On OS X, there is an actual significance to the distinction, which before lead e.g. to exception throwing in the below example being broken:
---
import core.exception;
void main() {
asm {
jmp Lfoo;
Lfoo:
;
}
throw cast(OutOfMemoryError)cast(void*)OutOfMemoryError.classinfo.init;
assert(0);
}
---
1) The last parameter of getGetElementPtr() has type bool. In some instances, a 2 is used as parameter. This is converted to true.
2) Several loops use int instead of unsigned. This causes warning about signed/unsigned mismatch.
Curiously, only Visual C++ complains about this. Nevertheless I think that the warnings should be fixed.
Several functions regarding file handling are deprecated. This commit replaces these functions with the new one from llvm::sys::fs and llvm::sys::path. It also removes some warnings about signed/unsigned mismatches.
This enables us to remove the whole ugly config file parsing hack.
Also includes some minor CMake beautifications (nobody should really be at 2.4 anymore, 2.6 is more than three years old now).
This solves the problem of LDC picking up a pre-existing system-wide config when druntime and Phobos are compiled as part of the build process.
This order was intended by Christian in 919f8d6, but then backed out again in eb3cd887 with no comment as to why – I hope this doesn't have any unwanted implications.
Unline older versions, llvm 3.0 provides atomic instructions, not intrinsics.
The patch wraps the instructions into intrinsics, so they will be accessible
from d code.
Before, _d_arraycatT was used to concatenate multiple arrays. That caused an issue when postblit
was called on a struct multiple times. The next code asserted due to the issue:
void main()
{
static struct S
{
int x;
int pad;
this(this)
{
++x;
}
}
auto sarr = new S[1];
auto sarr2 = sarr ~ sarr ~ sarr;
assert(sarr2[0].x == 1);
assert(sarr2[1].x == 1);
assert(sarr2[2].x == 1);
assert(sarr[0].x == 0);
}
