From: Gerd Hoffmann <kraxel@redhat.com> To: Avi Kivity <avi@redhat.com> Cc: qemu-devel@nongnu.org, Andrea Arcangeli <aarcange@redhat.com>, chrisw@redhat.com, kvm@vger.kernel.org Subject: Re: [Qemu-devel] Re: [PATCH 2 of 5] add can_dma/post_dma for direct IO Date: Mon, 05 Jan 2009 11:27:52 +0100 [thread overview] Message-ID: <4961E0A8.5040400@redhat.com> (raw) In-Reply-To: <494FEE78.6020006@redhat.com> Avi Kivity wrote: > Ian Jackson wrote: >> I think there is continued value in being able to emulate a guest >> whose physical address space exceeds the virtual address space in the >> host. The whole assumption that all guest accessible RAM is mapped at >> once, contiguously, is I think wrong. > > If Xen has a problem with 64-bit hosts, we can try to > accommodate it, Yes, this is the whole point. Flesh out the DMA api in a way that Xen can use it. Large memory guests are only one reason why Xen needs map/unmap. The other one is grant table support. > but to have 32-bit qemu/tcg or qemu/kvm support large > address spaces is pointless IMO. You don't have to support that. I think the DMA api should look like this in the end: struct qemu_dma_ops { map(...); xfer(...); unmap(...); }; The qemu/xen implementation would actually map/unmap. For qemu/tcg and qemu/kvm which has all guest memory permanently mapped map() would be a guest-physical -> host-virtual address translation and unmap() would be a nop (or refcount--). cheers, Gerd
WARNING: multiple messages have this Message-ID (diff)
From: Gerd Hoffmann <kraxel@redhat.com> To: Avi Kivity <avi@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com>, chrisw@redhat.com, qemu-devel@nongnu.org, kvm@vger.kernel.org Subject: Re: [Qemu-devel] Re: [PATCH 2 of 5] add can_dma/post_dma for direct IO Date: Mon, 05 Jan 2009 11:27:52 +0100 [thread overview] Message-ID: <4961E0A8.5040400@redhat.com> (raw) In-Reply-To: <494FEE78.6020006@redhat.com> Avi Kivity wrote: > Ian Jackson wrote: >> I think there is continued value in being able to emulate a guest >> whose physical address space exceeds the virtual address space in the >> host. The whole assumption that all guest accessible RAM is mapped at >> once, contiguously, is I think wrong. > > If Xen has a problem with 64-bit hosts, we can try to > accommodate it, Yes, this is the whole point. Flesh out the DMA api in a way that Xen can use it. Large memory guests are only one reason why Xen needs map/unmap. The other one is grant table support. > but to have 32-bit qemu/tcg or qemu/kvm support large > address spaces is pointless IMO. You don't have to support that. I think the DMA api should look like this in the end: struct qemu_dma_ops { map(...); xfer(...); unmap(...); }; The qemu/xen implementation would actually map/unmap. For qemu/tcg and qemu/kvm which has all guest memory permanently mapped map() would be a guest-physical -> host-virtual address translation and unmap() would be a nop (or refcount--). cheers, Gerd
next prev parent reply other threads:[~2009-01-05 10:28 UTC|newest] Thread overview: 111+ messages / expand[flat|nested] mbox.gz Atom feed top 2008-12-12 18:16 [PATCH 0 of 5] dma api v3 Andrea Arcangeli 2008-12-12 18:16 ` [Qemu-devel] " Andrea Arcangeli 2008-12-12 18:16 ` [PATCH 1 of 5] fix cpu_physical_memory len Andrea Arcangeli 2008-12-12 18:16 ` [Qemu-devel] " Andrea Arcangeli 2008-12-12 19:06 ` Anthony Liguori 2008-12-12 19:06 ` [Qemu-devel] " Anthony Liguori 2008-12-12 19:26 ` Andrea Arcangeli 2008-12-12 19:26 ` [Qemu-devel] " Andrea Arcangeli 2008-12-12 18:16 ` [PATCH 2 of 5] add can_dma/post_dma for direct IO Andrea Arcangeli 2008-12-12 18:16 ` [Qemu-devel] " Andrea Arcangeli 2008-12-12 19:00 ` Blue Swirl 2008-12-12 19:00 ` Blue Swirl 2008-12-12 19:18 ` Anthony Liguori 2008-12-12 19:18 ` Anthony Liguori 2008-12-12 20:05 ` Blue Swirl 2008-12-12 20:10 ` Anthony Liguori 2008-12-12 20:10 ` Anthony Liguori 2008-12-12 19:15 ` Anthony Liguori 2008-12-12 19:15 ` [Qemu-devel] " Anthony Liguori 2008-12-12 19:37 ` Andrea Arcangeli 2008-12-12 19:37 ` [Qemu-devel] " Andrea Arcangeli 2008-12-12 20:09 ` Anthony Liguori 2008-12-12 20:09 ` [Qemu-devel] " Anthony Liguori 2008-12-12 20:25 ` Gerd Hoffmann 2008-12-12 20:25 ` [Qemu-devel] " Gerd Hoffmann 2008-12-12 19:39 ` Anthony Liguori 2008-12-12 19:39 ` [Qemu-devel] " Anthony Liguori 2008-12-13 9:22 ` Avi Kivity 2008-12-13 9:22 ` Avi Kivity 2008-12-13 16:45 ` Anthony Liguori 2008-12-13 16:45 ` Anthony Liguori 2008-12-13 19:48 ` Avi Kivity 2008-12-13 19:48 ` Avi Kivity 2008-12-13 21:07 ` Anthony Liguori 2008-12-13 21:07 ` Anthony Liguori 2008-12-14 6:03 ` Avi Kivity 2008-12-14 6:03 ` Avi Kivity 2008-12-14 19:10 ` Anthony Liguori 2008-12-14 19:10 ` Anthony Liguori 2008-12-14 19:49 ` Avi Kivity 2008-12-14 19:49 ` Avi Kivity 2008-12-14 23:08 ` Anthony Liguori 2008-12-14 23:08 ` Anthony Liguori 2008-12-15 0:57 ` Paul Brook 2008-12-15 0:57 ` Paul Brook 2008-12-15 2:09 ` Anthony Liguori 2008-12-15 2:09 ` Anthony Liguori 2008-12-15 6:23 ` Avi Kivity 2008-12-15 6:23 ` Avi Kivity 2008-12-15 18:35 ` Blue Swirl 2008-12-15 18:35 ` Blue Swirl 2008-12-15 22:06 ` Anthony Liguori 2008-12-15 22:06 ` Anthony Liguori 2008-12-16 9:41 ` Avi Kivity 2008-12-16 9:41 ` Avi Kivity 2008-12-16 16:55 ` Blue Swirl 2008-12-16 16:55 ` Blue Swirl 2008-12-16 17:09 ` Avi Kivity 2008-12-16 17:09 ` Avi Kivity 2008-12-16 17:48 ` Anthony Liguori 2008-12-16 17:48 ` Anthony Liguori 2008-12-16 18:11 ` Blue Swirl 2008-12-16 18:11 ` Blue Swirl 2008-12-16 15:57 ` Blue Swirl 2008-12-16 15:57 ` Blue Swirl 2008-12-16 16:29 ` Paul Brook 2008-12-16 16:29 ` Paul Brook 2008-12-16 16:35 ` Blue Swirl 2008-12-16 16:35 ` Blue Swirl 2008-12-14 17:30 ` Jamie Lokier 2008-12-14 17:30 ` Jamie Lokier 2008-12-13 14:39 ` Andrea Arcangeli 2008-12-13 14:39 ` [Qemu-devel] " Andrea Arcangeli 2008-12-13 16:46 ` Anthony Liguori 2008-12-13 16:46 ` [Qemu-devel] " Anthony Liguori 2008-12-13 16:53 ` Andrea Arcangeli 2008-12-13 16:53 ` [Qemu-devel] " Andrea Arcangeli 2008-12-13 17:54 ` Andreas Färber 2008-12-13 21:11 ` Anthony Liguori 2008-12-13 21:11 ` [Qemu-devel] " Anthony Liguori 2008-12-14 16:47 ` Andrea Arcangeli 2008-12-14 16:47 ` [Qemu-devel] " Andrea Arcangeli 2008-12-14 17:01 ` Avi Kivity 2008-12-14 17:01 ` [Qemu-devel] " Avi Kivity 2008-12-14 17:15 ` Andrea Arcangeli 2008-12-14 17:15 ` [Qemu-devel] " Andrea Arcangeli 2008-12-14 19:59 ` Avi Kivity 2008-12-14 19:59 ` [Qemu-devel] " Avi Kivity 2008-12-22 16:44 ` Ian Jackson 2008-12-22 16:44 ` Ian Jackson 2008-12-22 19:44 ` Avi Kivity 2008-12-23 0:03 ` Thiemo Seufer 2008-12-23 1:02 ` Andrea Arcangeli 2008-12-23 1:02 ` Andrea Arcangeli 2008-12-23 17:31 ` Avi Kivity 2008-12-23 17:31 ` Avi Kivity 2008-12-22 19:46 ` Avi Kivity 2009-01-05 10:27 ` Gerd Hoffmann [this message] 2009-01-05 10:27 ` Gerd Hoffmann 2008-12-13 22:47 ` Anthony Liguori 2008-12-13 22:47 ` [Qemu-devel] " Anthony Liguori 2008-12-14 6:07 ` Avi Kivity 2008-12-14 6:07 ` [Qemu-devel] " Avi Kivity 2008-12-12 18:16 ` [PATCH 3 of 5] rename dma.c to isa_dma.c Andrea Arcangeli 2008-12-12 18:16 ` [Qemu-devel] " Andrea Arcangeli 2008-12-12 18:16 ` [PATCH 4 of 5] dma api Andrea Arcangeli 2008-12-12 18:16 ` [Qemu-devel] " Andrea Arcangeli 2008-12-12 18:55 ` Blue Swirl 2008-12-12 18:55 ` Blue Swirl 2008-12-12 18:16 ` [PATCH 5 of 5] bdrv_aio_readv/writev Andrea Arcangeli 2008-12-12 18:16 ` [Qemu-devel] " Andrea Arcangeli
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=4961E0A8.5040400@redhat.com \ --to=kraxel@redhat.com \ --cc=aarcange@redhat.com \ --cc=avi@redhat.com \ --cc=chrisw@redhat.com \ --cc=kvm@vger.kernel.org \ --cc=qemu-devel@nongnu.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.