Summary:
Explicitly deleting the copy constructor makes compiling the function
`ento::registerGenericTaintChecker` difficult with some compilers. When we
construct an `llvm::Optional<TaintConfig>`, the optional is constructed with a
const TaintConfig reference which it then uses to invoke the deleted TaintConfig
copy constructor.
I've observered this failing with clang 3.8 on Ubuntu 16.04.
Reviewers: compnerd, Szelethus, boga95, NoQ, alexshap
Subscribers: xazax.hun, baloghadamsoftware, szepet, a.sidorin, mikhail.ramalho, donat.nagy, dkrupp, Charusso, llvm-commits, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D66192
llvm-svn: 368779
When we're tracking a variable that is responsible for a null pointer
dereference or some other sinister programming error, we of course would like to
gather as much information why we think that the variable has that specific
value as possible. However, the newly introduced condition tracking shows that
tracking all values this thoroughly could easily cause an intolerable growth in
the bug report's length.
There are a variety of heuristics we discussed on the mailing list[1] to combat
this, all of them requiring to differentiate in between tracking a "regular
value" and a "condition".
This patch introduces the new `bugreporter::TrackingKind` enum, adds it to
several visitors as a non-optional argument, and moves some functions around to
make the code a little more coherent.
[1] http://lists.llvm.org/pipermail/cfe-dev/2019-June/062613.html
Differential Revision: https://reviews.llvm.org/D64270
llvm-svn: 368777
Summary: Thanks to Hui Huang and reviewers for all the help with this patch!
Reviewers: labath, jfb, clayborg
Reviewed By: labath
Subscribers: Hui, clayborg, dexonsmith, lldb-commits
Tags: #lldb
Differential Revision: https://reviews.llvm.org/D61687
llvm-svn: 368776
Summary:
This commit contains three small changes to enable lldb-server on Windows.
- Add lldb-server for Windows to the build
- Disable pty redirection on Windows for the initial lldb-server bring up
- Add a support to get the parent pid for a process on Windows
- Ifdef some signals which aren't supported on Windows
Thanks to Hui Huang for the help with this patch!
Reviewers: labath
Reviewed By: labath
Subscribers: JDevlieghere, compnerd, Hui, amccarth, xiaobai, srhines, mgorny, lldb-commits
Tags: #lldb
Differential Revision: https://reviews.llvm.org/D61686
llvm-svn: 368774
Summary:
The following code snippet taken from D64271#1572188 has an issue: namely,
because `flag`'s value isn't undef or a concrete int, it isn't being tracked.
int flag;
bool coin();
void foo() {
flag = coin();
}
void test() {
int *x = 0;
int local_flag;
flag = 1;
foo();
local_flag = flag;
if (local_flag)
x = new int;
foo();
local_flag = flag;
if (local_flag)
*x = 5;
}
This, in my opinion, makes no sense, other values may be interesting too.
Originally added by rC185608.
Differential Revision: https://reviews.llvm.org/D64287
llvm-svn: 368773
My previous change didn't fix the Windows bot. This patch is an attempt
to make guessing the path style more robust by first looking at the
compile dir and falling back to the actual file if that's unsuccessful.
llvm-svn: 368772
During the evaluation of D62883, I noticed a bunch of totally
meaningless notes with the pattern of "Calling 'A'" -> "Returning value"
-> "Returning from 'A'", which added no value to the report at all.
This patch (not only affecting tracked conditions mind you) prunes
diagnostic messages to functions that return a value not constrained to
be 0, and are also linear.
Differential Revision: https://reviews.llvm.org/D64232
llvm-svn: 368771
They're useful when trying to understand what's going on
inside your LazyCompoundValues.
Differential Revision: https://reviews.llvm.org/D65427
llvm-svn: 368769
When -trim-egraph is unavailable (say, when you're debugging a crash on
a real-world code that takes too long to reduce), it makes sense to view
the untrimmed graph up to the crashing node's predecessor, then dump the ID
(or a pointer) of the node in the attached debugger, and then trim
the dumped graph in order to keep only paths from the root to the node.
The newly added --to flag does exactly that:
$ exploded-graph-rewriter.py ExprEngine.dot --to 0x12229acd0
Multiple nodes can be specified. Stable IDs of nodes can be used
instead of pointers.
Differential Revision: https://reviews.llvm.org/D65345
llvm-svn: 368768
Explorers aren't the right abstraction. For the purposes of displaying svg files
we don't care in which order do we explore the nodes. We may care about this for
other analyses, but we're not there yet.
The function of cutting out chunks of the graph is performed poorly by
the explorers, because querying predecessors/successors on the explored nodes
yields original successors/predecessors even if they aren't being explored.
Introduce a new entity, "trimmers", that do one thing but to it right: cut out
chunks of the graph. Trimmers mutate the graph, so stale edges aren't even
visible to their consumers in the pipeline. Additionally, trimmers are
intrinsically composable: multiple trimmers can be applied to the graph
sequentially.
Refactor the single-path explorer into the single-path trimmer.
Rename the test file for consistency.
Differential Revision: https://reviews.llvm.org/D65344
llvm-svn: 368767
Change the default behavior: the tool no longer dumps the rewritten .dot file
to stdout, but instead it automatically converts it into an .html file
(which essentially wraps an .svg file) and immediately opens it with
the default web browser.
This means that the tool should now be fairly easy to use:
$ exploded-graph-rewriter.py /tmp/ExprEngine.dot
The benefits of wrapping the .svg file into an .html file are:
- It'll open in a web browser, which is the intended behavior.
An .svg file would be open with an image viewer/editor instead.
- It avoids the white background around the otherwise dark svg area
in dark mode.
The feature can be turned off by passing a flag '--rewrite-only'.
The LIT substitution is updated to enforce the old mode because
we don't want web browsers opening on our buildbots.
Differential Revision: https://reviews.llvm.org/D65250
llvm-svn: 368766
r367979 changed DirectoryWatcher::Create to return an llvm::Expected.
Adjust the Windows stub accordingly.
(upstreamed from github.com/apple/swift-clang)
llvm-svn: 368762
This is just a code skeleton for DirectoryWatcher-windows.cpp so the
build on Windows stops breaking.
(upstreamed from github.com/apple/swift-clang)
llvm-svn: 368761
Summary: Thanks to Hui Huang and the reviewers for all the help with this patch.
Reviewers: labath, Hui, jfb, clayborg, amccarth
Reviewed By: labath
Subscribers: amccarth, compnerd, dexonsmith, mgorny, jfb, teemperor, lldb-commits
Tags: #lldb
Differential Revision: https://reviews.llvm.org/D63165
llvm-svn: 368759
I am changing this to work around an issue that is being hit when
building with clang 3.8. Specifically, clang 3.8 requires that we have a user
defined default constructor for SectionRef for the default initialization of a
const SectionRef.
llvm-svn: 368758
Factor out commonly-used target code from the GlobalISelEmitter tests into
a GlobalISelEmitterCommon.td file. This is tested by the original
GlobalISelEmitter.td test.
This reduces the amount of boilerplate code necessary for tests like this.
Differential Revision: https://reviews.llvm.org/D65777
llvm-svn: 368757
I feel this is kinda important, because in a followup patch I'm adding different
kinds of interestingness, and propagating the correct kind in BugReporter.cpp is
just one less thing to worry about.
Differential Revision: https://reviews.llvm.org/D65578
llvm-svn: 368755
Apparently this does literally nothing.
When you think about this, it makes sense. If something is really important,
we're tracking it anyways, and that system is sophisticated enough to mark
actually interesting statements as such. I wouldn't say that it's even likely
that subexpressions are also interesting (array[10 - x + x]), so I guess even
if this produced any effects, its probably undesirable.
Differential Revision: https://reviews.llvm.org/D65487
llvm-svn: 368752
Summary:
[libomptarget] Factor architecture dependent code out of loop.cu
Related to the patch series starting D64217. Added subscribers to said series as reviewers. This effort is smaller in scope.
This patch factors out just enough architecture dependent code from loop.cu to allow the same source to be used with amdgcn, given a different target_impl.h. Testing is that the same bitcode (modulo variable names) is generated for libomptarget before and after the refactor, for nvptx and the out of tree amdgcn.
Reviewers: jdoerfert, ABataev, bollu, jfb, tra, grokos, Hahnfeld, guansong, xtian, gregrodgers, ronlieb, hfinkel, gtbercea, guraypp, arpith-jacob
Reviewed By: jdoerfert, ABataev
Subscribers: dexonsmith, openmp-commits
Tags: #openmp
Differential Revision: https://reviews.llvm.org/D65836
llvm-svn: 368751
Summary:
In the WebAssembly backend, when lowering variadic function calls, non-single
member aggregate type arguments are always passed by pointer.
However, when emitting va_arg code in clang, the arguments are instead read as
if they are passed directly. This results in the pointer being read as the
actual structure.
Fixes https://github.com/emscripten-core/emscripten/issues/9042.
Reviewers: tlively, sbc100, kripken, aheejin, dschuff
Reviewed By: dschuff
Subscribers: dschuff, jgravelle-google, sunfish, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D66168
llvm-svn: 368750
This patch implements global address lowering for 32/64 bit with small/large code models.
1.For 32bit large code model on AIX, there are newly added pseudo opcode LWZtocL & ADDIStocHA32, the support of which on MC layer will be
provided by future patches.
2.The default code model on AIX should be small code model.
3.Since AIX does not have medium code model, "report_fatal_error" when users specify it.
Differential Revision: https://reviews.llvm.org/D63547
llvm-svn: 368744
Summary:
- Moved the SourceExtraction header from lib to include so that it can be used in clangd.
Reviewers: arphaman
Subscribers: ilya-biryukov, dexonsmith, kadircet, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D65878
llvm-svn: 368743
The line number table header was substantially revised in DWARF 5 and is
not fully supported by LLDB's current debug line implementation.
This patch replaces the LLDB debug line parser with its counterpart in
LLVM. This was possible because of the limited contact surface between
the code to parse the DWARF debug line section and the rest of LLDB.
We pay a small cost in terms of performance and memory usage. This is
something we plan to address in the near future.
Differential revision: https://reviews.llvm.org/D62570
llvm-svn: 368742
Summary:
Ideally CompilerType would have no knowledge of clang or any individual
TypeSystem. Decoupling clang is relatively straightforward.
Differential Revision: https://reviews.llvm.org/D66102
llvm-svn: 368741
Summary:
As noted on Errc.h:
// * std::errc is just marked with is_error_condition_enum. This means that
// common patters like AnErrorCode == errc::no_such_file_or_directory take
// 4 virtual calls instead of two comparisons.
And on some libstdc++ those virtual functions conclude that
------------------------
int main() {
std::error_code foo = std::make_error_code(std::errc::no_such_file_or_directory);
return foo == std::errc::no_such_file_or_directory;
}
-------------------------
should exit with 0.
Reviewers: thakis, rnk, jfb
Reviewed By: thakis
Subscribers: lebedev.ri, dexonsmith, xbolva00, cfe-commits, caomhin
Tags: #clang
Differential Revision: https://reviews.llvm.org/D66143
llvm-svn: 368739
In D65379, I briefly described the construction of bug paths from an
ExplodedGraph. This patch is about refactoring the code processing the bug path
into a bug report.
A part of finding a valid bug report was running all visitors on the bug path,
so we already have a (possibly empty) set of diagnostics for each ExplodedNode
in it.
Then, for each diagnostic consumer, we construct non-visitor diagnostic pieces.
* We first construct the final diagnostic piece (the warning), then
* We start ascending the bug path from the error node's predecessor (since the
error node itself was used to construct the warning event). For each node
* We check the location (whether its a CallEnter, CallExit) etc. We simultaneously
keep track of where we are with the execution by pushing CallStack when we see a
CallExit (keep in mind that everything is happening in reverse!), popping it
when we find a CallEnter, compacting them into a single PathDiagnosticCallEvent.
void f() {
bar();
}
void g() {
f();
error(); // warning
}
=== The bug path ===
(root) -> f's CallEnter -> bar() -> f's CallExit -> (error node)
=== Constructed report ===
f's CallEnter -> bar() -> f's CallExit
^ /
\ V
(root) ---> f's CallEvent --> (error node)
* We also keep track of different PathPieces different location contexts
* (CallEvent::path in the above example has f's LocationContext, while the
CallEvent itself is in g's context) in a LocationContextMap object. Construct
whatever piece, if any, is needed for the note.
* If we need to generate edges (or arrows) do so. Make sure to also connect
these pieces with the ones that visitors emitted.
* Clean up the constructed PathDiagnostic by making arrows nicer, pruning
function calls, etc.
So I complained about mile long function invocations with seemingly the same
parameters being passed around. This problem, as I see it, a natural candidate
for creating classes and tying them all together.
I tried very hard to make the implementation feel natural, like, rolling off the
tongue. I introduced 2 new classes: PathDiagnosticBuilder (I mean, I kept the
name but changed almost everything in it) contains every contextual information
(owns the bug path, the diagnostics constructed but the visitors, the BugReport
itself, etc) needed for constructing a PathDiagnostic object, and is pretty much
completely immutable. BugReportContruct is the object containing every
non-contextual information (the PathDiagnostic object we're constructing, the
current location in the bug path, the location context map and the call stack I
meantioned earlier), and is passed around all over the place as a single entity
instead of who knows how many parameters.
I tried to used constness, asserts, limiting visibility of fields to my
advantage to clean up the code big time and dramatically improve safety. Also,
whenever I found the code difficult to understand, I added comments and/or
examples.
Here's a complete list of changes and my design philosophy behind it:
* Instead of construcing a ReportInfo object (added by D65379) after finding a
valid bug report, simply return an optional PathDiagnosticBuilder object straight
away. Move findValidReport into the class as a static method. I find
GRBugReporter::generatePathDiagnostics a joy to look at now.
* Rename generatePathDiagnosticForConsumer to generate (maybe not needed, but
felt that way in the moment) and moved it to PathDiagnosticBuilder. If we don't
need to generate diagnostics, bail out straight away, like we always should have.
After that, construct a BugReportConstruct object, leaving the rest of the logic
untouched.
* Move all static methods that would use contextual information into
PathDiagnosticBuilder, reduce their parameter count drastically by simply
passing around a BugReportConstruct object.
* Glance at the code I removed: Could you tell what the original
PathDiagnosticBuilder::LC object was for? It took a gooood long while for me to
realize that nothing really. It is always equal with the LocationContext
associated with our current position in the bug path. Remove it completely.
* The original code contains the following expression quite a bit:
LCM[&PD.getActivePath()], so what does it mean? I said that we collect the
contexts associated with different PathPieces, but why would we ever modify that,
shouldn't it be set? Well, theoretically yes, but in the implementation, the
address of PathDiagnostic::getActivePath doesn't change if we move to an outer,
previously unexplored function. Add both descriptive method names and
explanations to BugReportConstruct to help on this.
* Add plenty of asserts, both for safety and as a poor man's documentation.
Differential Revision: https://reviews.llvm.org/D65484
llvm-svn: 368737
That change (r363670) could leave a copy from vgpr to sgpr. Fixed.
Differential Revision: https://reviews.llvm.org/D66133
Change-Id: I00c3fe6fda2e8e1e36f53195b881b1449c777ea4
llvm-svn: 368736
When I'm new to a file/codebase, I personally find C++'s strong static type
system to be a great aid. BugReporter.cpp is still painful to read however:
function calls are made with mile long parameter lists, seemingly all of them
taken with a non-const reference/pointer. This patch fixes nothing but this:
make a few things const, and hammer it until it compiles.
Differential Revision: https://reviews.llvm.org/D65382
llvm-svn: 368735
This patch removes usage of FinalPhase from anywhere outside of the scope where
it is used to do argument handling. It also adds argument based trimming of
the Phase list pulled out of the Types.def table.
Differential Revision: https://reviews.llvm.org/D65993
llvm-svn: 368734
The MVE architecture has the idea of "beats", where a vector instruction can be
executed over several ticks of the architecture. This adds a similar system
into the Arm backend cost model, multiplying the cost of all vector
instructions by a factor.
This factor essentially becomes the expected difference between scalar code
and vector code, on average. MVE Vector instructions can also overlap so the a
true cost of them is often lower. But equally scalar instructions can in some
situations be dual issued, or have other optimisations such as unrolling or
make use of dsp instructions. The default is chosen as 2. This should not
prevent vectorisation is a most cases (as the vector instructions will still be
doing at least 4 times the work), but it will help prevent over vectorising in
cases where the benefits are less likely.
This adds things so far to the obvious places in ARMTargetTransformInfo, and
updates a few related costs like not treating float instructions as cost 2 just
because they are floats.
Differential Revision: https://reviews.llvm.org/D66005
llvm-svn: 368733
- Create ASTContext::attachCommentsToJustParsedDecls so we don't have to load external comments in Sema when trying to attach existing comments to just parsed Decls.
- Keep comments ordered and cache their decomposed location - faster SourceLoc-based searching.
- Optimize work with redeclarations.
- Keep one comment per redeclaration chain (represented by canonical Decl) instead of comment per redeclaration.
- For redeclaration chains with no comment attached keep just the last declaration in chain that had no comment instead of every comment-less redeclaration.
Differential Revision: https://reviews.llvm.org/D65301
llvm-svn: 368732
Jonas Devlieghere