For some drivers, that use the DMA API. This error message can be reached
several millions of times per second, causing spam to the kernel's printk
buffer and bringing the CPU usage up to 100% (so, it should be rate
limited). However, since there is at least one driver that is in the
mainline and
The pull request you sent on Fri, 10 Sep 2021 17:48:20 +0200:
> git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu.git
> tags/iommu-fixes-v5.15-rc0
has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/589e5cab170843b2f7f8260168ab2d77163d4384
Thank you!
--
Thanks. We did verify the correctness of this patch: we didn't see host
crash with guest reboot when this patch is applied.
Tested-by: Terry Bowman
Thanks,
-Wei
On 9/9/21 6:17 AM, Joerg Roedel wrote:
> Okay, after this triggered a defconfig compile warning, I squashed patch
> 1 and 2 into one
Hi Linus,
The following changes since commit d8768d7eb9c21ef928adb93402d9348bcc4a6915:
Merge branches 'apple/dart', 'arm/smmu', 'iommu/fixes', 'x86/amd', 'x86/vt-d'
and 'core' into next (2021-08-20 17:14:35 +0200)
are available in the Git repository at:
On Wed, Sep 08, 2021 at 05:58:33PM -0500, Tom Lendacky wrote:
> In prep for other confidential computing technologies, introduce a generic
preparation
> helper function, cc_platform_has(), that can be used to check for specific
> active confidential computing attributes, like memory encryption.
On 2021-09-10 13:05, Hamza Mahfooz wrote:
For some drivers, that call add_dma_entry() from somewhere down the call
stack.
Nit: strictly, drivers don't call add_dma_entry(). Drivers only call the
DMA API functions, and it is the DMA API internals which take a detour
through dma-debug when
For some drivers, that call add_dma_entry() from somewhere down the call
stack. If this error condition is triggered once, it causes the error
message to spam the kernel's printk buffer and bring the CPU usage up to
100%. Also, since there is at least one driver that is in the mainline
and suffers
On Wed, Sep 08, 2021 at 10:33:26PM -0500, Jeremy Linton wrote:
> PS, it might not hurt to rate limit/_once this somehow to avoid a runtime
> problem if it starts to trigger.
Yes, that might be a good idea. Care to prepare a patch?
___
iommu mailing
On Wed, Sep 08, 2021 at 10:33:26PM -0500, Jeremy Linton wrote:
> +DPAA2, netdev maintainers
> Hi,
>
> On 5/18/21 7:54 AM, Hamza Mahfooz wrote:
> > Since, overlapping mappings are not supported by the DMA API we should
> > report an error if active_cacheline_insert returns -EEXIST.
>
> It seems