mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Moger, Babu" <babu.moger@amd.com>
To: Reinette Chatre <reinette.chatre@intel.com>,
	corbet@lwn.net, tglx@linutronix.de, mingo@redhat.com,
	bp@alien8.de
Cc: fenghua.yu@intel.com, dave.hansen@linux.intel.com,
	x86@kernel.org, hpa@zytor.com, paulmck@kernel.org,
	akpm@linux-foundation.org, quic_neeraju@quicinc.com,
	rdunlap@infradead.org, damien.lemoal@opensource.wdc.com,
	songmuchun@bytedance.com, peterz@infradead.org,
	jpoimboe@kernel.org, pbonzini@redhat.com,
	chang.seok.bae@intel.com, pawan.kumar.gupta@linux.intel.com,
	jmattson@google.com, daniel.sneddon@linux.intel.com,
	sandipan.das@amd.com, tony.luck@intel.com, james.morse@arm.com,
	linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
	bagasdotme@gmail.com, eranian@google.com,
	christophe.leroy@csgroup.eu, jarkko@kernel.org,
	adrian.hunter@intel.com, quic_jiles@quicinc.com,
	peternewman@google.com
Subject: Re: [PATCH v8 8/8] x86/resctrl: Display hardware ids of resource groups
Date: Thu, 31 Aug 2023 18:58:20 -0500	[thread overview]
Message-ID: <012bf30d-7c41-b9a8-cdda-0922797d6a80@amd.com> (raw)
In-Reply-To: <5b75506b-80ac-a89d-4b70-f775c3b354d3@intel.com>

Hi Reinette,

On 8/31/23 12:42, Reinette Chatre wrote:
> Hi Babu,
> 
> On 8/30/2023 4:19 PM, Moger, Babu wrote:
>> On 8/29/23 15:14, Reinette Chatre wrote:
>>> On 8/21/2023 4:30 PM, Babu Moger wrote:
>>>
>>> ...
>>>
>>>> diff --git a/arch/x86/kernel/cpu/resctrl/internal.h b/arch/x86/kernel/cpu/resctrl/internal.h
>>>> index 2fae6d9e24d3..3fa32c1c2677 100644
>>>> --- a/arch/x86/kernel/cpu/resctrl/internal.h
>>>> +++ b/arch/x86/kernel/cpu/resctrl/internal.h
>>>> @@ -296,9 +296,15 @@ struct rdtgroup {
>>>>   *	--> RFTYPE_BASE (Files common for both MON and CTRL groups)
>>>>   *	    Files: cpus, cpus_list, tasks
>>>>   *
>>>> + *		--> RFTYPE_DEBUG (Files to help resctrl debugging)
>>>> + *		    File: mon_hw_id
>>>> + *
>>>
>>> This does not look right. I think mon_hw_id should have RFTYPE_MON
>>> (more below).
>>
>> I am not sure about this (more below).
>>
>>>
>>>>   *		--> RFTYPE_CTRL (Files only for CTRL group)
>>>>   *		    Files: mode, schemata, size
>>>>   *
>>>> + *			--> RFTYPE_DEBUG (Files to help resctrl debugging)
>>>> + *			    File: ctrl_hw_id
>>>> + *
>>>>   */
>>>>  #define RFTYPE_INFO			BIT(0)
>>>>  #define RFTYPE_BASE			BIT(1)
>>>> diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
>>>> index 94bd69f9964c..e0c2479acf49 100644
>>>> --- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
>>>> +++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
>>>> @@ -776,6 +776,38 @@ static int rdtgroup_tasks_show(struct kernfs_open_file *of,
>>>>  	return ret;
>>>>  }
>>>>  
>>>> +static int rdtgroup_closid_show(struct kernfs_open_file *of,
>>>> +				struct seq_file *s, void *v)
>>>> +{
>>>> +	struct rdtgroup *rdtgrp;
>>>> +	int ret = 0;
>>>> +
>>>> +	rdtgrp = rdtgroup_kn_lock_live(of->kn);
>>>> +	if (rdtgrp)
>>>> +		seq_printf(s, "%u\n", rdtgrp->closid);
>>>> +	else
>>>> +		ret = -ENOENT;
>>>> +	rdtgroup_kn_unlock(of->kn);
>>>> +
>>>> +	return ret;
>>>> +}
>>>> +
>>>> +static int rdtgroup_rmid_show(struct kernfs_open_file *of,
>>>> +			      struct seq_file *s, void *v)
>>>> +{
>>>> +	struct rdtgroup *rdtgrp;
>>>> +	int ret = 0;
>>>> +
>>>> +	rdtgrp = rdtgroup_kn_lock_live(of->kn);
>>>> +	if (rdtgrp)
>>>> +		seq_printf(s, "%u\n", rdtgrp->mon.rmid);
>>>> +	else
>>>> +		ret = -ENOENT;
>>>> +	rdtgroup_kn_unlock(of->kn);
>>>> +
>>>> +	return ret;
>>>> +}
>>>> +
>>>>  #ifdef CONFIG_PROC_CPU_RESCTRL
>>>>  
>>>>  /*
>>>> @@ -1837,6 +1869,13 @@ static struct rftype res_common_files[] = {
>>>>  		.seq_show	= rdtgroup_tasks_show,
>>>>  		.fflags		= RFTYPE_BASE,
>>>>  	},
>>>> +	{
>>>> +		.name		= "mon_hw_id",
>>>> +		.mode		= 0444,
>>>> +		.kf_ops		= &rdtgroup_kf_single_ops,
>>>> +		.seq_show	= rdtgroup_rmid_show,
>>>> +		.fflags		= RFTYPE_BASE | RFTYPE_DEBUG,
>>>
>>> I missed this earlier but I think this also needs a RFTYPE_MON.
>>> Perhaps patch 3 can introduce a new RFTYPE_MON_BASE to not
>>> have the flags of the two new files look too different?
>>
>> We have two types of files in base directory structure.
>>
>>  if (rtype == RDTCTRL_GROUP)
>>                 files = RFTYPE_BASE | RFTYPE_CTRL;
>>         else
>>                 files = RFTYPE_BASE | RFTYPE_MON;
>>
>> 1. RFTYPE_BASE | RFTYPE_CTRL;
>>    Files for the control group only.
>>
>> 2. RFTYPE_BASE | RFTYPE_MON;
>>    Files for both control and mon groups. However, RFTYPE_MON is not used
>> for any files. It is only RFTYPE_BASE.
>>
>> Because of the check in rdtgroup_add_files it all works fine.
>> For the control group it creates files with RFTYPE_BASE | RFTYPE_CTRL and
>> RFTYPE_BASE.
>>
>> For the mon group it creates files with RFTYPE_BASE only.
> 
> This describes current behavior because there are no resctrl
> files in base that are specific to monitoring, mon_hw_id is the
> first.
> 
> This does not mean that the new file mon_hw_id should just have
> RFTYPE_BASE because that would result in mon_hw_id being created
> for all control groups, even those that do not support monitoring
> Having mon_hw_id in resctrl for a group that does not support
> monitoring is not correct.
> 
> You should be able to reproduce this when booting your system
> with rdt=!cmt,!mbmlocal,!mbmtotal.

You are right. I reproduced it.

> 
>>
>> Adding FTYPE_MON_BASE will be a problem.
>>
> 
> Yes, this change does not just involve assigning the RFTYPE_MON
> to mon_hw_id. As you describe mkdir_rdt_prepare() does not take
> RFTYPE_MON into account when creating the files. Could this not just
> be a straightforward change to have it append RFTYPE_MON to the flags
> of files needing to be created for a CTRL_MON group? This would
> support new resource groups and then the default resource group
> would need to be taken into account also. What am I missing?
> 

It is not straight forward. We have have to handle few more things.
1. Base directory creation.
2. Mon directory creation after the base.

I got it working with this patches.  We may be able to clean it little
more or we can split also.

diff --git a/arch/x86/kernel/cpu/resctrl/internal.h
b/arch/x86/kernel/cpu/resctrl/internal.h
index 3fa32c1c2677..e2f3197f1c24 100644
--- a/arch/x86/kernel/cpu/resctrl/internal.h
+++ b/arch/x86/kernel/cpu/resctrl/internal.h
@@ -318,6 +318,7 @@ struct rdtgroup {
 #define RFTYPE_MON_INFO                        (RFTYPE_INFO | RFTYPE_MON)
 #define RFTYPE_TOP_INFO                        (RFTYPE_INFO | RFTYPE_TOP)
 #define RFTYPE_CTRL_BASE               (RFTYPE_BASE | RFTYPE_CTRL)
+#define RFTYPE_MON_BASE                        (RFTYPE_BASE | RFTYPE_MON)

 /* List of all resource groups */
 extern struct list_head rdt_all_groups;
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index e0c2479acf49..1f9abab7b9bd 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -1874,7 +1874,7 @@ static struct rftype res_common_files[] = {
                .mode           = 0444,
                .kf_ops         = &rdtgroup_kf_single_ops,
                .seq_show       = rdtgroup_rmid_show,
-               .fflags         = RFTYPE_BASE | RFTYPE_DEBUG,
+               .fflags         = RFTYPE_MON_BASE | RFTYPE_DEBUG,
        },
        {
                .name           = "schemata",
@@ -2558,6 +2558,7 @@ static void schemata_list_destroy(void)
 static int rdt_get_tree(struct fs_context *fc)
 {
        struct rdt_fs_context *ctx = rdt_fc2context(fc);
+       uint flags = RFTYPE_CTRL_BASE;
        struct rdt_domain *dom;
        struct rdt_resource *r;
        int ret;
@@ -2588,7 +2589,10 @@ static int rdt_get_tree(struct fs_context *fc)

        closid_init();

-       ret = rdtgroup_add_files(rdtgroup_default.kn, RFTYPE_CTRL_BASE);
+       if (rdt_mon_capable)
+               flags |= RFTYPE_MON;
+
+       ret = rdtgroup_add_files(rdtgroup_default.kn, flags);
        if (ret)
                goto out_schemata_free;

@@ -3336,6 +3340,9 @@ static int mkdir_rdt_prepare(struct kernfs_node
*parent_kn,
        else
                files = RFTYPE_BASE | RFTYPE_MON;

+       if (rdt_mon_capable)
+               files |= RFTYPE_MON;
+
        ret = rdtgroup_add_files(kn, files);
        if (ret) {
                rdt_last_cmd_puts("kernfs fill error\n");


-- 
Thanks
Babu Moger

  reply	other threads:[~2023-08-31 23:58 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-21 23:30 [PATCH v8 0/8] x86/resctrl: Miscellaneous resctrl features Babu Moger
2023-08-21 23:30 ` [PATCH v8 1/8] x86/resctrl: Add multiple tasks to the resctrl group at once Babu Moger
2023-09-01 22:13   ` Fenghua Yu
2023-09-06 14:56     ` Moger, Babu
2023-09-06 20:42       ` Fenghua Yu
2023-09-07 14:44         ` Moger, Babu
2023-09-07 14:51           ` Fenghua Yu
2023-08-21 23:30 ` [PATCH v8 2/8] x86/resctrl: Simplify rftype flag definitions Babu Moger
2023-09-01 22:14   ` Fenghua Yu
2023-08-21 23:30 ` [PATCH v8 3/8] x86/resctrl: Rename rftype flags for consistency Babu Moger
2023-09-01 22:15   ` Fenghua Yu
2023-08-21 23:30 ` [PATCH v8 4/8] x86/resctrl: Add comments on RFTYPE flags hierarchy Babu Moger
2023-08-23  7:03   ` Shaopeng Tan (Fujitsu)
2023-08-23 15:56     ` Moger, Babu
2023-08-29 20:08   ` Reinette Chatre
2023-08-30 16:30     ` Moger, Babu
2023-09-01 22:31   ` Fenghua Yu
2023-09-06 15:10     ` Moger, Babu
2023-08-21 23:30 ` [PATCH v8 5/8] x86/resctrl: Unwind the errors inside rdt_enable_ctx() Babu Moger
2023-08-29 20:10   ` Reinette Chatre
2023-08-30 16:38     ` Moger, Babu
2023-08-30 17:56       ` Reinette Chatre
2023-08-30 18:28         ` Moger, Babu
2023-09-01 23:33   ` Fenghua Yu
2023-08-21 23:30 ` [PATCH v8 6/8] x86/resctrl: Move default group file creation to mount Babu Moger
2023-08-29 20:11   ` Reinette Chatre
2023-08-30 19:50     ` Moger, Babu
2023-08-30 20:00       ` Reinette Chatre
2023-08-30 21:18         ` Moger, Babu
2023-08-30 22:05           ` Reinette Chatre
2023-08-30 23:24             ` Moger, Babu
2023-09-01 23:21   ` Fenghua Yu
2023-09-01 23:36     ` Reinette Chatre
2023-09-01 23:46       ` Fenghua Yu
2023-09-01 23:48         ` Reinette Chatre
2023-09-06 15:19     ` Moger, Babu
2023-08-21 23:30 ` [PATCH v8 7/8] x86/resctrl: Introduce "-o debug" mount option Babu Moger
2023-08-29 20:12   ` Reinette Chatre
2023-08-30 21:45     ` Moger, Babu
2023-09-01 22:35   ` Fenghua Yu
2023-08-21 23:30 ` [PATCH v8 8/8] x86/resctrl: Display hardware ids of resource groups Babu Moger
2023-08-29 20:14   ` Reinette Chatre
2023-08-30 23:19     ` Moger, Babu
2023-08-31 17:42       ` Reinette Chatre
2023-08-31 23:58         ` Moger, Babu [this message]
2023-09-01  0:43           ` Reinette Chatre
2023-09-01 17:28             ` Moger, Babu
2023-09-01 17:57               ` Reinette Chatre
2023-09-05 16:51                 ` Moger, Babu
2023-09-01 22:44   ` Fenghua Yu
2023-08-23  7:06 ` [PATCH v8 0/8] x86/resctrl: Miscellaneous resctrl features Shaopeng Tan (Fujitsu)
2023-08-23 15:12   ` 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=012bf30d-7c41-b9a8-cdda-0922797d6a80@amd.com \
    --to=babu.moger@amd.com \
    --cc=adrian.hunter@intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=bagasdotme@gmail.com \
    --cc=bp@alien8.de \
    --cc=chang.seok.bae@intel.com \
    --cc=christophe.leroy@csgroup.eu \
    --cc=corbet@lwn.net \
    --cc=damien.lemoal@opensource.wdc.com \
    --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=james.morse@arm.com \
    --cc=jarkko@kernel.org \
    --cc=jmattson@google.com \
    --cc=jpoimboe@kernel.org \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=paulmck@kernel.org \
    --cc=pawan.kumar.gupta@linux.intel.com \
    --cc=pbonzini@redhat.com \
    --cc=peternewman@google.com \
    --cc=peterz@infradead.org \
    --cc=quic_jiles@quicinc.com \
    --cc=quic_neeraju@quicinc.com \
    --cc=rdunlap@infradead.org \
    --cc=reinette.chatre@intel.com \
    --cc=sandipan.das@amd.com \
    --cc=songmuchun@bytedance.com \
    --cc=tglx@linutronix.de \
    --cc=tony.luck@intel.com \
    --cc=x86@kernel.org \
    /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

Powered by JetHome