Re: [GIT PULL 2/4] soc: drivers for 7.3
Linus Torvalds <[email protected]>
| Newsgroups | dev.linux.lists.soc,org.infradead.lists.linux-arm-kernel,org.kernel.vger.linux-kernel |
|---|---|
| Message-ID | <CAHk-=wh8khg-Odf9cg2jWe1jvSviTwGgKG9vUDzYNdMzgmxDhQ@mail.gmail.com> |
On Tue, 18 Aug 2026 at 14:17, Arnd Bergmann <[email protected]> wrote: > > .mailmap | 6 +- > Documentation/ABI/testing/debugfs-vfio | 26 + > .../ABI/testing/sysfs-bus-pci-drivers-xhci_hcd | 2 +- [...] > tools/virtio/linux/mod_devicetable.h | 14 + > tools/virtio/linux/virtio_features.h | 79 ++ > 2565 files changed, 15693 insertions(+), 8154 deletions(-) Well, that diffstat was just garbage. The real one looks something like this: 121 files changed, 4566 insertions(+), 1492 deletions(-) and actually mentions files that make sense. This garbage diffstat is normal when you have multiple merge bases - git will pick the first one it finds as the base for the diff, and the diff will then contain random other contents that came from entirely different trees. But the fact that you have multiple merge bases also means that you know how to merge with git, and so typically a quality pull request would involve doing a pre-merge test in order to get the proper "this is the diffstat you will see after a merge". I don't actually require that kind of better pull request, but it would make your PRs look nicer. Hint hint, Linus