linux-kernel - Re: [PATCH v18 00/33] x86,fs/resctrl: Support AMD Assignable Bandwidth Monitoring Counters (ABMC)

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 for Android: free password hash cracker in your pocket

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

Message-ID: <be1db2e1-b6e2-425c-81b0-00c8227b298d@amd.com>
Date: Tue, 16 Sep 2025 12:00:10 -0500
From: "Moger, Babu" <babu.moger@....com>
To: Reinette Chatre <reinette.chatre@...el.com>,
 Borislav Petkov <bp@...en8.de>
Cc: corbet@....net, tony.luck@...el.com, Dave.Martin@....com,
 james.morse@....com, tglx@...utronix.de, mingo@...hat.com,
 dave.hansen@...ux.intel.com, x86@...nel.org, hpa@...or.com, kas@...nel.org,
 rick.p.edgecombe@...el.com, akpm@...ux-foundation.org, paulmck@...nel.org,
 frederic@...nel.org, pmladek@...e.com, rostedt@...dmis.org, kees@...nel.org,
 arnd@...db.de, fvdl@...gle.com, seanjc@...gle.com, thomas.lendacky@....com,
 pawan.kumar.gupta@...ux.intel.com, perry.yuan@....com,
 manali.shukla@....com, sohil.mehta@...el.com, xin@...or.com,
 Neeraj.Upadhyay@....com, peterz@...radead.org, tiala@...rosoft.com,
 mario.limonciello@....com, dapeng1.mi@...ux.intel.com, michael.roth@....com,
 chang.seok.bae@...el.com, linux-doc@...r.kernel.org,
 linux-kernel@...r.kernel.org, linux-coco@...ts.linux.dev,
 kvm@...r.kernel.org, peternewman@...gle.com, eranian@...gle.com,
 gautham.shenoy@....com
Subject: Re: [PATCH v18 00/33] x86,fs/resctrl: Support AMD Assignable
 Bandwidth Monitoring Counters (ABMC)

Hi Boris/Reinette,

On 9/15/25 16:07, Reinette Chatre wrote:
> Hi Boris,
> 
> On 9/15/25 4:25 AM, Borislav Petkov wrote:
>> On Fri, Sep 05, 2025 at 04:33:59PM -0500, Babu Moger wrote:
>>>  .../admin-guide/kernel-parameters.txt         |    2 +-
>>>  Documentation/filesystems/resctrl.rst         |  325 ++++++
>>>  MAINTAINERS                                   |    1 +
>>>  arch/x86/include/asm/cpufeatures.h            |    1 +
>>>  arch/x86/include/asm/msr-index.h              |    2 +
>>>  arch/x86/include/asm/resctrl.h                |   16 -
>>>  arch/x86/kernel/cpu/resctrl/core.c            |   81 +-
>>>  arch/x86/kernel/cpu/resctrl/internal.h        |   56 +-
>>>  arch/x86/kernel/cpu/resctrl/monitor.c         |  248 +++-
>>>  arch/x86/kernel/cpu/scattered.c               |    1 +
>>>  fs/resctrl/ctrlmondata.c                      |   26 +-
>>>  fs/resctrl/internal.h                         |   58 +-
>>>  fs/resctrl/monitor.c                          | 1008 ++++++++++++++++-
>>>  fs/resctrl/rdtgroup.c                         |  252 ++++-
>>>  include/linux/resctrl.h                       |  148 ++-
>>>  include/linux/resctrl_types.h                 |   18 +-
>>>  16 files changed, 2019 insertions(+), 224 deletions(-)
>>
>> Ok, I've rebased and pushed out the pile into tip:x86/cache.
>>
>> Please run it one more time to make sure all is good.
> 
> Thank you very much.
> 
> I successfully completed as much testing as I can do without the hardware that has
> the feature. Will leave the actual feature sanity check to Babu.

I’ve completed the overnight test runs, and everything is working as expected.

However, I discovered an issue with automatic counter assignment that I
introduced during the v17 → v18 rebase. This is my bad. The automatic
merge mistakenly placed the code snippet in rdtgroup_unassign_cntrs()
instead of under mbm_assign_on_mkdir.

Here is the patch snippet. Will send the fix patch separately.

 diff --git a/fs/resctrl/monitor.c b/fs/resctrl/monitor.c
index 50c24460d992..4076336fbba6 100644
--- a/fs/resctrl/monitor.c
+++ b/fs/resctrl/monitor.c
@@ -1200,7 +1200,8 @@ void rdtgroup_assign_cntrs(struct rdtgroup *rdtgrp)
 {
 	struct rdt_resource *r = resctrl_arch_get_resource(RDT_RESOURCE_L3);

-	if (!r->mon_capable || !resctrl_arch_mbm_cntr_assign_enabled(r))
+	if (!r->mon_capable || !resctrl_arch_mbm_cntr_assign_enabled(r) ||
+	    !r->mon.mbm_assign_on_mkdir)
 		return;

 	if (resctrl_is_mon_event_enabled(QOS_L3_MBM_TOTAL_EVENT_ID))
@@ -1258,8 +1259,7 @@ void rdtgroup_unassign_cntrs(struct rdtgroup *rdtgrp)
 {
 	struct rdt_resource *r = resctrl_arch_get_resource(RDT_RESOURCE_L3);

-	if (!r->mon_capable || !resctrl_arch_mbm_cntr_assign_enabled(r) ||
-	    !r->mon.mbm_assign_on_mkdir)
+	if (!r->mon_capable || !resctrl_arch_mbm_cntr_assign_enabled(r))
 		return;

 	if (resctrl_is_mon_event_enabled(QOS_L3_MBM_TOTAL_EVENT_ID))
-- 
2.34.1

-- 
Thanks
Babu Moger