Remove custom Task implementations in base.

Remove custom Task implementations in base.

I left a few unconverted in this one, because they made my head hurt.

BUG=none
TEST=base_unittests


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=111407

[dcheng, 2011-11-23 01:01:27]

Please take a close look at the WaitableEvent changes. I believe them to be
correct, but the changes feel a little wonky.

[willchan no longer on Chromium, 2011-11-23 18:15:45]

http://codereview.chromium.org/8653006/diff/1/base/synchronization/waitable_e...
File base/synchronization/waitable_event_watcher_posix.cc (right):

http://codereview.chromium.org/8653006/diff/1/base/synchronization/waitable_e...
base/synchronization/waitable_event_watcher_posix.cc:94: void
AsyncCallbackHelper(scoped_refptr<Flag> flag,
Do Flag* instead of scoped_refptr<Flag>.

The type matters more at the base::Bind() callsite, not the function prototype.
This is because base::Bind() will be copying the argument into a storage
container for when Run() gets called. This storage must be a scoped_refptr.
However, when Run() is actually called, we'll do a scoped_refptr<T> to T*
implicit conversion, due to the operator* operator in scoped_refptr. Therefore,
there's no need to make the function prototype be scoped_refptr<T>, because
that'll just result in an extra unnecessary AddRef()/Release() pair.

http://codereview.chromium.org/8653006/diff/1/base/threading/worker_pool_unit...
File base/threading/worker_pool_unittest.cc (right):

http://codereview.chromium.org/8653006/diff/1/base/threading/worker_pool_unit...
base/threading/worker_pool_unittest.cc:24: void
PostTaskTestHelper(WaitableEvent* event) {
Why is this wrapper function necessary?

[dcheng, 2011-11-23 18:47:49]

http://codereview.chromium.org/8653006/diff/1/base/synchronization/waitable_e...
File base/synchronization/waitable_event_watcher_posix.cc (right):

http://codereview.chromium.org/8653006/diff/1/base/synchronization/waitable_e...
base/synchronization/waitable_event_watcher_posix.cc:94: void
AsyncCallbackHelper(scoped_refptr<Flag> flag,
On 2011/11/23 18:15:46, willchan wrote:
> Do Flag* instead of scoped_refptr<Flag>.
> 
> The type matters more at the base::Bind() callsite, not the function
prototype.
> This is because base::Bind() will be copying the argument into a storage
> container for when Run() gets called. This storage must be a scoped_refptr.
> However, when Run() is actually called, we'll do a scoped_refptr<T> to T*
> implicit conversion, due to the operator* operator in scoped_refptr.
Therefore,
> there's no need to make the function prototype be scoped_refptr<T>, because
> that'll just result in an extra unnecessary AddRef()/Release() pair.

Done, I wasn't sure if base::Bind() would automatically do the right thing if
the underlying type is refcounted.

http://codereview.chromium.org/8653006/diff/1/base/threading/worker_pool_unit...
File base/threading/worker_pool_unittest.cc (right):

http://codereview.chromium.org/8653006/diff/1/base/threading/worker_pool_unit...
base/threading/worker_pool_unittest.cc:24: void
PostTaskTestHelper(WaitableEvent* event) {
On 2011/11/23 18:15:46, willchan wrote:
> Why is this wrapper function necessary?

Done.

[willchan no longer on Chromium, 2011-11-23 18:53:04]

lgtm

[commit-bot: I haz the power, 2011-11-23 18:54:35]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/dcheng@chromium.org/8653006/6001

[commit-bot: I haz the power, 2011-11-23 20:45:45]

Change committed as 111407