On 2023/11/22 1:00, Jason Gunthorpe wrote:
> On Tue, Nov 21, 2023 at 03:47:48PM +0900, Hector Martin wrote:
>>> Which is sensitive only to !NULL fwspec, and if EPROBE_DEFER is
>>> returned fwspec will be freed and dev->iommu->fwspec will be NULL
>>> here.
>>>
>>> In the NULL case it does a 'bus pro
On Tue, Nov 21, 2023 at 03:47:48PM +0900, Hector Martin wrote:
> > Which is sensitive only to !NULL fwspec, and if EPROBE_DEFER is
> > returned fwspec will be freed and dev->iommu->fwspec will be NULL
> > here.
> >
> > In the NULL case it does a 'bus probe' with a NULL fwspec and all the
> > fwspe
On 2023/11/19 23:13, Jason Gunthorpe wrote:
> On Sun, Nov 19, 2023 at 06:19:43PM +0900, Hector Martin wrote:
+static int iommu_fwspec_assign_iommu(struct iommu_fwspec *fwspec,
+ struct device *dev,
+ struct fwnode_handle
On Sun, Nov 19, 2023 at 06:19:43PM +0900, Hector Martin wrote:
> >> +static int iommu_fwspec_assign_iommu(struct iommu_fwspec *fwspec,
> >> + struct device *dev,
> >> + struct fwnode_handle *iommu_fwnode)
> >> +{
> >> + const struct iommu
On 2023/11/19 17:10, Hector Martin wrote:
> On 2023/11/15 23:05, Jason Gunthorpe wrote:
>> Allow fwspec to exist independently from the dev->iommu by providing
>> functions to allow allocating and freeing the raw struct iommu_fwspec.
>>
>> Reflow the existing paths to call the new alloc/dealloc
On 2023/11/15 23:05, Jason Gunthorpe wrote:
> Allow fwspec to exist independently from the dev->iommu by providing
> functions to allow allocating and freeing the raw struct iommu_fwspec.
>
> Reflow the existing paths to call the new alloc/dealloc functions.
>
> Reviewed-by: Jerry Snitselaar
> S
