From: "Moger, Babu" <bmoger@amd.com>
To: Reinette Chatre <reinette.chatre@intel.com>,
Babu Moger <babu.moger@amd.com>,
corbet@lwn.net, fenghua.yu@intel.com, tglx@linutronix.de,
mingo@redhat.com, bp@alien8.de, dave.hansen@linux.intel.com
Cc: x86@kernel.org, hpa@zytor.com, paulmck@kernel.org,
rdunlap@infradead.org, tj@kernel.org, peterz@infradead.org,
yanjiewtw@gmail.com, kim.phillips@amd.com,
lukas.bulwahn@gmail.com, seanjc@google.com, jmattson@google.com,
leitao@debian.org, jpoimboe@kernel.org,
rick.p.edgecombe@intel.com, kirill.shutemov@linux.intel.com,
jithu.joseph@intel.com, kai.huang@intel.com,
kan.liang@linux.intel.com, daniel.sneddon@linux.intel.com,
pbonzini@redhat.com, sandipan.das@amd.com,
ilpo.jarvinen@linux.intel.com, peternewman@google.com,
maciej.wieczor-retman@intel.com, linux-doc@vger.kernel.org,
linux-kernel@vger.kernel.org, eranian@google.com,
james.morse@arm.com
Subject: Re: [PATCH v8 21/25] x86/resctrl: Introduce the interface to switch between monitor modes
Date: Fri, 18 Oct 2024 10:13:50 -0500 [thread overview]
Message-ID: <51b47b50-1b83-6565-e7da-14d4de0b15c8@amd.com> (raw)
In-Reply-To: <4d49182f-6185-49df-b9bf-29fbd17dccf6@intel.com>
Hi Reinette,
On 10/15/2024 10:36 PM, Reinette Chatre wrote:
> Hi Babu,
>
> On 10/9/24 10:39 AM, Babu Moger wrote:
>> Introduce interface to switch between mbm_cntr_assign and default modes.
>>
>> $ cat /sys/fs/resctrl/info/L3_MON/mbm_assign_mode
>> [mbm_cntr_assign]
>> default
>>
>> To enable the "mbm_cntr_assign" mode:
>> $ echo "mbm_cntr_assign" > /sys/fs/resctrl/info/L3_MON/mbm_assign_mode
>>
>> To enable the default monitoring mode:
>> $ echo "default" > /sys/fs/resctrl/info/L3_MON/mbm_assign_mode
>>
>> MBM event counters will reset when mbm_assign_mode is changed.
>>
>> Signed-off-by: Babu Moger <babu.moger@amd.com>
>> ---
>
> ...
>
>> ---
>> Documentation/arch/x86/resctrl.rst | 15 ++++++
>> arch/x86/kernel/cpu/resctrl/rdtgroup.c | 75 +++++++++++++++++++++++++-
>> 2 files changed, 89 insertions(+), 1 deletion(-)
>>
>> diff --git a/Documentation/arch/x86/resctrl.rst b/Documentation/arch/x86/resctrl.rst
>> index 99ee9c87952b..d9574078f735 100644
>> --- a/Documentation/arch/x86/resctrl.rst
>> +++ b/Documentation/arch/x86/resctrl.rst
>> @@ -291,6 +291,21 @@ with the following files:
>> that case reading the mbm_total_bytes and mbm_local_bytes may report
>> 'Unavailable' if there is no counter associated with that group.
>>
>> + * To enable "mbm_cntr_assign" mode:
>> + ::
>> +
>> + # echo "mbm_cntr_assign" > /sys/fs/resctrl/info/L3_MON/mbm_assign_mode
>
> extra spaces
Sure.
>
>> +
>> + * To enable default monitoring mode:
>> + ::
>> +
>> + # echo "default" > /sys/fs/resctrl/info/L3_MON/mbm_assign_mode
>
> extra spaces
Sure
>> +
>> + The MBM events (mbm_total_bytes and/or mbm_local_bytes) associated counters
>
> I did ask you not to copy the text verbatim
> https://lore.kernel.org/all/b38c93bf-4650-45d1-9aca-8b4c4d425886@intel.com/
Outch Yea.
The MBM events (mbm_total_bytes and/or mbm_local_bytes) associated with
counters may reset when mbm_assign_mode is changed.
>
>> + may reset when the mode is changed. Moving to mbm_cntr_assign mode will
>> + require users to assign the counters to the events. Otherwise, the MBM
>
> "will require" -> "require"
>
>> + event counters will return "Unassigned" when read.
>> +
>> "num_mbm_cntrs":
>> The number of monitoring counters available for assignment when the
>> architecture supports mbm_cntr_assign mode.
>> diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
>> index cb2c60c0319e..88eda3cf5c82 100644
>> --- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
>> +++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
>> @@ -888,6 +888,78 @@ static int rdtgroup_mbm_assign_mode_show(struct kernfs_open_file *of,
>> return 0;
>> }
>>
>> +static void mbm_cntr_reset(struct rdt_resource *r)
>> +{
>> + struct rdtgroup *prgrp, *crgrp;
>> + struct rdt_mon_domain *dom;
>> +
>> + /*
>> + * Hardware counters will reset after switching the monitor mode.
>> + * Reset the architectural state so that reading of hardware
>> + * counter is not considered as an overflow in the next update.
>> + * Also reset the domain counter bitmap.
>> + */
>> + list_for_each_entry(dom, &r->mon_domains, hdr.list) {
>> + bitmap_zero(dom->mbm_cntr_map, r->mon.num_mbm_cntrs);
>> + resctrl_arch_reset_rmid_all(r, dom);
>> + }
>> +
>> + /* Reset global MBM counter map */
>> + bitmap_fill(r->mon.mbm_cntr_free_map, r->mon.num_mbm_cntrs);
>> +
>> + /* Reset the cntr_id's for all the monitor groups */
>> + list_for_each_entry(prgrp, &rdt_all_groups, rdtgroup_list) {
>> + prgrp->mon.cntr_id[0] = MON_CNTR_UNSET;
>> + prgrp->mon.cntr_id[1] = MON_CNTR_UNSET;
>> + list_for_each_entry(crgrp, &prgrp->mon.crdtgrp_list,
>> + mon.crdtgrp_list) {
>> + crgrp->mon.cntr_id[0] = MON_CNTR_UNSET;
>> + crgrp->mon.cntr_id[1] = MON_CNTR_UNSET;
>> + }
>
> Please use MBM_EVENT_ARRAY_INDEX
Sure.
>
>> + }
>> +}
>> +
>> +static ssize_t rdtgroup_mbm_assign_mode_write(struct kernfs_open_file *of,
>> + char *buf, size_t nbytes, loff_t off)
>> +{
>> + struct rdt_resource *r = of->kn->parent->priv;
>> + int ret = 0;
>> + bool enable;
>> +
>> + /* Valid input requires a trailing newline */
>> + if (nbytes == 0 || buf[nbytes - 1] != '\n')
>> + return -EINVAL;
>> +
>> + buf[nbytes - 1] = '\0';
>> +
>> + cpus_read_lock();
>> + mutex_lock(&rdtgroup_mutex);
>> +
>> + rdt_last_cmd_clear();
>> +
>> + if (!strcmp(buf, "default")) {
>> + enable = 0;
>> + } else if (!strcmp(buf, "mbm_cntr_assign")) {
>> + enable = 1;
>> + } else {
>> + ret = -EINVAL;
>> + rdt_last_cmd_puts("Unsupported assign mode\n");
>> + goto write_exit;
>> + }
>
> Please keep two things in mind:
> * this file is always accessible, whether platform supports assignable
> counters or not.
> * this is resctrl fs code.
>
> So, considering above, how should user interpret the "Unsupported assign mode"?
> Shouldn't it also return this error if a user attempts to enable
> "mbm_cntr_assign" on a platform that does not support this mode?
>
>> +
>> + if (enable != resctrl_arch_mbm_cntr_assign_enabled(r)) {
>
> resctrl_arch_mbm_cntr_assign_enabled() returns true if mbm_cntr_assign
> mode is enabled, but when it returns false it could mean different things:
> platform supports mbm_cntr_assign mode, but it is disabled, or platform
> does not support mbm_cntr_assign mode.
>
> resctrl fs should not rely on all archs to duplicate the all the checking done
> in resctrl_arch_mbm_cntr_assign_set(). It should never ask arch to enable a mode
> that it knows the platform is not capable of.
Agree. This should take care of it. Good catch.
if (!strcmp(buf, "default")) {
enable = 0;
} else if (!strcmp(buf, "mbm_cntr_assign")) {
if (r->mon.mbm_cntr_assignable) {
enable = 1;
} else {
ret = -EINVAL;
rdt_last_cmd_puts("mbm_cntr_assign mode is not supported\n");
goto write_exit;
}
} else {
ret = -EINVAL;
rdt_last_cmd_puts("Unsupported assign mode\n");
goto write_exit;
}
>
>> + ret = resctrl_arch_mbm_cntr_assign_set(r, enable);
>> + if (!ret)
>> + mbm_cntr_reset(r);
>> + }
>> +
>> +write_exit:
>> + mutex_unlock(&rdtgroup_mutex);
>> + cpus_read_unlock();
>> +
>> + return ret ?: nbytes;
>> +}
>> +
>> static int rdtgroup_num_mbm_cntrs_show(struct kernfs_open_file *of,
>> struct seq_file *s, void *v)
>> {
>> @@ -2115,9 +2187,10 @@ static struct rftype res_common_files[] = {
>> },
>> {
>> .name = "mbm_assign_mode",
>> - .mode = 0444,
>> + .mode = 0644,
>> .kf_ops = &rdtgroup_kf_single_ops,
>> .seq_show = rdtgroup_mbm_assign_mode_show,
>> + .write = rdtgroup_mbm_assign_mode_write,
>> .fflags = RFTYPE_MON_INFO,
>> },
>> {
>
> Reinette
>
>
--
- Babu Moger
next prev parent reply other threads:[~2024-10-18 15:13 UTC|newest]
Thread overview: 124+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-09 17:39 [PATCH v8 00/25] x86/resctrl : Support AMD Assignable Bandwidth Monitoring Counters (ABMC) Babu Moger
2024-10-09 17:39 ` [PATCH v8 01/25] x86/cpufeatures: Add support for " Babu Moger
2024-10-09 17:39 ` [PATCH v8 02/25] x86/resctrl: Add ABMC feature in the command line options Babu Moger
2024-10-16 3:06 ` Reinette Chatre
2024-10-09 17:39 ` [PATCH v8 03/25] x86/resctrl: Consolidate monitoring related data from rdt_resource Babu Moger
2024-10-09 17:39 ` [PATCH v8 04/25] x86/resctrl: Detect Assignable Bandwidth Monitoring feature details Babu Moger
2024-10-16 3:06 ` Reinette Chatre
2024-10-09 17:39 ` [PATCH v8 05/25] x86/resctrl: Introduce resctrl_file_fflags_init() to initialize fflags Babu Moger
2024-10-09 17:39 ` [PATCH v8 06/25] x86/resctrl: Add support to enable/disable AMD ABMC feature Babu Moger
2024-10-11 18:14 ` Tony Luck
2024-10-11 20:53 ` Moger, Babu
2024-10-16 3:07 ` Reinette Chatre
2024-10-09 17:39 ` [PATCH v8 07/25] x86/resctrl: Introduce the interface to display monitor mode Babu Moger
2024-10-09 22:42 ` Tony Luck
2024-10-10 14:54 ` Moger, Babu
2024-10-10 15:07 ` Luck, Tony
2024-10-10 15:30 ` Moger, Babu
2024-10-10 16:02 ` Luck, Tony
2024-10-11 22:24 ` Reinette Chatre
2024-10-14 15:16 ` Moger, Babu
2024-10-16 3:12 ` Reinette Chatre
2024-10-16 15:57 ` Moger, Babu
2024-10-16 16:25 ` Reinette Chatre
2024-10-09 17:39 ` [PATCH v8 08/25] x86/resctrl: Introduce interface to display number of monitoring counters Babu Moger
2024-10-09 22:49 ` Tony Luck
2024-10-10 15:12 ` Moger, Babu
2024-10-10 15:58 ` Luck, Tony
2024-10-10 16:57 ` Moger, Babu
2024-10-10 17:08 ` Luck, Tony
2024-10-10 18:36 ` Moger, Babu
2024-10-10 18:57 ` Luck, Tony
2024-10-10 20:32 ` Moger, Babu
2024-10-11 17:44 ` Tony Luck
2024-10-11 20:49 ` Moger, Babu
2024-10-11 21:36 ` Tony Luck
2024-10-14 16:46 ` Reinette Chatre
2024-10-14 17:20 ` Moger, Babu
2024-10-14 17:49 ` Luck, Tony
2024-10-14 19:21 ` Moger, Babu
2024-10-14 19:51 ` Luck, Tony
2024-10-14 20:05 ` Reinette Chatre
2024-10-14 20:32 ` Moger, Babu
2024-10-24 17:29 ` Moger, Babu
2024-10-24 17:37 ` Luck, Tony
2024-10-25 20:31 ` Moger, Babu
2024-10-14 16:59 ` Reinette Chatre
2024-10-14 19:23 ` Moger, Babu
2024-10-14 16:25 ` Reinette Chatre
2024-10-14 17:46 ` Moger, Babu
2024-10-14 18:30 ` Reinette Chatre
2024-10-14 18:51 ` Moger, Babu
2024-10-09 17:39 ` [PATCH v8 09/25] x86/resctrl: Add __init attribute to dom_data_init() Babu Moger
2024-10-16 3:13 ` Reinette Chatre
2024-10-16 17:32 ` Moger, Babu
2024-10-16 18:55 ` Reinette Chatre
2024-10-16 20:18 ` Moger, Babu
2024-10-09 17:39 ` [PATCH v8 10/25] x86/resctrl: Introduce bitmap mbm_cntr_free_map to track assignable counters Babu Moger
2024-10-16 3:14 ` Reinette Chatre
2024-10-17 16:55 ` Moger, Babu
2024-10-09 17:39 ` [PATCH v8 11/25] x86/resctrl: Introduce mbm_total_cfg and mbm_local_cfg in struct rdt_hw_mon_domain Babu Moger
2024-10-16 3:15 ` Reinette Chatre
2024-10-09 17:39 ` [PATCH v8 12/25] x86/resctrl: Remove MSR reading of event configuration value Babu Moger
2024-10-16 3:16 ` Reinette Chatre
2024-10-17 17:59 ` Moger, Babu
2024-10-09 17:39 ` [PATCH v8 13/25] x86/resctrl: Introduce mbm_cntr_map to track assignable counters at domain Babu Moger
2024-10-16 3:19 ` Reinette Chatre
2024-10-09 17:39 ` [PATCH v8 14/25] x86/resctrl: Add data structures and definitions for ABMC assignment Babu Moger
2024-10-16 3:21 ` Reinette Chatre
2024-10-17 18:52 ` Moger, Babu
2024-10-17 21:13 ` Reinette Chatre
2024-10-17 23:02 ` Moger, Babu
2024-10-09 17:39 ` [PATCH v8 15/25] x86/resctrl: Introduce cntr_id in mongroup for assignments Babu Moger
2024-10-16 3:22 ` Reinette Chatre
2024-10-17 19:19 ` Moger, Babu
2024-10-09 17:39 ` [PATCH v8 16/25] x86/resctrl: Implement resctrl_arch_config_cntr() to assign a counter with ABMC Babu Moger
2024-10-16 3:23 ` Reinette Chatre
2024-10-17 22:44 ` Moger, Babu
2024-10-09 17:39 ` [PATCH v8 17/25] x86/resctrl: Add the interface to assign/update counter assignment Babu Moger
2024-10-16 3:25 ` Reinette Chatre
2024-10-17 22:56 ` Moger, Babu
2024-10-18 15:59 ` Reinette Chatre
2024-10-21 14:40 ` Moger, Babu
2024-10-21 15:31 ` Reinette Chatre
2024-10-22 1:15 ` Moger, Babu
2024-10-09 17:39 ` [PATCH v8 18/25] x86/resctrl: Add the interface to unassign a MBM counter Babu Moger
2024-10-16 3:29 ` Reinette Chatre
2024-10-17 23:11 ` Moger, Babu
2024-10-18 16:06 ` Reinette Chatre
2024-10-09 17:39 ` [PATCH v8 19/25] x86/resctrl: Auto assign/unassign counters when mbm_cntr_assign is enabled Babu Moger
2024-10-11 17:17 ` Tony Luck
2024-10-11 21:17 ` Moger, Babu
2024-10-11 21:33 ` Luck, Tony
2024-10-14 15:43 ` Moger, Babu
2024-10-14 16:18 ` Luck, Tony
2024-10-14 16:35 ` Moger, Babu
2024-10-15 2:39 ` Reinette Chatre
2024-10-15 15:43 ` Moger, Babu
2024-10-15 16:57 ` Luck, Tony
2024-10-15 17:18 ` Reinette Chatre
2024-10-15 20:42 ` Moger, Babu
2024-10-16 3:30 ` Reinette Chatre
2024-10-18 14:22 ` Moger, Babu
2024-10-09 17:39 ` [PATCH v8 20/25] x86/resctrl: Report "Unassigned" for MBM events in mbm_cntr_assign mode Babu Moger
2024-10-11 17:23 ` Tony Luck
2024-10-11 21:21 ` Moger, Babu
2024-10-16 3:31 ` Reinette Chatre
2024-10-18 14:31 ` Moger, Babu
2024-10-09 17:39 ` [PATCH v8 21/25] x86/resctrl: Introduce the interface to switch between monitor modes Babu Moger
2024-10-16 3:36 ` Reinette Chatre
2024-10-18 15:13 ` Moger, Babu [this message]
2024-10-09 17:39 ` [PATCH v8 22/25] x86/resctrl: Configure mbm_cntr_assign mode if supported Babu Moger
2024-10-09 17:39 ` [PATCH v8 23/25] x86/resctrl: Update assignments on event configuration changes Babu Moger
2024-10-16 3:40 ` Reinette Chatre
2024-10-18 15:50 ` Moger, Babu
2024-10-09 17:39 ` [PATCH v8 24/25] x86/resctrl: Introduce interface to list assignment states of all the groups Babu Moger
2024-10-16 3:40 ` Reinette Chatre
2024-10-21 14:56 ` Moger, Babu
2024-10-09 17:39 ` [PATCH v8 25/25] x86/resctrl: Introduce interface to modify assignment states of " Babu Moger
2024-10-16 3:43 ` Reinette Chatre
2024-10-21 17:04 ` Moger, Babu
2024-10-21 17:20 ` Reinette Chatre
2024-10-22 1:12 ` Moger, Babu
2024-10-16 3:05 ` [PATCH v8 00/25] x86/resctrl : Support AMD Assignable Bandwidth Monitoring Counters (ABMC) Reinette Chatre
2024-10-21 17:09 ` Moger, Babu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=51b47b50-1b83-6565-e7da-14d4de0b15c8@amd.com \
--to=bmoger@amd.com \
--cc=babu.moger@amd.com \
--cc=bp@alien8.de \
--cc=corbet@lwn.net \
--cc=daniel.sneddon@linux.intel.com \
--cc=dave.hansen@linux.intel.com \
--cc=eranian@google.com \
--cc=fenghua.yu@intel.com \
--cc=hpa@zytor.com \
--cc=ilpo.jarvinen@linux.intel.com \
--cc=james.morse@arm.com \
--cc=jithu.joseph@intel.com \
--cc=jmattson@google.com \
--cc=jpoimboe@kernel.org \
--cc=kai.huang@intel.com \
--cc=kan.liang@linux.intel.com \
--cc=kim.phillips@amd.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=leitao@debian.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lukas.bulwahn@gmail.com \
--cc=maciej.wieczor-retman@intel.com \
--cc=mingo@redhat.com \
--cc=paulmck@kernel.org \
--cc=pbonzini@redhat.com \
--cc=peternewman@google.com \
--cc=peterz@infradead.org \
--cc=rdunlap@infradead.org \
--cc=reinette.chatre@intel.com \
--cc=rick.p.edgecombe@intel.com \
--cc=sandipan.das@amd.com \
--cc=seanjc@google.com \
--cc=tglx@linutronix.de \
--cc=tj@kernel.org \
--cc=x86@kernel.org \
--cc=yanjiewtw@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).