On Mon, Jul 04, 2022 at 06:46:31PM +0100, Matthew Wilcox wrote:
On Mon, Jul 04, 2022 at 01:41:44PM -0400, Sasha Levin wrote:> Oh. That's super-annoying, because that wasn't part of my commit message. > Andrew has a bad habit of plucking text from the 0/n message and adding it > to the 1/n commit, even when it makes no sense to do so. This text was > to explain why I'm sending the patch series and why it should be merged. > It makes no sense to make it part of the patch commit message. The patch was picked up as a dependency for af2debd58bd7 ("s390/crash: make copy_oldmem_page() return number of bytes copied"), and not as a fix on it's own.But the backport of that patch should have been simply - return rc; + return csize; There was no need to bring in all this other gunk.
Possibly. In general we'll take the set of dependencies even if it means taking more code: it makes it easier to backport future patches, and is less prone to backport errors. If something is wrong we can always fix it up, this isn't permanent, right? -- Thanks, Sasha
