On Mon, Aug 21, 2023 at 04:36:49PM +0200, Joerg Roedel wrote:
> On Mon, Aug 21, 2023 at 09:51:13AM -0300, Jason Gunthorpe wrote:
> > So now that Joerg has dropped it - what is your big idea to make the
> > locking actually work right?
>
> I am not opposed to the general idea.
Well, I think Robin
On Mon, Aug 21, 2023 at 09:51:13AM -0300, Jason Gunthorpe wrote:
> So now that Joerg has dropped it - what is your big idea to make the
> locking actually work right?
I am not opposed to the general idea. When putting it into the tree I
wasn't aware how many users still need to be adapted to prope
On Mon, Aug 21, 2023 at 12:06:40PM +0100, Robin Murphy wrote:
> On 2023-08-18 22:32, Jason Gunthorpe wrote:
> > It is subtle and was never documented or enforced, but there has always
> > been an assumption that of_dma_configure_id() is not concurrent. It makes
> > several calls into the iommu laye
On Mon, Aug 21, 2023 at 02:27:47PM +0200, Joerg Roedel wrote:
> On Mon, Aug 21, 2023 at 12:06:40PM +0100, Robin Murphy wrote:
> > The solution is to drop those locking patches entirely and rethink the whole
> > thing.
>
> Agreed, that was exactly what I thought when looking at this patch.
>
> I d
On Mon, Aug 21, 2023 at 12:06:40PM +0100, Robin Murphy wrote:
> The solution is to drop those locking patches entirely and rethink the whole
> thing.
Agreed, that was exactly what I thought when looking at this patch.
I dropped the original 10 patches and the 4 fixes on-top from the IOMMU
tree. T
On 2023-08-18 22:32, Jason Gunthorpe wrote:
It turns out several drivers are calling of_dma_configure() outside the
expected bus_type.dma_configure op. This ends up being mis-locked and
triggers a lockdep assertion, or instance:
iommu_probe_device_locked+0xd4/0xe4
of_iommu_configure+0x10c/
On Fri, Aug 18, 2023 at 06:32:28PM -0300, Jason Gunthorpe wrote:
> It turns out several drivers are calling of_dma_configure() outside the
> expected bus_type.dma_configure op. This ends up being mis-locked and
> triggers a lockdep assertion, or instance:
Acked-by: Mark Brown
signature.asc
Desc
It turns out several drivers are calling of_dma_configure() outside the
expected bus_type.dma_configure op. This ends up being mis-locked and
triggers a lockdep assertion, or instance:
iommu_probe_device_locked+0xd4/0xe4
of_iommu_configure+0x10c/0x200
of_dma_configure_id+0x104/0x3b8
a6xx_g