Re: [PATCH] kernel/resource: optimize find_next_iomem_res

From: Ilpo Järvinen
Date: Mon Jun 03 2024 - 03:28:44 EST


On Fri, 31 May 2024, Chia-I Wu wrote:
> On Fri, May 31, 2024 at 1:57 AM Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx>
> wrote:
> On Thu, May 30, 2024 at 10:36:57PM -0700, Chia-I Wu wrote:
> > We can skip children resources when the parent resource does not cover
> > the range.
> >
> > This should help vmf_insert_* users on x86, such as several DRM
> drivers.
> > On my AMD Ryzen 5 7520C, when streaming data from cpu memory into
> amdgpu
> > bo, the throughput goes from 5.1GB/s to 6.6GB/s.  perf report says
> >
> >   34.69%--__do_fault
> >   34.60%--amdgpu_gem_fault
> >   34.00%--ttm_bo_vm_fault_reserved
> >   32.95%--vmf_insert_pfn_prot
> >   25.89%--track_pfn_insert
> >   24.35%--lookup_memtype
> >   21.77%--pat_pagerange_is_ram
> >   20.80%--walk_system_ram_range
> >   17.42%--find_next_iomem_res
> >
> > before this change, and
> >
> >   26.67%--__do_fault
> >   26.57%--amdgpu_gem_fault
> >   25.83%--ttm_bo_vm_fault_reserved
> >   24.40%--vmf_insert_pfn_prot
> >   14.30%--track_pfn_insert
> >   12.20%--lookup_memtype
> >   9.34%--pat_pagerange_is_ram
> >   8.22%--walk_system_ram_range
> >   5.09%--find_next_iomem_res
> >
> > after.
>
> Is there any documentation that explicitly says that the children
> resources
> must not overlap parent's one? Do we have some test cases? (Either way
> they
> needs to be added / expanded).
>
> I think it's the opposite.  The assumption here is that a child is always a subset of
> its parent.  Thus, if the range to be checked is not covered by a parent, we can skip
> the children.
>
> That's guaranteed by __request_resource.  I am less sure about __insert_resource but
> it appears to be the case too.  FWIW, resource_is_exclusive has the same assumption
> already.

Yes, the children resources are contained within the parent resource (at
least in PCI but given the code, I'd expect that to be general state of
affairs).

> It looks like I need to do some refactoring to add tests.
>
>
> P.S> I'm not so sure about this change. It needs a thoroughly testing,
> esp.
> in PCI case. Cc'ing to Ilpo.
>
> What's special about PCI?

--
i.