Martin Langhoff wrote: > > I understand the reasoning, but qualifying the build with a changed > kernel will kill us in QA across the system. Got to ruminate on that > one a bit...
When deployments report back to us "touchpad" problems are among the top 3 items. Usually #2 behind power. Due to the dynamic nature of the problem its difficult to access exactly how widespread the problems really are. The first reports usually are that its "all of the machines" but when I drill down to the details I get various conflicting info. Such is the nature of this particular annoyance. Recently (finally) the combination a consistent occurrence of the problem, good feedback paths, and the right set of debugging tools available has surfaced and we are getting information that will help answer some of these questions. I'm confident that we in the near future we will work this issue out as best as the hardware lets us. I consider touchpad issues above the "paper cuts" level of fixes. This is a change that will allow us to release regardless if the issue is fully solved and apply the Right Fix when its ready. So even if it does cause an increased QA load I think pgf's use of "must" is correct. We would be foolish to go though all the work do a new release and not include this suggestion. You have to QA the whole system regardless of what change you make so really it doesn't increase the QA that much anyway. -- Richard Smith <[email protected]> One Laptop Per Child _______________________________________________ Devel mailing list [email protected] http://lists.laptop.org/listinfo/devel
