[debugger] Move clearing of optimized code map out of GC.

[debugger] Move clearing of optimized code map out of GC.

This moves the clearing of all optimized code maps out of the GC and
into the debugger to where it is actually required. The main goal here
is to simplify the logic in the already complex visitor for our shared
function info objects.

R=yangguo@chromium.org

Committed: https://crrev.com/8daa7215d4c3b77a573584231a6b330715ba364f
Cr-Commit-Position: refs/heads/master@{#31888}

[Yang, 2015-11-05 18:46:49]

https://codereview.chromium.org/1423713018/diff/20001/src/debug/debug.cc
File src/debug/debug.cc (right):

https://codereview.chromium.org/1423713018/diff/20001/src/debug/debug.cc#newc...
src/debug/debug.cc:1310: // might have inlined the given function and don't
contain breakpoints.
See below. We already deal with inlining.

https://codereview.chromium.org/1423713018/diff/20001/src/debug/debug.cc#newc...
src/debug/debug.cc:1341: if (!function->Inlines(*shared)) continue;
We check for functions that inlines this function, so we can actually narrow
down functions for which we have to flush the optimized code map.

[Michael Starzinger, 2015-11-05 18:54:35]

https://codereview.chromium.org/1423713018/diff/20001/src/debug/debug.cc
File src/debug/debug.cc (right):

https://codereview.chromium.org/1423713018/diff/20001/src/debug/debug.cc#newc...
src/debug/debug.cc:1341: if (!function->Inlines(*shared)) continue;
On 2015/11/05 18:46:49, Yang wrote:
> We check for functions that inlines this function, so we can actually narrow
> down functions for which we have to flush the optimized code map.

Yes, but if there is no closure (i.e. JSFunction) but still a SharedFunctionInfo
which has entires in the optimized code map, then the code will not be
deoptmized and the optimized code map will not be cleared.

[Yang, 2015-11-09 10:59:25]

On 2015/11/05 18:54:35, Michael Starzinger wrote:
> https://codereview.chromium.org/1423713018/diff/20001/src/debug/debug.cc
> File src/debug/debug.cc (right):
> 
>
https://codereview.chromium.org/1423713018/diff/20001/src/debug/debug.cc#newc...
> src/debug/debug.cc:1341: if (!function->Inlines(*shared)) continue;
> On 2015/11/05 18:46:49, Yang wrote:
> > We check for functions that inlines this function, so we can actually narrow
> > down functions for which we have to flush the optimized code map.
> 
> Yes, but if there is no closure (i.e. JSFunction) but still a
SharedFunctionInfo
> which has entires in the optimized code map, then the code will not be
> deoptmized and the optimized code map will not be cleared.

lgtm, as discussed offline, please add a comment as to why we want to clear all
(shared functions with optimized code in the map that inlines the function where
we break, but no jsfunction instantiated).

[Michael Starzinger, 2015-11-09 14:20:17]

On 2015/11/09 10:59:25, Yang wrote:
> On 2015/11/05 18:54:35, Michael Starzinger wrote:
> > https://codereview.chromium.org/1423713018/diff/20001/src/debug/debug.cc
> > File src/debug/debug.cc (right):
> > 
> >
>
https://codereview.chromium.org/1423713018/diff/20001/src/debug/debug.cc#newc...
> > src/debug/debug.cc:1341: if (!function->Inlines(*shared)) continue;
> > On 2015/11/05 18:46:49, Yang wrote:
> > > We check for functions that inlines this function, so we can actually
narrow
> > > down functions for which we have to flush the optimized code map.
> > 
> > Yes, but if there is no closure (i.e. JSFunction) but still a
> SharedFunctionInfo
> > which has entires in the optimized code map, then the code will not be
> > deoptmized and the optimized code map will not be cleared.
> 
> lgtm, as discussed offline, please add a comment as to why we want to clear
all
> (shared functions with optimized code in the map that inlines the function
where
> we break, but no jsfunction instantiated).

Thanks! Done.

[Michael Starzinger, 2015-11-09 14:38:14]

The CQ bit was checked by mstarzinger@chromium.org

[Michael Starzinger, 2015-11-09 14:38:14]

The patchset sent to the CQ was uploaded after l-g-t-m from yangguo@chromium.org
Link to the patchset: https://codereview.chromium.org/1423713018/#ps60001
(title: "Addressed comments.")

[commit-bot: I haz the power, 2015-11-09 14:38:22]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1423713018/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1423713018/60001

[commit-bot: I haz the power, 2015-11-09 14:39:42]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-11-09 14:40:08]

Patchset 4 (id:??) landed as
https://crrev.com/8daa7215d4c3b77a573584231a6b330715ba364f
Cr-Commit-Position: refs/heads/master@{#31888}