On 15.12.2023 15:18, Roger Pau Monne wrote:
> The current loop that iterates from 0 to the maximum RAM address in order to
> setup the IOMMU mappings is highly inefficient, and it will get worse as the
> amount of RAM increases. It's also not accounting for any reserved regions
> past the last RAM address.
>
> Instead of iterating over memory addresses, iterate over the memory map
> regions
> and use a rangeset in order to keep track of which ranges need to be identity
> mapped in the hardware domain physical address space.
>
> On an AMD EPYC 7452 with 512GiB of RAM, the time to execute
> arch_iommu_hwdom_init() in nanoseconds is:
>
> x old
> + new
> N Min Max Median Avg Stddev
> x 5 2.2364154e+10 2.338244e+10 2.2474685e+10 2.2622409e+10 4.2949869e+08
> + 5 1025012 1033036 1026188 1028276.2 3623.1194
> Difference at 95.0% confidence
> -2.26214e+10 +/- 4.42931e+08
> -99.9955% +/- 9.05152e-05%
> (Student's t, pooled s = 3.03701e+08)
>
> Execution time of arch_iommu_hwdom_init() goes down from ~22s to ~0.001s.
>
> Note there's a change for HVM domains (ie: PVH dom0) that get switched to
> create the p2m mappings using map_mmio_regions() instead of
> p2m_add_identity_entry(), so that ranges can be mapped with a single function
> call if possible. Note that the interface of map_mmio_regions() doesn't
> allow creating read-only mappings, but so far there are no such mappings
> created for PVH dom0 in arch_iommu_hwdom_init().
>
> No change intended in the resulting mappings that a hardware domain gets.
>
> Signed-off-by: Roger Pau Monné <roger....@citrix.com>
Reviewed-by: Jan Beulich <jbeul...@suse.com>
> @@ -487,58 +416,45 @@ void __hwdom_init arch_iommu_hwdom_init(struct domain
> *d)
> if ( !map )
> panic("IOMMU init: unable to allocate rangeset\n");
>
> - max_pfn = (GB(4) >> PAGE_SHIFT) - 1;
> - top = max(max_pdx, pfn_to_pdx(max_pfn) + 1);
> + if ( iommu_hwdom_inclusive )
> + {
> + /* Add the whole range below 4GB, UNUSABLE regions will be removed.
> */
> + rc = rangeset_add_range(map, 0, PFN_DOWN(GB(4)) - 1);
> + if ( rc )
> + panic("IOMMU inclusive mappings can't be added: %d\n",
> + rc);
As in the earlier patch, perhaps no need to wrap lines here (and again
further down)?
Jan
