[lldb-dev] [Bug 28698] New: [lldb-mi] -break-insert with -f (pending flag) requires additional parameter
https://llvm.org/bugs/show_bug.cgi?id=28698 Bug ID: 28698 Summary: [lldb-mi] -break-insert with -f (pending flag) requires additional parameter Product: lldb Version: 3.8 Hardware: Macintosh OS: MacOS X Status: NEW Severity: normal Priority: P Component: All Bugs Assignee: lldb-dev@lists.llvm.org Reporter: pierson@microsoft.com CC: llvm-b...@lists.llvm.org Classification: Unclassified -break-insert with the pending flag (-f) requires a parameter when it should not require a parameter. -f should just set it up as a future break. error: > MI: Error: Command Args. Validation failed. Args missing additional > information: f ^error,msg="Command 'break-insert'. Command Args. Validation > failed. Args missing additional information: f" Documentation https://sourceware.org/gdb/onlinedocs/gdb/GDB_002fMI-Breakpoint-Commands.html#GDB_002fMI-Breakpoint-Commands -- You are receiving this mail because: You are the assignee for the bug. ___ lldb-dev mailing list lldb-dev@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev
Re: [lldb-dev] LLDB-MI from Eclipse hangs
I am glad to hear this. We will be happy to help with any issues you run into. If you have examples for any issues you run into, it would be great to get MI text snippets that we can run through the lldb-mi program on our end, it can help us to understand and reproduce any issues you run into. > 1. -breakpoint-insert doesn't handle locations that look like absolute > paths (because of leading /) Can you send an example? > 2. Executable path has to be absolute Does MI have the notion of setting a current working directory where things can then be relative to? If so, we should definitely fix this in LLDB. Example packet of setting the working directory and then the non working executable path would be nice. Greg > On Jul 17, 2016, at 2:26 PM, Marc-Andre Laperle via lldb-dev > wrote: > > Hi, > > I am currently working on integrating LLDB-MI in Eclipse CDT so that > it works more or less out of the box in the next CDT release due in > September (setting the lldb-mi path automatically, etc). For this > specific issue, I made it so that it simply does not send the > target-async command from Eclipse. I have worked around a few other > issues and I'll look at creating bugs in LLDB's Bugzilla and sending > patches to LLDB in the upcoming weeks. > > Here are some issues I worked around but would like to fix on the LLDB side: > 1. -breakpoint-insert doesn't handle locations that look like absolute > paths (because of leading /) > 2. Executable path has to be absolute > > This Eclipse integration is still very much in progress but in case > you are curious, the code is here: > https://github.com/MarkZ3/Eclipse-CDT-WIP/tree/lldb-mi > > Regards, > Marc-Andre > ___ > lldb-dev mailing list > lldb-dev@lists.llvm.org > http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev ___ lldb-dev mailing list lldb-dev@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev
[lldb-dev] [Bug 28702] New: LLDB-MI: pending break point set with command break-insert -f doesn't get resolved
https://llvm.org/bugs/show_bug.cgi?id=28702
Bug ID: 28702
Summary: LLDB-MI: pending break point set with command
break-insert -f doesn't get resolved
Product: lldb
Version: unspecified
Hardware: PC
OS: Linux
Status: NEW
Severity: normal
Priority: P
Component: All Bugs
Assignee: lldb-dev@lists.llvm.org
Reporter: a...@unlimitedcodeworks.xyz
CC: llvm-b...@lists.llvm.org
Classification: Unclassified
Pending break point set with command break-insert -f doesn't get resolved when
the library loads.
Step to reproduce:
1. start lldb-mi without loading a file
2. -break-insert -f main
3. -file-exec-and-symbols echo
4. -exec-run
Actual Result:
Program doesn't stop at the break point and exits normally with output like
*stopped,reason="exited-normally"
Expected result:
Program stops at the break point, with correct notification output like
*stopped,reason="breakpoint-hit",disp="del",bkptno="1",frame={level="0",addr="0x77a58650",func="__libc_start_main",args=[],file="??",fullname="??",line="-1"},thread-id="1",stopped-threads="all"
--
You are receiving this mail because:
You are the assignee for the bug.
___
lldb-dev mailing list
lldb-dev@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev
[lldb-dev] [Bug 28703] New: LLDB-MI: break-insert command flag -d (disabled) has no effect when combined with -f (pending)
https://llvm.org/bugs/show_bug.cgi?id=28703 Bug ID: 28703 Summary: LLDB-MI: break-insert command flag -d (disabled) has no effect when combined with -f (pending) Product: lldb Version: unspecified Hardware: PC OS: Linux Status: NEW Severity: normal Priority: P Component: All Bugs Assignee: lldb-dev@lists.llvm.org Reporter: a...@unlimitedcodeworks.xyz CC: llvm-b...@lists.llvm.org Classification: Unclassified break-insert command flag -d (disabled) has no effect when combined with -f (pending) These two flags should be able to work together. Specification: https://sourceware.org/gdb/onlinedocs/gdb/GDB_002fMI-Breakpoint-Commands.html#GDB_002fMI-Breakpoint-Commands Step to reproduce: 1. start lldb-mi 2. -file-exec-and-symbols 3. -break-insert -d -f main 4. -exec-run Actual result: Program stops at the breakpoint. Expected result: Program runs and exits normally because the breakpoint is disabled. -- You are receiving this mail because: You are the assignee for the bug. ___ lldb-dev mailing list lldb-dev@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev
[lldb-dev] [Bug 28709] New: lldb-mi: break-insert not working when using absolute paths
https://llvm.org/bugs/show_bug.cgi?id=28709
Bug ID: 28709
Summary: lldb-mi: break-insert not working when using absolute
paths
Product: lldb
Version: unspecified
Hardware: All
OS: All
Status: NEW
Severity: normal
Priority: P
Component: All Bugs
Assignee: lldb-dev@lists.llvm.org
Reporter: malape...@gmail.com
CC: llvm-b...@lists.llvm.org
Classification: Unclassified
Using lldb-mi from HEAD
Mac OS X and Linux
When specifying breakpoints with -break-insert, if an absolute path is used,
lldb does not stop at the breakpoint. It seems that the breakpoint was not
really inserted even if the response ^done was issued.
Here is an example code with steps.
main.cpp
#include
int main() {
std::cout << "hello world!" << std::endl;
}
1. Compile this program with debugging symbols (clang++ -g main.cpp)
2. Start lldb-mi (lldb-mi ./a.out)
3. Add a breakpoint on line 4 with
-break-insert /absolute/path/to/main.cpp:4
4. Run program (r)
The program exits normally (*stopped,reason="exited-normally")
If you try again with
-break-insert main.cpp:4
Then it works correctly and it stops at the breakpoint
(*stopped,reason="breakpoint-hit")
--
You are receiving this mail because:
You are the assignee for the bug.
___
lldb-dev mailing list
lldb-dev@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev
Re: [lldb-dev] LLDB-MI from Eclipse hangs
On Mon, Jul 25, 2016 at 4:46 PM, Greg Clayton wrote: > I am glad to hear this. We will be happy to help with any issues you run into. Thanks! > If you have examples for any issues you run into, it would be great to get MI text snippets that we can run through the lldb-mi program on our end, it can help us to understand and reproduce any issues you run into. > >> 1. -breakpoint-insert doesn't handle locations that look like absolute >> paths (because of leading /) > > Can you send an example? I opened this bug: https://llvm.org/bugs/show_bug.cgi?id=28709 > >> 2. Executable path has to be absolute > > Does MI have the notion of setting a current working directory where things can then be relative to? If so, we should definitely fix this in LLDB. Example packet of setting the working directory and then the non working executable path would be nice. This seems to work now on Mac. I'll have to double check on Linux. About the notion in MI, the command is -environment-cd I think. Regards, Marc-Andre > > Greg > > >> On Jul 17, 2016, at 2:26 PM, Marc-Andre Laperle via lldb-dev < lldb-dev@lists.llvm.org> wrote: >> >> Hi, >> >> I am currently working on integrating LLDB-MI in Eclipse CDT so that >> it works more or less out of the box in the next CDT release due in >> September (setting the lldb-mi path automatically, etc). For this >> specific issue, I made it so that it simply does not send the >> target-async command from Eclipse. I have worked around a few other >> issues and I'll look at creating bugs in LLDB's Bugzilla and sending >> patches to LLDB in the upcoming weeks. >> >> Here are some issues I worked around but would like to fix on the LLDB side: >> 1. -breakpoint-insert doesn't handle locations that look like absolute >> paths (because of leading /) >> 2. Executable path has to be absolute >> >> This Eclipse integration is still very much in progress but in case >> you are curious, the code is here: >> https://github.com/MarkZ3/Eclipse-CDT-WIP/tree/lldb-mi >> >> Regards, >> Marc-Andre >> ___ >> lldb-dev mailing list >> lldb-dev@lists.llvm.org >> http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev > ___ lldb-dev mailing list lldb-dev@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev
