Author: jingham
Date: Fri Sep 15 10:54:37 2017
New Revision: 313371
URL: http://llvm.org/viewvc/llvm-project?rev=313371&view=rev
Log:
Remove a couple of warnings pointed out by Ted Woodward.
Modified:
lldb/trunk/source/API/SBBreakpointName.cpp
Modified: lldb/trunk/source/API/SBBreakpointName
lemo created this revision.
Herald added a subscriber: ki.stfu.
The core of this change is the new CommandInterpreter::m_command_state, which
models the state transitions for interactive commands, including an
"interrupted" state transition.
In general, command interruption requires cooperation
amccarth added a comment.
I haven't looked at the whole patch yet, but it seems the SIGINT fix is well
isolated. That should probably be in a separate patch.
https://reviews.llvm.org/D37923
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
lemo created this revision.
Herald added a subscriber: ki.stfu.
1. Fix a data race (g_interrupt_sent flag usage was not thread safe, signals
can be handled on arbitrary threads)
2. exit() is not signal-safe, replaced it with the signal-safe equivalent
_exit()
https://reviews.llvm.org/D37926
F
lemo updated this revision to Diff 115476.
lemo added a comment.
Split the SIGINT handles fixes into a stanalone patch
https://reviews.llvm.org/D37923
Files:
include/lldb/Core/IOHandler.h
include/lldb/Interpreter/CommandInterpreter.h
source/Commands/CommandObjectTarget.cpp
source/Interp
zturner added a comment.
It turns out the function this called, `DispatchInputInterrupt`, already
acquires a mutex. Maybe put the synchronization in there? Then you don't have
to reproduce the synchronization in both MI and LLDB
https://reviews.llvm.org/D37926
How would that fix look? If the intention is to ignore nested SIGINT then
doing it directly in the handler seems cleaner and safer (re. safety, note
that m_input_reader_stack.GetMutex() is a std::*recursive_*mutex so
depending on it in signal handlers is a big no-no. The recursive_mutex
looks like
amccarth added inline comments.
Comment at: source/Commands/CommandObjectTarget.cpp:2058
+ break;
+}
num_dumped++;
Many LLVM developers prefer to omit the braces when the body of the
control-flow statement is a single statem
clayborg added a comment.
Looks fine to me.
https://reviews.llvm.org/D37926
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
fjricci created this revision.
This allows for the stack size to be configured, which isn't
possible with std::thread. Prevents overflowing the stack when
performing complex operations in the task pool on darwin,
where the default pthread stack size is only 512kb.
https://reviews.llvm.org/D37930
lemo added inline comments.
Comment at: source/Commands/CommandObjectTarget.cpp:2058
+ break;
+}
num_dumped++;
amccarth wrote:
> Many LLVM developers prefer to omit the braces when the body of the
> control-flow statement is
clayborg added inline comments.
Comment at: source/Commands/CommandObjectTarget.cpp:2056-2058
+if (m_interpreter.WasInterrupted()) {
+ break;
+}
Remove braces
Comment at: source/Commands/CommandObjectTarget.
lemo added inline comments.
Comment at: source/Interpreter/CommandInterpreter.cpp:2713
+ for (; chunk_size < size; ++chunk_size) {
+assert(data[chunk_size] != '\0');
+if (data[chunk_size] == '\n') {
amccarth wrote:
> Should we be that trustin
vadimcn created this revision.
vadimcn added a project: LLDB.
OpenOCD sends register classes as two separate nodes, fixed parser to
process both of them.
OpenOCD returns "l" in response to "qfThreadInfo", so IsUnsupportedResponse()
was false and we were ending up without any threads in the pro
Author: zturner
Date: Fri Sep 15 15:10:46 2017
New Revision: 313407
URL: http://llvm.org/viewvc/llvm-project?rev=313407&view=rev
Log:
Resubmit "[lit] Force site configs to run before source-tree configs"
This is a resubmission of r313270. It broke standalone builds of
compiler-rt because we were
clayborg added a comment.
The multiple feature fix is fine.
As for the qfThreadInfo fix, do you not have control over the OpenOCD GDB
server? I would be nice to modify it to return something valid in response to
qfThreadInfo? If you don't control or have access to modify the OpenOCD GDB
server
lemo updated this revision to Diff 115516.
lemo edited the summary of this revision.
lemo added a comment.
Incorporating CR feedback.
https://reviews.llvm.org/D37923
Files:
include/lldb/Core/IOHandler.h
include/lldb/Interpreter/CommandInterpreter.h
source/Commands/CommandObjectTarget.cpp
vadimcn added a comment.
This is what I see in the log:
< 16> send packet: $qfThreadInfo#bb
< 5> read packet: $l#6c
And here's code that generates it:
https://github.com/gnu-mcu-eclipse/openocd/blob/b21ab1d683aaee501d45fe8a509a2043123f16fd/src/rtos/rtos.c#L370
I agree, they should have
clayborg accepted this revision.
clayborg added a comment.
This revision is now accepted and ready to land.
As long as everyone agrees that no threads from qfThreadInfo means there is one
thread whose thread ID is 1 then this is ok.
Repository:
rL LLVM
https://reviews.llvm.org/D37934
Author: eugene
Date: Fri Sep 15 19:19:21 2017
New Revision: 313436
URL: http://llvm.org/viewvc/llvm-project?rev=313436&view=rev
Log:
More precise c library feature detection for Android.
Modified:
lldb/trunk/cmake/modules/LLDBGenerateConfig.cmake
lldb/trunk/source/Host/common/Socket.cpp
Author: eugene
Date: Fri Sep 15 19:58:49 2017
New Revision: 313437
URL: http://llvm.org/viewvc/llvm-project?rev=313437&view=rev
Log:
Check availability of accept4 in C++ instad of C code.
Modified:
lldb/trunk/cmake/modules/LLDBGenerateConfig.cmake
Modified: lldb/trunk/cmake/modules/LLDBGener
Author: vadimcn
Date: Fri Sep 15 20:53:13 2017
New Revision: 313442
URL: http://llvm.org/viewvc/llvm-project?rev=313442&view=rev
Log:
Fix compatibility with OpenOCD debug stub.
OpenOCD sends register classes as two separate nodes, fixed parser to
process both of them.
OpenOCD returns "l" in re
22 matches
Mail list logo
