Re: [PATCH] arm64: cpuinfo: Expose MIDR_EL1 and REVIDR_EL1 to sysfs

[Suzuki K Poulose]

On 10/06/16 18:02, Will Deacon wrote:
> On Fri, Jun 10, 2016 at 04:19:44PM +0100, Suzuki K Poulose wrote:
> > From: Steve Capper <steve.capper@xxxxxxxxxx>
> >
> > It can be useful for JIT software to be aware of MIDR_EL1 and
> > REVIDR_EL1 to ascertain the presence of any core errata that could
> > affect codegen.
> >
> > This patch exposes these registers through sysfs:
> >
> > /sys/devices/system/cpu/cpu$ID/identification/midr
> > /sys/devices/system/cpu/cpu$ID/identification/revidr


> > +
> > +#define CPUINFO_ATTR_RO(_name)							\
> > +	static ssize_t show_##_name (struct device *dev,			\
> > +			struct device_attribute *attr, char *buf)		\
> > +	{									\
> > +		struct cpuinfo_arm64 *info = &per_cpu(cpu_data, dev->id);	\
> > +		if (!cpu_present(dev->id))					\
> > +			return -ENODEV;						\
> > +										\
> > +		if (info->reg_midr)						\
> > +			return sprintf(buf, "0x%016x\n", info->reg_##_name);	\
>
> Should this be 0x%08x, as these are 32-bit registers?

Yes. Will change it. As per Mark's comments, I can change them to 64bit in a separate
patch.



> > +
> > +static int __init cpuinfo_regs_init(void)
> > +{
> > +	int cpu, finalcpu, ret;
> > +	struct device *dev;
> > +
> > +	for_each_present_cpu(cpu) {
> > +		dev = get_cpu_device(cpu);
> > +
> > +		if (!dev) {
> > +			ret = -ENODEV;
> > +			break;
> > +		}
> > +
> > +		ret = sysfs_create_group(&dev->kobj, &cpuregs_attr_group);
> > +		if (ret)
> > +			break;
> > +	}
> > +
> > +	if (!ret)
> > +		return 0;
> > +	/*
> > +	 * We were unable to put down sysfs groups for all the CPUs, revert
> > +	 * all the groups we have placed down s.t. none are visible.
> > +	 * Otherwise we could give a misleading picture of what's present.
> > +	 */
> > +	finalcpu = cpu;
> > +	for_each_present_cpu(cpu) {
> > +		if (cpu == finalcpu)
> > +			break;
> > +		dev = get_cpu_device(cpu);
> > +		if (dev)
> > +			sysfs_remove_group(&dev->kobj, &cpuregs_attr_group);
> > +	}
>
> Can CPUs be removed from underneath us using unregister_cpu? If so, I

Yes. Good point. Though this is done at early boot, nobody prevents
an unregister_cpu(). The safer way would be to wrap the code in
cpu_hotplug_disable()...enable().

I will respin it.



> don't think we should assume that get_cpu_device will succeed in the
> same places for both the loops.


Thanks
Suzuki