[lkml]   [2013]   [Feb]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH v3 00/22] PCI: Iterate pci host bridge instead of pci root bus
On Sat, Feb 2, 2013 at 9:05 PM, Yinghai Lu <> wrote:

>> Your solution is to introduce for_each_pci_host_bridge() as an
>> iterator through the known host bridges. There are two scenarios
>> where we use something like this:
>> 1) We want to perform an operation on every known host bridge.
>> 2) We want to initialize something for every host bridge.
>> In my opinion, the only instance of scenario 1) is bus_rescan_store(),
>> where we want to rescan all PCI host bridges.
>> In every other case, we're doing some kind of initialization of all
>> the host bridges. For these cases, for_each_pci_host_bridge() is the
>> wrong solution because it doesn't work for hot-added bridges. I think
>> these cases should be changed to use pcibios_root_bridge_prepare() or
>> something something else called in the host bridge add path.
> Yes, will check if those for_pci_host_bridge could be converted to
> adding calling in pcibios_root_bridge_prepare one by one.

I went through those patches again, here are findings:
1. run-time iteration includes at least 3 occurrence are addressed.
a. in pci-sysfs::bus_rescan_store
b. powerpc pci_64.c::sys_pciconfig_iobase system call
c. probe.c::pci_create_root_bus/pci_find_bus

2. edac: i7core_edac.c::i7core_pci_lastbus()
this one is not support host bridge hotplug, and in i7core_probe it will
bail-out if probed before overall.
To make edac with i7 to support host-bridge hotadd, will change the structure of
that i7core_probe.
But I'm not worried about that yet, because Nehalem-EX does not support EDAC,
and assume we will NOT have use case with new -EX intel cpu with
hotplug support to use EDAC.

3. pcibios_resource_survey/pcibios_assign_unassigned_resource in __init path.
it is with x86/frv/microblaze/mn10300.
for x86/acpi, we have per root bus calling in pci_root.c::acpi_pci_root_add().
other three does not support pci_root_bridge hotplug yet.

4. looks like all other changes are __init path, and those arch does not support
pci_root_bridge hotplug yet.

So looks like this patch set is ok now.

Also i updated

with new ordering of patches.

BTW, I would address unifying pcibios_reserource_survey between
x86/frv/microblaze/mn10300 later.



 \ /
  Last update: 2013-02-06 01:41    [W:0.130 / U:0.596 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site