[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <YqbTfi/h2P24ynQZ@infradead.org>
Date: Sun, 12 Jun 2022 23:04:46 -0700
From: Christoph Hellwig <hch@...radead.org>
To: Dongli Zhang <dongli.zhang@...cle.com>
Cc: Christoph Hellwig <hch@...radead.org>,
iommu@...ts.linux-foundation.org, xen-devel@...ts.xenproject.org,
x86@...nel.org, linuxppc-dev@...ts.ozlabs.org,
virtualization@...ts.linux-foundation.org,
linux-kernel@...r.kernel.org, m.szyprowski@...sung.com,
jgross@...e.com, tglx@...utronix.de, mingo@...hat.com,
bp@...en8.de, dave.hansen@...ux.intel.com, sstabellini@...nel.org,
mpe@...erman.id.au, konrad.wilk@...cle.com, mst@...hat.com,
jasowang@...hat.com, joe.jin@...cle.com
Subject: Re: [PATCH RFC v1 4/7] swiotlb: to implement io_tlb_high_mem
On Fri, Jun 10, 2022 at 02:56:08PM -0700, Dongli Zhang wrote:
> Since this patch file has 200+ lines, would you please help clarify what does
> 'this' indicate?
This indicates that any choice of a different swiotlb pools needs to
be hidden inside of Ń•wiotlb. The dma mapping API already provides
swiotlb the addressability requirement for the device. Similarly we
already have a SWIOTLB_ANY flag that switches to a 64-bit buffer
by default, which we can change to, or replace with a flag that
allocates an additional buffer that is not addressing limited.
Powered by blists - more mailing lists