linux-kernel - Re: [PATCH 1/2] x86/resctrl: Add io_alloc_min_cbm

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: <f5ac9fea-1c57-419a-befd-1caf946d6005@intel.com>
Date: Mon, 10 Nov 2025 13:40:32 -0800
From: Reinette Chatre <reinette.chatre@...el.com>
To: "Luck, Tony" <tony.luck@...el.com>, Babu Moger <bmoger@....com>, "Aaron
 Tomlin" <atomlin@...mlin.com>
CC: "Dave.Martin@....com" <Dave.Martin@....com>, "james.morse@....com"
	<james.morse@....com>, "babu.moger@....com" <babu.moger@....com>,
	"tglx@...utronix.de" <tglx@...utronix.de>, "mingo@...hat.com"
	<mingo@...hat.com>, "bp@...en8.de" <bp@...en8.de>,
	"dave.hansen@...ux.intel.com" <dave.hansen@...ux.intel.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 1/2] x86/resctrl: Add io_alloc_min_cbm_all interface for
 CBM reset

Hi Tony,

On 11/10/25 11:56 AM, Luck, Tony wrote:
>>> # echo "L2:*=fff" > schemata
>>>
>>> would work,. But
>>>
>>> # echo "L2:*=ffff" > schemata
>>>
>>> would try to set unimplemented bits on some cores and would fail.
>>
>>
>> I would consider this a user error, as the user is expected to know the
>> supported value for the domain.
>> This situation can occur even now — we simply report the error and exit.
> 
> Babu
> 
> Maybe it was a poor explanation on my part.
> 
> On a hybrid P-core/E-core system with different L2 cache topology schemata
> may look like this (8 L2 domains of one type, 4 L2 domains of other type.
> 
> $ cat schemata
> L2:0=ffff;1=ffff;2=ffff;3=ffff;4=ffff;5=ffff;6=ffff;7=ffff;8=7f;9=7f;10=7f;11=7f
> 
> The proposed wildcard syntax is only useful to set all domains to a value
> that is legal for all domains. It cannot be used for the "reset back to defaults"
> case because different domains have different defaults.

Disabling and re-enabling of io_alloc may be substitute for "reset back to defaults".

The '*' syntax would be useful to initialize domains to minimal allocations ...
assuming all cache instances support the same minimum. 

Would it be an issue if user attempts to assign a value that is not supported to
all domains? resctrl could fail on first invalid bitmask and the last_cmd_status can
then be expected to indicate the details of error.

Reinette