| Message ID | 20230425161456.21031-3-pmorel@linux.ibm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | s390x: CPU Topology | expand |
On 25/04/2023 18.14, Pierre Morel wrote: > The topology information are attributes of the CPU and are > specified during the CPU device creation. > > On hot plug we: > - calculate the default values for the topology for drawers, > books and sockets in the case they are not specified. > - verify the CPU attributes > - check that we have still room on the desired socket > > The possibility to insert a CPU in a mask is dependent on the > number of cores allowed in a socket, a book or a drawer, the > checking is done during the hot plug of the CPU to have an > immediate answer. > > If the complete topology is not specified, the core is added > in the physical topology based on its core ID and it gets > defaults values for the modifier attributes. > > This way, starting QEMU without specifying the topology can > still get some advantage of the CPU topology. > > Signed-off-by: Pierre Morel <pmorel@linux.ibm.com> > --- ... > diff --git a/hw/s390x/cpu-topology.c b/hw/s390x/cpu-topology.c > new file mode 100644 > index 0000000000..471e0e7292 > --- /dev/null > +++ b/hw/s390x/cpu-topology.c > @@ -0,0 +1,259 @@ > +/* SPDX-License-Identifier: GPL-2.0-or-later */ > +/* > + * CPU Topology Since you later introduce a file with almost the same name in the target/s390x/ folder, it would be fine to have some more explanation here what this file is all about (especially with regards to the other file in target/s390x/). > + * > + * Copyright IBM Corp. 2022,2023 > + * Author(s): Pierre Morel <pmorel@linux.ibm.com> > + * > + */ > + > +#include "qemu/osdep.h" > +#include "qapi/error.h" > +#include "qemu/error-report.h" > +#include "hw/qdev-properties.h" > +#include "hw/boards.h" > +#include "qemu/typedefs.h" qemu/typedefs.h is already included by osdep.h by default, no need to re-include it here. > +#include "target/s390x/cpu.h" > +#include "hw/s390x/s390-virtio-ccw.h" > +#include "hw/s390x/cpu-topology.h" ...> + > +/** > + * s390_topology_setup_cpu: > + * @ms: MachineState used to initialize the topology structure on > + * first call. > + * @cpu: the new S390CPU to insert in the topology structure > + * @errp: the error pointer > + * > + * Called from CPU Hotplug to check and setup the CPU attributes s/Hotplug/hotplug/ > + * before the CPU is inserted in the topology. > + * There is no need to update the MTCR explicitely here because it s/explicitely/explicitly/ > + * will be updated by KVM on creation of the new CPU. > + */ > +void s390_topology_setup_cpu(MachineState *ms, S390CPU *cpu, Error **errp) ... Thomas
On 4/27/23 15:38, Thomas Huth wrote: > On 25/04/2023 18.14, Pierre Morel wrote: >> The topology information are attributes of the CPU and are >> specified during the CPU device creation. >> >> On hot plug we: >> - calculate the default values for the topology for drawers, >> books and sockets in the case they are not specified. >> - verify the CPU attributes >> - check that we have still room on the desired socket >> >> The possibility to insert a CPU in a mask is dependent on the >> number of cores allowed in a socket, a book or a drawer, the >> checking is done during the hot plug of the CPU to have an >> immediate answer. >> >> If the complete topology is not specified, the core is added >> in the physical topology based on its core ID and it gets >> defaults values for the modifier attributes. >> >> This way, starting QEMU without specifying the topology can >> still get some advantage of the CPU topology. >> >> Signed-off-by: Pierre Morel <pmorel@linux.ibm.com> >> --- > ... >> diff --git a/hw/s390x/cpu-topology.c b/hw/s390x/cpu-topology.c >> new file mode 100644 >> index 0000000000..471e0e7292 >> --- /dev/null >> +++ b/hw/s390x/cpu-topology.c >> @@ -0,0 +1,259 @@ >> +/* SPDX-License-Identifier: GPL-2.0-or-later */ >> +/* >> + * CPU Topology > > Since you later introduce a file with almost the same name in the > target/s390x/ folder, it would be fine to have some more explanation > here what this file is all about (especially with regards to the other > file in target/s390x/). I first did put the interceptions in target/s390/ then moved them in target/s390x/kvm because it is KVM related then again only let STSI interception. But to be honest I do not see any reason why not put everything in hw/s390x/ if CPU topology is implemented for TCG I think the code will call insert_stsi_15_1_x() too. no? > >> + * >> + * Copyright IBM Corp. 2022,2023 >> + * Author(s): Pierre Morel <pmorel@linux.ibm.com> >> + * >> + */ >> + >> +#include "qemu/osdep.h" >> +#include "qapi/error.h" >> +#include "qemu/error-report.h" >> +#include "hw/qdev-properties.h" >> +#include "hw/boards.h" >> +#include "qemu/typedefs.h" > > qemu/typedefs.h is already included by osdep.h by default, no need to > re-include it here. OK , I remove it > >> +#include "target/s390x/cpu.h" >> +#include "hw/s390x/s390-virtio-ccw.h" >> +#include "hw/s390x/cpu-topology.h" > ...> + >> +/** >> + * s390_topology_setup_cpu: >> + * @ms: MachineState used to initialize the topology structure on >> + * first call. >> + * @cpu: the new S390CPU to insert in the topology structure >> + * @errp: the error pointer >> + * >> + * Called from CPU Hotplug to check and setup the CPU attributes > > s/Hotplug/hotplug/ yes > >> + * before the CPU is inserted in the topology. >> + * There is no need to update the MTCR explicitely here because it > > s/explicitely/explicitly/ thanx > >> + * will be updated by KVM on creation of the new CPU. >> + */ >> +void s390_topology_setup_cpu(MachineState *ms, S390CPU *cpu, Error >> **errp) > ... > > Thomas > > Regards, Pierre
On 4/25/23 18:14, Pierre Morel wrote: > The topology information are attributes of the CPU and are > specified during the CPU device creation. > > On hot plug we: > - calculate the default values for the topology for drawers, > books and sockets in the case they are not specified. > - verify the CPU attributes > - check that we have still room on the desired socket > > The possibility to insert a CPU in a mask is dependent on the > number of cores allowed in a socket, a book or a drawer, the > checking is done during the hot plug of the CPU to have an > immediate answer. > > If the complete topology is not specified, the core is added > in the physical topology based on its core ID and it gets > defaults values for the modifier attributes. > > This way, starting QEMU without specifying the topology can > still get some advantage of the CPU topology. > > Signed-off-by: Pierre Morel <pmorel@linux.ibm.com> > --- > MAINTAINERS | 6 + > include/hw/s390x/cpu-topology.h | 55 +++++++ > hw/s390x/cpu-topology.c | 259 ++++++++++++++++++++++++++++++++ > hw/s390x/s390-virtio-ccw.c | 22 ++- > hw/s390x/meson.build | 1 + > 5 files changed, 341 insertions(+), 2 deletions(-) > create mode 100644 include/hw/s390x/cpu-topology.h > create mode 100644 hw/s390x/cpu-topology.c > > diff --git a/MAINTAINERS b/MAINTAINERS > index 5340de0515..bb7b34d0d8 100644 > --- a/MAINTAINERS > +++ b/MAINTAINERS > @@ -1654,6 +1654,12 @@ F: hw/s390x/event-facility.c > F: hw/s390x/sclp*.c > L: qemu-s390x@nongnu.org > > +S390 CPU topology > +M: Pierre Morel <pmorel@linux.ibm.com> > +S: Supported > +F: include/hw/s390x/cpu-topology.h > +F: hw/s390x/cpu-topology.c > + > X86 Machines > ------------ > PC > diff --git a/include/hw/s390x/cpu-topology.h b/include/hw/s390x/cpu-topology.h > new file mode 100644 > index 0000000000..af36f634e0 > --- /dev/null > +++ b/include/hw/s390x/cpu-topology.h > @@ -0,0 +1,55 @@ > +/* SPDX-License-Identifier: GPL-2.0-or-later */ > +/* > + * CPU Topology > + * > + * Copyright IBM Corp. 2022,2023 > + * Author(s): Pierre Morel <pmorel@linux.ibm.com> > + * > + */ > +#ifndef HW_S390X_CPU_TOPOLOGY_H > +#define HW_S390X_CPU_TOPOLOGY_H > + > +#ifndef CONFIG_USER_ONLY > + > +#include "qemu/queue.h" > +#include "hw/boards.h" > +#include "qapi/qapi-types-machine-target.h" > + > +typedef struct S390Topology { > + uint8_t *cores_per_socket; > + CpuTopology *smp; > +} S390Topology; > + > +#ifdef CONFIG_KVM > +bool s390_has_topology(void); > +void s390_topology_setup_cpu(MachineState *ms, S390CPU *cpu, Error **errp); > +#else > +static inline bool s390_has_topology(void) > +{ > + return false; > +} > +static inline void s390_topology_setup_cpu(MachineState *ms, > + S390CPU *cpu, > + Error **errp) {} > +#endif > + > +extern S390Topology s390_topology; > + > +static inline int s390_std_socket(int n, CpuTopology *smp) > +{ > + return (n / smp->cores) % smp->sockets; > +} > + > +static inline int s390_std_book(int n, CpuTopology *smp) > +{ > + return (n / (smp->cores * smp->sockets)) % smp->books; > +} > + > +static inline int s390_std_drawer(int n, CpuTopology *smp) > +{ > + return (n / (smp->cores * smp->sockets * smp->books)) % smp->drawers; > +} > + > +#endif /* CONFIG_USER_ONLY */ > + > +#endif > diff --git a/hw/s390x/cpu-topology.c b/hw/s390x/cpu-topology.c > new file mode 100644 > index 0000000000..471e0e7292 > --- /dev/null > +++ b/hw/s390x/cpu-topology.c > @@ -0,0 +1,259 @@ > +/* SPDX-License-Identifier: GPL-2.0-or-later */ > +/* > + * CPU Topology > + * > + * Copyright IBM Corp. 2022,2023 > + * Author(s): Pierre Morel <pmorel@linux.ibm.com> > + * > + */ > + > +#include "qemu/osdep.h" > +#include "qapi/error.h" > +#include "qemu/error-report.h" > +#include "hw/qdev-properties.h" > +#include "hw/boards.h" > +#include "qemu/typedefs.h" > +#include "target/s390x/cpu.h" > +#include "hw/s390x/s390-virtio-ccw.h" > +#include "hw/s390x/cpu-topology.h" > + > +/* > + * s390_topology is used to keep the topology information. > + * .cores_per_socket: tracks information on the count of cores > + * per socket. > + * .smp: keeps track of the machine topology. > + * > + */ > +S390Topology s390_topology = { > + /* will be initialized after the cpu model is realized */ > + .cores_per_socket = NULL, > + .smp = NULL, > +}; > + > +/** > + * s390_socket_nb: > + * @cpu: s390x CPU > + * > + * Returns the socket number used inside the cores_per_socket array > + * for a topology tree entry > + */ > +static int __s390_socket_nb(int drawer_id, int book_id, int socket_id) > +{ > + return (drawer_id * s390_topology.smp->books + book_id) * > + s390_topology.smp->sockets + socket_id; > +} > + > +/** > + * s390_socket_nb: > + * @cpu: s390x CPU > + * > + * Returns the socket number used inside the cores_per_socket array > + * for a cpu. > + */ > +static int s390_socket_nb(S390CPU *cpu) > +{ > + return __s390_socket_nb(cpu->env.drawer_id, cpu->env.book_id, > + cpu->env.socket_id); > +} > + > +/** > + * s390_has_topology: > + * > + * Return value: if the topology is supported by the machine. > + */ > +bool s390_has_topology(void) > +{ > + return false; > +} > + > +/** > + * s390_topology_init: > + * @ms: the machine state where the machine topology is defined > + * > + * Keep track of the machine topology. > + * > + * Allocate an array to keep the count of cores per socket. > + * The index of the array starts at socket 0 from book 0 and > + * drawer 0 up to the maximum allowed by the machine topology. > + */ > +static void s390_topology_init(MachineState *ms) > +{ > + CpuTopology *smp = &ms->smp; > + > + s390_topology.smp = smp; I am not sure the 'smp' shortcut is necessary. 'MachineState *ms' is always available where 'CpuTopology *smp' is used. so it could be computed from a local variable AFAICT. It would reduce the risk of 'smp' being NULL in some (future) code path. Thanks, C. > + s390_topology.cores_per_socket = g_new0(uint8_t, smp->sockets * > + smp->books * smp->drawers); > +} > + > +/** > + * s390_topology_cpu_default: > + * @cpu: pointer to a S390CPU > + * @errp: Error pointer > + * > + * Setup the default topology if no attributes are already set. > + * Passing a CPU with some, but not all, attributes set is considered > + * an error. > + * > + * The function calculates the (drawer_id, book_id, socket_id) > + * topology by filling the cores starting from the first socket > + * (0, 0, 0) up to the last (smp->drawers, smp->books, smp->sockets). > + * > + * CPU type and dedication have defaults values set in the > + * s390x_cpu_properties, entitlement must be adjust depending on the > + * dedication. > + * > + * Returns false if it is impossible to setup a default topology > + * true otherwise. > + */ > +static bool s390_topology_cpu_default(S390CPU *cpu, Error **errp) > +{ > + CpuTopology *smp = s390_topology.smp; > + CPUS390XState *env = &cpu->env; > + > + /* All geometry topology attributes must be set or all unset */ > + if ((env->socket_id < 0 || env->book_id < 0 || env->drawer_id < 0) && > + (env->socket_id >= 0 || env->book_id >= 0 || env->drawer_id >= 0)) { > + error_setg(errp, > + "Please define all or none of the topology geometry attributes"); > + return false; > + } > + > + /* Check if one of the geometry topology is unset */ > + if (env->socket_id < 0) { > + /* Calculate default geometry topology attributes */ > + env->socket_id = s390_std_socket(env->core_id, smp); > + env->book_id = s390_std_book(env->core_id, smp); > + env->drawer_id = s390_std_drawer(env->core_id, smp); > + } > + > + /* > + * When the user specifies the entitlement as 'auto' on the command line, > + * qemu will set the entitlement as: > + * Medium when the CPU is not dedicated. > + * High when dedicated is true. > + */ > + if (env->entitlement == S390_CPU_ENTITLEMENT_AUTO) { > + if (env->dedicated) { > + env->entitlement = S390_CPU_ENTITLEMENT_HIGH; > + } else { > + env->entitlement = S390_CPU_ENTITLEMENT_MEDIUM; > + } > + } > + return true; > +} > + > +/** > + * s390_topology_check: > + * @socket_id: socket to check > + * @book_id: book to check > + * @drawer_id: drawer to check > + * @entitlement: entitlement to check > + * @dedicated: dedication to check > + * @errp: Error pointer > + * > + * The function checks if the topology > + * attributes fits inside the system topology. > + * > + * Returns false if the specified topology does not match with > + * the machine topology. > + */ > +static bool s390_topology_check(uint16_t socket_id, uint16_t book_id, > + uint16_t drawer_id, uint16_t entitlement, > + bool dedicated, Error **errp) > +{ > + CpuTopology *smp = s390_topology.smp; > + ERRP_GUARD(); > + > + if (socket_id >= smp->sockets) { > + error_setg(errp, "Unavailable socket: %d", socket_id); > + return false; > + } > + if (book_id >= smp->books) { > + error_setg(errp, "Unavailable book: %d", book_id); > + return false; > + } > + if (drawer_id >= smp->drawers) { > + error_setg(errp, "Unavailable drawer: %d", drawer_id); > + return false; > + } > + if (entitlement >= S390_CPU_ENTITLEMENT__MAX) { > + error_setg(errp, "Unknown entitlement: %d", entitlement); > + return false; > + } > + if (dedicated && (entitlement == S390_CPU_ENTITLEMENT_LOW || > + entitlement == S390_CPU_ENTITLEMENT_MEDIUM)) { > + error_setg(errp, "A dedicated cpu implies high entitlement"); > + return false; > + } > + return true; > +} > + > +/** > + * s390_update_cpu_props: > + * @ms: the machine state > + * @cpu: the CPU for which to update the properties from the environment. > + * > + */ > +static void s390_update_cpu_props(MachineState *ms, S390CPU *cpu) > +{ > + CpuInstanceProperties *props; > + > + props = &ms->possible_cpus->cpus[cpu->env.core_id].props; > + > + props->socket_id = cpu->env.socket_id; > + props->book_id = cpu->env.book_id; > + props->drawer_id = cpu->env.drawer_id; > +} > + > +/** > + * s390_topology_setup_cpu: > + * @ms: MachineState used to initialize the topology structure on > + * first call. > + * @cpu: the new S390CPU to insert in the topology structure > + * @errp: the error pointer > + * > + * Called from CPU Hotplug to check and setup the CPU attributes > + * before the CPU is inserted in the topology. > + * There is no need to update the MTCR explicitely here because it > + * will be updated by KVM on creation of the new CPU. > + */ > +void s390_topology_setup_cpu(MachineState *ms, S390CPU *cpu, Error **errp) > +{ > + ERRP_GUARD(); > + int entry; > + > + /* > + * We do not want to initialize the topology if the cpu model > + * does not support topology, consequently, we have to wait for > + * the first CPU to be realized, which realizes the CPU model > + * to initialize the topology structures. > + * > + * s390_topology_setup_cpu() is called from the cpu hotplug. > + */ > + if (!s390_topology.cores_per_socket) { > + s390_topology_init(ms); > + } > + > + if (!s390_topology_cpu_default(cpu, errp)) { > + return; > + } > + > + if (!s390_topology_check(cpu->env.socket_id, cpu->env.book_id, > + cpu->env.drawer_id, cpu->env.entitlement, > + cpu->env.dedicated, errp)) { > + return; > + } > + > + /* Do we still have space in the socket */ > + entry = s390_socket_nb(cpu); > + if (s390_topology.cores_per_socket[entry] >= s390_topology.smp->cores) { > + error_setg(errp, "No more space on this socket"); > + return; > + } > + > + /* Update the count of cores in sockets */ > + s390_topology.cores_per_socket[entry] += 1; > + > + /* topology tree is reflected in props */ > + s390_update_cpu_props(ms, cpu); > +} > diff --git a/hw/s390x/s390-virtio-ccw.c b/hw/s390x/s390-virtio-ccw.c > index 1a9bcda8b6..9df60ac447 100644 > --- a/hw/s390x/s390-virtio-ccw.c > +++ b/hw/s390x/s390-virtio-ccw.c > @@ -45,6 +45,7 @@ > #include "hw/s390x/pv.h" > #include "migration/blocker.h" > #include "qapi/visitor.h" > +#include "hw/s390x/cpu-topology.h" > > static Error *pv_mig_blocker; > > @@ -311,10 +312,18 @@ static void s390_cpu_plug(HotplugHandler *hotplug_dev, > { > MachineState *ms = MACHINE(hotplug_dev); > S390CPU *cpu = S390_CPU(dev); > + ERRP_GUARD(); > > g_assert(!ms->possible_cpus->cpus[cpu->env.core_id].cpu); > ms->possible_cpus->cpus[cpu->env.core_id].cpu = OBJECT(dev); > > + if (s390_has_topology()) { > + s390_topology_setup_cpu(ms, cpu, errp); > + if (*errp) { > + return; > + } > + } > + > if (dev->hotplugged) { > raise_irq_cpu_hotplug(); > } > @@ -554,11 +563,20 @@ static const CPUArchIdList *s390_possible_cpu_arch_ids(MachineState *ms) > sizeof(CPUArchId) * max_cpus); > ms->possible_cpus->len = max_cpus; > for (i = 0; i < ms->possible_cpus->len; i++) { > + CpuInstanceProperties *props = &ms->possible_cpus->cpus[i].props; > + > ms->possible_cpus->cpus[i].type = ms->cpu_type; > ms->possible_cpus->cpus[i].vcpus_count = 1; > ms->possible_cpus->cpus[i].arch_id = i; > - ms->possible_cpus->cpus[i].props.has_core_id = true; > - ms->possible_cpus->cpus[i].props.core_id = i; > + > + props->has_core_id = true; > + props->core_id = i; > + props->has_socket_id = true; > + props->socket_id = s390_std_socket(i, &ms->smp); > + props->has_book_id = true; > + props->book_id = s390_std_book(i, &ms->smp); > + props->has_drawer_id = true; > + props->drawer_id = s390_std_drawer(i, &ms->smp); > } > > return ms->possible_cpus; > diff --git a/hw/s390x/meson.build b/hw/s390x/meson.build > index f291016fee..58dfbdff4f 100644 > --- a/hw/s390x/meson.build > +++ b/hw/s390x/meson.build > @@ -24,6 +24,7 @@ s390x_ss.add(when: 'CONFIG_KVM', if_true: files( > 's390-stattrib-kvm.c', > 'pv.c', > 's390-pci-kvm.c', > + 'cpu-topology.c', > )) > s390x_ss.add(when: 'CONFIG_TCG', if_true: files( > 'tod-tcg.c',
On 5/2/23 14:30, Cédric Le Goater wrote: > On 4/25/23 18:14, Pierre Morel wrote: >> The topology information are attributes of the CPU and are >> specified during the CPU device creation. >> >> [...] >> + >> +/** >> + * s390_topology_init: >> + * @ms: the machine state where the machine topology is defined >> + * >> + * Keep track of the machine topology. >> + * >> + * Allocate an array to keep the count of cores per socket. >> + * The index of the array starts at socket 0 from book 0 and >> + * drawer 0 up to the maximum allowed by the machine topology. >> + */ >> +static void s390_topology_init(MachineState *ms) >> +{ >> + CpuTopology *smp = &ms->smp; >> + >> + s390_topology.smp = smp; > > I am not sure the 'smp' shortcut is necessary. 'MachineState *ms' is > always available where 'CpuTopology *smp' is used. so it could be > computed from a local variable AFAICT. It would reduce the risk of > 'smp' being NULL in some (future) code path. > > Thanks, > > C. So I will use directly current_machine->smp You are right it is more homogeneous, there is no need to keep a pointer here. Thanks Pierre [...]
On 28/04/2023 14.35, Pierre Morel wrote: > > On 4/27/23 15:38, Thomas Huth wrote: >> On 25/04/2023 18.14, Pierre Morel wrote: >>> The topology information are attributes of the CPU and are >>> specified during the CPU device creation. >>> >>> On hot plug we: >>> - calculate the default values for the topology for drawers, >>> books and sockets in the case they are not specified. >>> - verify the CPU attributes >>> - check that we have still room on the desired socket >>> >>> The possibility to insert a CPU in a mask is dependent on the >>> number of cores allowed in a socket, a book or a drawer, the >>> checking is done during the hot plug of the CPU to have an >>> immediate answer. >>> >>> If the complete topology is not specified, the core is added >>> in the physical topology based on its core ID and it gets >>> defaults values for the modifier attributes. >>> >>> This way, starting QEMU without specifying the topology can >>> still get some advantage of the CPU topology. >>> >>> Signed-off-by: Pierre Morel <pmorel@linux.ibm.com> >>> --- >> ... >>> diff --git a/hw/s390x/cpu-topology.c b/hw/s390x/cpu-topology.c >>> new file mode 100644 >>> index 0000000000..471e0e7292 >>> --- /dev/null >>> +++ b/hw/s390x/cpu-topology.c >>> @@ -0,0 +1,259 @@ >>> +/* SPDX-License-Identifier: GPL-2.0-or-later */ >>> +/* >>> + * CPU Topology >> >> Since you later introduce a file with almost the same name in the >> target/s390x/ folder, it would be fine to have some more explanation here >> what this file is all about (especially with regards to the other file in >> target/s390x/). > > > I first did put the interceptions in target/s390/ then moved them in > target/s390x/kvm because it is KVM related then again only let STSI > interception. > > But to be honest I do not see any reason why not put everything in hw/s390x/ > if CPU topology is implemented for TCG I think the code will call > insert_stsi_15_1_x() too. > > no? Oh well, it's all so borderline ... whether you rather think of this as part of the CPU (like the STSI instruction) or rather part of the machine (drawers, books, ...). I don't mind too much, as long as we don't have two files around with almost the same name (apart from "_" vs. "-"). So either keep the stsi part in target/s390x and use a better file name for that, or put everything together in one "cpu-topology.c" file. Or what do others think about it? Thomas
Hello, On 5/3/23 11:12, Thomas Huth wrote: > On 28/04/2023 14.35, Pierre Morel wrote: >> >> On 4/27/23 15:38, Thomas Huth wrote: >>> On 25/04/2023 18.14, Pierre Morel wrote: >>>> The topology information are attributes of the CPU and are >>>> specified during the CPU device creation. >>>> >>>> On hot plug we: >>>> - calculate the default values for the topology for drawers, >>>> books and sockets in the case they are not specified. >>>> - verify the CPU attributes >>>> - check that we have still room on the desired socket >>>> >>>> The possibility to insert a CPU in a mask is dependent on the >>>> number of cores allowed in a socket, a book or a drawer, the >>>> checking is done during the hot plug of the CPU to have an >>>> immediate answer. >>>> >>>> If the complete topology is not specified, the core is added >>>> in the physical topology based on its core ID and it gets >>>> defaults values for the modifier attributes. >>>> >>>> This way, starting QEMU without specifying the topology can >>>> still get some advantage of the CPU topology. >>>> >>>> Signed-off-by: Pierre Morel <pmorel@linux.ibm.com> >>>> --- >>> ... >>>> diff --git a/hw/s390x/cpu-topology.c b/hw/s390x/cpu-topology.c >>>> new file mode 100644 >>>> index 0000000000..471e0e7292 >>>> --- /dev/null >>>> +++ b/hw/s390x/cpu-topology.c >>>> @@ -0,0 +1,259 @@ >>>> +/* SPDX-License-Identifier: GPL-2.0-or-later */ >>>> +/* >>>> + * CPU Topology >>> >>> Since you later introduce a file with almost the same name in the target/s390x/ folder, it would be fine to have some more explanation here what this file is all about (especially with regards to the other file in target/s390x/). >> >> >> I first did put the interceptions in target/s390/ then moved them in target/s390x/kvm because it is KVM related then again only let STSI interception. >> >> But to be honest I do not see any reason why not put everything in hw/s390x/ if CPU topology is implemented for TCG I think the code will call insert_stsi_15_1_x() too. >> >> no? > > Oh well, it's all so borderline ... whether you rather think of this as part of the CPU (like the STSI instruction) or rather part of the machine (drawers, books, ...). > I don't mind too much, as long as we don't have two files around with almost the same name (apart from "_" vs. "-"). So either keep the stsi part in target/s390x and use a better file name for that, or put everything together in one "cpu-topology.c" file. > Or what do others think about it? Would it make sense to have a target/s390x/stsi.c file with the stsi routines to be called from TCG insn helpers and from the KVM backend ? This suggestion is based on the services found in the ioinst.c file. So, target/s390x/kvm/cpu_topology.c would become target/s390x/stsi.c and stsi services would be moved there, if that makes sense. Or target/s390x/kvm/stsi.c to start with because services are only active for KVM targets. Looking at hw/s390x/meson.build : s390x_ss.add(when: 'CONFIG_KVM', if_true: files( 'tod-kvm.c', 's390-skeys-kvm.c', 's390-stattrib-kvm.c', 'pv.c', 's390-pci-kvm.c', 'cpu-topology.c', )) It seems cpu-topology.c should be named cpu-topology-kvm.c to follow the same convention. However, I don't see much reason for the KVM condition, apart from the new polarization definitions in machine-target.json which depend on KVM. cpu-topology.c could well be compiled without the KVM #ifdef, all seems in place to detect support at runtime. In this file, we find a s390_handle_ptf() which is called from kvm_handle_ptf() in target/s390x/kvm/kvm.c. Is it the right place for it ? Shouldn't we move the service under target/s390x/kvm/kvm.c ? Thanks, C.
On 5/3/23 09:21, Pierre Morel wrote: > > On 5/2/23 14:30, Cédric Le Goater wrote: >> On 4/25/23 18:14, Pierre Morel wrote: >>> The topology information are attributes of the CPU and are >>> specified during the CPU device creation. >>> >>> > [...] >>> + >>> +/** >>> + * s390_topology_init: >>> + * @ms: the machine state where the machine topology is defined >>> + * >>> + * Keep track of the machine topology. >>> + * >>> + * Allocate an array to keep the count of cores per socket. >>> + * The index of the array starts at socket 0 from book 0 and >>> + * drawer 0 up to the maximum allowed by the machine topology. >>> + */ >>> +static void s390_topology_init(MachineState *ms) >>> +{ >>> + CpuTopology *smp = &ms->smp; >>> + >>> + s390_topology.smp = smp; >> >> I am not sure the 'smp' shortcut is necessary. 'MachineState *ms' is >> always available where 'CpuTopology *smp' is used. so it could be >> computed from a local variable AFAICT. It would reduce the risk of >> 'smp' being NULL in some (future) code path. >> >> Thanks, >> >> C. > > > So I will use directly current_machine->smp There are a couple of place where 'MachineState *ms' can be used directly. In the other places, routines will need an extra 'CpuTopology *smp' parameter. If it gets too ugly, current_machine->smp might a better alternative. Thanks, C.
On 5/3/23 12:23, Cédric Le Goater wrote: > Hello, > > On 5/3/23 11:12, Thomas Huth wrote: >> On 28/04/2023 14.35, Pierre Morel wrote: >>> >>> On 4/27/23 15:38, Thomas Huth wrote: >>>> On 25/04/2023 18.14, Pierre Morel wrote: >>>>> The topology information are attributes of the CPU and are >>>>> specified during the CPU device creation. >>>>> >>>>> On hot plug we: >>>>> - calculate the default values for the topology for drawers, >>>>> books and sockets in the case they are not specified. >>>>> - verify the CPU attributes >>>>> - check that we have still room on the desired socket >>>>> >>>>> The possibility to insert a CPU in a mask is dependent on the >>>>> number of cores allowed in a socket, a book or a drawer, the >>>>> checking is done during the hot plug of the CPU to have an >>>>> immediate answer. >>>>> >>>>> If the complete topology is not specified, the core is added >>>>> in the physical topology based on its core ID and it gets >>>>> defaults values for the modifier attributes. >>>>> >>>>> This way, starting QEMU without specifying the topology can >>>>> still get some advantage of the CPU topology. >>>>> >>>>> Signed-off-by: Pierre Morel <pmorel@linux.ibm.com> >>>>> --- >>>> ... >>>>> diff --git a/hw/s390x/cpu-topology.c b/hw/s390x/cpu-topology.c >>>>> new file mode 100644 >>>>> index 0000000000..471e0e7292 >>>>> --- /dev/null >>>>> +++ b/hw/s390x/cpu-topology.c >>>>> @@ -0,0 +1,259 @@ >>>>> +/* SPDX-License-Identifier: GPL-2.0-or-later */ >>>>> +/* >>>>> + * CPU Topology >>>> >>>> Since you later introduce a file with almost the same name in the >>>> target/s390x/ folder, it would be fine to have some more >>>> explanation here what this file is all about (especially with >>>> regards to the other file in target/s390x/). >>> >>> >>> I first did put the interceptions in target/s390/ then moved them in >>> target/s390x/kvm because it is KVM related then again only let STSI >>> interception. >>> >>> But to be honest I do not see any reason why not put everything in >>> hw/s390x/ if CPU topology is implemented for TCG I think the code >>> will call insert_stsi_15_1_x() too. >>> >>> no? >> >> Oh well, it's all so borderline ... whether you rather think of this >> as part of the CPU (like the STSI instruction) or rather part of the >> machine (drawers, books, ...). >> I don't mind too much, as long as we don't have two files around with >> almost the same name (apart from "_" vs. "-"). So either keep the >> stsi part in target/s390x and use a better file name for that, or put >> everything together in one "cpu-topology.c" file. >> Or what do others think about it? > > Would it make sense to have a target/s390x/stsi.c file with the stsi > routines to be called from TCG insn helpers and from the KVM backend ? > This suggestion is based on the services found in the ioinst.c file. > > So, target/s390x/kvm/cpu_topology.c would become target/s390x/stsi.c > and stsi services would be moved there, if that makes sense. > > Or target/s390x/kvm/stsi.c to start with because services are only > active for KVM targets. > > > Looking at hw/s390x/meson.build : > > s390x_ss.add(when: 'CONFIG_KVM', if_true: files( > 'tod-kvm.c', > 's390-skeys-kvm.c', > 's390-stattrib-kvm.c', > 'pv.c', > 's390-pci-kvm.c', > 'cpu-topology.c', > )) > > It seems cpu-topology.c should be named cpu-topology-kvm.c to follow > the same convention. > > However, I don't see much reason for the KVM condition, apart from > the new polarization definitions in machine-target.json which depend > on KVM. cpu-topology.c could well be compiled without the KVM #ifdef, > all seems in place to detect support at runtime. > > In this file, we find a s390_handle_ptf() which is called from > kvm_handle_ptf() in target/s390x/kvm/kvm.c. Is it the right place for > it ? Shouldn't we move the service under target/s390x/kvm/kvm.c ? > > Thanks, > > C. > > Hello Cédric, The reason to have the s390_handle_ptf() here is to be ready for TCG. We have the choice to let the cpu-topology.c file inside the KVM list of meson.build until TCG provides the CPU_TOPOLOGY facility and then move it to the general list or to move it now to the general list inside meson.build and keep code that will only be useful when TCG provides the CPU_TOPOLOGY facility The option taken here is to not let the code active if it is not in use. Did I answered the question or forgot something? Regards, Pierre
diff --git a/MAINTAINERS b/MAINTAINERS index 5340de0515..bb7b34d0d8 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1654,6 +1654,12 @@ F: hw/s390x/event-facility.c F: hw/s390x/sclp*.c L: qemu-s390x@nongnu.org +S390 CPU topology +M: Pierre Morel <pmorel@linux.ibm.com> +S: Supported +F: include/hw/s390x/cpu-topology.h +F: hw/s390x/cpu-topology.c + X86 Machines ------------ PC diff --git a/include/hw/s390x/cpu-topology.h b/include/hw/s390x/cpu-topology.h new file mode 100644 index 0000000000..af36f634e0 --- /dev/null +++ b/include/hw/s390x/cpu-topology.h @@ -0,0 +1,55 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * CPU Topology + * + * Copyright IBM Corp. 2022,2023 + * Author(s): Pierre Morel <pmorel@linux.ibm.com> + * + */ +#ifndef HW_S390X_CPU_TOPOLOGY_H +#define HW_S390X_CPU_TOPOLOGY_H + +#ifndef CONFIG_USER_ONLY + +#include "qemu/queue.h" +#include "hw/boards.h" +#include "qapi/qapi-types-machine-target.h" + +typedef struct S390Topology { + uint8_t *cores_per_socket; + CpuTopology *smp; +} S390Topology; + +#ifdef CONFIG_KVM +bool s390_has_topology(void); +void s390_topology_setup_cpu(MachineState *ms, S390CPU *cpu, Error **errp); +#else +static inline bool s390_has_topology(void) +{ + return false; +} +static inline void s390_topology_setup_cpu(MachineState *ms, + S390CPU *cpu, + Error **errp) {} +#endif + +extern S390Topology s390_topology; + +static inline int s390_std_socket(int n, CpuTopology *smp) +{ + return (n / smp->cores) % smp->sockets; +} + +static inline int s390_std_book(int n, CpuTopology *smp) +{ + return (n / (smp->cores * smp->sockets)) % smp->books; +} + +static inline int s390_std_drawer(int n, CpuTopology *smp) +{ + return (n / (smp->cores * smp->sockets * smp->books)) % smp->drawers; +} + +#endif /* CONFIG_USER_ONLY */ + +#endif diff --git a/hw/s390x/cpu-topology.c b/hw/s390x/cpu-topology.c new file mode 100644 index 0000000000..471e0e7292 --- /dev/null +++ b/hw/s390x/cpu-topology.c @@ -0,0 +1,259 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * CPU Topology + * + * Copyright IBM Corp. 2022,2023 + * Author(s): Pierre Morel <pmorel@linux.ibm.com> + * + */ + +#include "qemu/osdep.h" +#include "qapi/error.h" +#include "qemu/error-report.h" +#include "hw/qdev-properties.h" +#include "hw/boards.h" +#include "qemu/typedefs.h" +#include "target/s390x/cpu.h" +#include "hw/s390x/s390-virtio-ccw.h" +#include "hw/s390x/cpu-topology.h" + +/* + * s390_topology is used to keep the topology information. + * .cores_per_socket: tracks information on the count of cores + * per socket. + * .smp: keeps track of the machine topology. + * + */ +S390Topology s390_topology = { + /* will be initialized after the cpu model is realized */ + .cores_per_socket = NULL, + .smp = NULL, +}; + +/** + * s390_socket_nb: + * @cpu: s390x CPU + * + * Returns the socket number used inside the cores_per_socket array + * for a topology tree entry + */ +static int __s390_socket_nb(int drawer_id, int book_id, int socket_id) +{ + return (drawer_id * s390_topology.smp->books + book_id) * + s390_topology.smp->sockets + socket_id; +} + +/** + * s390_socket_nb: + * @cpu: s390x CPU + * + * Returns the socket number used inside the cores_per_socket array + * for a cpu. + */ +static int s390_socket_nb(S390CPU *cpu) +{ + return __s390_socket_nb(cpu->env.drawer_id, cpu->env.book_id, + cpu->env.socket_id); +} + +/** + * s390_has_topology: + * + * Return value: if the topology is supported by the machine. + */ +bool s390_has_topology(void) +{ + return false; +} + +/** + * s390_topology_init: + * @ms: the machine state where the machine topology is defined + * + * Keep track of the machine topology. + * + * Allocate an array to keep the count of cores per socket. + * The index of the array starts at socket 0 from book 0 and + * drawer 0 up to the maximum allowed by the machine topology. + */ +static void s390_topology_init(MachineState *ms) +{ + CpuTopology *smp = &ms->smp; + + s390_topology.smp = smp; + s390_topology.cores_per_socket = g_new0(uint8_t, smp->sockets * + smp->books * smp->drawers); +} + +/** + * s390_topology_cpu_default: + * @cpu: pointer to a S390CPU + * @errp: Error pointer + * + * Setup the default topology if no attributes are already set. + * Passing a CPU with some, but not all, attributes set is considered + * an error. + * + * The function calculates the (drawer_id, book_id, socket_id) + * topology by filling the cores starting from the first socket + * (0, 0, 0) up to the last (smp->drawers, smp->books, smp->sockets). + * + * CPU type and dedication have defaults values set in the + * s390x_cpu_properties, entitlement must be adjust depending on the + * dedication. + * + * Returns false if it is impossible to setup a default topology + * true otherwise. + */ +static bool s390_topology_cpu_default(S390CPU *cpu, Error **errp) +{ + CpuTopology *smp = s390_topology.smp; + CPUS390XState *env = &cpu->env; + + /* All geometry topology attributes must be set or all unset */ + if ((env->socket_id < 0 || env->book_id < 0 || env->drawer_id < 0) && + (env->socket_id >= 0 || env->book_id >= 0 || env->drawer_id >= 0)) { + error_setg(errp, + "Please define all or none of the topology geometry attributes"); + return false; + } + + /* Check if one of the geometry topology is unset */ + if (env->socket_id < 0) { + /* Calculate default geometry topology attributes */ + env->socket_id = s390_std_socket(env->core_id, smp); + env->book_id = s390_std_book(env->core_id, smp); + env->drawer_id = s390_std_drawer(env->core_id, smp); + } + + /* + * When the user specifies the entitlement as 'auto' on the command line, + * qemu will set the entitlement as: + * Medium when the CPU is not dedicated. + * High when dedicated is true. + */ + if (env->entitlement == S390_CPU_ENTITLEMENT_AUTO) { + if (env->dedicated) { + env->entitlement = S390_CPU_ENTITLEMENT_HIGH; + } else { + env->entitlement = S390_CPU_ENTITLEMENT_MEDIUM; + } + } + return true; +} + +/** + * s390_topology_check: + * @socket_id: socket to check + * @book_id: book to check + * @drawer_id: drawer to check + * @entitlement: entitlement to check + * @dedicated: dedication to check + * @errp: Error pointer + * + * The function checks if the topology + * attributes fits inside the system topology. + * + * Returns false if the specified topology does not match with + * the machine topology. + */ +static bool s390_topology_check(uint16_t socket_id, uint16_t book_id, + uint16_t drawer_id, uint16_t entitlement, + bool dedicated, Error **errp) +{ + CpuTopology *smp = s390_topology.smp; + ERRP_GUARD(); + + if (socket_id >= smp->sockets) { + error_setg(errp, "Unavailable socket: %d", socket_id); + return false; + } + if (book_id >= smp->books) { + error_setg(errp, "Unavailable book: %d", book_id); + return false; + } + if (drawer_id >= smp->drawers) { + error_setg(errp, "Unavailable drawer: %d", drawer_id); + return false; + } + if (entitlement >= S390_CPU_ENTITLEMENT__MAX) { + error_setg(errp, "Unknown entitlement: %d", entitlement); + return false; + } + if (dedicated && (entitlement == S390_CPU_ENTITLEMENT_LOW || + entitlement == S390_CPU_ENTITLEMENT_MEDIUM)) { + error_setg(errp, "A dedicated cpu implies high entitlement"); + return false; + } + return true; +} + +/** + * s390_update_cpu_props: + * @ms: the machine state + * @cpu: the CPU for which to update the properties from the environment. + * + */ +static void s390_update_cpu_props(MachineState *ms, S390CPU *cpu) +{ + CpuInstanceProperties *props; + + props = &ms->possible_cpus->cpus[cpu->env.core_id].props; + + props->socket_id = cpu->env.socket_id; + props->book_id = cpu->env.book_id; + props->drawer_id = cpu->env.drawer_id; +} + +/** + * s390_topology_setup_cpu: + * @ms: MachineState used to initialize the topology structure on + * first call. + * @cpu: the new S390CPU to insert in the topology structure + * @errp: the error pointer + * + * Called from CPU Hotplug to check and setup the CPU attributes + * before the CPU is inserted in the topology. + * There is no need to update the MTCR explicitely here because it + * will be updated by KVM on creation of the new CPU. + */ +void s390_topology_setup_cpu(MachineState *ms, S390CPU *cpu, Error **errp) +{ + ERRP_GUARD(); + int entry; + + /* + * We do not want to initialize the topology if the cpu model + * does not support topology, consequently, we have to wait for + * the first CPU to be realized, which realizes the CPU model + * to initialize the topology structures. + * + * s390_topology_setup_cpu() is called from the cpu hotplug. + */ + if (!s390_topology.cores_per_socket) { + s390_topology_init(ms); + } + + if (!s390_topology_cpu_default(cpu, errp)) { + return; + } + + if (!s390_topology_check(cpu->env.socket_id, cpu->env.book_id, + cpu->env.drawer_id, cpu->env.entitlement, + cpu->env.dedicated, errp)) { + return; + } + + /* Do we still have space in the socket */ + entry = s390_socket_nb(cpu); + if (s390_topology.cores_per_socket[entry] >= s390_topology.smp->cores) { + error_setg(errp, "No more space on this socket"); + return; + } + + /* Update the count of cores in sockets */ + s390_topology.cores_per_socket[entry] += 1; + + /* topology tree is reflected in props */ + s390_update_cpu_props(ms, cpu); +} diff --git a/hw/s390x/s390-virtio-ccw.c b/hw/s390x/s390-virtio-ccw.c index 1a9bcda8b6..9df60ac447 100644 --- a/hw/s390x/s390-virtio-ccw.c +++ b/hw/s390x/s390-virtio-ccw.c @@ -45,6 +45,7 @@ #include "hw/s390x/pv.h" #include "migration/blocker.h" #include "qapi/visitor.h" +#include "hw/s390x/cpu-topology.h" static Error *pv_mig_blocker; @@ -311,10 +312,18 @@ static void s390_cpu_plug(HotplugHandler *hotplug_dev, { MachineState *ms = MACHINE(hotplug_dev); S390CPU *cpu = S390_CPU(dev); + ERRP_GUARD(); g_assert(!ms->possible_cpus->cpus[cpu->env.core_id].cpu); ms->possible_cpus->cpus[cpu->env.core_id].cpu = OBJECT(dev); + if (s390_has_topology()) { + s390_topology_setup_cpu(ms, cpu, errp); + if (*errp) { + return; + } + } + if (dev->hotplugged) { raise_irq_cpu_hotplug(); } @@ -554,11 +563,20 @@ static const CPUArchIdList *s390_possible_cpu_arch_ids(MachineState *ms) sizeof(CPUArchId) * max_cpus); ms->possible_cpus->len = max_cpus; for (i = 0; i < ms->possible_cpus->len; i++) { + CpuInstanceProperties *props = &ms->possible_cpus->cpus[i].props; + ms->possible_cpus->cpus[i].type = ms->cpu_type; ms->possible_cpus->cpus[i].vcpus_count = 1; ms->possible_cpus->cpus[i].arch_id = i; - ms->possible_cpus->cpus[i].props.has_core_id = true; - ms->possible_cpus->cpus[i].props.core_id = i; + + props->has_core_id = true; + props->core_id = i; + props->has_socket_id = true; + props->socket_id = s390_std_socket(i, &ms->smp); + props->has_book_id = true; + props->book_id = s390_std_book(i, &ms->smp); + props->has_drawer_id = true; + props->drawer_id = s390_std_drawer(i, &ms->smp); } return ms->possible_cpus; diff --git a/hw/s390x/meson.build b/hw/s390x/meson.build index f291016fee..58dfbdff4f 100644 --- a/hw/s390x/meson.build +++ b/hw/s390x/meson.build @@ -24,6 +24,7 @@ s390x_ss.add(when: 'CONFIG_KVM', if_true: files( 's390-stattrib-kvm.c', 'pv.c', 's390-pci-kvm.c', + 'cpu-topology.c', )) s390x_ss.add(when: 'CONFIG_TCG', if_true: files( 'tod-tcg.c',
The topology information are attributes of the CPU and are specified during the CPU device creation. On hot plug we: - calculate the default values for the topology for drawers, books and sockets in the case they are not specified. - verify the CPU attributes - check that we have still room on the desired socket The possibility to insert a CPU in a mask is dependent on the number of cores allowed in a socket, a book or a drawer, the checking is done during the hot plug of the CPU to have an immediate answer. If the complete topology is not specified, the core is added in the physical topology based on its core ID and it gets defaults values for the modifier attributes. This way, starting QEMU without specifying the topology can still get some advantage of the CPU topology. Signed-off-by: Pierre Morel <pmorel@linux.ibm.com> --- MAINTAINERS | 6 + include/hw/s390x/cpu-topology.h | 55 +++++++ hw/s390x/cpu-topology.c | 259 ++++++++++++++++++++++++++++++++ hw/s390x/s390-virtio-ccw.c | 22 ++- hw/s390x/meson.build | 1 + 5 files changed, 341 insertions(+), 2 deletions(-) create mode 100644 include/hw/s390x/cpu-topology.h create mode 100644 hw/s390x/cpu-topology.c