commit 52cdbdd49853 ("driver core: correct device's shutdown order")
places an assumption of supplier<-consumer order on the process of probe.
But it turns out to break down the parent <- child order in some scene.
E.g in pci, a bridge is enabled by pci core, and behind it, the devices
have been probed. Then comes the bridge's module, which enables extra
feature(such as hotplug) on this bridge.
This will break the parent<-children order and cause failure when
"kexec -e" in some scenario.
I tried to fix this issue in pci subsystem, and it turns out to be wrong.
Thanks to Christoph Hellwig, he enlightens me that it should be a bug in
driver core.
To ease the review, I organize the patch as the following
[3/3] reflects the root cause of this bug. if [2/3] is not acceptable,
we still need some way to fix it.
[2/3] introduce a algorithm to reorder device
[1/3] some trivial help routine
Pingfan Liu (3):
drivers/base: introduce some help routines for reordering a group of
dev
drivers/base: reorder consumer and its children behind suppliers
drivers/base: only reordering consumer device when probing
drivers/base/base.h | 1 +
drivers/base/core.c | 135 ++++++++++++++++++++++++++++++++++++++++++++++++++++
drivers/base/dd.c | 9 +---
3 files changed, 138 insertions(+), 7 deletions(-)
Cc: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
Cc: Grygorii Strashko <grygorii.strashko@xxxxxx>
Cc: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Cc: Bjorn Helgaas <helgaas@xxxxxxxxxx>
Cc: Dave Young <dyoung@xxxxxxxxxx>
Cc: linux-pci@xxxxxxxxxxxxxxx
Cc: linuxppc-dev@xxxxxxxxxxxxxxxx
--
2.7.4
