From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:43283) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aiYu3-0005tH-KN for qemu-devel@nongnu.org; Tue, 22 Mar 2016 22:52:13 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1aiYu0-0001bH-8w for qemu-devel@nongnu.org; Tue, 22 Mar 2016 22:52:11 -0400 Date: Wed, 23 Mar 2016 13:53:16 +1100 From: David Gibson Message-ID: <20160323025315.GS23586@voom.redhat.com> References: <1458546426-26222-1-git-send-email-aik@ozlabs.ru> <1458546426-26222-18-git-send-email-aik@ozlabs.ru> <20160322051449.GG23586@voom.redhat.com> <56F0DDFF.7050308@ozlabs.ru> <20160323010844.GO23586@voom.redhat.com> <56F1FBAB.6090308@ozlabs.ru> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="UTgPAcoWnJ37G5eg" Content-Disposition: inline In-Reply-To: <56F1FBAB.6090308@ozlabs.ru> Subject: Re: [Qemu-devel] [PATCH qemu v14 17/18] vfio/spapr: Use VFIO_SPAPR_TCE_v2_IOMMU List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alexey Kardashevskiy Cc: Alex Williamson , qemu-ppc@nongnu.org, qemu-devel@nongnu.org --UTgPAcoWnJ37G5eg Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Mar 23, 2016 at 01:12:59PM +1100, Alexey Kardashevskiy wrote: > On 03/23/2016 12:08 PM, David Gibson wrote: > >On Tue, Mar 22, 2016 at 04:54:07PM +1100, Alexey Kardashevskiy wrote: > >>On 03/22/2016 04:14 PM, David Gibson wrote: > >>>On Mon, Mar 21, 2016 at 06:47:05PM +1100, Alexey Kardashevskiy wrote: > >>>>New VFIO_SPAPR_TCE_v2_IOMMU type supports dynamic DMA window manageme= nt. > >>>>This adds ability to VFIO common code to dynamically allocate/remove > >>>>DMA windows in the host kernel when new VFIO container is added/remov= ed. > >>>> > >>>>This adds VFIO_IOMMU_SPAPR_TCE_CREATE ioctl to vfio_listener_region_a= dd > >>>>and adds just created IOMMU into the host IOMMU list; the opposite > >>>>action is taken in vfio_listener_region_del. > >>>> > >>>>When creating a new window, this uses euristic to decide on the TCE t= able > >>>>levels number. > >>>> > >>>>This should cause no guest visible change in behavior. > >>>> > >>>>Signed-off-by: Alexey Kardashevskiy > >>>>--- > >>>>Changes: > >>>>v14: > >>>>* new to the series > >>>> > >>>>--- > >>>>TODO: > >>>>* export levels to PHB > >>>>--- > >>>> hw/vfio/common.c | 108 ++++++++++++++++++++++++++++++++++++++++++++= ++++++++--- > >>>> trace-events | 2 ++ > >>>> 2 files changed, 105 insertions(+), 5 deletions(-) > >>>> > >>>>diff --git a/hw/vfio/common.c b/hw/vfio/common.c > >>>>index 4e873b7..421d6eb 100644 > >>>>--- a/hw/vfio/common.c > >>>>+++ b/hw/vfio/common.c > >>>>@@ -279,6 +279,14 @@ static int vfio_host_iommu_add(VFIOContainer *co= ntainer, > >>>> return 0; > >>>> } > >>>> > >>>>+static void vfio_host_iommu_del(VFIOContainer *container, hwaddr min= _iova) > >>>>+{ > >>>>+ VFIOHostIOMMU *hiommu =3D vfio_host_iommu_lookup(container, min_= iova, 0x1000); > >>> > >>>The hard-coded 0x1000 looks dubious.. > >> > >>Well, that's the minimal page size... > > > >Really? Some BookE CPUs support 1KiB page size.. >=20 > Hm. For IOMMU? Ok. s/0x1000/1/ should do then :) Uh.. actually I don't think those CPUs generally had an IOMMU. But if it's been done for CPU MMU I wouldn't count on it not being done for IOMMU. 1 is a safer choice. >=20 >=20 > > > >>>>+ g_assert(hiommu); > >>>>+ QLIST_REMOVE(hiommu, hiommu_next); > >>>>+} > >>>>+ > >>>> static bool vfio_listener_skipped_section(MemoryRegionSection *sect= ion) > >>>> { > >>>> return (!memory_region_is_ram(section->mr) && > >>>>@@ -392,6 +400,61 @@ static void vfio_listener_region_add(MemoryListe= ner *listener, > >>>> } > >>>> end =3D int128_get64(llend); > >>>> > >>>>+ if (container->iommu_type =3D=3D VFIO_SPAPR_TCE_v2_IOMMU) { > >>> > >>>I think this would be clearer split out into a helper function, > >>>vfio_create_host_window() or something. > >> > >> > >>It is rather vfio_spapr_create_host_window() and we were avoiding > >>xxx_spapr_xxx so far. I'd cut-n-paste the SPAPR PCI AS listener to a > >>separate file but this usually triggers more discussion and never ends = well. > >> > >> > >> > >>>>+ unsigned entries, pages; > >>>>+ struct vfio_iommu_spapr_tce_create create =3D { .argsz =3D s= izeof(create) }; > >>>>+ > >>>>+ g_assert(section->mr->iommu_ops); > >>>>+ g_assert(memory_region_is_iommu(section->mr)); > >>> > >>>I don't think you need these asserts. AFAICT the same logic should > >>>work if a RAM MR was added directly to PCI address space - this would > >>>create the new host window, then the existing code for adding a RAM MR > >>>would map that block of RAM statically into the new window. > >> > >>In what configuration/machine can we do that on SPAPR? > > > >spapr guests won't ever do that. But you can run an x86 guest on a > >powernv host and this situation could come up. >=20 >=20 > I am pretty sure VFIO won't work in this case anyway. I'm not. There's no fundamental reason VFIO shouldn't work with TCG. > >In any case there's no point asserting if the code is correct anyway. >=20 > Assert here says (at least) "not tested" or "not expected to > happen". Hmmm.. >=20 >=20 > > > >>>>+ trace_vfio_listener_region_add_iommu(iova, end - 1); > >>>>+ /* > >>>>+ * FIXME: For VFIO iommu types which have KVM acceleration to > >>>>+ * avoid bouncing all map/unmaps through qemu this way, this > >>>>+ * would be the right place to wire that up (tell the KVM > >>>>+ * device emulation the VFIO iommu handles to use). > >>>>+ */ > >>>>+ create.window_size =3D memory_region_size(section->mr); > >>>>+ create.page_shift =3D > >>>>+ ctz64(section->mr->iommu_ops->get_page_sizes(section= ->mr)); > >>> > >>>Ah.. except that I guess you'd need to fall back to host page size > >>>here to handle a RAM MR. > >> > >>Can you give an example of such RAM MR being added to PCI AS on > >>SPAPR? > > > >On spapr, no. But you can run other machine types as guests (at least > >with TCG) on a host with the spapr IOMMU. > > > >>>>+ /* > >>>>+ * SPAPR host supports multilevel TCE tables, there is some > >>>>+ * euristic to decide how many levels we want for our table: > >>>>+ * 0..64 =3D 1; 65..4096 =3D 2; 4097..262144 =3D 3; 262145..= =3D 4 > >>>>+ */ > >>>>+ entries =3D create.window_size >> create.page_shift; > >>>>+ pages =3D (entries * sizeof(uint64_t)) / getpagesize(); > >>>>+ create.levels =3D ctz64(pow2ceil(pages) - 1) / 6 + 1; > >>>>+ > >>>>+ ret =3D ioctl(container->fd, VFIO_IOMMU_SPAPR_TCE_CREATE, &c= reate); > >>>>+ if (ret) { > >>>>+ error_report("Failed to create a window, ret =3D %d (%m)= ", ret); > >>>>+ goto fail; > >>>>+ } > >>>>+ > >>>>+ if (create.start_addr !=3D section->offset_within_address_sp= ace || > >>>>+ vfio_host_iommu_lookup(container, create.start_addr, > >>>>+ create.start_addr + create.window= _size - 1)) { > >>> > >>>Under what circumstances can this trigger? Is the kernel ioctl > >>>allowed to return a different window start address than the one > >>>requested? > >> > >>You already asked this some time ago :) The userspace cannot request > >>address, the host kernel returns one. > > > >Ok. For generality it would be nice if you could succeed here as long > >as the new host window covers the requested guest window, even if it > >doesn't match exactly. And for that matter to not request the new > >window if the host already has a window covering the guest region. >=20 >=20 > That would be dead code - when would it possibly work? I mean I could > instrument an artificial test but the actual user which might appear later > will likely be soooo different so this won't help anyway. Hmm, I suppose. It actually shouldn't be that hard to trigger a case like this, if you just bumped the bridge's dma64 base address property up a little bit - above the host kernel's base address, but small enough that you can still easily fit the guest memory in. > >>>The second check looks very strange - if it returns true doesn't that > >>>mean you *do* have host window which can accomodate this guest region, > >>>which is what you want? > >> > >>This should not happen, this is what this check is for. Can make it ass= ert() > >>or something like this. > > > >Oh.. I see. Because you've done the ioctl, but not recorded the new > >host window in the list yet. > > > >No, I think the correct approach is to look for an existing host > >window containing the requested guest window *before* you try to > >create a new host window. If one is already there, you can just carry > >on. >=20 > Right, I'll change this. >=20 >=20 > > > >>>>+ struct vfio_iommu_spapr_tce_remove remove =3D { > >>>>+ .argsz =3D sizeof(remove), > >>>>+ .start_addr =3D create.start_addr > >>>>+ }; > >>>>+ error_report("Host doesn't support DMA window at %"HWADD= R_PRIx", must be %"PRIx64, > >>>>+ section->offset_within_address_space, > >>>>+ create.start_addr); > >>>>+ ioctl(container->fd, VFIO_IOMMU_SPAPR_TCE_REMOVE, &remov= e); > >>>>+ ret =3D -EINVAL; > >>>>+ goto fail; > >>>>+ } > >>>>+ trace_vfio_spapr_create_window(create.page_shift, > >>>>+ create.window_size, > >>>>+ create.start_addr); > >>>>+ > >>>>+ vfio_host_iommu_add(container, create.start_addr, > >>>>+ create.start_addr + create.window_size -= 1, > >>>>+ 1ULL << create.page_shift); > >>>>+ } > >>>>+ > >>>> if (!vfio_host_iommu_lookup(container, iova, end - 1)) { > >>>> error_report("vfio: IOMMU container %p can't map guest IOVA= region" > >>>> " 0x%"HWADDR_PRIx"..0x%"HWADDR_PRIx, > >>>>@@ -525,6 +588,22 @@ static void vfio_listener_region_del(MemoryListe= ner *listener, > >>>> container, iova, end - iova, ret); > >>>> } > >>>> > >>>>+ if (container->iommu_type =3D=3D VFIO_SPAPR_TCE_v2_IOMMU) { > >>>>+ struct vfio_iommu_spapr_tce_remove remove =3D { > >>>>+ .argsz =3D sizeof(remove), > >>>>+ .start_addr =3D section->offset_within_address_space, > >>>>+ }; > >>>>+ ret =3D ioctl(container->fd, VFIO_IOMMU_SPAPR_TCE_REMOVE, &r= emove); > >>>>+ if (ret) { > >>>>+ error_report("Failed to remove window at %"PRIx64, > >>>>+ remove.start_addr); > >>>>+ } > >>>>+ > >>>>+ vfio_host_iommu_del(container, section->offset_within_addres= s_space); > >>>>+ > >>>>+ trace_vfio_spapr_remove_window(remove.start_addr); > >>>>+ } > >>>>+ > >>>> if (iommu && iommu->iommu_ops && iommu->iommu_ops->vfio_stop) { > >>>> iommu->iommu_ops->vfio_stop(section->mr); > >>>> } > >>>>@@ -928,11 +1007,30 @@ static int vfio_connect_container(VFIOGroup *g= roup, AddressSpace *as) > >>>> goto listener_release_exit; > >>>> } > >>>> > >>>>- /* The default table uses 4K pages */ > >>>>- vfio_host_iommu_add(container, info.dma32_window_start, > >>>>- info.dma32_window_start + > >>>>- info.dma32_window_size - 1, > >>>>- 0x1000); > >>>>+ if (v2) { > >>>>+ /* > >>>>+ * There is a default window in just created container. > >>>>+ * To make region_add/del simpler, we better remove this > >>>>+ * window now and let those iommu_listener callbacks > >>>>+ * create/remove them when needed. > >>>>+ */ > >>>>+ struct vfio_iommu_spapr_tce_remove remove =3D { > >>>>+ .argsz =3D sizeof(remove), > >>>>+ .start_addr =3D info.dma32_window_start, > >>>>+ }; > >>>>+ ret =3D ioctl(fd, VFIO_IOMMU_SPAPR_TCE_REMOVE, &remove); > >>>>+ if (ret) { > >>>>+ error_report("vfio: VFIO_IOMMU_SPAPR_TCE_REMOVE fail= ed: %m"); > >>>>+ ret =3D -errno; > >>>>+ goto free_container_exit; > >>>>+ } > >>>>+ } else { > >>>>+ /* The default table uses 4K pages */ > >>>>+ vfio_host_iommu_add(container, info.dma32_window_start, > >>>>+ info.dma32_window_start + > >>>>+ info.dma32_window_size - 1, > >>>>+ 0x1000); > >>>>+ } > >>>> } else { > >>>> error_report("vfio: No available IOMMU models"); > >>>> ret =3D -EINVAL; > >>>>diff --git a/trace-events b/trace-events > >>>>index cc619e1..f2b75a3 100644 > >>>>--- a/trace-events > >>>>+++ b/trace-events > >>>>@@ -1736,6 +1736,8 @@ vfio_region_finalize(const char *name, int inde= x) "Device %s, region %d" > >>>> vfio_region_mmaps_set_enabled(const char *name, bool enabled) "Regi= on %s mmaps enabled: %d" > >>>> vfio_ram_register(uint64_t va, uint64_t size, int ret) "va=3D%"PRIx= 64" size=3D%"PRIx64" ret=3D%d" > >>>> vfio_ram_unregister(uint64_t va, uint64_t size, int ret) "va=3D%"PR= Ix64" size=3D%"PRIx64" ret=3D%d" > >>>>+vfio_spapr_create_window(int ps, uint64_t ws, uint64_t off) "pageshi= ft=3D0x%x winsize=3D0x%"PRIx64" offset=3D0x%"PRIx64 > >>>>+vfio_spapr_remove_window(uint64_t off) "offset=3D%"PRIx64 > >>>> > >>>> # hw/vfio/platform.c > >>>> vfio_platform_base_device_init(char *name, int groupid) "%s belongs= to group #%d" > >>> > >> > >> > > >=20 >=20 --=20 David Gibson | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson --UTgPAcoWnJ37G5eg Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJW8gUbAAoJEGw4ysog2bOSIL4QAJHwGXmS75sBlnHQZ4RVLFWW GoXw15SGndKGvvwJrzI1r0YUjHtIYuAfWn2M258UVT5YwAVQKcEph50QUn8qxG5Z KFKB0jkptxYaXvN0hylaJZThIeZk5pXDUgotkQfwBeIo8G2Qr7+9EZNA4NrlbTfo pjO2REq0p63j0nEQ/mnbH8qtS0xB/NaT5qwBB108czfQqQv8xQXFVmgTZEu5hZh4 ndUSXz1lvYyKd8CEZ/vKKLyMIserpxUj5hJsq/MVSFrAARYRErajVNrNL/Peot6t HtNWKw5rS+CPtQoOTqrVH0lv12bnkoXP46U2kSBjoXhmeeHSSFfKgRErvKvbhH/i E/rrmjGEHtzidoHheUUvOvCGAInfDqaOtAl9kk8NBWZJZPWZAUdOA+rWaZg7h1Gx faQ5Dg/K9E2T4pq93dO8nGscgdMG2vdFkHZgZULo1Wq5N+ItnqYoO2aIQNOVzmzy KGgwb1MIDagBClqlb6dXNIYg/L5MvYu57rc82Ocqn8PmeemIIiKhjEUGDKitbMP9 TYTF+FjqiiTOSSVoapDgX0TuGUOxcVDtaO93zPQzf1lImoBj06KMg02V079B19ek vWi5oxWQlMNor7jCjuODPh93KpbkrSC5KQjCohUAPkMMeedlqvP2HZrVP0wGEJJg J634CGiLyqhsl5H0GYVI =OKX0 -----END PGP SIGNATURE----- --UTgPAcoWnJ37G5eg--