Each telemetry aggregator provides three status registers at the top
end of MMIO space after all the per-RMID per-event counters:

  agg_data_loss_count: This counts the number of times that this aggregator
  failed to accumulate a counter value supplied by a CPU core.

  agg_data_loss_timestamp: This is a "timestamp" from a free running
  25MHz uncore timer indicating when the most recent data loss occurred.

  last_update_timestamp: Another 25MHz timestamp indicating when the
  most recent counter update was successfully applied.

Create files in /sys/kernel/debug/resctrl/info/PERF_PKG_MON/x86_64/
to display the value of each of these status registers for each aggregator
in each enabled event group.

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/x86/kernel/cpu/resctrl/intel_aet.c | 53 +++++++++++++++++++++++++
 1 file changed, 53 insertions(+)

diff --git a/arch/x86/kernel/cpu/resctrl/intel_aet.c b/arch/x86/kernel/cpu/resctrl/intel_aet.c
index 4afc99e74cef..ff70f2b18165 100644
--- a/arch/x86/kernel/cpu/resctrl/intel_aet.c
+++ b/arch/x86/kernel/cpu/resctrl/intel_aet.c
@@ -13,6 +13,7 @@
 
 #include <linux/cleanup.h>
 #include <linux/cpu.h>
+#include <linux/debugfs.h>
 #include <linux/intel_vsec.h>
 #include <linux/io.h>
 #include <linux/minmax.h>
@@ -299,6 +300,55 @@ static bool get_pmt_feature(enum pmt_feature_id feature, struct event_group **ev
 	return false;
 }
 
+static int status_read(void *priv, u64 *val)
+{
+	void __iomem *info = (void __iomem *)priv;
+
+	*val = readq(info);
+
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(status_fops, status_read, NULL, "%llu\n");
+
+static void make_status_files(struct dentry *dir, struct event_group *e, int pkg, int instance)
+{
+	void *info = (void __force *)e->pkginfo[pkg]->addrs[instance] + e->mmio_size;
+	char name[64];
+
+	sprintf(name, "%s_pkg%d_agg%d_data_loss_count", e->name, pkg, instance);
+	debugfs_create_file(name, 0400, dir, info - 24, &status_fops);
+
+	sprintf(name, "%s_pkg%d_agg%d_data_loss_timestamp", e->name, pkg, instance);
+	debugfs_create_file(name, 0400, dir, info - 16, &status_fops);
+
+	sprintf(name, "%s_pkg%d_agg%d_last_update_timestamp", e->name, pkg, instance);
+	debugfs_create_file(name, 0400, dir, info - 8, &status_fops);
+}
+
+static void create_debug_event_status_files(struct dentry *dir, struct event_group *e)
+{
+	int num_pkgs = topology_max_packages();
+
+	for (int i = 0; i < num_pkgs; i++)
+		for (int j = 0; j < e->pkginfo[i]->num_regions; j++)
+			make_status_files(dir, e, i, j);
+}
+
+static void create_debugfs_status_file(void)
+{
+	struct rdt_resource *r = &rdt_resources_all[RDT_RESOURCE_PERF_PKG].r_resctrl;
+	struct event_group *evg;
+	struct dentry *infodir;
+
+	infodir = resctrl_debugfs_mon_info_arch_mkdir(r);
+	if (!infodir)
+		return;
+
+	list_for_each_entry(evg, &active_event_groups, list)
+		create_debug_event_status_files(infodir, evg);
+}
+
 /*
  * Ask OOBMSM discovery driver for all the RMID based telemetry groups
  * that it supports.
@@ -314,6 +364,9 @@ bool intel_aet_get_events(void)
 			       known_perf_event_groups,
 			       ARRAY_SIZE(known_perf_event_groups));
 
+	if (ret1 || ret2)
+		create_debugfs_status_file();
+
 	return ret1 || ret2;
 }
 
-- 
2.50.1

Hi Tony,

On 8/11/25 11:17 AM, Tony Luck wrote:
> Each telemetry aggregator provides three status registers at the top
> end of MMIO space after all the per-RMID per-event counters:
> 
>   agg_data_loss_count: This counts the number of times that this aggregator
>   failed to accumulate a counter value supplied by a CPU core.
> 
>   agg_data_loss_timestamp: This is a "timestamp" from a free running
>   25MHz uncore timer indicating when the most recent data loss occurred.
> 
>   last_update_timestamp: Another 25MHz timestamp indicating when the
>   most recent counter update was successfully applied.

With all three files being "per aggregator" it is not clear to me why
"last_update_timestamp" does not have the "agg_" prefix. Looking at the
debugfs files created all three are indeed created with the "agg" prefix
so this looks inconsistent.

> 
> Create files in /sys/kernel/debug/resctrl/info/PERF_PKG_MON/x86_64/
> to display the value of each of these status registers for each aggregator
> in each enabled event group.

To avoid reviewers needing to dig, this can add that these files are removed 
as part of the debugfs_remove_recursive() called during resctrl exit.

> 
> Signed-off-by: Tony Luck <tony.luck@intel.com>
> ---
>  arch/x86/kernel/cpu/resctrl/intel_aet.c | 53 +++++++++++++++++++++++++
>  1 file changed, 53 insertions(+)
> 
> diff --git a/arch/x86/kernel/cpu/resctrl/intel_aet.c b/arch/x86/kernel/cpu/resctrl/intel_aet.c
> index 4afc99e74cef..ff70f2b18165 100644
> --- a/arch/x86/kernel/cpu/resctrl/intel_aet.c
> +++ b/arch/x86/kernel/cpu/resctrl/intel_aet.c
> @@ -13,6 +13,7 @@
>  
>  #include <linux/cleanup.h>
>  #include <linux/cpu.h>
> +#include <linux/debugfs.h>
>  #include <linux/intel_vsec.h>
>  #include <linux/io.h>
>  #include <linux/minmax.h>
> @@ -299,6 +300,55 @@ static bool get_pmt_feature(enum pmt_feature_id feature, struct event_group **ev
>  	return false;
>  }
>  
> +static int status_read(void *priv, u64 *val)
> +{
> +	void __iomem *info = (void __iomem *)priv;
> +
> +	*val = readq(info);
> +
> +	return 0;
> +}
> +
> +DEFINE_SIMPLE_ATTRIBUTE(status_fops, status_read, NULL, "%llu\n");
> +
> +static void make_status_files(struct dentry *dir, struct event_group *e, int pkg, int instance)
> +{
> +	void *info = (void __force *)e->pkginfo[pkg]->addrs[instance] + e->mmio_size;
> +	char name[64];
> +
> +	sprintf(name, "%s_pkg%d_agg%d_data_loss_count", e->name, pkg, instance);
> +	debugfs_create_file(name, 0400, dir, info - 24, &status_fops);
> +
> +	sprintf(name, "%s_pkg%d_agg%d_data_loss_timestamp", e->name, pkg, instance);
> +	debugfs_create_file(name, 0400, dir, info - 16, &status_fops);
> +
> +	sprintf(name, "%s_pkg%d_agg%d_last_update_timestamp", e->name, pkg, instance);
> +	debugfs_create_file(name, 0400, dir, info - 8, &status_fops);
> +}
> +
> +static void create_debug_event_status_files(struct dentry *dir, struct event_group *e)
> +{
> +	int num_pkgs = topology_max_packages();
> +
> +	for (int i = 0; i < num_pkgs; i++)
> +		for (int j = 0; j < e->pkginfo[i]->num_regions; j++)
> +			make_status_files(dir, e, i, j);
> +}
> +
> +static void create_debugfs_status_file(void)
> +{
> +	struct rdt_resource *r = &rdt_resources_all[RDT_RESOURCE_PERF_PKG].r_resctrl;
> +	struct event_group *evg;
> +	struct dentry *infodir;
> +
> +	infodir = resctrl_debugfs_mon_info_arch_mkdir(r);
> +	if (!infodir)
> +		return;
> +
> +	list_for_each_entry(evg, &active_event_groups, list)
> +		create_debug_event_status_files(infodir, evg);

To avoid using/needing active_event_groups create_debug_event_status_files()
can be called at end of discover_events() and infodir can be a global that
is initialized there if it is NULL.

Reinette