search for: foo_t

Hi, I have the following program: // test.c #include <stdlib.h> struct foo_t { int x[1024]; }; __thread struct foo_t g_foo; void bar(struct foo_t* foo) { g_foo = *foo; } int main() { struct foo_t* f = (struct foo_t*)malloc(sizeof(struct foo_t)); bar(f); return 0; } When I compile it with clang I see that it inserts memcpy() in function bar(): $ clang -v clang ve...

...a header you can end up with two copies of this type, and their scope is completely unrelated. Here it is without using the type map: -- !0 = distinct !DICompileUnit() !1 = distinct !DISubprogram(name: "foo", unit: !0) !2 = distinct !DICompositeType(name: "T", identifier: "foo_T", scope: !1) !3 = distinct !DICompileUnit() !4 = distinct !DISubprogram(name: "foo", unit: !3) !5 = distinct !DICompositeType(name: "T", identifier: "foo_T", scope: !4) -- I had thought that when !DICompositeType had an 'identifier:' (followed ODR-rules)...

...n also just forward declare all these classes instead of #including their headers. +#include "assert.h" Please use the c++ version of c headers when possible: <cassert>. + MCDisassembler(MemoryObject& region, + raw_ostream& vStream) { Please use "foo_t &f" instead of "foo_t& f". + virtual ~MCDisassembler() { + } Make sure each class with a virtual method has at least one method defined out of line: http://llvm.org/docs/CodingStandards.html#ll_virtual_anch Some bigger stuff: include/llvm/MC/MCDisassembler.h seems...

On Fri, Apr 15, 2016 at 4:04 PM, Duncan P. N. Exon Smith < dexonsmith at apple.com> wrote: > > > On 2016-Apr-15, at 14:53, David Blaikie <dblaikie at gmail.com> wrote: > > > > > > > >> On Fri, Apr 15, 2016 at 2:27 PM, Duncan P. N. Exon Smith < > dexonsmith at apple.com> wrote: > >> > >> > On 2016-Apr-15, at 10:27, David

Bill, thanks for your comments. I'll respond to them individually. I've attached a new revision of the patch that addresses them. Patch built and tested against SVN 79487, with the additional attached fix that fixes an Intel table bug. Sean On 2009/08/18, at 0:57, Bill Wendling wrote: > 0. Watch out for tabs! Fixed. Thanks. > 1. Includes like this "#include

...some other things were more important at the time so I dropped the issue. Ultimately, there were three main issues: First, "unsigned longs" in the dom0 interface (but not the guest interface) should be converted[2] to a new type[3]: typedef unsigned long long __attribute__((aligned(8))) foo_t; Second, hypercalls that deal with frame numbers, e.g. DOM0_GETMEMLIST, also use "unsigned long". I''ve sent a separate mail about this, but I think it makes sense to define a DOM0_GETMEMLIST2 hypercall for this issue. I haven''t investigated this fully. Third, the tools...