[GIT PULL] arm64 updates for 6.1-rc1

[Catalin Marinas]

On Mon, Nov 14, 2022 at 02:09:04PM +0000, Will Deacon wrote:
> On Sat, Nov 12, 2022 at 12:48:20AM +0530, Manivannan Sadhasivam wrote:
> > I digged a little further and found that the crash was due to the secure
> > processor (XPU) violation. It happens because, CPU tried acccessing the memory
> > after sharing it with the modem for firmware metadata validation.
[...]
> > Sibi tried fixing this problem earlier by using a hack in the remoteproc driver
> > [1], but I guess that got negated due to c44094eee32f?
> 
> Performing a clean rather than a clean+invalidate when the buffer is
> allocated (which is what is achieved by c44094eee32f) shouldn't affect
> this afaict.

I agree. The DMA_ATTR_NO_KERNEL_MAPPING used in the qcom_q6v5_mss fix
only ensures that there is no non-cacheable DMA mapping but we still
have the kernel linear map in place that may be accessed speculatively
by the CPU at any point. I can't tell why the clean+inval makes any
difference but I suspect it's a matter of time before you'd hit similar
conditions again (maybe not at boot).

-- 
Catalin