Re: [dev-servo] Moving away from merge commits

2016-05-05 Thread Manish Goregaokar
> I don't remember all the arguments; but the point that stuck with me was > that such rebases are generally untested, > Nothing goes into master without being tested in the form master will be post-merge > Automatic patch merging works out surprisingly often -- but sometimes it > doesn't. If th

Re: [dev-servo] Moving away from merge commits

2016-05-05 Thread Olaf Buddenhagen
Hi, On Tue, Apr 26, 2016 at 11:20:17AM -0700, Gregory Szorc wrote: > The thing is, I'm not a huge fan of merge commits in version control, > especially for large projects. [...] > All I'm asking is that Servo and its immediately related projects > consider changing their ways. Please don't. This

Re: [dev-servo] when to switch to webrender by default

2016-05-05 Thread Patrick Walton
I'm pretty sure auto fallback would be doable. Patrick On Wed, May 4, 2016 at 1:31 PM, Jack Moffitt wrote: > > So is the idea that people without the ability to use webrender would > need > > to use the -c flag? The linux machine I use for Servo development yields > the > > "GL context creation

Re: [dev-servo] Output profiling data as a timeline

2016-05-05 Thread Josh Matthews
I added profiling markers for HTTP requests, JS execution, and synchronous DOM parsing and took a look at some sites with known pageload performance issues. The results have been... illuminating. Besides clear inefficiencies that became noticeable (reading max 1kb from HTTP response bodies at a