Changes in Kconfig

bca939d5 - vhost-scsi: Dynamically allocate scatterlists

Tue, 03 Dec 2024 19:15:12 +0000

vhost-scsi: Dynamically allocate scatterlistsWe currently preallocate scatterlists which have 2048 entries for eachcommand. For a small device with just 1 queue this results in:8 MB = 32 bytes per sg * 2048 entries * 128 cmdWhen mq is turned on and we increase the virtqueue_size so we can handlecommands from multiple queues in parallel, then this can sky rocket.This patch allows us to dynamically allocate the scatterlist like is donewith drivers like NVMe and SCSI.For small IO (4-16K) IOPs testing, we didn't see any regressions, butfor throughput testing we sometimes saw a 2-5% regression when thebackend device was very fast (8 NVMe drives in a MD RAID0 config or amemory backed device). As a result this patch makes the dynamicallocation feature a modparam so userspace can decide how it wants tobalance mem use and perf.Signed-off-by: Mike Christie Message-Id: <20241203191705.19431-6-michael.christie@oracle.com>Signed-off-by: Michael S. Tsirkin Acked-by: Stefan Hajnoczi List of files: /linux-6.15/drivers/vhost/Kconfig

e297cd54 - vhost_task: Allow vhost layer to use copy_process

Fri, 10 Mar 2023 22:03:30 +0000

vhost_task: Allow vhost layer to use copy_processQemu will create vhost devices in the kernel which perform network, SCSI,etc IO and management operations from worker threads created by thekthread API. Because the kthread API does a copy_process on the kthreaddthread, the vhost layer has to use kthread_use_mm to access the Qemuthread's memory and cgroup_attach_task_all to add itself to the Qemuthread's cgroups, and it bypasses the RLIMIT_NPROC limit which can resultin VMs creating more threads than the admin expected.This patch adds a new struct vhost_task which can be used instead ofkthreads. They allow the vhost layer to use copy_process and inheritthe userspace process's mm and cgroups, the task is accounted forunder the userspace's nproc count and can be seen in its process tree,and other features like namespaces work and are inherited by default.Signed-off-by: Mike Christie Acked-by: Michael S. Tsirkin Signed-off-by: Christian Brauner (Microsoft) Signed-off-by: Christian Brauner List of files: /linux-6.15/drivers/vhost/Kconfig

2cf1ba9a - vhost_vdpa: implement IRQ offloading in vhost_vdpa

Fri, 31 Jul 2020 06:55:31 +0000

vhost_vdpa: implement IRQ offloading in vhost_vdpaThis patch introduce a set of functions for setup/unsetupand update irq offloading respectively by register/unregisterand re-register the irq_bypass_producer.With these functions, this commit can setup/unsetupirq offloading through setting DRIVER_OK/!DRIVER_OK, andupdate irq offloading through SET_VRING_CALL.Signed-off-by: Zhu Lingshan Suggested-by: Jason Wang Link: https://lore.kernel.org/r/20200731065533.4144-5-lingshan.zhu@intel.comSigned-off-by: Michael S. Tsirkin List of files: /linux-6.15/drivers/vhost/Kconfig

a7f7f624 - treewide: replace '---help---' in Kconfig files with 'help'

Sat, 13 Jun 2020 16:50:22 +0000

treewide: replace '---help---' in Kconfig files with 'help'Since commit 84af7a6194e4 ("checkpatch: kconfig: prefer 'help' over'---help---'"), the number of '---help---' has been graduallydecreasing, but there are still more than 2400 instances.This commit finishes the conversion. While I touched the lines,I also fixed the indentation.There are a variety of indentation styles found. a) 4 spaces + '---help---' b) 7 spaces + '---help---' c) 8 spaces + '---help---' d) 1 space + 1 tab + '---help---' e) 1 tab + '---help---' (correct indentation) f) 1 tab + 1 space + '---help---' g) 1 tab + 2 spaces + '---help---'In order to convert all of them to 1 tab + 'help', I ran thefollowing commend: $ find . -name 'Kconfig*' | xargs sed -i 's/^[[:space:]]*---help---/\thelp/'Signed-off-by: Masahiro Yamada List of files: /linux-6.15/drivers/vhost/Kconfig

213e7721 - vhost: revert "vhost: disable for OABI"

Thu, 23 Apr 2020 12:36:57 +0000

vhost: revert "vhost: disable for OABI"This reverts commit d085eb8ce727 ("vhost: disable for OABI")With commit "virtio: force spec specified alignment on types"in place, we force proper alignment for all structures,so there's no longer a reason to blacklist OABI.Signed-off-by: Michael S. Tsirkin List of files: /linux-6.15/drivers/vhost/Kconfig

d085eb8c - vhost: disable for OABI

Mon, 06 Apr 2020 12:12:47 +0000

vhost: disable for OABIvhost is currently broken on the some ARM configs.The reason is that the ring element addresses are passed betweencomponents with different alignments assumptions. Thus, ifguest selects a pointer and host then gets and dereferencesit, then alignment assumed by the host's compiler might begreater than the actual alignment of the pointer.compiler on the host from assuming pointer is aligned.This actually triggers on ARM with -mabi=apcs-gnu - which is adeprecated configuration. With this OABI, compiler assumes thatall structures are 4 byte aligned - which is stronger thanvirtio guarantees for available and used rings, which aremerely 2 bytes. Thus a guest without -mabi=apcs-gnu runningon top of host with -mabi=apcs-gnu will be broken.The correct fix is to force alignment of structures - howeverthat is an intrusive fix that's best deferred until the next release.We didn't previously support such ancient systems at all - this surfacedafter vdpa support prompted removing dependency of vhost onVIRTULIZATION. So for now, let's just add something along the lines of depends on !ARM || AEABIto the virtio Kconfig declaration, and add a comment that it has to dowith struct member alignment.Note: we can't make VHOST and VHOST_RING themselves havea dependency since these are selected. Add a new symbol for that.We should be able to drop this dependency down the road.Fixes: 20c384f1ea1a0bc7 ("vhost: refine vhost and vringh kconfig")Suggested-by: Ard Biesheuvel Suggested-by: Richard Earnshaw Signed-off-by: Michael S. Tsirkin List of files: /linux-6.15/drivers/vhost/Kconfig

58ad1372 - vdpa: make vhost, virtio depend on menu

Sun, 12 Apr 2020 08:36:55 +0000

vdpa: make vhost, virtio depend on menuIf user did not configure any vdpa drivers, neither vhostnor virtio vdpa are going to be useful. So there's no pointin prompting for these and selecting vdpa core automatically.Simplify configuration by making virtio and vhost vdpadrivers depend on vdpa menu entry. Once done, we no longerneed a separate menu entry, so also get rid of this.While at it, fix up the IFC entry: VDPA->vDPA for consistencywith other places.Signed-off-by: Michael S. Tsirkin Acked-by: Jason Wang List of files: /linux-6.15/drivers/vhost/Kconfig

3302363a - virtio/test: fix up after IOTLB changes

Wed, 01 Apr 2020 16:46:22 +0000

virtio/test: fix up after IOTLB changesAllow building vringh without IOTLB (that's the case for userspacebuilds, will be useful for CAIF/VOD down the road too).Update for API tweaks.Don't include vringh with userspace builds.Cc: Jason Wang Cc: Eugenio Pérez Signed-off-by: Michael S. Tsirkin Acked-by: Jason Wang List of files: /linux-6.15/drivers/vhost/Kconfig

4c8cf318 - vhost: introduce vDPA-based backend

Thu, 26 Mar 2020 14:01:23 +0000

vhost: introduce vDPA-based backendThis patch introduces a vDPA-based vhost backend. This backend isbuilt on top of the same interface defined in virtio-vDPA and providesa generic vhost interface for userspace to accelerate the virtiodevices in guest.This backend is implemented as a vDPA device driver on top of the sameops used in virtio-vDPA. It will create char device entry namedvhost-vdpa-$index for userspace to use. Userspace can use vhost ioctlson top of this char device to setup the backend.Vhost ioctls are extended to make it type agnostic and behave like avirtio device, this help to eliminate type specific API like whatvhost_net/scsi/vsock did:- VHOST_VDPA_GET_DEVICE_ID: get the virtio device ID which is defined by virtio specification to differ from different type of devices- VHOST_VDPA_GET_VRING_NUM: get the maximum size of virtqueue supported by the vDPA device- VHSOT_VDPA_SET/GET_STATUS: set and get virtio status of vDPA device- VHOST_VDPA_SET/GET_CONFIG: access virtio config space- VHOST_VDPA_SET_VRING_ENABLE: enable a specific virtqueueFor memory mapping, IOTLB API is mandated for vhost-vDPA which meansuserspace drivers are required to useVHOST_IOTLB_UPDATE/VHOST_IOTLB_INVALIDATE to add or remove mapping fora specific userspace memory region.The vhost-vDPA API is designed to be type agnostic, but it allows netdevice only in current stage. Due to the lacking of control virtqueuesupport, some features were filter out by vhost-vdpa.We will enable more features and devices in the near future.Signed-off-by: Tiwei Bie Signed-off-by: Eugenio Pérez Signed-off-by: Jason Wang Link: https://lore.kernel.org/r/20200326140125.19794-8-jasowang@redhat.comSigned-off-by: Michael S. Tsirkin List of files: /linux-6.15/drivers/vhost/Kconfig

9ad9c49c - vringh: IOTLB support

Thu, 26 Mar 2020 14:01:20 +0000

vringh: IOTLB supportThis patch implements the third memory accessor for vringh besidescurrent kernel and userspace accessors. This idea is to allow vringhto do the address translation through an IOTLB which is implementedvia vhost_map interval tree. Users should setup and IOVA to PA mappingin this IOTLB.This allows us to:- Use vringh to access virtqueues with vIOMMU- Use vringh to implement software virtqueues for vDPA devicesSigned-off-by: Jason Wang Link: https://lore.kernel.org/r/20200326140125.19794-5-jasowang@redhat.comSigned-off-by: Michael S. Tsirkin List of files: /linux-6.15/drivers/vhost/Kconfig

0bbe3066 - vhost: factor out IOTLB

Thu, 26 Mar 2020 14:01:19 +0000

vhost: factor out IOTLBThis patch factors out IOTLB into a dedicated module in order to bereused by other modules like vringh. User may choose to enable theautomatic retiring by specifying VHOST_IOTLB_FLAG_RETIRE flag to fitfor the case of vhost device IOTLB implementation.Signed-off-by: Jason Wang Link: https://lore.kernel.org/r/20200326140125.19794-4-jasowang@redhat.comSigned-off-by: Michael S. Tsirkin List of files: /linux-6.15/drivers/vhost/Kconfig

20c384f1 - vhost: refine vhost and vringh kconfig

Thu, 26 Mar 2020 14:01:17 +0000

vhost: refine vhost and vringh kconfigCurrently, CONFIG_VHOST depends on CONFIG_VIRTUALIZATION. But vhost isnot necessarily for VM since it's a generic userspace and kernelcommunication protocol. Such dependency may prevent archs withoutvirtualization support from using vhost.To solve this, a dedicated vhost menu is created under drivers soCONIFG_VHOST can be decoupled out of CONFIG_VIRTUALIZATION.While at it, also squash Kconfig.vringh into vhost Kconfig file. Thisavoids the trick of conditional inclusion from VOP or CAIF. Then itwill be easier to introduce new vringh users and common dependency forboth vringh and vhost.Signed-off-by: Jason Wang Link: https://lore.kernel.org/r/20200326140125.19794-2-jasowang@redhat.comSigned-off-by: Michael S. Tsirkin List of files: /linux-6.15/drivers/vhost/Kconfig

ec8f24b7 - treewide: Add SPDX license identifier - Makefile/Kconfig

Sun, 19 May 2019 12:07:45 +0000

treewide: Add SPDX license identifier - Makefile/KconfigAdd SPDX license identifiers to all Make/Kconfig files which: - Have no license information of any formThese files fall under the project license, GPL v2 only. The resulting SPDXlicense identifier is: GPL-2.0-onlySigned-off-by: Thomas Gleixner Signed-off-by: Greg Kroah-Hartman List of files: /linux-6.15/drivers/vhost/Kconfig

84f3396b - vhost: allow vhost-scsi driver to be built-in

Fri, 10 Aug 2018 16:45:49 +0000

vhost: allow vhost-scsi driver to be built-inIt's useful to allow vhost-scsi to be built-in when testing vhost in L1+ L2 VMs and booting L1 VM with QEMU '-kernel' option.Signed-off-by: Greg Edwards Signed-off-by: Michael S. Tsirkin List of files: /linux-6.15/drivers/vhost/Kconfig

9a393b5d - tap: tap as an independent module

Sat, 11 Feb 2017 00:03:51 +0000

tap: tap as an independent moduleThis patch makes tap a separate module for other types of virtual interfaces, for example,ipvlan to use.Signed-off-by: Sainath Grandhi Signed-off-by: David S. Miller List of files: /linux-6.15/drivers/vhost/Kconfig

4d938245 - vhost: split out vringh Kconfig

Tue, 02 Aug 2016 00:03:35 +0000

vhost: split out vringh Kconfigvringh is pulled in by caif and mic, but the othervhost config does not need to be there.In particular, it makes no sense to have vhost net/scsi/sockunder caif/mic.Create a separate Kconfig file and put vringh bits there.Signed-off-by: Michael S. Tsirkin List of files: /linux-6.15/drivers/vhost/Kconfig

304ba62f - VSOCK: Add Makefile and Kconfig

Thu, 28 Jul 2016 14:36:35 +0000

VSOCK: Add Makefile and KconfigEnable virtio-vsock and vhost-vsock.Signed-off-by: Asias He Signed-off-by: Stefan Hajnoczi Signed-off-by: Michael S. Tsirkin List of files: /linux-6.15/drivers/vhost/Kconfig

6190efb0 - vhost: drop vringh dependency

Mon, 01 Aug 2016 22:53:13 +0000

vhost: drop vringh dependencyvringh isn't used by vhost net or scsi - it's usedby CAIF only at the moment. Drop the dependency.Signed-off-by: Michael S. Tsirkin List of files: /linux-6.15/drivers/vhost/Kconfig

8ac2837c - Revert "Merge branch 'vsock-virtio'"

Wed, 09 Dec 2015 02:51:12 +0000

Revert "Merge branch 'vsock-virtio'"This reverts commit 0d76d6e8b2507983a2cae4c09880798079007421 and mergecommit c402293bd76fbc93e52ef8c0947ab81eea3ae019, reversing changes madeto c89359a42e2a49656451569c382eed63e781153c.The virtio-vsock device specification is not finalized yet. MichaelTsirkin voiced concerned about merging this code when the hardwareinterface (and possibly the userspace interface) could still change.Signed-off-by: Stefan Hajnoczi Signed-off-by: David S. Miller List of files: /linux-6.15/drivers/vhost/Kconfig

8a2a2029 - VSOCK: Add Makefile and Kconfig

Wed, 02 Dec 2015 06:44:03 +0000

VSOCK: Add Makefile and KconfigEnable virtio-vsock and vhost-vsock.Signed-off-by: Asias He Signed-off-by: Stefan Hajnoczi Signed-off-by: David S. Miller List of files: /linux-6.15/drivers/vhost/Kconfig