That sounds very related to which features are enabled on which ports.
A per-directory map is probably easier to produce of course.
We have no good way to get any of this information out of our myriad
build and branch strategies employed by the numerous ports. :)
I too am in favor of [port] pref
On Wed, Oct 5, 2011 at 9:56 PM, Martin Robinson wrote:
> On Wed, Oct 5, 2011 at 6:09 PM, David Levin wrote:
>
> > btw, I'm not planning to work on this for now but would be happy to give
> > advice if someone took it up.
>
> A queryable mapping of which source directories and files apply to
> whi
On Wed, Oct 5, 2011 at 6:09 PM, David Levin wrote:
> btw, I'm not planning to work on this for now but would be happy to give
> advice if someone took it up.
A queryable mapping of which source directories and files apply to
which ports could be quite useful indeed.
--Martin
___
On Wed, Oct 5, 2011 at 6:07 PM, Alexey Proskuryakov wrote:
>
> 05.10.2011, в 17:47, James Robinson написал(а):
>
> That would certainly be nice, but it wouldn't be as useful as having the
> platform in the bug description. The "Platform" field is not very obvious
> in the bugzilla UI and doesn't
05.10.2011, в 17:47, James Robinson написал(а):
> That would certainly be nice, but it wouldn't be as useful as having the
> platform in the bug description. The "Platform" field is not very obvious in
> the bugzilla UI and doesn't show up at all in bugzilla-generated emails,
> ChangeLogs, or
On Oct 5, 2011, at 5:47 PM, James Robinson wrote:
> On Wed, Oct 5, 2011 at 5:43 PM, Simon Fraser wrote:
>
>> Why don't we just fix bugzilla to have a useful lists of ports in one of the
>> "Platform" popups?
>
> That would certainly be nice, but it wouldn't be as useful as having the
> platfo
On Wed, Oct 5, 2011 at 5:43 PM, Simon Fraser wrote:
> On Oct 5, 2011, at 5:37 PM, Darin Adler wrote:
>
> > Hi folks.
> >
> > It really helps frequent reviewers like me if you use platform prefixes
> on bugs. A platform prefix indicates that the bug only affects code on a
> particular platform. Fo
On Oct 5, 2011, at 5:37 PM, Darin Adler wrote:
> Hi folks.
>
> It really helps frequent reviewers like me if you use platform prefixes on
> bugs. A platform prefix indicates that the bug only affects code on a
> particular platform. For example:
>
>[Mac] Use four more named cursors if pres
Hi folks.
It really helps frequent reviewers like me if you use platform prefixes on
bugs. A platform prefix indicates that the bug only affects code on a
particular platform. For example:
[Mac] Use four more named cursors if present
The above title tells you the patch is Mac-specific. Or:
>> A simpler task that would be useful to GC would be to add to ClassInfo the
>> size of an object, so the collector can traverse objects without explicit
>> size class information.
>
> this looks interesting; I'll dig through the code-base and see what I
> can do.
Looks like Mark is currently
Hi!
I guess we could have an smart-pointer'ish thing (some kind of a GC
handle) which could take care of the relocation for native,
non-managed code. However,
> A simpler task that would be useful to GC would be to add to ClassInfo the
> size of an object, so the collector can traverse objects
Hi Sanjoy.
(+Oliver and Mark, who can help point you in the right direction.)
I don't think the system you described would work. To move an object, you need
to update all pointers to it. You also need to resolve other assumptions about
pointer identity. For example, you need to rework hash tabl
On Tue, Oct 4, 2011 at 10:46 PM, Dan Bernstein wrote:
>
> On Oct 4, 2011, at 10:37 PM, Adam Barth wrote:
>
>> About a week ago, the Chromium project measured a PLT regression on
>> Windows, Mac, and Linux:
>>
>> https://bugs.webkit.org/show_bug.cgi?id=69238
>>
>> I don't know whether the regressio
13 matches
Mail list logo