From: Mike Christie <michael.christie@oracle.com> To: target-devel@vger.kernel.org, linux-scsi@vger.kernel.org, stefanha@redhat.com, pbonzini@redhat.com, jasowang@redhat.com, mst@redhat.com, sgarzare@redhat.com, virtualization@lists.linux-foundation.org Subject: vhost: multiple worker support Date: Tue, 25 May 2021 13:05:51 -0500 [thread overview] Message-ID: <20210525180600.6349-1-michael.christie@oracle.com> (raw) The following patches apply over linus's tree or mst's vhost branch and my cleanup patchset: https://lists.linuxfoundation.org/pipermail/virtualization/2021-May/054354.html These patches allow us to support multiple vhost workers per device. I ended up just doing Stefan's original idea where userspace has the kernel create a worker and we pass back the pid. This has the benefit over the workqueue and userspace thread approach where we only have one'ish code path in the kernel during setup to detect old tools. The main IO paths and device/vq setup/teardown paths all use common code. The kernel patches here allow us to then do N workers device and also share workers across devices. I've also included a patch for qemu so you can get an idea of how it works. If we are ok with the kernel code then I'll break that up into a patchset and send to qemu-devel. Results: -------- When running with the null_blk driver and vhost-scsi I can get 1.2 million IOPs by just running a simple fio --filename=/dev/sda --direct=1 --rw=randrw --bs=4k --ioengine=libaio --iodepth=128 --numjobs=8 --time_based --group_reporting --name=iops --runtime=60 --eta-newline=1 The VM has 8 vCPUs and sda has 8 virtqueues and we can do a total of 1024 cmds per devices. To get 1.2 million IOPs I did have to tune and ran the virsh emulatorpin command so the vhost threads were running on different CPUs than the VM. If the vhost threads share CPUs then I get around 800K. For a more real device that are also CPU hogs like iscsi, I can still get 1 million IOPs using 1 dm-multipath device over 8 iscsi paths (natively it gets 1.1 million IOPs). Results/TODO Note: - I ported the vdpa sim code to support multiple workers and as-is now it made perf much worse. If I increase vdpa_sim_blk's num queues to 4-8 I get 700K IOPs with the fio command above. However with the multiple worker support it drops to 400K. The problem is the vdpa_sim lock and the iommu_lock. If I hack (like comment out locks or not worry about data corruption or crashes) then I can get around 1.2M - 1.6M IOPs with 8 queues and fio command above. So these patches could help other drivers, but it will just take more work to remove those types of locks. I was hoping the 2 items could be done indepentently since it helps vhost-scsi immediately. TODO: - Stefano has 2 questions about security issues passing the pid back to userspace and if we should do a feature bit. We are waiting to hear back from the list. v2: - change loop that we take a refcount to the worker in - replaced pid == -1 with define. - fixed tabbing/spacing coding style issue - use hash instead of list to lookup workers. - I dropped the patch that added an ioctl cmd to get a vq's worker's pid. I saw we might do a generic netlink interface instead.
WARNING: multiple messages have this Message-ID (diff)
From: Mike Christie <michael.christie@oracle.com> To: target-devel@vger.kernel.org, linux-scsi@vger.kernel.org, stefanha@redhat.com, pbonzini@redhat.com, jasowang@redhat.com, mst@redhat.com, sgarzare@redhat.com, virtualization@lists.linux-foundation.org Subject: vhost: multiple worker support Date: Tue, 25 May 2021 13:05:51 -0500 [thread overview] Message-ID: <20210525180600.6349-1-michael.christie@oracle.com> (raw) The following patches apply over linus's tree or mst's vhost branch and my cleanup patchset: https://lists.linuxfoundation.org/pipermail/virtualization/2021-May/054354.html These patches allow us to support multiple vhost workers per device. I ended up just doing Stefan's original idea where userspace has the kernel create a worker and we pass back the pid. This has the benefit over the workqueue and userspace thread approach where we only have one'ish code path in the kernel during setup to detect old tools. The main IO paths and device/vq setup/teardown paths all use common code. The kernel patches here allow us to then do N workers device and also share workers across devices. I've also included a patch for qemu so you can get an idea of how it works. If we are ok with the kernel code then I'll break that up into a patchset and send to qemu-devel. Results: -------- When running with the null_blk driver and vhost-scsi I can get 1.2 million IOPs by just running a simple fio --filename=/dev/sda --direct=1 --rw=randrw --bs=4k --ioengine=libaio --iodepth=128 --numjobs=8 --time_based --group_reporting --name=iops --runtime=60 --eta-newline=1 The VM has 8 vCPUs and sda has 8 virtqueues and we can do a total of 1024 cmds per devices. To get 1.2 million IOPs I did have to tune and ran the virsh emulatorpin command so the vhost threads were running on different CPUs than the VM. If the vhost threads share CPUs then I get around 800K. For a more real device that are also CPU hogs like iscsi, I can still get 1 million IOPs using 1 dm-multipath device over 8 iscsi paths (natively it gets 1.1 million IOPs). Results/TODO Note: - I ported the vdpa sim code to support multiple workers and as-is now it made perf much worse. If I increase vdpa_sim_blk's num queues to 4-8 I get 700K IOPs with the fio command above. However with the multiple worker support it drops to 400K. The problem is the vdpa_sim lock and the iommu_lock. If I hack (like comment out locks or not worry about data corruption or crashes) then I can get around 1.2M - 1.6M IOPs with 8 queues and fio command above. So these patches could help other drivers, but it will just take more work to remove those types of locks. I was hoping the 2 items could be done indepentently since it helps vhost-scsi immediately. TODO: - Stefano has 2 questions about security issues passing the pid back to userspace and if we should do a feature bit. We are waiting to hear back from the list. v2: - change loop that we take a refcount to the worker in - replaced pid == -1 with define. - fixed tabbing/spacing coding style issue - use hash instead of list to lookup workers. - I dropped the patch that added an ioctl cmd to get a vq's worker's pid. I saw we might do a generic netlink interface instead. _______________________________________________ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/virtualization
next reply other threads:[~2021-05-25 18:06 UTC|newest] Thread overview: 64+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-05-25 18:05 Mike Christie [this message] 2021-05-25 18:05 ` vhost: multiple worker support Mike Christie 2021-05-25 18:05 ` [PATCH 1/9] vhost: move worker thread fields to new struct Mike Christie 2021-05-25 18:05 ` Mike Christie 2021-06-03 10:16 ` Stefan Hajnoczi 2021-06-03 10:16 ` Stefan Hajnoczi 2021-05-25 18:05 ` [PATCH 2/9] vhost: move vhost worker creation to kick setup Mike Christie 2021-05-25 18:05 ` Mike Christie 2021-06-03 10:28 ` Stefan Hajnoczi 2021-06-03 10:28 ` Stefan Hajnoczi 2021-05-25 18:05 ` [PATCH 3/9] vhost: modify internal functions to take a vhost_worker Mike Christie 2021-05-25 18:05 ` Mike Christie 2021-06-03 10:45 ` Stefan Hajnoczi 2021-06-03 10:45 ` Stefan Hajnoczi 2021-05-25 18:05 ` [PATCH 4/9] vhost: allow vhost_polls to use different vhost_workers Mike Christie 2021-05-25 18:05 ` Mike Christie 2021-06-03 13:51 ` Stefan Hajnoczi 2021-06-03 13:51 ` Stefan Hajnoczi 2021-05-25 18:05 ` [PATCH 5/9] vhost-scsi: flush IO vqs then send TMF rsp Mike Christie 2021-05-25 18:05 ` Mike Christie 2021-06-03 13:54 ` Stefan Hajnoczi 2021-06-03 13:54 ` Stefan Hajnoczi 2021-05-25 18:05 ` [PATCH 6/9] vhost-scsi: make SCSI cmd completion per vq Mike Christie 2021-05-25 18:05 ` Mike Christie 2021-06-03 13:57 ` Stefan Hajnoczi 2021-06-03 13:57 ` Stefan Hajnoczi 2021-05-25 18:05 ` [PATCH 7/9] vhost: allow userspace to create workers Mike Christie 2021-05-25 18:05 ` Mike Christie 2021-06-03 14:30 ` Stefan Hajnoczi 2021-06-03 14:30 ` Stefan Hajnoczi 2021-06-05 23:53 ` michael.christie 2021-06-05 23:53 ` michael.christie 2021-06-07 15:19 ` Stefan Hajnoczi 2021-06-07 15:19 ` Stefan Hajnoczi 2021-06-09 21:03 ` Mike Christie 2021-06-09 21:03 ` Mike Christie 2021-06-10 8:06 ` Stefan Hajnoczi 2021-06-10 8:06 ` Stefan Hajnoczi 2021-06-18 2:49 ` Mike Christie 2021-06-18 2:49 ` Mike Christie 2021-06-21 13:41 ` Stefan Hajnoczi 2021-06-21 13:41 ` Stefan Hajnoczi 2021-05-25 18:05 ` [PATCH 8/9] vhost: add vhost_dev pointer to vhost_work Mike Christie 2021-05-25 18:05 ` Mike Christie 2021-06-03 14:31 ` Stefan Hajnoczi 2021-06-03 14:31 ` Stefan Hajnoczi 2021-05-25 18:06 ` [PATCH 9/9] vhost: support sharing workers across devs Mike Christie 2021-05-25 18:06 ` Mike Christie 2021-06-03 14:32 ` Stefan Hajnoczi 2021-06-03 14:32 ` Stefan Hajnoczi 2021-06-07 2:18 ` Jason Wang 2021-06-07 2:18 ` Jason Wang 2021-06-03 10:13 ` vhost: multiple worker support Stefan Hajnoczi 2021-06-03 10:13 ` Stefan Hajnoczi 2021-06-03 18:45 ` Mike Christie 2021-06-03 18:45 ` Mike Christie 2021-06-03 14:37 ` Stefan Hajnoczi 2021-06-03 14:37 ` Stefan Hajnoczi 2021-06-03 22:16 ` Mike Christie 2021-06-03 22:16 ` Mike Christie 2021-06-05 22:40 ` michael.christie 2021-06-05 22:40 ` michael.christie 2021-06-07 15:23 ` Stefan Hajnoczi 2021-06-07 15:23 ` Stefan Hajnoczi
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=20210525180600.6349-1-michael.christie@oracle.com \ --to=michael.christie@oracle.com \ --cc=jasowang@redhat.com \ --cc=linux-scsi@vger.kernel.org \ --cc=mst@redhat.com \ --cc=pbonzini@redhat.com \ --cc=sgarzare@redhat.com \ --cc=stefanha@redhat.com \ --cc=target-devel@vger.kernel.org \ --cc=virtualization@lists.linux-foundation.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.