[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: i915 "GPU HANG", bisected to a2daa27c0c61 "swiotlb: simplify swiotlb_max_segment"
- To: Juergen Gross <jgross@xxxxxxxx>
- From: Christoph Hellwig <hch@xxxxxx>
- Date: Tue, 18 Oct 2022 16:55:50 +0200
- Cc: Christoph Hellwig <hch@xxxxxx>, Jan Beulich <jbeulich@xxxxxxxx>, Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>, Anshuman Khandual <anshuman.khandual@xxxxxxx>, Stefano Stabellini <sstabellini@xxxxxxxxxx>, Oleksandr Tyshchenko <oleksandr_tyshchenko@xxxxxxxx>, regressions@xxxxxxxxxxxxxxx, xen-devel@xxxxxxxxxxxxxxxxxxxx, iommu@xxxxxxxxxxxxxxx, Robert Beckett <bob.beckett@xxxxxxxxxxxxx>, Jani Nikula <jani.nikula@xxxxxxxxxxxxxxx>, Joonas Lahtinen <joonas.lahtinen@xxxxxxxxxxxxxxx>, Rodrigo Vivi <rodrigo.vivi@xxxxxxxxx>, Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxxxxxxxx>, Matthew Auld <matthew.auld@xxxxxxxxx>, intel-gfx@xxxxxxxxxxxxxxxxxxxxx, dri-devel@xxxxxxxxxxxxxxxxxxxxx, Marek Marczykowski-Górecki <marmarek@xxxxxxxxxxxxxxxxxxxxxx>
- Delivery-date: Tue, 18 Oct 2022 14:55:56 +0000
- List-id: Xen developer discussion <xen-devel.lists.xenproject.org>
On Tue, Oct 18, 2022 at 04:53:50PM +0200, Juergen Gross wrote:
>> If we don't need the IS_ENABLED is not needed I'm all for dropping it.
>> But unless I misread the code, on arm/arm64 even PV guests are 1:1
>> mapped so that all Linux physically contigous memory also is Xen
>> contigous, so we don't need the hack.
>
> There are no PV guests on arm/arm64.
Ok, that's the part I was missing. In that case we should be fine
without the IS_ENABLED indeed.
|
