Added RefCountedMallocedMemory

Added RefCountedMallocedMemory

Which takes ownership over a chunk of memory allocated with malloc or calloc
and must then be freed using free().

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=236675

[João Eiras, 2013-11-20 18:05:46]

This class is handy for APIs that returned memory allocated with malloc, like
for instance the WebP encoding API.

[Elliot Glaysher, 2013-11-20 18:28:38]

On 2013/11/20 18:05:46, João Eiras wrote:
> This class is handy for APIs that returned memory allocated with malloc, like
> for instance the WebP encoding API.

Where are instances of this going to be fed to? (I assume you're doing this to
minimize copies with data that then gets fed into some chrome api that takes
RefCountedMemory instances.

[João Eiras, 2013-11-21 13:14:32]

On 2013/11/20 18:28:38, Elliot Glaysher wrote:
> On 2013/11/20 18:05:46, João Eiras wrote:
> > This class is handy for APIs that returned memory allocated with malloc,
like
> > for instance the WebP encoding API.
> 
> Where are instances of this going to be fed to? (I assume you're doing this to
> minimize copies with data that then gets fed into some chrome api that takes
> RefCountedMemory instances.

You assumed right. After grepping a bit for some common memory functions and use
of RefCountedMemory I did not manage to find anything special that would benefit
from this. However, it's utility code that we used in our project (with tests!)
which we thought would be useful for others.

[Elliot Glaysher, 2013-11-21 18:50:22]

On 2013/11/21 13:14:32, João Eiras wrote:
> On 2013/11/20 18:28:38, Elliot Glaysher wrote:
> > On 2013/11/20 18:05:46, João Eiras wrote:
> > > This class is handy for APIs that returned memory allocated with malloc,
> like
> > > for instance the WebP encoding API.
> > 
> > Where are instances of this going to be fed to? (I assume you're doing this
to
> > minimize copies with data that then gets fed into some chrome api that takes
> > RefCountedMemory instances.
> 
> You assumed right. After grepping a bit for some common memory functions and
use
> of RefCountedMemory I did not manage to find anything special that would
benefit
> from this. However, it's utility code that we used in our project (with
tests!)
> which we thought would be useful for others.

Odd thing is, I just came out of a hallway conversation where someone was
complaining about us not using webp for a bunch of our own internal images, and
I mentioned this patch. Basically, I suspect we'll need this now.

Or: lgtm.

[brettw, 2013-11-21 23:29:49]

lgtm

https://codereview.chromium.org/78603002/diff/1/base/memory/ref_counted_memory.h
File base/memory/ref_counted_memory.h (right):

https://codereview.chromium.org/78603002/diff/1/base/memory/ref_counted_memor...
base/memory/ref_counted_memory.h:116: // An implementation of RefCountedMemory,
that holds a chunk of memory
Grammar: no comma before "that"

[commit-bot: I haz the power, 2013-11-22 00:04:21]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/joaoe@opera.com/78603002/80001

[commit-bot: I haz the power, 2013-11-22 01:49:41]

Change committed as 236675

[Lei Zhang, 2015-12-19 00:57:22]

thestig@chromium.org changed reviewers:
+ thestig@chromium.org

[Lei Zhang, 2015-12-19 00:57:22]

So it's 2 years later, and I don't see any users of RefCountedMallocedMemory
except for the unit test. Did plans to use RefCountedMallocedMemory never
happen, or am I missing something?

[brettw, 2015-12-29 06:05:23]

Nice catch! Let's remove.

[Lei Zhang, 2015-12-29 19:11:37]

On 2015/12/29 06:05:23, brettw wrote:
> Nice catch! Let's remove.

https://codereview.chromium.org/1550963002