| e9262fc38f6fc3645a209fac7c4919e4d9cda576 |
|
04-Jan-2017 |
Colin Cross <ccross@android.com> |
Allow invokevirtual, invokespecial, invokestatic on interface methods
Allow default and static methods on interfaces if
--min-sdk-version >= 24.
Test: make checkbuild
Change-Id: I6b617a37256bdb95f4c11e58fe2ebf08cf7aa324
|
| 16fc52fde7c71ebdd14a03d0434058dbfa48b4ab |
|
26-Jan-2017 |
Chris Warrington <cmw@google.com> |
Make Dx argument parser reusable.
Test: will be part of tools/base tests.
Change-Id: Ib21c9ca2a95a5b78d0a229234a4f03a56cdba29c
|
| f809c2f0028b3df79787a5da5d5b35ce85331a32 |
|
23-Jan-2017 |
Chris Warrington <cmw@google.com> |
Use ConcurrentHashMaps for intern tables.
To reduce lock contention when predexing in process in parallel.
Test: existing.
Change-Id: Ia929bc10f57b1d2d4c62cffff4fc0f09cbf9e445
|
| 1c57b94a7dff8b47f8b1e52503834aed1baf2c36 |
|
23-Jan-2017 |
Chris Warrington <cmw@google.com> |
Dx: use context out/err everywhere.
Test: manual.
Change-Id: I14289dbd2e507372d82dd5cc64b4fac761026a5e
|
| e8bf740e4843a2e7a41bfa6c2f96f8ae4a7147fc |
|
20-Jan-2017 |
Chris Warrington <cmw@google.com> |
Javadoc fixes.
Test: Manual.
Change-Id: I306d61e3fe49353b64ebee24413a83062365968d
|
| 8f68769869e02895dc6474a5cd0bca20977e5ecd |
|
21-Nov-2016 |
Chris Warrington <cmw@google.com> |
Make dx code re-entrant
Remove static state, clear biggest caches after running.
Test: TBD.
Change-Id: Ib41a61687b48bfb46f7e104c41d784cad61f306a
|
| 900c3b497902357345a9444a972ac0a3bea1f09c |
|
10-Feb-2016 |
Liam Miller-Cushon <cushon@google.com> |
Improve dx error handling
If a class file cannot be parsed, print the error message and context
(including file name), and omit the stack trace unless --debug is
enabled.
Bug: 25075831
(cherry picked from commit bd156605a787bbbb5524b6b25254b37a67e6dd7f)
Change-Id: I1615591e6c389e2c99bbc024da22a59abe62a3a2
|
| bd156605a787bbbb5524b6b25254b37a67e6dd7f |
|
10-Feb-2016 |
Liam Miller-Cushon <cushon@google.com> |
Improve dx error handling
If a class file cannot be parsed, print the error message and context
(including file name), and omit the stack trace unless --debug is
enabled.
Bug: 25075831
Change-Id: I0a790a5b329640acb7185dbf94408b21dd17de8c
|
| 711bc49d66e3360ba818910b64e66ddd0ccc18ee |
|
17-Oct-2015 |
Rohit Agrawal <rohitagr@google.com> |
Add more context to dx if debugging.
Change-Id: Ib1fb4acd2ebac842acd7bd95952317ffa83fb54f
|
| f8ed7d3f0ea87b3fc1aa72859386c82d8d9bdbd4 |
|
12-Sep-2015 |
Esteban de la Canal <estebandlc@google.com> |
Fixes unnecessary multi-merge steps.
Changes merging from quadratic in the number of classes to linear.
This has a tremendous speed up while merging many dexes at the same time.
A sample test (iosched app) with 29 dexes goes from 6 seconds to 1.
Change-Id: Iff02a0dc44d098b0878e88d18f5f4083804a6495
|
| 259c2df556dd80998366fc23debb20cbd2d49be3 |
|
12-Sep-2015 |
Esteban de la Canal <estebandlc@google.com> |
Fixes unnecessary multi-merge steps.
Changes merging from quadratic in the number of classes to linear.
This has a tremendous speed up while merging many dexes at the same time.
A sample test (iosched app) with 29 dexes goes from 6 seconds to 1.
Change-Id: Iff02a0dc44d098b0878e88d18f5f4083804a6495
|
| 74fe74478c9112ab9fb042f1dfb0d9ac2b7ab9e2 |
|
11-Sep-2015 |
Nicolas Geoffray <ngeoffray@google.com> |
Revert "Fixes unnecessary multi-merge steps."
Breaks continuous libcore tests.
This reverts commit 964d093874aa3533aaf899429dc3820f0fdef4e6.
Change-Id: Ic22b473094425b4dbf5a15a6e1a4c9f66b1b4001
|
| 964d093874aa3533aaf899429dc3820f0fdef4e6 |
|
20-Aug-2015 |
estebandlc <estebandlc@google.com> |
Fixes unnecessary multi-merge steps.
Changes merging from quadratic in the number of classes to linear.
This has a tremendous speed up while merging many dexes at the same time.
A sample test (iosched app) with 29 dexes goes from 6 seconds to 1.
Change-Id: Ie287bdb32b76ba06f362437b05f928bb7b528493
|
| ab01c96b3c21098dffbba03248f801486c76bdc7 |
|
17-Aug-2015 |
Yohann Roussel <yroussel@google.com> |
Ensure deterministic multidex partitioning
This is done by sorting inputs before processing them.
Bug: 23305646
Change-Id: Ia55c481f4eb342bc925bc100299324ea470bf9a0
|
| 635aecf324d618ef45d82183d148d47e08383601 |
|
25-Mar-2015 |
Peter Jensen <jensenp@google.com> |
Fix potential dex index overflow in dx
Revert change to dex rotation algorithm made as part of commit:
845d9d0eed0f6556e11ee7f7204fda9c8dd41154
Bug: https://code.google.com/p/android/issues/detail?id=161887
Signed-off-by: Peter Jensen <jensenp@google.com>
(cherry picked from commit d024c06726c7e119ee76395611aa7cfe3b0a8b7e)
Change-Id: Iede742620854cef3542bf98b77e66aa8a305195e
|
| bd3b381a74023a63b3713749e4be02429467f789 |
|
13-Nov-2014 |
Peter Jensen <jensenp@google.com> |
Support --num-threads with --multi-dex (take 2)
With fix for regression introduced in original commit.
The current dx implementation supports options --multi-dex, for applications
not fitting within the dex format limitations; and --num-threads=N, triggers
concurrent processing of multiple input files. However, the implementation
has the following limitations:
The --num-threads option is disabled when used together with --multi-dex.
The --num-threads option implements concurrency at the level of classpath
entries, and does nothing when the classes to be translated are specified
with a single classpath element (e.g. single jar output from Proguard).
The existing --num-threads implementation may produce indeterministic output.
The heuristic used by the --multi-dex option to determine when to rotate the
dex output file is overly conservative.
The primary objective of this change is:
Concurrent translation of classes, independently of input specification format.
Support --num-threads=N in both mono- and multi-dex mode.
Deterministic class output order.
Near optimal use of dex file format capacity.
This is accomplished by reorganizing the dx workflow in a pipeline of
concurrent phases.
read-class | parse-class | translate-class | add-to-dex | convert-dex-to-byte[];
output-dex-files-or-jar
To manage dex file rotation (i.e. --multi-dex support), the parse-class and
add-to-dex phases are synchronized to prevent forwarding classes to the
translate-class phase if it could potentially result in breaking the dex
format limitations. The heuristic currently used to estimate the number of
indices needed for a class is improved, to minimize the amount of serialization
imposed by this feedback mechanism, and to improve the use of dex file capacity.
The translate-class and convert-dex-to-byte[] phases are further parallelized
with configurable (--num-threads=N option) thread pools. This allow translating
classes concurrently, while also performing output conversion in parallel.
Separate collector threads are used to collect results from the thread pools
in deterministic order.
Testing was performed on an Ubuntu system, with 6 cores and 12 hardware threads.
The taskset command was used to experimentally establish that running with more
than 8 hardware threads does not provide any additional benefit.
Experiments shows that the argument to --num-threads should not exceed the
lesser of the number of available hardware threads, and 5. Setting it to a
higher value results in no additional benefit.
The gain is generally larger for larger applications, and not significant for
small applications with less than a few thousands classes. Experiments with
generated classes shows that for large applications gains as high as 50% may
be possible.
For an existing real-life application with more than 11k classes, and requiring
2 dex files, a speed-up of 37% was achieved (--num-threads=5, 8 hardware
threads, 4g Java heap). A speedup of 31% was observed for another application
with ~7 classes.
For small applications, use of --num-threads=N>1 doesn’t provide significant
benefit. Running with --num-threads=1, the modified dx is slightly faster,
but no significant gain is observed unless the application requires multiple
dex files.
The one case where a significant regression may be observed is when using
--num-threads=N>1, with a single hardware thread. This is an inappropriate
configuration, even with the current implementation. However, because of
the limitations of the current implementation, such configurations may exist.
For instance, a configuration using both --multi-dex and --num-threads=5 will
currently generate a warning about using the two options together. With the
new implementation, the options can legitimately be used together, and could
result in an ~20% regression running on a single hardware thread.
Note: the current dx implementation, without --num-threads option, is already
approximately 50% slower with 1 hardware thread, compared to running with 2
or more. With 2 hardware threads the implementations are practically at par
(a little better, or a little worse, depending on the application).
Testing:
Tested with 6 existing applications ranging in size from 1K - 12K classes.
Updated and tested with relevant existing unit tests (one test changed to
account for better dex rotation heuristic).
Added unit test to test deterministic output.
Added unit performance test. By default run script merely validates that
--multi-dex and --num-threads can be used together (fast). However, the test
is configurable to perform performance test, over sets of generated classes.
Signed-off-by: Peter Jensen <jensenp@google.com>
(cherry picked from commit 845d9d0eed0f6556e11ee7f7204fda9c8dd41154)
(cherry picked from commit dd140a22d90495045024334a91770acaad8e065e)
Change-Id: I33a8ea0451efc0af7eb1d72e80cb926d6583d569
|
| dff33a1061ef7ac40fc7035bfbcd7e2512a3ca30 |
|
17-Dec-2014 |
Benoit Lamarche <benoitlamarche@google.com> |
Do not keep the DexFile field in Main between runs
Bug: 18486731
(cherry picked from commit bb404aa9c531c2e1d27b4f7c6d2b03c2b50fb871)
Change-Id: I048f6064cc49c78df21a1c42316761502a0a5366
|
| e345a3565597b76ee5bf6bfbf2c84ce9271be079 |
|
31-Oct-2014 |
Csaba Kozák <kozakcsabi@gmail.com> |
Command line option for disabling warnings
This commit adds a new command line option to dx:
--no-warning.
If the caller appends this to the arguments, dx won't print
warnings to System.err. This can be useful if the user
dexes external jars with old class format, so dx would
print lot of warning about those.
Bug: https://code.google.com/p/android/issues/detail?id=78285
Signed-off-by: Csaba Kozák <kozakcsabi@gmail.com>
(cherry picked from commit ef1de423e70704c478ee77956f44b0a040d8bede)
Change-Id: I6d09c684b5eb97aa28e0b12e3ef44b46293c7dec
|
| d024c06726c7e119ee76395611aa7cfe3b0a8b7e |
|
25-Mar-2015 |
Peter Jensen <jensenp@google.com> |
Fix potential dex index overflow in dx
Revert change to dex rotation algorithm made as part of commit:
845d9d0eed0f6556e11ee7f7204fda9c8dd41154
Bug: https://code.google.com/p/android/issues/detail?id=161887
Change-Id: I255aa52b74c921a78e77671ae6187157fef011e1
Signed-off-by: Peter Jensen <jensenp@google.com>
|
| dd140a22d90495045024334a91770acaad8e065e |
|
13-Nov-2014 |
Peter Jensen <jensenp@google.com> |
Support --num-threads with --multi-dex (take 2)
With fix for regression introduced in original commit.
The current dx implementation supports options --multi-dex, for applications
not fitting within the dex format limitations; and --num-threads=N, triggers
concurrent processing of multiple input files. However, the implementation
has the following limitations:
The --num-threads option is disabled when used together with --multi-dex.
The --num-threads option implements concurrency at the level of classpath
entries, and does nothing when the classes to be translated are specified
with a single classpath element (e.g. single jar output from Proguard).
The existing --num-threads implementation may produce indeterministic output.
The heuristic used by the --multi-dex option to determine when to rotate the
dex output file is overly conservative.
The primary objective of this change is:
Concurrent translation of classes, independently of input specification format.
Support --num-threads=N in both mono- and multi-dex mode.
Deterministic class output order.
Near optimal use of dex file format capacity.
This is accomplished by reorganizing the dx workflow in a pipeline of
concurrent phases.
read-class | parse-class | translate-class | add-to-dex | convert-dex-to-byte[];
output-dex-files-or-jar
To manage dex file rotation (i.e. --multi-dex support), the parse-class and
add-to-dex phases are synchronized to prevent forwarding classes to the
translate-class phase if it could potentially result in breaking the dex
format limitations. The heuristic currently used to estimate the number of
indices needed for a class is improved, to minimize the amount of serialization
imposed by this feedback mechanism, and to improve the use of dex file capacity.
The translate-class and convert-dex-to-byte[] phases are further parallelized
with configurable (--num-threads=N option) thread pools. This allow translating
classes concurrently, while also performing output conversion in parallel.
Separate collector threads are used to collect results from the thread pools
in deterministic order.
Testing was performed on an Ubuntu system, with 6 cores and 12 hardware threads.
The taskset command was used to experimentally establish that running with more
than 8 hardware threads does not provide any additional benefit.
Experiments shows that the argument to --num-threads should not exceed the
lesser of the number of available hardware threads, and 5. Setting it to a
higher value results in no additional benefit.
The gain is generally larger for larger applications, and not significant for
small applications with less than a few thousands classes. Experiments with
generated classes shows that for large applications gains as high as 50% may
be possible.
For an existing real-life application with more than 11k classes, and requiring
2 dex files, a speed-up of 37% was achieved (--num-threads=5, 8 hardware
threads, 4g Java heap). A speedup of 31% was observed for another application
with ~7 classes.
For small applications, use of --num-threads=N>1 doesn’t provide significant
benefit. Running with --num-threads=1, the modified dx is slightly faster,
but no significant gain is observed unless the application requires multiple
dex files.
The one case where a significant regression may be observed is when using
--num-threads=N>1, with a single hardware thread. This is an inappropriate
configuration, even with the current implementation. However, because of
the limitations of the current implementation, such configurations may exist.
For instance, a configuration using both --multi-dex and --num-threads=5 will
currently generate a warning about using the two options together. With the
new implementation, the options can legitimately be used together, and could
result in an ~20% regression running on a single hardware thread.
Note: the current dx implementation, without --num-threads option, is already
approximately 50% slower with 1 hardware thread, compared to running with 2
or more. With 2 hardware threads the implementations are practically at par
(a little better, or a little worse, depending on the application).
Testing:
Tested with 6 existing applications ranging in size from 1K - 12K classes.
Updated and tested with relevant existing unit tests (one test changed to
account for better dex rotation heuristic).
Added unit test to test deterministic output.
Added unit performance test. By default run script merely validates that
--multi-dex and --num-threads can be used together (fast). However, the test
is configurable to perform performance test, over sets of generated classes.
Signed-off-by: Peter Jensen <jensenp@google.com>
(cherry picked from commit 845d9d0eed0f6556e11ee7f7204fda9c8dd41154)
Change-Id: I721effa31c3b1a8b427d3a18ec554a19c5e9765b
|
| c8b036e3fb5e88eb501e953a8a8838b547f2dab4 |
|
09-Feb-2015 |
Benoit Lamarche <benoitlamarche@google.com> |
Revert "Support --num-threads with --multi-dex"
This reverts commit 845d9d0eed0f6556e11ee7f7204fda9c8dd41154.
Bug: 19313927
Change-Id: Ia6582a3914cc33762aef74da1f5a6a153c8c0ab2
|
| 845d9d0eed0f6556e11ee7f7204fda9c8dd41154 |
|
13-Nov-2014 |
Peter Jensen <jensenp@google.com> |
Support --num-threads with --multi-dex
The current dx implementation supports options --multi-dex, for applications
not fitting within the dex format limitations; and --num-threads=N, triggers
concurrent processing of multiple input files. However, the implementation
has the following limitations:
The --num-threads option is disabled when used together with --multi-dex.
The --num-threads option implements concurrency at the level of classpath
entries, and does nothing when the classes to be translated are specified
with a single classpath element (e.g. single jar output from Proguard).
The existing --num-threads implementation may produce indeterministic output.
The heuristic used by the --multi-dex option to determine when to rotate the
dex output file is overly conservative.
The primary objective of this change is:
Concurrent translation of classes, independently of input specification format.
Support --num-threads=N in both mono- and multi-dex mode.
Deterministic class output order.
Near optimal use of dex file format capacity.
This is accomplished by reorganizing the dx workflow in a pipeline of
concurrent phases.
read-class | parse-class | translate-class | add-to-dex | convert-dex-to-byte[];
output-dex-files-or-jar
To manage dex file rotation (i.e. --multi-dex support), the parse-class and
add-to-dex phases are synchronized to prevent forwarding classes to the
translate-class phase if it could potentially result in breaking the dex
format limitations. The heuristic currently used to estimate the number of
indices needed for a class is improved, to minimize the amount of serialization
imposed by this feedback mechanism, and to improve the use of dex file capacity.
The translate-class and convert-dex-to-byte[] phases are further parallelized
with configurable (--num-threads=N option) thread pools. This allow translating
classes concurrently, while also performing output conversion in parallel.
Separate collector threads are used to collect results from the thread pools
in deterministic order.
Testing was performed on an Ubuntu system, with 6 cores and 12 hardware threads.
The taskset command was used to experimentally establish that running with more
than 8 hardware threads does not provide any additional benefit.
Experiments shows that the argument to --num-threads should not exceed the
lesser of the number of available hardware threads, and 5. Setting it to a
higher value results in no additional benefit.
The gain is generally larger for larger applications, and not significant for
small applications with less than a few thousands classes. Experiments with
generated classes shows that for large applications gains as high as 50% may
be possible.
For an existing real-life application with more than 11k classes, and requiring
2 dex files, a speed-up of 37% was achieved (--num-threads=5, 8 hardware
threads, 4g Java heap). A speedup of 31% was observed for another application
with ~7 classes.
For small applications, use of --num-threads=N>1 doesn’t provide significant
benefit. Running with --num-threads=1, the modified dx is slightly faster,
but no significant gain is observed unless the application requires multiple
dex files.
The one case where a significant regression may be observed is when using
--num-threads=N>1, with a single hardware thread. This is an inappropriate
configuration, even with the current implementation. However, because of
the limitations of the current implementation, such configurations may exist.
For instance, a configuration using both --multi-dex and --num-threads=5 will
currently generate a warning about using the two options together. With the
new implementation, the options can legitimately be used together, and could
result in an ~20% regression running on a single hardware thread.
Note: the current dx implementation, without --num-threads option, is already
approximately 50% slower with 1 hardware thread, compared to running with 2
or more. With 2 hardware threads the implementations are practically at par
(a little better, or a little worse, depending on the application).
Testing:
Tested with 6 existing applications ranging in size from 1K - 12K classes.
Updated and tested with relevant existing unit tests (one test changed to
account for better dex rotation heuristic).
Added unit test to test deterministic output.
Added unit performance test. By default run script merely validates that
--multi-dex and --num-threads can be used together (fast). However, the test
is configurable to perform performance test, over sets of generated classes.
Change-Id: Ic2d11c422396e97171c2e6ceae9477113e261b8e
Signed-off-by: Peter Jensen <jensenp@google.com>
|
| bb404aa9c531c2e1d27b4f7c6d2b03c2b50fb871 |
|
17-Dec-2014 |
Benoit Lamarche <benoitlamarche@google.com> |
Do not keep the DexFile field in Main between runs
Bug: 18486731
Change-Id: I76dbdd709182c4663f2bbf42179f55d0184e97c2
|
| 2f99a0c9da7a067938dc2598c5dfebb5a2002b05 |
|
14-Nov-2014 |
Xavier Ducrohet <xav@google.com> |
Keep --input-list order.
Merging dex files starting with smaller ones can have
a big impact on performance as it potentially reduces
the number of compact events during the merge.
Gradle will now send the list sorted in ascending order
of file size, and dex should keep this order when parsing
the input list file.
Change-Id: Iff3ae4d30106fe70dca0764b33addfecceb84d09
|
| 9747950064e4711c15baf0986102320a2512097a |
|
14-Nov-2014 |
Xavier Ducrohet <xav@google.com> |
Keep --input-list order.
Merging dex files starting with smaller ones can have
a big impact on performance as it potentially reduces
the number of compact events during the merge.
Gradle will now send the list sorted in ascending order
of file size, and dex should keep this order when parsing
the input list file.
Change-Id: Iff3ae4d30106fe70dca0764b33addfecceb84d09
|
| ef1de423e70704c478ee77956f44b0a040d8bede |
|
31-Oct-2014 |
Csaba Kozák <kozakcsabi@gmail.com> |
Command line option for disabling warnings
This commit adds a new command line option to dx:
--no-warning.
If the caller appends this to the arguments, dx won't print
warnings to System.err. This can be useful if the user
dexes external jars with old class format, so dx would
print lot of warning about those.
Bug: https://code.google.com/p/android/issues/detail?id=78285
Change-Id: Ic2da274ec72e096bcf871c1f4512c54ddc228721
Signed-off-by: Csaba Kozák <kozakcsabi@gmail.com>
|
| 8a5d5ccc26c97d555024c36f54140f07a4b40b8c |
|
08-Aug-2014 |
John Fazekas <jfazekas@fd-software.com> |
Fixes command line too long error on Windows
On Windows machines the command line can only accept
commands that are 8192 characters or less. This can
cause a problem when a project has many dependencies
and many libraries are being passed to dex at once.
This fixes it by adding a new option to dex called
--input-list that can read inputs from a file.
This way we only have to pass a single file name
instead of a ton of jar files. This keeps the command
short so Windows will not complain about it.
Bug: 17438982
Signed-off-by: John Fazekas <jfazekas@fd-software.com>
(cherry picked from commit 914f2263ca57b21dc4ef1fb218e345a16a5a5db5)
Change-Id: I6edd1e7c1d1fddc8d6bf4c4963295d4f4e64032e
|
| 914f2263ca57b21dc4ef1fb218e345a16a5a5db5 |
|
08-Aug-2014 |
John Fazekas <jfazekas@fd-software.com> |
Fixes command line too long error on Windows
On Windows machines the command line can only accept
commands that are 8192 characters or less. This can
cause a problem when a project has many dependencies
and many libraries are being passed to dex at once.
This fixes it by adding a new option to dex called
--input-list that can read inputs from a file.
This way we only have to pass a single file name
instead of a ton of jar files. This keeps the command
short so Windows will not complain about it.
Change-Id: Ia62ed125c66b24ac81799d3bbac460ae07495b25
Signed-off-by: John Fazekas <jfazekas@fd-software.com>
|
| 084b7f109aa96a86a267074f5464732b8062bd5a |
|
21-Jul-2014 |
Yohann Roussel <yroussel@google.com> |
Do not generate empty dex in multidex.
This was possible when the first class considered for adding in a dex was too
big to be sure it will fit. Now DX will try to add the class in the current dex
and crash if the class causes one index to overflow.
(cherry picked from commit f15dafe674d2270cafaf4fb7993963ec474d3761)
Change-Id: I323f8111c2be6bace908d0539ff324c4bd033c05
|
| f15dafe674d2270cafaf4fb7993963ec474d3761 |
|
21-Jul-2014 |
Yohann Roussel <yroussel@google.com> |
Do not generate empty dex in multidex.
This was possible when the first class considered for adding in a dex was too
big to be sure it will fit. Now DX will try to add the class in the current dex
and crash if the class causes one index to overflow.
Change-Id: Ia1f8f733ee49e24bbb42db335ac9080312c1907a
|
| 4ec4dc5c35060cb2346770c84a6e705a45a45804 |
|
13-May-2014 |
Yohann Roussel <yroussel@google.com> |
Make a better detection of index limit in multidex.
- Intern annotation content sooner.
- Count declared methods separatly because it seems possible to have them
generate more method ids than they account in the constant pool if their
name and descriptor are the same as those of called methods.
- Take into account some possible ids that can be added during dex creation.
- Drop evaluation of type ids numbers because it was wrong, it's not simple
to do better and it's nearly impossible to have type ids overflow before
method ids.
Detection is still rough, only a max value is evaluated not the real value.
Bug 14633887
(cherry picked from commit 466d7d94b8b173a323fc5616837940219a216f9d)
Change-Id: Ib5e7b4a13a69ed76b3bbc19f7d138129fcc569c8
|
| 9cda6d302721b6943b32d150830917b17bb7deda |
|
28-Feb-2014 |
Benoit Lamarche <benoitlamarche@google.com> |
Fix detection of main dex overflow in multidex
Bug: 13239620
(cherry picked from commit c00fafdb91760a4cdfbe9e5323b5165ab4c01b76)
Change-Id: I1bdeee3b0279bed38e230ed3598c06227fc27477
|
| 466d7d94b8b173a323fc5616837940219a216f9d |
|
13-May-2014 |
Yohann Roussel <yroussel@google.com> |
Make a better detection of index limit in multidex.
- Intern annotation content sooner.
- Count declared methods separatly because it seems possible to have them
generate more method ids than they account in the constant pool if their
name and descriptor are the same as those of called methods.
- Take into account some possible ids that can be added during dex creation.
- Drop evaluation of type ids numbers because it was wrong, it's not simple
to do better and it's nearly impossible to have type ids overflow before
method ids.
Detection is still rough, only a max value is evaluated not the real value.
Bug 14633887
Change-Id: I2e3180662cb4b1c58bbc743b69270ffaa64f3106
|
| cecacd3a4b5ab7a45340f3843522a42a0a4cf287 |
|
18-Apr-2014 |
Benoit Lamarche <benoitlamarche@google.com> |
Return the correct error code in multi-thread mode
Bug: 14062905
In "--num-threads" mode, dx did not return a result code different
than zero in case of a Throwable.
This CL aims to have the same behaviour in multi-threaded mode as in
normal mode.
There are 2 cases:
- When there is an Exception, it is supposed to be handled by
ClassPathOpener.Consumer.onException(). The number of errors is then
incremented, which is used in "com.android.dx.command.dexer.Main" to
return the error code. With "--num-threads" it couldn't be handled by
onException(). So I've modified the ParallelProcessor to use the
ClassPathOpener, so that the Exception is handled inside the thread.
- When there is an Error it is handled using the Futures of
ParallelProcessors through ExecutorService.
Change-Id: Idc9d8d8bc737587ca518f8f9166e5e511ec22908
|
| c00fafdb91760a4cdfbe9e5323b5165ab4c01b76 |
|
28-Feb-2014 |
Benoit Lamarche <benoitlamarche@google.com> |
Fix detection of main dex overflow in multidex
Bug: 13239620
Change-Id: Ie897070cbbe0605f38fd6db7b6a50c91f2d1364b
|
| 3530d30f6adfd8b8607d6a0bf7988833945f5cf4 |
|
17-Jan-2014 |
Jean-Marie Henaff <jmhenaff@google.com> |
Remove code related to extended-opcode.
- This support was removed one year ago.
(cherry-pick from commit 18b65cbb3ec6634618bd3240a692507432e634b5)
Change-Id: I1154e518dc8a15220f2fcb163056570030492bb0
|
| 0820aa265488e7104ad43dc0584c01f0bfb01c7a |
|
16-Jan-2014 |
Jean-Marie Henaff <jmhenaff@google.com> |
Fix dx test 034.
- Test 034 requires that a dex file is dumped in a human readable form
even if the dex file does not contain ClassDef.
Consequently, the command line behavior is modified to satisfy this requirement.
(cherry-pick from commit 64f4aad1e6ef58b29f860b59512ca1a676ac6eb8)
Change-Id: Id64b563745db2b22ec98fba9be25413192c07719
|
| 43ee5e97c443949aa68c529e249ed6e1e5057de5 |
|
22-Oct-2013 |
Benoit Lamarche <benoitlamarche@google.com> |
Deactivate multi-thread in multi-dex mode
It should already have been the case, but it wasn't and it crashed.
Change-Id: Icb461d6e43ef2d113f59038d8c2327f3cfaff86c
(cherry picked from commit cf069ca52f2af9f70fcbeac99443f00401d43b4a)
|
| cf069ca52f2af9f70fcbeac99443f00401d43b4a |
|
22-Oct-2013 |
Benoit Lamarche <benoitlamarche@google.com> |
Deactivate multi-thread in multi-dex mode
It should already have been the case, but it wasn't and it crashed.
Change-Id: Icb461d6e43ef2d113f59038d8c2327f3cfaff86c
|
| 418c84770acea60b3b13aa37ac21563dba4c8854 |
|
22-Aug-2013 |
Yohann Roussel <yroussel@google.com> |
Make a more verbose too many id error message.
This restores the old message, replacing "ids" by "references" and
advertising for multidex otpions.
Updates dalvik/tests/089-many-methods accordingly.
(cherry picked from commit d352de04f03b848e0246119344ea13e7233018f0)
Change-Id: I5934bc9fb7812b66c0e38935946cacdf1c35a402
|
| 3abdb48fdefcc3136a8d7f9c168a305d9be9690e |
|
21-Aug-2013 |
Yohann Roussel <yroussel@google.com> |
Fix 2 filtering problems in multi dex + main list.
- Classes in archives were added in all dexes.
- Adding resources could be done twice in non "strictNameCheck" mode
(when using BestEffortMainDexListFilter) if there was 2 secondary
dex.
(cherry picked from commit ad27e19691c6723a801b127f8802f5cdf4f44359)
Change-Id: I5cbc1b273518e868fdd3d642d50488ddcbec2af6
|
| 4c656e4ec2f5c5036dc67fb4034c1e7ff7abf343 |
|
31-Jul-2013 |
Benoit Lamarche <benoitlamarche@google.com> |
Introduce --multi-dex option in dx.
Add --multi-dex options to dx command line to allow the generation of
several dex files when method index limit is about to be reached.
Also add
--main-dex-list allowing to force some classes in the main dex.
--minimal-main-dex to keep in main edx only classes specified by
main-dex-list.
--set-max-idx-number to set an arbitrary idx limit for the
splitting.
(cherry picked from commit c7daf656da3a4854296b6a8bb702e3ee418450e5)
Change-Id: I2b42272be91484a75783eb94cd30581159948975
|
| d352de04f03b848e0246119344ea13e7233018f0 |
|
22-Aug-2013 |
Yohann Roussel <yroussel@google.com> |
Make a more verbose too many id error message.
This restores the old message, replacing "ids" by "references" and
advertising for multidex otpions.
Updates dalvik/tests/089-many-methods accordingly.
Change-Id: Ibc4ad1c4606dba59b515b793d729e4e982709699
|
| ad27e19691c6723a801b127f8802f5cdf4f44359 |
|
21-Aug-2013 |
Yohann Roussel <yroussel@google.com> |
Fix 2 filtering problems in multi dex + main list.
- Classes in archives were added in all dexes.
- Adding resources could be done twice in non "strictNameCheck" mode
(when using BestEffortMainDexListFilter) if there was 2 secondary
dex.
Change-Id: Id7ecca9965a3adefc93f09d95235befbd2b1fe00
|
| c7daf656da3a4854296b6a8bb702e3ee418450e5 |
|
31-Jul-2013 |
Benoit Lamarche <benoitlamarche@google.com> |
Introduce --multi-dex option in dx.
Add --multi-dex options to dx command line to allow the generation of
several dex files when method index limit is about to be reached.
Also add
--main-dex-list allowing to force some classes in the main dex.
--minimal-main-dex to keep in main edx only classes specified by
main-dex-list.
--set-max-idx-number to set an arbitrary idx limit for the
splitting.
Change-Id: I57d0860b6156e53355e08151b64c1320b7984004
|
| 0b44e47c51b5c735283cc99b0c08cac82dba8a00 |
|
26-Apr-2013 |
delphinemartin <delphinemartin@google.com> |
Report problems as errors in dx (with exit codes)
(cherry picked from commit 593a7233e13d6eeb8ae5d894e29c6a27cbc7b7f0)
Change-Id: I6421f95ceb97a596a19781e46cb4767bfb21cdbd
|
| fe107fb6e3f308ac5174ebdc5a794ee880c741d9 |
|
17-Sep-2011 |
Jesse Wilson <jessewilson@google.com> |
Mechanical refactoring of dx into two parts.
This splits off a new package, com.android.dex that contains
code for parsing and modelling dex files. This code is usable
both at build time (for compilation and analysis) and at runtime
(for introspection).
The original package, com.android.dx is the tool that compiles
.class files into .dex files. That package also includes utilities
for merging and querying dex files, all visible to the dx command
line application.
In a follow up change I'll move the new com.android.dex package
into the libcore/ project, and configure build rules so that it's
included in both dx.jar and core.jar. The core.jar will then be
able to use it for annotation processing.
Change-Id: I3c58cf87e728e4dda1925a0992c1ee7b8130e81a
(cherry picked from commit bab4abb07335d162ecdb8091fc395cf84803a580)
Conflicts:
dx/junit-tests/com/android/dx/util/BitIntSetTest.java
dx/junit-tests/com/android/dx/util/BitsTest.java
dx/junit-tests/com/android/dx/util/IntListTest.java
dx/junit-tests/com/android/dx/util/ListIntSetTest.java
dx/src/com/android/dex/Dex.java
dx/src/com/android/dex/TableOfContents.java
dx/src/com/android/dx/annotations/AnnotationAccess.java
dx/src/com/android/dx/command/Main.java
dx/src/com/android/dx/dex/code/OutputFinisher.java
dx/src/com/android/dx/dex/code/form/Form32s.java
dx/src/com/android/dx/dex/code/form/Form33x.java
dx/src/com/android/dx/dex/code/form/Form41c.java
dx/src/com/android/dx/dex/code/form/Form52c.java
dx/src/com/android/dx/gen/DexGenerator.java
dx/src/com/android/dx/merge/DexMerger.java
|
| f870f2dce9300c8dec620613371f08e5c234245b |
|
16-Oct-2012 |
jeffhao <jeffhao@google.com> |
Add dx option to always generate const-string/jumbo.do not merge.
This allows large dex files with many strings to be merged properly.
(cherry picked from commit 266f45ff7da18022faf5f77df76c69f8cdad313f)
Change-Id: I5fe4c55d84a91101a4f89f590117aa6dc0bfc0f2
|
| 9db54b1e21b8e994658890328e57caa822e444a7 |
|
16-Oct-2012 |
jeffhao <jeffhao@google.com> |
Add dx option to always generate const-string/jumbo. do not merge.
This allows large dex files with many strings to be merged properly.
(cherry picked from commit 266f45ff7da18022faf5f77df76c69f8cdad313f)
Change-Id: I0277b1a0a81b3a77092a9e9f8f648ff190e29723
|
| 69dd411c7ef47e3ea3f6ff8be9a4d903f545f0d5 |
|
16-Oct-2012 |
jeffhao <jeffhao@google.com> |
Add dx option to always generate const-string/jumbo. do not merge.
This allows large dex files with many strings to be merged properly.
(cherry picked from commit 266f45ff7da18022faf5f77df76c69f8cdad313f)
Change-Id: I5e85346610c4c5280ecdd9e97c63eaef4420642e
|
| 266f45ff7da18022faf5f77df76c69f8cdad313f |
|
16-Oct-2012 |
jeffhao <jeffhao@google.com> |
Add dx option to always generate const-string/jumbo.
This allows large dex files with many strings to be merged properly.
Change-Id: I2c32b03ddb42735d16bc6155c1a1096159fb15ed
|
| 45ffaef820370a176097e2987a27e9169d1f8869 |
|
12-Jun-2012 |
Xavier Ducrohet <xav@android.com> |
Make dex merging stateless. do not merge.
This is required to run dx in Eclipse which doesn't unload the dx
library after each run.
(cherry-pick of 7aa5ce7e990dc3766eba97cd0932b62e4de21503.)
(cherry picked from commit 5e90165d5918f18aec0964bc475f4064e9143599)
Change-Id: Id5818f516539d13423b7cef4b2b1af30a9686b45
|
| 5e90165d5918f18aec0964bc475f4064e9143599 |
|
12-Jun-2012 |
Xavier Ducrohet <xav@android.com> |
Make dex merging stateless.
This is required to run dx in Eclipse which doesn't unload the dx
library after each run.
(cherry-pick of 7aa5ce7e990dc3766eba97cd0932b62e4de21503.)
Change-Id: I7a69a3015d448ddd5558c307cd01346156cbc739
|
| 7aa5ce7e990dc3766eba97cd0932b62e4de21503 |
|
12-Jun-2012 |
Xavier Ducrohet <xav@android.com> |
Make dex merging stateless.
This is required to run dx in Eclipse which doesn't unload the dx
library after each run.
Change-Id: I84df49d424ea465f70613eb602d1a4890804296a
|
| 36db01eeaa04e3af3401c5bd6a399d3fb7b040e3 |
|
23-Aug-2011 |
jeffhao <jeffhao@google.com> |
Removed parsing and output for unused --target-api switch in dx.
This is cherry-picked from dalvik-dev.
Change-Id: I74627e4c261e086e8c2c7cf38591cd3e9f5707e6
|
| 1e4c4bebc1feb4b68155e9c2e7e6f2c056ef8e3b |
|
12-Jul-2011 |
Raphael Moll <ralf@android.com> |
Dex for Windows: Support paths with spaces.
1- Change the format of the command line parsed by dx.jar.
Parameters with values are usually specified with one
single argument e.g.
--output=/some/path with spaces
That breaks for a variety of reasons on windows, so the first
step is to allow a new 2-parameter form:
--output "/some/path with spaces"
2- The next step is fixing the way the dx.bat wrapper
extracts the -J parameters to be able to cope with
arguments that contain spaces.
SDK bugs: 4567, 18388
Change-Id: Ia837e7307998c0bf41ae39abb36abebfc27da104
|
| a7ab2182266fb9ea1e4e8d1347fc6d6da35760d4 |
|
16-Jun-2011 |
Elliott Hughes <enh@google.com> |
Fix unbundled apps on pre-ICS devices.
Change-Id: I4aa58b5395423ec2f05be3dcb1783d0e35eabb2c
|
| 7baeded82e40352ebfb43b716e0b1dd752e72e7e |
|
15-Jun-2011 |
jeffhao <jeffhao@google.com> |
Dx defaults to emitting dexfiles without jumbo opcodes.
There's compatibility issues with going back and forth between versions,
and it's best to have dx default to a stable point and allow
optimizations to be controlled by switches. For jumbo opcodes to be
enabled, dx can be given the switch "--target-api=14".
Change-Id: I824092c795638ca329325b14c435bc41c1ab56d5
|
| 9fdbd91288a237eb58e18e4de9c729c3c268c318 |
|
25-May-2011 |
Dan Bornstein <danfuzz@android.com> |
Update dex file magic number.
Even though the dex format was technically resilient with respect to
the addition of new opcodes, consensus is that the errors one sees
when trying to use a new dex file on an old build were sufficiently
inscrutable that it was worth the effort to update the version number
embedded in the dex format magic.
This change updates dx to produce the new version number when extended
opcodes are enabled (which is the default, but may be overridden by
targeting an older API level).
This also updates the vm to recognize and accept both the new current
version number as well as the immediately previous one. Note: It won't
reject an old-version file if it happens to use the new opcodes; that
would just be a gratuitous and pointless failure.
Bug: 4364986
Change-Id: If8febbb0b91c1719df4247bf69c511251362d91f
|
| 333201833d506a3accdeac6ceb7caba8d4b95797 |
|
13-Apr-2011 |
Jesse Wilson <jessewilson@google.com> |
Combine CstUtf8 and CstString.
The only benefit we were seeing is that one wrapped its toHuman
in quotes, the other didn't.
It was far too easy to use the wrong one. We had code defending
against that in NameValuePair and CstArray.
Change-Id: Ib2e6a1596b97decced37952d46e1831b7bcd0d5d
|
| afe002d1474d7c2ef2ac80de32f21f1949af2a41 |
|
07-Apr-2011 |
Dan Bornstein <danfuzz@android.com> |
Update the interpretation of --target-api.
Change-Id: I4ab158384d09470d84097d944ae1ddadeb2ad045
|
| 09dc1b7301e8b03fc6772b022d21afe9cb077d89 |
|
22-Mar-2011 |
Dan Bornstein <danfuzz@android.com> |
Make sure both options objects get created.
Previously, the new DexOptions object would only get created when
calling dx as a command directly, but not when it was used as a
library (via a call to Main.run()).
Change-Id: Ie4f37ff9140539a4f2c7af50228ac6eaf6bebe05
|
| 09d308b04b7c4fd3ed83a8f8f4c07be67a25478c |
|
17-Mar-2011 |
Jesse Wilson <jessewilson@google.com> |
Add support for pre-dexed libraries to dx.
http://b/4090053
Change-Id: Ifb8b9230939cb2a72a767128f79cbfbd042e95a6
|
| 3dfda9ad1964510e4a7948a240b30cd710e86341 |
|
17-Mar-2011 |
Dan Bornstein <danfuzz@android.com> |
Add --target-api=N option to dx.
This change adds the option and plumbs it into where it
needs to go, but doesn't add any code to take action on it.
That will come in a follow-up.
Bug: 4094709
Change-Id: I9c796e176e125b0bcee18af56d9e6da802dfa081
|
| d2b3f499be6f5b1d4076b1244be3d1501d178315 |
|
08-Mar-2011 |
Jesse Wilson <jessewilson@google.com> |
Permit dx to build resource-only dex files.
Previously it would build a .jar file containing a classes.dex
that itself contained no classes.
Change-Id: I9c36a5dcdd17d2821be15f87656d72ef0048c113
|
| dc520adfdcf0f0e9190cdb90605c42a7cc8fa98f |
|
07-Feb-2011 |
Dan Bornstein <danfuzz@android.com> |
Fix dex processing when no output file is specified.
This case is mainly exercised in test code.
Change-Id: Ifbcbec3f001cae925090796ce8ed69f6a986384c
|
| d0900d13fb7bb1706aff5205af7e2d6517335bee |
|
21-Jan-2011 |
jeffhao <jeffhao@google.com> |
Made handling of sim exceptions nicer.
Change-Id: I089512c046d90ea37da4a33433edc4ec308a628d
|
| dc86cd9edc8b80953c8b698a83cdaebf6825d798 |
|
18-Jan-2011 |
Jesse Wilson <jessewilson@google.com> |
Add an undocumented --incremental option to dx.
Change-Id: I48879b2f724e9b92c99c669803f9c8de01487327
|
| 9028bfd131e42a58255699d1471c914d58893c61 |
|
13-Dec-2010 |
jeffhao <jeffhao@google.com> |
Adding javax.sip to the dx blacklist.
Change-Id: I5be068821862ccf073cb758e60b686f46a3c97c1
|
| 590f78fc4b6e3268149c5eacb7c8b7dd33497f44 |
|
22-Oct-2010 |
jeffhao <jeffhao@google.com> |
Added multi-threaded support to processing of class files in dx.
Change-Id: Ideb54c414073a9651b21ce0697e6444fa80f146d
|
| 85dc40d2a0cb19792bf3ee6f6c57fed08eb91ea4 |
|
27-Sep-2010 |
Dan Bornstein <danfuzz@android.com> |
Take another tilt at the "--core-library" messaging windmill.
This time, I added a note about what appears to be the most common
inadvertent triggering, and I point people at JarJar as a possible
solution when the core class inclusion was intentional.
Change-Id: I2d576df0b8944882838cbb6e2b050d59434a238b
|
| de75089fb7216d19e9c22cce4dc62a49513477d3 |
|
09-Jun-2010 |
Carl Shapiro <cshapiro@google.com> |
Remove trailing whitespace.
Change-Id: I95534bb2b88eaf48f2329282041118cd034c812b
|
| 72e93344b4d1ffc71e9c832ec23de0657e5b04a5 |
|
13-Nov-2009 |
Jean-Baptiste Queru <jbq@google.com> |
eclair snapshot
|
| 55423dcd081e30c4fc27b997f127db7b00f1b981 |
|
15-May-2009 |
Dan Bornstein <danfuzz@android.com> |
Fix a bunch of little problems noticed by FindBugs, and cleaned up some
spacing issues that I happened to notice along the way.
|
| de1ab3724b284ab053322db7ff723f54d7826c96 |
|
31-Mar-2009 |
Dan Bornstein <> |
AI 143732: Tweaks, in response to public critique.
Automated import of CL 143732
|
| 291fe153ba3b64dfb0472f1f53cee59c53f5da04 |
|
30-Mar-2009 |
Dan Bornstein <> |
AI 143492: Another crank of the --core-library wheel.
Automated import of CL 143492
|
| 99409883d9c4c0ffb49b070ce307bb33a9dfe9f1 |
|
19-Mar-2009 |
The Android Open Source Project <initial-contribution@android.com> |
auto import //branches/master/...@140412
|
| f6c387128427e121477c1b32ad35cdcaa5101ba3 |
|
04-Mar-2009 |
The Android Open Source Project <initial-contribution@android.com> |
auto import from //depot/cupcake/@135843
|
| f72d5de56a522ac3be03873bdde26f23a5eeeb3c |
|
04-Mar-2009 |
The Android Open Source Project <initial-contribution@android.com> |
auto import from //depot/cupcake/@135843
|
| 2ad60cfc28e14ee8f0bb038720836a4696c478ad |
|
21-Oct-2008 |
The Android Open Source Project <initial-contribution@android.com> |
Initial Contribution
|