On Thu, 25 Dec 2008, David Brownell wrote: > On Thursday 25 December 2008, Alan Stern wrote: > > I guess it was never made clear that this needed to be pushed to Linus > > because it fixed problems in the mainline kernel. Sorry about the > > mixup. > > I'm not sure how much of that is in the "not yet used by > mainline" category you mentioned in your fix-misspellings > patch, though. Maybe none of it. The spelling bugs are > certainly in mainline. The misspelled macros are in mainline, but no code in mainline uses those macros. That's why the misspellings were able to go undetected for a while. > Clearly there's an issue in Greg's current patch queue; > I didn't know that fix-misspellings patch existed, and Greg categorizes his pending patches according to subsystem (USB, driver core, staging). That becomes a little awkward in situations like this, where the new macros and the spelling corrections were submitted as changes to the driver core but then were used by patches affecting USB. I generally base my work on the gregkh-all-*.patch file, which includes all the others, precisely in order to avoid this sort of problem. Usually separating out patches by subsystem results in independent sub-series that can build just fine. But sometimes, like now, it doesn't. This seems to be unavoidable. It gets even worse when there are inter-related patches affecting multiple subsystems that don't all have the same maintainer! > it seems to fix some of the issues coming from the patch > that starts to use those symbols. You've got cause and effect reversed. The spelling correction patch was submitted first and the patch using the corrected macros was submitted second. So it didn't exactly "fix" the issues -- more like, it prevented them from becoming issues in the first place. > Do you just want to provide an update to that patch of > yours (usb-enhance-usage-of-pm_message_t.patch) which > seems to be the proximate cause of the build bugs? We can handle this any way Greg deems best. Since the fix-misspellings patch is already in his tree, the easiest approach is to do nothing and wait for it to be sent in along with everything else during the upcoming merge window. If you'd prefer some other course of action, it will probably be okay with me. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
