hw/acpi/aml-build.c | 75 +++++++++++++++ hw/arm/virt-acpi-build.c | 8 +- hw/arm/virt.c | 171 +++++++++++++++++++++++++++++++++-- hw/core/machine.c | 7 ++ hw/i386/pc.c | 7 ++ include/hw/acpi/aml-build.h | 7 ++ include/hw/boards.h | 1 + include/sysemu/device_tree.h | 1 + qemu-options.hx | 24 +++-- softmmu/device_tree.c | 44 ++++++++- softmmu/vl.c | 3 + 11 files changed, 326 insertions(+), 22 deletions(-)
Hi, This is v4 of the series [1] that I posted to introduce support for generating cpu topology descriptions to guest. Comments are welcome! Description: Once the view of an accurate virtual cpu topology is provided to guest, with a well-designed vCPU pinning to the pCPU we may get a huge benefit, e.g., the scheduling performance improvement. See Dario Faggioli's research and the related performance tests in [2] for reference. So here we go, this patch series introduces cpu topology support for ARM platform. In this series, instead of quietly enforcing the support for the latest machine type, a new parameter "expose=on|off" in -smp command line is introduced to leave QEMU users a choice to decide whether to enable the feature or not. This will allow the feature to work on different machine types and also ideally compat with already in-use -smp command lines. Also we make much stricter requirement for the topology configuration with "expose=on". Furthermore, both cpu-map in DT and ACPI PPTT table are introduced to present cpu topology to the guest. And an ARM-specific -smp parsing function virt_smp_parse is introduced, which shares the same logic with smp_parse() when "expose=off" and follow the stricter parsing rule when "expose=on". [1] https://patchwork.kernel.org/project/qemu-devel/cover/20210516102900.28036-1-wangyanan55@huawei.com/ [2] https://kvmforum2020.sched.com/event/eE1y/virtual-topology-for-virtual-machines-friend-or-foe-dario-faggioli-suse Test results about exposure of topology: After applying this patch series, launch an ACPI guest with virt-6.1 on an ARM server. 1) Enable the support: With cmdline: -smp 96,sockets=2,cores=48,threads=1,expose=on or cmdline: -smp 96,maxcpus=96,sockets=2,cores=48,threads=1,expose=on we get: Architecture: aarch64 Byte Order: Little Endian CPU(s): 96 On-line CPU(s) list: 0-95 Thread(s) per core: 1 Core(s) per socket: 48 Socket(s): 2 NUMA node(s): 1 Vendor ID: 0x48 Model: 0 Stepping: 0x1 BogoMIPS: 200.00 NUMA node0 CPU(s): 0-95 2) Disable the support: With cmdline: -smp 96 or cmdline: -smp 96,expose=off we get: Architecture: aarch64 Byte Order: Little Endian CPU(s): 96 On-line CPU(s) list: 0-95 Thread(s) per core: 1 Core(s) per socket: 96 Socket(s): 1 NUMA node(s): 1 Vendor ID: 0x48 Model: 0 Stepping: 0x1 BogoMIPS: 200.00 NUMA node0 CPU(s): 0-95 --- Changelogs: v3->v4: - add new -smp parameter "expose=on|off" for users to enable/disable the feature - add stricter -smp cmdline parsing rules on "expose=on" case - move build_pptt to generic aml-build.c - add default cluster node in the cpu-map - rebase on top of latest upstream master - v3: https://patchwork.kernel.org/project/qemu-devel/cover/20210516102900.28036-1-wangyanan55@huawei.com/ v2->v3: - address comments from David, Philippe, and Andrew. Thanks! - split some change into separate commits for ease of review - adjust parsing rules of virt_smp_parse to be more strict (after discussion with Andrew) - adjust author credit for the patches - v2: https://patchwork.kernel.org/project/qemu-devel/cover/20210413080745.33004-1-wangyanan55@huawei.com/ v1->v2: - Address Andrew Jones's comments - Address Michael S. Tsirkin's comments - v1: https://patchwork.kernel.org/project/qemu-devel/cover/20210225085627.2263-1-fangying1@huawei.com/ --- Andrew Jones (2): hw/arm/virt: Add cpu-map to device tree hw/acpi/aml-build: Generate PPTT table Yanan Wang (5): vl: Add expose=on|off option support in -smp command line hw/arm/virt: Add separate -smp parsing function for ARM machines machine: disallow -smp expose=on for non-ARM machines device_tree: Add qemu_fdt_add_path hw/acpi/aml-build: Add Processor hierarchy node structure hw/acpi/aml-build.c | 75 +++++++++++++++ hw/arm/virt-acpi-build.c | 8 +- hw/arm/virt.c | 171 +++++++++++++++++++++++++++++++++-- hw/core/machine.c | 7 ++ hw/i386/pc.c | 7 ++ include/hw/acpi/aml-build.h | 7 ++ include/hw/boards.h | 1 + include/sysemu/device_tree.h | 1 + qemu-options.hx | 24 +++-- softmmu/device_tree.c | 44 ++++++++- softmmu/vl.c | 3 + 11 files changed, 326 insertions(+), 22 deletions(-) -- 2.23.0
On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > Hi, > > This is v4 of the series [1] that I posted to introduce support for > generating cpu topology descriptions to guest. Comments are welcome! > > Description: > Once the view of an accurate virtual cpu topology is provided to guest, > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > e.g., the scheduling performance improvement. See Dario Faggioli's > research and the related performance tests in [2] for reference. So here > we go, this patch series introduces cpu topology support for ARM platform. > > In this series, instead of quietly enforcing the support for the latest > machine type, a new parameter "expose=on|off" in -smp command line is > introduced to leave QEMU users a choice to decide whether to enable the > feature or not. This will allow the feature to work on different machine > types and also ideally compat with already in-use -smp command lines. > Also we make much stricter requirement for the topology configuration > with "expose=on". Seeing this 'expose=on' parameter feels to me like we're adding a "make-it-work=yes" parameter. IMHO this is just something that should be done by default for the current machine type version and beyond. I don't see the need for a parameter to turnthis on, especially since it is being made architecture specific. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > > Hi, > > > > This is v4 of the series [1] that I posted to introduce support for > > generating cpu topology descriptions to guest. Comments are welcome! > > > > Description: > > Once the view of an accurate virtual cpu topology is provided to guest, > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > > e.g., the scheduling performance improvement. See Dario Faggioli's > > research and the related performance tests in [2] for reference. So here > > we go, this patch series introduces cpu topology support for ARM platform. > > > > In this series, instead of quietly enforcing the support for the latest > > machine type, a new parameter "expose=on|off" in -smp command line is > > introduced to leave QEMU users a choice to decide whether to enable the > > feature or not. This will allow the feature to work on different machine > > types and also ideally compat with already in-use -smp command lines. > > Also we make much stricter requirement for the topology configuration > > with "expose=on". > > Seeing this 'expose=on' parameter feels to me like we're adding a > "make-it-work=yes" parameter. IMHO this is just something that should > be done by default for the current machine type version and beyond. > I don't see the need for a parameter to turnthis on, especially since > it is being made architecture specific. > I agree. Yanan, we never discussed an "expose" parameter in the previous versions of this series. We discussed a "strict" parameter though, which would allow existing command lines to "work" using assumptions of what the user meant and strict=on users to get what they mean or an error saying that they asked for something that won't work or would require unreasonable assumptions. Why was this changed to an "expose" parameter? Thanks, drew
On 2021/6/22 19:46, Andrew Jones wrote: > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: >> On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: >>> Hi, >>> >>> This is v4 of the series [1] that I posted to introduce support for >>> generating cpu topology descriptions to guest. Comments are welcome! >>> >>> Description: >>> Once the view of an accurate virtual cpu topology is provided to guest, >>> with a well-designed vCPU pinning to the pCPU we may get a huge benefit, >>> e.g., the scheduling performance improvement. See Dario Faggioli's >>> research and the related performance tests in [2] for reference. So here >>> we go, this patch series introduces cpu topology support for ARM platform. >>> >>> In this series, instead of quietly enforcing the support for the latest >>> machine type, a new parameter "expose=on|off" in -smp command line is >>> introduced to leave QEMU users a choice to decide whether to enable the >>> feature or not. This will allow the feature to work on different machine >>> types and also ideally compat with already in-use -smp command lines. >>> Also we make much stricter requirement for the topology configuration >>> with "expose=on". >> Seeing this 'expose=on' parameter feels to me like we're adding a >> "make-it-work=yes" parameter. IMHO this is just something that should >> be done by default for the current machine type version and beyond. >> I don't see the need for a parameter to turnthis on, especially since >> it is being made architecture specific. >> > I agree. > > Yanan, we never discussed an "expose" parameter in the previous versions > of this series. We discussed a "strict" parameter though, which would > allow existing command lines to "work" using assumptions of what the user > meant and strict=on users to get what they mean or an error saying that > they asked for something that won't work or would require unreasonable > assumptions. Why was this changed to an "expose" parameter? Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] of this series. [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ And in the discussion, we hoped things would work like below with "strict" parameter: Users who want to describe cpu topology should provide cmdline like -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 and in this case we require an more accurate -smp configuration and then generate the cpu topology description through ACPI/DT. While without a strict description, no cpu topology description would be generated, so they get nothing through ACPI/DT. It seems to me that the "strict" parameter actually serves as a knob to turn on/off the exposure of topology, and this is the reason I changed the name. Thanks, Yanan . > Thanks, > drew > > .
On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: > > > On 2021/6/22 19:46, Andrew Jones wrote: > > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: > > > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > > > > Hi, > > > > > > > > This is v4 of the series [1] that I posted to introduce support for > > > > generating cpu topology descriptions to guest. Comments are welcome! > > > > > > > > Description: > > > > Once the view of an accurate virtual cpu topology is provided to guest, > > > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > > > > e.g., the scheduling performance improvement. See Dario Faggioli's > > > > research and the related performance tests in [2] for reference. So here > > > > we go, this patch series introduces cpu topology support for ARM platform. > > > > > > > > In this series, instead of quietly enforcing the support for the latest > > > > machine type, a new parameter "expose=on|off" in -smp command line is > > > > introduced to leave QEMU users a choice to decide whether to enable the > > > > feature or not. This will allow the feature to work on different machine > > > > types and also ideally compat with already in-use -smp command lines. > > > > Also we make much stricter requirement for the topology configuration > > > > with "expose=on". > > > Seeing this 'expose=on' parameter feels to me like we're adding a > > > "make-it-work=yes" parameter. IMHO this is just something that should > > > be done by default for the current machine type version and beyond. > > > I don't see the need for a parameter to turnthis on, especially since > > > it is being made architecture specific. > > > > > I agree. > > > > Yanan, we never discussed an "expose" parameter in the previous versions > > of this series. We discussed a "strict" parameter though, which would > > allow existing command lines to "work" using assumptions of what the user > > meant and strict=on users to get what they mean or an error saying that > > they asked for something that won't work or would require unreasonable > > assumptions. Why was this changed to an "expose" parameter? > Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] > of this series. > [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ > > And in the discussion, we hoped things would work like below with "strict" > parameter: > Users who want to describe cpu topology should provide cmdline like > > -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 > > and in this case we require an more accurate -smp configuration and > then generate the cpu topology description through ACPI/DT. > > While without a strict description, no cpu topology description would > be generated, so they get nothing through ACPI/DT. > > It seems to me that the "strict" parameter actually serves as a knob to > turn on/off the exposure of topology, and this is the reason I changed > the name. Yes, the use of 'strict=on' is no better than expose=on IMHO. If I give QEMU a cli -smp cpus=4,sockets=2,cores=2,threads=1 then I expect that topology to be exposed to the guest. I shouldn't have to add extra flags to make that happen. Looking at the thread, it seems the concern was around the fact that the settings were not honoured historically and thus the CLI values could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 A similar problem existed on x86 platforms. When we made that stricter we had cde that issued a warning for a few releases, essentially deprecating the config. EVentually it was turned into a fatal error. This gave applications time to fix their broken configs, while having correct configs "just work". I'd suggest doing the same for arm. If the -smp args are semantically valid then expose the topology automatically (for new machine type). If the -smp args are semantically broken, then issue a warning. In a few releases time, turn this warning into an error. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
Hi Daniel, On 2021/6/22 20:41, Daniel P. Berrangé wrote: > On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: >> >> On 2021/6/22 19:46, Andrew Jones wrote: >>> On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: >>>> On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: >>>>> Hi, >>>>> >>>>> This is v4 of the series [1] that I posted to introduce support for >>>>> generating cpu topology descriptions to guest. Comments are welcome! >>>>> >>>>> Description: >>>>> Once the view of an accurate virtual cpu topology is provided to guest, >>>>> with a well-designed vCPU pinning to the pCPU we may get a huge benefit, >>>>> e.g., the scheduling performance improvement. See Dario Faggioli's >>>>> research and the related performance tests in [2] for reference. So here >>>>> we go, this patch series introduces cpu topology support for ARM platform. >>>>> >>>>> In this series, instead of quietly enforcing the support for the latest >>>>> machine type, a new parameter "expose=on|off" in -smp command line is >>>>> introduced to leave QEMU users a choice to decide whether to enable the >>>>> feature or not. This will allow the feature to work on different machine >>>>> types and also ideally compat with already in-use -smp command lines. >>>>> Also we make much stricter requirement for the topology configuration >>>>> with "expose=on". >>>> Seeing this 'expose=on' parameter feels to me like we're adding a >>>> "make-it-work=yes" parameter. IMHO this is just something that should >>>> be done by default for the current machine type version and beyond. >>>> I don't see the need for a parameter to turnthis on, especially since >>>> it is being made architecture specific. >>>> >>> I agree. >>> >>> Yanan, we never discussed an "expose" parameter in the previous versions >>> of this series. We discussed a "strict" parameter though, which would >>> allow existing command lines to "work" using assumptions of what the user >>> meant and strict=on users to get what they mean or an error saying that >>> they asked for something that won't work or would require unreasonable >>> assumptions. Why was this changed to an "expose" parameter? >> Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] >> of this series. >> [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ >> >> And in the discussion, we hoped things would work like below with "strict" >> parameter: >> Users who want to describe cpu topology should provide cmdline like >> >> -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 >> >> and in this case we require an more accurate -smp configuration and >> then generate the cpu topology description through ACPI/DT. >> >> While without a strict description, no cpu topology description would >> be generated, so they get nothing through ACPI/DT. >> >> It seems to me that the "strict" parameter actually serves as a knob to >> turn on/off the exposure of topology, and this is the reason I changed >> the name. > Yes, the use of 'strict=on' is no better than expose=on IMHO. > > If I give QEMU a cli > > -smp cpus=4,sockets=2,cores=2,threads=1 > > then I expect that topology to be exposed to the guest. I shouldn't > have to add extra flags to make that happen. > > Looking at the thread, it seems the concern was around the fact that > the settings were not honoured historically and thus the CLI values > could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong configuration, and the parsing function already report error for this case. We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" or "-smp 4, cores=1" are not acceptable any more because we are starting to expose the topology. > A similar problem existed on x86 platforms. When we made that stricter > we had cde that issued a warning for a few releases, essentially > deprecating the config. EVentually it was turned into a fatal error. > This gave applications time to fix their broken configs, while having > correct configs "just work". I understand this solution. Stop exposing topology for unqualified -smp config and report a warning message at the transitional phase, and finally incur an error for them. BTW, just want to be sure, it this a common method in QEMU development to solve this kind of compatibility issues? > I'd suggest doing the same for arm. If the -smp args are semantically > valid then expose the topology automatically (for new machine type). > If the -smp args are semantically broken, then issue a warning. In > a few releases time, turn this warning into an error. So this topology feature will only work for the current machine type and the following versions, right? Thanks, Yanan . > Regards, > Daniel
On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote:
> Hi Daniel,
>
> On 2021/6/22 20:41, Daniel P. Berrangé wrote:
> > On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote:
> > >
> > > On 2021/6/22 19:46, Andrew Jones wrote:
> > > > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote:
> > > > > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote:
> > > > > > Hi,
> > > > > >
> > > > > > This is v4 of the series [1] that I posted to introduce support for
> > > > > > generating cpu topology descriptions to guest. Comments are welcome!
> > > > > >
> > > > > > Description:
> > > > > > Once the view of an accurate virtual cpu topology is provided to guest,
> > > > > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit,
> > > > > > e.g., the scheduling performance improvement. See Dario Faggioli's
> > > > > > research and the related performance tests in [2] for reference. So here
> > > > > > we go, this patch series introduces cpu topology support for ARM platform.
> > > > > >
> > > > > > In this series, instead of quietly enforcing the support for the latest
> > > > > > machine type, a new parameter "expose=on|off" in -smp command line is
> > > > > > introduced to leave QEMU users a choice to decide whether to enable the
> > > > > > feature or not. This will allow the feature to work on different machine
> > > > > > types and also ideally compat with already in-use -smp command lines.
> > > > > > Also we make much stricter requirement for the topology configuration
> > > > > > with "expose=on".
> > > > > Seeing this 'expose=on' parameter feels to me like we're adding a
> > > > > "make-it-work=yes" parameter. IMHO this is just something that should
> > > > > be done by default for the current machine type version and beyond.
> > > > > I don't see the need for a parameter to turnthis on, especially since
> > > > > it is being made architecture specific.
> > > > >
> > > > I agree.
> > > >
> > > > Yanan, we never discussed an "expose" parameter in the previous versions
> > > > of this series. We discussed a "strict" parameter though, which would
> > > > allow existing command lines to "work" using assumptions of what the user
> > > > meant and strict=on users to get what they mean or an error saying that
> > > > they asked for something that won't work or would require unreasonable
> > > > assumptions. Why was this changed to an "expose" parameter?
> > > Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1]
> > > of this series.
> > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/
> > >
> > > And in the discussion, we hoped things would work like below with "strict"
> > > parameter:
> > > Users who want to describe cpu topology should provide cmdline like
> > >
> > > -smp strict=on,cpus=4,sockets=2,cores=2,threads=1
> > >
> > > and in this case we require an more accurate -smp configuration and
> > > then generate the cpu topology description through ACPI/DT.
> > >
> > > While without a strict description, no cpu topology description would
> > > be generated, so they get nothing through ACPI/DT.
> > >
> > > It seems to me that the "strict" parameter actually serves as a knob to
> > > turn on/off the exposure of topology, and this is the reason I changed
> > > the name.
> > Yes, the use of 'strict=on' is no better than expose=on IMHO.
> >
> > If I give QEMU a cli
> >
> > -smp cpus=4,sockets=2,cores=2,threads=1
> >
> > then I expect that topology to be exposed to the guest. I shouldn't
> > have to add extra flags to make that happen.
> >
> > Looking at the thread, it seems the concern was around the fact that
> > the settings were not honoured historically and thus the CLI values
> > could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9
> This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong
> configuration, and the parsing function already report error for this case.
>
> We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1"
> for exposure of topology, and the incomplete ones like "-smp 4,sockets=1"
> or "-smp 4, cores=1" are not acceptable any more because we are starting
> to expose the topology.
Incomplete specified topologies *are* acceptable.
The smp_parse method will automatically fill in any missing values.
ie,
-smp 4,cores=1
-smp cores=1
-smp threads=1
-smp sockets=4
are all functionally identical to
-smp 4,sockets=4,cores=1,dies=1,threads=1
The QEMU man page says this explicitly
For the PC target, the number of cores per die, the
number of threads per cores, the number of dies per packages and the
total number of sockets can be specified. Missing values will be
computed. If any on the three values is given, the total number of
CPUs n can be omitted.
note this qemu-options.hx doc will require updating since it will apply
to more than just the PC target.
Regards,
Daniel
--
|: https://berrange.com -o- https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org -o- https://fstop138.berrange.com :|
|: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
On Tue, 22 Jun 2021 at 15:11, Daniel P. Berrangé <berrange@redhat.com> wrote: > The QEMU man page says this explicitly > > For the PC target, the number of cores per die, the > number of threads per cores, the number of dies per packages and the > total number of sockets can be specified. Missing values will be > computed. If any on the three values is given, the total number of > CPUs n can be omitted. Anybody feel like submitting a patch to fix the typo? Should read "If any of"... -- PMM
On Tue, Jun 22, 2021 at 03:15:18PM +0100, Peter Maydell wrote: > On Tue, 22 Jun 2021 at 15:11, Daniel P. Berrangé <berrange@redhat.com> wrote: > > The QEMU man page says this explicitly > > > > For the PC target, the number of cores per die, the > > number of threads per cores, the number of dies per packages and the > > total number of sockets can be specified. Missing values will be > > computed. If any on the three values is given, the total number of > > CPUs n can be omitted. > > Anybody feel like submitting a patch to fix the typo? Should read > "If any of"... Actually looking at the broader text for -smp, the whole thing just needs to be rewritten from scratch IMHO. I'll send a patch.... Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
Hi Daniel, On 2021/6/22 22:28, Daniel P. Berrangé wrote: > On Tue, Jun 22, 2021 at 03:15:18PM +0100, Peter Maydell wrote: >> On Tue, 22 Jun 2021 at 15:11, Daniel P. Berrangé <berrange@redhat.com> wrote: >>> The QEMU man page says this explicitly >>> >>> For the PC target, the number of cores per die, the >>> number of threads per cores, the number of dies per packages and the >>> total number of sockets can be specified. Missing values will be >>> computed. If any on the three values is given, the total number of >>> CPUs n can be omitted. >> Anybody feel like submitting a patch to fix the typo? Should read >> "If any of"... > Actually looking at the broader text for -smp, the whole thing just needs > to be rewritten from scratch IMHO. I'll send a patch.... Do you have any plan to rewrite the -smp text in qemu-options.hx recently? I suggest that we explain in more detail how the missing values will be computed, so that qemu users can have a clear image of what a -smp parameter collection would be parsed out if they are using an incomplete cmdline and have read the man page. Also if we all agree to prefer cores over sockets for all arches since machine type 6.2, I'll be glad to add this change for the man page and update the parsing helpers in next version of this patch series. Thanks, Yanan .
On Mon, Jun 28, 2021 at 07:14:02PM +0800, wangyanan (Y) wrote: > Hi Daniel, > > On 2021/6/22 22:28, Daniel P. Berrangé wrote: > > On Tue, Jun 22, 2021 at 03:15:18PM +0100, Peter Maydell wrote: > > > On Tue, 22 Jun 2021 at 15:11, Daniel P. Berrangé <berrange@redhat.com> wrote: > > > > The QEMU man page says this explicitly > > > > > > > > For the PC target, the number of cores per die, the > > > > number of threads per cores, the number of dies per packages and the > > > > total number of sockets can be specified. Missing values will be > > > > computed. If any on the three values is given, the total number of > > > > CPUs n can be omitted. > > > Anybody feel like submitting a patch to fix the typo? Should read > > > "If any of"... > > Actually looking at the broader text for -smp, the whole thing just needs > > to be rewritten from scratch IMHO. I'll send a patch.... > Do you have any plan to rewrite the -smp text in qemu-options.hx recently? > > I suggest that we explain in more detail how the missing values will > be computed, so that qemu users can have a clear image of what a > -smp parameter collection would be parsed out if they are using an > incomplete cmdline and have read the man page. > > Also if we all agree to prefer cores over sockets for all arches since > machine type 6.2, I'll be glad to add this change for the man page > and update the parsing helpers in next version of this patch series. I wrote a docs update on friday, which I've just sent out as a short patch series with yourself CC'd. I wrote it such that we can easily update it again, if we want to prefer cores over sockets. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
On 2021/6/28 19:31, Daniel P. Berrangé wrote: > On Mon, Jun 28, 2021 at 07:14:02PM +0800, wangyanan (Y) wrote: >> Hi Daniel, >> >> On 2021/6/22 22:28, Daniel P. Berrangé wrote: >>> On Tue, Jun 22, 2021 at 03:15:18PM +0100, Peter Maydell wrote: >>>> On Tue, 22 Jun 2021 at 15:11, Daniel P. Berrangé <berrange@redhat.com> wrote: >>>>> The QEMU man page says this explicitly >>>>> >>>>> For the PC target, the number of cores per die, the >>>>> number of threads per cores, the number of dies per packages and the >>>>> total number of sockets can be specified. Missing values will be >>>>> computed. If any on the three values is given, the total number of >>>>> CPUs n can be omitted. >>>> Anybody feel like submitting a patch to fix the typo? Should read >>>> "If any of"... >>> Actually looking at the broader text for -smp, the whole thing just needs >>> to be rewritten from scratch IMHO. I'll send a patch.... >> Do you have any plan to rewrite the -smp text in qemu-options.hx recently? >> >> I suggest that we explain in more detail how the missing values will >> be computed, so that qemu users can have a clear image of what a >> -smp parameter collection would be parsed out if they are using an >> incomplete cmdline and have read the man page. >> >> Also if we all agree to prefer cores over sockets for all arches since >> machine type 6.2, I'll be glad to add this change for the man page >> and update the parsing helpers in next version of this patch series. > I wrote a docs update on friday, which I've just sent out as a short > patch series with yourself CC'd. I wrote it such that we can easily > update it again, if we want to prefer cores over sockets. > Great, thanks. I'm going to have a look. Thanks, Yanan .
On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: > On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: > > Hi Daniel, > > > > On 2021/6/22 20:41, Daniel P. Berrangé wrote: > > > On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: > > > > > > > > On 2021/6/22 19:46, Andrew Jones wrote: > > > > > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: > > > > > > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > > > > > > > Hi, > > > > > > > > > > > > > > This is v4 of the series [1] that I posted to introduce support for > > > > > > > generating cpu topology descriptions to guest. Comments are welcome! > > > > > > > > > > > > > > Description: > > > > > > > Once the view of an accurate virtual cpu topology is provided to guest, > > > > > > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > > > > > > > e.g., the scheduling performance improvement. See Dario Faggioli's > > > > > > > research and the related performance tests in [2] for reference. So here > > > > > > > we go, this patch series introduces cpu topology support for ARM platform. > > > > > > > > > > > > > > In this series, instead of quietly enforcing the support for the latest > > > > > > > machine type, a new parameter "expose=on|off" in -smp command line is > > > > > > > introduced to leave QEMU users a choice to decide whether to enable the > > > > > > > feature or not. This will allow the feature to work on different machine > > > > > > > types and also ideally compat with already in-use -smp command lines. > > > > > > > Also we make much stricter requirement for the topology configuration > > > > > > > with "expose=on". > > > > > > Seeing this 'expose=on' parameter feels to me like we're adding a > > > > > > "make-it-work=yes" parameter. IMHO this is just something that should > > > > > > be done by default for the current machine type version and beyond. > > > > > > I don't see the need for a parameter to turnthis on, especially since > > > > > > it is being made architecture specific. > > > > > > > > > > > I agree. > > > > > > > > > > Yanan, we never discussed an "expose" parameter in the previous versions > > > > > of this series. We discussed a "strict" parameter though, which would > > > > > allow existing command lines to "work" using assumptions of what the user > > > > > meant and strict=on users to get what they mean or an error saying that > > > > > they asked for something that won't work or would require unreasonable > > > > > assumptions. Why was this changed to an "expose" parameter? > > > > Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] > > > > of this series. > > > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ > > > > > > > > And in the discussion, we hoped things would work like below with "strict" > > > > parameter: > > > > Users who want to describe cpu topology should provide cmdline like > > > > > > > > -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > and in this case we require an more accurate -smp configuration and > > > > then generate the cpu topology description through ACPI/DT. > > > > > > > > While without a strict description, no cpu topology description would > > > > be generated, so they get nothing through ACPI/DT. > > > > > > > > It seems to me that the "strict" parameter actually serves as a knob to > > > > turn on/off the exposure of topology, and this is the reason I changed > > > > the name. > > > Yes, the use of 'strict=on' is no better than expose=on IMHO. > > > > > > If I give QEMU a cli > > > > > > -smp cpus=4,sockets=2,cores=2,threads=1 > > > > > > then I expect that topology to be exposed to the guest. I shouldn't > > > have to add extra flags to make that happen. > > > > > > Looking at the thread, it seems the concern was around the fact that > > > the settings were not honoured historically and thus the CLI values > > > could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 > > This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong > > configuration, and the parsing function already report error for this case. > > > > We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" > > for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" > > or "-smp 4, cores=1" are not acceptable any more because we are starting > > to expose the topology. > > Incomplete specified topologies *are* acceptable. > > The smp_parse method will automatically fill in any missing values. > > ie, > > -smp 4,cores=1 > -smp cores=1 > -smp threads=1 > -smp sockets=4 > > are all functionally identical to > > -smp 4,sockets=4,cores=1,dies=1,threads=1 > > > The QEMU man page says this explicitly > > For the PC target, the number of cores per die, the > number of threads per cores, the number of dies per packages and the > total number of sockets can be specified. Missing values will be > computed. If any on the three values is given, the total number of > CPUs n can be omitted. It doesn't say how it will compute them though, which for the default smp_parse and for x86 is to prefer sockets over cores over threads. That's not necessarily what the user expects. IMO, we need a 'strict=on' parameter that doesn't allow any collection of smp parameters which require unreasonable assumptions. Reasonable assumptions are threads=1, when threads is not specified and the rest of the math adds up. Also, maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not as reasonable to decide how to divide cores among sockets or to assume threads=1 when only sockets and cores are given. How do we know the user didn't forget to specify threads if we can't check the math? Thanks, drew > > note this qemu-options.hx doc will require updating since it will apply > to more than just the PC target. > > Regards, > Daniel > -- > |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| > |: https://libvirt.org -o- https://fstop138.berrange.com :| > |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :| >
On Tue, Jun 22, 2021 at 04:29:15PM +0200, Andrew Jones wrote: > On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: > > On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: > > > Hi Daniel, > > > > > > On 2021/6/22 20:41, Daniel P. Berrangé wrote: > > > > On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: > > > > > > > > > > On 2021/6/22 19:46, Andrew Jones wrote: > > > > > > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: > > > > > > > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > > > > > > > > Hi, > > > > > > > > > > > > > > > > This is v4 of the series [1] that I posted to introduce support for > > > > > > > > generating cpu topology descriptions to guest. Comments are welcome! > > > > > > > > > > > > > > > > Description: > > > > > > > > Once the view of an accurate virtual cpu topology is provided to guest, > > > > > > > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > > > > > > > > e.g., the scheduling performance improvement. See Dario Faggioli's > > > > > > > > research and the related performance tests in [2] for reference. So here > > > > > > > > we go, this patch series introduces cpu topology support for ARM platform. > > > > > > > > > > > > > > > > In this series, instead of quietly enforcing the support for the latest > > > > > > > > machine type, a new parameter "expose=on|off" in -smp command line is > > > > > > > > introduced to leave QEMU users a choice to decide whether to enable the > > > > > > > > feature or not. This will allow the feature to work on different machine > > > > > > > > types and also ideally compat with already in-use -smp command lines. > > > > > > > > Also we make much stricter requirement for the topology configuration > > > > > > > > with "expose=on". > > > > > > > Seeing this 'expose=on' parameter feels to me like we're adding a > > > > > > > "make-it-work=yes" parameter. IMHO this is just something that should > > > > > > > be done by default for the current machine type version and beyond. > > > > > > > I don't see the need for a parameter to turnthis on, especially since > > > > > > > it is being made architecture specific. > > > > > > > > > > > > > I agree. > > > > > > > > > > > > Yanan, we never discussed an "expose" parameter in the previous versions > > > > > > of this series. We discussed a "strict" parameter though, which would > > > > > > allow existing command lines to "work" using assumptions of what the user > > > > > > meant and strict=on users to get what they mean or an error saying that > > > > > > they asked for something that won't work or would require unreasonable > > > > > > assumptions. Why was this changed to an "expose" parameter? > > > > > Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] > > > > > of this series. > > > > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ > > > > > > > > > > And in the discussion, we hoped things would work like below with "strict" > > > > > parameter: > > > > > Users who want to describe cpu topology should provide cmdline like > > > > > > > > > > -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > and in this case we require an more accurate -smp configuration and > > > > > then generate the cpu topology description through ACPI/DT. > > > > > > > > > > While without a strict description, no cpu topology description would > > > > > be generated, so they get nothing through ACPI/DT. > > > > > > > > > > It seems to me that the "strict" parameter actually serves as a knob to > > > > > turn on/off the exposure of topology, and this is the reason I changed > > > > > the name. > > > > Yes, the use of 'strict=on' is no better than expose=on IMHO. > > > > > > > > If I give QEMU a cli > > > > > > > > -smp cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > then I expect that topology to be exposed to the guest. I shouldn't > > > > have to add extra flags to make that happen. > > > > > > > > Looking at the thread, it seems the concern was around the fact that > > > > the settings were not honoured historically and thus the CLI values > > > > could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 > > > This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong > > > configuration, and the parsing function already report error for this case. > > > > > > We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" > > > for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" > > > or "-smp 4, cores=1" are not acceptable any more because we are starting > > > to expose the topology. > > > > Incomplete specified topologies *are* acceptable. > > > > The smp_parse method will automatically fill in any missing values. > > > > ie, > > > > -smp 4,cores=1 > > -smp cores=1 > > -smp threads=1 > > -smp sockets=4 > > > > are all functionally identical to > > > > -smp 4,sockets=4,cores=1,dies=1,threads=1 > > > > > > The QEMU man page says this explicitly > > > > For the PC target, the number of cores per die, the > > number of threads per cores, the number of dies per packages and the > > total number of sockets can be specified. Missing values will be > > computed. If any on the three values is given, the total number of > > CPUs n can be omitted. > > It doesn't say how it will compute them though, which for the default > smp_parse and for x86 is to prefer sockets over cores over threads. > That's not necessarily what the user expects. IMO, we need a 'strict=on' > parameter that doesn't allow any collection of smp parameters which > require unreasonable assumptions. Reasonable assumptions are threads=1, > when threads is not specified and the rest of the math adds up. Also, > maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not > as reasonable to decide how to divide cores among sockets or to assume > threads=1 when only sockets and cores are given. How do we know the user > didn't forget to specify threads if we can't check the math? It is a tradeoff between convenience and error reporting. The current approach is obviousy implemented from the POV that the convenience of defaulting parameters is more broadly useful than error reporting in fact of mistakes. If strongly caring about the specific break down of sockets, dies, cores, threads then all must be provided explicitly so the built-in computing of omitted parameters is skipped. If the user can forget to give a threads=NN param, then they can also forget to give a strict=on param, so that's not especially compelling way to improve error reporting IMHO. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
On Tue, 22 Jun 2021 16:29:15 +0200 Andrew Jones <drjones@redhat.com> wrote: > On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: > > On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: > > > Hi Daniel, > > > > > > On 2021/6/22 20:41, Daniel P. Berrangé wrote: > > > > On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: > > > > > > > > > > On 2021/6/22 19:46, Andrew Jones wrote: > > > > > > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: > > > > > > > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > > > > > > > > Hi, > > > > > > > > > > > > > > > > This is v4 of the series [1] that I posted to introduce support for > > > > > > > > generating cpu topology descriptions to guest. Comments are welcome! > > > > > > > > > > > > > > > > Description: > > > > > > > > Once the view of an accurate virtual cpu topology is provided to guest, > > > > > > > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > > > > > > > > e.g., the scheduling performance improvement. See Dario Faggioli's > > > > > > > > research and the related performance tests in [2] for reference. So here > > > > > > > > we go, this patch series introduces cpu topology support for ARM platform. > > > > > > > > > > > > > > > > In this series, instead of quietly enforcing the support for the latest > > > > > > > > machine type, a new parameter "expose=on|off" in -smp command line is > > > > > > > > introduced to leave QEMU users a choice to decide whether to enable the > > > > > > > > feature or not. This will allow the feature to work on different machine > > > > > > > > types and also ideally compat with already in-use -smp command lines. > > > > > > > > Also we make much stricter requirement for the topology configuration > > > > > > > > with "expose=on". > > > > > > > Seeing this 'expose=on' parameter feels to me like we're adding a > > > > > > > "make-it-work=yes" parameter. IMHO this is just something that should > > > > > > > be done by default for the current machine type version and beyond. > > > > > > > I don't see the need for a parameter to turnthis on, especially since > > > > > > > it is being made architecture specific. > > > > > > > > > > > > > I agree. > > > > > > > > > > > > Yanan, we never discussed an "expose" parameter in the previous versions > > > > > > of this series. We discussed a "strict" parameter though, which would > > > > > > allow existing command lines to "work" using assumptions of what the user > > > > > > meant and strict=on users to get what they mean or an error saying that > > > > > > they asked for something that won't work or would require unreasonable > > > > > > assumptions. Why was this changed to an "expose" parameter? > > > > > Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] > > > > > of this series. > > > > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ > > > > > > > > > > And in the discussion, we hoped things would work like below with "strict" > > > > > parameter: > > > > > Users who want to describe cpu topology should provide cmdline like > > > > > > > > > > -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > and in this case we require an more accurate -smp configuration and > > > > > then generate the cpu topology description through ACPI/DT. > > > > > > > > > > While without a strict description, no cpu topology description would > > > > > be generated, so they get nothing through ACPI/DT. > > > > > > > > > > It seems to me that the "strict" parameter actually serves as a knob to > > > > > turn on/off the exposure of topology, and this is the reason I changed > > > > > the name. > > > > Yes, the use of 'strict=on' is no better than expose=on IMHO. > > > > > > > > If I give QEMU a cli > > > > > > > > -smp cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > then I expect that topology to be exposed to the guest. I shouldn't > > > > have to add extra flags to make that happen. > > > > > > > > Looking at the thread, it seems the concern was around the fact that > > > > the settings were not honoured historically and thus the CLI values > > > > could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 > > > This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong > > > configuration, and the parsing function already report error for this case. > > > > > > We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" > > > for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" > > > or "-smp 4, cores=1" are not acceptable any more because we are starting > > > to expose the topology. > > > > Incomplete specified topologies *are* acceptable. > > > > The smp_parse method will automatically fill in any missing values. > > > > ie, > > > > -smp 4,cores=1 > > -smp cores=1 > > -smp threads=1 > > -smp sockets=4 > > > > are all functionally identical to > > > > -smp 4,sockets=4,cores=1,dies=1,threads=1 > > > > > > The QEMU man page says this explicitly > > > > For the PC target, the number of cores per die, the > > number of threads per cores, the number of dies per packages and the > > total number of sockets can be specified. Missing values will be > > computed. If any on the three values is given, the total number of > > CPUs n can be omitted. > > It doesn't say how it will compute them though, which for the default > smp_parse and for x86 is to prefer sockets over cores over threads. > That's not necessarily what the user expects. IMO, we need a 'strict=on' > parameter that doesn't allow any collection of smp parameters which > require unreasonable assumptions. Reasonable assumptions are threads=1, > when threads is not specified and the rest of the math adds up. Also, > maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not > as reasonable to decide how to divide cores among sockets or to assume > threads=1 when only sockets and cores are given. How do we know the user > didn't forget to specify threads if we can't check the math? or just outlaw all invalid topologies incl. incomplete by default (without requiring extra option), and permit them only for old machine types ()using compat machinery) without topo info provided to guest. And maybe later deprecate invalid topologies altogether. > > Thanks, > drew > > > > > note this qemu-options.hx doc will require updating since it will apply > > to more than just the PC target. > > > > Regards, > > Daniel > > -- > > |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| > > |: https://libvirt.org -o- https://fstop138.berrange.com :| > > |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :| > > >
On Tue, Jun 22, 2021 at 05:40:13PM +0200, Igor Mammedov wrote: > On Tue, 22 Jun 2021 16:29:15 +0200 > Andrew Jones <drjones@redhat.com> wrote: > > > On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: > > > On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: > > > > Hi Daniel, > > > > > > > > On 2021/6/22 20:41, Daniel P. Berrangé wrote: > > > > > On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: > > > > > > > > > > > > On 2021/6/22 19:46, Andrew Jones wrote: > > > > > > > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: > > > > > > > > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > > > > > > > > > Hi, > > > > > > > > > > > > > > > > > > This is v4 of the series [1] that I posted to introduce support for > > > > > > > > > generating cpu topology descriptions to guest. Comments are welcome! > > > > > > > > > > > > > > > > > > Description: > > > > > > > > > Once the view of an accurate virtual cpu topology is provided to guest, > > > > > > > > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > > > > > > > > > e.g., the scheduling performance improvement. See Dario Faggioli's > > > > > > > > > research and the related performance tests in [2] for reference. So here > > > > > > > > > we go, this patch series introduces cpu topology support for ARM platform. > > > > > > > > > > > > > > > > > > In this series, instead of quietly enforcing the support for the latest > > > > > > > > > machine type, a new parameter "expose=on|off" in -smp command line is > > > > > > > > > introduced to leave QEMU users a choice to decide whether to enable the > > > > > > > > > feature or not. This will allow the feature to work on different machine > > > > > > > > > types and also ideally compat with already in-use -smp command lines. > > > > > > > > > Also we make much stricter requirement for the topology configuration > > > > > > > > > with "expose=on". > > > > > > > > Seeing this 'expose=on' parameter feels to me like we're adding a > > > > > > > > "make-it-work=yes" parameter. IMHO this is just something that should > > > > > > > > be done by default for the current machine type version and beyond. > > > > > > > > I don't see the need for a parameter to turnthis on, especially since > > > > > > > > it is being made architecture specific. > > > > > > > > > > > > > > > I agree. > > > > > > > > > > > > > > Yanan, we never discussed an "expose" parameter in the previous versions > > > > > > > of this series. We discussed a "strict" parameter though, which would > > > > > > > allow existing command lines to "work" using assumptions of what the user > > > > > > > meant and strict=on users to get what they mean or an error saying that > > > > > > > they asked for something that won't work or would require unreasonable > > > > > > > assumptions. Why was this changed to an "expose" parameter? > > > > > > Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] > > > > > > of this series. > > > > > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ > > > > > > > > > > > > And in the discussion, we hoped things would work like below with "strict" > > > > > > parameter: > > > > > > Users who want to describe cpu topology should provide cmdline like > > > > > > > > > > > > -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > > > and in this case we require an more accurate -smp configuration and > > > > > > then generate the cpu topology description through ACPI/DT. > > > > > > > > > > > > While without a strict description, no cpu topology description would > > > > > > be generated, so they get nothing through ACPI/DT. > > > > > > > > > > > > It seems to me that the "strict" parameter actually serves as a knob to > > > > > > turn on/off the exposure of topology, and this is the reason I changed > > > > > > the name. > > > > > Yes, the use of 'strict=on' is no better than expose=on IMHO. > > > > > > > > > > If I give QEMU a cli > > > > > > > > > > -smp cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > then I expect that topology to be exposed to the guest. I shouldn't > > > > > have to add extra flags to make that happen. > > > > > > > > > > Looking at the thread, it seems the concern was around the fact that > > > > > the settings were not honoured historically and thus the CLI values > > > > > could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 > > > > This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong > > > > configuration, and the parsing function already report error for this case. > > > > > > > > We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" > > > > for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" > > > > or "-smp 4, cores=1" are not acceptable any more because we are starting > > > > to expose the topology. > > > > > > Incomplete specified topologies *are* acceptable. > > > > > > The smp_parse method will automatically fill in any missing values. > > > > > > ie, > > > > > > -smp 4,cores=1 > > > -smp cores=1 > > > -smp threads=1 > > > -smp sockets=4 > > > > > > are all functionally identical to > > > > > > -smp 4,sockets=4,cores=1,dies=1,threads=1 > > > > > > > > > The QEMU man page says this explicitly > > > > > > For the PC target, the number of cores per die, the > > > number of threads per cores, the number of dies per packages and the > > > total number of sockets can be specified. Missing values will be > > > computed. If any on the three values is given, the total number of > > > CPUs n can be omitted. > > > > It doesn't say how it will compute them though, which for the default > > smp_parse and for x86 is to prefer sockets over cores over threads. > > That's not necessarily what the user expects. IMO, we need a 'strict=on' > > parameter that doesn't allow any collection of smp parameters which > > require unreasonable assumptions. Reasonable assumptions are threads=1, > > when threads is not specified and the rest of the math adds up. Also, > > maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not > > as reasonable to decide how to divide cores among sockets or to assume > > threads=1 when only sockets and cores are given. How do we know the user > > didn't forget to specify threads if we can't check the math? > > or just outlaw all invalid topologies incl. incomplete by default > (without requiring extra option), and permit them only for old machine > types ()using compat machinery) without topo info provided to guest. > And maybe later deprecate invalid topologies altogether. I like this proposal, but Peter generally does not want currently working command lines to stop working. The 'virt' machine type always points to the latest machine type, so '-M virt -smp sockets=2,cores=4' will currently work (even though no topology is generated), but, if we were to merge patches that outlawed that using the compat machinery for 6.2, then it'll stop working, possibly breaking scripts. The 'strict' parameter allows one to opt into the strict parsing and actually get the topology described. Everyone else will still have working command lines without topology no matter what they use. If Peter agrees to making the smp parser strict from 6.2 on (possibly by issuing a warning for a release or two instead of an error, like Daniel suggested), then that's indeed better. If not, then I don't know how to get a useful -smp command line parser without the additional parameter. Thanks, drew > > > > > > Thanks, > > drew > > > > > > > > note this qemu-options.hx doc will require updating since it will apply > > > to more than just the PC target. > > > > > > Regards, > > > Daniel > > > -- > > > |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| > > > |: https://libvirt.org -o- https://fstop138.berrange.com :| > > > |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :| > > > > > >
On Tue, Jun 22, 2021 at 05:40:13PM +0200, Igor Mammedov wrote: > On Tue, 22 Jun 2021 16:29:15 +0200 > Andrew Jones <drjones@redhat.com> wrote: > > > On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: > > > On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: > > > > Hi Daniel, > > > > > > > > On 2021/6/22 20:41, Daniel P. Berrangé wrote: > > > > > On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: > > > > > > > > > > > > On 2021/6/22 19:46, Andrew Jones wrote: > > > > > > > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: > > > > > > > > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > > > > > > > > > Hi, > > > > > > > > > > > > > > > > > > This is v4 of the series [1] that I posted to introduce support for > > > > > > > > > generating cpu topology descriptions to guest. Comments are welcome! > > > > > > > > > > > > > > > > > > Description: > > > > > > > > > Once the view of an accurate virtual cpu topology is provided to guest, > > > > > > > > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > > > > > > > > > e.g., the scheduling performance improvement. See Dario Faggioli's > > > > > > > > > research and the related performance tests in [2] for reference. So here > > > > > > > > > we go, this patch series introduces cpu topology support for ARM platform. > > > > > > > > > > > > > > > > > > In this series, instead of quietly enforcing the support for the latest > > > > > > > > > machine type, a new parameter "expose=on|off" in -smp command line is > > > > > > > > > introduced to leave QEMU users a choice to decide whether to enable the > > > > > > > > > feature or not. This will allow the feature to work on different machine > > > > > > > > > types and also ideally compat with already in-use -smp command lines. > > > > > > > > > Also we make much stricter requirement for the topology configuration > > > > > > > > > with "expose=on". > > > > > > > > Seeing this 'expose=on' parameter feels to me like we're adding a > > > > > > > > "make-it-work=yes" parameter. IMHO this is just something that should > > > > > > > > be done by default for the current machine type version and beyond. > > > > > > > > I don't see the need for a parameter to turnthis on, especially since > > > > > > > > it is being made architecture specific. > > > > > > > > > > > > > > > I agree. > > > > > > > > > > > > > > Yanan, we never discussed an "expose" parameter in the previous versions > > > > > > > of this series. We discussed a "strict" parameter though, which would > > > > > > > allow existing command lines to "work" using assumptions of what the user > > > > > > > meant and strict=on users to get what they mean or an error saying that > > > > > > > they asked for something that won't work or would require unreasonable > > > > > > > assumptions. Why was this changed to an "expose" parameter? > > > > > > Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] > > > > > > of this series. > > > > > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ > > > > > > > > > > > > And in the discussion, we hoped things would work like below with "strict" > > > > > > parameter: > > > > > > Users who want to describe cpu topology should provide cmdline like > > > > > > > > > > > > -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > > > and in this case we require an more accurate -smp configuration and > > > > > > then generate the cpu topology description through ACPI/DT. > > > > > > > > > > > > While without a strict description, no cpu topology description would > > > > > > be generated, so they get nothing through ACPI/DT. > > > > > > > > > > > > It seems to me that the "strict" parameter actually serves as a knob to > > > > > > turn on/off the exposure of topology, and this is the reason I changed > > > > > > the name. > > > > > Yes, the use of 'strict=on' is no better than expose=on IMHO. > > > > > > > > > > If I give QEMU a cli > > > > > > > > > > -smp cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > then I expect that topology to be exposed to the guest. I shouldn't > > > > > have to add extra flags to make that happen. > > > > > > > > > > Looking at the thread, it seems the concern was around the fact that > > > > > the settings were not honoured historically and thus the CLI values > > > > > could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 > > > > This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong > > > > configuration, and the parsing function already report error for this case. > > > > > > > > We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" > > > > for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" > > > > or "-smp 4, cores=1" are not acceptable any more because we are starting > > > > to expose the topology. > > > > > > Incomplete specified topologies *are* acceptable. > > > > > > The smp_parse method will automatically fill in any missing values. > > > > > > ie, > > > > > > -smp 4,cores=1 > > > -smp cores=1 > > > -smp threads=1 > > > -smp sockets=4 > > > > > > are all functionally identical to > > > > > > -smp 4,sockets=4,cores=1,dies=1,threads=1 > > > > > > > > > The QEMU man page says this explicitly > > > > > > For the PC target, the number of cores per die, the > > > number of threads per cores, the number of dies per packages and the > > > total number of sockets can be specified. Missing values will be > > > computed. If any on the three values is given, the total number of > > > CPUs n can be omitted. > > > > It doesn't say how it will compute them though, which for the default > > smp_parse and for x86 is to prefer sockets over cores over threads. > > That's not necessarily what the user expects. IMO, we need a 'strict=on' > > parameter that doesn't allow any collection of smp parameters which > > require unreasonable assumptions. Reasonable assumptions are threads=1, > > when threads is not specified and the rest of the math adds up. Also, > > maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not > > as reasonable to decide how to divide cores among sockets or to assume > > threads=1 when only sockets and cores are given. How do we know the user > > didn't forget to specify threads if we can't check the math? > > or just outlaw all invalid topologies incl. incomplete by default > (without requiring extra option), and permit them only for old machine > types ()using compat machinery) without topo info provided to guest. > And maybe later deprecate invalid topologies altogether. This feels like it is creating pain for users to fix a problem that isn't shown to actually be causing any common issues. We've supposed that users are having problems when forgetting to specify "threads" and not having the compute value be desirable, but where are the bug reports to back this up ? The partial topologies are valid and have well defined semantics. Those semantics may not match everyone's preference, but that doesn't make them invalid. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
On Tue, Jun 22, 2021 at 06:14:25PM +0100, Daniel P. Berrangé wrote: > On Tue, Jun 22, 2021 at 05:40:13PM +0200, Igor Mammedov wrote: > > On Tue, 22 Jun 2021 16:29:15 +0200 > > Andrew Jones <drjones@redhat.com> wrote: > > > > > On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: > > > > On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: > > > > > Hi Daniel, > > > > > > > > > > On 2021/6/22 20:41, Daniel P. Berrangé wrote: > > > > > > On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: > > > > > > > > > > > > > > On 2021/6/22 19:46, Andrew Jones wrote: > > > > > > > > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: > > > > > > > > > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > > > > > > > > > > Hi, > > > > > > > > > > > > > > > > > > > > This is v4 of the series [1] that I posted to introduce support for > > > > > > > > > > generating cpu topology descriptions to guest. Comments are welcome! > > > > > > > > > > > > > > > > > > > > Description: > > > > > > > > > > Once the view of an accurate virtual cpu topology is provided to guest, > > > > > > > > > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > > > > > > > > > > e.g., the scheduling performance improvement. See Dario Faggioli's > > > > > > > > > > research and the related performance tests in [2] for reference. So here > > > > > > > > > > we go, this patch series introduces cpu topology support for ARM platform. > > > > > > > > > > > > > > > > > > > > In this series, instead of quietly enforcing the support for the latest > > > > > > > > > > machine type, a new parameter "expose=on|off" in -smp command line is > > > > > > > > > > introduced to leave QEMU users a choice to decide whether to enable the > > > > > > > > > > feature or not. This will allow the feature to work on different machine > > > > > > > > > > types and also ideally compat with already in-use -smp command lines. > > > > > > > > > > Also we make much stricter requirement for the topology configuration > > > > > > > > > > with "expose=on". > > > > > > > > > Seeing this 'expose=on' parameter feels to me like we're adding a > > > > > > > > > "make-it-work=yes" parameter. IMHO this is just something that should > > > > > > > > > be done by default for the current machine type version and beyond. > > > > > > > > > I don't see the need for a parameter to turnthis on, especially since > > > > > > > > > it is being made architecture specific. > > > > > > > > > > > > > > > > > I agree. > > > > > > > > > > > > > > > > Yanan, we never discussed an "expose" parameter in the previous versions > > > > > > > > of this series. We discussed a "strict" parameter though, which would > > > > > > > > allow existing command lines to "work" using assumptions of what the user > > > > > > > > meant and strict=on users to get what they mean or an error saying that > > > > > > > > they asked for something that won't work or would require unreasonable > > > > > > > > assumptions. Why was this changed to an "expose" parameter? > > > > > > > Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] > > > > > > > of this series. > > > > > > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ > > > > > > > > > > > > > > And in the discussion, we hoped things would work like below with "strict" > > > > > > > parameter: > > > > > > > Users who want to describe cpu topology should provide cmdline like > > > > > > > > > > > > > > -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > > > > > and in this case we require an more accurate -smp configuration and > > > > > > > then generate the cpu topology description through ACPI/DT. > > > > > > > > > > > > > > While without a strict description, no cpu topology description would > > > > > > > be generated, so they get nothing through ACPI/DT. > > > > > > > > > > > > > > It seems to me that the "strict" parameter actually serves as a knob to > > > > > > > turn on/off the exposure of topology, and this is the reason I changed > > > > > > > the name. > > > > > > Yes, the use of 'strict=on' is no better than expose=on IMHO. > > > > > > > > > > > > If I give QEMU a cli > > > > > > > > > > > > -smp cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > > > then I expect that topology to be exposed to the guest. I shouldn't > > > > > > have to add extra flags to make that happen. > > > > > > > > > > > > Looking at the thread, it seems the concern was around the fact that > > > > > > the settings were not honoured historically and thus the CLI values > > > > > > could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 > > > > > This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong > > > > > configuration, and the parsing function already report error for this case. > > > > > > > > > > We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" > > > > > for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" > > > > > or "-smp 4, cores=1" are not acceptable any more because we are starting > > > > > to expose the topology. > > > > > > > > Incomplete specified topologies *are* acceptable. > > > > > > > > The smp_parse method will automatically fill in any missing values. > > > > > > > > ie, > > > > > > > > -smp 4,cores=1 > > > > -smp cores=1 > > > > -smp threads=1 > > > > -smp sockets=4 > > > > > > > > are all functionally identical to > > > > > > > > -smp 4,sockets=4,cores=1,dies=1,threads=1 > > > > > > > > > > > > The QEMU man page says this explicitly > > > > > > > > For the PC target, the number of cores per die, the > > > > number of threads per cores, the number of dies per packages and the > > > > total number of sockets can be specified. Missing values will be > > > > computed. If any on the three values is given, the total number of > > > > CPUs n can be omitted. > > > > > > It doesn't say how it will compute them though, which for the default > > > smp_parse and for x86 is to prefer sockets over cores over threads. > > > That's not necessarily what the user expects. IMO, we need a 'strict=on' > > > parameter that doesn't allow any collection of smp parameters which > > > require unreasonable assumptions. Reasonable assumptions are threads=1, > > > when threads is not specified and the rest of the math adds up. Also, > > > maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not > > > as reasonable to decide how to divide cores among sockets or to assume > > > threads=1 when only sockets and cores are given. How do we know the user > > > didn't forget to specify threads if we can't check the math? > > > > or just outlaw all invalid topologies incl. incomplete by default > > (without requiring extra option), and permit them only for old machine > > types ()using compat machinery) without topo info provided to guest. > > And maybe later deprecate invalid topologies altogether. > > This feels like it is creating pain for users to fix a problem that > isn't shown to actually be causing any common issues. > > We've supposed that users are having problems when forgetting to > specify "threads" and not having the compute value be desirable, > but where are the bug reports to back this up ? > > The partial topologies are valid and have well defined semantics. > Those semantics may not match everyone's preference, but that > doesn't make them invalid. > If we adopt the [undocumented] semantics of x86 for arm, then we may surprise some users that expect e.g. '-smp 16' to give them a single socket with 16 cores, because they'll start getting 16 sockets with 1 core each. That's because if we don't describe a topology to an arm linux guest then it assumes cores. Maybe we shouldn't worry about this, but I'd prefer we require explicit inputs from users and, if necessary, for them to explicitly opt-in to requiring those explicit inputs. Thanks, drew
On Tue, Jun 22, 2021 at 07:29:34PM +0200, Andrew Jones wrote: > On Tue, Jun 22, 2021 at 06:14:25PM +0100, Daniel P. Berrangé wrote: > > On Tue, Jun 22, 2021 at 05:40:13PM +0200, Igor Mammedov wrote: > > > On Tue, 22 Jun 2021 16:29:15 +0200 > > > Andrew Jones <drjones@redhat.com> wrote: > > > > > > > On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: > > > > > On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: > > > > > > Hi Daniel, > > > > > > > > > > > > On 2021/6/22 20:41, Daniel P. Berrangé wrote: > > > > > > > On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: > > > > > > > > > > > > > > > > On 2021/6/22 19:46, Andrew Jones wrote: > > > > > > > > > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: > > > > > > > > > > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > > > > > > > > > > > Hi, > > > > > > > > > > > > > > > > > > > > > > This is v4 of the series [1] that I posted to introduce support for > > > > > > > > > > > generating cpu topology descriptions to guest. Comments are welcome! > > > > > > > > > > > > > > > > > > > > > > Description: > > > > > > > > > > > Once the view of an accurate virtual cpu topology is provided to guest, > > > > > > > > > > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > > > > > > > > > > > e.g., the scheduling performance improvement. See Dario Faggioli's > > > > > > > > > > > research and the related performance tests in [2] for reference. So here > > > > > > > > > > > we go, this patch series introduces cpu topology support for ARM platform. > > > > > > > > > > > > > > > > > > > > > > In this series, instead of quietly enforcing the support for the latest > > > > > > > > > > > machine type, a new parameter "expose=on|off" in -smp command line is > > > > > > > > > > > introduced to leave QEMU users a choice to decide whether to enable the > > > > > > > > > > > feature or not. This will allow the feature to work on different machine > > > > > > > > > > > types and also ideally compat with already in-use -smp command lines. > > > > > > > > > > > Also we make much stricter requirement for the topology configuration > > > > > > > > > > > with "expose=on". > > > > > > > > > > Seeing this 'expose=on' parameter feels to me like we're adding a > > > > > > > > > > "make-it-work=yes" parameter. IMHO this is just something that should > > > > > > > > > > be done by default for the current machine type version and beyond. > > > > > > > > > > I don't see the need for a parameter to turnthis on, especially since > > > > > > > > > > it is being made architecture specific. > > > > > > > > > > > > > > > > > > > I agree. > > > > > > > > > > > > > > > > > > Yanan, we never discussed an "expose" parameter in the previous versions > > > > > > > > > of this series. We discussed a "strict" parameter though, which would > > > > > > > > > allow existing command lines to "work" using assumptions of what the user > > > > > > > > > meant and strict=on users to get what they mean or an error saying that > > > > > > > > > they asked for something that won't work or would require unreasonable > > > > > > > > > assumptions. Why was this changed to an "expose" parameter? > > > > > > > > Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] > > > > > > > > of this series. > > > > > > > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ > > > > > > > > > > > > > > > > And in the discussion, we hoped things would work like below with "strict" > > > > > > > > parameter: > > > > > > > > Users who want to describe cpu topology should provide cmdline like > > > > > > > > > > > > > > > > -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > > > > > > > and in this case we require an more accurate -smp configuration and > > > > > > > > then generate the cpu topology description through ACPI/DT. > > > > > > > > > > > > > > > > While without a strict description, no cpu topology description would > > > > > > > > be generated, so they get nothing through ACPI/DT. > > > > > > > > > > > > > > > > It seems to me that the "strict" parameter actually serves as a knob to > > > > > > > > turn on/off the exposure of topology, and this is the reason I changed > > > > > > > > the name. > > > > > > > Yes, the use of 'strict=on' is no better than expose=on IMHO. > > > > > > > > > > > > > > If I give QEMU a cli > > > > > > > > > > > > > > -smp cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > > > > > then I expect that topology to be exposed to the guest. I shouldn't > > > > > > > have to add extra flags to make that happen. > > > > > > > > > > > > > > Looking at the thread, it seems the concern was around the fact that > > > > > > > the settings were not honoured historically and thus the CLI values > > > > > > > could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 > > > > > > This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong > > > > > > configuration, and the parsing function already report error for this case. > > > > > > > > > > > > We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" > > > > > > for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" > > > > > > or "-smp 4, cores=1" are not acceptable any more because we are starting > > > > > > to expose the topology. > > > > > > > > > > Incomplete specified topologies *are* acceptable. > > > > > > > > > > The smp_parse method will automatically fill in any missing values. > > > > > > > > > > ie, > > > > > > > > > > -smp 4,cores=1 > > > > > -smp cores=1 > > > > > -smp threads=1 > > > > > -smp sockets=4 > > > > > > > > > > are all functionally identical to > > > > > > > > > > -smp 4,sockets=4,cores=1,dies=1,threads=1 > > > > > > > > > > > > > > > The QEMU man page says this explicitly > > > > > > > > > > For the PC target, the number of cores per die, the > > > > > number of threads per cores, the number of dies per packages and the > > > > > total number of sockets can be specified. Missing values will be > > > > > computed. If any on the three values is given, the total number of > > > > > CPUs n can be omitted. > > > > > > > > It doesn't say how it will compute them though, which for the default > > > > smp_parse and for x86 is to prefer sockets over cores over threads. > > > > That's not necessarily what the user expects. IMO, we need a 'strict=on' > > > > parameter that doesn't allow any collection of smp parameters which > > > > require unreasonable assumptions. Reasonable assumptions are threads=1, > > > > when threads is not specified and the rest of the math adds up. Also, > > > > maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not > > > > as reasonable to decide how to divide cores among sockets or to assume > > > > threads=1 when only sockets and cores are given. How do we know the user > > > > didn't forget to specify threads if we can't check the math? > > > > > > or just outlaw all invalid topologies incl. incomplete by default > > > (without requiring extra option), and permit them only for old machine > > > types ()using compat machinery) without topo info provided to guest. > > > And maybe later deprecate invalid topologies altogether. > > > > This feels like it is creating pain for users to fix a problem that > > isn't shown to actually be causing any common issues. > > > > We've supposed that users are having problems when forgetting to > > specify "threads" and not having the compute value be desirable, > > but where are the bug reports to back this up ? > > > > The partial topologies are valid and have well defined semantics. > > Those semantics may not match everyone's preference, but that > > doesn't make them invalid. > > > > If we adopt the [undocumented] semantics of x86 for arm, then we may > surprise some users that expect e.g. '-smp 16' to give them a single > socket with 16 cores, because they'll start getting 16 sockets with 1 > core each. That's because if we don't describe a topology to an arm linux > guest then it assumes cores. Maybe we shouldn't worry about this, but I'd > prefer we require explicit inputs from users and, if necessary, for them > to explicitly opt-in to requiring those explicit inputs. Even for x86, defaulting to maximising sockets over cores is sub-optimal. In real world x86 hardware it is very rare to have sockets > 2 or 4. For large CPU counts, you generally have large cores-per-socket counts on x86. The QEMU preference for sockets over cores on x86 (and PPC too IIUC) is a fairly arbitrary historical decision. It can cause problems with guest OS licensing because both Windows and RHEL have been known to charge differently for sockets vs cores, with high core counts being cheaper. We are not tied into the precise behaviour of the computed topology values, as we have no made any promises. All that's required is that we keep ABI compat for existing machine types. So we could decide to change the computed topology so that it prefers high core counts, over sockets, whem using new machine types only. That would seem to benefit all arches, by making QEMU more reflective of real world CPUs topology. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
Hi, On 2021/6/23 1:39, Daniel P. Berrangé wrote: > On Tue, Jun 22, 2021 at 07:29:34PM +0200, Andrew Jones wrote: >> On Tue, Jun 22, 2021 at 06:14:25PM +0100, Daniel P. Berrangé wrote: >>> On Tue, Jun 22, 2021 at 05:40:13PM +0200, Igor Mammedov wrote: >>>> On Tue, 22 Jun 2021 16:29:15 +0200 >>>> Andrew Jones <drjones@redhat.com> wrote: >>>> >>>>> On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: >>>>>> On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: >>>>>>> Hi Daniel, >>>>>>> >>>>>>> On 2021/6/22 20:41, Daniel P. Berrangé wrote: >>>>>>>> On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: >>>>>>>>> On 2021/6/22 19:46, Andrew Jones wrote: >>>>>>>>>> On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: >>>>>>>>>>> On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: >>>>>>>>>>>> Hi, >>>>>>>>>>>> >>>>>>>>>>>> This is v4 of the series [1] that I posted to introduce support for >>>>>>>>>>>> generating cpu topology descriptions to guest. Comments are welcome! >>>>>>>>>>>> >>>>>>>>>>>> Description: >>>>>>>>>>>> Once the view of an accurate virtual cpu topology is provided to guest, >>>>>>>>>>>> with a well-designed vCPU pinning to the pCPU we may get a huge benefit, >>>>>>>>>>>> e.g., the scheduling performance improvement. See Dario Faggioli's >>>>>>>>>>>> research and the related performance tests in [2] for reference. So here >>>>>>>>>>>> we go, this patch series introduces cpu topology support for ARM platform. >>>>>>>>>>>> >>>>>>>>>>>> In this series, instead of quietly enforcing the support for the latest >>>>>>>>>>>> machine type, a new parameter "expose=on|off" in -smp command line is >>>>>>>>>>>> introduced to leave QEMU users a choice to decide whether to enable the >>>>>>>>>>>> feature or not. This will allow the feature to work on different machine >>>>>>>>>>>> types and also ideally compat with already in-use -smp command lines. >>>>>>>>>>>> Also we make much stricter requirement for the topology configuration >>>>>>>>>>>> with "expose=on". >>>>>>>>>>> Seeing this 'expose=on' parameter feels to me like we're adding a >>>>>>>>>>> "make-it-work=yes" parameter. IMHO this is just something that should >>>>>>>>>>> be done by default for the current machine type version and beyond. >>>>>>>>>>> I don't see the need for a parameter to turnthis on, especially since >>>>>>>>>>> it is being made architecture specific. >>>>>>>>>>> >>>>>>>>>> I agree. >>>>>>>>>> >>>>>>>>>> Yanan, we never discussed an "expose" parameter in the previous versions >>>>>>>>>> of this series. We discussed a "strict" parameter though, which would >>>>>>>>>> allow existing command lines to "work" using assumptions of what the user >>>>>>>>>> meant and strict=on users to get what they mean or an error saying that >>>>>>>>>> they asked for something that won't work or would require unreasonable >>>>>>>>>> assumptions. Why was this changed to an "expose" parameter? >>>>>>>>> Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] >>>>>>>>> of this series. >>>>>>>>> [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ >>>>>>>>> >>>>>>>>> And in the discussion, we hoped things would work like below with "strict" >>>>>>>>> parameter: >>>>>>>>> Users who want to describe cpu topology should provide cmdline like >>>>>>>>> >>>>>>>>> -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 >>>>>>>>> >>>>>>>>> and in this case we require an more accurate -smp configuration and >>>>>>>>> then generate the cpu topology description through ACPI/DT. >>>>>>>>> >>>>>>>>> While without a strict description, no cpu topology description would >>>>>>>>> be generated, so they get nothing through ACPI/DT. >>>>>>>>> >>>>>>>>> It seems to me that the "strict" parameter actually serves as a knob to >>>>>>>>> turn on/off the exposure of topology, and this is the reason I changed >>>>>>>>> the name. >>>>>>>> Yes, the use of 'strict=on' is no better than expose=on IMHO. >>>>>>>> >>>>>>>> If I give QEMU a cli >>>>>>>> >>>>>>>> -smp cpus=4,sockets=2,cores=2,threads=1 >>>>>>>> >>>>>>>> then I expect that topology to be exposed to the guest. I shouldn't >>>>>>>> have to add extra flags to make that happen. >>>>>>>> >>>>>>>> Looking at the thread, it seems the concern was around the fact that >>>>>>>> the settings were not honoured historically and thus the CLI values >>>>>>>> could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 >>>>>>> This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong >>>>>>> configuration, and the parsing function already report error for this case. >>>>>>> >>>>>>> We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" >>>>>>> for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" >>>>>>> or "-smp 4, cores=1" are not acceptable any more because we are starting >>>>>>> to expose the topology. >>>>>> Incomplete specified topologies *are* acceptable. >>>>>> >>>>>> The smp_parse method will automatically fill in any missing values. >>>>>> >>>>>> ie, >>>>>> >>>>>> -smp 4,cores=1 >>>>>> -smp cores=1 >>>>>> -smp threads=1 >>>>>> -smp sockets=4 >>>>>> >>>>>> are all functionally identical to >>>>>> >>>>>> -smp 4,sockets=4,cores=1,dies=1,threads=1 >>>>>> >>>>>> >>>>>> The QEMU man page says this explicitly >>>>>> >>>>>> For the PC target, the number of cores per die, the >>>>>> number of threads per cores, the number of dies per packages and the >>>>>> total number of sockets can be specified. Missing values will be >>>>>> computed. If any on the three values is given, the total number of >>>>>> CPUs n can be omitted. >>>>> It doesn't say how it will compute them though, which for the default >>>>> smp_parse and for x86 is to prefer sockets over cores over threads. >>>>> That's not necessarily what the user expects. IMO, we need a 'strict=on' >>>>> parameter that doesn't allow any collection of smp parameters which >>>>> require unreasonable assumptions. Reasonable assumptions are threads=1, >>>>> when threads is not specified and the rest of the math adds up. Also, >>>>> maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not >>>>> as reasonable to decide how to divide cores among sockets or to assume >>>>> threads=1 when only sockets and cores are given. How do we know the user >>>>> didn't forget to specify threads if we can't check the math? >>>> or just outlaw all invalid topologies incl. incomplete by default >>>> (without requiring extra option), and permit them only for old machine >>>> types ()using compat machinery) without topo info provided to guest. >>>> And maybe later deprecate invalid topologies altogether. >>> This feels like it is creating pain for users to fix a problem that >>> isn't shown to actually be causing any common issues. >>> >>> We've supposed that users are having problems when forgetting to >>> specify "threads" and not having the compute value be desirable, >>> but where are the bug reports to back this up ? >>> >>> The partial topologies are valid and have well defined semantics. >>> Those semantics may not match everyone's preference, but that >>> doesn't make them invalid. >>> >> If we adopt the [undocumented] semantics of x86 for arm, then we may >> surprise some users that expect e.g. '-smp 16' to give them a single >> socket with 16 cores, because they'll start getting 16 sockets with 1 >> core each. That's because if we don't describe a topology to an arm linux >> guest then it assumes cores. Maybe we shouldn't worry about this, but I'd >> prefer we require explicit inputs from users and, if necessary, for them >> to explicitly opt-in to requiring those explicit inputs. > Even for x86, defaulting to maximising sockets over cores is sub-optimal. > In real world x86 hardware it is very rare to have sockets > 2 or 4. For > large CPU counts, you generally have large cores-per-socket counts on x86. > > The QEMU preference for sockets over cores on x86 (and PPC too IIUC) > is a fairly arbitrary historical decision. > > It can cause problems with guest OS licensing because both Windows > and RHEL have been known to charge differently for sockets vs cores, > with high core counts being cheaper. > > We are not tied into the precise behaviour of the computed topology > values, as we have no made any promises. All that's required is that > we keep ABI compat for existing machine types. If based on this point of view that we haven't made any promises for the precise behavior of the computed topology, things may get much easier. I have the following understanding (also a proposal): We will introduce the support for exposing cpu topology since machine type 6.2 and we will also describe the computed topology for the guest. We will not make any stricter parsing logic, however the -smp content in qemu-options.hx should be rearranged to clearly explain how the missing values will exactly be computed. And this is what QEMU is responsible for. We know that a well designed cpu topology configuration can gain much benefit for the guest, while a badly designed one will also probably cause negative impact. But the users should be responsible for the design of the -smp cmdlines. If they are using an incomplete cmdline for a 6.2 machine, then they should have known what the computed values will be and that the computed topology will be exposed to the guest. > > So we could decide to change the computed topology so that it prefers > high core counts, over sockets, whem using new machine types only. > That would seem to benefit all arches, by making QEMU more reflective > of real world CPUs topology. If we really decide to prefer cores over sockets over threads for new machine types, then I think we should also record this change in qemu-option.hx. Thanks, Yanan . > Regards, > Daniel
On Mon, Jun 28, 2021 at 04:43:05PM +0800, wangyanan (Y) wrote: > Hi, > On 2021/6/23 1:39, Daniel P. Berrangé wrote: > > On Tue, Jun 22, 2021 at 07:29:34PM +0200, Andrew Jones wrote: > > > On Tue, Jun 22, 2021 at 06:14:25PM +0100, Daniel P. Berrangé wrote: > > > > On Tue, Jun 22, 2021 at 05:40:13PM +0200, Igor Mammedov wrote: > > > > > On Tue, 22 Jun 2021 16:29:15 +0200 > > > > > Andrew Jones <drjones@redhat.com> wrote: > > > > > > > > > > > On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: > > > > > > > On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: > > > > > > > > Hi Daniel, > > > > > > > > > > > > > > > > On 2021/6/22 20:41, Daniel P. Berrangé wrote: > > > > > > > > > On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: > > > > > > > > > > On 2021/6/22 19:46, Andrew Jones wrote: > > > > > > > > > > > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: > > > > > > > > > > > > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > > > > > > > > > > > > > Hi, > > > > > > > > > > > > > > > > > > > > > > > > > > This is v4 of the series [1] that I posted to introduce support for > > > > > > > > > > > > > generating cpu topology descriptions to guest. Comments are welcome! > > > > > > > > > > > > > > > > > > > > > > > > > > Description: > > > > > > > > > > > > > Once the view of an accurate virtual cpu topology is provided to guest, > > > > > > > > > > > > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > > > > > > > > > > > > > e.g., the scheduling performance improvement. See Dario Faggioli's > > > > > > > > > > > > > research and the related performance tests in [2] for reference. So here > > > > > > > > > > > > > we go, this patch series introduces cpu topology support for ARM platform. > > > > > > > > > > > > > > > > > > > > > > > > > > In this series, instead of quietly enforcing the support for the latest > > > > > > > > > > > > > machine type, a new parameter "expose=on|off" in -smp command line is > > > > > > > > > > > > > introduced to leave QEMU users a choice to decide whether to enable the > > > > > > > > > > > > > feature or not. This will allow the feature to work on different machine > > > > > > > > > > > > > types and also ideally compat with already in-use -smp command lines. > > > > > > > > > > > > > Also we make much stricter requirement for the topology configuration > > > > > > > > > > > > > with "expose=on". > > > > > > > > > > > > Seeing this 'expose=on' parameter feels to me like we're adding a > > > > > > > > > > > > "make-it-work=yes" parameter. IMHO this is just something that should > > > > > > > > > > > > be done by default for the current machine type version and beyond. > > > > > > > > > > > > I don't see the need for a parameter to turnthis on, especially since > > > > > > > > > > > > it is being made architecture specific. > > > > > > > > > > > I agree. > > > > > > > > > > > > > > > > > > > > > > Yanan, we never discussed an "expose" parameter in the previous versions > > > > > > > > > > > of this series. We discussed a "strict" parameter though, which would > > > > > > > > > > > allow existing command lines to "work" using assumptions of what the user > > > > > > > > > > > meant and strict=on users to get what they mean or an error saying that > > > > > > > > > > > they asked for something that won't work or would require unreasonable > > > > > > > > > > > assumptions. Why was this changed to an "expose" parameter? > > > > > > > > > > Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] > > > > > > > > > > of this series. > > > > > > > > > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ > > > > > > > > > > > > > > > > > > > > And in the discussion, we hoped things would work like below with "strict" > > > > > > > > > > parameter: > > > > > > > > > > Users who want to describe cpu topology should provide cmdline like > > > > > > > > > > > > > > > > > > > > -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > > > > > > > > > > > and in this case we require an more accurate -smp configuration and > > > > > > > > > > then generate the cpu topology description through ACPI/DT. > > > > > > > > > > > > > > > > > > > > While without a strict description, no cpu topology description would > > > > > > > > > > be generated, so they get nothing through ACPI/DT. > > > > > > > > > > > > > > > > > > > > It seems to me that the "strict" parameter actually serves as a knob to > > > > > > > > > > turn on/off the exposure of topology, and this is the reason I changed > > > > > > > > > > the name. > > > > > > > > > Yes, the use of 'strict=on' is no better than expose=on IMHO. > > > > > > > > > > > > > > > > > > If I give QEMU a cli > > > > > > > > > > > > > > > > > > -smp cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > > > > > > > > > then I expect that topology to be exposed to the guest. I shouldn't > > > > > > > > > have to add extra flags to make that happen. > > > > > > > > > > > > > > > > > > Looking at the thread, it seems the concern was around the fact that > > > > > > > > > the settings were not honoured historically and thus the CLI values > > > > > > > > > could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 > > > > > > > > This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong > > > > > > > > configuration, and the parsing function already report error for this case. > > > > > > > > > > > > > > > > We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" > > > > > > > > for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" > > > > > > > > or "-smp 4, cores=1" are not acceptable any more because we are starting > > > > > > > > to expose the topology. > > > > > > > Incomplete specified topologies *are* acceptable. > > > > > > > > > > > > > > The smp_parse method will automatically fill in any missing values. > > > > > > > > > > > > > > ie, > > > > > > > > > > > > > > -smp 4,cores=1 > > > > > > > -smp cores=1 > > > > > > > -smp threads=1 > > > > > > > -smp sockets=4 > > > > > > > > > > > > > > are all functionally identical to > > > > > > > > > > > > > > -smp 4,sockets=4,cores=1,dies=1,threads=1 > > > > > > > > > > > > > > > > > > > > > The QEMU man page says this explicitly > > > > > > > > > > > > > > For the PC target, the number of cores per die, the > > > > > > > number of threads per cores, the number of dies per packages and the > > > > > > > total number of sockets can be specified. Missing values will be > > > > > > > computed. If any on the three values is given, the total number of > > > > > > > CPUs n can be omitted. > > > > > > It doesn't say how it will compute them though, which for the default > > > > > > smp_parse and for x86 is to prefer sockets over cores over threads. > > > > > > That's not necessarily what the user expects. IMO, we need a 'strict=on' > > > > > > parameter that doesn't allow any collection of smp parameters which > > > > > > require unreasonable assumptions. Reasonable assumptions are threads=1, > > > > > > when threads is not specified and the rest of the math adds up. Also, > > > > > > maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not > > > > > > as reasonable to decide how to divide cores among sockets or to assume > > > > > > threads=1 when only sockets and cores are given. How do we know the user > > > > > > didn't forget to specify threads if we can't check the math? > > > > > or just outlaw all invalid topologies incl. incomplete by default > > > > > (without requiring extra option), and permit them only for old machine > > > > > types ()using compat machinery) without topo info provided to guest. > > > > > And maybe later deprecate invalid topologies altogether. > > > > This feels like it is creating pain for users to fix a problem that > > > > isn't shown to actually be causing any common issues. > > > > > > > > We've supposed that users are having problems when forgetting to > > > > specify "threads" and not having the compute value be desirable, > > > > but where are the bug reports to back this up ? > > > > > > > > The partial topologies are valid and have well defined semantics. > > > > Those semantics may not match everyone's preference, but that > > > > doesn't make them invalid. > > > > > > > If we adopt the [undocumented] semantics of x86 for arm, then we may > > > surprise some users that expect e.g. '-smp 16' to give them a single > > > socket with 16 cores, because they'll start getting 16 sockets with 1 > > > core each. That's because if we don't describe a topology to an arm linux > > > guest then it assumes cores. Maybe we shouldn't worry about this, but I'd > > > prefer we require explicit inputs from users and, if necessary, for them > > > to explicitly opt-in to requiring those explicit inputs. > > Even for x86, defaulting to maximising sockets over cores is sub-optimal. > > In real world x86 hardware it is very rare to have sockets > 2 or 4. For > > large CPU counts, you generally have large cores-per-socket counts on x86. > > > > The QEMU preference for sockets over cores on x86 (and PPC too IIUC) > > is a fairly arbitrary historical decision. > > > > It can cause problems with guest OS licensing because both Windows > > and RHEL have been known to charge differently for sockets vs cores, > > with high core counts being cheaper. > > > > We are not tied into the precise behaviour of the computed topology > > values, as we have no made any promises. All that's required is that > > we keep ABI compat for existing machine types. > If based on this point of view that we haven't made any promises for the > precise behavior of the computed topology, things may get much easier. > I have the following understanding (also a proposal): > > We will introduce the support for exposing cpu topology since machine > type 6.2 and we will also describe the computed topology for the guest. > We will not make any stricter parsing logic, however the -smp content in > qemu-options.hx should be rearranged to clearly explain how the missing > values will exactly be computed. And this is what QEMU is responsible for. > > We know that a well designed cpu topology configuration can gain much > benefit for the guest, while a badly designed one will also probably cause > negative impact. But the users should be responsible for the design of the > -smp cmdlines. If they are using an incomplete cmdline for a 6.2 machine, > then they should have known what the computed values will be and that > the computed topology will be exposed to the guest. > > > > So we could decide to change the computed topology so that it prefers > > high core counts, over sockets, whem using new machine types only. > > That would seem to benefit all arches, by making QEMU more reflective > > of real world CPUs topology. > If we really decide to prefer cores over sockets over threads for new > machine > types, then I think we should also record this change in qemu-option.hx. > I agree. The proposal sounds good to me. I'd like to hear Eduardo's opinion too (CC'ed). Thanks, drew
On 2021/6/28 16:58, Andrew Jones wrote: > On Mon, Jun 28, 2021 at 04:43:05PM +0800, wangyanan (Y) wrote: >> Hi, >> On 2021/6/23 1:39, Daniel P. Berrangé wrote: >>> On Tue, Jun 22, 2021 at 07:29:34PM +0200, Andrew Jones wrote: >>>> On Tue, Jun 22, 2021 at 06:14:25PM +0100, Daniel P. Berrangé wrote: >>>>> On Tue, Jun 22, 2021 at 05:40:13PM +0200, Igor Mammedov wrote: >>>>>> On Tue, 22 Jun 2021 16:29:15 +0200 >>>>>> Andrew Jones <drjones@redhat.com> wrote: >>>>>> >>>>>>> On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: >>>>>>>> On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: >>>>>>>>> Hi Daniel, >>>>>>>>> >>>>>>>>> On 2021/6/22 20:41, Daniel P. Berrangé wrote: >>>>>>>>>> On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: >>>>>>>>>>> On 2021/6/22 19:46, Andrew Jones wrote: >>>>>>>>>>>> On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: >>>>>>>>>>>>> On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: >>>>>>>>>>>>>> Hi, >>>>>>>>>>>>>> >>>>>>>>>>>>>> This is v4 of the series [1] that I posted to introduce support for >>>>>>>>>>>>>> generating cpu topology descriptions to guest. Comments are welcome! >>>>>>>>>>>>>> >>>>>>>>>>>>>> Description: >>>>>>>>>>>>>> Once the view of an accurate virtual cpu topology is provided to guest, >>>>>>>>>>>>>> with a well-designed vCPU pinning to the pCPU we may get a huge benefit, >>>>>>>>>>>>>> e.g., the scheduling performance improvement. See Dario Faggioli's >>>>>>>>>>>>>> research and the related performance tests in [2] for reference. So here >>>>>>>>>>>>>> we go, this patch series introduces cpu topology support for ARM platform. >>>>>>>>>>>>>> >>>>>>>>>>>>>> In this series, instead of quietly enforcing the support for the latest >>>>>>>>>>>>>> machine type, a new parameter "expose=on|off" in -smp command line is >>>>>>>>>>>>>> introduced to leave QEMU users a choice to decide whether to enable the >>>>>>>>>>>>>> feature or not. This will allow the feature to work on different machine >>>>>>>>>>>>>> types and also ideally compat with already in-use -smp command lines. >>>>>>>>>>>>>> Also we make much stricter requirement for the topology configuration >>>>>>>>>>>>>> with "expose=on". >>>>>>>>>>>>> Seeing this 'expose=on' parameter feels to me like we're adding a >>>>>>>>>>>>> "make-it-work=yes" parameter. IMHO this is just something that should >>>>>>>>>>>>> be done by default for the current machine type version and beyond. >>>>>>>>>>>>> I don't see the need for a parameter to turnthis on, especially since >>>>>>>>>>>>> it is being made architecture specific. >>>>>>>>>>>> I agree. >>>>>>>>>>>> >>>>>>>>>>>> Yanan, we never discussed an "expose" parameter in the previous versions >>>>>>>>>>>> of this series. We discussed a "strict" parameter though, which would >>>>>>>>>>>> allow existing command lines to "work" using assumptions of what the user >>>>>>>>>>>> meant and strict=on users to get what they mean or an error saying that >>>>>>>>>>>> they asked for something that won't work or would require unreasonable >>>>>>>>>>>> assumptions. Why was this changed to an "expose" parameter? >>>>>>>>>>> Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] >>>>>>>>>>> of this series. >>>>>>>>>>> [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ >>>>>>>>>>> >>>>>>>>>>> And in the discussion, we hoped things would work like below with "strict" >>>>>>>>>>> parameter: >>>>>>>>>>> Users who want to describe cpu topology should provide cmdline like >>>>>>>>>>> >>>>>>>>>>> -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 >>>>>>>>>>> >>>>>>>>>>> and in this case we require an more accurate -smp configuration and >>>>>>>>>>> then generate the cpu topology description through ACPI/DT. >>>>>>>>>>> >>>>>>>>>>> While without a strict description, no cpu topology description would >>>>>>>>>>> be generated, so they get nothing through ACPI/DT. >>>>>>>>>>> >>>>>>>>>>> It seems to me that the "strict" parameter actually serves as a knob to >>>>>>>>>>> turn on/off the exposure of topology, and this is the reason I changed >>>>>>>>>>> the name. >>>>>>>>>> Yes, the use of 'strict=on' is no better than expose=on IMHO. >>>>>>>>>> >>>>>>>>>> If I give QEMU a cli >>>>>>>>>> >>>>>>>>>> -smp cpus=4,sockets=2,cores=2,threads=1 >>>>>>>>>> >>>>>>>>>> then I expect that topology to be exposed to the guest. I shouldn't >>>>>>>>>> have to add extra flags to make that happen. >>>>>>>>>> >>>>>>>>>> Looking at the thread, it seems the concern was around the fact that >>>>>>>>>> the settings were not honoured historically and thus the CLI values >>>>>>>>>> could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 >>>>>>>>> This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong >>>>>>>>> configuration, and the parsing function already report error for this case. >>>>>>>>> >>>>>>>>> We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" >>>>>>>>> for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" >>>>>>>>> or "-smp 4, cores=1" are not acceptable any more because we are starting >>>>>>>>> to expose the topology. >>>>>>>> Incomplete specified topologies *are* acceptable. >>>>>>>> >>>>>>>> The smp_parse method will automatically fill in any missing values. >>>>>>>> >>>>>>>> ie, >>>>>>>> >>>>>>>> -smp 4,cores=1 >>>>>>>> -smp cores=1 >>>>>>>> -smp threads=1 >>>>>>>> -smp sockets=4 >>>>>>>> >>>>>>>> are all functionally identical to >>>>>>>> >>>>>>>> -smp 4,sockets=4,cores=1,dies=1,threads=1 >>>>>>>> >>>>>>>> >>>>>>>> The QEMU man page says this explicitly >>>>>>>> >>>>>>>> For the PC target, the number of cores per die, the >>>>>>>> number of threads per cores, the number of dies per packages and the >>>>>>>> total number of sockets can be specified. Missing values will be >>>>>>>> computed. If any on the three values is given, the total number of >>>>>>>> CPUs n can be omitted. >>>>>>> It doesn't say how it will compute them though, which for the default >>>>>>> smp_parse and for x86 is to prefer sockets over cores over threads. >>>>>>> That's not necessarily what the user expects. IMO, we need a 'strict=on' >>>>>>> parameter that doesn't allow any collection of smp parameters which >>>>>>> require unreasonable assumptions. Reasonable assumptions are threads=1, >>>>>>> when threads is not specified and the rest of the math adds up. Also, >>>>>>> maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not >>>>>>> as reasonable to decide how to divide cores among sockets or to assume >>>>>>> threads=1 when only sockets and cores are given. How do we know the user >>>>>>> didn't forget to specify threads if we can't check the math? >>>>>> or just outlaw all invalid topologies incl. incomplete by default >>>>>> (without requiring extra option), and permit them only for old machine >>>>>> types ()using compat machinery) without topo info provided to guest. >>>>>> And maybe later deprecate invalid topologies altogether. >>>>> This feels like it is creating pain for users to fix a problem that >>>>> isn't shown to actually be causing any common issues. >>>>> >>>>> We've supposed that users are having problems when forgetting to >>>>> specify "threads" and not having the compute value be desirable, >>>>> but where are the bug reports to back this up ? >>>>> >>>>> The partial topologies are valid and have well defined semantics. >>>>> Those semantics may not match everyone's preference, but that >>>>> doesn't make them invalid. >>>>> >>>> If we adopt the [undocumented] semantics of x86 for arm, then we may >>>> surprise some users that expect e.g. '-smp 16' to give them a single >>>> socket with 16 cores, because they'll start getting 16 sockets with 1 >>>> core each. That's because if we don't describe a topology to an arm linux >>>> guest then it assumes cores. Maybe we shouldn't worry about this, but I'd >>>> prefer we require explicit inputs from users and, if necessary, for them >>>> to explicitly opt-in to requiring those explicit inputs. >>> Even for x86, defaulting to maximising sockets over cores is sub-optimal. >>> In real world x86 hardware it is very rare to have sockets > 2 or 4. For >>> large CPU counts, you generally have large cores-per-socket counts on x86. >>> >>> The QEMU preference for sockets over cores on x86 (and PPC too IIUC) >>> is a fairly arbitrary historical decision. >>> >>> It can cause problems with guest OS licensing because both Windows >>> and RHEL have been known to charge differently for sockets vs cores, >>> with high core counts being cheaper. >>> >>> We are not tied into the precise behaviour of the computed topology >>> values, as we have no made any promises. All that's required is that >>> we keep ABI compat for existing machine types. >> If based on this point of view that we haven't made any promises for the >> precise behavior of the computed topology, things may get much easier. >> I have the following understanding (also a proposal): >> >> We will introduce the support for exposing cpu topology since machine >> type 6.2 and we will also describe the computed topology for the guest. >> We will not make any stricter parsing logic, however the -smp content in >> qemu-options.hx should be rearranged to clearly explain how the missing >> values will exactly be computed. And this is what QEMU is responsible for. >> >> We know that a well designed cpu topology configuration can gain much >> benefit for the guest, while a badly designed one will also probably cause >> negative impact. But the users should be responsible for the design of the >> -smp cmdlines. If they are using an incomplete cmdline for a 6.2 machine, >> then they should have known what the computed values will be and that >> the computed topology will be exposed to the guest. >>> So we could decide to change the computed topology so that it prefers >>> high core counts, over sockets, whem using new machine types only. >>> That would seem to benefit all arches, by making QEMU more reflective >>> of real world CPUs topology. >> If we really decide to prefer cores over sockets over threads for new >> machine >> types, then I think we should also record this change in qemu-option.hx. >> > I agree. The proposal sounds good to me. I'd like to hear Eduardo's > opinion too (CC'ed). Thanks, Drew. I also hope that more people will reach an agreement on this issue, so that I can prepare v5 with confidence, although it will probably be late for 6.1. Thanks, Yanan . > Thanks, > drew > > > .
Hi Drew, Igor, I have a question below, hope for some explanation... :) I'm trying to rearrange the smp_parse() helper to make it more scalable. But I wonder why we are currently using maxcpus to calculate the missing sockets while using *cpus* to calculate the missing cores and threads? This makes the following cmdlines work fine, -smp cpus=8,maxcpus=12 <==> -smp cpus=8,maxcpus=12,sockets=12,cores=1,threads=1 -smp cpus=8,maxcpus=12,cores=6 <==> -smp cpus=8,maxcpus=12,sockets=2,cores=6,threads=1 -smp cpus=8,maxcpus=12,threads=2 <==> -smp cpus=8,maxcpus=12,sockets=6,cores=1,threads=2 but the following ones break the invalid CPU topology check: -smp cpus=8,maxcpus=12,sockets=2 <==> -smp cpus=8,maxcpus=12,sockets=2,cores=4,threads=1 -smp cpus=8,maxcpus=12,sockets=4,threads=1 <==> -smp cpus=8,maxcpus=12,sockets=4,cores=2,threads=1 -smp maxcpus=12 <==> -smp cpus=1,maxcpus=12,sockets=1,cores=1,threads=1 -smp maxcpus=12,sockets=2 <==> -smp cpus=2,maxcpus=12,sockets=2,cores=1,threads=1 IMO we should uniformly use maxcpus to calculate the missing sockets also cores and threads, which will allow all the above cmdlines work. Or maybe I missed something? I read the related discussion in [1] but didn't get an unambiguous conclusion. [1] https://patchwork.kernel.org/project/qemu-devel/patch/1535553121-80352-1-git-send-email-imammedo@redhat.com/ Regards, Yanan . On 2021/6/28 16:58, Andrew Jones wrote: > On Mon, Jun 28, 2021 at 04:43:05PM +0800, wangyanan (Y) wrote: >> Hi, >> On 2021/6/23 1:39, Daniel P. Berrangé wrote: >>> On Tue, Jun 22, 2021 at 07:29:34PM +0200, Andrew Jones wrote: >>>> On Tue, Jun 22, 2021 at 06:14:25PM +0100, Daniel P. Berrangé wrote: >>>>> On Tue, Jun 22, 2021 at 05:40:13PM +0200, Igor Mammedov wrote: >>>>>> On Tue, 22 Jun 2021 16:29:15 +0200 >>>>>> Andrew Jones <drjones@redhat.com> wrote: >>>>>> >>>>>>> On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: >>>>>>>> On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: >>>>>>>>> Hi Daniel, >>>>>>>>> >>>>>>>>> On 2021/6/22 20:41, Daniel P. Berrangé wrote: >>>>>>>>>> On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: >>>>>>>>>>> On 2021/6/22 19:46, Andrew Jones wrote: >>>>>>>>>>>> On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: >>>>>>>>>>>>> On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: >>>>>>>>>>>>>> Hi, >>>>>>>>>>>>>> >>>>>>>>>>>>>> This is v4 of the series [1] that I posted to introduce support for >>>>>>>>>>>>>> generating cpu topology descriptions to guest. Comments are welcome! >>>>>>>>>>>>>> >>>>>>>>>>>>>> Description: >>>>>>>>>>>>>> Once the view of an accurate virtual cpu topology is provided to guest, >>>>>>>>>>>>>> with a well-designed vCPU pinning to the pCPU we may get a huge benefit, >>>>>>>>>>>>>> e.g., the scheduling performance improvement. See Dario Faggioli's >>>>>>>>>>>>>> research and the related performance tests in [2] for reference. So here >>>>>>>>>>>>>> we go, this patch series introduces cpu topology support for ARM platform. >>>>>>>>>>>>>> >>>>>>>>>>>>>> In this series, instead of quietly enforcing the support for the latest >>>>>>>>>>>>>> machine type, a new parameter "expose=on|off" in -smp command line is >>>>>>>>>>>>>> introduced to leave QEMU users a choice to decide whether to enable the >>>>>>>>>>>>>> feature or not. This will allow the feature to work on different machine >>>>>>>>>>>>>> types and also ideally compat with already in-use -smp command lines. >>>>>>>>>>>>>> Also we make much stricter requirement for the topology configuration >>>>>>>>>>>>>> with "expose=on". >>>>>>>>>>>>> Seeing this 'expose=on' parameter feels to me like we're adding a >>>>>>>>>>>>> "make-it-work=yes" parameter. IMHO this is just something that should >>>>>>>>>>>>> be done by default for the current machine type version and beyond. >>>>>>>>>>>>> I don't see the need for a parameter to turnthis on, especially since >>>>>>>>>>>>> it is being made architecture specific. >>>>>>>>>>>> I agree. >>>>>>>>>>>> >>>>>>>>>>>> Yanan, we never discussed an "expose" parameter in the previous versions >>>>>>>>>>>> of this series. We discussed a "strict" parameter though, which would >>>>>>>>>>>> allow existing command lines to "work" using assumptions of what the user >>>>>>>>>>>> meant and strict=on users to get what they mean or an error saying that >>>>>>>>>>>> they asked for something that won't work or would require unreasonable >>>>>>>>>>>> assumptions. Why was this changed to an "expose" parameter? >>>>>>>>>>> Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] >>>>>>>>>>> of this series. >>>>>>>>>>> [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ >>>>>>>>>>> >>>>>>>>>>> And in the discussion, we hoped things would work like below with "strict" >>>>>>>>>>> parameter: >>>>>>>>>>> Users who want to describe cpu topology should provide cmdline like >>>>>>>>>>> >>>>>>>>>>> -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 >>>>>>>>>>> >>>>>>>>>>> and in this case we require an more accurate -smp configuration and >>>>>>>>>>> then generate the cpu topology description through ACPI/DT. >>>>>>>>>>> >>>>>>>>>>> While without a strict description, no cpu topology description would >>>>>>>>>>> be generated, so they get nothing through ACPI/DT. >>>>>>>>>>> >>>>>>>>>>> It seems to me that the "strict" parameter actually serves as a knob to >>>>>>>>>>> turn on/off the exposure of topology, and this is the reason I changed >>>>>>>>>>> the name. >>>>>>>>>> Yes, the use of 'strict=on' is no better than expose=on IMHO. >>>>>>>>>> >>>>>>>>>> If I give QEMU a cli >>>>>>>>>> >>>>>>>>>> -smp cpus=4,sockets=2,cores=2,threads=1 >>>>>>>>>> >>>>>>>>>> then I expect that topology to be exposed to the guest. I shouldn't >>>>>>>>>> have to add extra flags to make that happen. >>>>>>>>>> >>>>>>>>>> Looking at the thread, it seems the concern was around the fact that >>>>>>>>>> the settings were not honoured historically and thus the CLI values >>>>>>>>>> could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 >>>>>>>>> This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong >>>>>>>>> configuration, and the parsing function already report error for this case. >>>>>>>>> >>>>>>>>> We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" >>>>>>>>> for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" >>>>>>>>> or "-smp 4, cores=1" are not acceptable any more because we are starting >>>>>>>>> to expose the topology. >>>>>>>> Incomplete specified topologies *are* acceptable. >>>>>>>> >>>>>>>> The smp_parse method will automatically fill in any missing values. >>>>>>>> >>>>>>>> ie, >>>>>>>> >>>>>>>> -smp 4,cores=1 >>>>>>>> -smp cores=1 >>>>>>>> -smp threads=1 >>>>>>>> -smp sockets=4 >>>>>>>> >>>>>>>> are all functionally identical to >>>>>>>> >>>>>>>> -smp 4,sockets=4,cores=1,dies=1,threads=1 >>>>>>>> >>>>>>>> >>>>>>>> The QEMU man page says this explicitly >>>>>>>> >>>>>>>> For the PC target, the number of cores per die, the >>>>>>>> number of threads per cores, the number of dies per packages and the >>>>>>>> total number of sockets can be specified. Missing values will be >>>>>>>> computed. If any on the three values is given, the total number of >>>>>>>> CPUs n can be omitted. >>>>>>> It doesn't say how it will compute them though, which for the default >>>>>>> smp_parse and for x86 is to prefer sockets over cores over threads. >>>>>>> That's not necessarily what the user expects. IMO, we need a 'strict=on' >>>>>>> parameter that doesn't allow any collection of smp parameters which >>>>>>> require unreasonable assumptions. Reasonable assumptions are threads=1, >>>>>>> when threads is not specified and the rest of the math adds up. Also, >>>>>>> maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not >>>>>>> as reasonable to decide how to divide cores among sockets or to assume >>>>>>> threads=1 when only sockets and cores are given. How do we know the user >>>>>>> didn't forget to specify threads if we can't check the math? >>>>>> or just outlaw all invalid topologies incl. incomplete by default >>>>>> (without requiring extra option), and permit them only for old machine >>>>>> types ()using compat machinery) without topo info provided to guest. >>>>>> And maybe later deprecate invalid topologies altogether. >>>>> This feels like it is creating pain for users to fix a problem that >>>>> isn't shown to actually be causing any common issues. >>>>> >>>>> We've supposed that users are having problems when forgetting to >>>>> specify "threads" and not having the compute value be desirable, >>>>> but where are the bug reports to back this up ? >>>>> >>>>> The partial topologies are valid and have well defined semantics. >>>>> Those semantics may not match everyone's preference, but that >>>>> doesn't make them invalid. >>>>> >>>> If we adopt the [undocumented] semantics of x86 for arm, then we may >>>> surprise some users that expect e.g. '-smp 16' to give them a single >>>> socket with 16 cores, because they'll start getting 16 sockets with 1 >>>> core each. That's because if we don't describe a topology to an arm linux >>>> guest then it assumes cores. Maybe we shouldn't worry about this, but I'd >>>> prefer we require explicit inputs from users and, if necessary, for them >>>> to explicitly opt-in to requiring those explicit inputs. >>> Even for x86, defaulting to maximising sockets over cores is sub-optimal. >>> In real world x86 hardware it is very rare to have sockets > 2 or 4. For >>> large CPU counts, you generally have large cores-per-socket counts on x86. >>> >>> The QEMU preference for sockets over cores on x86 (and PPC too IIUC) >>> is a fairly arbitrary historical decision. >>> >>> It can cause problems with guest OS licensing because both Windows >>> and RHEL have been known to charge differently for sockets vs cores, >>> with high core counts being cheaper. >>> >>> We are not tied into the precise behaviour of the computed topology >>> values, as we have no made any promises. All that's required is that >>> we keep ABI compat for existing machine types. >> If based on this point of view that we haven't made any promises for the >> precise behavior of the computed topology, things may get much easier. >> I have the following understanding (also a proposal): >> >> We will introduce the support for exposing cpu topology since machine >> type 6.2 and we will also describe the computed topology for the guest. >> We will not make any stricter parsing logic, however the -smp content in >> qemu-options.hx should be rearranged to clearly explain how the missing >> values will exactly be computed. And this is what QEMU is responsible for. >> >> We know that a well designed cpu topology configuration can gain much >> benefit for the guest, while a badly designed one will also probably cause >> negative impact. But the users should be responsible for the design of the >> -smp cmdlines. If they are using an incomplete cmdline for a 6.2 machine, >> then they should have known what the computed values will be and that >> the computed topology will be exposed to the guest. >>> So we could decide to change the computed topology so that it prefers >>> high core counts, over sockets, whem using new machine types only. >>> That would seem to benefit all arches, by making QEMU more reflective >>> of real world CPUs topology. >> If we really decide to prefer cores over sockets over threads for new >> machine >> types, then I think we should also record this change in qemu-option.hx. >> > I agree. The proposal sounds good to me. I'd like to hear Eduardo's > opinion too (CC'ed). > > Thanks, > drew > > > .
On Wed, Jun 30, 2021 at 02:36:31PM +0800, wangyanan (Y) wrote: > Hi Drew, Igor, > > I have a question below, hope for some explanation... :) > > I'm trying to rearrange the smp_parse() helper to make it more scalable. > But I wonder why we are currently using maxcpus to calculate the missing > sockets while using *cpus* to calculate the missing cores and threads? > > This makes the following cmdlines work fine, > -smp cpus=8,maxcpus=12 <==> -smp > cpus=8,maxcpus=12,sockets=12,cores=1,threads=1 > -smp cpus=8,maxcpus=12,cores=6 <==> -smp > cpus=8,maxcpus=12,sockets=2,cores=6,threads=1 > -smp cpus=8,maxcpus=12,threads=2 <==> -smp > cpus=8,maxcpus=12,sockets=6,cores=1,threads=2 > > but the following ones break the invalid CPU topology check: > -smp cpus=8,maxcpus=12,sockets=2 <==> -smp > cpus=8,maxcpus=12,sockets=2,cores=4,threads=1 > -smp cpus=8,maxcpus=12,sockets=4,threads=1 <==> -smp > cpus=8,maxcpus=12,sockets=4,cores=2,threads=1 > -smp maxcpus=12 <==> -smp cpus=1,maxcpus=12,sockets=1,cores=1,threads=1 > -smp maxcpus=12,sockets=2 <==> -smp > cpus=2,maxcpus=12,sockets=2,cores=1,threads=1 > > IMO we should uniformly use maxcpus to calculate the missing sockets > also cores and threads, which will allow all the above cmdlines work. > Or maybe I missed something? I read the related discussion in [1] but > didn't get an unambiguous conclusion. > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/1535553121-80352-1-git-send-email-imammedo@redhat.com/ I agree that maxcpus should be used for all calculations. I think we need to write -smp parsing from scratch using a set of clean requirements and then use the machine compat stuff to switch to it. And also properly document it with something like "Since 6.2..." Thanks, drew > > Regards, > Yanan > . > > On 2021/6/28 16:58, Andrew Jones wrote: > > On Mon, Jun 28, 2021 at 04:43:05PM +0800, wangyanan (Y) wrote: > > > Hi, > > > On 2021/6/23 1:39, Daniel P. Berrangé wrote: > > > > On Tue, Jun 22, 2021 at 07:29:34PM +0200, Andrew Jones wrote: > > > > > On Tue, Jun 22, 2021 at 06:14:25PM +0100, Daniel P. Berrangé wrote: > > > > > > On Tue, Jun 22, 2021 at 05:40:13PM +0200, Igor Mammedov wrote: > > > > > > > On Tue, 22 Jun 2021 16:29:15 +0200 > > > > > > > Andrew Jones <drjones@redhat.com> wrote: > > > > > > > > > > > > > > > On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: > > > > > > > > > On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: > > > > > > > > > > Hi Daniel, > > > > > > > > > > > > > > > > > > > > On 2021/6/22 20:41, Daniel P. Berrangé wrote: > > > > > > > > > > > On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: > > > > > > > > > > > > On 2021/6/22 19:46, Andrew Jones wrote: > > > > > > > > > > > > > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: > > > > > > > > > > > > > > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > > > > > > > > > > > > > > > Hi, > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > This is v4 of the series [1] that I posted to introduce support for > > > > > > > > > > > > > > > generating cpu topology descriptions to guest. Comments are welcome! > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > Description: > > > > > > > > > > > > > > > Once the view of an accurate virtual cpu topology is provided to guest, > > > > > > > > > > > > > > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > > > > > > > > > > > > > > > e.g., the scheduling performance improvement. See Dario Faggioli's > > > > > > > > > > > > > > > research and the related performance tests in [2] for reference. So here > > > > > > > > > > > > > > > we go, this patch series introduces cpu topology support for ARM platform. > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > In this series, instead of quietly enforcing the support for the latest > > > > > > > > > > > > > > > machine type, a new parameter "expose=on|off" in -smp command line is > > > > > > > > > > > > > > > introduced to leave QEMU users a choice to decide whether to enable the > > > > > > > > > > > > > > > feature or not. This will allow the feature to work on different machine > > > > > > > > > > > > > > > types and also ideally compat with already in-use -smp command lines. > > > > > > > > > > > > > > > Also we make much stricter requirement for the topology configuration > > > > > > > > > > > > > > > with "expose=on". > > > > > > > > > > > > > > Seeing this 'expose=on' parameter feels to me like we're adding a > > > > > > > > > > > > > > "make-it-work=yes" parameter. IMHO this is just something that should > > > > > > > > > > > > > > be done by default for the current machine type version and beyond. > > > > > > > > > > > > > > I don't see the need for a parameter to turnthis on, especially since > > > > > > > > > > > > > > it is being made architecture specific. > > > > > > > > > > > > > I agree. > > > > > > > > > > > > > > > > > > > > > > > > > > Yanan, we never discussed an "expose" parameter in the previous versions > > > > > > > > > > > > > of this series. We discussed a "strict" parameter though, which would > > > > > > > > > > > > > allow existing command lines to "work" using assumptions of what the user > > > > > > > > > > > > > meant and strict=on users to get what they mean or an error saying that > > > > > > > > > > > > > they asked for something that won't work or would require unreasonable > > > > > > > > > > > > > assumptions. Why was this changed to an "expose" parameter? > > > > > > > > > > > > Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] > > > > > > > > > > > > of this series. > > > > > > > > > > > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ > > > > > > > > > > > > > > > > > > > > > > > > And in the discussion, we hoped things would work like below with "strict" > > > > > > > > > > > > parameter: > > > > > > > > > > > > Users who want to describe cpu topology should provide cmdline like > > > > > > > > > > > > > > > > > > > > > > > > -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > > > > > > > > > > > > > > > and in this case we require an more accurate -smp configuration and > > > > > > > > > > > > then generate the cpu topology description through ACPI/DT. > > > > > > > > > > > > > > > > > > > > > > > > While without a strict description, no cpu topology description would > > > > > > > > > > > > be generated, so they get nothing through ACPI/DT. > > > > > > > > > > > > > > > > > > > > > > > > It seems to me that the "strict" parameter actually serves as a knob to > > > > > > > > > > > > turn on/off the exposure of topology, and this is the reason I changed > > > > > > > > > > > > the name. > > > > > > > > > > > Yes, the use of 'strict=on' is no better than expose=on IMHO. > > > > > > > > > > > > > > > > > > > > > > If I give QEMU a cli > > > > > > > > > > > > > > > > > > > > > > -smp cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > > > > > > > > > > > > > then I expect that topology to be exposed to the guest. I shouldn't > > > > > > > > > > > have to add extra flags to make that happen. > > > > > > > > > > > > > > > > > > > > > > Looking at the thread, it seems the concern was around the fact that > > > > > > > > > > > the settings were not honoured historically and thus the CLI values > > > > > > > > > > > could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 > > > > > > > > > > This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong > > > > > > > > > > configuration, and the parsing function already report error for this case. > > > > > > > > > > > > > > > > > > > > We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" > > > > > > > > > > for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" > > > > > > > > > > or "-smp 4, cores=1" are not acceptable any more because we are starting > > > > > > > > > > to expose the topology. > > > > > > > > > Incomplete specified topologies *are* acceptable. > > > > > > > > > > > > > > > > > > The smp_parse method will automatically fill in any missing values. > > > > > > > > > > > > > > > > > > ie, > > > > > > > > > > > > > > > > > > -smp 4,cores=1 > > > > > > > > > -smp cores=1 > > > > > > > > > -smp threads=1 > > > > > > > > > -smp sockets=4 > > > > > > > > > > > > > > > > > > are all functionally identical to > > > > > > > > > > > > > > > > > > -smp 4,sockets=4,cores=1,dies=1,threads=1 > > > > > > > > > > > > > > > > > > > > > > > > > > > The QEMU man page says this explicitly > > > > > > > > > > > > > > > > > > For the PC target, the number of cores per die, the > > > > > > > > > number of threads per cores, the number of dies per packages and the > > > > > > > > > total number of sockets can be specified. Missing values will be > > > > > > > > > computed. If any on the three values is given, the total number of > > > > > > > > > CPUs n can be omitted. > > > > > > > > It doesn't say how it will compute them though, which for the default > > > > > > > > smp_parse and for x86 is to prefer sockets over cores over threads. > > > > > > > > That's not necessarily what the user expects. IMO, we need a 'strict=on' > > > > > > > > parameter that doesn't allow any collection of smp parameters which > > > > > > > > require unreasonable assumptions. Reasonable assumptions are threads=1, > > > > > > > > when threads is not specified and the rest of the math adds up. Also, > > > > > > > > maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not > > > > > > > > as reasonable to decide how to divide cores among sockets or to assume > > > > > > > > threads=1 when only sockets and cores are given. How do we know the user > > > > > > > > didn't forget to specify threads if we can't check the math? > > > > > > > or just outlaw all invalid topologies incl. incomplete by default > > > > > > > (without requiring extra option), and permit them only for old machine > > > > > > > types ()using compat machinery) without topo info provided to guest. > > > > > > > And maybe later deprecate invalid topologies altogether. > > > > > > This feels like it is creating pain for users to fix a problem that > > > > > > isn't shown to actually be causing any common issues. > > > > > > > > > > > > We've supposed that users are having problems when forgetting to > > > > > > specify "threads" and not having the compute value be desirable, > > > > > > but where are the bug reports to back this up ? > > > > > > > > > > > > The partial topologies are valid and have well defined semantics. > > > > > > Those semantics may not match everyone's preference, but that > > > > > > doesn't make them invalid. > > > > > > > > > > > If we adopt the [undocumented] semantics of x86 for arm, then we may > > > > > surprise some users that expect e.g. '-smp 16' to give them a single > > > > > socket with 16 cores, because they'll start getting 16 sockets with 1 > > > > > core each. That's because if we don't describe a topology to an arm linux > > > > > guest then it assumes cores. Maybe we shouldn't worry about this, but I'd > > > > > prefer we require explicit inputs from users and, if necessary, for them > > > > > to explicitly opt-in to requiring those explicit inputs. > > > > Even for x86, defaulting to maximising sockets over cores is sub-optimal. > > > > In real world x86 hardware it is very rare to have sockets > 2 or 4. For > > > > large CPU counts, you generally have large cores-per-socket counts on x86. > > > > > > > > The QEMU preference for sockets over cores on x86 (and PPC too IIUC) > > > > is a fairly arbitrary historical decision. > > > > > > > > It can cause problems with guest OS licensing because both Windows > > > > and RHEL have been known to charge differently for sockets vs cores, > > > > with high core counts being cheaper. > > > > > > > > We are not tied into the precise behaviour of the computed topology > > > > values, as we have no made any promises. All that's required is that > > > > we keep ABI compat for existing machine types. > > > If based on this point of view that we haven't made any promises for the > > > precise behavior of the computed topology, things may get much easier. > > > I have the following understanding (also a proposal): > > > > > > We will introduce the support for exposing cpu topology since machine > > > type 6.2 and we will also describe the computed topology for the guest. > > > We will not make any stricter parsing logic, however the -smp content in > > > qemu-options.hx should be rearranged to clearly explain how the missing > > > values will exactly be computed. And this is what QEMU is responsible for. > > > > > > We know that a well designed cpu topology configuration can gain much > > > benefit for the guest, while a badly designed one will also probably cause > > > negative impact. But the users should be responsible for the design of the > > > -smp cmdlines. If they are using an incomplete cmdline for a 6.2 machine, > > > then they should have known what the computed values will be and that > > > the computed topology will be exposed to the guest. > > > > So we could decide to change the computed topology so that it prefers > > > > high core counts, over sockets, whem using new machine types only. > > > > That would seem to benefit all arches, by making QEMU more reflective > > > > of real world CPUs topology. > > > If we really decide to prefer cores over sockets over threads for new > > > machine > > > types, then I think we should also record this change in qemu-option.hx. > > > > > I agree. The proposal sounds good to me. I'd like to hear Eduardo's > > opinion too (CC'ed). > > > > Thanks, > > drew > > > > > > . >
On 2021/6/30 16:30, Andrew Jones wrote: > On Wed, Jun 30, 2021 at 02:36:31PM +0800, wangyanan (Y) wrote: >> Hi Drew, Igor, >> >> I have a question below, hope for some explanation... :) >> >> I'm trying to rearrange the smp_parse() helper to make it more scalable. >> But I wonder why we are currently using maxcpus to calculate the missing >> sockets while using *cpus* to calculate the missing cores and threads? >> >> This makes the following cmdlines work fine, >> -smp cpus=8,maxcpus=12 <==> -smp >> cpus=8,maxcpus=12,sockets=12,cores=1,threads=1 >> -smp cpus=8,maxcpus=12,cores=6 <==> -smp >> cpus=8,maxcpus=12,sockets=2,cores=6,threads=1 >> -smp cpus=8,maxcpus=12,threads=2 <==> -smp >> cpus=8,maxcpus=12,sockets=6,cores=1,threads=2 >> >> but the following ones break the invalid CPU topology check: >> -smp cpus=8,maxcpus=12,sockets=2 <==> -smp >> cpus=8,maxcpus=12,sockets=2,cores=4,threads=1 >> -smp cpus=8,maxcpus=12,sockets=4,threads=1 <==> -smp >> cpus=8,maxcpus=12,sockets=4,cores=2,threads=1 >> -smp maxcpus=12 <==> -smp cpus=1,maxcpus=12,sockets=1,cores=1,threads=1 >> -smp maxcpus=12,sockets=2 <==> -smp >> cpus=2,maxcpus=12,sockets=2,cores=1,threads=1 >> >> IMO we should uniformly use maxcpus to calculate the missing sockets >> also cores and threads, which will allow all the above cmdlines work. >> Or maybe I missed something? I read the related discussion in [1] but >> didn't get an unambiguous conclusion. >> >> [1] https://patchwork.kernel.org/project/qemu-devel/patch/1535553121-80352-1-git-send-email-imammedo@redhat.com/ > I agree that maxcpus should be used for all calculations. Thanks. From my view uniformly using maxcpus to calculate the missing values won't break any existing working cmdlines, but will allow some now being invalid and incomplete cmdlines to be valid. I will use maxcpus and test the parser for all possible parameter collections. > I think we need > to write -smp parsing from scratch using a set of clean requirements and > then use the machine compat stuff to switch to it. And also properly > document it with something like "Since 6.2..." I agree to rewrite the -smp parsing. But what's the meaning of clean requirements? Sorry I didn't get it. Thanks, Yanan . > >> Regards, >> Yanan >> . >> >> On 2021/6/28 16:58, Andrew Jones wrote: >>> On Mon, Jun 28, 2021 at 04:43:05PM +0800, wangyanan (Y) wrote: >>>> Hi, >>>> On 2021/6/23 1:39, Daniel P. Berrangé wrote: >>>>> On Tue, Jun 22, 2021 at 07:29:34PM +0200, Andrew Jones wrote: >>>>>> On Tue, Jun 22, 2021 at 06:14:25PM +0100, Daniel P. Berrangé wrote: >>>>>>> On Tue, Jun 22, 2021 at 05:40:13PM +0200, Igor Mammedov wrote: >>>>>>>> On Tue, 22 Jun 2021 16:29:15 +0200 >>>>>>>> Andrew Jones <drjones@redhat.com> wrote: >>>>>>>> >>>>>>>>> On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: >>>>>>>>>> On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: >>>>>>>>>>> Hi Daniel, >>>>>>>>>>> >>>>>>>>>>> On 2021/6/22 20:41, Daniel P. Berrangé wrote: >>>>>>>>>>>> On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: >>>>>>>>>>>>> On 2021/6/22 19:46, Andrew Jones wrote: >>>>>>>>>>>>>> On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: >>>>>>>>>>>>>>> On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: >>>>>>>>>>>>>>>> Hi, >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> This is v4 of the series [1] that I posted to introduce support for >>>>>>>>>>>>>>>> generating cpu topology descriptions to guest. Comments are welcome! >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> Description: >>>>>>>>>>>>>>>> Once the view of an accurate virtual cpu topology is provided to guest, >>>>>>>>>>>>>>>> with a well-designed vCPU pinning to the pCPU we may get a huge benefit, >>>>>>>>>>>>>>>> e.g., the scheduling performance improvement. See Dario Faggioli's >>>>>>>>>>>>>>>> research and the related performance tests in [2] for reference. So here >>>>>>>>>>>>>>>> we go, this patch series introduces cpu topology support for ARM platform. >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> In this series, instead of quietly enforcing the support for the latest >>>>>>>>>>>>>>>> machine type, a new parameter "expose=on|off" in -smp command line is >>>>>>>>>>>>>>>> introduced to leave QEMU users a choice to decide whether to enable the >>>>>>>>>>>>>>>> feature or not. This will allow the feature to work on different machine >>>>>>>>>>>>>>>> types and also ideally compat with already in-use -smp command lines. >>>>>>>>>>>>>>>> Also we make much stricter requirement for the topology configuration >>>>>>>>>>>>>>>> with "expose=on". >>>>>>>>>>>>>>> Seeing this 'expose=on' parameter feels to me like we're adding a >>>>>>>>>>>>>>> "make-it-work=yes" parameter. IMHO this is just something that should >>>>>>>>>>>>>>> be done by default for the current machine type version and beyond. >>>>>>>>>>>>>>> I don't see the need for a parameter to turnthis on, especially since >>>>>>>>>>>>>>> it is being made architecture specific. >>>>>>>>>>>>>> I agree. >>>>>>>>>>>>>> >>>>>>>>>>>>>> Yanan, we never discussed an "expose" parameter in the previous versions >>>>>>>>>>>>>> of this series. We discussed a "strict" parameter though, which would >>>>>>>>>>>>>> allow existing command lines to "work" using assumptions of what the user >>>>>>>>>>>>>> meant and strict=on users to get what they mean or an error saying that >>>>>>>>>>>>>> they asked for something that won't work or would require unreasonable >>>>>>>>>>>>>> assumptions. Why was this changed to an "expose" parameter? >>>>>>>>>>>>> Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] >>>>>>>>>>>>> of this series. >>>>>>>>>>>>> [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ >>>>>>>>>>>>> >>>>>>>>>>>>> And in the discussion, we hoped things would work like below with "strict" >>>>>>>>>>>>> parameter: >>>>>>>>>>>>> Users who want to describe cpu topology should provide cmdline like >>>>>>>>>>>>> >>>>>>>>>>>>> -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 >>>>>>>>>>>>> >>>>>>>>>>>>> and in this case we require an more accurate -smp configuration and >>>>>>>>>>>>> then generate the cpu topology description through ACPI/DT. >>>>>>>>>>>>> >>>>>>>>>>>>> While without a strict description, no cpu topology description would >>>>>>>>>>>>> be generated, so they get nothing through ACPI/DT. >>>>>>>>>>>>> >>>>>>>>>>>>> It seems to me that the "strict" parameter actually serves as a knob to >>>>>>>>>>>>> turn on/off the exposure of topology, and this is the reason I changed >>>>>>>>>>>>> the name. >>>>>>>>>>>> Yes, the use of 'strict=on' is no better than expose=on IMHO. >>>>>>>>>>>> >>>>>>>>>>>> If I give QEMU a cli >>>>>>>>>>>> >>>>>>>>>>>> -smp cpus=4,sockets=2,cores=2,threads=1 >>>>>>>>>>>> >>>>>>>>>>>> then I expect that topology to be exposed to the guest. I shouldn't >>>>>>>>>>>> have to add extra flags to make that happen. >>>>>>>>>>>> >>>>>>>>>>>> Looking at the thread, it seems the concern was around the fact that >>>>>>>>>>>> the settings were not honoured historically and thus the CLI values >>>>>>>>>>>> could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 >>>>>>>>>>> This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong >>>>>>>>>>> configuration, and the parsing function already report error for this case. >>>>>>>>>>> >>>>>>>>>>> We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" >>>>>>>>>>> for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" >>>>>>>>>>> or "-smp 4, cores=1" are not acceptable any more because we are starting >>>>>>>>>>> to expose the topology. >>>>>>>>>> Incomplete specified topologies *are* acceptable. >>>>>>>>>> >>>>>>>>>> The smp_parse method will automatically fill in any missing values. >>>>>>>>>> >>>>>>>>>> ie, >>>>>>>>>> >>>>>>>>>> -smp 4,cores=1 >>>>>>>>>> -smp cores=1 >>>>>>>>>> -smp threads=1 >>>>>>>>>> -smp sockets=4 >>>>>>>>>> >>>>>>>>>> are all functionally identical to >>>>>>>>>> >>>>>>>>>> -smp 4,sockets=4,cores=1,dies=1,threads=1 >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> The QEMU man page says this explicitly >>>>>>>>>> >>>>>>>>>> For the PC target, the number of cores per die, the >>>>>>>>>> number of threads per cores, the number of dies per packages and the >>>>>>>>>> total number of sockets can be specified. Missing values will be >>>>>>>>>> computed. If any on the three values is given, the total number of >>>>>>>>>> CPUs n can be omitted. >>>>>>>>> It doesn't say how it will compute them though, which for the default >>>>>>>>> smp_parse and for x86 is to prefer sockets over cores over threads. >>>>>>>>> That's not necessarily what the user expects. IMO, we need a 'strict=on' >>>>>>>>> parameter that doesn't allow any collection of smp parameters which >>>>>>>>> require unreasonable assumptions. Reasonable assumptions are threads=1, >>>>>>>>> when threads is not specified and the rest of the math adds up. Also, >>>>>>>>> maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not >>>>>>>>> as reasonable to decide how to divide cores among sockets or to assume >>>>>>>>> threads=1 when only sockets and cores are given. How do we know the user >>>>>>>>> didn't forget to specify threads if we can't check the math? >>>>>>>> or just outlaw all invalid topologies incl. incomplete by default >>>>>>>> (without requiring extra option), and permit them only for old machine >>>>>>>> types ()using compat machinery) without topo info provided to guest. >>>>>>>> And maybe later deprecate invalid topologies altogether. >>>>>>> This feels like it is creating pain for users to fix a problem that >>>>>>> isn't shown to actually be causing any common issues. >>>>>>> >>>>>>> We've supposed that users are having problems when forgetting to >>>>>>> specify "threads" and not having the compute value be desirable, >>>>>>> but where are the bug reports to back this up ? >>>>>>> >>>>>>> The partial topologies are valid and have well defined semantics. >>>>>>> Those semantics may not match everyone's preference, but that >>>>>>> doesn't make them invalid. >>>>>>> >>>>>> If we adopt the [undocumented] semantics of x86 for arm, then we may >>>>>> surprise some users that expect e.g. '-smp 16' to give them a single >>>>>> socket with 16 cores, because they'll start getting 16 sockets with 1 >>>>>> core each. That's because if we don't describe a topology to an arm linux >>>>>> guest then it assumes cores. Maybe we shouldn't worry about this, but I'd >>>>>> prefer we require explicit inputs from users and, if necessary, for them >>>>>> to explicitly opt-in to requiring those explicit inputs. >>>>> Even for x86, defaulting to maximising sockets over cores is sub-optimal. >>>>> In real world x86 hardware it is very rare to have sockets > 2 or 4. For >>>>> large CPU counts, you generally have large cores-per-socket counts on x86. >>>>> >>>>> The QEMU preference for sockets over cores on x86 (and PPC too IIUC) >>>>> is a fairly arbitrary historical decision. >>>>> >>>>> It can cause problems with guest OS licensing because both Windows >>>>> and RHEL have been known to charge differently for sockets vs cores, >>>>> with high core counts being cheaper. >>>>> >>>>> We are not tied into the precise behaviour of the computed topology >>>>> values, as we have no made any promises. All that's required is that >>>>> we keep ABI compat for existing machine types. >>>> If based on this point of view that we haven't made any promises for the >>>> precise behavior of the computed topology, things may get much easier. >>>> I have the following understanding (also a proposal): >>>> >>>> We will introduce the support for exposing cpu topology since machine >>>> type 6.2 and we will also describe the computed topology for the guest. >>>> We will not make any stricter parsing logic, however the -smp content in >>>> qemu-options.hx should be rearranged to clearly explain how the missing >>>> values will exactly be computed. And this is what QEMU is responsible for. >>>> >>>> We know that a well designed cpu topology configuration can gain much >>>> benefit for the guest, while a badly designed one will also probably cause >>>> negative impact. But the users should be responsible for the design of the >>>> -smp cmdlines. If they are using an incomplete cmdline for a 6.2 machine, >>>> then they should have known what the computed values will be and that >>>> the computed topology will be exposed to the guest. >>>>> So we could decide to change the computed topology so that it prefers >>>>> high core counts, over sockets, whem using new machine types only. >>>>> That would seem to benefit all arches, by making QEMU more reflective >>>>> of real world CPUs topology. >>>> If we really decide to prefer cores over sockets over threads for new >>>> machine >>>> types, then I think we should also record this change in qemu-option.hx. >>>> >>> I agree. The proposal sounds good to me. I'd like to hear Eduardo's >>> opinion too (CC'ed). >>> >>> Thanks, >>> drew >>> >>> >>> . > > .
On Wed, Jun 30, 2021 at 05:37:42PM +0800, wangyanan (Y) wrote: > On 2021/6/30 16:30, Andrew Jones wrote: > > On Wed, Jun 30, 2021 at 02:36:31PM +0800, wangyanan (Y) wrote: > > > Hi Drew, Igor, > > > > > > I have a question below, hope for some explanation... :) > > > > > > I'm trying to rearrange the smp_parse() helper to make it more scalable. > > > But I wonder why we are currently using maxcpus to calculate the missing > > > sockets while using *cpus* to calculate the missing cores and threads? > > > > > > This makes the following cmdlines work fine, > > > -smp cpus=8,maxcpus=12 <==> -smp > > > cpus=8,maxcpus=12,sockets=12,cores=1,threads=1 > > > -smp cpus=8,maxcpus=12,cores=6 <==> -smp > > > cpus=8,maxcpus=12,sockets=2,cores=6,threads=1 > > > -smp cpus=8,maxcpus=12,threads=2 <==> -smp > > > cpus=8,maxcpus=12,sockets=6,cores=1,threads=2 > > > > > > but the following ones break the invalid CPU topology check: > > > -smp cpus=8,maxcpus=12,sockets=2 <==> -smp > > > cpus=8,maxcpus=12,sockets=2,cores=4,threads=1 > > > -smp cpus=8,maxcpus=12,sockets=4,threads=1 <==> -smp > > > cpus=8,maxcpus=12,sockets=4,cores=2,threads=1 > > > -smp maxcpus=12 <==> -smp cpus=1,maxcpus=12,sockets=1,cores=1,threads=1 > > > -smp maxcpus=12,sockets=2 <==> -smp > > > cpus=2,maxcpus=12,sockets=2,cores=1,threads=1 > > > > > > IMO we should uniformly use maxcpus to calculate the missing sockets > > > also cores and threads, which will allow all the above cmdlines work. > > > Or maybe I missed something? I read the related discussion in [1] but > > > didn't get an unambiguous conclusion. > > > > > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/1535553121-80352-1-git-send-email-imammedo@redhat.com/ > > I agree that maxcpus should be used for all calculations. > Thanks. From my view uniformly using maxcpus to calculate the missing > values won't break any existing working cmdlines, but will allow some now > being invalid and incomplete cmdlines to be valid. I will use maxcpus and > test the parser for all possible parameter collections. > > I think we need > > to write -smp parsing from scratch using a set of clean requirements and > > then use the machine compat stuff to switch to it. And also properly > > document it with something like "Since 6.2..." > I agree to rewrite the -smp parsing. But what's the meaning of clean > requirements? > Sorry I didn't get it. I think -smp evolved without all the details considered up front. Now that we've considered the details/requirements more completely, then let's apply our knowledge of them to an implementation that gets them all covered. Here's a quick list from the top of my head, there might be some missing - maxcpus should be used for computation of missing values - we should assume cores over sockets over threads - we should allow extending the topology with arch-specific members, such as dies, which will always default to 1 when not provided, rather than be computed - we should store the results in the smp machine properties Thanks, drew > > Thanks, > Yanan > . > > > > > Regards, > > > Yanan > > > . > > > > > > On 2021/6/28 16:58, Andrew Jones wrote: > > > > On Mon, Jun 28, 2021 at 04:43:05PM +0800, wangyanan (Y) wrote: > > > > > Hi, > > > > > On 2021/6/23 1:39, Daniel P. Berrangé wrote: > > > > > > On Tue, Jun 22, 2021 at 07:29:34PM +0200, Andrew Jones wrote: > > > > > > > On Tue, Jun 22, 2021 at 06:14:25PM +0100, Daniel P. Berrangé wrote: > > > > > > > > On Tue, Jun 22, 2021 at 05:40:13PM +0200, Igor Mammedov wrote: > > > > > > > > > On Tue, 22 Jun 2021 16:29:15 +0200 > > > > > > > > > Andrew Jones <drjones@redhat.com> wrote: > > > > > > > > > > > > > > > > > > > On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: > > > > > > > > > > > On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: > > > > > > > > > > > > Hi Daniel, > > > > > > > > > > > > > > > > > > > > > > > > On 2021/6/22 20:41, Daniel P. Berrangé wrote: > > > > > > > > > > > > > On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: > > > > > > > > > > > > > > On 2021/6/22 19:46, Andrew Jones wrote: > > > > > > > > > > > > > > > On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: > > > > > > > > > > > > > > > > On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: > > > > > > > > > > > > > > > > > Hi, > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > This is v4 of the series [1] that I posted to introduce support for > > > > > > > > > > > > > > > > > generating cpu topology descriptions to guest. Comments are welcome! > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > Description: > > > > > > > > > > > > > > > > > Once the view of an accurate virtual cpu topology is provided to guest, > > > > > > > > > > > > > > > > > with a well-designed vCPU pinning to the pCPU we may get a huge benefit, > > > > > > > > > > > > > > > > > e.g., the scheduling performance improvement. See Dario Faggioli's > > > > > > > > > > > > > > > > > research and the related performance tests in [2] for reference. So here > > > > > > > > > > > > > > > > > we go, this patch series introduces cpu topology support for ARM platform. > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > In this series, instead of quietly enforcing the support for the latest > > > > > > > > > > > > > > > > > machine type, a new parameter "expose=on|off" in -smp command line is > > > > > > > > > > > > > > > > > introduced to leave QEMU users a choice to decide whether to enable the > > > > > > > > > > > > > > > > > feature or not. This will allow the feature to work on different machine > > > > > > > > > > > > > > > > > types and also ideally compat with already in-use -smp command lines. > > > > > > > > > > > > > > > > > Also we make much stricter requirement for the topology configuration > > > > > > > > > > > > > > > > > with "expose=on". > > > > > > > > > > > > > > > > Seeing this 'expose=on' parameter feels to me like we're adding a > > > > > > > > > > > > > > > > "make-it-work=yes" parameter. IMHO this is just something that should > > > > > > > > > > > > > > > > be done by default for the current machine type version and beyond. > > > > > > > > > > > > > > > > I don't see the need for a parameter to turnthis on, especially since > > > > > > > > > > > > > > > > it is being made architecture specific. > > > > > > > > > > > > > > > I agree. > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > Yanan, we never discussed an "expose" parameter in the previous versions > > > > > > > > > > > > > > > of this series. We discussed a "strict" parameter though, which would > > > > > > > > > > > > > > > allow existing command lines to "work" using assumptions of what the user > > > > > > > > > > > > > > > meant and strict=on users to get what they mean or an error saying that > > > > > > > > > > > > > > > they asked for something that won't work or would require unreasonable > > > > > > > > > > > > > > > assumptions. Why was this changed to an "expose" parameter? > > > > > > > > > > > > > > Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] > > > > > > > > > > > > > > of this series. > > > > > > > > > > > > > > [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ > > > > > > > > > > > > > > > > > > > > > > > > > > > > And in the discussion, we hoped things would work like below with "strict" > > > > > > > > > > > > > > parameter: > > > > > > > > > > > > > > Users who want to describe cpu topology should provide cmdline like > > > > > > > > > > > > > > > > > > > > > > > > > > > > -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > > > > > > > > > > > > > > > > > > > and in this case we require an more accurate -smp configuration and > > > > > > > > > > > > > > then generate the cpu topology description through ACPI/DT. > > > > > > > > > > > > > > > > > > > > > > > > > > > > While without a strict description, no cpu topology description would > > > > > > > > > > > > > > be generated, so they get nothing through ACPI/DT. > > > > > > > > > > > > > > > > > > > > > > > > > > > > It seems to me that the "strict" parameter actually serves as a knob to > > > > > > > > > > > > > > turn on/off the exposure of topology, and this is the reason I changed > > > > > > > > > > > > > > the name. > > > > > > > > > > > > > Yes, the use of 'strict=on' is no better than expose=on IMHO. > > > > > > > > > > > > > > > > > > > > > > > > > > If I give QEMU a cli > > > > > > > > > > > > > > > > > > > > > > > > > > -smp cpus=4,sockets=2,cores=2,threads=1 > > > > > > > > > > > > > > > > > > > > > > > > > > then I expect that topology to be exposed to the guest. I shouldn't > > > > > > > > > > > > > have to add extra flags to make that happen. > > > > > > > > > > > > > > > > > > > > > > > > > > Looking at the thread, it seems the concern was around the fact that > > > > > > > > > > > > > the settings were not honoured historically and thus the CLI values > > > > > > > > > > > > > could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 > > > > > > > > > > > > This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong > > > > > > > > > > > > configuration, and the parsing function already report error for this case. > > > > > > > > > > > > > > > > > > > > > > > > We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" > > > > > > > > > > > > for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" > > > > > > > > > > > > or "-smp 4, cores=1" are not acceptable any more because we are starting > > > > > > > > > > > > to expose the topology. > > > > > > > > > > > Incomplete specified topologies *are* acceptable. > > > > > > > > > > > > > > > > > > > > > > The smp_parse method will automatically fill in any missing values. > > > > > > > > > > > > > > > > > > > > > > ie, > > > > > > > > > > > > > > > > > > > > > > -smp 4,cores=1 > > > > > > > > > > > -smp cores=1 > > > > > > > > > > > -smp threads=1 > > > > > > > > > > > -smp sockets=4 > > > > > > > > > > > > > > > > > > > > > > are all functionally identical to > > > > > > > > > > > > > > > > > > > > > > -smp 4,sockets=4,cores=1,dies=1,threads=1 > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > The QEMU man page says this explicitly > > > > > > > > > > > > > > > > > > > > > > For the PC target, the number of cores per die, the > > > > > > > > > > > number of threads per cores, the number of dies per packages and the > > > > > > > > > > > total number of sockets can be specified. Missing values will be > > > > > > > > > > > computed. If any on the three values is given, the total number of > > > > > > > > > > > CPUs n can be omitted. > > > > > > > > > > It doesn't say how it will compute them though, which for the default > > > > > > > > > > smp_parse and for x86 is to prefer sockets over cores over threads. > > > > > > > > > > That's not necessarily what the user expects. IMO, we need a 'strict=on' > > > > > > > > > > parameter that doesn't allow any collection of smp parameters which > > > > > > > > > > require unreasonable assumptions. Reasonable assumptions are threads=1, > > > > > > > > > > when threads is not specified and the rest of the math adds up. Also, > > > > > > > > > > maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not > > > > > > > > > > as reasonable to decide how to divide cores among sockets or to assume > > > > > > > > > > threads=1 when only sockets and cores are given. How do we know the user > > > > > > > > > > didn't forget to specify threads if we can't check the math? > > > > > > > > > or just outlaw all invalid topologies incl. incomplete by default > > > > > > > > > (without requiring extra option), and permit them only for old machine > > > > > > > > > types ()using compat machinery) without topo info provided to guest. > > > > > > > > > And maybe later deprecate invalid topologies altogether. > > > > > > > > This feels like it is creating pain for users to fix a problem that > > > > > > > > isn't shown to actually be causing any common issues. > > > > > > > > > > > > > > > > We've supposed that users are having problems when forgetting to > > > > > > > > specify "threads" and not having the compute value be desirable, > > > > > > > > but where are the bug reports to back this up ? > > > > > > > > > > > > > > > > The partial topologies are valid and have well defined semantics. > > > > > > > > Those semantics may not match everyone's preference, but that > > > > > > > > doesn't make them invalid. > > > > > > > > > > > > > > > If we adopt the [undocumented] semantics of x86 for arm, then we may > > > > > > > surprise some users that expect e.g. '-smp 16' to give them a single > > > > > > > socket with 16 cores, because they'll start getting 16 sockets with 1 > > > > > > > core each. That's because if we don't describe a topology to an arm linux > > > > > > > guest then it assumes cores. Maybe we shouldn't worry about this, but I'd > > > > > > > prefer we require explicit inputs from users and, if necessary, for them > > > > > > > to explicitly opt-in to requiring those explicit inputs. > > > > > > Even for x86, defaulting to maximising sockets over cores is sub-optimal. > > > > > > In real world x86 hardware it is very rare to have sockets > 2 or 4. For > > > > > > large CPU counts, you generally have large cores-per-socket counts on x86. > > > > > > > > > > > > The QEMU preference for sockets over cores on x86 (and PPC too IIUC) > > > > > > is a fairly arbitrary historical decision. > > > > > > > > > > > > It can cause problems with guest OS licensing because both Windows > > > > > > and RHEL have been known to charge differently for sockets vs cores, > > > > > > with high core counts being cheaper. > > > > > > > > > > > > We are not tied into the precise behaviour of the computed topology > > > > > > values, as we have no made any promises. All that's required is that > > > > > > we keep ABI compat for existing machine types. > > > > > If based on this point of view that we haven't made any promises for the > > > > > precise behavior of the computed topology, things may get much easier. > > > > > I have the following understanding (also a proposal): > > > > > > > > > > We will introduce the support for exposing cpu topology since machine > > > > > type 6.2 and we will also describe the computed topology for the guest. > > > > > We will not make any stricter parsing logic, however the -smp content in > > > > > qemu-options.hx should be rearranged to clearly explain how the missing > > > > > values will exactly be computed. And this is what QEMU is responsible for. > > > > > > > > > > We know that a well designed cpu topology configuration can gain much > > > > > benefit for the guest, while a badly designed one will also probably cause > > > > > negative impact. But the users should be responsible for the design of the > > > > > -smp cmdlines. If they are using an incomplete cmdline for a 6.2 machine, > > > > > then they should have known what the computed values will be and that > > > > > the computed topology will be exposed to the guest. > > > > > > So we could decide to change the computed topology so that it prefers > > > > > > high core counts, over sockets, whem using new machine types only. > > > > > > That would seem to benefit all arches, by making QEMU more reflective > > > > > > of real world CPUs topology. > > > > > If we really decide to prefer cores over sockets over threads for new > > > > > machine > > > > > types, then I think we should also record this change in qemu-option.hx. > > > > > > > > > I agree. The proposal sounds good to me. I'd like to hear Eduardo's > > > > opinion too (CC'ed). > > > > > > > > Thanks, > > > > drew > > > > > > > > > > > > . > > > > . >
On 2021/6/30 19:56, Andrew Jones wrote: > On Wed, Jun 30, 2021 at 05:37:42PM +0800, wangyanan (Y) wrote: >> On 2021/6/30 16:30, Andrew Jones wrote: >>> On Wed, Jun 30, 2021 at 02:36:31PM +0800, wangyanan (Y) wrote: >>>> Hi Drew, Igor, >>>> >>>> I have a question below, hope for some explanation... :) >>>> >>>> I'm trying to rearrange the smp_parse() helper to make it more scalable. >>>> But I wonder why we are currently using maxcpus to calculate the missing >>>> sockets while using *cpus* to calculate the missing cores and threads? >>>> >>>> This makes the following cmdlines work fine, >>>> -smp cpus=8,maxcpus=12 <==> -smp >>>> cpus=8,maxcpus=12,sockets=12,cores=1,threads=1 >>>> -smp cpus=8,maxcpus=12,cores=6 <==> -smp >>>> cpus=8,maxcpus=12,sockets=2,cores=6,threads=1 >>>> -smp cpus=8,maxcpus=12,threads=2 <==> -smp >>>> cpus=8,maxcpus=12,sockets=6,cores=1,threads=2 >>>> >>>> but the following ones break the invalid CPU topology check: >>>> -smp cpus=8,maxcpus=12,sockets=2 <==> -smp >>>> cpus=8,maxcpus=12,sockets=2,cores=4,threads=1 >>>> -smp cpus=8,maxcpus=12,sockets=4,threads=1 <==> -smp >>>> cpus=8,maxcpus=12,sockets=4,cores=2,threads=1 >>>> -smp maxcpus=12 <==> -smp cpus=1,maxcpus=12,sockets=1,cores=1,threads=1 >>>> -smp maxcpus=12,sockets=2 <==> -smp >>>> cpus=2,maxcpus=12,sockets=2,cores=1,threads=1 >>>> >>>> IMO we should uniformly use maxcpus to calculate the missing sockets >>>> also cores and threads, which will allow all the above cmdlines work. >>>> Or maybe I missed something? I read the related discussion in [1] but >>>> didn't get an unambiguous conclusion. >>>> >>>> [1] https://patchwork.kernel.org/project/qemu-devel/patch/1535553121-80352-1-git-send-email-imammedo@redhat.com/ >>> I agree that maxcpus should be used for all calculations. >> Thanks. From my view uniformly using maxcpus to calculate the missing >> values won't break any existing working cmdlines, but will allow some now >> being invalid and incomplete cmdlines to be valid. I will use maxcpus and >> test the parser for all possible parameter collections. >>> I think we need >>> to write -smp parsing from scratch using a set of clean requirements and >>> then use the machine compat stuff to switch to it. And also properly >>> document it with something like "Since 6.2..." >> I agree to rewrite the -smp parsing. But what's the meaning of clean >> requirements? >> Sorry I didn't get it. > I think -smp evolved without all the details considered up front. Now that > we've considered the details/requirements more completely, then let's > apply our knowledge of them to an implementation that gets them all > covered. Got it now. > Here's a quick list from the top of my head, there might be > some missing > > - maxcpus should be used for computation of missing values > - we should assume cores over sockets over threads > - we should allow extending the topology with arch-specific > members, such as dies, which will always default to 1 when > not provided, rather than be computed > - we should store the results in the smp machine properties Right! This is a good summary of what we have discussed recently. Thanks, Yanan . > >> Thanks, >> Yanan >> . >>>> Regards, >>>> Yanan >>>> . >>>> >>>> On 2021/6/28 16:58, Andrew Jones wrote: >>>>> On Mon, Jun 28, 2021 at 04:43:05PM +0800, wangyanan (Y) wrote: >>>>>> Hi, >>>>>> On 2021/6/23 1:39, Daniel P. Berrangé wrote: >>>>>>> On Tue, Jun 22, 2021 at 07:29:34PM +0200, Andrew Jones wrote: >>>>>>>> On Tue, Jun 22, 2021 at 06:14:25PM +0100, Daniel P. Berrangé wrote: >>>>>>>>> On Tue, Jun 22, 2021 at 05:40:13PM +0200, Igor Mammedov wrote: >>>>>>>>>> On Tue, 22 Jun 2021 16:29:15 +0200 >>>>>>>>>> Andrew Jones <drjones@redhat.com> wrote: >>>>>>>>>> >>>>>>>>>>> On Tue, Jun 22, 2021 at 03:10:57PM +0100, Daniel P. Berrangé wrote: >>>>>>>>>>>> On Tue, Jun 22, 2021 at 10:04:52PM +0800, wangyanan (Y) wrote: >>>>>>>>>>>>> Hi Daniel, >>>>>>>>>>>>> >>>>>>>>>>>>> On 2021/6/22 20:41, Daniel P. Berrangé wrote: >>>>>>>>>>>>>> On Tue, Jun 22, 2021 at 08:31:22PM +0800, wangyanan (Y) wrote: >>>>>>>>>>>>>>> On 2021/6/22 19:46, Andrew Jones wrote: >>>>>>>>>>>>>>>> On Tue, Jun 22, 2021 at 11:18:09AM +0100, Daniel P. Berrangé wrote: >>>>>>>>>>>>>>>>> On Tue, Jun 22, 2021 at 05:34:06PM +0800, Yanan Wang wrote: >>>>>>>>>>>>>>>>>> Hi, >>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>> This is v4 of the series [1] that I posted to introduce support for >>>>>>>>>>>>>>>>>> generating cpu topology descriptions to guest. Comments are welcome! >>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>> Description: >>>>>>>>>>>>>>>>>> Once the view of an accurate virtual cpu topology is provided to guest, >>>>>>>>>>>>>>>>>> with a well-designed vCPU pinning to the pCPU we may get a huge benefit, >>>>>>>>>>>>>>>>>> e.g., the scheduling performance improvement. See Dario Faggioli's >>>>>>>>>>>>>>>>>> research and the related performance tests in [2] for reference. So here >>>>>>>>>>>>>>>>>> we go, this patch series introduces cpu topology support for ARM platform. >>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>> In this series, instead of quietly enforcing the support for the latest >>>>>>>>>>>>>>>>>> machine type, a new parameter "expose=on|off" in -smp command line is >>>>>>>>>>>>>>>>>> introduced to leave QEMU users a choice to decide whether to enable the >>>>>>>>>>>>>>>>>> feature or not. This will allow the feature to work on different machine >>>>>>>>>>>>>>>>>> types and also ideally compat with already in-use -smp command lines. >>>>>>>>>>>>>>>>>> Also we make much stricter requirement for the topology configuration >>>>>>>>>>>>>>>>>> with "expose=on". >>>>>>>>>>>>>>>>> Seeing this 'expose=on' parameter feels to me like we're adding a >>>>>>>>>>>>>>>>> "make-it-work=yes" parameter. IMHO this is just something that should >>>>>>>>>>>>>>>>> be done by default for the current machine type version and beyond. >>>>>>>>>>>>>>>>> I don't see the need for a parameter to turnthis on, especially since >>>>>>>>>>>>>>>>> it is being made architecture specific. >>>>>>>>>>>>>>>> I agree. >>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>> Yanan, we never discussed an "expose" parameter in the previous versions >>>>>>>>>>>>>>>> of this series. We discussed a "strict" parameter though, which would >>>>>>>>>>>>>>>> allow existing command lines to "work" using assumptions of what the user >>>>>>>>>>>>>>>> meant and strict=on users to get what they mean or an error saying that >>>>>>>>>>>>>>>> they asked for something that won't work or would require unreasonable >>>>>>>>>>>>>>>> assumptions. Why was this changed to an "expose" parameter? >>>>>>>>>>>>>>> Yes, we indeed discuss a new "strict" parameter but not a "expose" in v2 [1] >>>>>>>>>>>>>>> of this series. >>>>>>>>>>>>>>> [1] https://patchwork.kernel.org/project/qemu-devel/patch/20210413080745.33004-6-wangyanan55@huawei.com/ >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> And in the discussion, we hoped things would work like below with "strict" >>>>>>>>>>>>>>> parameter: >>>>>>>>>>>>>>> Users who want to describe cpu topology should provide cmdline like >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> -smp strict=on,cpus=4,sockets=2,cores=2,threads=1 >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> and in this case we require an more accurate -smp configuration and >>>>>>>>>>>>>>> then generate the cpu topology description through ACPI/DT. >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> While without a strict description, no cpu topology description would >>>>>>>>>>>>>>> be generated, so they get nothing through ACPI/DT. >>>>>>>>>>>>>>> >>>>>>>>>>>>>>> It seems to me that the "strict" parameter actually serves as a knob to >>>>>>>>>>>>>>> turn on/off the exposure of topology, and this is the reason I changed >>>>>>>>>>>>>>> the name. >>>>>>>>>>>>>> Yes, the use of 'strict=on' is no better than expose=on IMHO. >>>>>>>>>>>>>> >>>>>>>>>>>>>> If I give QEMU a cli >>>>>>>>>>>>>> >>>>>>>>>>>>>> -smp cpus=4,sockets=2,cores=2,threads=1 >>>>>>>>>>>>>> >>>>>>>>>>>>>> then I expect that topology to be exposed to the guest. I shouldn't >>>>>>>>>>>>>> have to add extra flags to make that happen. >>>>>>>>>>>>>> >>>>>>>>>>>>>> Looking at the thread, it seems the concern was around the fact that >>>>>>>>>>>>>> the settings were not honoured historically and thus the CLI values >>>>>>>>>>>>>> could be garbage. ie -smp cpus=4,sockets=8,cores=3,thread=9 >>>>>>>>>>>>> This "-smp cpus=4,sockets=8,cores=3,threads=9" behaviors as a wrong >>>>>>>>>>>>> configuration, and the parsing function already report error for this case. >>>>>>>>>>>>> >>>>>>>>>>>>> We hope more complete config like "-smp 4,sockets=2,cores=2,threads=1" >>>>>>>>>>>>> for exposure of topology, and the incomplete ones like "-smp 4,sockets=1" >>>>>>>>>>>>> or "-smp 4, cores=1" are not acceptable any more because we are starting >>>>>>>>>>>>> to expose the topology. >>>>>>>>>>>> Incomplete specified topologies *are* acceptable. >>>>>>>>>>>> >>>>>>>>>>>> The smp_parse method will automatically fill in any missing values. >>>>>>>>>>>> >>>>>>>>>>>> ie, >>>>>>>>>>>> >>>>>>>>>>>> -smp 4,cores=1 >>>>>>>>>>>> -smp cores=1 >>>>>>>>>>>> -smp threads=1 >>>>>>>>>>>> -smp sockets=4 >>>>>>>>>>>> >>>>>>>>>>>> are all functionally identical to >>>>>>>>>>>> >>>>>>>>>>>> -smp 4,sockets=4,cores=1,dies=1,threads=1 >>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>>> The QEMU man page says this explicitly >>>>>>>>>>>> >>>>>>>>>>>> For the PC target, the number of cores per die, the >>>>>>>>>>>> number of threads per cores, the number of dies per packages and the >>>>>>>>>>>> total number of sockets can be specified. Missing values will be >>>>>>>>>>>> computed. If any on the three values is given, the total number of >>>>>>>>>>>> CPUs n can be omitted. >>>>>>>>>>> It doesn't say how it will compute them though, which for the default >>>>>>>>>>> smp_parse and for x86 is to prefer sockets over cores over threads. >>>>>>>>>>> That's not necessarily what the user expects. IMO, we need a 'strict=on' >>>>>>>>>>> parameter that doesn't allow any collection of smp parameters which >>>>>>>>>>> require unreasonable assumptions. Reasonable assumptions are threads=1, >>>>>>>>>>> when threads is not specified and the rest of the math adds up. Also, >>>>>>>>>>> maxcpus == cpus when maxcpus isn't specified is reasonable. But, it's not >>>>>>>>>>> as reasonable to decide how to divide cores among sockets or to assume >>>>>>>>>>> threads=1 when only sockets and cores are given. How do we know the user >>>>>>>>>>> didn't forget to specify threads if we can't check the math? >>>>>>>>>> or just outlaw all invalid topologies incl. incomplete by default >>>>>>>>>> (without requiring extra option), and permit them only for old machine >>>>>>>>>> types ()using compat machinery) without topo info provided to guest. >>>>>>>>>> And maybe later deprecate invalid topologies altogether. >>>>>>>>> This feels like it is creating pain for users to fix a problem that >>>>>>>>> isn't shown to actually be causing any common issues. >>>>>>>>> >>>>>>>>> We've supposed that users are having problems when forgetting to >>>>>>>>> specify "threads" and not having the compute value be desirable, >>>>>>>>> but where are the bug reports to back this up ? >>>>>>>>> >>>>>>>>> The partial topologies are valid and have well defined semantics. >>>>>>>>> Those semantics may not match everyone's preference, but that >>>>>>>>> doesn't make them invalid. >>>>>>>>> >>>>>>>> If we adopt the [undocumented] semantics of x86 for arm, then we may >>>>>>>> surprise some users that expect e.g. '-smp 16' to give them a single >>>>>>>> socket with 16 cores, because they'll start getting 16 sockets with 1 >>>>>>>> core each. That's because if we don't describe a topology to an arm linux >>>>>>>> guest then it assumes cores. Maybe we shouldn't worry about this, but I'd >>>>>>>> prefer we require explicit inputs from users and, if necessary, for them >>>>>>>> to explicitly opt-in to requiring those explicit inputs. >>>>>>> Even for x86, defaulting to maximising sockets over cores is sub-optimal. >>>>>>> In real world x86 hardware it is very rare to have sockets > 2 or 4. For >>>>>>> large CPU counts, you generally have large cores-per-socket counts on x86. >>>>>>> >>>>>>> The QEMU preference for sockets over cores on x86 (and PPC too IIUC) >>>>>>> is a fairly arbitrary historical decision. >>>>>>> >>>>>>> It can cause problems with guest OS licensing because both Windows >>>>>>> and RHEL have been known to charge differently for sockets vs cores, >>>>>>> with high core counts being cheaper. >>>>>>> >>>>>>> We are not tied into the precise behaviour of the computed topology >>>>>>> values, as we have no made any promises. All that's required is that >>>>>>> we keep ABI compat for existing machine types. >>>>>> If based on this point of view that we haven't made any promises for the >>>>>> precise behavior of the computed topology, things may get much easier. >>>>>> I have the following understanding (also a proposal): >>>>>> >>>>>> We will introduce the support for exposing cpu topology since machine >>>>>> type 6.2 and we will also describe the computed topology for the guest. >>>>>> We will not make any stricter parsing logic, however the -smp content in >>>>>> qemu-options.hx should be rearranged to clearly explain how the missing >>>>>> values will exactly be computed. And this is what QEMU is responsible for. >>>>>> >>>>>> We know that a well designed cpu topology configuration can gain much >>>>>> benefit for the guest, while a badly designed one will also probably cause >>>>>> negative impact. But the users should be responsible for the design of the >>>>>> -smp cmdlines. If they are using an incomplete cmdline for a 6.2 machine, >>>>>> then they should have known what the computed values will be and that >>>>>> the computed topology will be exposed to the guest. >>>>>>> So we could decide to change the computed topology so that it prefers >>>>>>> high core counts, over sockets, whem using new machine types only. >>>>>>> That would seem to benefit all arches, by making QEMU more reflective >>>>>>> of real world CPUs topology. >>>>>> If we really decide to prefer cores over sockets over threads for new >>>>>> machine >>>>>> types, then I think we should also record this change in qemu-option.hx. >>>>>> >>>>> I agree. The proposal sounds good to me. I'd like to hear Eduardo's >>>>> opinion too (CC'ed). >>>>> >>>>> Thanks, >>>>> drew >>>>> >>>>> >>>>> . >>> . > > .
© 2016 - 2024 Red Hat, Inc.