On Sun, May 15, 2011 at 05:50:54PM +0100, Ben Hutchings wrote: > Greg, if I send you a patch with an extra note in the commit message > that it has been adjusted for 2.6.32, there is usually a good reason for > that. If you require explanation, please ask for it rather than quietly > substituting your own version.
If you say "adjusted", I think "fuzz handled, or something like that", I don't think "changed the logic" like you are saying you have done. > Commit 7e8e5d9718744b817bfea6f020586d7035cc89f4 'netxen: module firmware > hints' added a MODULE_FIRMWARE reference to the 'unified' firmware files > for this driver, which the version in 2.6.32 does not support. For a > similar reason, it doesn't apply cleanly. In the version I sent to you > for 2.6.32.y, I fixed this up and noted that I had adjusted the patch. > But 2.6.32.40 includes a different version, where you have made a trivial > (and wrong) textual adjustment without even mentioning it in the commit > message. I took the original commit, and resolved the diff conflict, nothing more, nothing less. I didn't realize that you had changed the strings, or the logic, as you didn't say that. Whereever possible, I take the original commit, as that way I know that someone hasn't modified the patch in ways they did not describe (as you have done here, and as others have done in the past.) I've been burned before, and don't want to again. That is why, again, I would love just a list of git commit ids, not the patches, wherever possible. And whan you do make a change, like you did above, you need to specify it, in detail, what you did and why you did it differently from the original commit, in order for me to be able to accept it. > I think the effect in this case is minor, but now I have to worry whether > you are quietly changing other patches. Now I wonder if you are quietly changing other patches as well, it goes both ways :) For the above problem, please just send me a fix-up patch to apply to the .32 tree to resolve the issue. thanks, greg k-h _______________________________________________ stable mailing list [email protected] http://linux.kernel.org/mailman/listinfo/stable
