Message ID | 20140429131101.GI28159@titan.lakedaemon.net |
---|---|
State | New |
Headers | show |
On Tue, Apr 29, 2014 at 09:11:01AM -0400, Jason Cooper wrote: > arm-soc guys, > > This rather large (for fixes) changeset resolves multiple problems with > the mvebu PCIe driver. While it works fine for simple use cases, more > complicated uses highlighted multiple problems with MSI handling and mbus > window creation. > > These changes are Acked by the appropriate maintainers and have been in > -next for about a week. Merged all three. We usually don't split up fixes in separate branches, so no need to do so in the future. -Olof
On Sun, May 04, 2014 at 10:40:33PM -0700, Olof Johansson wrote: > On Tue, Apr 29, 2014 at 09:11:01AM -0400, Jason Cooper wrote: > > arm-soc guys, > > > > This rather large (for fixes) changeset resolves multiple problems with > > the mvebu PCIe driver. While it works fine for simple use cases, more > > complicated uses highlighted multiple problems with MSI handling and mbus > > window creation. > > > > These changes are Acked by the appropriate maintainers and have been in > > -next for about a week. > > Merged all three. We usually don't split up fixes in separate branches, so no > need to do so in the future. Noted. The mbus/pci/irqchip fix series was a complicated one, but I could've (should've) merged it into mvebu/fixes before sending the pull request. Will do so in the future. We've (mvebu, and other sub-archs afaict) been splitting /fixes and /dt-fixes for a while now. Which I think helps reinforce the independence of the two. Do you want those merged together as well, now? thx, Jason.
On Mon, May 5, 2014 at 4:49 AM, Jason Cooper <jason@lakedaemon.net> wrote: > On Sun, May 04, 2014 at 10:40:33PM -0700, Olof Johansson wrote: >> On Tue, Apr 29, 2014 at 09:11:01AM -0400, Jason Cooper wrote: >> > arm-soc guys, >> > >> > This rather large (for fixes) changeset resolves multiple problems with >> > the mvebu PCIe driver. While it works fine for simple use cases, more >> > complicated uses highlighted multiple problems with MSI handling and mbus >> > window creation. >> > >> > These changes are Acked by the appropriate maintainers and have been in >> > -next for about a week. >> >> Merged all three. We usually don't split up fixes in separate branches, so no >> need to do so in the future. > > Noted. The mbus/pci/irqchip fix series was a complicated one, but I > could've (should've) merged it into mvebu/fixes before sending the pull > request. Will do so in the future. > > We've (mvebu, and other sub-archs afaict) been splitting /fixes and > /dt-fixes for a while now. Which I think helps reinforce the > independence of the two. Do you want those merged together as well, > now? Depends on volume. If it's a lot of fixes, separate is fine. If it's just a few patches, then you're mostly causing extra work by separating. This set seemed a bit on the small side for splitting out. -Olof