[Lldb-commits] [PATCH] D129814: Fix stepping over watchpoints in architectures that raise the exception before executing the instruction

This revision was automatically updated to reflect the committed changes. Closed by commit rG5778ada8e54e: Make hit point counts reliable for architectures that stop before evaluation. (authored by jingham). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.o

[Lldb-commits] [PATCH] D129338: Tell the user which pathname was invalid...

This revision was automatically updated to reflect the committed changes. Closed by commit rGe83d47f6b7bd: When the module path for `command script import` is invalid, echo the path. (authored by jingham). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org

[Lldb-commits] [PATCH] D130098: [LLDB][NFC][Reliability] Fix uninitialized variables from Coverity scan

jingham requested changes to this revision. jingham added a comment. This revision now requires changes to proceed. Thanks for finding these uninitialized variables. That's a valuable exercise. Couple of comments; Most of the time you set lldb::addr_t's to 0, they probably should have been se

[Lldb-commits] [PATCH] D130098: [LLDB][NFC][Reliability] Fix uninitialized variables from Coverity scan

jingham added inline comments. Comment at: lldb/source/Plugins/Language/ObjC/NSDictionary.cpp:140 DataDescriptor_64 *m_data_64 = nullptr; - lldb::addr_t m_data_ptr; + lldb::addr_t m_data_ptr = {}; CompilerType m_pair_type; jingham wrote: > Invalid address

[Lldb-commits] [PATCH] D130098: [LLDB][NFC][Reliability] Fix uninitialized variables from Coverity scan

jingham added inline comments. Comment at: lldb/source/Commands/CommandObjectTarget.cpp:4685 +lldb::tid_t m_thread_id = LLDB_INVALID_THREAD_ID; +uint32_t m_thread_index = 0; std::string m_thread_name; clayborg wrote: > jingham wrote: > > Because of t

[Lldb-commits] [PATCH] D129078: [LLDB][ClangExpression] Allow expression evaluation from within C++ Lambdas

jingham accepted this revision. jingham added a comment. This looks good to me. I had a trivial comment about comments. You added a test for the condition in lambda as well, that's great! Comment at: lldb/include/lldb/Expression/Materializer.h:88 + /// \param[in] name Name

[Lldb-commits] [PATCH] D130524: Don't hold the StackFrame mutex while getting a ValueObject Dynamic value

jingham created this revision. jingham added reviewers: JDevlieghere, clayborg, augusto2112. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. We've noticed occasional hangs in the test TestGuiExpandThread

[Lldb-commits] [PATCH] D130524: Don't hold the StackFrame mutex while getting a ValueObject Dynamic value

This revision was automatically updated to reflect the committed changes. Closed by commit rG8b7775a472e3: StackFrame::GetValueObjectForFrameVariable holds the StackFrame lock too long. (authored by jingham). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.

[Lldb-commits] [PATCH] D130674: Accurate watchpoint hit counts redux

jingham created this revision. jingham added reviewers: labath, clayborg, jasonmolenda, JDevlieghere. Herald added a subscriber: kristof.beyls. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. This is a r

[Lldb-commits] [PATCH] D130674: Accurate watchpoint hit counts redux

jingham added a comment. Note, just so it isn't confusing, there are some other bits of the ThreadPlan machinery where a ThreadPlan or the Thread itself can say "Even though the thread plan computation says I should stop, I'm going to force a run". However, those controls were all just about t

[Lldb-commits] [PATCH] D131036: [lldb/crashlog] Add `-s|--skip-status` option to interactive mode

jingham added a comment. The code looks fine to me. This is a new capability, and I don't see how we can guess with certainty whether the clients want this output or not, so I think we should add it as an explicit option, as in the current patch. Comment at: lldb/examples/py

[Lldb-commits] [PATCH] D131159: [WIP][lldb] Use WSAEventSelect for MainLoop polling

jingham added a comment. This is a WIP, presumably in the final version there won't be prominent #ifdef _WIN32 in a file in the "Host/common" directory. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D131159/new/ https://reviews.llvm.org/D131159 __

[Lldb-commits] [PATCH] D131138: [lldb] Dynamically generate enum names in lldbutil

jingham added a comment. In D131138#3698508 , @mib wrote: > This is awesome 🤩 ! I was also thinking of changing the way enums are exposed > to python: instead of having everything added to the `lldb` python module, we > could create a class per enum and

[Lldb-commits] [PATCH] D130674: Accurate watchpoint hit counts redux

This revision was automatically updated to reflect the committed changes. Closed by commit rG0948f1cf8177: Reapply the commits to enable accurate hit-count detection for watchpoints. (authored by jingham). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org

[Lldb-commits] [PATCH] D131130: [lldb] Improve EXC_RESOURCE exception reason

jingham accepted this revision. jingham added a comment. This revision is now accepted and ready to land. LGTM CHANGES SINCE LAST ACTION https://reviews.llvm.org/D131130/new/ https://reviews.llvm.org/D131130 ___ lldb-commits mailing list lldb-commi

[Lldb-commits] [PATCH] D131294: [LLDB][NFC] Reliability fixes to TCPSocket code

jingham added a comment. We don't use the constant first so you don't accidentally use "=" rather than "==" convention anywhere else in lldb, so it looks weird here. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D131294/new/ https://reviews.llvm.or

[Lldb-commits] [PATCH] D131996: Use a SmallPtrSet rather than a SmallVector in ClusterManager.

jingham created this revision. jingham added reviewers: JDevlieghere, jasonmolenda, kastiglione, labath. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. The m_objects store in ClusterManager holds pointe

[Lldb-commits] [PATCH] D131996: Use a SmallPtrSet rather than a SmallVector in ClusterManager.

jingham added a comment. In D131996#3728253 , @labath wrote: > In case you have an idea of what's the "typical" number of objects in shared > cluster, then we can tune the integer template parameter (in case we usually > have more than 16 objects, then

[Lldb-commits] [PATCH] D131996: Use a SmallPtrSet rather than a SmallVector in ClusterManager.

This revision was automatically updated to reflect the committed changes. Closed by commit rG33722848fcb5: Use a SmallPtrSet rather than a SmallVector in ClusterManager. (authored by jingham). Changed prior to commit: https://reviews.llvm.org/D131996?vs=453127&id=453685#toc Repository: rG LL

[Lldb-commits] [PATCH] D132191: Treat a UUID of all zeros consistently to mean "Invalid UUID"

jingham created this revision. jingham added reviewers: JDevlieghere, clayborg, labath. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. At present, there are two ways to get data into a UUID object, an u

[Lldb-commits] [PATCH] D132191: Treat a UUID of all zeros consistently to mean "Invalid UUID"

jingham added a comment. In D132191#3734014 , @clayborg wrote: > Minidump files have UUID values that are zeroed out. We will need to do > something for these that can sense all zeroes and return a default > constructed one. I'm confused. The MinidumP

[Lldb-commits] [PATCH] D132191: Treat a UUID of all zeros consistently to mean "Invalid UUID"

jingham added a comment. Remember, all this patch does is make a universal rule that "All zero UUID's are placeholders and not meant to be matched against." The only platforms/ObjectFile readers that weren't already following that rule were Darwin - where this was definitely just historical ac

[Lldb-commits] [PATCH] D132415: [LLDB] Add data formatter for std::coroutine_handle

jingham added inline comments. Comment at: lldb/test/API/functionalities/data-formatter/data-formatter-stl/generic/coroutine_handle/TestCoroutineHandle.py:15 + +class TestCoroutineHandle(TestBase): +def do_test(self, stdlib_type): aprantl wrote: > Nice test!

[Lldb-commits] [PATCH] D132415: [LLDB] Add data formatter for std::coroutine_handle

jingham added inline comments. Comment at: lldb/test/API/functionalities/data-formatter/data-formatter-stl/generic/coroutine_handle/TestCoroutineHandle.py:44 + +lldbutil.run_to_source_breakpoint(self, '// Break after co_yield', +lldb.SBFileSpec("main.cpp"

[Lldb-commits] [PATCH] D132397: [LLDB] Clean up after command fails

jingham added a comment. Each new command invocation is expected to prime the ExecutionContext before invoking the command. I don't think there's any situation where one command should use a previous command's execution context on purpose. So if we're getting to Execute without the CommandInt

[Lldb-commits] [PATCH] D132397: [LLDB] Clean up after command fails

jingham accepted this revision. jingham added a comment. This revision is now accepted and ready to land. Actually, I think this is 100% correct, and corrects an error I made when I added argument number checking a little while back. The comments on CheckRequirements say: Every command should c

[Lldb-commits] [PATCH] D132598: [lldb] Add more dylib paths for exception breakpoints

jingham accepted this revision. jingham added a comment. This revision is now accepted and ready to land. LGTM too. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D132598/new/ https://reviews.llvm.org/D132598

[Lldb-commits] [PATCH] D121818: [lldb] Expose diagnostic events through the SB API

jingham added a comment. Yes, if we had had the SBStructuredData class from the start, that would have been a clearer, more robust and flexible way to do the event data in general. But at this point the event data word meanings are API so we can't really remove the old way. CHANGES SINCE LAS

[Lldb-commits] [PATCH] D121977: [lldb/test] Add events listener helper class to lldbtest

jingham added inline comments. Comment at: lldb/packages/Python/lldbsuite/test/lldbutil.py:1624 +# == +# Utility class to setup a progress event listener +# == JDevlie

[Lldb-commits] [PATCH] D122025: [lldb] Remove lldbassert from CommandInterpreter::PrintCommandOutput

jingham added a comment. I think it makes sense to require that the CommandResultObject's command output is UTF-8 string with no embedded '\0'-s. We wouldn't do smart things with rendering command output with nulls in it, and I don't see any reason to support such a thing. OTOH, this seems li

[Lldb-commits] [PATCH] D122283: [lldb] Add a Stream variant that escapes null bytes

jingham accepted this revision. jingham added a comment. This revision is now accepted and ready to land. Nice! Thanks for doing this. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D122283/new/ https://reviews.llvm.org/D122283 ___ lldb-commit

[Lldb-commits] [PATCH] D122680: Add a setting to force overwriting commands in "command script add"

jingham created this revision. jingham added a reviewer: JDevlieghere. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. When I added containers it seemed like it would be easier to accidentally overwrite

[Lldb-commits] [PATCH] D122684: [lldb] Use the selected and host platform to disambiguate between fat binary architectures

jingham added inline comments. Comment at: lldb/source/Target/Platform.cpp:1252 +if (selected_platform_sp) { + if (selected_platform_sp->IsCompatibleArchitecture( + arch, process_host_arch, false, nullptr)) { Why are you passing process_host

[Lldb-commits] [PATCH] D122684: [lldb] Use the selected and host platform to disambiguate between fat binary architectures

jingham added inline comments. Comment at: lldb/source/Target/Platform.cpp:1252 +if (selected_platform_sp) { + if (selected_platform_sp->IsCompatibleArchitecture( + arch, process_host_arch, false, nullptr)) { jingham wrote: > Why are you pas

[Lldb-commits] [PATCH] D122680: Add a setting to force overwriting commands in "command script add"

This revision was automatically updated to reflect the committed changes. Closed by commit rG1f7b58f2a504: Add a setting to not require --overwrite to overwrite commands. (authored by jingham). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D122680/new

[Lldb-commits] [PATCH] D123421: Handle the new "selector-specific stub" ObjC Method dispatch method

jingham created this revision. jingham added a reviewer: JDevlieghere. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. Clang is adding a feature to ObjC code generation, where instead of calling objc_msg

[Lldb-commits] [PATCH] D123421: Handle the new "selector-specific stub" ObjC Method dispatch method

jingham updated this revision to Diff 421638. jingham added a comment. -U on the patch. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D123421/new/ https://reviews.llvm.org/D123421 Files: lldb/source/Plugins/LanguageRuntime/ObjC/AppleObjCRunt

[Lldb-commits] [PATCH] D123421: Handle the new "selector-specific stub" ObjC Method dispatch method

jingham updated this revision to Diff 421651. jingham added a comment. Address review comments Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D123421/new/ https://reviews.llvm.org/D123421 Files: lldb/source/Plugins/LanguageRuntime/ObjC/AppleObjCR

[Lldb-commits] [PATCH] D123421: Handle the new "selector-specific stub" ObjC Method dispatch method

jingham added a comment. I also made one functional change. Some code was checking `!m_lookup_implementation_function_code.empty()` to see if we had found an implementation function. That won't work if I put the common prefix into this ivar in the constructor, so I change that to leave the mem

[Lldb-commits] [PATCH] D123421: Handle the new "selector-specific stub" ObjC Method dispatch method

jingham marked 6 inline comments as done. jingham added inline comments. Comment at: lldb/source/Plugins/LanguageRuntime/ObjC/AppleObjCRuntime/AppleObjCTrampolineHandler.cpp:1012-1015 +if (log) { + LLDB_LOG(log, "Resolving call for class - {0} and selector - {1}

[Lldb-commits] [PATCH] D104629: [lldb] Escape semicolons for all shells

jingham added a comment. This is good in general, but I'm not sure we want to escape $. One of the uses of shell expansion is to get shell variables expanded. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D104629/new/ https://reviews.llvm.org/D104

[Lldb-commits] [PATCH] D123690: [lldb] Expand $ when using tcsh

jingham accepted this revision. jingham added a comment. This revision is now accepted and ready to land. tcsh is different from the other shells in that a non-existent variable is trapped by the shell, and an error produced BEFORE running the command, whereas in other shells it just gets passed

[Lldb-commits] [PATCH] D123746: [lldb] Prevent crash when adding a stop hook with --shlib

jingham added a comment. This is fine, though the fact that it crashed when you tried to specify the shared library means we don't have any tests that the shared library gets copied over from the dummy target correctly. Might be nice to extend the test to have a "file not_called_test" and "bre

[Lldb-commits] [PATCH] D124028: Add some docs on how to use container commands to the python reference doc

jingham created this revision. jingham added a reviewer: JDevlieghere. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. I added this feature a while back but didn't put an example in the python reference

[Lldb-commits] [PATCH] D124029: Add a mutex to the ThreadPlanStackMap to prevent it getting corrupted

jingham created this revision. jingham added a reviewer: JDevlieghere. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. We were seeing very occasional crashes that looked like corrupted ThreadPlans. Th

[Lldb-commits] [PATCH] D124198: [LLDB][Unwind] Add stack scanning as fallback unwind plan if no symbol file is available.

jingham added a comment. I also wonder if this shouldn't require a separate flag to turn on this search, or be a separate command in the way Greg did. People rely on backtraces being accurate. It's fine to add something more like a desperation play - something is sometimes better than nothing

[Lldb-commits] [PATCH] D124579: Make partial function name matching match on context boundaries

jingham created this revision. jingham added reviewers: JDevlieghere, labath, clayborg, shafik. Herald added a subscriber: mgorny. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. We don't require users t

[Lldb-commits] [PATCH] D123502: [lldb][NFC] Refactor printing of short options in help

jingham accepted this revision. jingham added a comment. This revision is now accepted and ready to land. Looks like it will do the job, and this version is definitely easier to read. Were there any tests for the help output? If not, maybe we should add one. Other than that, LGTM. Repository:

[Lldb-commits] [PATCH] D123500: [lldb][NFC] Add more tests for GenerateOptionsUsage

jingham accepted this revision. jingham added a comment. This revision is now accepted and ready to land. Ah, tests for help that I asked for in another review... Excellent. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D123500/new/ https://reviews

[Lldb-commits] [PATCH] D123204: [lldb] Fix initialization of LazyBool/bool variables m_overwrite/m_overwrite_lazy. NFCI.

jingham accepted this revision. jingham added a comment. This revision is now accepted and ready to land. Thanks for these cleanups. In actual fact, the values set for options in the constructor for CommandOptions are never used, only the values in OptionParsingStarted matter. But it's still c

[Lldb-commits] [PATCH] D117383: [lldb] Expose std::pair children for unordered_map

jingham added a comment. Herald added a project: All. Sorry for the delay here. This output is definitely nicer than the previous version, so this change is desirable. If you really are switching on unordered_map vrs. unordered_set behavior, can you check the backend ValueObject's type to det

[Lldb-commits] [PATCH] D124597: Select the correct DWARF location list entry when looking for variable locations

jingham accepted this revision. jingham added a comment. This revision is now accepted and ready to land. The change seems correct. Since the test only requires -01, and the test code is quite straightforward, the test seems fine. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTIO

[Lldb-commits] [PATCH] D124029: Add a mutex to the ThreadPlanStackMap to prevent it getting corrupted

This revision was automatically updated to reflect the committed changes. Closed by commit rGdca2bc408186: Add a mutex to the ThreadPlanStackMap class. (authored by jingham). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D124029/new/ https://reviews.

[Lldb-commits] [PATCH] D124028: Add some docs on how to use container commands to the python reference doc

This revision was automatically updated to reflect the committed changes. Closed by commit rGaa7470a1b313: Add a paragraph showing how to use container commands. (authored by jingham). Changed prior to commit: https://reviews.llvm.org/D124028?vs=423691&id=426128#toc Repository: rG LLVM Githu

[Lldb-commits] [PATCH] D119997: Add a little test for simple breaking on overloaded functions by name

jingham added a comment. Herald added a project: All. This was submitted (dd8490d207d3a1612091abbea04bf660f133a89f ) but I neglected to put the "Differential Revision" tag in the commit message so automation didn't pick that u

[Lldb-commits] [PATCH] D124579: Make partial function name matching match on context boundaries

jingham updated this revision to Diff 426170. jingham added a comment. Address review comments. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D124579/new/ https://reviews.llvm.org/D124579 Files: lldb/include/lldb/Target/Language.h lldb/source/C

[Lldb-commits] [PATCH] D124579: Make partial function name matching match on context boundaries

jingham marked 6 inline comments as done. jingham added inline comments. Comment at: lldb/include/lldb/Target/Language.h:225 + // symbol was really B::A::my_function. We want that to be + // a match. But we wouldn't want this to match AnotherB::A::my_function. The + // use

[Lldb-commits] [PATCH] D124947: Allow `target create` with no local file

jingham created this revision. jingham added reviewers: JDevlieghere, labath, clayborg, jasonmolenda. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. In 92eaad2dd7adb5ee92f397cef85ab11f2612294e

[Lldb-commits] [PATCH] D124731: [lldb] Consider binary as module of last resort

jingham requested changes to this revision. jingham added a comment. This revision now requires changes to proceed. This seems like a reasonable fallback, and the implementation looks fine. You need to add a test case setting an address breakpoint in the executable and making sure that works. Sh

[Lldb-commits] [PATCH] D125148: Add an example command that runs to one of a set of breakpoints

jingham created this revision. jingham added reviewers: JDevlieghere, kastiglione. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. In a situation where you have a lot of breakpoints set, some enabled and

[Lldb-commits] [PATCH] D125325: Pass plugin_name in SBProcess::SaveCore

jingham added a comment. In D125325#3505305 , @PatriosTheGreat wrote: > Hi Greg. > > Thanks for the review. > I fixed the feedback. > I forgot the default initialization of plugin_name parameter in SBProcess.i > in previous version, but I assume it's st

[Lldb-commits] [PATCH] D125434: Make a more convenient way to allow Darwin users to ignore certain Mach Exceptions

jingham created this revision. jingham added reviewers: JDevlieghere, jasonmolenda, labath. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. For relatively uninteresting reasons, it is not possible when r

[Lldb-commits] [PATCH] D125437: [lldb/API] Add SBCompileUnit::GetIndexForLineEntry method to SB API

jingham requested changes to this revision. jingham added a comment. This revision now requires changes to proceed. This needs a doctoring for the API and some test. Comment at: lldb/source/API/SBCompileUnit.cpp:86 + uint32_t index = UINT32_MAX; + if (m_opaque_ptr && line_ent

[Lldb-commits] [PATCH] D124579: Make partial function name matching match on context boundaries

This revision was automatically updated to reflect the committed changes. jingham marked 4 inline comments as done. Closed by commit rG3339000e0bda: We don't require users to type out the full context of a function, for (authored by jingham). Repository: rG LLVM Github Monorepo CHANGES SINCE L

[Lldb-commits] [PATCH] D125437: [lldb/API] Add SBCompileUnit::GetIndexForLineEntry method to SB API

jingham accepted this revision. jingham added a comment. This revision is now accepted and ready to land. LGTM CHANGES SINCE LAST ACTION https://reviews.llvm.org/D125437/new/ https://reviews.llvm.org/D125437 ___ lldb-commits mailing list lldb-commi

[Lldb-commits] [PATCH] D125505: Cache tilde expansions to avoid redundant lookups

jingham requested changes to this revision. jingham added a comment. This revision now requires changes to proceed. I don't think the cache belongs in a global in the TildeExpressionResolver class. The derived ExpressionResolvers might very well resolve the same ~foo to different paths (e.g. if

[Lldb-commits] [PATCH] D124947: Allow `target create` with no local file

This revision was automatically updated to reflect the committed changes. Closed by commit rG2a21700bc5be: In 92eaad2dd7adb5ee92f397cef85ab11f2612294e I made it possible to run (authored by jingham). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D1249

[Lldb-commits] [PATCH] D108228: Fix error handling in the libcxx std::string formatter

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG7afd257ff8a4: Fix the std::string formatter to report errors in the case where the (authored by jingham). Herald added a project: All. Changed prior

[Lldb-commits] [PATCH] D125434: Make a more convenient way to allow Darwin users to ignore certain Mach Exceptions

jingham added a comment. Addressed review comments. Comment at: lldb/source/Plugins/Platform/MacOSX/PlatformDarwin.cpp:80-111 +// Static Variables +static uint32_t g_initialize_count = 0; + +void PlatformDarwin::Initialize() { + Platform::Initialize(); + + if (g_initialize_co

[Lldb-commits] [PATCH] D125434: Make a more convenient way to allow Darwin users to ignore certain Mach Exceptions

jingham updated this revision to Diff 430435. jingham added a comment. Address review comments Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D125434/new/ https://reviews.llvm.org/D125434 Files: lldb/include/lldb/Interpreter/OptionValueString.h

[Lldb-commits] [PATCH] D125434: Make a more convenient way to allow Darwin users to ignore certain Mach Exceptions

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGbff4673b4178: Add a darwin platform setting to specify which exceptions debugserver (authored by jingham). Repository: rG LLVM Github Monorepo CH

[Lldb-commits] [PATCH] D126057: [lldb] Fix that empty target.run-args are not actually used when launching process

jingham accepted this revision. jingham added a comment. GetPropertyAtIndexAsArg isn't actually documented, so we don't know what the original intent for the return value was. But in almost all the cases where we look up a property using this function, when we use the enum directly we discard

[Lldb-commits] [PATCH] D126259: Add the ability to specify signal actions (pass, stop or notify) for a signal before a process is created

jingham created this revision. jingham added reviewers: JDevlieghere, labath, chaoren, clayborg, kastiglione. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. One of the weaknesses of the signal handling

[Lldb-commits] [PATCH] D126259: Add the ability to specify signal actions (pass, stop or notify) for a signal before a process is created

jingham marked 3 inline comments as done. jingham added a comment. This patch treats the Signal structure held in the target differently from UnixSignals::Signal. The latter always has to have values for all three actions and needs to have a signal number. The former doesn't actually claim to

[Lldb-commits] [PATCH] D126259: Add the ability to specify signal actions (pass, stop or notify) for a signal before a process is created

jingham marked 3 inline comments as done. jingham added a comment. In D126259#3534729 , @clayborg wrote: > So the "process handle" command allows us to set signals by signal number as > well. Does this patch support this? It seems like it wouldn't be too

[Lldb-commits] [PATCH] D126259: Add the ability to specify signal actions (pass, stop or notify) for a signal before a process is created

jingham updated this revision to Diff 431745. jingham added a comment. Don't allow setting signal actions by signal number before you have a process. We don't know what signal 20 is going to end up being till we have a process, so allowing this by number doesn't make sense. Repository: rG LL

[Lldb-commits] [PATCH] D126259: Add the ability to specify signal actions (pass, stop or notify) for a signal before a process is created

jingham added a comment. In D126259#3534944 , @clayborg wrote: > In D126259#3534919 , @jingham wrote: > >> In D126259#3534729 , @clayborg >> wrote: >> >>> So the "process

[Lldb-commits] [PATCH] D126259: Add the ability to specify signal actions (pass, stop or notify) for a signal before a process is created

jingham added a comment. In D126259#3535004 , @clayborg wrote: > In D126259#3534997 , @jingham wrote: > >> Don't allow setting signal actions by signal number before you have a >> process. > > I understand > >> W

[Lldb-commits] [PATCH] D125148: Add an example command that runs to one of a set of breakpoints

jingham updated this revision to Diff 431837. jingham added a comment. Added support for breakpoint names as well as ID's. Addressed other review comments. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D125148/new/ https://reviews.llvm.org/D125148

[Lldb-commits] [PATCH] D125148: Add an example command that runs to one of a set of breakpoints

jingham marked 7 inline comments as done. jingham added a comment. The patch gets a little hard to read with all the no-longer in the right places comments, but I think I addressed everything. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D125148/ne

[Lldb-commits] [PATCH] D126259: Add the ability to specify signal actions (pass, stop or notify) for a signal before a process is created

jingham updated this revision to Diff 432145. jingham added a comment. Address Jonas' review comments Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D126259/new/ https://reviews.llvm.org/D126259 Files: lldb/include/lldb/Target/Target.h lldb/incl

[Lldb-commits] [PATCH] D126259: Add the ability to specify signal actions (pass, stop or notify) for a signal before a process is created

jingham added inline comments. Comment at: lldb/include/lldb/Target/Target.h:1451 + /// Print all the signals set in this target. + void PrintDummySignals(Stream &strm, Args signals, size_t num_signals); + JDevlieghere wrote: > Was `Args` supposed to be a refer

[Lldb-commits] [PATCH] D68671: Add the ability to pass extra args to a Python breakpoint command function

jingham added a comment. Herald added a project: All. Just for history's sake, this change was committed (738af7a6241c9). I put in the Differential Revision line in the changelog, so I'm not sure why this didn't get marked as closed... Repository: rLLDB LLDB CHANGES SINCE LAST ACTION htt

[Lldb-commits] [PATCH] D126259: Add the ability to specify signal actions (pass, stop or notify) for a signal before a process is created

jingham added inline comments. Comment at: lldb/source/Commands/CommandObjectProcess.cpp:1477-1479 +bool only_target_values; +bool do_clear; +bool dummy; JDevlieghere wrote: > jingham wrote: > > JDevlieghere wrote: > > > Let's initialize these to the

[Lldb-commits] [PATCH] D126259: Add the ability to specify signal actions (pass, stop or notify) for a signal before a process is created

This revision was automatically updated to reflect the committed changes. Closed by commit rG134d7f9a4b97: Store a by name list of signals with their actions in the Target (authored by jingham). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D126259/ne

[Lldb-commits] [PATCH] D126513: Add -b (--continue-to-breakpoint) option to the "process continue" command

jingham created this revision. jingham added reviewers: JDevlieghere, kastiglione. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. Sometimes you have a complex set of breakpoints, some enabled and some n

[Lldb-commits] [PATCH] D126730: Fix a bug where `break com add -s py -o "some_python" BKPT_NAME` only added the command to the first breakpoint

jingham created this revision. jingham added reviewers: kastiglione, JDevlieghere. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. This was a "probably should have copy-pasted" error. A "!" was omitted

[Lldb-commits] [PATCH] D126730: Fix a bug where `break com add -s py -o "some_python" BKPT_NAME` only added the command to the first breakpoint

This revision was automatically updated to reflect the committed changes. Closed by commit rGd92f7f790c8e: Fix a copy-paste error in "br com add -s py -o 'some_python' BKPT_NAME" (authored by jingham). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D12

[Lldb-commits] [PATCH] D126789: Stop regex commands from double entry into the history

jingham created this revision. jingham added reviewers: kastiglione, clayborg. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. At present, if you run a regex command (e.g. the `b` command) you will get t

[Lldb-commits] [PATCH] D126789: Stop regex commands from double entry into the history

jingham added a comment. Note, the original code tries to finesse this by passing eLazyBoolCalculate, which then resolves to a check against the current command depth. But that check was wrong for SBDebugger.HandleCommand, which is considered at depth 0, and anyway, I don't think it ever makes

[Lldb-commits] [PATCH] D126789: Stop regex commands from double entry into the history

jingham added a comment. In D126789#3550572 , @kastiglione wrote: > thank you for fixing! > > Q: why does this test have this thin `test_X`/`do_x_test` separation? We did that separation originally (and the sample test reflects that) so that you could

[Lldb-commits] [PATCH] D126789: Stop regex commands from double entry into the history

jingham added inline comments. Comment at: lldb/source/Commands/CommandObjectRegexCommand.cpp:74-76 // Pass in true for "no context switching". The command that called us // should have set up the context appropriately, we shouldn't have to // redo that. ---

[Lldb-commits] [PATCH] D126789: Stop regex commands from double entry into the history

jingham added inline comments. Comment at: lldb/source/Commands/CommandObjectRegexCommand.cpp:74-76 // Pass in true for "no context switching". The command that called us // should have set up the context appropriately, we shouldn't have to // redo that. ---

[Lldb-commits] [PATCH] D126789: Stop regex commands from double entry into the history

jingham updated this revision to Diff 434082. jingham added a comment. Fix a comment that hadn't been correct for a while... Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D126789/new/ https://reviews.llvm.org/D126789 Files: lldb/source/Commands/C

[Lldb-commits] [PATCH] D126789: Stop regex commands from double entry into the history

This revision was automatically updated to reflect the committed changes. Closed by commit rG8cc8b36f24d6: CommandObjectRegexCommand shouldn't put two commands on the history stack. (authored by jingham). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/

[Lldb-commits] [PATCH] D127038: Add some documentation for the "breakpoint name" feature

jingham created this revision. jingham added reviewers: kastiglione, mib, JDevlieghere, clayborg. Herald added a project: All. jingham requested review of this revision. Herald added a project: LLDB. Herald added a subscriber: lldb-commits. breakpoint names have a bunch of uses, some obvious and s

[Lldb-commits] [PATCH] D127016: [lldb] Prevent crash due to reading memory from page zero.

jingham added a comment. More generally, you can tell which elements of the exe_ctx you need to check in any given command by looking at requirements for the command which are passed into the constructor for the command. Any entity that's required there, you don't need to check for, but otherw

[Lldb-commits] [PATCH] D127038: Add some documentation for the "breakpoint name" feature

jingham updated this revision to Diff 434222. jingham added a comment. Address review comments. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D127038/new/ https://reviews.llvm.org/D127038 Files: lldb/source/Commands/CommandObjectBreakpoint.cpp I

[Lldb-commits] [PATCH] D127038: Add some documentation for the "breakpoint name" feature

jingham added a comment. Mostly reworking the paragraph on breakpoint hiding. Comment at: lldb/source/Commands/CommandObjectBreakpoint.cpp:2088 +The "liveness" of these settings is one way, from name to breakpoint. +If you use break modify to change an option that is also con

[Lldb-commits] [PATCH] D126513: Add -b (--continue-to-breakpoint) option to the "process continue" command

jingham updated this revision to Diff 435013. jingham added a comment. Allow for specifying breakpoint locations as well as breakpoints. Also fix the sync setting & resetting. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D126513/new/ https://review

<    3   4   5   6   7   8   9   10   11   12   >