On Friday, October 25, 2013 07:25:45 PM Viresh Kumar wrote: > On 25 October 2013 18:26, Rafael J. Wysocki <rjw@xxxxxxx> wrote: > > Having considered that a bit I think that I'd prefer one patch doing all of > > these changes in one go (and with all applicable ACKs collected), one of the > > reasons being that if it is necessary to revert that stuff, whatever the > > reason, it will be much easier to do that with just one commit than with > > 34 of them. > > With a similar reason I think the probability is more that a revert might > be required for individual drivers as they may need to switch back to > ->target() instead of ->target_index() and so keeping them separate > might be better. > > In case we need to revert all patches due to some breakage, we can > always do that in a single commit if required. > > What do you say? If I need to revert the first patch, then I'll need to revert all of them. Also, if you do the same change in multiple places it actually is easier to handle it pretty much regardless of the angle you look at that from if that's done in one patch. Please do that. -- I speak only for myself. Rafael J. Wysocki, Intel Open Source Technology Center. -- To unsubscribe from this list: send the line "unsubscribe cpufreq" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
