On Monday, January 26, 2009 8:46 am Greg KH wrote: > On Tue, Jan 27, 2009 at 12:00:00AM +0800, Yu Zhao wrote: > > On Sun, Jan 25, 2009 at 05:25:02AM +0800, Greg KH wrote: > > > On Sun, Jan 25, 2009 at 12:41:34AM +0800, Yu Zhao wrote: > > > > Signed-off-by: Yu Zhao <yu.zhao@xxxxxxxxx> > > > > --- > > > > drivers/pci/remove.c | 3 +++ > > > > 1 files changed, 3 insertions(+), 0 deletions(-) > > > > > > > > diff --git a/drivers/pci/remove.c b/drivers/pci/remove.c > > > > index 042e089..caf8e1e 100644 > > > > --- a/drivers/pci/remove.c > > > > +++ b/drivers/pci/remove.c > > > > @@ -71,6 +71,9 @@ void pci_remove_bus(struct pci_bus *pci_bus) > > > > down_write(&pci_bus_sem); > > > > list_del(&pci_bus->node); > > > > up_write(&pci_bus_sem); > > > > + if (!pci_bus->is_added) > > > > + return; > > > > > > What race condition is this solving? > > > > pci_bus_add_child() sets the 'is_added' if a bus is successfully > > registered, so I guess the reverse operation should check it too. > > Otherwise, the 'is_added' is useless and should be removed :-) > > Then I suggest you document this in the changelog message for the patch > > :) Zhao, can you send an updated patch? Also is this 2.6.29 material (i.e. does it fix a reported bug)? -- Jesse Barnes, Intel Open Source Technology Center -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
