we raise the priority and allow it to
try some more.
No attempt is made to put the thread priority back. (The HeapWorker
thread doesn't do anything in an idle process, and what it does do is
geared toward freeing one kind of resource or another.)
For bug
2492196.
Change-Id: Ic734c2e2819b9d60d20b2961f2a75085d5879495
Elliott Hughes [Tue, 9 Mar 2010 22:49:47 +0000 (14:49 -0800)]
Merge "Minor documentation improvements."
Elliott Hughes [Tue, 9 Mar 2010 00:54:42 +0000 (16:54 -0800)]
Minor documentation improvements.
Based on user-submitted bugs that were just misunderstandings, plus
implementation bugs caused by the intended behavior being somewhat
subtle.
Change-Id: Ic2606b5e57dadc95a35c2d0a977c01434a2fa28a
Brian Carlstrom [Tue, 9 Mar 2010 22:23:08 +0000 (14:23 -0800)]
Merge "Fix certimport.sh to check for Bouncy Castle provider installation (and add 1.6 JDK to PATH)"
Brian Carlstrom [Tue, 9 Mar 2010 19:30:13 +0000 (11:30 -0800)]
Fix certimport.sh to check for Bouncy Castle provider installation (and add 1.6 JDK to PATH)
Now if certimport.sh is run on a machine without the
BouncyCastleProvider installed, it will suggest how to apt-get install
the proper package.
At enh's suggestion, I tried running with out own local Bouncy Castle
classes to see what would happen, but the code ended up depending on
our NativeCrypto JNI code and there that isn't proper JNI code to be
loading into a RI JDK.
Also at enh's suggestion, we now prepend a JDK 1.6 bin directory to
the path for correctly correct default behavior. I do make sure it
exists and warn if it does not.
Change-Id: Ic936a6cc69fa3795e917c052ed79d19b2e66b5a1
Colin Cross [Tue, 9 Mar 2010 21:23:40 +0000 (13:23 -0800)]
Add armv7-a-neon build target
Change-Id: I981d55b53f6b3c185fe93384924bdbe18057132c
Carl Shapiro [Tue, 9 Mar 2010 20:25:51 +0000 (12:25 -0800)]
Merge "Hoist shape discrimination above thin lock owner test in the lock procedure. It is unsafe to reckon a thin lock owner without having first determined that the bit pattern of the lock word corresponds to that of a thin lock. Without proper ordering, a monitor lock can and will be created which, excluding the shape bit, corresponds to a lock owned by the calling thread."
Jean-Baptiste Queru [Tue, 9 Mar 2010 02:04:06 +0000 (18:04 -0800)]
Add an empty CleanSpec.mk
Change-Id: I43d2404e71aa2bc0d3d6aada35d613bb484129e7
Jesse Wilson [Tue, 9 Mar 2010 00:55:58 +0000 (16:55 -0800)]
Merge "Adding support for getUserData() and setUserData() to DOM nodes."
Jesse Wilson [Tue, 9 Mar 2010 00:19:39 +0000 (16:19 -0800)]
Adding support for getUserData() and setUserData() to DOM nodes.
Also making sure both Document and DocumentType get assigned
a Document value when possible; this is necessary to store the
user data objects.
Carl Shapiro [Mon, 8 Mar 2010 22:38:42 +0000 (14:38 -0800)]
Hoist shape discrimination above thin lock owner test in the lock
procedure. It is unsafe to reckon a thin lock owner without having
first determined that the bit pattern of the lock word corresponds to
that of a thin lock. Without proper ordering, a monitor lock can and
will be created which, excluding the shape bit, corresponds to a lock
owned by the calling thread.
In addition, move compiler barriers so they immediately preceed base
address publication. Also, kill cargo-cult volatiles that confounded
my bug hunt.
Bill Buzbee [Mon, 8 Mar 2010 19:30:19 +0000 (11:30 -0800)]
Jit: fix for
2483131 - VM daemon thread shutdown with JIT enabled
As part of shutdown, the Jit needs to unchain all translations so that
they can respond to suspend requests. This change introduces a
dvmJitUnchainAll() into shutdown, and also removes some previous
housecleaning that freed our key tables. We can't actually free those
because suspended threads in the process of shutting down may hold
references.
Change-Id: I7aad332e7195a94970c25a25b2d9fda5607bec08
Bill Buzbee [Sun, 7 Mar 2010 07:30:57 +0000 (23:30 -0800)]
Jit: Make most Jit compile failures non-fatal; just abort offending translation
Issue
2175597 Jit compile failures should abort translation, but not the VM
Added new dvmCompileAbort() to replace uses of dvmAbort() when something goes
wrong during the compliation of a trace. In that case, we'll abort the translation
and set it's head to the interpret-only "translation".
Carl Shapiro [Sat, 6 Mar 2010 01:34:05 +0000 (17:34 -0800)]
Merge "Jit: Fix for issue
2487769, Simplify in-line thin lock release"
Ben Cheng [Fri, 5 Mar 2010 23:37:30 +0000 (15:37 -0800)]
Merge "Collect more JIT stats in the assert build."
Ben Cheng [Fri, 5 Mar 2010 23:27:21 +0000 (15:27 -0800)]
Collect more JIT stats in the assert build.
New stuff includes breakdown of callsite types (ie monomorphic vs polymorphic
vs monoporphic resolved to native), total time spent in JIT'ing, and average
JIT time per compilation.
Example output:
D/dalvikvm( 840): 4042 compilations using 1976 + 329108 bytes
D/dalvikvm( 840): Compiler arena uses 10 blocks (8100 bytes each)
D/dalvikvm( 840): Compiler work queue length is 0/36
D/dalvikvm( 840): size if 8192, entries used is 4137
D/dalvikvm( 840): JIT: 4137 traces, 8192 slots, 1099 chains, 40 thresh, Non-blocking
D/dalvikvm( 840): JIT: Lookups:
1128780 hits, 168564 misses; 179520 normal, 6 punt
D/dalvikvm( 840): JIT: noChainExit: 528464 IC miss, 194708 interp callsite, 0 switch overflow
D/dalvikvm( 840): JIT: Invoke: 507 mono, 988 poly, 72 native, 1038 return
D/dalvikvm( 840): JIT: Total compilation time: 2342 ms
D/dalvikvm( 840): JIT: Avg unit compilation time: 579 us
D/dalvikvm( 840): JIT: 3357 Translation chains, 97 interp stubs
D/dalvikvm( 840): dalvik.vm.jit.op = 0-2,4-5,7-8,a-c,e-16,19-1a,1c-23,26,28-29,2b-2f,31-3d,44-4b,4d-51,60,62-63,68-69,70-72,76-78,7b,81-82,84,87,89,8d-93,95-98,a1,a3,a6,a8-a9,b0-b3,b5-b6,bb-bf,c6-c8,d0,d2-d6,d8,da-e2,ee-f0,f2-fb,
D/dalvikvm( 840): Code size stats: 50666/105126 (compiled/total Dalvik), 329108 (native)
Costin Manolache [Fri, 5 Mar 2010 22:56:05 +0000 (14:56 -0800)]
Merge "Fix server side SSLEngine ServerKeyExchange signature."
Bill Buzbee [Fri, 5 Mar 2010 18:56:37 +0000 (10:56 -0800)]
Jit: Fix for issue
2487769, Simplify in-line thin lock release
The Jit was using ldrex/strex to clear an owned thin lock in the
fast path. This was not necessary - an integer store works and is
much faster.
Jesse Wilson [Fri, 5 Mar 2010 18:21:16 +0000 (10:21 -0800)]
Merge "Implementing the Java 7 APIs for DeflaterOutputStreams."
Jesse Wilson [Fri, 5 Mar 2010 01:18:44 +0000 (17:18 -0800)]
Implementing the Java 7 APIs for DeflaterOutputStreams.
See bug
1729487.
Costin Manolache [Tue, 2 Mar 2010 21:47:01 +0000 (13:47 -0800)]
Fix server side SSLEngine ServerKeyExchange signature.
Code using SSLEngine for non-blocking SSL can't talk with openssl as a client,
since the signature is computed on different content (and openssl checks it,
unlike java). The fix is to use strip the 0x00 prefix when signing - like
it is done when generating the message, refactored both to use a common
method. We also include the length in the signature, it was also missing.
Elliott Hughes [Thu, 4 Mar 2010 05:35:41 +0000 (21:35 -0800)]
Prevent java.text.Normalizer from getting a default constructor.
The dalvik continuous build's new jdiff run caught this.
Elliott Hughes [Thu, 4 Mar 2010 02:14:40 +0000 (18:14 -0800)]
Merge "Don't call a method that can be overridden from File's constructors."
Andy McFadden [Thu, 4 Mar 2010 01:53:00 +0000 (17:53 -0800)]
Merge "Update local copy of getSchedulerGroup."
Elliott Hughes [Thu, 4 Mar 2010 01:39:14 +0000 (17:39 -0800)]
Don't call a method that can be overridden from File's constructors.
Bug:
2486943
Andy McFadden [Wed, 3 Mar 2010 23:37:10 +0000 (15:37 -0800)]
Update local copy of getSchedulerGroup.
The getSchedulerGroup() in libcutils was updated, but for some reason we
have our own copy in the VM. This change brings it up to date so we
stop crashing every time something requests a virtual stack (e.g. with a
"kill -3 <pid>").
(The function is now an exact clone -- the previous version was slightly
modified.)
Bill Buzbee [Wed, 3 Mar 2010 23:30:35 +0000 (15:30 -0800)]
Merge "Jit: Sapphire tuning - mostly scheduling."
Bill Buzbee [Wed, 3 Mar 2010 00:14:41 +0000 (16:14 -0800)]
Jit: Sapphire tuning - mostly scheduling.
Re-enabled load/store motion that had inadvertently been turned off for
non-armv7 targets. Tagged memory references with the kind of memory
they touch (Dalvik frame, literal pool, heap) to enable more aggressive
load hoisting. Eliminated some largely duplicate code in the target
specific files. Reworked temp register allocation code to allocate next
temp round-robin (to improve scheduling opportunities).
Overall, nice gain for Sapphire. Shows 5% to 15% on some benchmarks, and
measurable improvements for Passion.
Elliott Hughes [Wed, 3 Mar 2010 21:23:46 +0000 (13:23 -0800)]
am
b16b78db: am
275acdff: Copy Froyo\'s version of MulticastSocketTest to Eclair for CTS.
Merge commit '
b16b78db118036d7df2cf5705baa18f34bd42eec'
* commit '
b16b78db118036d7df2cf5705baa18f34bd42eec':
Copy Froyo's version of MulticastSocketTest to Eclair for CTS.
Elliott Hughes [Wed, 3 Mar 2010 21:12:27 +0000 (13:12 -0800)]
am
275acdff: Copy Froyo\'s version of MulticastSocketTest to Eclair for CTS.
Merge commit '
275acdff64a58f407cec552e7027cb07414b48ef' into eclair-plus-aosp
* commit '
275acdff64a58f407cec552e7027cb07414b48ef':
Copy Froyo's version of MulticastSocketTest to Eclair for CTS.
Elliott Hughes [Wed, 3 Mar 2010 07:58:39 +0000 (23:58 -0800)]
Copy Froyo's version of MulticastSocketTest to Eclair for CTS.
Partners are seeing test failures from the old tests, which were very bogus.
Andy McFadden [Tue, 2 Mar 2010 23:21:18 +0000 (15:21 -0800)]
Merge "Switch thread's cgroup during spin-on-suspend."
Carl Shapiro [Tue, 2 Mar 2010 23:12:53 +0000 (15:12 -0800)]
Merge "Print the self thread id and the owning thread id in the description of the IllegalMonitorExcpetion throw by unlockMonitor."
Elliott Hughes [Tue, 2 Mar 2010 23:04:09 +0000 (15:04 -0800)]
Use assertEquals rather than assertTrue/assertFalse.
Andy McFadden [Tue, 2 Mar 2010 22:23:04 +0000 (14:23 -0800)]
Switch thread's cgroup during spin-on-suspend.
Dalvik uses a safe-point suspend mechanism, in which threads are asked
to suspend and do so at their earliest convenience. The thread doing
the asking will sleep-spin until all threads have complied. If a thread
keeps on running, we get a "spin on suspend" situation.
The VM tries to correct matters by raising the priority of the spinning
thread if it's below normal. Until now it was just fiddling with the
"nice" value, but it appears that isn't enough (e.g. bug
2467653). We
now also change the thread's process group from "background" to
"foreground" in an attempt to get it some CPU.
Also includes some changes as part of the Log Reduction Act:
- Don't log the thread stacks on the first spin-on-suspend iteration.
The act of logging can itself be quite a drag.
- Removed the redundant and unnecessary "dumping state" line, as well
as some dead glibc-only code.
- While waiting for daemon threads to shut down, only show the "not
ready yet" message on the first iteration.
Elliott Hughes [Tue, 2 Mar 2010 21:24:31 +0000 (13:24 -0800)]
Add our home-grown Normalizer tests.
Originally https://android-git.corp.google.com/g/42517.
Elliott Hughes [Tue, 2 Mar 2010 04:07:07 +0000 (20:07 -0800)]
Merge "Implement (but @hide) java.text.Normalizer from Java 6."
Jesse Wilson [Tue, 2 Mar 2010 03:15:57 +0000 (19:15 -0800)]
Merge "Implementing almost all of Document.normalizeDocument()."
Jesse Wilson [Tue, 2 Mar 2010 01:31:53 +0000 (17:31 -0800)]
Implementing almost all of Document.normalizeDocument().
This follows the rules specified by DOMConfiguration. In particular:
- replacing CDATA nodes with text
- splitting CDATA nodes
- merging text nodes
- stripping comments
- detecting invalid characters
I haven't added the normalization code for XML validation or namespaces.
Elliott Hughes [Tue, 2 Mar 2010 00:28:05 +0000 (16:28 -0800)]
Implement (but @hide) java.text.Normalizer from Java 6.
Based on https://android-git.corp.google.com/g/42516.
Includes the harmony tests from their Java 6 branch.
Bug: 719001
David 'Digit' Turner [Mon, 1 Mar 2010 23:45:58 +0000 (15:45 -0800)]
Merge "Allow unpriviledged gdbserver to attach to debuggable processes."
David 'Digit' Turner [Tue, 23 Feb 2010 00:56:58 +0000 (16:56 -0800)]
Allow unpriviledged gdbserver to attach to debuggable processes.
This patch is used to allow an unprivileged gdbserver binary to
perform a ptrace attach to a debuggable process.
The kernel will only allow the attach to succeed under the following
conditions:
- gdbserver and the process are run under the exact same uids/gids
- the process is dumpable
The second condition requires this patch. The first one is ensured
by different changes to the platform (that don't directly touch Dalvik).
This is part of the work performed to allow native debugging of
applications on production devices.
Andy McFadden [Mon, 1 Mar 2010 21:39:38 +0000 (13:39 -0800)]
Merge "Added a simple check for stray tracked refs."
Andy McFadden [Mon, 1 Mar 2010 20:40:06 +0000 (12:40 -0800)]
Merge "Added an x86-atom "TODO" list."
Andy McFadden [Mon, 1 Mar 2010 20:16:12 +0000 (12:16 -0800)]
Added an x86-atom "TODO" list.
Besides the obvious things (unimplemented opcodes), this points out a
couple of changes that were made to the interpreter and require a
corresponding change for x86-atom.
Andy McFadden [Mon, 1 Mar 2010 18:52:00 +0000 (10:52 -0800)]
Added a simple check for stray tracked refs.
The idea is to make sure the main thread is getting through the VM init
process with an empty tracked references table. If we find any, we
print a warning and dump the table, but continue on anyway.
My test was a trivial dvmMalloc(), which turned up a flaw in the ref
table dump code: it didn't deal with "raw" dvmMalloc sections. So
that's fixed as well.
Jesse Wilson [Mon, 1 Mar 2010 18:21:05 +0000 (10:21 -0800)]
Merge "New implementation for DOMConfiguration."
Jesse Wilson [Sat, 27 Feb 2010 08:00:54 +0000 (00:00 -0800)]
New implementation for DOMConfiguration.
This API is disgusting. Its not regular, not typesafe, sparsely
implemented and overly specific. I'm implementing the minimum I can
get away with - exactly the same route taken by the RI.
The Parameter stuff feels a little bit like overkill, but it allowed
me to group related chunks of code together and avoid long chains of
equalsIgnoreCase if statements.
This is necessary to implement Document.normalize(), one of the
last remaining APIs in DOMv3.
Also fixing our implementation of Node.normalize() and adding tests
to discover a bug in that code. Previously a document like this:
"<foo>abc<br>def</foo>"
Would be normalized to this:
"<foo>abcdef<br></foo>"
Which is a horrible bug! Yuck.
Implementation and tests for Document.normalize() are forthcoming.
Elliott Hughes [Mon, 1 Mar 2010 18:19:46 +0000 (10:19 -0800)]
Merge "Fix a FormatFlagsConversionMismatchException bug I introduced in Froyo."
Elliott Hughes [Sun, 28 Feb 2010 19:10:10 +0000 (11:10 -0800)]
Fix a FormatFlagsConversionMismatchException bug I introduced in Froyo.
We want to show the characters representing the flags, not the decimal
values of their ASCII representation. "Use overloading judiciously",
as the man says, or cause errors like this.
Seen in FRE56B (see https://android-git.corp.google.com/g/42528).
Andy McFadden [Fri, 26 Feb 2010 22:25:32 +0000 (14:25 -0800)]
Merge "Don't set x86 arch variant."
Andy McFadden [Fri, 26 Feb 2010 21:34:01 +0000 (13:34 -0800)]
Don't set x86 arch variant.
This was a workaround for a missing bit in the core build. No longer
needed.
Jesse Wilson [Fri, 26 Feb 2010 21:16:39 +0000 (13:16 -0800)]
Merge "Adding tests to parse document attributes from the DOM."
Jesse Wilson [Fri, 26 Feb 2010 20:17:32 +0000 (12:17 -0800)]
Adding tests to parse document attributes from the DOM.
This adds implementation and tests for these methods:
getInputEncoding(), getDocumentURI().
And tests for these methods:
getXmlEncoding(), getXmlVersion(), getXmlStandalone().
Andy McFadden [Fri, 26 Feb 2010 18:16:38 +0000 (10:16 -0800)]
Merge "Update a few things in x86-atom."
Andy McFadden [Fri, 26 Feb 2010 15:35:20 +0000 (07:35 -0800)]
Update a few things in x86-atom.
This replaced unuses opcodes EC and EF with C stubs for the breakpoint
and execute-inline/range instructions, which were added after the last
bunch of stuff we rolled out to eclair open-source. I also rebuilt the
"out" dir to pick up the various other changes.
Ben Cheng [Thu, 25 Feb 2010 04:58:44 +0000 (20:58 -0800)]
Tweak the interpreter entries and 2nd level trace filter to capture more traces.
Real changes:
1) Add a new entry point from JIT to the interpreter to request hot traces w/o
doing chaining.
2) Increase the granularity of the secondary profile filter to match 64-byte
chunks using 64 entries.
The remaining are just cosmetic changes.
Brian Carlstrom [Fri, 26 Feb 2010 00:15:50 +0000 (16:15 -0800)]
Merge "First pass at reorganizing org.apache.harmony.xnet.provider.jsse native code into a single file."
Elliott Hughes [Fri, 26 Feb 2010 00:04:11 +0000 (16:04 -0800)]
Merge "Fix a couple of typos in JIT function names."
Brian Carlstrom [Thu, 11 Feb 2010 06:56:47 +0000 (22:56 -0800)]
First pass at reorganizing org.apache.harmony.xnet.provider.jsse native code into a single file.
At enh's suggestion, I'm consoldiating the OpenSSL related native code
into a single wrapper class NativeCrypto. This changes is the firs
step, combining the cpp code into a single NativeCode.cpp. The next
step will involved introducting a single SSL_CTX in NativeCode and
cleaning up SSL_CTX use. As part of this, I'll start moving the native
wrappers to from various OpenSSL*.java classes into NativeCode.
Bill Buzbee [Thu, 25 Feb 2010 23:43:39 +0000 (15:43 -0800)]
Merge "Jit: fix typo that caused incorrect def mask for blr's."
Elliott Hughes [Thu, 25 Feb 2010 23:41:42 +0000 (15:41 -0800)]
Fix a couple of typos in JIT function names.
(I saw these the other day, but preferred a separate patch.)
Bill Buzbee [Thu, 25 Feb 2010 23:38:40 +0000 (15:38 -0800)]
Jit: fix typo that caused incorrect def mask for blr's.
Elliott Hughes [Thu, 25 Feb 2010 22:59:44 +0000 (14:59 -0800)]
Merge "Document that Class.getSigners() always returns null."
Elliott Hughes [Thu, 25 Feb 2010 22:42:17 +0000 (14:42 -0800)]
Document that Class.getSigners() always returns null.
Bug:
1594865
Bug: http://code.google.com/p/android/issues/detail?id=1766
Andy McFadden [Thu, 25 Feb 2010 22:12:32 +0000 (14:12 -0800)]
Merge "Rearrange JDWP cleanup."
Andy McFadden [Thu, 25 Feb 2010 20:31:04 +0000 (12:31 -0800)]
Rearrange JDWP cleanup.
This moves the "get rid of the breakpoints" part before the "tell the
rest of the VM that there's no debugger attached" part. Hitting a
breakpoint when there's no debugger attached is not expected.
For bug
2461667.
Elliott Hughes [Thu, 25 Feb 2010 16:22:52 +0000 (08:22 -0800)]
Merge "Keep the order from getaddrinfo() unchanged if preferIPv6Addresses() is true."
Carl Shapiro [Wed, 24 Feb 2010 10:30:55 +0000 (02:30 -0800)]
Print the self thread id and the owning thread id in the description
of the IllegalMonitorExcpetion throw by unlockMonitor.
Ben Cheng [Thu, 25 Feb 2010 02:34:25 +0000 (18:34 -0800)]
Merge "Print "JIT" in the thread dump if the top frame is in JIT'ed code."
Andy McFadden [Thu, 25 Feb 2010 02:28:14 +0000 (18:28 -0800)]
Merge "Added printf attribute to exception function."
Ben Cheng [Thu, 25 Feb 2010 01:27:01 +0000 (17:27 -0800)]
Print "JIT" in the thread dump if the top frame is in JIT'ed code.
Elliott Hughes [Thu, 25 Feb 2010 00:36:18 +0000 (16:36 -0800)]
Optimize more easy multiplications by constants.
Rather than make these changes in the libraries (*10 being a common case),
let's do them once and for all in the JIT.
The 2^n-1 case could be better if we generated RSB instructions, but the
current "fake" RSB is still better than a full multiply.
Thumb doesn't support reg/reg/reg/shift instructions, so we can't optimize
the "population count <= 2" cases (such as *10) there.
Tested on sholes, passion, and passion-running-sapphire (and visually
inspected to check we weren't trying to generate Thumb2 instructions there).
Also tested with the self-verifier.
Ben Cheng [Wed, 24 Feb 2010 23:00:40 +0000 (15:00 -0800)]
Enhance the jit profiler to print more statistics and be more verbose.
Specifically, the per-trace invocation percentage is printed now.
D/dalvikvm( 671): TRACEPROFILE 0x42b6eed8
15449863 2.13% [0(+2), 253] Ljava/nio/Buffer;position;()I
D/dalvikvm( 671): TRACEPROFILE 0x42b7b3ac
15410023 2.13% [0x4(+3), 450] Lorg/apache/xml/serializer/CharInfo;shouldMapTextChar;(I)Z
D/dalvikvm( 671): TRACEPROFILE 0x42b7b2cc
15346582 2.12% [0(+2), 449] Lorg/apache/xml/serializer/CharInfo;shouldMapTextChar;(I)Z
D/dalvikvm( 671): TRACEPROFILE 0x42b7b40c
15013650 2.07% [0xb4(+2), 1527] Lorg/apache/xml/serializer/ToStream;characters;([CII)V
D/dalvikvm( 671): TRACEPROFILE 0x42b7b4f0
14907957 2.06% [0xc7(+2), 1537] Lorg/apache/xml/serializer/ToStream;characters;([CII)V
D/dalvikvm( 671): TRACEPROFILE 0x42b7b308
14898588 2.06% [0xad(+3), 1525] Lorg/apache/xml/serializer/ToStream;characters;([CII)V
D/dalvikvm( 671): TRACEPROFILE 0x42b7b52c
14798762 2.04% [0xc4(+2), 1523] Lorg/apache/xml/serializer/ToStream;characters;([CII)V
D/dalvikvm( 671): TRACEPROFILE 0x42b7b570
14763770 2.04% [0xf3(+2), 1577] Lorg/apache/xml/serializer/ToStream;characters;([CII)V
And the top 10 traces will be recompiled in verbose mode to facilitate code
quality analysis.
Andy McFadden [Wed, 24 Feb 2010 19:18:33 +0000 (11:18 -0800)]
am
cf4e8b02: Merge "This is a contribution of x86-atom targeted assembly for the fast byte-code interpreter engine. This is an initial contribution with minimal optimizations that target the Intel ATOM processor. We expect to continuously improve this code. It is expe
Merge commit '
cf4e8b02ecfb5ec17a27ed4188dd7a9db14ab7bf'
* commit '
cf4e8b02ecfb5ec17a27ed4188dd7a9db14ab7bf':
This is a contribution of x86-atom targeted assembly for the fast byte-code interpreter engine. This is an initial contribution with minimal optimizations that target the Intel ATOM processor. We expect to continuously improve this code. It is expected that there will be a discussion on the potential merge of this code and similar efforts (i.e. the mterp/x86 directory first included with cupcake). While this code is intended to target ATOM and not a generic X-86 processor, we were able to show the following improvements over the c-portable interpreter using the simulator build:
Dan Bornstein [Wed, 24 Feb 2010 18:54:08 +0000 (10:54 -0800)]
Clarify a few items in the dex spec.
In particular, do a bit of "may" vs. "must" maintenance, and be specific
about references as arguments to filled-new-array.
Change-Id: Iae4cb734e05cd29230e2c9a343108bbfb3c20193
Andy McFadden [Wed, 24 Feb 2010 17:39:52 +0000 (09:39 -0800)]
Added printf attribute to exception function.
Andy McFadden [Wed, 24 Feb 2010 17:39:09 +0000 (09:39 -0800)]
Merge "Add class init stats to alloc counters (API change)."