Second codegen test passing on ARM (simulated).

Second codegen test passing on ARM (simulated).
This required support on ARM for:
- compilation of static calls
- patching of static calls
- stub to call into runtime
- redirection support for calls from simulator to host runtime
- stack frame iteration
Committed: https://code.google.com/p/dart/source/detail?r=19405

[regis, 2013-03-01 06:56:14]

See you on Monday,
Regis

[zra, 2013-03-01 17:32:49]

https://codereview.chromium.org/12381034/diff/1/runtime/vm/assembler_arm.cc
File runtime/vm/assembler_arm.cc (right):

https://codereview.chromium.org/12381034/diff/1/runtime/vm/assembler_arm.cc#n...
runtime/vm/assembler_arm.cc:1051: void Assembler::Drop(intptr_t stack_elements)
{
Maybe change name to DropFromStack?

https://codereview.chromium.org/12381034/diff/1/runtime/vm/stack_frame_arm.cc
File runtime/vm/stack_frame_arm.cc (right):

https://codereview.chromium.org/12381034/diff/1/runtime/vm/stack_frame_arm.cc...
runtime/vm/stack_frame_arm.cc:19: static const int kEntrypointMarkerOffetFromFp
= 2 * kWordSize;
Missing s in Offet?

https://codereview.chromium.org/12381034/diff/1/runtime/vm/stack_frame_arm.cc...
runtime/vm/stack_frame_arm.cc:28: intptr_t
StackFrame::EntrypointMarkerOffetFromFp() {
Here, too.

https://codereview.chromium.org/12381034/diff/1/runtime/vm/stack_frame_ia32.cc
File runtime/vm/stack_frame_ia32.cc (right):

https://codereview.chromium.org/12381034/diff/1/runtime/vm/stack_frame_ia32.c...
runtime/vm/stack_frame_ia32.cc:19: static const int kEntrypointMarkerOffetFromFp
= -1 * kWordSize;
Offet, again

https://codereview.chromium.org/12381034/diff/1/runtime/vm/stack_frame_ia32.c...
runtime/vm/stack_frame_ia32.cc:28: intptr_t
StackFrame::EntrypointMarkerOffetFromFp() {
and here. More in same files for other archs

[srdjan, 2013-03-01 17:46:28]

https://codereview.chromium.org/12381034/diff/1/runtime/vm/assembler_arm.cc
File runtime/vm/assembler_arm.cc (right):

https://codereview.chromium.org/12381034/diff/1/runtime/vm/assembler_arm.cc#n...
runtime/vm/assembler_arm.cc:1051: void Assembler::Drop(intptr_t stack_elements)
{
On 2013/03/01 17:32:49, zra wrote:
> Maybe change name to DropFromStack?

We are already using the name Drop for other architectures, and it is quite a
bit shorter.

https://codereview.chromium.org/12381034/diff/1/runtime/vm/instructions_arm.cc
File runtime/vm/instructions_arm.cc (right):

https://codereview.chromium.org/12381034/diff/1/runtime/vm/instructions_arm.c...
runtime/vm/instructions_arm.cc:70: const intptr_t signed_target_address =
static_cast<intptr_t>(target_address);
I'd like to discuss this code in person.

https://codereview.chromium.org/12381034/diff/1/runtime/vm/runtime_entry_arm.cc
File runtime/vm/runtime_entry_arm.cc (right):

https://codereview.chromium.org/12381034/diff/1/runtime/vm/runtime_entry_arm....
runtime/vm/runtime_entry_arm.cc:38: __ BranchLink(&label);
Think positive:
if (leaf) {
} else {
}

[regis, 2013-03-04 17:52:18]

Thanks!

https://codereview.chromium.org/12381034/diff/1/runtime/vm/assembler_arm.cc
File runtime/vm/assembler_arm.cc (right):

https://codereview.chromium.org/12381034/diff/1/runtime/vm/assembler_arm.cc#n...
runtime/vm/assembler_arm.cc:1051: void Assembler::Drop(intptr_t stack_elements)
{
On 2013/03/01 17:46:28, srdjan wrote:
> On 2013/03/01 17:32:49, zra wrote:
> > Maybe change name to DropFromStack?
> 
> We are already using the name Drop for other architectures, and it is quite a
> bit shorter.

Not changing the name, but not too fond of it either.

https://codereview.chromium.org/12381034/diff/1/runtime/vm/instructions_arm.cc
File runtime/vm/instructions_arm.cc (right):

https://codereview.chromium.org/12381034/diff/1/runtime/vm/instructions_arm.c...
runtime/vm/instructions_arm.cc:70: const intptr_t signed_target_address =
static_cast<intptr_t>(target_address);
On 2013/03/01 17:46:28, srdjan wrote:
> I'd like to discuss this code in person.

As explained in person, the signed shift right is canceled out by the signed
shift left in Smi::New. I've changed the code not to use Smi::New, to make it
clearer.

https://codereview.chromium.org/12381034/diff/1/runtime/vm/runtime_entry_arm.cc
File runtime/vm/runtime_entry_arm.cc (right):

https://codereview.chromium.org/12381034/diff/1/runtime/vm/runtime_entry_arm....
runtime/vm/runtime_entry_arm.cc:38: __ BranchLink(&label);
On 2013/03/01 17:46:28, srdjan wrote:
> Think positive:
> if (leaf) {
> } else {
> }

Done.

https://codereview.chromium.org/12381034/diff/1/runtime/vm/stack_frame_arm.cc
File runtime/vm/stack_frame_arm.cc (right):

https://codereview.chromium.org/12381034/diff/1/runtime/vm/stack_frame_arm.cc...
runtime/vm/stack_frame_arm.cc:19: static const int kEntrypointMarkerOffetFromFp
= 2 * kWordSize;
On 2013/03/01 17:32:49, zra wrote:
> Missing s in Offet?

Done.

https://codereview.chromium.org/12381034/diff/1/runtime/vm/stack_frame_arm.cc...
runtime/vm/stack_frame_arm.cc:28: intptr_t
StackFrame::EntrypointMarkerOffetFromFp() {
On 2013/03/01 17:32:49, zra wrote:
> Here, too.

Done.

https://codereview.chromium.org/12381034/diff/1/runtime/vm/stack_frame_ia32.cc
File runtime/vm/stack_frame_ia32.cc (right):

https://codereview.chromium.org/12381034/diff/1/runtime/vm/stack_frame_ia32.c...
runtime/vm/stack_frame_ia32.cc:19: static const int kEntrypointMarkerOffetFromFp
= -1 * kWordSize;
On 2013/03/01 17:32:49, zra wrote:
> Offet, again

Done.

https://codereview.chromium.org/12381034/diff/1/runtime/vm/stack_frame_ia32.c...
runtime/vm/stack_frame_ia32.cc:28: intptr_t
StackFrame::EntrypointMarkerOffetFromFp() {
On 2013/03/01 17:32:49, zra wrote:
> and here. More in same files for other archs

Done.

[regis, 2013-03-04 18:08:51]

Committed patchset #2 manually as r19405 (presubmit successful).