On Thu, Sep 07, 2023 at 01:12:23PM +0200, Petr Tesařík wrote:
> Hi all,
>
> sorry for my late reply; I've been away from my work setup for a
> month...
Please take a look at:
https://lore.kernel.org/linux-iommu/20230905064441.127588-1-...@lst.de/T/#u
Hi all,
sorry for my late reply; I've been away from my work setup for a
month...
On Wed, 30 Aug 2023 08:55:51 -0600
Jonathan Corbet wrote:
> So it seems this code got merged without this question ever being
> answered. Sorry if it's a dumb one, but I don't think this
> functionality works as
On Wed, Aug 09, 2023 at 03:20:43PM -0600, Jonathan Corbet wrote:
> > spin_unlock_irqrestore(&dev->dma_io_tlb_lock, flags);
> >
> > - /* Pairs with smp_rmb() in swiotlb_find_pool(). */
> > - smp_wmb();
> > found:
> > + dev->dma_uses_io_tlb = true;
> > + /* Pairs with smp_rmb() in is_s
So it seems this code got merged without this question ever being
answered. Sorry if it's a dumb one, but I don't think this
functionality works as advertised...
Thanks,
jon
Jonathan Corbet writes:
> Petr Tesarik writes:
>
>> From: Petr Tesarik
>>
>> Skip searching the software IO TLB if a
Petr Tesarik writes:
> From: Petr Tesarik
>
> Skip searching the software IO TLB if a device has never used it, making
> sure these devices are not affected by the introduction of multiple IO TLB
> memory pools.
>
> Additional memory barrier is required to ensure that the new value of the
> flag
From: Petr Tesarik
Skip searching the software IO TLB if a device has never used it, making
sure these devices are not affected by the introduction of multiple IO TLB
memory pools.
Additional memory barrier is required to ensure that the new value of the
flag is visible to other CPUs after mappi