Page MenuHomePhabricator
Feed Advanced Search

Yesterday

compnerd accepted D69076: [cmake] Pass external project source directories to sub-configures.

I’m not sure that this is the best direction long term. This is specific to the LLVM build, the runtimes builds do not take these parameters, and they are the primary things we’re trying to cross compile with this infrastructure. But the monorepo makes this more complicated and makes the configure the same, so, yes, I suppose this needs to be preserved. *sigh*

Thu, Oct 17, 11:58 AM · Restricted Project
compnerd accepted D69126: [Object] Fix the return type of getOffset/getSize.

I wish that we had a better type. For the time being, the 64-bit with is sufficient for current needs.

Thu, Oct 17, 11:40 AM · Restricted Project

Wed, Oct 16

compnerd added a comment to D69013: [AArch64][X86] Don't assume __powidf2 is available on Windows..

I do believe that we need it for all the targets (x86, x86_64, armv7, aarch64).

Wed, Oct 16, 8:37 AM · Restricted Project

Tue, Oct 15

compnerd accepted D68971: [libunwind][Android] Improve workaround for PIE zero-dlpi_addr bug.

Thanks, this should be more robust. Nice catch with the image base being explicitly set and using p_vaddr rather than p_offset, that was something that I failed to account for. I really wish that we had a more complete test suite, but that is *way* beyond this change.

Tue, Oct 15, 10:20 AM · Restricted Project
compnerd accepted D68966: [llvm-lipo] Add missing cast.

I don't think that generating a file of that size makes sense for the test case. Generating a small file and overwriting the file size would be viable, but in this case, we need to use the file size itself, so that doesn't work either.

Tue, Oct 15, 10:11 AM · Restricted Project

Mon, Oct 14

compnerd accepted D68915: [clang][IFS] Escape mangled name in-order to not break llvm-ifs with names mangled using MS ABI.
Mon, Oct 14, 10:07 AM · Restricted Project

Fri, Oct 11

compnerd accepted D68859: Fixing crash in clang IFS for enum, record, and typedef decls..
Fri, Oct 11, 7:55 AM · Restricted Project

Thu, Oct 10

compnerd added a comment to D68833: [CMake] Re-order runtimes in the order of dependencies.

I feel like Im not understanding something. The point of CMake is that it tracks dependencies. We are we manually tracking dependencies?

Thu, Oct 10, 8:25 PM · Restricted Project

Tue, Oct 8

Herald added a reviewer for D66904: [libunwind] Fix memory leak in handling of DW_CFA_remember_state and DW_CFA_restore_state: mclow.lists.
Tue, Oct 8, 9:04 AM · Restricted Project
compnerd added inline comments to D68614: [LLDB] Remove standalone build dep on llvm-strip.
Tue, Oct 8, 8:45 AM · Restricted Project

Thu, Oct 3

compnerd accepted D68319: [llvm-lipo] Relax the check of the specified input file architecture.
Thu, Oct 3, 9:27 AM · Restricted Project
compnerd accepted D68387: [libunwind] Adjust libunwind_01.pass.cpp test for ARM EHABI.

*sigh* I thought it used the size of the function to mark the end of the region. Thanks.

Thu, Oct 3, 9:24 AM · Restricted Project

Wed, Oct 2

compnerd accepted D68339: [llvm] [cmake] Add install targets for missing distribution components.
Wed, Oct 2, 8:48 AM · Restricted Project

Tue, Oct 1

compnerd accepted D63978: Clang Interface Stubs merger plumbing for Driver.
Tue, Oct 1, 9:43 PM · Restricted Project, Restricted Project

Mon, Sep 30

compnerd accepted D68196: [llvm-readobj/llvm-readelf] Delete --arm-attributes (alias for --arch-specific).

Thanks for cleaning this up

Mon, Sep 30, 1:12 PM · Restricted Project
compnerd committed rG95679409048c: build: serialise `LLVM_ENABLE_UNWIND_TABLES` into LLVMConfig (authored by compnerd).
build: serialise `LLVM_ENABLE_UNWIND_TABLES` into LLVMConfig
Mon, Sep 30, 1:04 PM
compnerd committed rL373253: build: serialise `LLVM_ENABLE_UNWIND_TABLES` into LLVMConfig.
build: serialise `LLVM_ENABLE_UNWIND_TABLES` into LLVMConfig
Mon, Sep 30, 1:02 PM
compnerd accepted D68192: Fix PR40710: Outlined Function has token parameter but isn't an intrinsic.
Mon, Sep 30, 8:31 AM · Restricted Project

Fri, Sep 27

compnerd accepted D68116: [llvm-lipo] Add support for -arch.
Fri, Sep 27, 9:24 AM · Restricted Project

Wed, Sep 25

compnerd added a comment to D66904: [libunwind] Fix memory leak in handling of DW_CFA_remember_state and DW_CFA_restore_state.

I think that using assembly for the test is fine

Wed, Sep 25, 10:10 AM · Restricted Project
compnerd added a comment to D67999: Fix `compiler_rt_logbf_test.c` test failure for Builtins-i386-darwin test suite..

@rupprecht as much as I really want to agree with you in that this isn't needed to match the behaviour of the system's libm, the problem is that __builtin_logb can be misused to implement a library. So, really, I would really urge Apple to consider maintaining similarity across all the platforms for the builtins - there is no official guarantee that is made on the behaviour of the builtin operations, so If there is a strong reason to have the divergence, please do share that, and I think that we can probably accommodate that with the small tweak that @rupprecht points out:

Wed, Sep 25, 8:49 AM · Restricted Project, Restricted Project

Tue, Sep 24

compnerd committed rG56eae602da92: Host: use the platform identifiers from LLVM (NFC) (authored by compnerd).
Host: use the platform identifiers from LLVM (NFC)
Tue, Sep 24, 3:56 PM
compnerd committed rL372788: Host: use the platform identifiers from LLVM (NFC).
Host: use the platform identifiers from LLVM (NFC)
Tue, Sep 24, 3:56 PM
compnerd accepted D67954: [LLDB] [Windows] Initial support for ARM64 register contexts.

Honestly, this is just setting up the register context for ARM64. I dont think that there is much of a test for this. I mean, I suppose you could test this by instantiating the context and trying to read it through the interface. But, I question the value of such a test. Whether you go with the in-process or out-of-process approach and whether you are doing DWARF of CodeView debugging this is going to be needed. As to running the test suite - you can cross-compile and run the tests remotely.

Tue, Sep 24, 8:31 AM · Restricted Project, Restricted Project
compnerd added inline comments to D63978: Clang Interface Stubs merger plumbing for Driver.
Tue, Sep 24, 8:07 AM · Restricted Project, Restricted Project

Mon, Sep 23

compnerd committed rG082f895b1a9b: HotColdSplitting: invalidate the AssumptionCache on split (authored by compnerd).
HotColdSplitting: invalidate the AssumptionCache on split
Mon, Sep 23, 3:28 PM
compnerd committed rL372667: HotColdSplitting: invalidate the AssumptionCache on split.
HotColdSplitting: invalidate the AssumptionCache on split
Mon, Sep 23, 3:24 PM
compnerd closed D67936: HotColdSplitting: invalidate the AssumptionCache on split.

r372667

Mon, Sep 23, 3:24 PM · Restricted Project
compnerd accepted D67911: [LLDB] [Windows] Add missing ifdefs to fix building for non-x86 architectures.
Mon, Sep 23, 3:09 PM · Restricted Project, Restricted Project
compnerd created D67936: HotColdSplitting: invalidate the AssumptionCache on split.
Mon, Sep 23, 2:44 PM · Restricted Project
compnerd added a comment to D67911: [LLDB] [Windows] Add missing ifdefs to fix building for non-x86 architectures.

I think that printf is quite an amazing notification :-)

Mon, Sep 23, 1:35 PM · Restricted Project, Restricted Project
compnerd added a comment to D67911: [LLDB] [Windows] Add missing ifdefs to fix building for non-x86 architectures.

What do you think of adding some sort of notification that hardware breakpoints are currently unsupported and that we are falling back to software breakpoints for the #else case?

Mon, Sep 23, 9:03 AM · Restricted Project, Restricted Project
compnerd accepted D67913: [LLDB] [Windows] Map COFF ARM machine ids to the right triplet architectures.

Nit: triple, not triplet.

Mon, Sep 23, 9:03 AM · Restricted Project, Restricted Project
compnerd accepted D67912: [LLDB] [PECOFF] Recognize arm64 executables.
Mon, Sep 23, 9:03 AM · Restricted Project, Restricted Project

Sat, Sep 21

compnerd added a comment to D67873: Allow llvm-objdump to run on executable files.

Please post patches with context.

Sat, Sep 21, 7:00 AM · Restricted Project

Fri, Sep 20

compnerd accepted D67863: [LLDB] Cast -1 (as invalid socket) to the socket type before comparing.
Fri, Sep 20, 5:44 PM · Restricted Project, Restricted Project
compnerd accepted D67758: [llvm-lipo] Add support for archives.
Fri, Sep 20, 3:30 PM · Restricted Project
compnerd added inline comments to D67863: [LLDB] Cast -1 (as invalid socket) to the socket type before comparing.
Fri, Sep 20, 2:36 PM · Restricted Project, Restricted Project
compnerd accepted D67862: [LLDB] Use SetErrorStringWithFormatv for cases that use LLVM style format strings.
Fri, Sep 20, 2:36 PM · Restricted Project, Restricted Project
compnerd committed rG0ae02fd71a38: Unwind: avoid warning about unused typedef (authored by compnerd).
Unwind: avoid warning about unused typedef
Fri, Sep 20, 1:47 PM
compnerd committed rL372427: Unwind: avoid warning about unused typedef.
Unwind: avoid warning about unused typedef
Fri, Sep 20, 1:45 PM
compnerd closed D67804: [libunwind] Avoid warning about unused typedef.

SVN r372427

Fri, Sep 20, 1:45 PM · Restricted Project
compnerd accepted D67858: [LLDB] Check for the GCC/MinGW compatible arch defines for windows, in addition to MSVC defines.
Fri, Sep 20, 1:44 PM · Restricted Project, Restricted Project
compnerd added inline comments to D67758: [llvm-lipo] Add support for archives.
Fri, Sep 20, 9:02 AM · Restricted Project
compnerd committed rGae685e7aab7f: Unwind: prevent unw_get_proc_info from returning stale data (authored by compnerd).
Unwind: prevent unw_get_proc_info from returning stale data
Fri, Sep 20, 8:53 AM
compnerd committed rL372407: Unwind: prevent unw_get_proc_info from returning stale data.
Unwind: prevent unw_get_proc_info from returning stale data
Fri, Sep 20, 8:52 AM
compnerd closed D66889: [libunwind] Fix unw_get_proc_info sometimes returning stale data.

SVN r372407

Fri, Sep 20, 8:52 AM
compnerd accepted D67804: [libunwind] Avoid warning about unused typedef.

Sure, this seems reasonable.

Fri, Sep 20, 8:47 AM · Restricted Project

Wed, Sep 18

compnerd accepted D67700: [Object] Extend MachOUniversalBinary::getObjectForArch.

Okay, both @mtrent and @alexshap bring up good points. I *suppose* that is okay, but I shall string grumble. @mtrent, I think we could also simply make getObjectForArch be internal to the type, and force users to go down either getMachOArchiveForArch or getMachOObjectForArch.

Wed, Sep 18, 3:00 PM · Restricted Project
compnerd committed rG8f16cc46f7d8: unwind: remove a could of extraneous `else` (NFC) (authored by compnerd).
unwind: remove a could of extraneous `else` (NFC)
Wed, Sep 18, 9:16 AM
compnerd committed rL372233: unwind: remove a could of extraneous `else` (NFC).
unwind: remove a could of extraneous `else` (NFC)
Wed, Sep 18, 9:16 AM
compnerd accepted D66889: [libunwind] Fix unw_get_proc_info sometimes returning stale data.

Would be nice if we could have a test case for this.

Wed, Sep 18, 9:16 AM
compnerd added inline comments to D67700: [Object] Extend MachOUniversalBinary::getObjectForArch.
Wed, Sep 18, 9:07 AM · Restricted Project

Sep 13 2019

compnerd added inline comments to D63978: Clang Interface Stubs merger plumbing for Driver.
Sep 13 2019, 4:29 PM · Restricted Project, Restricted Project
compnerd committed rL371908: request commit access for compnerd.
request commit access for compnerd
Sep 13 2019, 4:18 PM

Sep 11 2019

compnerd added a comment to D58702: Enabled Backtraces with just ENABLE_BACKTRACES.

@yifeihe, you should be able to build with backtraces disabled to get the previous behaviour. The fact that the behaviour was a certain way previously was just an accident, not an explicit design point.

Sep 11 2019, 10:52 AM · Restricted Project

Sep 9 2019

compnerd added a comment to D67195: Adding support for overriding LLVM_ENABLE_RUNTIMES for runtimes builds..

@beanz - I must be mistaken about something. The condition is if LLVM_ENABLE_RUNTIMES is NOT specified for the current target, we pass in the global value (which was previously passed via the PASSTHROUGH_PREFIXES), and if it is specified, we will pass in the specified value. So, the only behaviour change that is visible is if the user specified a target specific LLVM_ENABLE_RUNTIMES value it is now honoured.

Sep 9 2019, 4:41 PM · Restricted Project

Sep 4 2019

compnerd added a comment to D67195: Adding support for overriding LLVM_ENABLE_RUNTIMES for runtimes builds..

@beanz - I can see where you are coming from there. But, the thing is that when you are building for a wide variety of targets at once, that doesn't make sense since not all the runtimes are compatible with all targets (e.g. unwind doesn't support Windows). Calculating the set difference and passing that along seems more complicated, though plausible. Note that this approach preserves the existing behaviour - unless you *explicitly* state the runtimes for a specific target, the LLVM_ENABLE_RUNTIMES passed globally will continue to be passed through, so from a user's perspective nothing has changed (and that is intentional - no one should need to update their CMake caches for this change).

Sep 4 2019, 5:09 PM · Restricted Project
compnerd added a comment to D66970: [libc++] Explicitly provide a list of exported symbols for libc++.

I think that controlling via annotations is actually very useful. It will actually still be needed to ensure that libc++ continues to build on Windows properly (which really does want the annotations). The annotations being applied there makes it easy to then handle properly on the other platforms as well. Additionally, diffs explicitly call out the interface in the change, which again helps review the changes. Finally, there are some symbols which are synthesized by the compiler and the visibility for those are controlled by means of the annotation on the associated class. This again is far easier to manage with the explicit annotation rather than the explicit export list.

Sep 4 2019, 12:52 PM · Restricted Project

Sep 2 2019

compnerd added a comment to D67061: [mir-canon][NFCi] Adding opt arg to enable vreg renaming only mode..

This is not a NFCI change - it is adding new functionality. Does it make sense perhaps to have this be a separate pass?

Sep 2 2019, 2:27 PM · Restricted Project

Aug 30 2019

compnerd added a comment to D66834: Driver tests: set `--sysroot=""` to support clang with `DEFAULT_SYSROOT`.

I think that this is pretty easy to forget. Fortunately, last argument wins. Why not sink this into the %clang substitution in lit? That ensures that we run with an empty sysroot and then when the test needs to adjust the sysroot, it can do so explicitly.

Aug 30 2019, 9:02 AM · Restricted Project, Restricted Project
compnerd accepted D66405: [llvm-ifs] llvm Interface Stubs merging + object file generation tool..
Aug 30 2019, 8:55 AM · Restricted Project

Aug 29 2019

compnerd committed rGbe638099a476: DebugInfo: add CodeView register mapping for ARM NT (authored by compnerd).
DebugInfo: add CodeView register mapping for ARM NT
Aug 29 2019, 5:17 PM
compnerd committed rL370423: DebugInfo: add CodeView register mapping for ARM NT.
DebugInfo: add CodeView register mapping for ARM NT
Aug 29 2019, 5:17 PM
compnerd added a comment to D66405: [llvm-ifs] llvm Interface Stubs merging + object file generation tool..

Please rename the tests to be more meaningful so that others can possibly identify what they are meant to test and can address things which may break when other changes are made.

Aug 29 2019, 12:39 PM · Restricted Project

Aug 27 2019

compnerd created D66858: POSIX DYLD: add workaround for android L loader.
Aug 27 2019, 6:33 PM · Restricted Project
compnerd added a comment to D66405: [llvm-ifs] llvm Interface Stubs merging + object file generation tool..

I think it may make some sense to strip the user label prefix from the symbols that are listed in the interface, and then add that in when generating the ELF/TBD.

Aug 27 2019, 5:05 PM · Restricted Project

Aug 26 2019

compnerd added a comment to D66405: [llvm-ifs] llvm Interface Stubs merging + object file generation tool..

I suppose that the test cases that I think are valuable:

  • 1 file
  • 1+ file
  • functions
  • objects
  • functions w/ different alignments
  • objects w/ different alignments
  • functions w/ different sizes
  • objects w/ different sizes
  • functions w/ different alignments, sizes
  • objects w/ different alignments, sizes
  • weak symbols
  • mismatch (error cases)
Aug 26 2019, 5:13 PM · Restricted Project
compnerd added a comment to D66405: [llvm-ifs] llvm Interface Stubs merging + object file generation tool..

Probably is worth having a test for the error case as well.

Aug 26 2019, 5:10 PM · Restricted Project

Aug 23 2019

compnerd added inline comments to D66405: [llvm-ifs] llvm Interface Stubs merging + object file generation tool..
Aug 23 2019, 1:21 PM · Restricted Project
compnerd added a comment to D66405: [llvm-ifs] llvm Interface Stubs merging + object file generation tool..

This definitely needs tests.

Aug 23 2019, 1:21 PM · Restricted Project
compnerd committed rGe7823a53164b: Windows: explicitly cast constants to `DWORD` (authored by compnerd).
Windows: explicitly cast constants to `DWORD`
Aug 23 2019, 11:04 AM
compnerd committed rL369788: Windows: explicitly cast constants to `DWORD`.
Windows: explicitly cast constants to `DWORD`
Aug 23 2019, 11:03 AM
compnerd added a reviewer for D66648: [compiler-rt] Respect CMAKE_C_FLAGS when building builtins in-tree: beanz.
Aug 23 2019, 11:03 AM · Restricted Project, Restricted Project
compnerd closed D66445: Explicitly Cast Constants to DWORD.

SVN r369788

Aug 23 2019, 11:03 AM · Restricted Project
compnerd added a comment to D66648: [compiler-rt] Respect CMAKE_C_FLAGS when building builtins in-tree.

Im not sure if this works. This sets the flags from the global flags, which may not be what we want as compiler-rt can be built as part of llvm itself.

Aug 23 2019, 11:03 AM · Restricted Project, Restricted Project

Aug 22 2019

compnerd accepted D66573: [clang][ifs] Dropping older experimental interface stub formats..

Please add a diagnostic to indicate that the provided value is invalid for the argument (or at at least extend the existing the tests to ensure that the old values are diagnosed in the driver rather hitting the assert in the frontend.

Aug 22 2019, 10:23 AM · Restricted Project, Restricted Project
compnerd accepted D66446: [clang][IFS] Adding new Interface Stubs format. .

Please add a test case for read/write, otherwise LGTM.

Aug 22 2019, 10:15 AM · Restricted Project, Restricted Project

Aug 19 2019

compnerd committed rG2e8b57558df2: Windows: Include "windows" Instead of "Windows" (authored by compnerd).
Windows: Include "windows" Instead of "Windows"
Aug 19 2019, 3:48 PM
compnerd committed rL369307: Windows: Include "windows" Instead of "Windows".
Windows: Include "windows" Instead of "Windows"
Aug 19 2019, 3:47 PM
compnerd closed D66448: Include "windows" Instead of "Windows".

SVN r369307

Aug 19 2019, 3:47 PM · Restricted Project
compnerd accepted D66445: Explicitly Cast Constants to DWORD.
Aug 19 2019, 3:39 PM · Restricted Project
compnerd accepted D66448: Include "windows" Instead of "Windows".
Aug 19 2019, 3:37 PM · Restricted Project

Aug 16 2019

compnerd added a comment to D65983: Autogenerate the shebang lines for tools/opt-viewer.

Python eggs may be another option which might be an easier way forward.

Aug 16 2019, 3:44 PM · Restricted Project
compnerd added a comment to D65983: Autogenerate the shebang lines for tools/opt-viewer.

At least for the from-source distribution, where the destination of the image is the build host, Python_EXECUTABLE is definitely preferable (e.g. #!@Python_EXECUTABLE@). Furthermore, there is no reason to assume that /usr/bin/env python3 will give you the same python 3 as what was given to cmake (you can control that with -DPython_EXECUTABLE=...). I think that requiring the user to specify the tool is possibly the best option - that is completely remove the shebang as that ensures that shifts the burden of selection to the user. Another option may be to use something like p2exe and create self-contained tools.

Aug 16 2019, 3:42 PM · Restricted Project

Aug 13 2019

compnerd added a comment to D66058: [NFC][clang] Move much of the argument handling code from Driver::BuildActions to Driver::handleArguments..

Ugh, this is still not the most structured handling of the arguments. But, yeah, this seems like it should be equivalent. Fine by me if @aaron.ballman has no more comments.

Aug 13 2019, 2:59 PM · Restricted Project, Restricted Project
compnerd added a comment to D63978: Clang Interface Stubs merger plumbing for Driver.

Tests?

Aug 13 2019, 2:49 PM · Restricted Project, Restricted Project

Aug 10 2019

compnerd added inline comments to D65993: [NFC][clang] Adding argument based Phase list filtering to getComplicationPhases.
Aug 10 2019, 9:35 PM · Restricted Project, Restricted Project

Aug 9 2019

compnerd committed rGa5af238343e7: CodeGen: ensure 8-byte aligned String Swift CF ABI (authored by compnerd).
CodeGen: ensure 8-byte aligned String Swift CF ABI
Aug 9 2019, 12:30 PM
compnerd committed rL368471: CodeGen: ensure 8-byte aligned String Swift CF ABI.
CodeGen: ensure 8-byte aligned String Swift CF ABI
Aug 9 2019, 12:28 PM

Aug 8 2019

compnerd accepted D65969: [clang][NFC] Consolidating usage of "FinalPhase" in Driver::BuildActions.

Seems like it should be equivalent.

Aug 8 2019, 2:16 PM · Restricted Project, Restricted Project

Aug 7 2019

compnerd added inline comments to D65863: [ARM] Add support for the s,j,x,N,O inline asm constraints.
Aug 7 2019, 3:19 PM · Restricted Project, Restricted Project
compnerd added a comment to D65831: libc++: add `-include` or `/FI` to the interface compile definitions.

Hmm, yes, it is for use from the build tree. I don't see how INTERFACE_INCLUDE_DIRECTORIES can do a -include though. It has to be INTERFACE_COMPILE_OPTIONS or INTERFACE_COMPILE_DEFINITIONS.

Aug 7 2019, 1:29 PM

Aug 6 2019

compnerd added inline comments to D65839: [Driver] Add verbatim dry run option.
Aug 6 2019, 8:37 PM · Restricted Project
compnerd created D65831: libc++: add `-include` or `/FI` to the interface compile definitions.
Aug 6 2019, 3:07 PM
compnerd added a comment to D65829: [clang][DirectoryWatcher][NFC] Swapping asserts for llvm fatal_error in ::create..

@lhames - I like the logAllUnhandledErrors!

Aug 6 2019, 2:55 PM · Restricted Project, Restricted Project
compnerd added a comment to D65829: [clang][DirectoryWatcher][NFC] Swapping asserts for llvm fatal_error in ::create..

@gribozavr I think that this usage here is actually useful because it
a) tests the actual behaviour
b) provides example code for other users

Aug 6 2019, 2:44 PM · Restricted Project, Restricted Project
compnerd accepted D65676: [llvm-lipo] Update llvm-lipo docs for -info -thin -create -replace -segalign flags.
Aug 6 2019, 1:22 PM · Restricted Project
compnerd added inline comments to D65798: [lldb][CMake] Infer `Clang_DIR` if not passed explicitly.
Aug 6 2019, 1:08 PM · Restricted Project, Restricted Project
compnerd added inline comments to D65798: [lldb][CMake] Infer `Clang_DIR` if not passed explicitly.
Aug 6 2019, 8:26 AM · Restricted Project, Restricted Project