The Linux Kernel Mailing List
 help / color / mirror / Atom feed
From: Reinette Chatre <reinette.chatre@intel.com>
To: Chen Yu <yu.c.chen@intel.com>, <tony.luck@intel.com>
Cc: <x86@kernel.org>, <linux-kernel@vger.kernel.org>,
	<tglx@kernel.org>, <bp@alien8.de>, <mingo@redhat.com>,
	<dave.hansen@linux.intel.com>, <hpa@zytor.com>,
	<fenghuay@nvidia.com>, <babu.moger@amd.com>,
	<anil.keshavamurthy@broadcom.com>, <chen.yu@linux.dev>,
	Hongyu Ning <hongyu.ning@linux.intel.com>
Subject: Re: [PATCH v5 05/10] x86/resctrl: Parse ACPI CMRC table
Date: Fri, 10 Jul 2026 16:46:24 -0700	[thread overview]
Message-ID: <87255829-a70d-4b64-9fd7-9771677d1933@intel.com> (raw)
In-Reply-To: <a8adfc6ddb622229d1a7000980f83a6763427d99.1782866200.git.yu.c.chen@intel.com>

Hi Chenyu,

On 7/1/26 6:46 AM, Chen Yu wrote:
> The CMRC (Cache Monitoring Registers for CPU Agents Description)
> sub-table of ERDT describes the MMIO registers used to read
> cache monitoring counters (e.g. LLC occupancy) for an RMD.

Please use the entire line length available (please check all changelogs).
Some context for this request:
https://lore.kernel.org/lkml/20250916105447.GCaMlB976WLxHHeNMD@fat_crate.local/

> 
> Parse each CMRC sub-table, ioremap its register window, and save
> the CMRC pointer in the corresponding ERDT domain entry so that
> later monitoring code can read the counters via MMIO.

Is the "save the CMRC pointer ..." referring to how resctrl makes
a copy of the CMRC table? If so, that is very different from
"saving a pointer".

> 
> Suggested-by: Tony Luck <tony.luck@intel.com>
> Reviewed-by: Thomas Gleixner <tglx@kernel.org>
> Tested-by: Hongyu Ning <hongyu.ning@linux.intel.com>
> Signed-off-by: Chen Yu <yu.c.chen@intel.com>
> ---
> v4->v5:
>   No change.    
> ---
>  arch/x86/kernel/cpu/resctrl/erdt.c     | 44 ++++++++++++++++++++++++++
>  arch/x86/kernel/cpu/resctrl/internal.h |  1 +
>  2 files changed, 45 insertions(+)
> 
> diff --git a/arch/x86/kernel/cpu/resctrl/erdt.c b/arch/x86/kernel/cpu/resctrl/erdt.c
> index 6c1df7e43eab..a5754d64fcc1 100644
> --- a/arch/x86/kernel/cpu/resctrl/erdt.c
> +++ b/arch/x86/kernel/cpu/resctrl/erdt.c
> @@ -26,6 +26,7 @@ static LIST_HEAD(domain_info_list);
>  static bool __erdt_enabled;
>  
>  #define ERDT_VALID_VERSION		1
> +#define CMRC_SUPPORTED_INDEX_FN		1
>  #define RMDD_FLAG_CPU_L3_DOMAIN		BIT(0)
>  
>  /* Bitmask of valid sub-tables found in the first RMDD, used to ensure all RMDDs match. */
> @@ -79,6 +80,7 @@ static void cleanup_one_domain(struct erdt_domain_info *d)
>  {
>  	erdt_iounmap_domain(d);
>  	free_cpumask_var(d->cpu_mask);
> +	kfree(d->cmrc);
>  	kfree(d);
>  }
>  
> @@ -112,6 +114,41 @@ static __init int cacd_init(struct acpi_subtbl_hdr_16 *subtbl,
>  	return 0;
>  }
>  
> +static __init int cmrc_init(struct acpi_subtbl_hdr_16 *subtbl,
> +			    struct erdt_domain_info *domain_info)
> +{
> +	struct acpi_erdt_cmrc *cmrc = (struct acpi_erdt_cmrc *)subtbl;
> +
> +	if (subtbl->length < sizeof(*cmrc)) {

I think it will be helpful if the ERDT parsing code uses consistent patterns. Here,
for example, now that a pointer of accurate type is available via cmrc, can cmrc
be referenced instead to be consistent with cacd_init()?

> +		pr_warn(FW_BUG "Truncated CMRC subtable\n");
> +		return -EIO;
> +	}
> +
> +	if (cmrc->index_fn != CMRC_SUPPORTED_INDEX_FN) {
> +		pr_info("Unsupported CMRC index function %d\n", cmrc->index_fn);
> +		return -EIO;
> +	}
> +
> +	if (!cmrc->clump_size) {
> +		pr_warn(FW_BUG "CMRC clump_size is zero\n");
> +		return -EIO;
> +	}
> +
> +	domain_info->base[ERDT_MMIO_CMRC_BASE] =
> +		erdt_ioremap(cmrc->cmt_reg_base, cmrc->cmt_reg_size, "CMRC base");
> +	if (!domain_info->base[ERDT_MMIO_CMRC_BASE])
> +		return -EIO;
> +
> +	domain_info->cmrc = kmemdup(cmrc, subtbl->length, GFP_KERNEL);
> +	if (!domain_info->cmrc) {
> +		iounmap(domain_info->base[ERDT_MMIO_CMRC_BASE]);
> +		domain_info->base[ERDT_MMIO_CMRC_BASE] = NULL;
> +		return -ENOMEM;
> +	}
> +
> +	return 0;
> +}
> +
>  static inline struct acpi_subtbl_hdr_16 *rmdd_subtbl(struct acpi_erdt_rmdd *rmdd)
>  {
>  	return (void *)rmdd + sizeof(*rmdd);
> @@ -178,6 +215,13 @@ static __init bool parse_rmdd_entry(struct acpi_subtbl_hdr_16 *rmdd_hdr)
>  				goto cleanup;
>  
>  			subtbl_mask |= BIT(ACPI_ERDT_TYPE_CACD);
> +			break;
> +		case ACPI_ERDT_TYPE_CMRC:
> +			/* TBD: Only 1 CMRR per domain is allowed? */

Who are you asking this question to?

> +			if (!(subtbl_mask & BIT(ACPI_ERDT_TYPE_CMRC)) &&
> +			    !cmrc_init(subtbl, domain_info))
> +				subtbl_mask |= BIT(ACPI_ERDT_TYPE_CMRC);
> +
>  			break;
>  		default:
>  			break;
> diff --git a/arch/x86/kernel/cpu/resctrl/internal.h b/arch/x86/kernel/cpu/resctrl/internal.h
> index 7d9100b7648f..3dab05c38003 100644
> --- a/arch/x86/kernel/cpu/resctrl/internal.h
> +++ b/arch/x86/kernel/cpu/resctrl/internal.h
> @@ -36,6 +36,7 @@ enum erdt_mmio_type {
>  
>  struct erdt_domain_info {
>  	void __iomem		*base[ERDT_MMIO_NUM_TYPES];
> +	struct acpi_erdt_cmrc	*cmrc;
>  	cpumask_var_t		cpu_mask;
>  	int			max_rmid;
>  	struct list_head	list;

Reinette


  reply	other threads:[~2026-07-10 23:46 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-01 13:44 [PATCH v5 00/10] Introduce MMIO-based CMT access for Enhanced RDT Chen Yu
2026-07-01 13:45 ` [PATCH v5 01/10] x86/resctrl: Require 64-bit x86 for resctrl support Chen Yu
2026-07-01 13:45 ` [PATCH v5 02/10] x86/topology: Export topo_lookup_cpuid() for resctrl use Chen Yu
2026-07-10 23:35   ` Reinette Chatre
2026-07-01 13:45 ` [PATCH v5 03/10] x86/resctrl: Parse ACPI ERDT table and save CACD cpumask for RMDD domains Chen Yu
2026-07-10 23:42   ` Reinette Chatre
2026-07-01 13:45 ` [PATCH v5 04/10] x86/resctrl: Attach ACPI ERDT information to L3 mon domain on CPU online Chen Yu
2026-07-10 23:45   ` Reinette Chatre
2026-07-01 13:46 ` [PATCH v5 05/10] x86/resctrl: Parse ACPI CMRC table Chen Yu
2026-07-10 23:46   ` Reinette Chatre [this message]
2026-07-01 13:46 ` [PATCH v5 06/10] x86/resctrl: Replace "msr" in monitoring data identifiers Chen Yu
2026-07-10 23:47   ` Reinette Chatre
2026-07-01 13:46 ` [PATCH v5 07/10] x86/resctrl: Refactor the monitor read function Chen Yu
2026-07-01 13:47 ` [PATCH v5 08/10] fs/resctrl: Do not invoke smp_processor_id() in preemptible context Chen Yu
2026-07-10 23:48   ` Reinette Chatre
2026-07-01 13:47 ` [PATCH v5 09/10] x86/resctrl: Introduce helpers to read L3 occupancy via MMIO Chen Yu
2026-07-10 23:53   ` Reinette Chatre
2026-07-01 13:47 ` [PATCH v5 10/10] x86/resctrl: Enable " Chen Yu
2026-07-10 23:55   ` Reinette Chatre

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=87255829-a70d-4b64-9fd7-9771677d1933@intel.com \
    --to=reinette.chatre@intel.com \
    --cc=anil.keshavamurthy@broadcom.com \
    --cc=babu.moger@amd.com \
    --cc=bp@alien8.de \
    --cc=chen.yu@linux.dev \
    --cc=dave.hansen@linux.intel.com \
    --cc=fenghuay@nvidia.com \
    --cc=hongyu.ning@linux.intel.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=tglx@kernel.org \
    --cc=tony.luck@intel.com \
    --cc=x86@kernel.org \
    --cc=yu.c.chen@intel.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