On Sun, Feb 19, 2017 at 09:22:35PM -0700, Logan Gunthorpe wrote:
> Really, in any situation where there's a cdev and a device in the same
> structure, the life cycles of the two become linked but their reference
> counts are not and that is the problem here.
Yes, the cdev must hold a kref on the
On 19/02/17 02:43 PM, Dan Williams wrote:
> Is this race present for all file operations? I've only seen it with
> mmap() and late faults. So if these other drivers do not support mmap
> it's not clear they can trigger the failure.
I don't see how it's related to mmap only. I think there's a num
On Sat, Feb 18, 2017 at 12:22 PM, Logan Gunthorpe wrote:
> Hi,
>
> Please don't apply this patch and instead apply the switchtec driver as
> we submitted in v2. As per the discussion in [1], not using the cdev's
> kobj parent results in incorrect reference counting and a possible use
> of the cdev
Hi,
Please don't apply this patch and instead apply the switchtec driver as
we submitted in v2. As per the discussion in [1], not using the cdev's
kobj parent results in incorrect reference counting and a possible use
of the cdev after its containing structure is freed.
I've also done a quick rev
Per a suggestion from Greg Kroah-Hartman [1], don't set the cdev's
kobject parent. This allows us to use device_register instead of init
and add.
[1] https://lkml.org/lkml/2017/2/10/370
---
drivers/pci/switch/switchtec.c | 15 +++
1 file changed, 7 insertions(+), 8 deletions(-)
diff
