From: Alex Williamson <alex.williamson@redhat.com> To: Jason Gunthorpe <jgg@nvidia.com> Cc: Yi Liu <yi.l.liu@intel.com>, "Tian, Kevin" <kevin.tian@intel.com>, "Peng, Chao P" <chao.p.peng@intel.com>, "Sun, Yi Y" <yi.y.sun@intel.com>, "qemu-devel@nongnu.org" <qemu-devel@nongnu.org>, "david@gibson.dropbear.id.au" <david@gibson.dropbear.id.au>, "thuth@redhat.com" <thuth@redhat.com>, "farman@linux.ibm.com" <farman@linux.ibm.com>, "mjrosato@linux.ibm.com" <mjrosato@linux.ibm.com>, "akrowiak@linux.ibm.com" <akrowiak@linux.ibm.com>, "pasic@linux.ibm.com" <pasic@linux.ibm.com>, "jjherne@linux.ibm.com" <jjherne@linux.ibm.com>, "jasowang@redhat.com" <jasowang@redhat.com>, "kvm@vger.kernel.org" <kvm@vger.kernel.org>, "nicolinc@nvidia.com" <nicolinc@nvidia.com>, "eric.auger@redhat.com" <eric.auger@redhat.com>, "eric.auger.pro@gmail.com" <eric.auger.pro@gmail.com>, "peterx@redhat.com" <peterx@redhat.com> Subject: Re: [RFC 15/18] vfio/iommufd: Implement iommufd backend Date: Tue, 26 Apr 2022 14:59:31 -0600 [thread overview] Message-ID: <20220426145931.23cb976b.alex.williamson@redhat.com> (raw) In-Reply-To: <20220426192703.GS2125828@nvidia.com> On Tue, 26 Apr 2022 16:27:03 -0300 Jason Gunthorpe <jgg@nvidia.com> wrote: > On Tue, Apr 26, 2022 at 12:45:41PM -0600, Alex Williamson wrote: > > On Tue, 26 Apr 2022 11:11:56 -0300 > > Jason Gunthorpe <jgg@nvidia.com> wrote: > > > > > On Tue, Apr 26, 2022 at 10:08:30PM +0800, Yi Liu wrote: > > > > > > > > I think it is strange that the allowed DMA a guest can do depends on > > > > > the order how devices are plugged into the guest, and varys from > > > > > device to device? > > > > > > > > > > IMHO it would be nicer if qemu would be able to read the new reserved > > > > > regions and unmap the conflicts before hot plugging the new device. We > > > > > don't have a kernel API to do this, maybe we should have one? > > > > > > > > For userspace drivers, it is fine to do it. For QEMU, it's not quite easy > > > > since the IOVA is GPA which is determined per the e820 table. > > > > > > Sure, that is why I said we may need a new API to get this data back > > > so userspace can fix the address map before attempting to attach the > > > new device. Currently that is not possible at all, the device attach > > > fails and userspace has no way to learn what addresses are causing > > > problems. > > > > We have APIs to get the IOVA ranges, both with legacy vfio and the > > iommufd RFC, QEMU could compare these, but deciding to remove an > > existing mapping is not something to be done lightly. > > Not quite, you can get the IOVA ranges after you attach the device, > but device attach will fail if the new range restrictions intersect > with the existing mappings. So we don't have an easy way to learn the > new range restriction in a way that lets userspace ensure an attach > will not fail due to reserved ranged overlapping with mappings. > > The best you could do is make a dummy IOAS then attach the device, > read the mappings, detatch, and then do your unmaps. Right, the same thing the kernel does currently. > I'm imagining something like IOMMUFD_DEVICE_GET_RANGES that can be > called prior to attaching on the device ID. Something like /sys/kernel/iommu_groups/$GROUP/reserved_regions? > > We must be absolutely certain that there is no DMA to that range > > before doing so. > > Yes, but at the same time if the VM thinks it can DMA to that memory > then it is quite likely to DMA to it with the new device that doesn't > have it mapped in the first place. Sorry, this assertion doesn't make sense to me. We can't assume a vIOMMU on x86, so QEMU typically maps the entire VM address space (ie. device address space == system memory). Some of those mappings are likely DMA targets (RAM), but only a tiny fraction of the address space may actually be used for DMA. Some of those mappings are exceedingly unlikely P2P DMA targets (device memory), so we don't consider mapping failures to be fatal to attaching the device. If we have a case where a range failed for one device but worked for a previous, we're in the latter scenario, because we should have failed the device attach otherwise. Your assertion would require that there are existing devices (plural) making use of this mapping and that the new device is also likely to make use of this mapping. I have a hard time believing that evidence exists to support that statement. > It is also a bit odd that the behavior depends on the order the > devices are installed as if you plug the narrower device first then > the next device will happily use the narrower ranges, but viceversa > will get a different result. P2P use cases are sufficiently rare that this hasn't been an issue. I think there's also still a sufficient healthy dose of FUD whether a system supports P2P that drivers do some validation before relying on it. > This is why I find it bit strange that qemu doesn't check the > ranges. eg I would expect that anything declared as memory in the E820 > map has to be mappable to the iommu_domain or the device should not > attach at all. You have some interesting assumptions around associating MemoryRegionSegments from the device AddressSpace to something like an x86 specific E820 table. The currently used rule of thumb is that if we think it's memory, mapping failure is fatal to the device, otherwise it's not. If we want each device to have the most complete mapping possible, then we'd use a container per device, but that implies a lot of extra overhead. Instead we try to attach the device to an existing container within the address space and assume if it was good enough there, it's good enough here. > The P2P is a bit trickier, and I know we don't have a good story > because we lack ACPI description, but I would have expected the same > kind of thing. Anything P2Pable should be in the iommu_domain or the > device should not attach. As with system memory there are only certain > parts of the E820 map that an OS would use for P2P. > > (ideally ACPI would indicate exactly what combinations of devices are > P2Pable and then qemu would use that drive the mandatory address > ranges in the IOAS) How exactly does ACPI indicate that devices can do P2P? How can we rely on ACPI for a problem that's not unique to platforms that implement ACPI? > > > > yeah. qemu can filter the P2P BAR mapping and just stop it in qemu. We > > > > haven't added it as it is something you will add in future. so didn't > > > > add it in this RFC. :-) Please let me know if it feels better to filter > > > > it from today. > > > > > > I currently hope it will use a different map API entirely and not rely > > > on discovering the P2P via the VMA. eg using a DMABUF FD or something. > > > > > > So blocking it in qemu feels like the right thing to do. > > > > Wait a sec, so legacy vfio supports p2p between devices, which has a > > least a couple known use cases, primarily involving GPUs for at least > > one of the peers, and we're not going to make equivalent support a > > feature requirement for iommufd? > > I said "different map API" - something like IOMMU_FD_MAP_DMABUF > perhaps. For future support, yes, but your last sentence above states to outright block it for now, which would be a visible feature regression vs legacy vfio. > The trouble with taking in a user pointer to MMIO memory is that it > becomes quite annoying to go from a VMA back to the actual owner > object so we can establish proper refcounting and lifetime of struct-page-less > memory. Requiring userspace to make that connection via a FD > simplifies and generalizes this. > > So, qemu would say 'oh this memory is exported by VFIO, I will do > VFIO_EXPORT_DMA_BUF, then do IOMMU_FD_MAP_DMABUF, then close the FD' > > For vfio_compat we'd have to build some hacky compat approach to > discover the dmabuf for vfio-pci from the VMA. > > But if qemu is going this way with a new implementation I would prefer > the new implementation use the new way, when we decide what it should > be. > > As I mentioned before I would like to use DMABUF since I already have > a use-case to expose DMABUF from vfio-pci to connect to RDMA. I will > post the vfio DMABUF patch I have already. I'm not suggesting there aren't issues with P2P mappings, we all know that legacy vfio has various issues currently. I'm only stating that there are use cases for it and if we cannot support those use cases then we can't do a transparent switch to iommufd when it's available. Switching would depend not only on kernel/QEMU support, but the necessary features for the VM, where we have no means to programmatically determine the latter. Thanks, Alex
WARNING: multiple messages have this Message-ID (diff)
From: Alex Williamson <alex.williamson@redhat.com> To: Jason Gunthorpe <jgg@nvidia.com> Cc: "akrowiak@linux.ibm.com" <akrowiak@linux.ibm.com>, "jjherne@linux.ibm.com" <jjherne@linux.ibm.com>, "Tian, Kevin" <kevin.tian@intel.com>, "Peng, Chao P" <chao.p.peng@intel.com>, "kvm@vger.kernel.org" <kvm@vger.kernel.org>, "mjrosato@linux.ibm.com" <mjrosato@linux.ibm.com>, "farman@linux.ibm.com" <farman@linux.ibm.com>, "jasowang@redhat.com" <jasowang@redhat.com>, "qemu-devel@nongnu.org" <qemu-devel@nongnu.org>, "peterx@redhat.com" <peterx@redhat.com>, "pasic@linux.ibm.com" <pasic@linux.ibm.com>, "eric.auger@redhat.com" <eric.auger@redhat.com>, "Sun, Yi Y" <yi.y.sun@intel.com>, Yi Liu <yi.l.liu@intel.com>, "nicolinc@nvidia.com" <nicolinc@nvidia.com>, "thuth@redhat.com" <thuth@redhat.com>, "eric.auger.pro@gmail.com" <eric.auger.pro@gmail.com>, "david@gibson.dropbear.id.au" <david@gibson.dropbear.id.au> Subject: Re: [RFC 15/18] vfio/iommufd: Implement iommufd backend Date: Tue, 26 Apr 2022 14:59:31 -0600 [thread overview] Message-ID: <20220426145931.23cb976b.alex.williamson@redhat.com> (raw) In-Reply-To: <20220426192703.GS2125828@nvidia.com> On Tue, 26 Apr 2022 16:27:03 -0300 Jason Gunthorpe <jgg@nvidia.com> wrote: > On Tue, Apr 26, 2022 at 12:45:41PM -0600, Alex Williamson wrote: > > On Tue, 26 Apr 2022 11:11:56 -0300 > > Jason Gunthorpe <jgg@nvidia.com> wrote: > > > > > On Tue, Apr 26, 2022 at 10:08:30PM +0800, Yi Liu wrote: > > > > > > > > I think it is strange that the allowed DMA a guest can do depends on > > > > > the order how devices are plugged into the guest, and varys from > > > > > device to device? > > > > > > > > > > IMHO it would be nicer if qemu would be able to read the new reserved > > > > > regions and unmap the conflicts before hot plugging the new device. We > > > > > don't have a kernel API to do this, maybe we should have one? > > > > > > > > For userspace drivers, it is fine to do it. For QEMU, it's not quite easy > > > > since the IOVA is GPA which is determined per the e820 table. > > > > > > Sure, that is why I said we may need a new API to get this data back > > > so userspace can fix the address map before attempting to attach the > > > new device. Currently that is not possible at all, the device attach > > > fails and userspace has no way to learn what addresses are causing > > > problems. > > > > We have APIs to get the IOVA ranges, both with legacy vfio and the > > iommufd RFC, QEMU could compare these, but deciding to remove an > > existing mapping is not something to be done lightly. > > Not quite, you can get the IOVA ranges after you attach the device, > but device attach will fail if the new range restrictions intersect > with the existing mappings. So we don't have an easy way to learn the > new range restriction in a way that lets userspace ensure an attach > will not fail due to reserved ranged overlapping with mappings. > > The best you could do is make a dummy IOAS then attach the device, > read the mappings, detatch, and then do your unmaps. Right, the same thing the kernel does currently. > I'm imagining something like IOMMUFD_DEVICE_GET_RANGES that can be > called prior to attaching on the device ID. Something like /sys/kernel/iommu_groups/$GROUP/reserved_regions? > > We must be absolutely certain that there is no DMA to that range > > before doing so. > > Yes, but at the same time if the VM thinks it can DMA to that memory > then it is quite likely to DMA to it with the new device that doesn't > have it mapped in the first place. Sorry, this assertion doesn't make sense to me. We can't assume a vIOMMU on x86, so QEMU typically maps the entire VM address space (ie. device address space == system memory). Some of those mappings are likely DMA targets (RAM), but only a tiny fraction of the address space may actually be used for DMA. Some of those mappings are exceedingly unlikely P2P DMA targets (device memory), so we don't consider mapping failures to be fatal to attaching the device. If we have a case where a range failed for one device but worked for a previous, we're in the latter scenario, because we should have failed the device attach otherwise. Your assertion would require that there are existing devices (plural) making use of this mapping and that the new device is also likely to make use of this mapping. I have a hard time believing that evidence exists to support that statement. > It is also a bit odd that the behavior depends on the order the > devices are installed as if you plug the narrower device first then > the next device will happily use the narrower ranges, but viceversa > will get a different result. P2P use cases are sufficiently rare that this hasn't been an issue. I think there's also still a sufficient healthy dose of FUD whether a system supports P2P that drivers do some validation before relying on it. > This is why I find it bit strange that qemu doesn't check the > ranges. eg I would expect that anything declared as memory in the E820 > map has to be mappable to the iommu_domain or the device should not > attach at all. You have some interesting assumptions around associating MemoryRegionSegments from the device AddressSpace to something like an x86 specific E820 table. The currently used rule of thumb is that if we think it's memory, mapping failure is fatal to the device, otherwise it's not. If we want each device to have the most complete mapping possible, then we'd use a container per device, but that implies a lot of extra overhead. Instead we try to attach the device to an existing container within the address space and assume if it was good enough there, it's good enough here. > The P2P is a bit trickier, and I know we don't have a good story > because we lack ACPI description, but I would have expected the same > kind of thing. Anything P2Pable should be in the iommu_domain or the > device should not attach. As with system memory there are only certain > parts of the E820 map that an OS would use for P2P. > > (ideally ACPI would indicate exactly what combinations of devices are > P2Pable and then qemu would use that drive the mandatory address > ranges in the IOAS) How exactly does ACPI indicate that devices can do P2P? How can we rely on ACPI for a problem that's not unique to platforms that implement ACPI? > > > > yeah. qemu can filter the P2P BAR mapping and just stop it in qemu. We > > > > haven't added it as it is something you will add in future. so didn't > > > > add it in this RFC. :-) Please let me know if it feels better to filter > > > > it from today. > > > > > > I currently hope it will use a different map API entirely and not rely > > > on discovering the P2P via the VMA. eg using a DMABUF FD or something. > > > > > > So blocking it in qemu feels like the right thing to do. > > > > Wait a sec, so legacy vfio supports p2p between devices, which has a > > least a couple known use cases, primarily involving GPUs for at least > > one of the peers, and we're not going to make equivalent support a > > feature requirement for iommufd? > > I said "different map API" - something like IOMMU_FD_MAP_DMABUF > perhaps. For future support, yes, but your last sentence above states to outright block it for now, which would be a visible feature regression vs legacy vfio. > The trouble with taking in a user pointer to MMIO memory is that it > becomes quite annoying to go from a VMA back to the actual owner > object so we can establish proper refcounting and lifetime of struct-page-less > memory. Requiring userspace to make that connection via a FD > simplifies and generalizes this. > > So, qemu would say 'oh this memory is exported by VFIO, I will do > VFIO_EXPORT_DMA_BUF, then do IOMMU_FD_MAP_DMABUF, then close the FD' > > For vfio_compat we'd have to build some hacky compat approach to > discover the dmabuf for vfio-pci from the VMA. > > But if qemu is going this way with a new implementation I would prefer > the new implementation use the new way, when we decide what it should > be. > > As I mentioned before I would like to use DMABUF since I already have > a use-case to expose DMABUF from vfio-pci to connect to RDMA. I will > post the vfio DMABUF patch I have already. I'm not suggesting there aren't issues with P2P mappings, we all know that legacy vfio has various issues currently. I'm only stating that there are use cases for it and if we cannot support those use cases then we can't do a transparent switch to iommufd when it's available. Switching would depend not only on kernel/QEMU support, but the necessary features for the VM, where we have no means to programmatically determine the latter. Thanks, Alex
next prev parent reply other threads:[~2022-04-26 20:59 UTC|newest] Thread overview: 125+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-04-14 10:46 [RFC 00/18] vfio: Adopt iommufd Yi Liu 2022-04-14 10:46 ` Yi Liu 2022-04-14 10:46 ` [RFC 01/18] scripts/update-linux-headers: Add iommufd.h Yi Liu 2022-04-14 10:46 ` Yi Liu 2022-04-14 10:46 ` [RFC 02/18] linux-headers: Import latest vfio.h and iommufd.h Yi Liu 2022-04-14 10:46 ` Yi Liu 2022-04-14 10:46 ` [RFC 03/18] hw/vfio/pci: fix vfio_pci_hot_reset_result trace point Yi Liu 2022-04-14 10:46 ` Yi Liu 2022-04-14 10:46 ` [RFC 04/18] vfio/pci: Use vbasedev local variable in vfio_realize() Yi Liu 2022-04-14 10:46 ` Yi Liu 2022-04-14 10:46 ` [RFC 05/18] vfio/common: Rename VFIOGuestIOMMU::iommu into ::iommu_mr Yi Liu 2022-04-14 10:46 ` Yi Liu 2022-04-14 10:46 ` [RFC 06/18] vfio/common: Split common.c into common.c, container.c and as.c Yi Liu 2022-04-14 10:46 ` [RFC 07/18] vfio: Add base object for VFIOContainer Yi Liu 2022-04-14 10:46 ` Yi Liu 2022-04-29 6:29 ` David Gibson 2022-04-29 6:29 ` David Gibson 2022-05-03 13:05 ` Yi Liu 2022-04-14 10:47 ` [RFC 08/18] vfio/container: Introduce vfio_[attach/detach]_device Yi Liu 2022-04-14 10:47 ` Yi Liu 2022-04-14 10:47 ` [RFC 09/18] vfio/platform: Use vfio_[attach/detach]_device Yi Liu 2022-04-14 10:47 ` Yi Liu 2022-04-14 10:47 ` [RFC 10/18] vfio/ap: " Yi Liu 2022-04-14 10:47 ` Yi Liu 2022-04-14 10:47 ` [RFC 11/18] vfio/ccw: " Yi Liu 2022-04-14 10:47 ` Yi Liu 2022-04-14 10:47 ` [RFC 12/18] vfio/container-obj: Introduce [attach/detach]_device container callbacks Yi Liu 2022-04-14 10:47 ` Yi Liu 2022-04-14 10:47 ` [RFC 13/18] vfio/container-obj: Introduce VFIOContainer reset callback Yi Liu 2022-04-14 10:47 ` Yi Liu 2022-04-14 10:47 ` [RFC 14/18] hw/iommufd: Creation Yi Liu 2022-04-14 10:47 ` Yi Liu 2022-04-14 10:47 ` [RFC 15/18] vfio/iommufd: Implement iommufd backend Yi Liu 2022-04-14 10:47 ` Yi Liu 2022-04-22 14:58 ` Jason Gunthorpe 2022-04-22 21:33 ` Alex Williamson 2022-04-22 21:33 ` Alex Williamson 2022-04-26 9:55 ` Yi Liu 2022-04-26 9:55 ` Yi Liu 2022-04-26 10:41 ` Tian, Kevin 2022-04-26 10:41 ` Tian, Kevin 2022-04-26 13:41 ` Jason Gunthorpe 2022-04-26 14:08 ` Yi Liu 2022-04-26 14:08 ` Yi Liu 2022-04-26 14:11 ` Jason Gunthorpe 2022-04-26 18:45 ` Alex Williamson 2022-04-26 18:45 ` Alex Williamson 2022-04-26 19:27 ` Jason Gunthorpe 2022-04-26 20:59 ` Alex Williamson [this message] 2022-04-26 20:59 ` Alex Williamson 2022-04-26 23:08 ` Jason Gunthorpe 2022-04-26 13:53 ` Jason Gunthorpe 2022-04-14 10:47 ` [RFC 16/18] vfio/iommufd: Add IOAS_COPY_DMA support Yi Liu 2022-04-14 10:47 ` Yi Liu 2022-04-14 10:47 ` [RFC 17/18] vfio/as: Allow the selection of a given iommu backend Yi Liu 2022-04-14 10:47 ` Yi Liu 2022-04-14 10:47 ` [RFC 18/18] vfio/pci: Add an iommufd option Yi Liu 2022-04-14 10:47 ` Yi Liu 2022-04-15 8:37 ` [RFC 00/18] vfio: Adopt iommufd Nicolin Chen 2022-04-17 10:30 ` Eric Auger 2022-04-17 10:30 ` Eric Auger 2022-04-19 3:26 ` Nicolin Chen 2022-04-25 19:40 ` Eric Auger 2022-04-25 19:40 ` Eric Auger 2022-04-18 8:49 ` Tian, Kevin 2022-04-18 8:49 ` Tian, Kevin 2022-04-18 12:09 ` Yi Liu 2022-04-18 12:09 ` Yi Liu 2022-04-25 19:51 ` Eric Auger 2022-04-25 19:51 ` Eric Auger 2022-04-25 19:55 ` Eric Auger 2022-04-25 19:55 ` Eric Auger 2022-04-26 8:39 ` Tian, Kevin 2022-04-26 8:39 ` Tian, Kevin 2022-04-22 22:09 ` Alex Williamson 2022-04-22 22:09 ` Alex Williamson 2022-04-25 10:10 ` Daniel P. Berrangé 2022-04-25 10:10 ` Daniel P. Berrangé 2022-04-25 13:36 ` Jason Gunthorpe 2022-04-25 14:37 ` Alex Williamson 2022-04-25 14:37 ` Alex Williamson 2022-04-26 8:37 ` Tian, Kevin 2022-04-26 8:37 ` Tian, Kevin 2022-04-26 12:33 ` Jason Gunthorpe 2022-04-26 16:21 ` Alex Williamson 2022-04-26 16:21 ` Alex Williamson 2022-04-26 16:42 ` Jason Gunthorpe 2022-04-26 19:24 ` Alex Williamson 2022-04-26 19:24 ` Alex Williamson 2022-04-26 19:36 ` Jason Gunthorpe 2022-04-28 3:21 ` Tian, Kevin 2022-04-28 3:21 ` Tian, Kevin 2022-04-28 14:24 ` Alex Williamson 2022-04-28 14:24 ` Alex Williamson 2022-04-28 16:20 ` Daniel P. Berrangé 2022-04-28 16:20 ` Daniel P. Berrangé 2022-04-29 0:45 ` Tian, Kevin 2022-04-29 0:45 ` Tian, Kevin 2022-04-25 20:23 ` Eric Auger 2022-04-25 20:23 ` Eric Auger 2022-04-25 22:53 ` Alex Williamson 2022-04-25 22:53 ` Alex Williamson 2022-04-26 9:47 ` Shameerali Kolothum Thodi via 2022-04-26 9:47 ` Shameerali Kolothum Thodi 2022-04-26 11:44 ` Eric Auger 2022-04-26 11:44 ` Eric Auger 2022-04-26 12:43 ` Shameerali Kolothum Thodi 2022-04-26 12:43 ` Shameerali Kolothum Thodi via 2022-04-26 16:35 ` Alex Williamson 2022-04-26 16:35 ` Alex Williamson 2022-05-09 14:24 ` Zhangfei Gao 2022-05-10 3:17 ` Yi Liu 2022-05-10 6:51 ` Eric Auger 2022-05-10 12:35 ` Zhangfei Gao 2022-05-10 12:45 ` Jason Gunthorpe 2022-05-10 14:08 ` Yi Liu 2022-05-11 14:17 ` zhangfei.gao 2022-05-12 9:01 ` zhangfei.gao 2022-05-17 8:55 ` Yi Liu 2022-05-18 7:22 ` zhangfei.gao 2022-05-18 14:00 ` Yi Liu 2022-06-28 8:14 ` Shameerali Kolothum Thodi 2022-06-28 8:14 ` Shameerali Kolothum Thodi via 2022-06-28 8:58 ` Eric Auger 2022-05-17 8:52 ` Yi Liu
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=20220426145931.23cb976b.alex.williamson@redhat.com \ --to=alex.williamson@redhat.com \ --cc=akrowiak@linux.ibm.com \ --cc=chao.p.peng@intel.com \ --cc=david@gibson.dropbear.id.au \ --cc=eric.auger.pro@gmail.com \ --cc=eric.auger@redhat.com \ --cc=farman@linux.ibm.com \ --cc=jasowang@redhat.com \ --cc=jgg@nvidia.com \ --cc=jjherne@linux.ibm.com \ --cc=kevin.tian@intel.com \ --cc=kvm@vger.kernel.org \ --cc=mjrosato@linux.ibm.com \ --cc=nicolinc@nvidia.com \ --cc=pasic@linux.ibm.com \ --cc=peterx@redhat.com \ --cc=qemu-devel@nongnu.org \ --cc=thuth@redhat.com \ --cc=yi.l.liu@intel.com \ --cc=yi.y.sun@intel.com \ /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.