From mboxrd@z Thu Jan 1 00:00:00 1970 From: arnd@arndb.de (Arnd Bergmann) Date: Thu, 25 Oct 2012 15:13:43 +0000 Subject: [GIT PULL] first set of ux500 for ARM SoC for v3.8 In-Reply-To: References: <201210251344.44546.arnd@arndb.de> Message-ID: <201210251513.43931.arnd@arndb.de> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Thursday 25 October 2012, Linus Walleij wrote: > On Thu, Oct 25, 2012 at 3:44 PM, Arnd Bergmann wrote: > > > Please remember to write the pull request in a way that lets us reuse some > > of the text for the merge changeset comment. Ideally in a signed tag > > so git puts it in there automatically. > > So what was wrong with this signed tag, looking like that? > > ---------------------------------------------------------------- > A first set of Ux500 core patches for v3.8: > - Support the HREF520 board > - Add the device serial number to the entropy pool > > ---------------------------------------------------------------- > > I guess what I'm after is what makes it reusable if this one > isn't. The tag looks great, but you were sending the pull request for the branch "ux500-core", not for the tag "ux500-core-for-arm-soc". This has happened to me a few times before, especially when pushing the tag just before sending the pull request. In that case, git-request-pull cannot find the tag on the server because it has not been mirrored back from the git+ssh server to the publically accessible one that it reads the tags from. Arnd