Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1356195
| From | Yinghai Lu <yinghai@kernel.org> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [PATCH v10 01/59] PCI: Add pci_find_root_bus_resource() |
| Date | 2016-03-11 22:30 +0100 |
| Message-ID | <rbCNc-3SY-9@gated-at.bofh.it> (permalink) |
| References | <r5TH3-5ok-3@gated-at.bofh.it> <r5U0p-5vw-11@gated-at.bofh.it> <raZVw-1bV-9@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
Hi Bjorn, Thanks for taking time to look at those patches! On Wed, Mar 9, 2016 at 7:54 PM, Bjorn Helgaas <helgaas@kernel.org> wrote: > On Wed, Feb 24, 2016 at 06:11:52PM -0800, Yinghai Lu wrote: >> Add pci_find_root_bus_resource() to return root bus resource >> for input resource. >> >> For sparc mem64 support, We need that host bridge window res: >> 1. we need direct parent for request_resource_conflict() calling in >> pci_register_legacy_regions(). >> 2. check if return is NULL to decide if region is valid or not. > > Whenever somebody proposes a new interface that's only required for > one architecture, the obvious question is "what is unique about that > architecture that makes the new interface necessary?" Good point. I was thinking to keep the current behavior in sparc64 arch, as I don't want to break old setups. Later we can try to make it more like arch. > > Neither this patch nor the follow-on patch that adds the use of > pci_find_root_bus_resource() answers that question. > would be better to keep those two in same patch? Thanks Yinghai
Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread
Re: [PATCH v10 01/59] PCI: Add pci_find_root_bus_resource() Bjorn Helgaas <helgaas@kernel.org> - 2016-03-10 05:00 +0100 Re: [PATCH v10 01/59] PCI: Add pci_find_root_bus_resource() Yinghai Lu <yinghai@kernel.org> - 2016-03-11 22:30 +0100
csiph-web