2012/5/28 Chris Wilson <chris at chris-wilson.co.uk>: > On Mon, 28 May 2012 16:42:49 -0300, Paulo Zanoni <przanoni at gmail.com> wrote: >> This patch adds the code to fix the problem, but it depends on the >> removal of some code that can't be removed right now and will come >> later in the patch series. The patch that we need is: >> ? - drm/i915: don't write 0 to DIP control at HDMI init > > I was going to grumble a bit more and ask if these could be split into > generational patches so that a bisect doesn't land on a commit affecting > them all. However, it sounds like bisecting through this series is going > to be problematic anyway... Just to be sure, we are not introducing > issues to be resolved in later patches? No. The idea of depending on a patch that comes later in the series is because I tried to avoid any regressions in the middle of the series and also tried to avoid moving code to the "set_infoframes" function only to remove it later. The "don't write 0 to DIP" patch needs the set_infoframes function to completely set the state of the control register, so it basically needed all the previous patches. Any regression introduced in the middle of the series and then removed later is an accident. I could split patches into generational patches, but then we'd have like 60 patches, most of them one-liners... > -Chris > > -- > Chris Wilson, Intel Open Source Technology Centre -- Paulo Zanoni
