linux-kernel - Re: [PATCH v5] perf/x86/amd/power: Add AMD accumulated power reporting mechanism

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <alpine.DEB.2.11.1602291125510.3638@nanos>
Date:	Mon, 29 Feb 2016 12:30:47 +0100 (CET)
From:	Thomas Gleixner <tglx@...utronix.de>
To:	Huang Rui <ray.huang@....com>
cc:	Borislav Petkov <bp@...e.de>,
	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...nel.org>,
	Andy Lutomirski <luto@...capital.net>,
	Robert Richter <rric@...nel.org>,
	Jacob Shin <jacob.w.shin@...il.com>,
	Arnaldo Carvalho de Melo <acme@...nel.org>,
	Kan Liang <kan.liang@...el.com>, linux-kernel@...r.kernel.org,
	spg_linux_kernel@....com, x86@...nel.org,
	Suravee Suthikulpanit <suravee.suthikulpanit@....com>,
	Aravind Gopalakrishnan <Aravind.Gopalakrishnan@....com>,
	Borislav Petkov <bp@...en8.de>,
	Fengguang Wu <fengguang.wu@...el.com>,
	Guenter Roeck <linux@...ck-us.net>
Subject: Re: [PATCH v5] perf/x86/amd/power: Add AMD accumulated power reporting
 mechanism

On Tue, 1 Mar 2016, Huang Rui wrote:
> On Fri, Feb 26, 2016 at 11:18:28AM +0100, Thomas Gleixner wrote:
> > > +static int __init amd_power_pmu_init(void)
> > > +{
> > > +	int i, ret;
> > > +	u64 tmp;
> > > +
> > > +	if (!x86_match_cpu(cpu_match))
> > > +		return 0;
> > > +
> > > +	if (!boot_cpu_has(X86_FEATURE_ACC_POWER))
> > > +		return -ENODEV;
> > > +
> > > +	cores_per_cu = amd_get_cores_per_cu();
> > > +	cu_num = boot_cpu_data.x86_max_cores / cores_per_cu;
> > 
> > Please use the new package management functions which are on the way to tip.
> > 
> Can you give me some hints?

http://git.kernel.org/tip/1f12e32f4cd5243ae46d8b933181be0d022c6793

topology_max_packages() is what you want.
 
> > You cannot issue the CPU STARTING callback on present, but not online cpus.
> > 
> 
> Do you mean we should change for_each_present_cpu to
> for_each_online_cpu?
> 
> My orignal intent here, it's to allocate data structures of
> "power_pmu_masks" for each core.

Sure, you can allocate stuff for present cpus, but you cannot call the
CPU_STARTING callback for offline cpus.

> But now, I think we needn't below codes, right?

-ENOPARSE
 
> for_each_online_cpu(i)
>         power_cpu_init(i);

You still need that.

> > > +
> > > +	__perf_cpu_notifier(power_cpu_notifier);
> > > +
> > > +	ret = perf_pmu_register(&pmu_class, "power", -1);
> > > +	if (WARN_ON(ret)) {
> > > +		pr_warn("AMD Power PMU registration failed\n");
> > 
> > 
> > So that leaves the notifier installed and leaks all the allocated memory.
> > 
> 
> OK, do you mean "issue CPU_STARTING callback on present cpus" will
> cause the memory leak here? Could you please explain more?

Oh well, can you actually read your own code? You allocate a gazillion of
memory in power_cpu_prepare() and on error you just leak it.

Thanks,

	tglx