[Lldb-commits] [PATCH] D82857: [LLDB] Add per-thread register infos shared pointer in gdb-remote

omjavaid updated this revision to Diff 304728. omjavaid added a comment. Minor update after changes to D82863 . CHANGES SINCE LAST ACTION https://reviews.llvm.org/D82857/new/ https://reviews.llvm.org/D82857 Files:

[Lldb-commits] [PATCH] D82863: [LLDB] Add support to resize SVE registers at run-time

omjavaid updated this revision to Diff 304726. omjavaid added a comment. This update addresses comments about last revision. Also sending offset field in qRegisterInfo and XML register description has been disabled D91241 CHANGES SINCE LAST ACTION

[Lldb-commits] [lldb] 856fd98 - Generalize regex matching std::string variants to compensate for recent

Author: Richard Smith Date: 2020-11-11T17:55:47-08:00 New Revision: 856fd98a176240470dcc2b8ad54b5c17ef6a75b3 URL: https://github.com/llvm/llvm-project/commit/856fd98a176240470dcc2b8ad54b5c17ef6a75b3 DIFF: https://github.com/llvm/llvm-project/commit/856fd98a176240470dcc2b8ad54b5c17ef6a75b3.diff

[Lldb-commits] [PATCH] D77153: [lldb/DataFormatters] Display null C++ pointers as nullptr

jingham accepted this revision. jingham added a comment. It seems weird that even if you had a summary formatter for some pointer type that was trying to print "DANGER WILL ROBINSON" when the pointer value was 0x0, we will override that and print "nullptr" in a C++ context or "nil" in an ObjC

[Lldb-commits] [PATCH] D90729: [trace][intel-pt] Scaffold the 'thread trace start | stop' commands

wallace updated this revision to Diff 304692. wallace added a comment. - Use CommandObjectProxy and delete the Delegate that I created - I had to add a few methods to the proxy to match with that I need. In any case, the implementation is not more complete. Repository: rG LLVM Github

[Lldb-commits] [PATCH] D91130: [crashlog] Implement parser for JSON encoded crashlogs

JDevlieghere updated this revision to Diff 304685. JDevlieghere marked an inline comment as done. JDevlieghere added a comment. Add tests for JSON and text format CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91130/new/ https://reviews.llvm.org/D91130 Files:

[Lldb-commits] [lldb] 0783ad9 - [lldb] Switch expect to runCmd in TestRecursiveTypes (NFC)

Author: Dave Lee Date: 2020-11-11T16:17:38-08:00 New Revision: 0783ad9e6a2c1156ca0bf5493473d6c98f006593 URL: https://github.com/llvm/llvm-project/commit/0783ad9e6a2c1156ca0bf5493473d6c98f006593 DIFF: https://github.com/llvm/llvm-project/commit/0783ad9e6a2c1156ca0bf5493473d6c98f006593.diff

[Lldb-commits] [PATCH] D91206: [lldb] Switch expect to runCmd in TestRecursiveTypes (NFC)

This revision was automatically updated to reflect the committed changes. Closed by commit rG0783ad9e6a2c: [lldb] Switch expect to runCmd in TestRecursiveTypes (NFC) (authored by kastiglione). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91206/new/

[Lldb-commits] [PATCH] D90729: [trace][intel-pt] Scaffold the 'thread trace start | stop' commands

clayborg added a comment. Walter and I discussed some issues offline and came to the conclusion we need to use CommandObjectProxy and get rid of CommandObjectDelegate. This will simplify the patch and use the pre-existing class that already does what we needed CommandObjectDelegate to do...

[Lldb-commits] [PATCH] D91238: Recognize __builtin_debugtrap on arm64, advance pc past it so users can continue easily

jingham added a comment. The attraction of having stub fix up the PC in the stop after hitting the trap for breakpoints (in this case by moving the PC before the trap on architectures that execute the trap) was that this decision could be made simply in the stub, but if lldb had to check

[Lldb-commits] [PATCH] D90729: [trace][intel-pt] Scaffold the 'thread trace start | stop' commands

clayborg added inline comments. Comment at: lldb/include/lldb/Interpreter/CommandObjectDelegate.h:30-32 + virtual lldb::CommandObjectSP DoGetDelegateCommand() = 0; + + CommandObject *GetDelegateCommand(); So currently you are storing the command object into

[Lldb-commits] [PATCH] D69589: [lldb] Refactor all POST_BUILD commands into targets

hhb closed this revision. hhb added a comment. I gave up on this.  Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69589/new/ https://reviews.llvm.org/D69589 ___ lldb-commits mailing list

[Lldb-commits] [PATCH] D91130: [crashlog] Implement parser for JSON encoded crashlogs

jasonmolenda added a comment. I guess the register context part of the crashlog.json would be the most architecture-dependent, but I bet crashlog.json doesn't use any of that when printing the backtrace -- it only uses the stack trace's pc values. So you could have a "x86_64" crashlog.json,

[Lldb-commits] [PATCH] D91130: [crashlog] Implement parser for JSON encoded crashlogs

jasonmolenda added a comment. In D91130#2389867 , @clayborg wrote: > > Crash logs can be loaded on other machines as well. Nothing stopping the test > from running on linux, windows, or other systems. The main thing we want to > test is if the JSON

[Lldb-commits] [PATCH] D91238: Recognize __builtin_debugtrap on arm64, advance pc past it so users can continue easily

clayborg added a comment. In D91238#2389877 , @jasonmolenda wrote: > Yeah, I don't know where it's best to centralize the logic of (1) recognize a > builtin_debugtrap instruction, and (2) set the pc so we step past it, know > how to step past it when

[Lldb-commits] [PATCH] D91238: Recognize __builtin_debugtrap on arm64, advance pc past it so users can continue easily

jasonmolenda added a comment. Yeah, I don't know where it's best to centralize the logic of (1) recognize a builtin_debugtrap instruction, and (2) set the pc so we step past it, know how to step past it when the user asks to resume control. debugserver (and lldb-server I'm sure) are already

[Lldb-commits] [PATCH] D91130: [crashlog] Implement parser for JSON encoded crashlogs

clayborg added a comment. In D91130#2389831 , @jasonmolenda wrote: > You could also make a fakey dsymForUUID script like we do in some other > corefile tests. Compile a C file with foo(), bar(), baz(), substitute the > addresses of those symbols from

[Lldb-commits] [PATCH] D90729: [trace][intel-pt] Scaffold the 'thread trace start | stop' commands

wallace marked 10 inline comments as done. wallace added inline comments. Comment at: lldb/include/lldb/Target/ProcessTrace.h:1 //===-- ProcessTrace.h --*- C++ -*-===// // clayborg wrote: > Should ProcessTrace and

[Lldb-commits] [PATCH] D91130: [crashlog] Implement parser for JSON encoded crashlogs

jasonmolenda added a comment. You could also make a fakey dsymForUUID script like we do in some other corefile tests. Compile a C file with foo(), bar(), baz(), substitute the addresses of those symbols from the binary into the crashlog.json, substitute the UUID of the compiled program in the

[Lldb-commits] [PATCH] D90729: [trace][intel-pt] Scaffold the 'thread trace start | stop' commands

wallace updated this revision to Diff 304651. wallace added a comment. Address comments: - Don't cache the command delegate. It was overcomplicating the code for little gain. - Create a RecordedProcess base class for trace and coredump processes, so that we can put the IsLiveDebugSession

[Lldb-commits] [PATCH] D91130: [crashlog] Implement parser for JSON encoded crashlogs

clayborg added a comment. In D91130#2387692 , @JDevlieghere wrote: > In D91130#2387390 , @clayborg wrote: > >> JSON crash logs are almost here??? I remember asking for those back when I >> was at Apple! Great

[Lldb-commits] [PATCH] D91238: Recognize __builtin_debugtrap on arm64, advance pc past it so users can continue easily

clayborg added a comment. Another possible way to do this would be to make a new stop reason in LLDB. debugserver currently can return a "reason" whose values are one of "trap", "breakpoint", "watchpoint", "exception", or "exec". What if we added "debug-trap"? Then back in LLDB we could add

[Lldb-commits] [PATCH] D69589: [lldb] Refactor all POST_BUILD commands into targets

JDevlieghere added a comment. Ha, I looked at the date but didn't realize this was a review from fall last year :-) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69589/new/ https://reviews.llvm.org/D69589

[Lldb-commits] [PATCH] D69589: [lldb] Refactor all POST_BUILD commands into targets

JDevlieghere added a subscriber: sgraenitz. JDevlieghere added a comment. In D69589#1769663 , @sgraenitz wrote: > I'd recommend to be careful when making a change of this size. There are a > lot of possible configurations. > If you split it up into a few

[Lldb-commits] [PATCH] D91293: [lldb] [Process/FreeBSDRemote] Modernize and simplify YMM logic

mgorny created this revision. mgorny added reviewers: labath, emaste, krytarowski. mgorny requested review of this revision. Eliminate the remaining swith-case code for register getters, and migrate YMM registers to regset-oriented model. Since these registers are recombined from XMM and

[Lldb-commits] [lldb] d2f18e6 - Fix 21555fff4de811309ea7935f9cb65578c957d77f

Author: Walter Erquinigo Date: 2020-11-11T12:30:24-08:00 New Revision: d2f18e6b1e0bed97f5218af499c4e0b88c4dd361 URL: https://github.com/llvm/llvm-project/commit/d2f18e6b1e0bed97f5218af499c4e0b88c4dd361 DIFF:

[Lldb-commits] [PATCH] D90490: [intel-pt][trace] Implement a "get supported trace type" packet

wallace added a comment. @stella.stamenova I'm working on a fix right now. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D90490/new/ https://reviews.llvm.org/D90490 ___ lldb-commits mailing list

[Lldb-commits] [PATCH] D77153: [lldb/DataFormatters] Display null C++ pointers as nullptr

JDevlieghere updated this revision to Diff 304614. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D77153/new/ https://reviews.llvm.org/D77153 Files: lldb/include/lldb/Target/Language.h lldb/source/DataFormatters/ValueObjectPrinter.cpp lldb/source/Expression/UserExpression.cpp

[Lldb-commits] [PATCH] D90490: [intel-pt][trace] Implement a "get supported trace type" packet

stella.stamenova added a comment. It looks like this broke the windows lldb buildbot: http://lab.llvm.org:8011/#/builders/83/builds/693 Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D90490/new/ https://reviews.llvm.org/D90490

[Lldb-commits] [lldb] fc8c1ea - [lldb/test] Add missing decorators import

Author: Vedant Kumar Date: 2020-11-11T10:48:27-08:00 New Revision: fc8c1ea9aff376526e5e18f60caaf84b81785886 URL: https://github.com/llvm/llvm-project/commit/fc8c1ea9aff376526e5e18f60caaf84b81785886 DIFF: https://github.com/llvm/llvm-project/commit/fc8c1ea9aff376526e5e18f60caaf84b81785886.diff

[Lldb-commits] [lldb] b7c06dc - [ThreadPlan] Add a test for `thread step-in -r`, NFC (reapply)

Author: Vedant Kumar Date: 2020-11-11T10:43:38-08:00 New Revision: b7c06dcb739f8a30c8dc9d4cc8832a9c939a1c49 URL: https://github.com/llvm/llvm-project/commit/b7c06dcb739f8a30c8dc9d4cc8832a9c939a1c49 DIFF: https://github.com/llvm/llvm-project/commit/b7c06dcb739f8a30c8dc9d4cc8832a9c939a1c49.diff

[Lldb-commits] [PATCH] D90490: [intel-pt][trace] Implement a "get supported trace type" packet

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG21555fff4de8: [intel-pt][trace] Implement a get supported trace type packet (authored by wallace). Changed prior to commit:

[Lldb-commits] [lldb] 21555ff - [intel-pt][trace] Implement a "get supported trace type" packet

Author: Walter Erquinigo Date: 2020-11-11T10:35:58-08:00 New Revision: 21555fff4de811309ea7935f9cb65578c957d77f URL: https://github.com/llvm/llvm-project/commit/21555fff4de811309ea7935f9cb65578c957d77f DIFF:

[Lldb-commits] [PATCH] D91103: [tooling] Add support for fixits that indicate code will need reformatting

njames93 updated this revision to Diff 304574. njames93 marked an inline comment as done. njames93 added a comment. Address nit by replacing optional. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91103/new/ https://reviews.llvm.org/D91103 Files:

Re: [Lldb-commits] [lldb] d9624f4 - Revert "[ThreadPlan] Add a test for `thread step-in -r`, NFC"

Thanks for the revert. Do you know anyone actively working on ThreadPlan support for Windows? vedant > On Nov 11, 2020, at 9:10 AM, Stella Stamenova via lldb-commits > wrote: > > > Author: Stella Stamenova > Date: 2020-11-11T09:09:43-08:00 > New Revision:

[Lldb-commits] [PATCH] D91206: [lldb] Switch expect to runCmd in TestRecursiveTypes (NFC)

kastiglione updated this revision to Diff 304568. kastiglione added a comment. Remove bogus message. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91206/new/ https://reviews.llvm.org/D91206 Files: lldb/test/API/types/TestRecursiveTypes.py

[Lldb-commits] [PATCH] D91193: [lldb] Fine tune expect() validation

kastiglione added a comment. Thanks for pointing this in the right direction @labath, @teemperor. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91193/new/ https://reviews.llvm.org/D91193 ___

[Lldb-commits] [lldb] d9624f4 - Revert "[ThreadPlan] Add a test for `thread step-in -r`, NFC"

Author: Stella Stamenova Date: 2020-11-11T09:09:43-08:00 New Revision: d9624f444807bdac92e37f85ab07db1eb8a2bdf2 URL: https://github.com/llvm/llvm-project/commit/d9624f444807bdac92e37f85ab07db1eb8a2bdf2 DIFF:

[Lldb-commits] [PATCH] D91268: [lldb] [Process/FreeBSDRemote] Access debug registers via offsets

mgorny created this revision. mgorny added reviewers: labath, emaste, krytarowski. Herald added a subscriber: pengfei. mgorny requested review of this revision. Use offset-based method to access x86 debug registers. This also involves adding a test for the correctness of these offsets, and

[Lldb-commits] [PATCH] D91264: [lldb] [test] Add a minimal test for x86 dbreg reading

mgorny created this revision. mgorny added reviewers: labath, krytarowski, emaste. Herald added a subscriber: pengfei. mgorny requested review of this revision. Add a test verifying that after the 'watchpoint' command, new values of x86 debug registers can be read back correctly. The primary

[Lldb-commits] [PATCH] D91254: [lldb] [Process/Utility] Fix DR offsets for FreeBSD

mgorny created this revision. mgorny added reviewers: labath, krytarowski, emaste. Herald added a subscriber: arichardson. mgorny requested review of this revision. Fix Debug Register offsets to be specified relatively to UserArea on FreeBSD/amd64 and FreeBSD/i386, and add them to UserArea on

[Lldb-commits] [PATCH] D91248: [lldb] [Process/FreeBSDRemote] Access FPR via RegisterInfo offsets

mgorny updated this revision to Diff 304479. mgorny added a comment. Rename `ASSERT_FPR` to `ASSERT_OFF` since it's a more generic offset-based assert. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91248/new/ https://reviews.llvm.org/D91248 Files:

[Lldb-commits] [PATCH] D91103: [tooling] Add support for fixits that indicate code will need reformatting

kadircet added inline comments. Comment at: clang-tools-extra/clangd/Diagnostics.cpp:677 // If requested and possible, create a message like "change 'foo' to 'bar'". -if (SyntheticMessage && FixIts.size() == 1) { - const auto = FixIts.front(); +if

[Lldb-commits] [PATCH] D91103: [tooling] Add support for fixits that indicate code will need reformatting

njames93 updated this revision to Diff 304477. njames93 marked 2 inline comments as done. njames93 added a comment. Removed the first loop for clangd diagnostic, turns out it didnt make the following code that much messier. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[Lldb-commits] [PATCH] D91248: [lldb] [Process/FreeBSDRemote] Access FPR via RegisterInfo offsets

mgorny created this revision. mgorny added reviewers: labath, krytarowski, emaste. mgorny requested review of this revision. Use offset-based method to access base x87 FPU registers, using offsets relative to the position of 'struct FPR', as determined by the location of first register in it

[Lldb-commits] [PATCH] D91103: [tooling] Add support for fixits that indicate code will need reformatting

njames93 added inline comments. Comment at: clang-tools-extra/clangd/Diagnostics.cpp:636 +// Filter out any reformat fixits, we don't handle these. +// FIXME: Can we? +llvm::erase_if(FixIts, kadircet wrote: > in theory yes, as we have access to

[Lldb-commits] [PATCH] D91216: [lldb] [Process/FreeBSDRemote] Access GPR via reginfo offsets

mgorny updated this revision to Diff 304448. mgorny added a comment. Add a helper `ASSERT_REG` macro. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91216/new/ https://reviews.llvm.org/D91216 Files: lldb/source/Plugins/Process/FreeBSDRemote/NativeRegisterContextFreeBSD_x86_64.cpp

[Lldb-commits] [PATCH] D91216: [lldb] [Process/FreeBSDRemote] Access GPR via reginfo offsets

mgorny updated this revision to Diff 304445. mgorny added a comment. clang-format CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91216/new/ https://reviews.llvm.org/D91216 Files: lldb/source/Plugins/Process/FreeBSDRemote/NativeRegisterContextFreeBSD_x86_64.cpp

[Lldb-commits] [PATCH] D87442: [lldb] Show flags for memory regions

DavidSpickett updated this revision to Diff 304441. DavidSpickett added a comment. - Remove unused assert header - Simplify error handling in LinuxProcMapsTest Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D87442/new/

[Lldb-commits] [PATCH] D87442: [lldb] Show flags for memory regions

DavidSpickett added inline comments. Comment at: lldb/test/API/linux/aarch64/mte_memory_region/main.c:9-14 + if (!(getauxval(AT_HWCAP2) & HWCAP2_MTE)) +return 1; + + int got = prctl(PR_SET_TAGGED_ADDR_CTRL, PR_TAGGED_ADDR_ENABLE, 0, 0, 0); + if (got) +return 1;

[Lldb-commits] [PATCH] D91241: [LLDB] Make offset field optional in RegisterInfo packet for Arm64

omjavaid created this revision. omjavaid added a reviewer: labath. Herald added a subscriber: kristof.beyls. omjavaid requested review of this revision. This patch carries forward our aim to remove offset field from qRegisterInfo packets and XML register description. I have created a new

[Lldb-commits] [PATCH] D91216: [lldb] [Process/FreeBSDRemote] Access GPR via reginfo offsets

mgorny updated this revision to Diff 304433. mgorny edited the summary of this revision. mgorny added a comment. Updated to use unittests instead of static asserts. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91216/new/ https://reviews.llvm.org/D91216 Files:

[Lldb-commits] [PATCH] D91206: [lldb] Switch expect to runCmd in TestRecursiveTypes (NFC)

teemperor accepted this revision. teemperor added a comment. This revision is now accepted and ready to land. I don't think we should re-add the bogus `RUN_SUCCEEDED` messages here. `self.runCmd("print tpi")` is undocumented, but still better than wrong documentation. Beside that this LGTM,

[Lldb-commits] [PATCH] D77153: [lldb/DataFormatters] Display null C++ pointers as nullptr

teemperor requested changes to this revision. teemperor added a comment. This revision now requires changes to proceed. A few comments, but otherwise this seems good. Comment at: lldb/include/lldb/Target/Language.h:214 + virtual llvm::StringRef NilReferenceSummaryString() {

[Lldb-commits] [PATCH] D87442: [lldb] Show flags for memory regions

labath added a comment. I think this looks pretty good now. Some questions inline... Comment at: lldb/source/Target/MemoryRegionInfo.cpp:10 #include "lldb/Target/MemoryRegionInfo.h" +#include "lldb/Utility/LLDBAssert.h" unused? Comment

[Lldb-commits] [PATCH] D91103: [tooling] Add support for fixits that indicate code will need reformatting

kadircet added inline comments. Comment at: clang-tools-extra/clangd/Diagnostics.cpp:636 +// Filter out any reformat fixits, we don't handle these. +// FIXME: Can we? +llvm::erase_if(FixIts, in theory yes, as we have access to source manager, we can

[Lldb-commits] [PATCH] D91238: Recognize __builtin_debugtrap on arm64, advance pc past it so users can continue easily

labath added a reviewer: jingham. labath added a comment. Including Jim, as he was involved in the discussion last time this came around http://lists.llvm.org/pipermail/lldb-dev/2020-March/016017.html. This functionality would not work on linux/lldb-server currently, though I would love if it

[Lldb-commits] [PATCH] D77153: [lldb/DataFormatters] Display null C++ pointers as nullptr

JDevlieghere updated this revision to Diff 304415. JDevlieghere added a comment. Fix the C test CHANGES SINCE LAST ACTION https://reviews.llvm.org/D77153/new/ https://reviews.llvm.org/D77153 Files: lldb/include/lldb/Target/Language.h lldb/source/DataFormatters/ValueObjectPrinter.cpp

[Lldb-commits] [PATCH] D91216: [lldb] [Process/FreeBSDRemote] Access GPR via reginfo offsets

labath added a comment. Regarding the asserts, I'm thinking if it wouldn't be better to check the compare the resulting RegisterInfo structs instead of the raw struct layout. The reasons for that are: - it will be checking the thing that we're actually relying on (the register info macros is

[Lldb-commits] [PATCH] D91193: [lldb] Fine tune expect() validation

labath added a comment. Yes, runCmd is the way to go here... Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91193/new/ https://reviews.llvm.org/D91193 ___ lldb-commits mailing list

[Lldb-commits] [lldb] 66ae40e - [lldb][test] Remove not_remote_testsuite_ready in favor of skipIfRemote decorator

Author: Raphael Isemann Date: 2020-11-11T09:14:54+01:00 New Revision: 66ae40ebfb83c8beb2080123d3866bd08520fcef URL: https://github.com/llvm/llvm-project/commit/66ae40ebfb83c8beb2080123d3866bd08520fcef DIFF:

[Lldb-commits] [PATCH] D91238: Recognize __builtin_debugtrap on arm64, advance pc past it so users can continue easily

jasonmolenda created this revision. jasonmolenda added reviewers: clayborg, labath. jasonmolenda added a project: LLDB. Herald added subscribers: omjavaid, JDevlieghere, pengfei, kristof.beyls. jasonmolenda requested review of this revision. This is a change I need on Darwin systems, so I'm