Jordan Rose [Thu, 6 Dec 2012 18:58:18 +0000 (18:58 +0000)]
[analyzer] Simplify RetainCountChecker's handling of dead symbols.
Previously we made three passes over the set of dead symbols, and removed
them from the state /twice/. Now we combine the autorelease pass and the
symbol death pass, and only have to remove the bindings for the symbols
that leaked.
Jordan Rose [Thu, 6 Dec 2012 18:58:15 +0000 (18:58 +0000)]
[analyzer] Use a smarter algorithm to find the last block in an inlined call.
Previously we would search for the last statement, then back up to the
entrance of the block that contained that statement. Now, while we're
scanning for the statement, we just keep track of which blocks are being
exited (in reverse order).
Jordan Rose [Thu, 6 Dec 2012 18:58:06 +0000 (18:58 +0000)]
[analyzer] Aggressively cut back on the canonicalization in RegionStore.
Whenever we touch a single bindings cluster multiple times, we can delay
canonicalizing it until the final access. This has some interesting
implications, in particular that we shouldn't remove an /empty/ cluster
from the top-level map until canonicalization.
This is good for a 2% speedup or so on the test case in
<rdar://problem/12810842>
Jordan Rose [Thu, 6 Dec 2012 18:58:01 +0000 (18:58 +0000)]
[analyzer] Remove bindExprAndLocation, which does extra work for no gain.
This feature was probably intended to improve diagnostics, but was currently
only used when dumping the Environment. It shows what location a given value
was loaded from, e.g. when evaluating an LValueToRValue cast.
Rework the bitfield access IR generation to address PR13619 and
generally support the C++11 memory model requirements for bitfield
accesses by relying more heavily on LLVM's memory model.
The primary change this introduces is to move from a manually aligned
and strided access pattern across the bits of the bitfield to a much
simpler lump access of all bits in the bitfield followed by math to
extract the bits relevant for the particular field.
This simplifies the code significantly, but relies on LLVM to
intelligently lowering these integers.
I have tested LLVM's lowering both synthetically and in benchmarks. The
lowering appears to be functional, and there are no really significant
performance regressions. Different code patterns accessing bitfields
will vary in how this impacts them. The only real regressions I'm seeing
are a few patterns where the LLVM code generation for loads that feed
directly into a mask operation don't take advantage of the x86 ability
to do a smaller load and a cheap zero-extension. This doesn't regress
any benchmark in the nightly test suite on my box past the noise
threshold, but my box is quite noisy. I'll be watching the LNT numbers,
and will look into further improvements to the LLVM lowering as needed.
Don't require that, during template deduction, a template specialization type
as a function parameter has at least as many template arguments as one used in
a function argument (not even if the argument has been resolved to an exact
type); the additional parameters might be provided by default template
arguments in the template. We don't need this check, since we now implement
[temp.deduct.call]p4 with an additional check after deduction.
Chad Rosier [Wed, 5 Dec 2012 23:08:09 +0000 (23:08 +0000)]
[driver, ms-inline asm] Have -fms-extensions enable the AsmBlocks language
option. MS-style inline asm can now be enabled by either -fasm-blocks or
-fms-extensions.
rdar://12808010
[c-index-test] Introduce '-index-compile-db' which accepts a compilation database file
and does an '-index-file' for all compile commands in the database.
Testing C++ declarations embedded in
<declaration> tag of Comment XML.
Added DeclPrint support for constructors
and fix tests accordingly.
This is wip. // rdar://12378714
Jordan Rose [Wed, 5 Dec 2012 18:44:49 +0000 (18:44 +0000)]
Format strings: offer a cast to 'unichar' for %C in Objective-C contexts.
For most cases where a conversion specifier doesn't match an argument,
we usually guess that the conversion specifier is wrong. However, if
the argument is an integer type and the specifier is %C, it's likely
the user really did mean to print the integer as a character.
(This is more common than %c because there is no way to specify a unichar
literal -- you have to write an integer literal, such as '0x2603',
and then cast it to unichar.)
This does not change the behavior of %S, since there are fewer cases
where printing a literal Unicode *string* is necessary, but this could
easily be changed in the future.
Jordan Rose [Wed, 5 Dec 2012 18:44:40 +0000 (18:44 +0000)]
Format strings: a character literal should be printed with %c, not %d.
The type of a character literal is 'int' in C, but if the user writes a
character /as/ a literal, we should assume they meant it to be a
character and not a numeric value, and thus offer %c as a correction
rather than %d.
There's a special case for multi-character literals (like 'MooV'), which
have implementation-defined value and usually cannot be printed with %c.
These still use %d as the suggestion.
In C++, the type of a character literal is 'char', and so this problem
doesn't exist.
Daniel Jasper [Wed, 5 Dec 2012 14:57:28 +0000 (14:57 +0000)]
Indentation fixes for clang-format.
- Fix behavior of memoization together with optimization
- Correctly attribute the PenaltyIndentLevel (breaking directly after "(" did
not count towards the inner level)
- Recognize more tokens as assignments
Reuse an existing diagnostic for tsan/msan needing -pie error.
Add a diagnosting for -fsanitize=memory conflicting with other sanitizers.
Extend tests.
Richard Smith [Wed, 5 Dec 2012 11:34:06 +0000 (11:34 +0000)]
In C++, if we hit an error in the class-head, don't try to parse the class body.
Our error recovery path may have made the class anonymous, and that has a pretty
disastrous impact on any attempt to parse a class body containing constructors.
Richard Smith [Wed, 5 Dec 2012 11:04:55 +0000 (11:04 +0000)]
PR14049: Don't say "expanded from macro 'foo'" when 'foo' just happens to be
the LHS of a token paste. Use "expanded from here" instead when we're not sure
it's actually a macro.
Richard Smith [Wed, 5 Dec 2012 06:20:58 +0000 (06:20 +0000)]
Simplify slightly by seperating out the responsibility for emission of a caret
diagnostic from the emission of macro backtraces. Incidentally, we now get the
displayed source location for a diagnostic and the location for the caret from
the same place, rather than computing them separately. No functionality change.
Testing C++ declarations embedded in
<declaration> tag of Comment XML and
added support for friend declaration printing.
This is wip. // rdar://12378714
Thread safety analysis: Add a new "beta" warning flag: -Wthread-safety-beta.
As the analysis improves, it will continue to add new warnings that are
potentially disruptive to existing users. From now on, such warnings will
first be introduced under the "beta" flag. Such warnings are not turned on by
default; their purpose is to allow users to test their code against future
planned changes, before those changes are actually made. After a suitable
migration period, beta warnings will be folded into the standard
-Wthread-safety.
Add -whole-archive around the ASan runtime archive in the link command.
This ensures that even though it comes first, we pick up its .o files.
Note that if we can use this (or something similar / equivalent) on
other platforms, we could potentially remove
ReplaceOperatorsNewAndDelete from the ASan runtimes.
We should probably do something similar for TSan and MSan as well.
Bill Wendling [Tue, 4 Dec 2012 21:33:58 +0000 (21:33 +0000)]
Use the 'count' attribute to calculate the upper bound of an array.
The count attribute is more accurate with regards to the size of an array. It
also obviates the upper bound attribute in the subrange. We can also better
handle an unbound array by setting the count to -1 instead of the lower bound to
1 and upper bound to 0.
Currently, with -fsanitize=address, the driver appends libclang_rt.asan.a to
the link command. This all works fine when the driver is also responsible for
adding -lstdc++ to the link command. But, if -lstdc++ (or libstdc++.a, etc) is
passed explicitly to the driver, the ASan runtime will appear in the link
command after the standard library, leading to multiple-definition errors for
the global 'operator new' and 'operator delete'. Fix this in a painfully
simple way, by inserting libclang_rt.asan.a at the start of the link command
instead of the end.
If we need to do something more clever, we can walk the link command looking
for something that resembles libstdc++ and insert libclang_rt.asan.a as late
as possible, but the simple solution works for now.
Testing C declarations embedded in
<declaration> tag of Comment XML and fixed a
missing block literal printout as result of the testing.
// rdar://12378714
objective-c blocks: Consider padding due to alignment
after the fixed size block header when generating
captured block variable info. // rdar://12773256
Autotools has the same include guard for both Clang and LLVM's config.h.
Shuffling order causes the wrong one to win.
CMake didn't exhibit this problem because Clang's has *no* guards.
I'll fix this properly tomorrow when Eric and I can check both build
systems and get them to DTRT, but for now unbreak some bots by hoisting
this header.
projects / clang / log