| Message ID | 20221007141207.30635-1-jiaxun.yang@flygoat.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | [v5] MIPS: Expose prid and globalnumber to sysfs | expand |
On Fri, Oct 07, 2022 at 03:12:07PM +0100, Jiaxun Yang wrote: > Some application would like to know precise model and rev of processor > to do errata workaround or optimization. > > Expose them in sysfs as: > /sys/devices/system/cpu/cpuX/regs/identification/prid > /sys/devices/system/cpu/cpuX/regs/identification/globalnumber > > Reusing AArch64 CPU registers directory. > > Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com> > --- > v2: Drop static qualifier for kobj (gregkh) > v3: Use kzalloc to allocate struct cpuregs. > note: When Greg mentioned about static I was thinking about > static qualifier of percpu variable. After reading documents > again it turns out kobjs should be allocated at runtime. Arm64's > cpuinfo kobj is also on a percpu variable... I guess that was a > intentional use? > v4: Properly handle err of kobj creation. (gregkh) > v5: Drop invalid kfree > --- > .../ABI/testing/sysfs-devices-system-cpu | 11 +++ > arch/mips/kernel/topology.c | 99 +++++++++++++++++++ > 2 files changed, 110 insertions(+) > > diff --git a/Documentation/ABI/testing/sysfs-devices-system-cpu b/Documentation/ABI/testing/sysfs-devices-system-cpu > index 5bf61881f012..9fdfe2de0f76 100644 > --- a/Documentation/ABI/testing/sysfs-devices-system-cpu > +++ b/Documentation/ABI/testing/sysfs-devices-system-cpu > @@ -512,6 +512,17 @@ Description: information about CPUs heterogeneity. > > cpu_capacity: capacity of cpuX. > > +What: /sys/devices/system/cpu/cpuX/regs/ > + /sys/devices/system/cpu/cpuX/regs/identification/ > + /sys/devices/system/cpu/cpuX/regs/identification/prid > + /sys/devices/system/cpu/cpuX/regs/identification/globalnumber > +Date: October 2022 > +Contact: Linux MIPS Kernel Mailing list <linux-mips@vger.kernel.org> > +Description: MIPS CPU registers > + > + 'identification' directory exposes the Processor ID and Global Number > + registers for identifying model and revision of the CPU. > + > What: /sys/devices/system/cpu/vulnerabilities > /sys/devices/system/cpu/vulnerabilities/meltdown > /sys/devices/system/cpu/vulnerabilities/spectre_v1 > diff --git a/arch/mips/kernel/topology.c b/arch/mips/kernel/topology.c > index 9429d85a4703..80aaaca3cfbc 100644 > --- a/arch/mips/kernel/topology.c > +++ b/arch/mips/kernel/topology.c > @@ -5,6 +5,8 @@ > #include <linux/node.h> > #include <linux/nodemask.h> > #include <linux/percpu.h> > +#include <linux/seq_file.h> > +#include <linux/smp.h> > > static DEFINE_PER_CPU(struct cpu, cpu_devices); > > @@ -26,3 +28,100 @@ static int __init topology_init(void) > } > > subsys_initcall(topology_init); > + > +static struct kobj_type cpuregs_kobj_type = { > + .sysfs_ops = &kobj_sysfs_ops, > +}; > + > +struct cpureg { > + struct kobject kobj; > + struct cpuinfo_mips *info; > +}; > +static DEFINE_PER_CPU(struct cpureg *, cpuregs); > + > +#define kobj_to_cpureg(kobj) container_of(kobj, struct cpureg, kobj) > +#define CPUREGS_ATTR_RO(_name, _field) \ > + static ssize_t _name##_show(struct kobject *kobj, \ > + struct kobj_attribute *attr, char *buf) \ > + { \ > + struct cpuinfo_mips *info = kobj_to_cpureg(kobj)->info; \ > + \ > + return sprintf(buf, "0x%08x\n", info->_field); \ sysfs_emit() please. > + } \ > + static struct kobj_attribute cpuregs_attr_##_name = __ATTR_RO(_name) > + > +CPUREGS_ATTR_RO(prid, processor_id); > +CPUREGS_ATTR_RO(globalnumber, globalnumber); > + > +static struct attribute *cpuregs_id_attrs[] = { > + &cpuregs_attr_prid.attr, > + &cpuregs_attr_globalnumber.attr, > + NULL > +}; > + > +static const struct attribute_group cpuregs_attr_group = { > + .attrs = cpuregs_id_attrs, > + .name = "identification" > +}; > + > +static int cpuregs_cpu_online(unsigned int cpu) > +{ > + int rc; > + struct device *dev; > + struct cpureg *reg; > + > + dev = get_cpu_device(cpu); > + if (!dev) { > + rc = -ENODEV; > + goto out; > + } > + reg = kzalloc(sizeof(struct cpureg), GFP_KERNEL); > + if (!reg) { > + rc = -ENOMEM; > + goto out; > + } > + rc = kobject_init_and_add(®->kobj, &cpuregs_kobj_type, > + &dev->kobj, "regs"); > + if (rc) > + goto out_kobj; > + rc = sysfs_create_group(®->kobj, &cpuregs_attr_group); > + if (rc) > + goto out_kobj; > + > + return 0; > +out_kobj: > + kobject_put(®->kobj); > +out: > + return rc; > +} > + > +static int cpuregs_cpu_offline(unsigned int cpu) > +{ > + struct device *dev; > + struct cpureg *reg = per_cpu(cpuregs, cpu); > + > + dev = get_cpu_device(cpu); > + if (!dev || !reg) > + return -ENODEV; > + if (reg->kobj.parent) { Why are you looking at the parent of a kobject? Why not just always remove the kobject if you have a reference to it now? How does the parent matter? thanks, greg k-h
在2022年10月7日十月 下午3:52，Greg KH写道： [...] >> + >> +static int cpuregs_cpu_offline(unsigned int cpu) >> +{ >> + struct device *dev; >> + struct cpureg *reg = per_cpu(cpuregs, cpu); >> + >> + dev = get_cpu_device(cpu); >> + if (!dev || !reg) >> + return -ENODEV; >> + if (reg->kobj.parent) { > > Why are you looking at the parent of a kobject? Why not just always > remove the kobject if you have a reference to it now? How does the > parent matter? Another dummy copy from Arm64 code... kobject_put should be enough here? Thanks > > thanks, > > greg k-h
On Fri, Oct 07, 2022 at 05:37:34PM +0100, Jiaxun Yang wrote: > > > 在2022年10月7日十月 下午3:52，Greg KH写道： > [...] > >> + > >> +static int cpuregs_cpu_offline(unsigned int cpu) > >> +{ > >> + struct device *dev; > >> + struct cpureg *reg = per_cpu(cpuregs, cpu); > >> + > >> + dev = get_cpu_device(cpu); > >> + if (!dev || !reg) > >> + return -ENODEV; > >> + if (reg->kobj.parent) { > > > > Why are you looking at the parent of a kobject? Why not just always > > remove the kobject if you have a reference to it now? How does the > > parent matter? > > Another dummy copy from Arm64 code... kobject_put should be enough here? Why would it not be enough?
Hi Jiaxun,
I love your patch! Perhaps something to improve:
[auto build test WARNING on linus/master]
[also build test WARNING on v6.0 next-20221007]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Jiaxun-Yang/MIPS-Expose-prid-and-globalnumber-to-sysfs/20221007-221349
base: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 4c86114194e644b6da9107d75910635c9e87179e
reproduce:
# https://github.com/intel-lab-lkp/linux/commit/630e2c4b72708ec85e8abb5a045bcec6e15ec1f0
git remote add linux-review https://github.com/intel-lab-lkp/linux
git fetch --no-tags linux-review Jiaxun-Yang/MIPS-Expose-prid-and-globalnumber-to-sysfs/20221007-221349
git checkout 630e2c4b72708ec85e8abb5a045bcec6e15ec1f0
make menuconfig
# enable CONFIG_COMPILE_TEST, CONFIG_WARN_MISSING_DOCUMENTS, CONFIG_WARN_ABI_ERRORS
make htmldocs
If you fix the issue, kindly add following tag where applicable
| Reported-by: kernel test robot <lkp@intel.com>
All warnings (new ones prefixed by >>):
>> Warning: /sys/devices/system/cpu/cpuX/regs/ is defined 2 times: ./Documentation/ABI/testing/sysfs-devices-system-cpu:486 ./Documentation/ABI/testing/sysfs-devices-system-cpu:514
diff --git a/Documentation/ABI/testing/sysfs-devices-system-cpu b/Documentation/ABI/testing/sysfs-devices-system-cpu index 5bf61881f012..9fdfe2de0f76 100644 --- a/Documentation/ABI/testing/sysfs-devices-system-cpu +++ b/Documentation/ABI/testing/sysfs-devices-system-cpu @@ -512,6 +512,17 @@ Description: information about CPUs heterogeneity. cpu_capacity: capacity of cpuX. +What: /sys/devices/system/cpu/cpuX/regs/ + /sys/devices/system/cpu/cpuX/regs/identification/ + /sys/devices/system/cpu/cpuX/regs/identification/prid + /sys/devices/system/cpu/cpuX/regs/identification/globalnumber +Date: October 2022 +Contact: Linux MIPS Kernel Mailing list <linux-mips@vger.kernel.org> +Description: MIPS CPU registers + + 'identification' directory exposes the Processor ID and Global Number + registers for identifying model and revision of the CPU. + What: /sys/devices/system/cpu/vulnerabilities /sys/devices/system/cpu/vulnerabilities/meltdown /sys/devices/system/cpu/vulnerabilities/spectre_v1 diff --git a/arch/mips/kernel/topology.c b/arch/mips/kernel/topology.c index 9429d85a4703..80aaaca3cfbc 100644 --- a/arch/mips/kernel/topology.c +++ b/arch/mips/kernel/topology.c @@ -5,6 +5,8 @@ #include <linux/node.h> #include <linux/nodemask.h> #include <linux/percpu.h> +#include <linux/seq_file.h> +#include <linux/smp.h> static DEFINE_PER_CPU(struct cpu, cpu_devices); @@ -26,3 +28,100 @@ static int __init topology_init(void) } subsys_initcall(topology_init); + +static struct kobj_type cpuregs_kobj_type = { + .sysfs_ops = &kobj_sysfs_ops, +}; + +struct cpureg { + struct kobject kobj; + struct cpuinfo_mips *info; +}; +static DEFINE_PER_CPU(struct cpureg *, cpuregs); + +#define kobj_to_cpureg(kobj) container_of(kobj, struct cpureg, kobj) +#define CPUREGS_ATTR_RO(_name, _field) \ + static ssize_t _name##_show(struct kobject *kobj, \ + struct kobj_attribute *attr, char *buf) \ + { \ + struct cpuinfo_mips *info = kobj_to_cpureg(kobj)->info; \ + \ + return sprintf(buf, "0x%08x\n", info->_field); \ + } \ + static struct kobj_attribute cpuregs_attr_##_name = __ATTR_RO(_name) + +CPUREGS_ATTR_RO(prid, processor_id); +CPUREGS_ATTR_RO(globalnumber, globalnumber); + +static struct attribute *cpuregs_id_attrs[] = { + &cpuregs_attr_prid.attr, + &cpuregs_attr_globalnumber.attr, + NULL +}; + +static const struct attribute_group cpuregs_attr_group = { + .attrs = cpuregs_id_attrs, + .name = "identification" +}; + +static int cpuregs_cpu_online(unsigned int cpu) +{ + int rc; + struct device *dev; + struct cpureg *reg; + + dev = get_cpu_device(cpu); + if (!dev) { + rc = -ENODEV; + goto out; + } + reg = kzalloc(sizeof(struct cpureg), GFP_KERNEL); + if (!reg) { + rc = -ENOMEM; + goto out; + } + rc = kobject_init_and_add(®->kobj, &cpuregs_kobj_type, + &dev->kobj, "regs"); + if (rc) + goto out_kobj; + rc = sysfs_create_group(®->kobj, &cpuregs_attr_group); + if (rc) + goto out_kobj; + + return 0; +out_kobj: + kobject_put(®->kobj); +out: + return rc; +} + +static int cpuregs_cpu_offline(unsigned int cpu) +{ + struct device *dev; + struct cpureg *reg = per_cpu(cpuregs, cpu); + + dev = get_cpu_device(cpu); + if (!dev || !reg) + return -ENODEV; + if (reg->kobj.parent) { + sysfs_remove_group(®->kobj, &cpuregs_attr_group); + kobject_put(®->kobj); + } + + return 0; +} + +static int __init cpuinfo_regs_init(void) +{ + int ret; + + ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "mips/topology:online", + cpuregs_cpu_online, cpuregs_cpu_offline); + if (ret < 0) { + pr_err("cpuinfo: failed to register hotplug callbacks.\n"); + return ret; + } + return 0; +} + +device_initcall(cpuinfo_regs_init);
Some application would like to know precise model and rev of processor to do errata workaround or optimization. Expose them in sysfs as: /sys/devices/system/cpu/cpuX/regs/identification/prid /sys/devices/system/cpu/cpuX/regs/identification/globalnumber Reusing AArch64 CPU registers directory. Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com> --- v2: Drop static qualifier for kobj (gregkh) v3: Use kzalloc to allocate struct cpuregs. note: When Greg mentioned about static I was thinking about static qualifier of percpu variable. After reading documents again it turns out kobjs should be allocated at runtime. Arm64's cpuinfo kobj is also on a percpu variable... I guess that was a intentional use? v4: Properly handle err of kobj creation. (gregkh) v5: Drop invalid kfree --- .../ABI/testing/sysfs-devices-system-cpu | 11 +++ arch/mips/kernel/topology.c | 99 +++++++++++++++++++ 2 files changed, 110 insertions(+)