[Lldb-commits] [PATCH] D141629: Run address expression argument values through ABI::FixCodeAddress to strip TBI/pointer auth bytes on AArch64

jasonmolenda created this revision. jasonmolenda added a reviewer: DavidSpickett. jasonmolenda added a project: LLDB. Herald added subscribers: Michael137, kristof.beyls. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. This is a

[Lldb-commits] [PATCH] D141972: Delay loading of qProcessInfo-informed binaries until later in the Process setup

jasonmolenda created this revision. jasonmolenda added a reviewer: jingham. jasonmolenda added a project: LLDB. Herald added a subscriber: JDevlieghere. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. I handle two types of binar

[Lldb-commits] [PATCH] D141972: Delay loading of qProcessInfo-informed binaries until later in the Process setup

jasonmolenda updated this revision to Diff 490232. jasonmolenda added a comment. Add a comment about the intentional placement of binary loading/load address setting in DidLaunchOrAttach to address Jim's feedback. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://review

[Lldb-commits] [PATCH] D141972: Delay loading of qProcessInfo-informed binaries until later in the Process setup

jasonmolenda updated this revision to Diff 490258. jasonmolenda added a comment. Thanks for the feedback Jonas, updated patch. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D141972/new/ https://reviews.llvm.org/D141972 Files: lldb/source/Plugins/

[Lldb-commits] [PATCH] D141972: Delay loading of qProcessInfo-informed binaries until later in the Process setup

This revision was automatically updated to reflect the committed changes. Closed by commit rGdb223b7f01f7: Do qProcessInfo-hint binary loading later in Process setup (authored by jasonmolenda). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D141972/new

[Lldb-commits] [PATCH] D142067: Remove the "help" subcommand - its undocumented, undiscoverable and doesn't work....

jasonmolenda added a comment. Yeah this is a good cleanup, but fwiw maybe because using git, but I sometimes expect `break set --help` etc to work, forgetting which style lldb uses. just throwing it out there as something I get tripped up by once in a while, not really related to this change,

[Lldb-commits] [PATCH] D141629: Run address expression argument values through ABI::FixCodeAddress to strip TBI/pointer auth bytes on AArch64

This revision was automatically updated to reflect the committed changes. Closed by commit rG484bc2bcc799: Run cmdline address expressions through ABI's FixAddress (authored by jasonmolenda). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D141629/new/

[Lldb-commits] [PATCH] D142662: Allow bytes-only mach-o corefile to load into lldb

jasonmolenda created this revision. jasonmolenda added reviewers: JDevlieghere, mib. jasonmolenda added a project: LLDB. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. `ProcessMachCore::DoLoadCore` has an error check that if it

[Lldb-commits] [PATCH] D142663: Improve disable-language-runtime-unwindplans description to be more searchable

jasonmolenda created this revision. jasonmolenda added a reviewer: kastiglione. jasonmolenda added a project: LLDB. Herald added a subscriber: JDevlieghere. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. The current description

[Lldb-commits] [PATCH] D142663: Improve disable-language-runtime-unwindplans description to be more searchable

This revision was automatically updated to reflect the committed changes. Closed by commit rG8112bd2cb484: disable-language-runtime-unwindplans desc rewrite to be searchable (authored by jasonmolenda). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D14

[Lldb-commits] [PATCH] D142662: Allow bytes-only mach-o corefile to load into lldb

This revision was automatically updated to reflect the committed changes. Closed by commit rG841b26f1d80f: Don't flag memory-only mach-o corefiles as invalid (authored by jasonmolenda). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D142662/new/ https

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

jasonmolenda created this revision. jasonmolenda added a reviewer: DavidSpickett. jasonmolenda added a project: LLDB. Herald added subscribers: omjavaid, JDevlieghere, kristof.beyls. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

jasonmolenda added a comment. Thanks for the nits, good points. Comment at: lldb/source/API/SBValue.cpp:935-940 +ProcessSP process_sp = m_opaque_sp->GetProcessSP(); +if (!process_sp) + return ret_val; +ABISP abi_sp = process_sp->GetABI(); +if (abi_sp) +

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

jasonmolenda added inline comments. Comment at: lldb/source/API/SBValue.cpp:935-940 +ProcessSP process_sp = m_opaque_sp->GetProcessSP(); +if (!process_sp) + return ret_val; +ABISP abi_sp = process_sp->GetABI(); +if (abi_sp) + return abi_sp->FixCodeAddres

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

jasonmolenda updated this revision to Diff 492951. jasonmolenda added a comment. Fix the nits Jonas pointed out, thanks Jonas. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D142792/new/ https://reviews.llvm.org/D142792 Files: lldb/bindings/interf

[Lldb-commits] [PATCH] D142793: Remove -py3 command line arg from swig invocation; generates warning with swig 4.1.1

jasonmolenda created this revision. jasonmolenda added a reviewer: JDevlieghere. jasonmolenda added a project: LLDB. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. When I touch the bindings and build with ninja, I see [5/15]

[Lldb-commits] [PATCH] D142793: Remove -py3 command line arg from swig invocation; generates warning with swig 4.1.1

jasonmolenda abandoned this revision. jasonmolenda added a comment. hasn't landed it, hasn't beaten me!!! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D142793/new/ https://reviews.llvm.org/D142793 ___ ll

[Lldb-commits] [PATCH] D143012: Add a bit of error checking to SBProcess::ReadMemory to check if a nullptr destination buffer was provided, return error

jasonmolenda created this revision. jasonmolenda added a reviewer: JDevlieghere. jasonmolenda added a project: LLDB. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. I helped a developer using SB API yesterday who had a bug and t

[Lldb-commits] [PATCH] D143215: Separate Process::GetWatchpointSupportInfo into two methods to get the two separate pieces of information

jasonmolenda created this revision. jasonmolenda added a reviewer: jingham. jasonmolenda added a project: LLDB. Herald added subscribers: JDevlieghere, kristof.beyls. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. There are two

[Lldb-commits] [PATCH] D143215: Separate Process::GetWatchpointSupportInfo into two methods to get the two separate pieces of information

jasonmolenda updated this revision to Diff 494428. jasonmolenda added a comment. Ah, I wasn't paying close enough attention and got the logic for MIPS/PPC64 watchpoint exceptions backwards in my update to GDBRemoteCommunicationClient::GetWatchpointReportedAfter, update patch for that. Reposito

[Lldb-commits] [PATCH] D143215: Separate Process::GetWatchpointSupportInfo into two methods to get the two separate pieces of information

jasonmolenda added a comment. tl;dr: watchpoints don't work on things like arm targets with a JTAG gdb stub that doesn't support `qWatchpointSupportInfo:` or `qHostInfo` with the `watchpoint_exceptions_received` key. Nearly all of this patch is mechanical, and IMO the question is whether I sho

[Lldb-commits] [PATCH] D143232: Return an error when the CFA resolves to no known register, instead of segfaulting

jasonmolenda created this revision. jasonmolenda added a reviewer: bulbazord. jasonmolenda added a project: LLDB. Herald added a subscriber: JDevlieghere. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. I'm working on a bug repo

[Lldb-commits] [PATCH] D143232: Return an error when the CFA resolves to no known register, instead of segfaulting

jasonmolenda added a comment. In D143232#4101396 , @bulbazord wrote: > LGTM. I wonder if there's a good way to exercise this with a test? Like maybe > we can create some bogus unwind information and see if LLDB falls over when > consuming it? I'm not s

[Lldb-commits] [PATCH] D143232: Return an error when the CFA resolves to no known register, instead of segfaulting

This revision was automatically updated to reflect the committed changes. Closed by commit rG701030c3ecba: In InitializeZerothFrame check for a CFA/AFA or error out (authored by jasonmolenda). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D143232/new/

[Lldb-commits] [PATCH] D143215: Separate Process::GetWatchpointSupportInfo into two methods to get the two separate pieces of information

jasonmolenda updated this revision to Diff 495385. jasonmolenda edited the summary of this revision. jasonmolenda added a comment. David's feedback was spot on, overhaul patch to address. Change Process::GetWatchpointSlotCount to return an optional count, for when it is available, instead of dep

[Lldb-commits] [PATCH] D143215: Separate Process::GetWatchpointSupportInfo into two methods to get the two separate pieces of information

jasonmolenda added a comment. I phabracated wrong and the end of the last msg was truncated. I meant to end it with this: If the number of watchpoints could not be fetched with an lldb-extension packet, Process:GetWatchpointSupportInfo would not report whether breakpoints are received before

[Lldb-commits] [PATCH] D143012: Add a bit of error checking to SBProcess::ReadMemory to check if a nullptr destination buffer was provided, return error

jasonmolenda updated this revision to Diff 495634. jasonmolenda added a comment. Update patch for Jonas' suggestions. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D143012/new/ https://reviews.llvm.org/D143012 Files: lldb/source/API/SBProcess.cpp

[Lldb-commits] [PATCH] D143012: Add a bit of error checking to SBProcess::ReadMemory to check if a nullptr destination buffer was provided, return error

This revision was automatically updated to reflect the committed changes. Closed by commit rG62c747517cd9: Check if null buffer handed to SBProcess::ReadMemory (authored by jasonmolenda). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D143012/new/ htt

[Lldb-commits] [PATCH] D143215: Separate Process::GetWatchpointSupportInfo into two methods to get the two separate pieces of information

jasonmolenda updated this revision to Diff 495686. jasonmolenda added a comment. Went over the rewrite one more time, fixing up comments, fixing a couple of logic bugs I wrote, and having tested it with a debugserver I hacked to remove the `watchpoint_exceptions_received:before;` key in the qHos

[Lldb-commits] [PATCH] D143215: Separate Process::GetWatchpointSupportInfo into two methods to get the two separate pieces of information

jasonmolenda added a comment. If I was going to describe this patch briefly at this point, I would say: "Change lldb from depending on the remote gdb stub to tell it whether watchpoints are received before or after the instruction, to knowing the architectures where watchpoint exceptions are re

[Lldb-commits] [PATCH] D142715: [LLDB] Apply FixCodeAddress to all forms of address arguments

jasonmolenda accepted this revision. jasonmolenda added a comment. This revision is now accepted and ready to land. LGTM. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D142715/new/ https://reviews.llvm.org/D142715 __

[Lldb-commits] [PATCH] D143215: Separate Process::GetWatchpointSupportInfo into two methods to get the two separate pieces of information

jasonmolenda added inline comments. Comment at: lldb/source/Target/Process.cpp:2373 +reported_after = false; + return reported_after; +} Emmmer wrote: > DavidSpickett wrote: > > DavidSpickett wrote: > > > Would this be any clearer with multiple returns? Or o

[Lldb-commits] [PATCH] D143215: lldb can know architecture-dependent watchpoint behaviors, instead of depending on the remote stub

jasonmolenda updated this revision to Diff 495976. jasonmolenda retitled this revision from "Separate Process::GetWatchpointSupportInfo into two methods to get the two separate pieces of information" to "lldb can know architecture-dependent watchpoint behaviors, instead of depending on the remot

[Lldb-commits] [PATCH] D143932: Further delay simulator SDK filepath discovery until needed

jasonmolenda created this revision. jasonmolenda added a reviewer: aprantl. jasonmolenda added a project: LLDB. Herald added a subscriber: JDevlieghere. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. In https://reviews.llvm.org

[Lldb-commits] [PATCH] D143932: Further delay simulator SDK filepath discovery until needed

This revision was not accepted when it landed; it landed in state "Needs Review". This revision was automatically updated to reflect the committed changes. Closed by commit rGee88f1177003: Find SDK path more lazily in Apple Simulator platforms (authored by jasonmolenda). Changed prior to commit:

[Lldb-commits] [PATCH] D143215: lldb can know architecture-dependent watchpoint behaviors, instead of depending on the remote stub

This revision was automatically updated to reflect the committed changes. Closed by commit rGeaeb8ddd4a9d: [LLDB] add arch-specific watchpoint behavior defaults to lldb (authored by jasonmolenda). Changed prior to commit: https://reviews.llvm.org/D143215?vs=495976&id=497401#toc Repository: r

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

jasonmolenda added inline comments. Comment at: lldb/source/API/SBValue.cpp:936 + if (ABISP abi_sp = process_sp->GetABI()) +return abi_sp->FixCodeAddress(ret_val); +return ret_val; DavidSpickett wrote: > We've talked about this before, and for no

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

jasonmolenda updated this revision to Diff 497461. jasonmolenda added a comment. Updated patch to address David's feedback. Thanks David! Updated comments and SB API doc to conform to English. Not decided on whether we should use FixCodeAddress for our strippings here; the difference between

[Lldb-commits] [PATCH] D144142: [debugserver] Initialize logging earlier in the startup sequence

jasonmolenda added inline comments. Comment at: lldb/tools/debugserver/source/debugserver.cpp:1342 DNBLogSetLogMask(log_flags); } Should this be moved out of this `log_file != NULL` block so that the logging flags are set based on `-f` value? Otherwis

[Lldb-commits] [PATCH] D144528: [lldb] Warn when Mach-O files have overlapping segments

jasonmolenda accepted this revision. jasonmolenda added a comment. I'm OK with giving this a try. SectionLoadList::SetSectionLoadAddress specifically notes that there are cases were sections overlap in the virtual address space. All of the binaries in the shared cache have a single LINKEDIT s

[Lldb-commits] [PATCH] D144528: [lldb] Warn when Mach-O files have overlapping segments

jasonmolenda added a comment. In D144528#4145610 , @bulbazord wrote: > In D144528#4145593 , @jasonmolenda > wrote: > >> >> tl;dr it's fine if LINKEDIT's "overlap" because lldb will never need to take >> an a

[Lldb-commits] [PATCH] D144311: [debugserver] Add one additional sleep before attaching after waiting

jasonmolenda accepted this revision. jasonmolenda added a comment. This revision is now accepted and ready to land. Ah, sorry I missed this one, LGTM. Attaching to a process very early in startup during this transition between dyld's is especially difficult to recover from; letting the process

[Lldb-commits] [PATCH] D144937: [LLDB] Expose several methods in SBWatchpoint

jasonmolenda added a comment. Hi Dan, I hadn't looked this over very closely but one thing that jumped out is that you're adding a member to SBWatchpoint, and we can't do that, it's an API breaking change. In cases where we need to store additional information than a weak pointer to an lldb pr

[Lldb-commits] [PATCH] D144937: [LLDB] Expose several methods in SBWatchpoint

jasonmolenda added a comment. In D144937#4159324 , @jasonmolenda wrote: > In cases where we need to store additional information than a weak pointer to > an lldb private object, we traditionally add an Impl class which has the > additional member(s) an

[Lldb-commits] [PATCH] D145020: Hoist debugserver arch-dep sources out of a side CMakeLists into the debugserver main CMakeLists

jasonmolenda created this revision. jasonmolenda added a reviewer: JDevlieghere. jasonmolenda added a project: LLDB. Herald added a subscriber: kristof.beyls. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. There was historical

[Lldb-commits] [PATCH] D145020: Hoist debugserver arch-dep sources out of a side CMakeLists into the debugserver main CMakeLists

This revision was automatically updated to reflect the committed changes. Closed by commit rGf136ca848309: Put the arch-dep debugserver files in main CMakeLists.txt (authored by jasonmolenda). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D145020/new/

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

jasonmolenda updated this revision to Diff 501574. jasonmolenda added a comment. Clean up the test case - the C file and python file - to address David's feedback and test it more thoroughly. Fix `ValueObjectPrinter::PrintValueAndSummaryIfNeeded()` to only add the `actual=hex` if this is a poi

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

This revision was automatically updated to reflect the committed changes. Closed by commit rG4d635be2dbad: Add SBValue::GetValueAsAddress API for removing non-addressing metadata (authored by jasonmolenda). Changed prior to commit: https://reviews.llvm.org/D142792?vs=501574&id=501962#toc Repos

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

jasonmolenda updated this revision to Diff 501995. jasonmolenda added a comment. Updating patch with the final version, barring some fix for the armv7 ubuntu bot and TestDataFormatterSmartArray.py line 229 where it has a c-string pointer that gets its 0th bit cleared, Got output: (char *) s

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

jasonmolenda added a comment. Need to figure out something for the armv7 case before I try re-landing. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D142792/new/ https://reviews.llvm.org/D142792 ___ lldb-

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

jasonmolenda added a comment. Maybe the right answer is to use FixDataAddress. It's possible someone could make a AArch64 FixCodeAddress that clears the low 2 bits. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D142792/new/ https://reviews.llvm.or

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

jasonmolenda added a comment. Hm, I've also got some problems with some ObjC tagged pointers in the testsuite, regardless of the FixCodeAddress/FixDataAddress issue. Need to look at this a bit more. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D1

[Lldb-commits] [PATCH] D145547: When setting load addresses on darwin kernel kexts, handle case where in-memory load commands are not updated

jasonmolenda created this revision. jasonmolenda added a reviewer: JDevlieghere. jasonmolenda added a project: LLDB. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. The Darwin kernel has a number of kernel extensions, kexts, aki

[Lldb-commits] [PATCH] D145547: When setting load addresses on darwin kernel kexts, handle case where in-memory load commands are not updated

jasonmolenda updated this revision to Diff 503497. jasonmolenda added a comment. Show an alternate approach, where we check the file types of everything in the kernel+kext list, and if something is not a kernel/kext, don't try to load it at all. v. the first change in `DynamicLoaderDarwinKerne

[Lldb-commits] [PATCH] D145547: When setting load addresses on darwin kernel kexts, handle case where in-memory load commands are not updated

jasonmolenda updated this revision to Diff 503550. jasonmolenda added a comment. Add an additional bit to the "don't load non-kexts, non-kernels" addition. I refined this by checking to see if the binary is already registered in the Target, and already has a load address set. If the binary is,

[Lldb-commits] [PATCH] D145547: When setting load addresses on darwin kernel kexts, handle case where in-memory load commands are not updated

jasonmolenda marked an inline comment as done. jasonmolenda added inline comments. Comment at: lldb/source/Plugins/DynamicLoader/Darwin-Kernel/DynamicLoaderDarwinKernel.cpp:835 + m_name.c_str(), m_uuid.GetAsString().c_str()); +m_module_sp.reset()

[Lldb-commits] [PATCH] D145547: When setting load addresses on darwin kernel kexts, handle case where in-memory load commands are not updated

jasonmolenda updated this revision to Diff 503569. jasonmolenda marked an inline comment as done. jasonmolenda added a comment. Update for Jonas' suggestions. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D145547/new/ https://reviews.llvm.org/D14554

[Lldb-commits] [PATCH] D145547: When setting load addresses on darwin kernel kexts, handle case where in-memory load commands are not updated

This revision was automatically updated to reflect the committed changes. Closed by commit rG794d2089cd74: Don't load non-kexts in darwin kernel debug; handle unslid segs (authored by jasonmolenda). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D14554

[Lldb-commits] [PATCH] D145568: [lldb] Add dummy field to RegisterInfo for register flags use later

jasonmolenda accepted this revision. jasonmolenda added a comment. This revision is now accepted and ready to land. LGTM. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D145568/new/ https://reviews.llvm.org/D145568 __

[Lldb-commits] [PATCH] D145566: [lldb] Add RegisterFlags class

jasonmolenda accepted this revision. jasonmolenda added a comment. This revision is now accepted and ready to land. LGTM. I'm not really advocating to changing `m_type` to be an enum, but curious how that might fit into the total context of this feature, and if it makes any sense. ==

[Lldb-commits] [PATCH] D145580: [lldb] Show register fields using bitfield struct types

jasonmolenda added a comment. This all looks good to me. the phab says there's a missing newline at the end of TestXMLRegisterFlags.py. Given that we can construct fake debug sessions with a gdb_remote_client test, do you think it would be worth having a test that's says it's a big-endian tar

[Lldb-commits] [PATCH] D145580: [lldb] Show register fields using bitfield struct types

jasonmolenda accepted this revision. jasonmolenda added a comment. This revision is now accepted and ready to land. I'll mark it as accepted, I don't know if Jim or Pavel have any comments. I think this is good. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews

[Lldb-commits] [PATCH] D145574: [lldb] Read register fields from target XML

jasonmolenda accepted this revision. jasonmolenda added a comment. This revision is now accepted and ready to land. This looks good to me. Comment at: lldb/source/Plugins/Process/gdb-remote/ProcessGDBRemote.cpp:4066 +"ProcessGDBRemote::ParseFlags Invalid start %

[Lldb-commits] [PATCH] D142792: Add SBValue::GetValueAsAddress(), strip off ptrauth, TBI, MTE bits on AArch64 systems

jasonmolenda planned changes to this revision. jasonmolenda added a comment. OK, I understand the objc testsuite failures. ObjC has tagged pointers, where the address of an object is either a virtual address, or a big old bag of bits that can contain a small value, like instead of a pointer to

[Lldb-commits] [PATCH] D145580: [lldb] Show register fields using bitfield struct types

jasonmolenda added a comment. This gdb_remote_client test looks real nice with this update. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D145580/new/ https://reviews.llvm.org/D145580 ___ lldb-commits mai

[Lldb-commits] [PATCH] D147462: Use kernel's global variable indicating how many bits are used in addressing when loading Darwin xnu kernel

jasonmolenda created this revision. jasonmolenda added a reviewer: JDevlieghere. jasonmolenda added a project: LLDB. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. The Darwin xnu kernel has a global variable which has the numbe

[Lldb-commits] [PATCH] D147462: Use kernel's global variable indicating how many bits are used in addressing when loading Darwin xnu kernel

jasonmolenda added inline comments. Comment at: lldb/source/Plugins/DynamicLoader/Darwin-Kernel/DynamicLoaderDarwinKernel.cpp:1093 + // don't mark those as addressable. + m_process->SetVirtualAddressableBits((wordsize * 8) - 9); + addr_t sym_addr = sym

[Lldb-commits] [PATCH] D147462: Use kernel's global variable indicating how many bits are used in addressing when loading Darwin xnu kernel

jasonmolenda updated this revision to Diff 510586. jasonmolenda added a comment. Update patch to address some of Jonas' comments. I was adding a check that the ptrsize of the architecture was 64-bit, but we don't work on 32-bit xnu kernels for a few years now; we can assume 64-bit safely. I am

[Lldb-commits] [PATCH] D147462: Use kernel's global variable indicating how many bits are used in addressing when loading Darwin xnu kernel

jasonmolenda added a comment. Yeah, I agree that messing with the global setting of addressable bits here is not ideal, but at this point in time the fact that we probably have an un-set/invalid value means it won't make things worse. Comment at: lldb/source/Plugins/DynamicL

[Lldb-commits] [PATCH] D147462: Use kernel's global variable indicating how many bits are used in addressing when loading Darwin xnu kernel

This revision was automatically updated to reflect the committed changes. Closed by commit rG8b092714c304: Using global variable in xnu kernel, set # of addressable bits (authored by jasonmolenda). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D147462

[Lldb-commits] [PATCH] D147674: Interpret ESR/FAR bits directly on watchpoint exceptions in debugserver, clarify how watchpoint descriptions in stop packets work

jasonmolenda created this revision. jasonmolenda added reviewers: labath, JDevlieghere, omjavaid, DavidSpickett. jasonmolenda added a project: LLDB. Herald added subscribers: atanasyan, kristof.beyls, arichardson, sdardis. Herald added a project: All. jasonmolenda requested review of this revision.

[Lldb-commits] [PATCH] D147674: Interpret ESR/FAR bits directly on watchpoint exceptions in debugserver, clarify how watchpoint descriptions in stop packets work

jasonmolenda added a comment. One possible criticism I have of the current reason:watchpoint + description-up-to-three-integers is that we should add key-value pairs to the stop-info packet `wp-address:` `wp-index:` `wp-hit-address` and honestly, `silently-continue:` instead of doing an archite

[Lldb-commits] [PATCH] D147482: [lldb] Add an overload to SetModuleLoadAddress that takes an unsigned value

jasonmolenda accepted this revision. jasonmolenda added a comment. This revision is now accepted and ready to land. LGTM. We can detect this This is one of those cases where I'd like to mark SBTarget::SetModuleLoadAddress as deprecated. I like that you detect this situation in SBTarget::SetLo

[Lldb-commits] [PATCH] D147816: Clarify how watchpoint description in stop packets work, fix AArch64 unintended behavior

jasonmolenda created this revision. jasonmolenda added reviewers: labath, omjavaid, DavidSpickett. jasonmolenda added a project: LLDB. Herald added subscribers: JDevlieghere, atanasyan, kristof.beyls, arichardson, sdardis. Herald added a project: All. jasonmolenda requested review of this revision

[Lldb-commits] [PATCH] D147820: debugserver: move AArch64 watchpoint traps within a watchpointed region, parse ESR flags and send them to lldb

jasonmolenda created this revision. jasonmolenda added a reviewer: JDevlieghere. jasonmolenda added a project: LLDB. Herald added subscribers: omjavaid, atanasyan, kristof.beyls, arichardson, sdardis. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subsc

[Lldb-commits] [PATCH] D147674: Interpret ESR/FAR bits directly on watchpoint exceptions in debugserver, clarify how watchpoint descriptions in stop packets work

jasonmolenda abandoned this revision. jasonmolenda added a comment. Intermixing generic/linux AArch64 lldb watchpoint StopInfo changes in with a change to debugserver made this a difficult patch to ask anyone to review in total. I've separated it into one phab for the lldb watchpoint StopInfo

[Lldb-commits] [PATCH] D147816: Clarify how watchpoint description in stop packets work, fix AArch64 unintended behavior

jasonmolenda added a comment. Thanks Omair! Yeah we have the same problem on Darwin - AArch64 doesn't require that the address we get on a watchpoint trap in the FAR register point within the address range we are watching. I never fixed this on Darwin so I know the failure mode - lldb says "t

[Lldb-commits] [PATCH] D147833: [lldb] Change return type of EventData::GetFlavor

jasonmolenda added a comment. Am I missing something, how does this work when we have uses like `event_data->GetFlavor() == TargetEventData::GetFlavorString()` - is this changing from a one-time construction of a ConstString to a construction of a ConstString every time it's called by implicit

[Lldb-commits] [PATCH] D147833: [lldb] Change return type of EventData::GetFlavor

jasonmolenda added a comment. In D147833#4252681 , @bulbazord wrote: > In D147833#4252651 , @jasonmolenda > wrote: > >> Am I missing something, how does this work when we have uses like >> `event_data->GetFlavor

[Lldb-commits] [PATCH] D147833: [lldb] Change return type of EventData::GetFlavor

jasonmolenda added a comment. idk maybe I'm over-thinking it, but this does make me a little uncomfortable. We have at least one instance where someone did `platform_sp->GetPluginName() == "ios-simulator"` in ClangExpressionSourceCode.cpp (probably to avoid a dependency on an an apple platform

[Lldb-commits] [PATCH] D147833: [lldb] Change return type of EventData::GetFlavor

jasonmolenda added a comment. In D147833#4253416 , @jasonmolenda wrote: > idk maybe I'm over-thinking it, but this does make me a little uncomfortable. > We have at least one instance where someone did > `platform_sp->GetPluginName() == "ios-simulator

[Lldb-commits] [PATCH] D147833: [lldb] Change return type of EventData::GetFlavor

jasonmolenda added a comment. In D147833#4253499 , @bulbazord wrote: > > Yes, we definitely do that in many places. I see your point though, comparing > const c-string pointers can get pretty tricky if somebody refactors this in > the future. In tha

[Lldb-commits] [PATCH] D147820: debugserver: move AArch64 watchpoint traps within a watchpointed region, parse ESR flags and send them to lldb

jasonmolenda added a comment. Thanks for the review. Will update the patch. Comment at: lldb/tools/debugserver/source/MacOSX/MachException.cpp:162-169 + if (exc_type == EXC_BREAKPOINT && exc_data[0] == EXC_ARM_DA_DEBUG && + exc_data.size() > 1) { +stop_info->reason =

[Lldb-commits] [PATCH] D147820: debugserver: move AArch64 watchpoint traps within a watchpointed region, parse ESR flags and send them to lldb

jasonmolenda added inline comments. Comment at: lldb/tools/debugserver/source/MacOSX/MachException.cpp:162-169 + if (exc_type == EXC_BREAKPOINT && exc_data[0] == EXC_ARM_DA_DEBUG && + exc_data.size() > 1) { +stop_info->reason = eStopTypeWatchpoint; +stop_info->detai

[Lldb-commits] [PATCH] D147820: debugserver: move AArch64 watchpoint traps within a watchpointed region, parse ESR flags and send them to lldb

jasonmolenda added inline comments. Comment at: lldb/test/API/commands/watchpoints/unaligned-watchpoint/main.c:8-11 + printf("%lld %lld\n", (*u64_p)++, (*u64_p)++); + printf("%lld %lld\n", (*u64_p)++, (*u64_p)++); + printf("%lld %lld\n", (*u64_p)++, (*u64_p)++); + printf(

[Lldb-commits] [PATCH] D147820: debugserver: move AArch64 watchpoint traps within a watchpointed region, parse ESR flags and send them to lldb

jasonmolenda updated this revision to Diff 512577. jasonmolenda added a comment. Update patch to incorporate Jonas' & David's feedback. I'm holding this patch until I can get the lldb changes to fix the reason:watchpoint `description` handling corrected, or the test case in this patch will fail

[Lldb-commits] [PATCH] D147816: Clarify how watchpoint description in stop packets work, fix AArch64 unintended behavior

jasonmolenda added inline comments. Comment at: lldb/docs/lldb-gdb-remote.txt:1609 +// There may be false-positive watchpoint hits on AArch64 as well, +// in the SVE Streaming Mode, but that is less common (v. ESR +// register flag "WPF", "Watchpoin

[Lldb-commits] [PATCH] D147816: Clarify how watchpoint description in stop packets work, fix AArch64 unintended behavior

jasonmolenda updated this revision to Diff 512610. jasonmolenda added a comment. Update the patch to incorporate David's first feedback. I know this is a bit of a tricky patch to review because it is 95% comments/documentation/renaming of variables, followed by the one key change of removing `c

[Lldb-commits] [PATCH] D147816: Clarify how watchpoint description in stop packets work, fix AArch64 unintended behavior

jasonmolenda added a comment. FWIW to show the motivation for this patch (which, again comes down to removing `core >= ArchSpec::eCore_arm_generic && core <= ArchSpec::eCore_arm_aarch64` when creating the StopInfo from the watchpoint description), the test case I wrote in https://reviews.llvm.o

[Lldb-commits] [PATCH] D147816: Clarify how watchpoint description in stop packets work, fix AArch64 unintended behavior

This revision was automatically updated to reflect the committed changes. Closed by commit rGe38b0fa83a93: Remove AArch64 out of MIPS watchpoint-skip, doc wp description (authored by jasonmolenda). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D147816

[Lldb-commits] [PATCH] D147820: debugserver: move AArch64 watchpoint traps within a watchpointed region, parse ESR flags and send them to lldb

This revision was automatically updated to reflect the committed changes. Closed by commit rGe76cfaca70be: AArch64 debugserver parse ESR register for watchpoints (authored by jasonmolenda). Changed prior to commit: https://reviews.llvm.org/D147820?vs=512577&id=513036#toc Repository: rG LLVM

[Lldb-commits] [PATCH] D148175: [lldb] Add `operator StringRef` to ConstString

jasonmolenda accepted this revision. jasonmolenda added a comment. This revision is now accepted and ready to land. In D148175#4263217 , @aprantl wrote: > I understand that the opposite direction is explicit because actual work is > being done. In this d

[Lldb-commits] [PATCH] D148603: Remove hardcoding of addressing bits in ABIMacOSX_arm64

jasonmolenda created this revision. jasonmolenda added a reviewer: JDevlieghere. jasonmolenda added a project: LLDB. Herald added subscribers: kristof.beyls, arichardson. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. Early in

[Lldb-commits] [PATCH] D148603: Remove hardcoding of addressing bits in ABIMacOSX_arm64

This revision was automatically updated to reflect the committed changes. Closed by commit rG53430bfd5c9d: Remove hardcoded address bits in ABIMacOS_arm64 (authored by jasonmolenda). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D148603/new/ https://

[Lldb-commits] [PATCH] D148715: [LLDB] Discard register flags where the size doesn't match the register

jasonmolenda accepted this revision. jasonmolenda added a comment. This revision is now accepted and ready to land. LGTM. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D148715/new/ https://reviews.llvm.org/D148715 __

[Lldb-commits] [PATCH] D148790: [LLDB] Don't print register fields when asked for a specific format

jasonmolenda accepted this revision. jasonmolenda added a comment. This revision is now accepted and ready to land. This seems fine to me; it's nice to have a way to quiet these when they get in the way, even if it's maybe not the first thing you'd look for, it's an uncommon enough use case that

[Lldb-commits] [PATCH] D149040: Refactor and generalize debugserver code for setting hardware watchpoints on AArch64

jasonmolenda created this revision. jasonmolenda added a reviewer: JDevlieghere. jasonmolenda added a project: LLDB. Herald added a subscriber: kristof.beyls. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. I am adding MASK (pow

[Lldb-commits] [PATCH] D149111: [lldb] Add support for specifying language when setting watchpoint by expression

jasonmolenda accepted this revision. jasonmolenda added a comment. This revision is now accepted and ready to land. LGTM. Comment at: lldb/source/Interpreter/OptionGroupWatchpoint.cpp:67 + "Number of bytes to use to watch a region."}, +{LLDB_OPT_SET_2, + false,

[Lldb-commits] [PATCH] D149224: Fix an off-by-one error with armv7 mach-o corefile register contexts (LC_THREADs)

jasonmolenda created this revision. jasonmolenda added reviewers: fixathon, JDevlieghere. jasonmolenda added a project: LLDB. Herald added subscribers: omjavaid, kristof.beyls. Herald added a project: All. jasonmolenda requested review of this revision. Herald added a subscriber: lldb-commits. Sla

[Lldb-commits] [PATCH] D149284: [lldb] Fix finding .Bundle directories in PlatformDarwinKernel

jasonmolenda added a comment. I added this in 2016 via 243bd763ecad466fac748b565ea52c661107684d and I have no idea what it is for. I know of no such prefix related to kernel debugging. Let's remove it instead. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews

<    1   2   3   4   5   6   7   8   9   10   >