base: Fix MemoryPressureListener documentation

base: Fix MemoryPressureListener documentation

ObserverListThreadSafe will call PostTask on all observer callbacks,
even those on the same thread. So correct the comment in
MemoryPressureListener which states callbacks on the same thread as the
system broadcast are called synchronously.

BUG=
NOTRY=true

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=226497

[boliu, 2013-10-02 03:34:34]

Marcus, do you think we should fix ObserverListThreadSafe to make this guarantee
correct?

[bulach, 2013-10-02 16:56:28]

lgtm, I think it's fine to just update the documentation...

the graphs we're collecting shows that this seems to be working fine, I don't
think all the memory needs to be trimmed synchronously (and even it had to be
synchronously, we wouldn't want to block for long running prunes anyways), so I
guess your new proposal is fine...

[boliu, 2013-10-02 17:05:33]

On 2013/10/02 16:56:28, bulach wrote:
> lgtm, I think it's fine to just update the documentation...
> 
> the graphs we're collecting shows that this seems to be working fine, I don't
> think all the memory needs to be trimmed synchronously (and even it had to be
> synchronously, we wouldn't want to block for long running prunes anyways), so
I
> guess your new proposal is fine...

Just a note, one of the hidden guarantees for android onTrimMemory is that the
view egl context is current during the call, so GL resources can be cleaned up.
It's only relevant for webview, so we are registering/unregistering our own
callbacks for it.

[boliu, 2013-10-02 17:08:39]

+thakis for base owners

[Nico, 2013-10-02 17:51:20]

lgtm

[commit-bot: I haz the power, 2013-10-02 17:55:11]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/boliu@chromium.org/25676002/1

[commit-bot: I haz the power, 2013-10-02 17:57:12]

Change committed as 226497