linux-kernel - Re: [PATCH v15 11/32] x86,fs/resctrl: Handle events that can be read from any CPU

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] [day] [month] [year] [list]

Message-ID: <33a2ada3-d523-4ed8-9313-74814daded0b@intel.com>
Date: Tue, 9 Dec 2025 13:31:09 -0800
From: Reinette Chatre <reinette.chatre@...el.com>
To: "Luck, Tony" <tony.luck@...el.com>
CC: Fenghua Yu <fenghuay@...dia.com>, Maciej Wieczor-Retman
	<maciej.wieczor-retman@...el.com>, Peter Newman <peternewman@...gle.com>,
	James Morse <james.morse@....com>, Babu Moger <babu.moger@....com>, "Drew
 Fustini" <dfustini@...libre.com>, Dave Martin <Dave.Martin@....com>, Chen Yu
	<yu.c.chen@...el.com>, <x86@...nel.org>, <linux-kernel@...r.kernel.org>,
	<patches@...ts.linux.dev>
Subject: Re: [PATCH v15 11/32] x86,fs/resctrl: Handle events that can be read
 from any CPU

Hi Tony,

On 12/9/25 12:37 PM, Luck, Tony wrote:
> On Tue, Dec 09, 2025 at 11:39:56AM -0800, Reinette Chatre wrote:
>> Hi Tony,
>>
>> On 12/4/25 12:53 PM, Tony Luck wrote:
>>> resctrl assumes that monitor events can only be read from a CPU in the
>>> cpumask_t set of each domain.  This is true for x86 events accessed with an
>>> MSR interface, but may not be true for other access methods such as MMIO.
>>>
>>> Introduce and use flag mon_evt::any_cpu, settable by architecture, that
>>> indicates there are no restrictions on which CPU can read that event.
>>
>> I propose to append (something like): "This flag is not supported by the L3
>> event reading that requires to be run on a CPU that belongs to the L3
>> domain of the event being read."
> 
> Perahps simpler?
> 
> This flag is not supported for RDT_RESOURCE_L3 monitoring events.

Simpler, but missing "why". The changelog starts by highlighting that it is x86 events
accessed with an MSR interface that can "only be read from a CPU in the cpumask_t set
of each domain" but worth highlighting is that __l3_mon_event_count()/__l3_mon_event_count_sum()
is resctrl fs code that, independent from x86, requires to be run on a CPU that belongs to
the L3 domain.

> 
>>
>>> diff --git a/fs/resctrl/monitor.c b/fs/resctrl/monitor.c
>>> index 340b847ab397..081ff659b52c 100644
>>> --- a/fs/resctrl/monitor.c
>>> +++ b/fs/resctrl/monitor.c
>>> @@ -518,10 +518,12 @@ static int __mon_event_count(struct rdtgroup *rdtgrp, struct rmid_read *rr)
>>>  {
>>>  	switch (rr->r->rid) {
>>>  	case RDT_RESOURCE_L3:
>>> -		if (rr->hdr)
>>> +		if (rr->hdr) {
>>> +			WARN_ON_ONCE(rr->evt->any_cpu);
>>
>> I appreciate that this addresses the feedback from v13 but it needs to be adjusted for the
>> code changes made since then. This means that this warning applies to both
>> __l3_mon_event_count() and __l3_mon_event_count_sum() , no?
> 
> I'l move the WARN_ON_ONCE() before the "if"

Thank you.

Reinette