On Mon, Aug 26, 2019 at 04:17:03PM +1000, Alexey Kardashevskiy wrote:
> Invalidating a TCE cache entry for each updated TCE is quite expensive.
> This makes use of the new iommu_table_ops::xchg_no_kill()/tce_kill()
> callbacks to bring down the time spent in mapping a huge guest DMA window;
>
Invalidating a TCE cache entry for each updated TCE is quite expensive.
This makes use of the new iommu_table_ops::xchg_no_kill()/tce_kill()
callbacks to bring down the time spent in mapping a huge guest DMA window;
roughly 20s to 10s for each guest's 100GB of DMA space.
Signed-off-by: Alexey
