| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously we recreated this intern table during runtime startup.
This added 50-100ms of boot time.
Fixed bug where we didn't copy over hashcodes into the image.
Deleted some stale code.
Bug: 20727525
Bug: 19569780
Change-Id: I08959e9aa2a73cedb52f393033e2ffea3a26e76b
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Optimizing + quick tests are passing, devices boot.
TODO: Test and fix bugs in mips64.
Saves 16 bytes per most ArtMethod, 7.5MB reduction in system PSS.
Some of the savings are from removal of virtual methods and direct
methods object arrays.
Bug: 19264997
(cherry picked from commit e401d146407d61eeb99f8d6176b2ac13c4df1e33)
Change-Id: I622469a0cfa0e7082a2119f3d6a9491eb61e3f3d
Fix some ArtMethod related bugs
Added root visiting for runtime methods, not currently required
since the GcRoots in these methods are null.
Added missing GetInterfaceMethodIfProxy in GetMethodLine, fixes
--trace run-tests 005, 044.
Fixed optimizing compiler bug where we used a normal stack location
instead of double on ARM64, this fixes the debuggable tests.
TODO: Fix JDWP tests.
Bug: 19264997
Change-Id: I7c55f69c61d1b45351fd0dc7185ffe5efad82bd3
ART: Fix casts for 64-bit pointers on 32-bit compiler.
Bug: 19264997
Change-Id: Ief45cdd4bae5a43fc8bfdfa7cf744e2c57529457
Fix JDWP tests after ArtMethod change
Fixes Throwable::GetStackDepth for exception event detection after
internal stack trace representation change.
Adds missing ArtMethod::GetInterfaceMethodIfProxy call in case of
proxy method.
Bug: 19264997
Change-Id: I363e293796848c3ec491c963813f62d868da44d2
Fix accidental IMT and root marking regression
Was always using the conflict trampoline. Also included fix for
regression in GC time caused by extra roots. Most of the regression
was IMT.
Fixed bug in DumpGcPerformanceInfo where we would get SIGABRT due to
detached thread.
EvaluateAndApplyChanges:
From ~2500 -> ~1980
GC time: 8.2s -> 7.2s due to 1s less of MarkConcurrentRoots
Bug: 19264997
Change-Id: I4333e80a8268c2ed1284f87f25b9f113d4f2c7e0
Fix bogus image test assert
Previously we were comparing the size of the non moving space to
size of the image file.
Now we properly compare the size of the image space against the size
of the image file.
Bug: 19264997
Change-Id: I7359f1f73ae3df60c5147245935a24431c04808a
[MIPS64] Fix art_quick_invoke_stub argument offsets.
ArtMethod reference's size got bigger, so we need to move other args
and leave enough space for ArtMethod* and 'this' pointer.
This fixes mips64 boot.
Bug: 19264997
Change-Id: I47198d5f39a4caab30b3b77479d5eedaad5006ab
|
|
|
|
|
|
|
| |
Also fixed some lines that were too long, and a few other minor
details.
Change-Id: I6efba5fb6e03eb5d0a300fddb2a75bf8e2f175cb
|
|
|
|
|
|
|
|
| |
Don't pollute hash table with duplicates.
Check for existing value before storing a new one.
Change-Id: If43ea5e98a3f3ba187b8dea66f3712ec62bd3f60
Signed-off-by: nikolay serdjuk <nikolay.y.serdjuk@intel.com>
|
|
|
|
|
|
| |
Also some cleanup.
Change-Id: Ia3de8f2d409770be3619ec116e8b06ecd82338fe
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Changed GcRoot to use compressed references. Changed root visiting to
use virtual functions instead of function pointers. Changed root visting
interface to be an array of roots instead of a single root at a time.
Added buffered root marking helper to avoid dispatch overhead.
Root marking seems a bit faster on EvaluateAndApplyChanges due to batch
marking. Pause times unaffected.
Mips64 is untested but might work, maybe.
Before:
MarkConcurrentRoots: Sum: 67.678ms 99% C.I. 2us-664.999us Avg: 161.138us Max: 671us
After:
MarkConcurrentRoots: Sum: 54.806ms 99% C.I. 2us-499.986us Avg: 136.333us Max: 602us
Bug: 19264997
Change-Id: I0a71ebb5928f205b9b3f7945b25db6489d5657ca
|
|
|
|
|
| |
Bug: 12687968
Change-Id: I62f70274d47df6d6cab714df95c518b750ce3105
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Refactored old root callback to use a new class called RootInfo.
RootInfo contains all the relevant info related to the root
associated with the callback. The MarkSweep::VerifyRoot function
now uses this info to print the StackVisitor's described location
if the GC root is of the type kRootJavaFrame.
Some other cleanup.
Example output:
E/art (12167): Tried to mark 0x123 not contained by any spaces
E/art (12167): Attempting see if it's a bad root
E/art (12167): Found invalid root: 0x123 with type RootJavaFrame
E/art (12167): Location=Visiting method
'void java.lang.Runtime.gc()' at dex PC 0xffffffff (native PC 0x0)
vreg=0
(cherry picked from commit 12f7423a2bb4bfab76700d84eb6d4338d211983a)
Bug: 18588862
Change-Id: Ic5a2781f704e931265ffb3621c2eab4b2e25f60f
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
imageless
If we fail to create the image we use imageless running, this was
breaking AddImageStringsToTable since we didn't check for this
case. Added a check which avoids calling AddImageStringsToTable and
MoveImageClassesToClassTable if we don't have an image.
Bug: 18412472
(cherry picked from commit ab1953f35880f94752e1fcde3f6b3f0d3532125b)
Change-Id: Ida18e1fcd04d752d55923e65c3122c0ac0bde728
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
More memory efficient than libcxx since we do not box the values.
Change intern table to use new hash set. Clean up intern table by
removing const casts and deleting unnecessary code.
Changed the class linker to use a hash set, also added a pre-zygote
class table.
5 samples of:
adb shell stop && adb shell start && sleep 60 && adb shell dumpsys meminfo
Before:
165929 kB: Native
175859 kB: Native
168434 kB: Native
166559 kB: Native
169958 kB: Native
After:
160972 kB: Native
159439 kB: Native
157204 kB: Native
165093 kB: Native
163039 kB: Native
TODO: Add HashTable which is implemented by using a HashSet.
TODO: Use for DexFile::find_class_def_misses_.
TODO: Investigate using mem maps instead of native heap.
Bug: 17808975
Change-Id: I93e376cf6eb9628cf52f4aefdadb6157acfb799a
(cherry picked from commit e05d1d5fd86867afc7513b1c546375dba11eee50)
|
|
|
|
|
|
|
| |
Enable -Wno-conversion-null, -Wredundant-decls and -Wshadow in general,
and -Wunused-but-set-parameter for GCC builds.
Change-Id: I81bbdd762213444673c65d85edae594a523836e5
|
|
|
|
|
|
|
|
|
|
|
| |
Fix associated errors about unused paramenters and implict sign conversions.
For sign conversion this was largely in the area of enums, so add ostream
operators for the effected enums and fix tools/generate-operator-out.py.
Tidy arena allocation code and arena allocated data types, rather than fixing
new and delete operators.
Remove dead code.
Change-Id: I5b433e722d2f75baacfacae4d32aef4a828bfe1b
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When we create the runtime, we now add the image strings to the
intern table if we are the zygote. This caused some memory bloat,
so I added an extra unordered set to the intern table.
There is now two unordered sets (hash talbe). One for pre-zygote
interns and one for post-zygote interns. This helps since the
pre-zygote hash table doesn't get dirtied. Even with adding
the image strings, we get total memory savings of around 5-7 MB
native PSS after device boot.
FB launch Before:
2.20% art::DexFile::FindStringId(char const*) const
TotalTime: 2069
TotalTime: 1985
TotalTime: 2088
TotalTime: 2003
TotalTime: 2034
TotalTime: 2049
After boot native PSS: 175585 kB: Native
After:
0.27% art::DexFile::FindStringId(char const*) const
TotalTime: 1682
TotalTime: 1756
TotalTime: 1825
TotalTime: 1751
TotalTime: 1666
TotalTime: 1813
After boot native PSS: 167089 kB: Native
Bug: 18054905
Bug: 16828525
Bug: 17808975
(cherry picked from commit b6e292bf7eac9d73c6b79b1e9b7b87beb02436c9)
Change-Id: Ie367f3222f8c4db409ec49c3845276908b51e9c9
|
|
|
|
|
|
| |
Also deleted GcRoot::Assign.
Change-Id: Ib1ea739cf79c72fc92a8628cd9095c11b660e018
|
|
|
|
|
|
|
|
|
|
|
| |
Intern table active used bytes goes from 430k to 317k on system
server. Similar %wise savings on other apps.
Bug: 16238192
(cherry picked from commit d910fcef539e12ab181e56ec80684f39c4e95733)
Change-Id: Ic70395124435c6f420a77e6d8639404a160f395a
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Introduce a value-type wrapper around Object* for GC roots so that 1)
we won't have to directly add the read barrier code in many places and
2) we can avoid accidentally bypassing/missing read barriers on GC
roots (the GcRoot interface ensures that the read barrier is executed
on a read).
The jdwp test passed.
Bug: 12687968
Change-Id: Ib167c7c325b3c7e3900133578815f04d219972a1
|
|
|
|
|
|
|
|
|
| |
Add read barriers for the strong roots in the intern table and the
(strong) roots in the class table to make possible concurrent scanning
of them.
Bug: 12687968
Change-Id: If6edc33a37e65a8494e66dc3b144138b1530367f
|
|
|
|
|
| |
Bug: 12687968
Change-Id: I424f1df76a7e3d7154fb9f3c951c973d19bd640f
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When we have a hash code collision but the string is not yet
in the intern table, the old Lookup() would iterate until
the end of the table, i.e. it was up to linear in the size
of the table rather than linear in the number of colliding
hash codes.
And though the multimap::find() is implemented in terms of
lower_bound() in the standard library we're using, this
requirement doesn't seem to be in the C++ standard. It was
therefore wrong to assume that we will iterate across all
hits when starting from the iterator returned by find().
Change-Id: Ie24aaea6e55750a7aafbed24b136878c5dba66eb
|
|
|
|
|
|
|
| |
Also remove the Android.libcxx.mk and other bits of stlport compatibility
mechanics.
Change-Id: Icdf7188ba3c79cdf5617672c1cfd0a68ae596a61
|
|
|
|
|
|
|
|
|
| |
Use ART_WITH_STLPORT (enabled for the target) to cause the use of UniquePtr,
for the host switch to std::unique_ptr. For now the type remains called
UniquePtr.
Make dalvik compile with clang on the host, move its build to C++11.
Change-Id: I5ba8d2757904bc089ed62047ea03de3c0853fb12
|
|
|
|
|
|
|
|
|
|
| |
If you pass in a null utf string it should not be the same behavior
as out of memory.
This previously caused serious problems in:
https://android-review.googlesource.com/#/c/80768/
Change-Id: I9dfb710b57f6cc91064812f52a3db64254769461
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The new root visiting logic has a concept of a root log which holds
new roots which were added since the start of the GC. This is an
optimization since it lets us only mark these newly added roots
during the pause (or pre-cleaning) since the other roots intern table
and class linker roots were marked concurrently at the start of the
GC.
Before (EvaluateAndApplyChanges):
MarkConcurrentRoots: Sum: 605.193ms
After:
MarkConcurrentRoots: Sum: 271.858ms
This should also reduce pathological GC pauses which used to be able
to happen when the intern table or class linker became "dirty"
during the concurrent GC.
Change-Id: I433fab021f2c339d50c35aaae7161a50a0901dec
|
|
|
|
|
|
|
| |
Simplifies code and improves the performance of root visiting since
we usually don't need to check to see if the object moved.
Change-Id: Iba998f5a15ae1fa1b53ca5226dd2168a411196cf
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Removes the class initialization blacklist and use transaction to detect and
revert class initialization attempting to invoke native method. This only
concerns class initialization happening at compilation time when generating an
image (like boot.art for the system).
In transactional mode, we log every object's field assignment and array update.
Therefore we're able to abort a transaction to restore values of fields and
array as they were before the transaction starts. We also log changes to the
intern string table so we can restore its state prior to transaction start.
Since transactional mode only happens at compilation time, we don't need to log
all these changes at runtime. In order to reduce the overhead of testing if
transactional mode is on/off, we templatize interfaces of mirror::Object and
mirror::Array, respectively responsible for setting a field and setting an
array element.
For various reasons, we skip some specific fields from transaction:
- Object's class and array's length must remain unchanged so garbage collector
can compute object's size.
- Immutable fields only set during class loading: list of fields, method,
dex caches, vtables, ... as all classes have been loaded and verified before a
transaction occurs.
- Object's monitor for performance reason.
Before generating the image, we browse the heap to collect objects that need to
be written into it. Since the heap may still holds references to unreachable
objects due to aborted transactions, we trigger one collection at the end of
the class preinitialization phase.
Since the transaction is held by the runtime and all compilation threads share
the same runtime, we need to ensure only one compilation thread has exclusive
access to the runtime. To workaround this issue, we force class initialization
phase to run with only one thread. Note this is only done when generating image
so application compilation is not impacted. This issue will be addressed in a
separate CL.
Bug: 9676614
Change-Id: I221910a9183a5ba6c2b99a277f5a5a68bc69b5f9
|
|
|
|
|
|
|
| |
Enables us to pass the root type and thread id to hprof.
Bug: 12680863
Change-Id: I6a0f1f9e3aa8f9b4033d695818ae7ca3460d67cb
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The compacting collector is currently similar to semispace. It works by
copying objects back and forth between two bump pointer spaces. There
are types of objects which are "non-movable" due to current runtime
limitations. These are Classes, Methods, and Fields.
Bump pointer spaces are a new type of continuous alloc space which have
no lock in the allocation code path. When you allocate from these it uses
atomic operations to increase an index. Traversing the objects in the bump
pointer space relies on Object::SizeOf matching the allocated size exactly.
Runtime changes:
JNI::GetArrayElements returns copies objects if you attempt to get the
backing data of a movable array. For GetArrayElementsCritical, we return
direct backing storage for any types of arrays, but temporarily disable
the GC until the critical region is completed.
Added a new runtime call called VisitObjects, this is used in place of
the old pattern which was flushing the allocation stack and walking
the bitmaps.
Changed image writer to be compaction safe and use object monitor word
for forwarding addresses.
Added a bunch of added SIRTs to ClassLinker, MethodLinker, etc..
TODO: Enable switching allocators, compacting on background, etc..
Bug: 8981901
Change-Id: I3c886fd322a6eef2b99388d19a765042ec26ab99
|
|\
| |
| |
| | |
Change-Id: I31e0c05c5e917cf3275296a7f65bd8b78ec9b78b
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Enabled by disallowing new system weaks during the pause and
re-allowing it after the system weaks have been swept. Reduces
GC pause by ~1ms.
Fixes pause regression caused by fix for
Bug: 10626133
Change-Id: If49d33e7ef19cb728ed3cef5187acfa53b9b05d8
|
|\ \
| | |
| | |
| | | |
dalvik-dev
|
| | |
| | |
| | |
| | |
| | |
| | | |
Required for moving collectors.
Change-Id: Ib97ba4a05af1139f8d388077a15e62bcb9534855
|
|\ \ \
| |/ /
|/| /
| |/ |
Change-Id: I61734cbf585decf9a50ba23f008b341d025339ff
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
There was a race which caused the class linker / intern table to not
become dirty after adding a root. We now guard the is dirty flag by
the corresponding locks to prevent this from occuring. This was
causing roots to be occasionally missed.
Also fixes the bug where we occasionally scan more cards than
needed.
Bug: 10626133
Change-Id: I0f6e72d92035ff463954d66988ef610ea0df61be
|
|/
|
|
|
|
| |
Needed for copying collectors.
Change-Id: Icc4a342a57e0cfb79587edb02ef8c85e08808877
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Bug 10432288.
Find Classes and Strings from dex caches lazily rather than when the image is
loaded.
Make class status changes do notifies when there can be waiters.
For Class lookup there's a pathology if we always search dex caches and
so after 1000 failures move all classes into the class table.
Be consistent in using "const char*" for class linker descriptors as this
most easily agrees with the type in the dex file.
Improve the intern run-test so that it has a case of a literal contained in the
image.
Modify image_test to allow any valid lock word rather than expecting 0, ideally
we wouldn't see inflated monitors but we do due to NotifyAll (see bug 6961405).
Change-Id: Ia9bfa748eeccb9b4498784b97c6823141b1f6db8
|
|
|
|
|
|
|
|
|
|
|
|
| |
Needed to update cpplint to handle const auto.
Fixed a few cpplint errors that were being missed before.
Replaced most of the TODO c++0x with ranged based loops. Loops which
do not have a descriptive container name have a concrete type instead
of auto.
Change-Id: Id7cc0f27030f56057c544e94277300b3f298c9c5
|
|
|
|
| |
Change-Id: Iae286862c85fb8fd8901eae1204cd6d271d69496
|
|
The runtime, compiler, dex2oat, and oatdump now are in seperate trees
to prevent dependency creep. They can now be individually built
without rebuilding the rest of the art projects. dalvikvm and jdwpspy
were already this way. Builds in the art directory should behave as
before, building everything including tests.
Change-Id: Ic6b1151e5ed0f823c3dd301afd2b13eb2d8feb81
|