Types defined in function prototype are diagnosed earlier in C++ compilation.
They are put into declaration context where the prototype is introduced. Later on,
when FunctionDecl object is created, these types are moved into the function context.
This patch fixes PR19018 and PR18963.
Differential Revision: http://reviews.llvm.org/D4145
llvm-svn: 211718
The C++ language requires that the address of a function be the same
across all translation units. To make __declspec(dllimport) useful,
this means that a dllimported function must also obey this rule. MSVC
implements this by dynamically querying the import address table located
in the linked executable. This means that the address of such a
function in C++ is not constant (which violates other rules).
However, the C language has no notion of ODR nor does it permit dynamic
initialization whatsoever. This requires implementations to _not_
dynamically query the import address table and instead utilize a wrapper
function that will be synthesized by the linker which will eventually
query the import address table. The effect this has is, to say the
least, perplexing.
Consider the following C program:
__declspec(dllimport) void f(void);
typedef void (*fp)(void);
static const fp var = &f;
const fp fun() { return &f; }
int main() { return fun() == var; }
MSVC will statically initialize "var" with the address of the wrapper
function and "fun" returns the address of the actual imported function.
This means that "main" will return false!
Note that LLVM's optimizers are strong enough to figure out that "main"
should return true. However, this result is dependent on having
optimizations enabled!
N.B. This change also permits the usage of dllimport declarators inside
of template arguments; they are sufficiently constant for such a
purpose. Add tests to make sure we don't regress here.
llvm-svn: 211677
and is unrelated to the NEON intrinsics in arm_neon.h. On little
endian machines it works fine, however on big endian machines it
exhibits surprising behaviour:
uint32x2_t x = {42, 64};
return vget_lane_u32(x, 0); // Will return 64.
Because of this, explicitly call out that it is unsupported on big
endian machines.
This patch will emit the following warning in big-endian mode:
test.c:3:15: warning: vector initializers are a GNU extension and are not compatible with NEON intrinsics [-Wgnu]
int32x4_t x = {0, 1, 2, 3};
^
test.c:3:15: note: consider using vld1q_s32() to initialize a vector from memory, or vcombine_s32(vcreate_s32(), vcreate_s32()) to initialize from integer constants
1 warning generated.
llvm-svn: 211362
CL permits static redeclarations to follow extern declarations. The
storage specifier on the latter declaration has no effect.
This fixes PR20034.
Differential Revision: http://reviews.llvm.org/D4149
llvm-svn: 211238
There comes a time in the life of any amateur code generator when dumb string
concatenation just won't cut it any more. For NeonEmitter.cpp, that time has
come.
There were a bunch of magic type codes which meant different things depending on
the context. There were a bunch of special cases that really had no reason to be
there but the whole thing was so creaky that removing them would cause something
weird to fall over. There was a 1000 line switch statement for code generation
involving string concatenation, which actually did lexical scoping to an extent
(!!) with a bunch of semi-repeated cases.
I tried to refactor this three times in three different ways without
success. The only way forward was to rewrite the entire thing. Luckily the
testing coverage on this stuff is absolutely massive, both with regression tests
and the "emperor" random test case generator.
The main change is that previously, in arm_neon.td a bunch of "Operation"s were
defined with special names. NeonEmitter.cpp knew about these Operations and
would emit code based on a huge switch. Actually this doesn't make much sense -
the type information was held as strings, so type checking was impossible. Also
TableGen's DAG type actually suits this sort of code generation very well
(surprising that...)
So now every operation is defined in terms of TableGen DAGs. There are a bunch
of operators to use, including "op" (a generic unary or binary operator), "call"
(to call other intrinsics) and "shuffle" (take a guess...). One of the main
advantages of this apart from making it more obvious what is going on, is that
we have proper type inference. This has two obvious advantages:
1) TableGen can error on bad intrinsic definitions easier, instead of just
generating wrong code.
2) Calls to other intrinsics are typechecked too. So
we no longer need to work out whether the thing we call needs to be the Q-lane
version or the D-lane version - TableGen knows that itself!
Here's an example: before:
case OpAbdl: {
std::string abd = MangleName("vabd", typestr, ClassS) + "(__a, __b)";
if (typestr[0] != 'U') {
// vabd results are always unsigned and must be zero-extended.
std::string utype = "U" + typestr.str();
s += "(" + TypeString(proto[0], typestr) + ")";
abd = "(" + TypeString('d', utype) + ")" + abd;
s += Extend(utype, abd) + ";";
} else {
s += Extend(typestr, abd) + ";";
}
break;
}
after:
def OP_ABDL : Op<(cast "R", (call "vmovl", (cast $p0, "U",
(call "vabd", $p0, $p1))))>;
As an example of what happens if you do something wrong now, here's what happens
if you make $p0 unsigned before the call to "vabd" - that is, $p0 -> (cast "U",
$p0):
arm_neon.td:574:1: error: No compatible intrinsic found - looking up intrinsic 'vabd(uint8x8_t, int8x8_t)'
Available overloads:
- float64x2_t vabdq_v(float64x2_t, float64x2_t)
- float64x1_t vabd_v(float64x1_t, float64x1_t)
- float64_t vabdd_f64(float64_t, float64_t)
- float32_t vabds_f32(float32_t, float32_t)
... snip ...
This makes it seriously easy to work out what you've done wrong in fairly nasty
intrinsics.
As part of this I've massively beefed up the documentation in arm_neon.td too.
Things still to do / on the radar:
- Testcase generation. This was implemented in the previous version and not in
the new one, because
- Autogenerated tests are not being run. The testcase in test/ differs from
the autogenerated version.
- There were a whole slew of special cases in the testcase generation that just
felt (and looked) like hacks.
If someone really feels strongly about this, I can try and reimplement it too.
- Big endian. That's coming soon and should be a very small diff on top of this one.
llvm-svn: 211101
This patch adds support for pointer types in global named registers variables.
It'll be lowered as a pair of read/write_register and inttoptr/ptrtoint calls.
Also adds some early checks on types on SemaDecl to avoid the assert.
Tests changed accordingly. (PR19837)
llvm-svn: 210274
to the normal non-placement ::operator new and ::operator delete, but allow
optimizations like new-expressions and delete-expressions do.
llvm-svn: 210137
This allows us to be more careful when dealing with enums whose fixed
underlying type requires special handling in a format string, like
NSInteger.
A refinement of r163266 from a year and a half ago, which added the
special handling for NSInteger and friends in the first place.
<rdar://problem/16616623>
llvm-svn: 209966
A few (mostly CodeGen) parts of Clang were tightly coupled to the
AArch64 backend. Now that it's gone, they will not even compile.
I've also deduplicated RUN lines in many of the AArch64 tests. This
might improve "make check-all" time noticably: some of those NEON
tests were monsters.
llvm-svn: 209578
This patch implements global named registers in Clang, lowering to the just
created intrinsics in LLVM (@llvm.read/write_register). A new type of LValue
had to be created (Register), which just adds support to carry the metadata
node containing the name of the register. Two new methods to emit loads and
stores interoperate with another to emit the named metadata node.
No guarantees are being made and only non-allocatable global variable named
registers are being supported. Local named register support is unchanged.
llvm-svn: 209149
Now that llvm cannot represent alias cycles, we have to diagnose erros just
before trying to close the cycle. This degrades the errors a bit. The real
solution is what it was before: if we want to provide good errors for these
cases, we have to be able to find a clang level decl given a mangled name
and produce the error from Sema.
llvm-svn: 209008
We were emitting dynamic initializers for __thread variables if there
was no explicit initializer, as in this test case:
struct S { S(); };
__thread S s;
llvm-svn: 207675
glibc expects that stddef.h only defines a single thing if either of these
defines is set. For example, before this change, a C file containing
#include <stdlib.h>
int ptrdiff_t = 0;
would compile with gcc but not with clang. Now it compiles with clang too.
This also fixes PR12997, where older versions of the Linux headers would define
NULL incorrectly, and glibc would define __need_NULL and expect stddef.h to
redefine NULL with the correct definition.
llvm-svn: 207606
We never aka vector types because our attributed syntax for it is less
comprehensible than the typedefs. This leaves the user in the dark when
the typedef isn't named that well.
Example:
v2s v; v4f w;
w = v;
The naming in this cases isn't even that bad, but the error we give is
useless without looking up the actual typedefs.
t.c:6:5: error: assigning to 'v4f' from incompatible type 'v2s'
Now:
t.c:6:5: error: assigning to 'v4f' (vector of 4 'float' values) from
incompatible type 'v2s' (vector of 2 'int' values)
We do this for all diagnostics that print a vector type.
llvm-svn: 207267
through to the output even if the input comment comes from an untrusted source
Attribute filtering is currently based on a blacklist, which right now includes
all event handler attributes (they contain JavaScipt code). It should be
switched to a whitelist, but going over all of the HTML5 spec requires a
significant amount of time.
llvm-svn: 206882
This is a partial revert of 183015.
By not recognizing things like _setjmp we lose (returns_twice) attribute on
them, which leads to incorrect code generation.
Fixes PR16138.
llvm-svn: 206362
When instantiating an array that has an alignment attribute on it, we
were looking through the array type and only considering the element
type for the resulting alignment. We need to make sure we take the
array's requirements into account too.
llvm-svn: 206317
This patch adds support for the msvc pragmas section, bss_seg, code_seg,
const_seg and data_seg as well as support for __declspec(allocate()).
Additionally it corrects semantics and adds diagnostics for
__attribute__((section())) and the interaction between the attribute
and the msvc pragmas and declspec. In general conflicts should now be
well diganosed within and among these features.
In supporting the pragmas new machinery for uniform lexing for
msvc pragmas was introduced. The new machinery always lexes the
entire pragma and stores it on an annotation token. The parser
is responsible for parsing the pragma when the handling the
annotation token.
There is a known outstanding bug in this implementation in C mode.
Because these attributes and pragmas apply _only_ to definitions, we
process them at the time we detect a definition. Due to tentative
definitions in C, we end up processing the definition late. This means
that in C mode, everything that ends up in a BSS section will end up in
the _last_ BSS section rather than the one that was live at the time of
tentative definition, even if that turns out to be the point of actual
definition. This issue is not known to impact anything as of yet
because we are not aware of a clear use or use case for #pragma bss_seg
but should be fixed at some point.
Differential Revision=http://reviews.llvm.org/D3065#inline-16241
llvm-svn: 205810
which warns on compound conditionals that always evaluate to the same value.
For instance, (x > 5 && x < 3) will always be false since no value for x can
satisfy both conditions.
This patch also changes the CFG to use these tautological values for better
branch analysis. The test for -Wunreachable-code shows how this change catches
additional dead code.
Patch by Anders Rönnholm.
llvm-svn: 205665
better. This warning will now trigger on the following conditionals:
bool b;
int i;
if (b > 1) {} // always false
if (0 <= (i > 5)) {} // always true
if (-1 > b) {} // always false
Patch by Per Viberg.
llvm-svn: 205608
Serge Pavlov