linux-kernel - Re: [PATCH v15-RFC 1/8] x86/resctrl: Split the RDT_RESOURCE

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: <ZcZyqs5hnQqZ5ZV0@agluck-desk3>
Date: Fri, 9 Feb 2024 10:44:58 -0800
From: Tony Luck <tony.luck@...el.com>
To: "Moger, Babu" <babu.moger@....com>
Cc: Fenghua Yu <fenghua.yu@...el.com>,
	Reinette Chatre <reinette.chatre@...el.com>,
	Peter Newman <peternewman@...gle.com>,
	Jonathan Corbet <corbet@....net>,
	Shuah Khan <skhan@...uxfoundation.org>, x86@...nel.org,
	Shaopeng Tan <tan.shaopeng@...itsu.com>,
	James Morse <james.morse@....com>,
	Jamie Iles <quic_jiles@...cinc.com>,
	Randy Dunlap <rdunlap@...radead.org>,
	Drew Fustini <dfustini@...libre.com>, linux-kernel@...r.kernel.org,
	linux-doc@...r.kernel.org, patches@...ts.linux.dev
Subject: Re: [PATCH v15-RFC 1/8] x86/resctrl: Split the RDT_RESOURCE_L3
 resource

On Fri, Feb 09, 2024 at 09:28:16AM -0600, Moger, Babu wrote:
> >  enum resctrl_res_level {
> > +	RDT_RESOURCE_L3_MON,
> >  	RDT_RESOURCE_L3,
> 
> How about?
> RDT_RESOURCE_L3,
> RDT_RESOURCE_L3_MON,

Does the order matter? I put the L3_MON one first because historically
cache occupancy was the first resctrl tool. But if you have a better
argument for the order, then I can change it.

> >  	RDT_RESOURCE_L2,
> >  	RDT_RESOURCE_MBA,
> > diff --git a/arch/x86/kernel/cpu/resctrl/core.c b/arch/x86/kernel/cpu/resctrl/core.c
> > index aa9810a64258..c50f55d7790e 100644
> > --- a/arch/x86/kernel/cpu/resctrl/core.c
> > +++ b/arch/x86/kernel/cpu/resctrl/core.c
> > @@ -60,6 +60,16 @@ mba_wrmsr_amd(struct rdt_domain *d, struct msr_param *m,
> >  #define domain_init(id) LIST_HEAD_INIT(rdt_resources_all[id].r_resctrl.domains)
> >  
> >  struct rdt_hw_resource rdt_resources_all[] = {
> > +	[RDT_RESOURCE_L3_MON] =
> > +	{
> > +		.r_resctrl = {
> > +			.rid			= RDT_RESOURCE_L3_MON,
> > +			.name			= "L3",
> 
> L3_MON ?

That was my first choice too. But I found:

$ ls /sys/fs/resctrl/info
L3  L3_MON_MON  last_cmd_status  MB

This would be easy to fix ... just change this code to not append
an extra "_MON" to the directory name:

        for_each_mon_capable_rdt_resource(r) {
                fflags = r->fflags | RFTYPE_MON_INFO;
                sprintf(name, "%s_MON", r->name);
                ret = rdtgroup_mkdir_info_resdir(r, name, fflags);
                if (ret)
                        goto out_destroy;
        }

But I also saw this:
$ ls /sys/fs/resctrl/mon_data/
mon_L3_MON_00  mon_L3_MON_01

which didn't seem to have an easy fix. So I took the easy route and left
the ".name" field as "L3_MON".

-Tony