Traverse inlined frames lazily when printing the stacktrace.

Traverse inlined frames lazily when printing the stacktrace. No need to carry separate function and code array, since function can always be extracted from code.

R=asiva@google.com

Committed: https://code.google.com/p/dart/source/detail?r=27209

[siva, 2013-09-04 22:25:20]

As discussed offline on OOM cases we should make sure we don't allocate anymore
heap memory.

https://codereview.chromium.org/23964003/diff/25001/runtime/lib/stacktrace.cc
File runtime/lib/stacktrace.cc (right):

https://codereview.chromium.org/23964003/diff/25001/runtime/lib/stacktrace.cc...
runtime/lib/stacktrace.cc:55: if (func.is_visible()) {
I think we should do this filtering in the print functionality as this is more
related to printing than collecting the stack frames.

https://codereview.chromium.org/23964003/diff/25001/runtime/vm/object.cc
File runtime/vm/object.cc (right):

https://codereview.chromium.org/23964003/diff/25001/runtime/vm/object.cc#newc...
runtime/vm/object.cc:66: DECLARE_FLAG(bool, verbose_stacktrace);
Are these flags now in alphabetical order?

https://codereview.chromium.org/23964003/diff/25001/runtime/vm/object.cc#newc...
runtime/vm/object.cc:14690: } else if (function.is_visible() ||
FLAG_verbose_stacktrace) {
I think we should be skipping over this inline frame iterations for stack traces
which are collected on OOM and stack overflow cases.

Maybe you could have a field in the stack trace object which indicates if it is
a pre allocated stack trace.

[srdjan, 2013-09-05 18:14:09]

PTAL

https://codereview.chromium.org/23964003/diff/25001/runtime/lib/stacktrace.cc
File runtime/lib/stacktrace.cc (right):

https://codereview.chromium.org/23964003/diff/25001/runtime/lib/stacktrace.cc...
runtime/lib/stacktrace.cc:55: if (func.is_visible()) {
On 2013/09/04 22:25:21, siva wrote:
> I think we should do this filtering in the print functionality as this is more
> related to printing than collecting the stack frames.

Done.

https://codereview.chromium.org/23964003/diff/25001/runtime/vm/object.cc
File runtime/vm/object.cc (right):

https://codereview.chromium.org/23964003/diff/25001/runtime/vm/object.cc#newc...
runtime/vm/object.cc:66: DECLARE_FLAG(bool, verbose_stacktrace);
On 2013/09/04 22:25:21, siva wrote:
> Are these flags now in alphabetical order? 

Yes, split into DEFINEs and DECLAREs

https://codereview.chromium.org/23964003/diff/25001/runtime/vm/object.cc#newc...
runtime/vm/object.cc:14690: } else if (function.is_visible() ||
FLAG_verbose_stacktrace) {
On 2013/09/04 22:25:21, siva wrote:
> I think we should be skipping over this inline frame iterations for stack
traces
> which are collected on OOM and stack overflow cases.
> 
> Maybe you could have a field in the stack trace object which indicates if it
is
> a pre allocated stack trace.

Done.

[siva, 2013-09-05 19:52:45]

lgtm

https://codereview.chromium.org/23964003/diff/69001/runtime/vm/raw_object.h
File runtime/vm/raw_object.h (right):

https://codereview.chromium.org/23964003/diff/69001/runtime/vm/raw_object.h#n...
runtime/vm/raw_object.h:1485: bool expand_inlined_;  // False for special
exceptions.
// False for pre allocated stack trace (used in OOM and Stack overflow).

[srdjan, 2013-09-05 20:24:50]

https://codereview.chromium.org/23964003/diff/69001/runtime/vm/raw_object.h
File runtime/vm/raw_object.h (right):

https://codereview.chromium.org/23964003/diff/69001/runtime/vm/raw_object.h#n...
runtime/vm/raw_object.h:1485: bool expand_inlined_;  // False for special
exceptions.
On 2013/09/05 19:52:45, siva wrote:
> // False for pre allocated stack trace (used in OOM and Stack overflow).

Done.

[srdjan, 2013-09-05 20:25:07]

Committed patchset #12 manually as r27209 (presubmit successful).