lp:~pali/llvm/lldb-trunk

Created by Pali on 2014-04-11 and last modified on 2019-10-16
Get this branch:
bzr branch lp:~pali/llvm/lldb-trunk

Branch merges

Related bugs

Related blueprints

Branch information

Owner:
Pali
Project:
LLVM
Status:
Development

Import details

Import Status: Reviewed

This branch is an import of the Subversion branch from http://llvm.org/svn/llvm-project/lldb/trunk.

The next import is scheduled to run in 1 hour.

Last successful import was 4 hours ago.

Import started 4 hours ago on izar and finished 4 hours ago taking 2 minutes — see the log
Import started 10 hours ago on izar and finished 10 hours ago taking 2 minutes — see the log
Import started 16 hours ago on alnitak and finished 16 hours ago taking 2 minutes — see the log
Import started 22 hours ago on izar and finished 22 hours ago taking 2 minutes — see the log
Import started on 2019-10-15 on izar and finished on 2019-10-15 taking 2 minutes — see the log
Import started on 2019-10-15 on alnitak and finished on 2019-10-15 taking 2 minutes — see the log
Import started on 2019-10-15 on izar and finished on 2019-10-15 taking 2 minutes — see the log
Import started on 2019-10-14 on izar and finished on 2019-10-14 taking 2 minutes — see the log
Import started on 2019-10-14 on alnitak and finished on 2019-10-14 taking 2 minutes — see the log
Import started on 2019-10-14 on alnitak and finished on 2019-10-14 taking 2 minutes — see the log

Recent revisions

20817. By lawrence_danna 20 hours ago

update ScriptInterpreterPython to use File, not FILE*

Summary:
ScriptInterpreterPython needs to save and restore sys.stdout and
friends when LLDB runs a python script.

It currently does this using FILE*, which is not optimal. If
whatever was in sys.stdout can not be represented as a FILE*, then
it will not be restored correctly when the script is finished.

It also means that if the debugger's own output stream is not
representable as a file, ScriptInterpreterPython will not be able
to redirect python's output correctly.

This patch updates ScriptInterpreterPython to represent files with
lldb_private::File, and to represent whatever the user had in
sys.stdout as simply a PythonObject.

This will make lldb interoperate better with other scripts or programs
that need to manipulate sys.stdout.

Reviewers: JDevlieghere, jasonmolenda, labath

Reviewed By: labath

Subscribers: lldb-commits

Tags: #lldb

Differential Revision: https://reviews.llvm.org/D68962

20816. By hhb on 2019-10-15

[lldb] Creates _liblldb symlink from cmake

Summary:
This is another attempt of D67993.

This change removed hard coded relative paths. This way we can generate correct result when get_python_lib() returns a different path, or LLDB_PYTHON_RELATIVE_PATH is specified directly.

By moving things out of python, we are also able to correctly process more cross compile situations. E.g. .pyd vs .so for Windows.

Subscribers: lldb-commits, mgorny

Tags: #lldb

Differential Revision: https://reviews.llvm.org/D68858

20815. By lawrence_danna on 2019-10-15

remove FILE* usage from SBStream.i

Summary:
This patch removes FILE* and replaces it with SBFile and FileSP the
SWIG interface for `SBStream.i`. And this is the last one. With
this change, nothing in the python API will can access a FILE* method
on the C++ side.

Reviewers: JDevlieghere, jasonmolenda, labath

Reviewed By: labath

Subscribers: lldb-commits

Tags: #lldb

Differential Revision: https://reviews.llvm.org/D68960

20814. By lawrence_danna on 2019-10-15

eliminate virtual methods from PythonDataObjects

Summary:
This patch eliminates a bunch of boilerplate from
PythonDataObjects, as well as the use of virtual methods.
In my opinion it also makes the Reset logic a lot more
clear and easy to follow. The price is yet another
template. I think it's worth it.

Reviewers: JDevlieghere, jasonmolenda, labath, zturner

Reviewed By: JDevlieghere, labath

Subscribers: lldb-commits

Tags: #lldb

Differential Revision: https://reviews.llvm.org/D68918

20813. By lawrence_danna on 2019-10-15

convert SBDebugger::***FileHandle() wrappers to native files.

Summary:
This patch converts the swig wrappers for SetInputFileHandle() and friends
to emulate the old behavior using SetInputFile().

This will clear the way for deleting the FILE* typemaps altogether.

Reviewers: JDevlieghere, jasonmolenda, labath

Reviewed By: labath

Subscribers: mehdi_amini, dexonsmith, lldb-commits

Tags: #lldb

Differential Revision: https://reviews.llvm.org/D68856

20812. By lawrence_danna on 2019-10-15

SBFile::GetFile: convert SBFile back into python native files.

Summary:
This makes SBFile::GetFile public and adds a SWIG typemap to convert
the result back into a python native file.

If the underlying File itself came from a python file, it is returned
identically. Otherwise a new python file object is created using
the file descriptor.

Reviewers: JDevlieghere, jasonmolenda, labath

Reviewed By: labath

Subscribers: lldb-commits

Tags: #lldb

Differential Revision: https://reviews.llvm.org/D68737

20811. By adrian on 2019-10-15

Increase gdbremote_testcase timeouts when running under ASAN.

20810. By adrian on 2019-10-15

Skip PExpect tests under ASAN, I can't get them to work reliably.

20809. By aleksandr.urakov on 2019-10-15

[Windows][NFC] Fix tests after r374528.

Differential Revision: https://reviews.llvm.org/D67347

20808. By mstorsjo on 2019-10-15

[LLDB] [PECOFF] Use a "pc" vendor name in aarch64 triples

This matches all other architectures listed in the same file.

This fixes debugging aarch64 executables with lldb-server, which
otherwise fails, with log messages like these:

Target::SetArchitecture changing architecture to aarch64 (aarch64-pc-windows-msvc)
Target::SetArchitecture Trying to select executable file architecture aarch64 (aarch64-pc-windows-msvc)

ArchSpec::SetArchitecture sets the vendor to llvm::Triple::PC
for any coff/win32 combination, and if this doesn't match the triple
set by the PECOFF module, things doesn't seem to work with when
using lldb-server.

Differential Revision: https://reviews.llvm.org/D68939

Branch metadata

Branch format:
Branch format 7
Repository format:
Bazaar repository format 2a (needs bzr 1.16 or later)
This branch contains Public information 
Everyone can see this information.

Subscribers

No subscribers.