Author: labath
Date: Thu May 10 03:46:03 2018
New Revision: 331970
URL: http://llvm.org/viewvc/llvm-project?rev=331970=rev
Log:
Convert all RunShellCommand functions to use the Timeout class
this completes the Timeout migration started in r331880 with the
Predicate class.
Modified:
labath added a comment.
The test you copied this from predates `lldb-test` (and it tests some behavior
quirks which are not nicely expressible in text output). We should have nicer
ways of testing things like this now. It should be sufficient to add a print
line which displays the triple in
labath created this revision.
labath added reviewers: zturner, clayborg.
now that llvm supports host-agnostic path manipulation functions (and
most of their kinks have been ironed out), we can remove our copies of
the path parsing functions in favour of the llvm ones.
This should be NFC except
Author: labath
Date: Thu May 10 01:59:17 2018
New Revision: 331966
URL: http://llvm.org/viewvc/llvm-project?rev=331966=rev
Log:
CPlusPlusLanguage: Add unit tests for the FindAlternateFunctionManglings method
I was considering modifying this function, so I wrote some tests to make
sure I don't
labath added a comment.
The patch looks much better now, but I think we still need to discuss the data
extractor sliding issue, as right now that's very hacky.
https://reviews.llvm.org/D32167
___
lldb-commits mailing list
probinson added a comment.
So, I've been attacking the type-units-in-.debug_info problem in LLVM, and what
I've done (so far) in the LLVM parser is to factor out a DWARFUnitHeader class,
and made DWARFUnit derive from that. This makes it feasible to parse a unit
header before deciding what
On Thu, 10 May 2018 at 16:19, wrote:
> > Anyway, the issue is that the PS4 targets don't emit DW_AT_linkage_name
> > attributes in the debug info, but we are still emitting an index entry
> > with
> > the mangled names. This caused the verifier to complain.
> >
> > Two
> Anyway, the issue is that the PS4 targets don't emit DW_AT_linkage_name
> attributes in the debug info, but we are still emitting an index entry
> with
> the mangled names. This caused the verifier to complain.
>
> Two interesting questions follow from this:
> 1. (mainly for Paul, I guess)
clayborg added inline comments.
Comment at: source/Utility/FileSpec.cpp:277-281
+ auto style = LLVMPathSyntax(syntax);
+ m_filename.SetString(llvm::sys::path::filename(resolved, style));
+ llvm::StringRef dir = llvm::sys::path::parent_path(resolved, style);
+ if
clayborg added a comment.
I really need to get this in as well. It is a very important feature and I am
trying to do the right thing and get it into open source so we don't end up
keeping the patch locally. So any help on expediting any comments and
iterations would be great.
clayborg added a comment.
In https://reviews.llvm.org/D32167#1094173, @labath wrote:
> The patch looks much better now, but I think we still need to discuss the
> data extractor sliding issue, as right now that's very hacky.
It makes things just work with .debug_info as is pretends that the
labath added inline comments.
Comment at: source/Utility/FileSpec.cpp:277-281
+ auto style = LLVMPathSyntax(syntax);
+ m_filename.SetString(llvm::sys::path::filename(resolved, style));
+ llvm::StringRef dir = llvm::sys::path::parent_path(resolved, style);
+ if (!dir.empty())
polyakov.alex added a comment.
So, what do you think about the patch? There is no any decision about lldb-mi
tests and I do not know what to do with this.
Repository:
rL LLVM
https://reviews.llvm.org/D46588
___
lldb-commits mailing list
aprantl updated this revision to Diff 146234.
aprantl added a comment.
Much smaller test thanks to Pavel's suggestion!
https://reviews.llvm.org/D46669
Files:
lit/Modules/lc_version_min.yaml
source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp
Index:
aprantl updated this revision to Diff 146235.
aprantl added a comment.
Added a REQUIRES: darwin
https://reviews.llvm.org/D46669
Files:
lit/Modules/lc_version_min.yaml
source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp
Index: source/Plugins/ObjectFile/Mach-O/ObjectFileMachO.cpp
compnerd created this revision.
compnerd added reviewers: davide, labath, vsk, zturner.
Herald added a subscriber: mgorny.
Use proper cmake techniques to detect where the libedit package resides.
This allows for the use of libedit from an alternative location which is
needed for supporting
aprantl added a comment.
The code looks good — I think it would be worthwhile to try and convert this
test to a LIT test. You can try to model it after
`lit/Expr/TestCallUserDefinedFunction.test`. You will also need to modify
`lit/lit.cfg` to define a `%lldb-mi` variable, but that should be
aprantl added inline comments.
Comment at:
packages/Python/lldbsuite/test/tools/lldb-mi/breakpoint/TestMiBreak.py:157
+self.runCmd("-break-insert main")
+# FIXME function name is unknown on Darwin, fullname should be ??,
line is -1
+self.expect(
clayborg requested changes to this revision.
clayborg added inline comments.
This revision now requires changes to proceed.
Comment at: source/Plugins/Platform/POSIX/PlatformPOSIX.cpp:1046-1047
+ {
+static std::mutex do_dlopen_mutex;
+std::lock_guard
friss added inline comments.
Comment at: source/Plugins/Platform/POSIX/PlatformPOSIX.cpp:1046-1047
+ {
+static std::mutex do_dlopen_mutex;
+std::lock_guard lock(do_dlopen_mutex);
+
clayborg wrote:
> We should put the mutex, or better yet a
clayborg added inline comments.
Comment at: source/Plugins/Platform/POSIX/PlatformPOSIX.cpp:1047
+static std::mutex do_dlopen_mutex;
+std::lock_guard lock(do_dlopen_mutex);
+
Accessor would be fine. The other reason for putting this in the process is
aprantl created this revision.
aprantl added reviewers: friss, jingham.
Herald added a subscriber: mgorny.
Inside Xcode and in Xcode toolchains LLDB is always in lockstep
with the Swift compiler, so it can reuse its Clang resource
directory. This allows LLDB and the Swift compiler to share the
friss created this revision.
friss added a reviewer: jingham.
Herald added a subscriber: emaste.
Multiple threads could be calling into DoLoadImage concurrently,
only one should be allowed to create the UtilityFunction.
https://reviews.llvm.org/D46733
Files:
jingham accepted this revision.
jingham added a comment.
That looks fine to me.
https://reviews.llvm.org/D46733
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
jasonmolenda accepted this revision.
jasonmolenda added a comment.
This revision is now accepted and ready to land.
Looks good.
https://reviews.llvm.org/D46669
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
Author: adrian
Date: Thu May 10 17:41:20 2018
New Revision: 332067
URL: http://llvm.org/viewvc/llvm-project?rev=332067=rev
Log:
Retrieve the deployment target when retrieving an object file's triple.
Getting the deployment target can be significant information when
rebuilding clang modules since
This revision was automatically updated to reflect the committed changes.
Closed by commit rL332067: Retrieve the deployment target when retrieving an
object files triple. (authored by adrian, committed by ).
Herald added a subscriber: llvm-commits.
Changed prior to commit:
friss updated this revision to Diff 146268.
friss added a comment.
I heted the idea of hosting the once_flag in Process but using
it in PlatformPOSIX. I did a bigger refactoring where platforms
pass a factory lambda to the accessor and the thread-safe init
is done in the Process class itself
probinson added a comment.
I can't coment on the xcode and python parts. The C++ all looks plausible.
I think the "sliding" trick will be trickier in the LLVM parser, as it has to
handle .o files (which might have multiple instances of .debug_types in v4, and
multiple instances of .debug_info
29 matches
Mail list logo