| Message ID | 1502359259-24966-1-git-send-email-tianyu.lan@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Hi Konrad: Thanks for your review. On 2017年08月11日 01:50, Konrad Rzeszutek Wilk wrote: > On Thu, Aug 10, 2017 at 06:00:59PM +0800, Lan Tianyu wrote: >> Intel Xeon phi chip will support 352 logical threads. For HPC usage >> case, it will create a huge VM with vcpu number as same as host cpus. This >> patch is to increase max vcpu number to 352. > > Why not 1024 or 4096? This is on demand. We can set a higher number since KVM already has x2apic and vIOMMU interrupt remapping support. > > Are there any issues with increasing the value from 288 to 352 right now? No found. > > Also perhaps this should be made in an Kconfig entry? That will be anther option but I find different platforms will define different MAX_VCPU. If we introduce a generic Kconfig entry, different platforms should have different range. Radim & Paolo, Could you give some input? In qemu thread, we will set max vcpu to 8192 for x86 VM. In KVM, The length of vcpu pointer array in struct kvm and dest_vcpu_bitmap in kvm_irq_delivery_to_apic() are specified by KVM_MAX_VCPUS. Should we keep align with Qemu?
On 11.08.2017 09:49, Lan Tianyu wrote: > Hi Konrad: > Thanks for your review. > > On 2017年08月11日 01:50, Konrad Rzeszutek Wilk wrote: >> On Thu, Aug 10, 2017 at 06:00:59PM +0800, Lan Tianyu wrote: >>> Intel Xeon phi chip will support 352 logical threads. For HPC usage >>> case, it will create a huge VM with vcpu number as same as host cpus. This >>> patch is to increase max vcpu number to 352. >> >> Why not 1024 or 4096? > > This is on demand. We can set a higher number since KVM already has > x2apic and vIOMMU interrupt remapping support. > >> >> Are there any issues with increasing the value from 288 to 352 right now? > > No found. > >> >> Also perhaps this should be made in an Kconfig entry? > > That will be anther option but I find different platforms will define > different MAX_VCPU. If we introduce a generic Kconfig entry, different > platforms should have different range. > > Radim & Paolo, Could you give some input? In qemu thread, we will set > max vcpu to 8192 for x86 VM. In KVM, The length of vcpu pointer array in > struct kvm and dest_vcpu_bitmap in kvm_irq_delivery_to_apic() are > specified by KVM_MAX_VCPUS. Should we keep align with Qemu? > commit 682f732ecf7396e9d6fe24d44738966699fae6c0 Author: Radim Krčmář <rkrcmar@redhat.com> Date: Tue Jul 12 22:09:29 2016 +0200 KVM: x86: bump MAX_VCPUS to 288 288 is in high demand because of Knights Landing CPU. We cannot set the limit to 640k, because that would be wasting space. I think we want to keep it small as long as possible. I remember a patch series from Radim which would dynamically allocate memory for these arrays (using a new VM creation ioctl, specifying the max # of vcpus). Wonder what happened to that (I remember requesting a simply remalloc instead of a new VM creation ioctl :] ).
2017-08-11 10:11+0200, David Hildenbrand: > On 11.08.2017 09:49, Lan Tianyu wrote: >> Hi Konrad: >> Thanks for your review. >> >> On 2017年08月11日 01:50, Konrad Rzeszutek Wilk wrote: >>> On Thu, Aug 10, 2017 at 06:00:59PM +0800, Lan Tianyu wrote: >>>> Intel Xeon phi chip will support 352 logical threads. For HPC usage >>>> case, it will create a huge VM with vcpu number as same as host cpus. This >>>> patch is to increase max vcpu number to 352. >>> >>> Why not 1024 or 4096? >> >> This is on demand. We can set a higher number since KVM already has >> x2apic and vIOMMU interrupt remapping support. >> >>> >>> Are there any issues with increasing the value from 288 to 352 right now? >> >> No found. Yeah, the only issue until around 2^20 (when we reach the maximum of logical x2APIC addressing) should be the size of per-VM arrays when only few VCPUs are going to be used. >>> Also perhaps this should be made in an Kconfig entry? >> >> That will be anther option but I find different platforms will define >> different MAX_VCPU. If we introduce a generic Kconfig entry, different >> platforms should have different range. >> >> Radim & Paolo, Could you give some input? In qemu thread, we will set >> max vcpu to 8192 for x86 VM. In KVM, The length of vcpu pointer array in >> struct kvm and dest_vcpu_bitmap in kvm_irq_delivery_to_apic() are >> specified by KVM_MAX_VCPUS. Should we keep align with Qemu? That would be great. > commit 682f732ecf7396e9d6fe24d44738966699fae6c0 > Author: Radim Krčmář <rkrcmar@redhat.com> > Date: Tue Jul 12 22:09:29 2016 +0200 > > KVM: x86: bump MAX_VCPUS to 288 > > 288 is in high demand because of Knights Landing CPU. > We cannot set the limit to 640k, because that would be wasting space. > > I think we want to keep it small as long as possible. I remember a patch > series from Radim which would dynamically allocate memory for these > arrays (using a new VM creation ioctl, specifying the max # of vcpus). > Wonder what happened to that (I remember requesting a simply remalloc > instead of a new VM creation ioctl :] ). Eh, I forgot about them ... I didn't like the dynamic allocation as we would need to protect the memory, which would result in a much bigger changeset, or fragile macros. I can't recall the disgust now, so I'll send a RFC with the dynamic version to see how it turned out. Thanks.
On Fri, Aug 11, 2017 at 03:00:20PM +0200, Radim Krčmář wrote: > 2017-08-11 10:11+0200, David Hildenbrand: > > On 11.08.2017 09:49, Lan Tianyu wrote: > >> Hi Konrad: > >> Thanks for your review. > >> > >> On 2017年08月11日 01:50, Konrad Rzeszutek Wilk wrote: > >>> On Thu, Aug 10, 2017 at 06:00:59PM +0800, Lan Tianyu wrote: > >>>> Intel Xeon phi chip will support 352 logical threads. For HPC usage > >>>> case, it will create a huge VM with vcpu number as same as host cpus. This > >>>> patch is to increase max vcpu number to 352. > >>> > >>> Why not 1024 or 4096? > >> > >> This is on demand. We can set a higher number since KVM already has > >> x2apic and vIOMMU interrupt remapping support. > >> > >>> > >>> Are there any issues with increasing the value from 288 to 352 right now? > >> > >> No found. > > Yeah, the only issue until around 2^20 (when we reach the maximum of > logical x2APIC addressing) should be the size of per-VM arrays when only > few VCPUs are going to be used. Migration with 352 CPUs all being busy dirtying memory and also poking at various I/O ports (say all of them dirtying the VGA) is no problem? > > >>> Also perhaps this should be made in an Kconfig entry? > >> > >> That will be anther option but I find different platforms will define > >> different MAX_VCPU. If we introduce a generic Kconfig entry, different > >> platforms should have different range. By different platforms you mean q35 vs the older one, and such? Not whether the underlaying accelerator is tcg, Xen, KVM, or bHyve? What I was trying to understand whether it makes even sense for the platforms to have such limits in the first place - and instead the accelerators should be the ones setting it? > >> > >> Radim & Paolo, Could you give some input? In qemu thread, we will set > >> max vcpu to 8192 for x86 VM. In KVM, The length of vcpu pointer array in > >> struct kvm and dest_vcpu_bitmap in kvm_irq_delivery_to_apic() are > >> specified by KVM_MAX_VCPUS. Should we keep align with Qemu? > > That would be great. > > > commit 682f732ecf7396e9d6fe24d44738966699fae6c0 > > Author: Radim Krčmář <rkrcmar@redhat.com> > > Date: Tue Jul 12 22:09:29 2016 +0200 > > > > KVM: x86: bump MAX_VCPUS to 288 > > > > 288 is in high demand because of Knights Landing CPU. > > We cannot set the limit to 640k, because that would be wasting space. > > > > I think we want to keep it small as long as possible. I remember a patch > > series from Radim which would dynamically allocate memory for these > > arrays (using a new VM creation ioctl, specifying the max # of vcpus). > > Wonder what happened to that (I remember requesting a simply remalloc > > instead of a new VM creation ioctl :] ). > > Eh, I forgot about them ... I didn't like the dynamic allocation as we > would need to protect the memory, which would result in a much bigger > changeset, or fragile macros. > > I can't recall the disgust now, so I'll send a RFC with the dynamic > version to see how it turned out. > > Thanks.
On Thu, Aug 10, 2017 at 12:00 PM, Lan Tianyu <tianyu.lan@intel.com> wrote: > Intel Xeon phi chip will support 352 logical threads. For HPC usage > case, it will create a huge VM with vcpu number as same as host cpus. This > patch is to increase max vcpu number to 352. This number was bumped in the past to 288 to accommodate Knights Landing, but KNL's max designed thread number is actually 304: the on-die interconnect mesh is 6*7, with four cells taken for interconnect and memory controllers, there are 38 CPU cells. Each CPU cell has two cores with shared L2. Each core is SMT4. 38*8 = 304. Intel fuses two cells (or more), so 288 is the largest number of threads on a KNL you can buy, but 304 thread KNLs most probably also exist (however they may be rather rare since they require completely defect-free die). I think it's better if Linux would support those too. What is the design maximum for these new "nominally 352 thread" Xeon Phis which are "nominally 352 thread"? 360? (If the mesh is 7*7 and the same 4 cells are taked for non-CPU needs)
On 2017年08月12日 03:35, Konrad Rzeszutek Wilk wrote: > On Fri, Aug 11, 2017 at 03:00:20PM +0200, Radim Krčmář wrote: >> 2017-08-11 10:11+0200, David Hildenbrand: >>> On 11.08.2017 09:49, Lan Tianyu wrote: >>>> Hi Konrad: >>>> Thanks for your review. >>>> >>>> On 2017年08月11日 01:50, Konrad Rzeszutek Wilk wrote: >>>>> On Thu, Aug 10, 2017 at 06:00:59PM +0800, Lan Tianyu wrote: >>>>>> Intel Xeon phi chip will support 352 logical threads. For HPC usage >>>>>> case, it will create a huge VM with vcpu number as same as host cpus. This >>>>>> patch is to increase max vcpu number to 352. >>>>> >>>>> Why not 1024 or 4096? >>>> >>>> This is on demand. We can set a higher number since KVM already has >>>> x2apic and vIOMMU interrupt remapping support. >>>> >>>>> >>>>> Are there any issues with increasing the value from 288 to 352 right now? >>>> >>>> No found. >> >> Yeah, the only issue until around 2^20 (when we reach the maximum of >> logical x2APIC addressing) should be the size of per-VM arrays when only >> few VCPUs are going to be used. > > Migration with 352 CPUs all being busy dirtying memory and also poking > at various I/O ports (say all of them dirtying the VGA) is no problem? This depends on what kind of workload is running during migration. I think this may affect service down time since there maybe a lot of dirty memory data to transfer after stopping vcpus. This also depends on how user sets "migrate_set_downtime" for qemu. But I think increasing vcpus will break migration function. > > >> >>>>> Also perhaps this should be made in an Kconfig entry? >>>> >>>> That will be anther option but I find different platforms will define >>>> different MAX_VCPU. If we introduce a generic Kconfig entry, different >>>> platforms should have different range. > > > By different platforms you mean q35 vs the older one, and such? I meant x86, arm, sparc and other vendors' code define different max vcpu number. > Not whether the underlaying accelerator is tcg, Xen, KVM, or bHyve? > > What I was trying to understand whether it makes even sense for > the platforms to have such limits in the first place - and instead the > accelerators should be the ones setting it? > > >>>> >>>> Radim & Paolo, Could you give some input? In qemu thread, we will set >>>> max vcpu to 8192 for x86 VM. In KVM, The length of vcpu pointer array in >>>> struct kvm and dest_vcpu_bitmap in kvm_irq_delivery_to_apic() are >>>> specified by KVM_MAX_VCPUS. Should we keep align with Qemu? >> >> That would be great. >> >>> commit 682f732ecf7396e9d6fe24d44738966699fae6c0 >>> Author: Radim Krčmář <rkrcmar@redhat.com> >>> Date: Tue Jul 12 22:09:29 2016 +0200 >>> >>> KVM: x86: bump MAX_VCPUS to 288 >>> >>> 288 is in high demand because of Knights Landing CPU. >>> We cannot set the limit to 640k, because that would be wasting space. >>> >>> I think we want to keep it small as long as possible. I remember a patch >>> series from Radim which would dynamically allocate memory for these >>> arrays (using a new VM creation ioctl, specifying the max # of vcpus). >>> Wonder what happened to that (I remember requesting a simply remalloc >>> instead of a new VM creation ioctl :] ). >> >> Eh, I forgot about them ... I didn't like the dynamic allocation as we >> would need to protect the memory, which would result in a much bigger >> changeset, or fragile macros. >> >> I can't recall the disgust now, so I'll send a RFC with the dynamic >> version to see how it turned out. >> >> Thanks.
On Tue, Aug 15, 2017 at 11:00:04AM +0800, Lan Tianyu wrote: > On 2017年08月12日 03:35, Konrad Rzeszutek Wilk wrote: > > On Fri, Aug 11, 2017 at 03:00:20PM +0200, Radim Krčmář wrote: > >> 2017-08-11 10:11+0200, David Hildenbrand: > >>> On 11.08.2017 09:49, Lan Tianyu wrote: > >>>> Hi Konrad: > >>>> Thanks for your review. > >>>> > >>>> On 2017年08月11日 01:50, Konrad Rzeszutek Wilk wrote: > >>>>> On Thu, Aug 10, 2017 at 06:00:59PM +0800, Lan Tianyu wrote: > >>>>>> Intel Xeon phi chip will support 352 logical threads. For HPC usage > >>>>>> case, it will create a huge VM with vcpu number as same as host cpus. This > >>>>>> patch is to increase max vcpu number to 352. > >>>>> > >>>>> Why not 1024 or 4096? > >>>> > >>>> This is on demand. We can set a higher number since KVM already has > >>>> x2apic and vIOMMU interrupt remapping support. > >>>> > >>>>> > >>>>> Are there any issues with increasing the value from 288 to 352 right now? > >>>> > >>>> No found. > >> > >> Yeah, the only issue until around 2^20 (when we reach the maximum of > >> logical x2APIC addressing) should be the size of per-VM arrays when only > >> few VCPUs are going to be used. > > > > Migration with 352 CPUs all being busy dirtying memory and also poking > > at various I/O ports (say all of them dirtying the VGA) is no problem? > > This depends on what kind of workload is running during migration. I > think this may affect service down time since there maybe a lot of dirty > memory data to transfer after stopping vcpus. This also depends on how > user sets "migrate_set_downtime" for qemu. But I think increasing vcpus > will break migration function. OK, so let me take a step back. I see this nice 'supported' CPU count that is exposed in kvm module. Then there is QEMU throwing out a warning if you crank up the CPU count above that number. Red Hat's web-pages talk about CPU count as well. And I am assuming all of those are around what has been tested and what has shown to work. And one of those test-cases surely must be migration. Ergo, if the vCPU count increase will break migration, then it is a regression. Or a fix/work needs to be done to support a higher CPU count for migrating? Is my understanding incorrect? > > > > > > >> > >>>>> Also perhaps this should be made in an Kconfig entry? > >>>> > >>>> That will be anther option but I find different platforms will define > >>>> different MAX_VCPU. If we introduce a generic Kconfig entry, different > >>>> platforms should have different range. > > > > > > By different platforms you mean q35 vs the older one, and such? > > I meant x86, arm, sparc and other vendors' code define different max > vcpu number. Right, and? > > > Not whether the underlaying accelerator is tcg, Xen, KVM, or bHyve? > > > > What I was trying to understand whether it makes even sense for > > the platforms to have such limits in the first place - and instead the > > accelerators should be the ones setting it? > > > > > >>>> > >>>> Radim & Paolo, Could you give some input? In qemu thread, we will set > >>>> max vcpu to 8192 for x86 VM. In KVM, The length of vcpu pointer array in > >>>> struct kvm and dest_vcpu_bitmap in kvm_irq_delivery_to_apic() are > >>>> specified by KVM_MAX_VCPUS. Should we keep align with Qemu? > >> > >> That would be great. > >> > >>> commit 682f732ecf7396e9d6fe24d44738966699fae6c0 > >>> Author: Radim Krčmář <rkrcmar@redhat.com> > >>> Date: Tue Jul 12 22:09:29 2016 +0200 > >>> > >>> KVM: x86: bump MAX_VCPUS to 288 > >>> > >>> 288 is in high demand because of Knights Landing CPU. > >>> We cannot set the limit to 640k, because that would be wasting space. > >>> > >>> I think we want to keep it small as long as possible. I remember a patch > >>> series from Radim which would dynamically allocate memory for these > >>> arrays (using a new VM creation ioctl, specifying the max # of vcpus). > >>> Wonder what happened to that (I remember requesting a simply remalloc > >>> instead of a new VM creation ioctl :] ). > >> > >> Eh, I forgot about them ... I didn't like the dynamic allocation as we > >> would need to protect the memory, which would result in a much bigger > >> changeset, or fragile macros. > >> > >> I can't recall the disgust now, so I'll send a RFC with the dynamic > >> version to see how it turned out. > >> > >> Thanks. > > > -- > Best regards > Tianyu Lan
2017-08-15 11:00+0800, Lan Tianyu: > On 2017年08月12日 03:35, Konrad Rzeszutek Wilk wrote: >> On Fri, Aug 11, 2017 at 03:00:20PM +0200, Radim Krčmář wrote: >>> 2017-08-11 10:11+0200, David Hildenbrand: >>>> On 11.08.2017 09:49, Lan Tianyu wrote: >>>>> On 2017年08月11日 01:50, Konrad Rzeszutek Wilk wrote: >>>>>> Are there any issues with increasing the value from 288 to 352 right now? >>>>> >>>>> No found. >>> >>> Yeah, the only issue until around 2^20 (when we reach the maximum of >>> logical x2APIC addressing) should be the size of per-VM arrays when only >>> few VCPUs are going to be used. (I was talking only about the KVM side.) >> Migration with 352 CPUs all being busy dirtying memory and also poking >> at various I/O ports (say all of them dirtying the VGA) is no problem? > > This depends on what kind of workload is running during migration. I > think this may affect service down time since there maybe a lot of dirty > memory data to transfer after stopping vcpus. This also depends on how > user sets "migrate_set_downtime" for qemu. But I think increasing vcpus > will break migration function. Utilizing post-copy in the last migration phase should make migration of busy big guests possible. (I agree that pre-copy in not going to be feasible.)
(Missed this mail before my last reply.) 2017-08-15 10:10-0400, Konrad Rzeszutek Wilk: > On Tue, Aug 15, 2017 at 11:00:04AM +0800, Lan Tianyu wrote: > > On 2017年08月12日 03:35, Konrad Rzeszutek Wilk wrote: > > > Migration with 352 CPUs all being busy dirtying memory and also poking > > > at various I/O ports (say all of them dirtying the VGA) is no problem? > > > > This depends on what kind of workload is running during migration. I > > think this may affect service down time since there maybe a lot of dirty > > memory data to transfer after stopping vcpus. This also depends on how > > user sets "migrate_set_downtime" for qemu. But I think increasing vcpus > > will break migration function. > > OK, so let me take a step back. > > I see this nice 'supported' CPU count that is exposed in kvm module. > > Then there is QEMU throwing out a warning if you crank up the CPU count > above that number. I find the range between "recommended max" and "hard max" VCPU count confusing at best ... IIUC, it was there because KVM internals had problems with scaling and we will hit more in the future because some loops still are linear on VCPU count. The exposed value doesn't say whether migration will work, because that is a userspace thing and we're not aware of bottlenecks on the KVM side. > Red Hat's web-pages talk about CPU count as well. > > And I am assuming all of those are around what has been tested and > what has shown to work. And one of those test-cases surely must > be migration. Right, Red Hat will only allow/support what it has tested, even if upstream has a practically unlimited count. I think the upstream number used to be raised by Red Hat, which is why upstream isn't at the hard implementation limit ... > Ergo, if the vCPU count increase will break migration, then it is > a regression. Raising the limit would not break existing guests, but I would rather avoid adding higher VCPU count as a feature that disables migration. > Or a fix/work needs to be done to support a higher CPU count for > migrating? Post-copy migration should handle higher CPU count and it is the default fallback on QEMU. Asking the question on a userspace list would yield better answers, though. Thanks.
On 2017年08月15日 22:10, Konrad Rzeszutek Wilk wrote: > On Tue, Aug 15, 2017 at 11:00:04AM +0800, Lan Tianyu wrote: >> On 2017年08月12日 03:35, Konrad Rzeszutek Wilk wrote: >>> On Fri, Aug 11, 2017 at 03:00:20PM +0200, Radim Krčmář wrote: >>>> 2017-08-11 10:11+0200, David Hildenbrand: >>>>> On 11.08.2017 09:49, Lan Tianyu wrote: >>>>>> Hi Konrad: >>>>>> Thanks for your review. >>>>>> >>>>>> On 2017年08月11日 01:50, Konrad Rzeszutek Wilk wrote: >>>>>>> On Thu, Aug 10, 2017 at 06:00:59PM +0800, Lan Tianyu wrote: >>>>>>>> Intel Xeon phi chip will support 352 logical threads. For HPC usage >>>>>>>> case, it will create a huge VM with vcpu number as same as host cpus. This >>>>>>>> patch is to increase max vcpu number to 352. >>>>>>> >>>>>>> Why not 1024 or 4096? >>>>>> >>>>>> This is on demand. We can set a higher number since KVM already has >>>>>> x2apic and vIOMMU interrupt remapping support. >>>>>> >>>>>>> >>>>>>> Are there any issues with increasing the value from 288 to 352 right now? >>>>>> >>>>>> No found. >>>> >>>> Yeah, the only issue until around 2^20 (when we reach the maximum of >>>> logical x2APIC addressing) should be the size of per-VM arrays when only >>>> few VCPUs are going to be used. >>> >>> Migration with 352 CPUs all being busy dirtying memory and also poking >>> at various I/O ports (say all of them dirtying the VGA) is no problem? >> >> This depends on what kind of workload is running during migration. I >> think this may affect service down time since there maybe a lot of dirty >> memory data to transfer after stopping vcpus. This also depends on how >> user sets "migrate_set_downtime" for qemu. But I think increasing vcpus >> will break migration function. > > OK, so let me take a step back. > > I see this nice 'supported' CPU count that is exposed in kvm module. > > Then there is QEMU throwing out a warning if you crank up the CPU count > above that number. > > Red Hat's web-pages talk about CPU count as well. > > And I am assuming all of those are around what has been tested and > what has shown to work. And one of those test-cases surely must > be migration. > Sorry. This is a typo. I originally meant increasing vcpu shouldn't break migration function and just affect service downtime. If there was such issue, we should fix it. > Ergo, if the vCPU count increase will break migration, then it is > a regression. > > Or a fix/work needs to be done to support a higher CPU count for > migrating? > > > Is my understanding incorrect? You are right. > >> >>> >>> >>>> >>>>>>> Also perhaps this should be made in an Kconfig entry? >>>>>> >>>>>> That will be anther option but I find different platforms will define >>>>>> different MAX_VCPU. If we introduce a generic Kconfig entry, different >>>>>> platforms should have different range. >>> >>> >>> By different platforms you mean q35 vs the older one, and such? >> >> I meant x86, arm, sparc and other vendors' code define different max >> vcpu number. > > Right, and? If we introduce a general kconfig of max vcpus for all vendors, it should have different max vcpu range for different vendor.
On Tue, Aug 15, 2017 at 06:13:29PM +0200, Radim Krčmář wrote: > (Missed this mail before my last reply.) > > 2017-08-15 10:10-0400, Konrad Rzeszutek Wilk: > > On Tue, Aug 15, 2017 at 11:00:04AM +0800, Lan Tianyu wrote: > > > On 2017年08月12日 03:35, Konrad Rzeszutek Wilk wrote: > > > > Migration with 352 CPUs all being busy dirtying memory and also poking > > > > at various I/O ports (say all of them dirtying the VGA) is no problem? > > > > > > This depends on what kind of workload is running during migration. I > > > think this may affect service down time since there maybe a lot of dirty > > > memory data to transfer after stopping vcpus. This also depends on how > > > user sets "migrate_set_downtime" for qemu. But I think increasing vcpus > > > will break migration function. > > > > OK, so let me take a step back. > > > > I see this nice 'supported' CPU count that is exposed in kvm module. > > > > Then there is QEMU throwing out a warning if you crank up the CPU count > > above that number. > > I find the range between "recommended max" and "hard max" VCPU count > confusing at best ... IIUC, it was there because KVM internals had > problems with scaling and we will hit more in the future because some > loops still are linear on VCPU count. Is that documented somewhere? There are some folks would be interested in looking at that if it was known what exactly to look for.. > > The exposed value doesn't say whether migration will work, because that > is a userspace thing and we're not aware of bottlenecks on the KVM side. > > > Red Hat's web-pages talk about CPU count as well. > > > > And I am assuming all of those are around what has been tested and > > what has shown to work. And one of those test-cases surely must > > be migration. > > Right, Red Hat will only allow/support what it has tested, even if > upstream has a practically unlimited count. I think the upstream number > used to be raised by Red Hat, which is why upstream isn't at the hard > implementation limit ... Aim for the sky! Perhaps then lets crank it up to 4096 upstream and let each vendor/distro/cloud decide the right number based on their testing. And also have more folks report issues as they try running say running these huge vCPU guests? > > > Ergo, if the vCPU count increase will break migration, then it is > > a regression. > > Raising the limit would not break existing guests, but I would rather > avoid adding higher VCPU count as a feature that disables migration. > > > Or a fix/work needs to be done to support a higher CPU count for > > migrating? > > Post-copy migration should handle higher CPU count and it is the default > fallback on QEMU. Asking the question on a userspace list would yield > better answers, though. > > Thanks.
On Wed, Aug 16, 2017 at 11:07:55AM +0800, Lan Tianyu wrote: > On 2017年08月15日 22:10, Konrad Rzeszutek Wilk wrote: > > On Tue, Aug 15, 2017 at 11:00:04AM +0800, Lan Tianyu wrote: > >> On 2017年08月12日 03:35, Konrad Rzeszutek Wilk wrote: > >>> On Fri, Aug 11, 2017 at 03:00:20PM +0200, Radim Krčmář wrote: > >>>> 2017-08-11 10:11+0200, David Hildenbrand: > >>>>> On 11.08.2017 09:49, Lan Tianyu wrote: > >>>>>> Hi Konrad: > >>>>>> Thanks for your review. > >>>>>> > >>>>>> On 2017年08月11日 01:50, Konrad Rzeszutek Wilk wrote: > >>>>>>> On Thu, Aug 10, 2017 at 06:00:59PM +0800, Lan Tianyu wrote: > >>>>>>>> Intel Xeon phi chip will support 352 logical threads. For HPC usage > >>>>>>>> case, it will create a huge VM with vcpu number as same as host cpus. This > >>>>>>>> patch is to increase max vcpu number to 352. > >>>>>>> > >>>>>>> Why not 1024 or 4096? > >>>>>> > >>>>>> This is on demand. We can set a higher number since KVM already has > >>>>>> x2apic and vIOMMU interrupt remapping support. > >>>>>> > >>>>>>> > >>>>>>> Are there any issues with increasing the value from 288 to 352 right now? > >>>>>> > >>>>>> No found. > >>>> > >>>> Yeah, the only issue until around 2^20 (when we reach the maximum of > >>>> logical x2APIC addressing) should be the size of per-VM arrays when only > >>>> few VCPUs are going to be used. > >>> > >>> Migration with 352 CPUs all being busy dirtying memory and also poking > >>> at various I/O ports (say all of them dirtying the VGA) is no problem? > >> > >> This depends on what kind of workload is running during migration. I > >> think this may affect service down time since there maybe a lot of dirty > >> memory data to transfer after stopping vcpus. This also depends on how > >> user sets "migrate_set_downtime" for qemu. But I think increasing vcpus > >> will break migration function. > > > > OK, so let me take a step back. > > > > I see this nice 'supported' CPU count that is exposed in kvm module. > > > > Then there is QEMU throwing out a warning if you crank up the CPU count > > above that number. > > > > Red Hat's web-pages talk about CPU count as well. > > > > And I am assuming all of those are around what has been tested and > > what has shown to work. And one of those test-cases surely must > > be migration. > > > > Sorry. This is a typo. I originally meant increasing vcpu shouldn't > break migration function and just affect service downtime. If there was > such issue, we should fix it. > > > > Ergo, if the vCPU count increase will break migration, then it is > > a regression. > > > > Or a fix/work needs to be done to support a higher CPU count for > > migrating? > > > > > > Is my understanding incorrect? > > You are right. > > > > >> > >>> > >>> > >>>> > >>>>>>> Also perhaps this should be made in an Kconfig entry? > >>>>>> > >>>>>> That will be anther option but I find different platforms will define > >>>>>> different MAX_VCPU. If we introduce a generic Kconfig entry, different > >>>>>> platforms should have different range. > >>> > >>> > >>> By different platforms you mean q35 vs the older one, and such? > >> > >> I meant x86, arm, sparc and other vendors' code define different max > >> vcpu number. > > > > Right, and? > > If we introduce a general kconfig of max vcpus for all vendors, it > should have different max vcpu range for different vendor. Sounds sensible as well. But based on this thread it seems that the issue of what is 'supported' vs what is in the code is completely at odds of each other. Meaning you may as well go forth and put in a huge amount and it would be OK with the maintainers? > > > > > -- > Best regards > Tianyu Lan
2017-08-18 09:57-0400, Konrad Rzeszutek Wilk: > On Tue, Aug 15, 2017 at 06:13:29PM +0200, Radim Krčmář wrote: > > (Missed this mail before my last reply.) > > > > 2017-08-15 10:10-0400, Konrad Rzeszutek Wilk: > > > On Tue, Aug 15, 2017 at 11:00:04AM +0800, Lan Tianyu wrote: > > > > On 2017年08月12日 03:35, Konrad Rzeszutek Wilk wrote: > > > > > Migration with 352 CPUs all being busy dirtying memory and also poking > > > > > at various I/O ports (say all of them dirtying the VGA) is no problem? > > > > > > > > This depends on what kind of workload is running during migration. I > > > > think this may affect service down time since there maybe a lot of dirty > > > > memory data to transfer after stopping vcpus. This also depends on how > > > > user sets "migrate_set_downtime" for qemu. But I think increasing vcpus > > > > will break migration function. > > > > > > OK, so let me take a step back. > > > > > > I see this nice 'supported' CPU count that is exposed in kvm module. > > > > > > Then there is QEMU throwing out a warning if you crank up the CPU count > > > above that number. > > > > I find the range between "recommended max" and "hard max" VCPU count > > confusing at best ... IIUC, it was there because KVM internals had > > problems with scaling and we will hit more in the future because some > > loops still are linear on VCPU count. > > Is that documented somewhere? There are some folks would be interested > in looking at that if it was known what exactly to look for.. Not really, Documentation/virtual/kvm/api.txt says: The recommended max_vcpus value can be retrieved using the KVM_CAP_NR_VCPUS of the KVM_CHECK_EXTENSION ioctl() at run-time. And "recommended" is not explained any further. We can only state that the value has no connection with userspace functionality, because it is provided by KVM. Red Hat was raising the KVM_CAP_NR_VCPUS after testing on a machine that had enough physical cores. (PLE had to be slightly optimized when going to 240.) > > The exposed value doesn't say whether migration will work, because that > > is a userspace thing and we're not aware of bottlenecks on the KVM side. > > > > > Red Hat's web-pages talk about CPU count as well. > > > > > > And I am assuming all of those are around what has been tested and > > > what has shown to work. And one of those test-cases surely must > > > be migration. > > > > Right, Red Hat will only allow/support what it has tested, even if > > upstream has a practically unlimited count. I think the upstream number > > used to be raised by Red Hat, which is why upstream isn't at the hard > > implementation limit ... > > Aim for the sky! Perhaps then lets crank it up to 4096 upstream and let > each vendor/distro/cloud decide the right number based on their > testing. And hit the ceiling. :) NR_CPUS seems like a good number upstream.
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 87ac4fb..2cdc043 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -35,7 +35,7 @@ #include <asm/asm.h> #include <asm/kvm_page_track.h> -#define KVM_MAX_VCPUS 288 +#define KVM_MAX_VCPUS 352 #define KVM_SOFT_MAX_VCPUS 240 #define KVM_MAX_VCPU_ID 1023 #define KVM_USER_MEM_SLOTS 509
Intel Xeon phi chip will support 352 logical threads. For HPC usage case, it will create a huge VM with vcpu number as same as host cpus. This patch is to increase max vcpu number to 352. Signed-off-by: Lan Tianyu <tianyu.lan@intel.com> --- arch/x86/include/asm/kvm_host.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)