Perf stat metric grouping with hardware information
Subsystems (2)
v2 cover Perf stat metric grouping with hardware information
From: Weilin Wang <[email protected]>
Chanages in v2:
- Updated the ordering of patches and removed bug fixing patches [Kan]
- Move code into dedicated patches or more related patches [Kan]
- Removed unnecessary code [Kan, Yang]
Grouping algorithm details:
- Currently, this grouping method groups a combined list of events, which is a unique list of events for the required metrics.
- The grouping process traverses the event list to build up a event_info_list that includes counter restriction rules of each event. It also creates a pmu_info_list for the counter info - the number of gp counters and fixed counters available in each PMU in current system. When grouping is in progress, there are two bitmaps used to represent usages of gp counters and fixed counters in each group. These two bitmaps are initialized based on the two counter sizes of the corresponding PMU.
Assumption: this method reads counter availablility from data described in JSON files and builds up groups based on this info.
- The grouping step traverses the event_info_list. For each event, it checks bitmaps to get counter availability in current group. It sets the bit and inserts event into the group if found a vacant counter that fits the restrictions of the event. Otherwise, it creates a new group, places event in it, and then inserts it into the group list.
v1: https://lore.kernel.org/all/[email protected]/
Weilin Wang (17):
perf stat: Add hardware-grouping cmd option to perf stat perf stat: Add basic functions for the hardware-grouping stat cmd
option
perf pmu-events: Add functions in jevent.py to parse counter and event
info for hardware aware grouping
perf pmu-events: Add counter info into JSON files for SapphireRapids perf pmu-events: Add event counter data for Cascadelakex perf pmu-events: Add event counter data for Icelakex
perf stat: Add functions to set counter bitmaps for hardware-grouping
method
perf stat: Add functions to get counter info
perf stat: Add functions to create new group and assign events into
groups for hardware-grouping method
perf stat: Add build string function and topdown events handling in
hardware-grouping
perf stat: Add function to handle special events in hardware-grouping perf stat: Add function to combine metrics for hardware-grouping perf stat: Handle taken alone in hardware-grouping
perf stat: Handle NMI in hardware-grouping
perf stat: Code refactoring in hardware-grouping
perf stat: Add tool events support in hardware-grouping perf pmu-events: Add event counter data for Tigerlake
tools/lib/bitmap.c | 20 + tools/perf/builtin-stat.c | 7 + .../arch/x86/cascadelakex/cache.json | 1237 ++++++++++++ .../arch/x86/cascadelakex/counter.json | 17 + .../arch/x86/cascadelakex/floating-point.json | 16 + .../arch/x86/cascadelakex/frontend.json | 68 + .../arch/x86/cascadelakex/memory.json | 751 ++++++++ .../arch/x86/cascadelakex/other.json | 168 ++ .../arch/x86/cascadelakex/pipeline.json | 102 + .../arch/x86/cascadelakex/uncore-cache.json | 1138 +++++++++++ .../x86/cascadelakex/uncore-interconnect.json | 1272 +++++++++++++ .../arch/x86/cascadelakex/uncore-io.json | 394 ++++ .../arch/x86/cascadelakex/uncore-memory.json | 509 +++++ .../arch/x86/cascadelakex/uncore-power.json | 25 + .../arch/x86/cascadelakex/virtual-memory.json | 28 + .../pmu-events/arch/x86/icelakex/cache.json | 98 + .../pmu-events/arch/x86/icelakex/counter.json | 17 + .../arch/x86/icelakex/floating-point.json | 13 + .../arch/x86/icelakex/frontend.json | 55 + .../pmu-events/arch/x86/icelakex/memory.json | 53 + .../pmu-events/arch/x86/icelakex/other.json | 52 + .../arch/x86/icelakex/pipeline.json | 92 + .../arch/x86/icelakex/uncore-cache.json | 965 ++++++++++ .../x86/icelakex/uncore-interconnect.json | 1667 +++++++++++++++++ .../arch/x86/icelakex/uncore-io.json | 966 ++++++++++ .../arch/x86/icelakex/uncore-memory.json | 186 ++ .../arch/x86/icelakex/uncore-power.json | 26 + .../arch/x86/icelakex/virtual-memory.json | 22 + .../arch/x86/sapphirerapids/cache.json | 104 + .../arch/x86/sapphirerapids/counter.json | 17 + .../x86/sapphirerapids/floating-point.json | 25 + .../arch/x86/sapphirerapids/frontend.json | 98 +- .../arch/x86/sapphirerapids/memory.json | 44 + .../arch/x86/sapphirerapids/other.json | 40 + .../arch/x86/sapphirerapids/pipeline.json | 118 ++ .../arch/x86/sapphirerapids/uncore-cache.json | 534 +++++- .../arch/x86/sapphirerapids/uncore-cxl.json | 56 + .../sapphirerapids/uncore-interconnect.json | 476 +++++ .../arch/x86/sapphirerapids/uncore-io.json | 373 ++++ .../x86/sapphirerapids/uncore-memory.json | 391 ++++ .../arch/x86/sapphirerapids/uncore-power.json | 24 + .../x86/sapphirerapids/virtual-memory.json | 20 + .../pmu-events/arch/x86/tigerlake/cache.json | 65 + .../arch/x86/tigerlake/counter.json | 7 + .../arch/x86/tigerlake/floating-point.json | 13 + .../arch/x86/tigerlake/frontend.json | 56 + .../pmu-events/arch/x86/tigerlake/memory.json | 31 + .../pmu-events/arch/x86/tigerlake/other.json | 4 + .../arch/x86/tigerlake/pipeline.json | 96 + .../x86/tigerlake/uncore-interconnect.json | 11 + .../arch/x86/tigerlake/uncore-memory.json | 6 + .../arch/x86/tigerlake/uncore-other.json | 1 + .../arch/x86/tigerlake/virtual-memory.json | 20 + tools/perf/pmu-events/jevents.py | 177 +- tools/perf/pmu-events/pmu-events.h | 26 +- tools/perf/util/metricgroup.c | 947 ++++++++++ tools/perf/util/metricgroup.h | 106 ++ tools/perf/util/pmu.c | 5 + tools/perf/util/pmu.h | 1 + tools/perf/util/stat.h | 1 + 60 files changed, 13832 insertions(+), 25 deletions(-) create mode 100644 tools/perf/pmu-events/arch/x86/cascadelakex/counter.json create mode 100644 tools/perf/pmu-events/arch/x86/icelakex/counter.json create mode 100644 tools/perf/pmu-events/arch/x86/sapphirerapids/counter.json create mode 100644 tools/perf/pmu-events/arch/x86/tigerlake/counter.json
--
2.39.3
RFC PATCH v2 1/17 perf stat: Add hardware-grouping cmd option to perf stat
From: Weilin Wang <[email protected]>
Hardware counter and event information could be used to help creating event groups that better utilize hardware counters and improve multiplexing.
Add a cmd option to allow user to choose this new metric grouping method.
$ perf stat -M TopdownL1 -a --hardware-grouping
--- tools/perf/builtin-stat.c | 7 +++++++ tools/perf/util/metricgroup.c | 5 +++++ tools/perf/util/metricgroup.h | 1 + tools/perf/util/stat.h | 1 + 4 files changed, 14 insertions(+)
tools/perf/builtin-stat.c +7 -0
@@ -1256,6 +1256,8 @@ static struct option stat_options[] = { "don't try to share events between metrics in a group"),OPT_BOOLEAN (0 ,"metric-no-threshold" , &stat_config.metric_no_threshold,"don't try to share events between metrics in a group " ), +OPT_BOOLEAN (0 ,"hardware-grouping" , &stat_config.hardware_aware_grouping, +"Use hardware aware metric grouping method" ),OPT_BOOLEAN (0 ,"topdown" , &topdown_run,"measure top-down statistics" ), OPT_UINTEGER(0, "td-level", &stat_config.topdown_level,
@@ -2061,6 +2063,7 @@ static int add_default_attributes(void) stat_config.metric_no_threshold, stat_config.user_requested_cpu_list, stat_config.system_wide, + stat_config.hardware_aware_grouping, &stat_config.metric_events); }
@@ -2094,6 +2097,7 @@ static int add_default_attributes(void) stat_config.metric_no_threshold, stat_config.user_requested_cpu_list, stat_config.system_wide, + stat_config.hardware_aware_grouping, &stat_config.metric_events); }
@@ -2128,6 +2132,7 @@ static int add_default_attributes(void)/*metric_no_threshold=*/ true, stat_config.user_requested_cpu_list, stat_config.system_wide, + stat_config.hardware_aware_grouping, &stat_config.metric_events) < 0) return -1; }
@@ -2169,6 +2174,7 @@ static int add_default_attributes(void)/*metric_no_threshold=*/ true, stat_config.user_requested_cpu_list, stat_config.system_wide, + stat_config.hardware_aware_grouping, &stat_config.metric_events) < 0) return -1;
@@ -2702,6 +2708,7 @@ int cmd_stat(int argc, const char **argv) stat_config.metric_no_threshold, stat_config.user_requested_cpu_list, stat_config.system_wide, + stat_config.hardware_aware_grouping, &stat_config.metric_events);zfree (&metrics); }
tools/perf/util/metricgroup.c +5 -0
@@ -1690,12 +1690,17 @@ int metricgroup__parse_groups(struct evlist *perf_evlist,bool metric_no_threshold, const char *user_requested_cpu_list, bool system_wide, + bool hardware_aware_grouping, struct rblist *metric_events) { const struct pmu_metrics_table *table = pmu_metrics_table__find();if (!table)return -EINVAL ; +if (hardware_aware_grouping) { +pr_debug ("Use hardware aware grouping instead of traditional metric grouping method\n" ); + } + return parse_groups(perf_evlist, pmu, str, metric_no_group, metric_no_merge, metric_no_threshold, user_requested_cpu_list, system_wide,
tools/perf/util/metricgroup.h +1 -0
@@ -77,6 +77,7 @@ int metricgroup__parse_groups(struct evlist *perf_evlist,bool metric_no_threshold, const char *user_requested_cpu_list, bool system_wide, + bool hardware_aware_grouping, struct rblist *metric_events); int metricgroup__parse_groups_test(struct evlist *evlist, const struct pmu_metrics_table *table,
tools/perf/util/stat.h +1 -0
@@ -87,6 +87,7 @@ struct perf_stat_config {bool metric_no_group;bool metric_no_merge;bool metric_no_threshold; +bool hardware_aware_grouping;bool stop_read_counter;bool iostat_run;char *user_requested_cpu_list;
Signature · 1 line
2.39.3
RFC PATCH v2 2/17 perf stat: Add basic functions for the hardware-grouping stat cmd option
In reply to [email protected]From: Weilin Wang <[email protected]>
Add the first set of functions for the hardware-grouping method. Function hw_awre_parse_groups() is the entry point of this metric grouping method. It does metric grouping on a combined list of events and will create a list of grouping strings as final results of the grouping method. These grouping strings will be used in the same mannor as existing metric grouping process.
This method will fall back to normal grouping when hardware aware grouping return with err so that perf stat still executes and returns with correct result.
--- tools/perf/util/metricgroup.c | 210 +++++++++++++++++++++++++++++++++- tools/perf/util/metricgroup.h | 9 ++ 2 files changed, 218 insertions(+), 1 deletion(-)
tools/perf/util/metricgroup.c +209 -1
@@ -1432,6 +1432,101 @@ static int build_combined_expr_ctx(const struct list_head *metric_list,return ret; } +/** +* hw_aware_build_grouping - Build event groupings by reading counter +* requirement of the events and counter available on the system from +* pmu-events. +* @ctx: the event identifiers parsed from metrics. +* @groupings: header to the list of final event grouping. +* @modifier: any modifiers added to the events. +*/ +static int hw_aware_build_grouping (struct expr_parse_ctx *ctx__maybe_unused , +struct list_head *groupings__maybe_unused , +const char *modifier__maybe_unused ) +{ +int ret =0 ; + +pr_debug ("This is a placeholder\n" ); +return ret; +} + +static void group_str_free (struct metricgroup__group_strs *g) +{ +if (!g) +return ; + +strbuf_release (&g->grouping_str); +free (g); +} + +static void metricgroup__free_grouping_strs (struct list_head + *grouping_strs) +{ +struct metricgroup__group_strs *g, *tmp; + +list_for_each_entry_safe (g, tmp, grouping_strs, nd) { +list_del_init (&g->nd); +group_str_free (g); + } +} + +/** +* hw_aware_parse_ids - Build the event string for the ids and parse them +* creating an evlist. The encoded metric_ids are decoded. Events are placed +* into groups based on event counter requirements and counter availabilities of +* the system. +* @metric_no_merge: is metric sharing explicitly disabled. +* @fake_pmu: used when testing metrics not supported by the current CPU. +* @ids: the event identifiers parsed from a metric. +* @modifier: any modifiers added to the events. +* @out_evlist: the created list of events. +*/ +static int hw_aware_parse_ids (struct perf_pmu *fake_pmu, +struct expr_parse_ctx *ids,const char *modifier, +struct evlist **out_evlist) +{ +struct parse_events_error parse_error; +struct evlist *parsed_evlist; +LIST_HEAD (groupings); +struct metricgroup__group_strs *group; +int ret; + + *out_evlist =NULL ; + ret =hw_aware_build_grouping (ids, &groupings, modifier); +if (ret) { +metricgroup__free_grouping_strs (&groupings); +return ret; + } + + parsed_evlist =evlist__new (); +if (!parsed_evlist) { + ret = -ENOMEM ; + goto err_out; + } +list_for_each_entry (group, &groupings, nd) { +struct strbuf *events = &group->grouping_str; + +pr_debug ("Parsing metric events '%s'\n" , events->buf); +parse_events_error__init (&parse_error); + ret =__parse_events (parsed_evlist, events->buf,/*pmu_filter=*/ NULL , + &parse_error, fake_pmu,/*warn_if_reordered=*/ false); +if (ret) { +parse_events_error__print (&parse_error, events->buf); + goto err_out; + } + ret =decode_all_metric_ids (parsed_evlist, modifier); +if (ret) + goto err_out; + } + *out_evlist = parsed_evlist; + parsed_evlist =NULL ; +err_out: +parse_events_error__exit (&parse_error); +evlist__delete (parsed_evlist); +metricgroup__free_grouping_strs (&groupings); +return ret; +} +/** * parse_ids - Build the event string for the ids and parse them creating an * evlist. The encoded metric_ids are decoded.
@@ -1520,6 +1615,114 @@ static int parse_ids(bool metric_no_merge, struct perf_pmu *fake_pmu,return ret; } +static int hw_aware_parse_groups (struct evlist *perf_evlist, +const char *pmu,const char *str, +bool metric_no_threshold, +const char *user_requested_cpu_list, +bool system_wide, +struct perf_pmu *fake_pmu, +struct rblist *metric_events_list, +const struct pmu_metrics_table *table) +{ +struct evlist *combined_evlist =NULL ; +LIST_HEAD (metric_list); +struct metric *m; +int ret; +bool metric_no_group = false; +bool metric_no_merge = false; + +if (metric_events_list->nr_entries ==0 ) +metricgroup__rblist_init (metric_events_list); + ret =metricgroup__add_metric_list (pmu, str, metric_no_group, metric_no_threshold, + user_requested_cpu_list, + system_wide, &metric_list, table); +if (ret) + goto out; + +/* Sort metrics from largest to smallest. */ +list_sort (NULL , &metric_list, metric_list_cmp); + +if (!metric_no_merge) { +struct expr_parse_ctx *combined =NULL ; + + ret =build_combined_expr_ctx (&metric_list, &combined); + +if (!ret && combined &&hashmap__size (combined->ids)) { + ret =hw_aware_parse_ids (fake_pmu, combined, +/*modifier=*/ NULL , + &combined_evlist); + } + +if (ret) + goto out; + +if (combined) +expr__ctx_free (combined); + } + +list_for_each_entry (m, &metric_list, nd) { +struct metric_expr *expr; +struct metric_event *me; +struct evsel **metric_events; + + ret =setup_metric_events (fake_pmu ?"all" : m->pmu, m->pctx->ids, + combined_evlist, &metric_events); +if (ret) { +pr_debug ("Cannot resolve IDs for %s: %s\n" , + m->metric_name, m->metric_expr); + goto out; + } + + me =metricgroup__lookup (metric_events_list, metric_events[0 ], true); + + expr =malloc (sizeof (struct metric_expr )); +if (!expr) { + ret = -ENOMEM ; +free (metric_events); + goto out; + } + + expr->metric_refs = m->metric_refs; + m->metric_refs =NULL ; + expr->metric_expr = m->metric_expr; +if (m->modifier) { +char *tmp; + +if (asprintf (&tmp,"%s:%s" , m->metric_name, m->modifier) <0 ) + expr->metric_name =NULL ; +else + expr->metric_name = tmp; + }else { + expr->metric_name =strdup (m->metric_name); + } + +if (!expr->metric_name) { + ret = -ENOMEM ; +free (metric_events); + goto out; + } + expr->metric_threshold = m->metric_threshold; + expr->metric_unit = m->metric_unit; + expr->metric_events = metric_events; + expr->runtime = m->pctx->sctx.runtime; +list_add (&expr->nd, &me->head); + } + +if (combined_evlist) { +evlist__splice_list_tail (perf_evlist, &combined_evlist->core.entries); +evlist__delete (combined_evlist); + } + +list_for_each_entry (m, &metric_list, nd) { +if (m->evlist) +evlist__splice_list_tail (perf_evlist, &m->evlist->core.entries); + } + +out: +metricgroup__free_metrics (&metric_list); +return ret; +} + static int parse_groups(struct evlist *perf_evlist, const char *pmu, const char *str, bool metric_no_group,
@@ -1698,10 +1901,15 @@ int metricgroup__parse_groups(struct evlist *perf_evlist,if (!table)return -EINVAL ;if (hardware_aware_grouping) { +int ret;pr_debug ("Use hardware aware grouping instead of traditional metric grouping method\n" ); + ret =hw_aware_parse_groups (perf_evlist, pmu, str, + metric_no_threshold, user_requested_cpu_list, system_wide, +/*fake_pmu=*/ NULL , metric_events, table); +if (!ret) +return 0 ; } -return parse_groups (perf_evlist, pmu, str, metric_no_group, metric_no_merge, metric_no_threshold, user_requested_cpu_list, system_wide,/*fake_pmu=*/ NULL , metric_events, table);
tools/perf/util/metricgroup.h +9 -0
@@ -6,6 +6,7 @@#include <linux/rbtree.h> #include <stdbool.h> #include "pmu-events/pmu-events.h" +#include "strbuf.h" struct evlist ;struct evsel ;
@@ -66,6 +67,14 @@ struct metric_expr {int runtime; }; +/** +* Each group is one node in the group string list. +*/ +struct metricgroup__group_strs { +struct list_head nd; +struct strbuf grouping_str; +}; +struct metric_event *metricgroup__lookup (struct rblist *metric_events,struct evsel *evsel,bool create);
Signature · 1 line
2.39.3
Re: [RFC PATCH v2 02/17] perf stat: Add basic functions for the hardware-grouping stat cmd option
Quoted 4 lines · On 2023-10-13 9:51 p.m., [email protected] wrote:
From: Weilin Wang <[email protected]>
Add the first set of functions for the hardware-grouping method. Function
hw_awre_parse_groups()is the entry point of this metric grouping method.
hw_aware_parse_groups()
Quoted 3 lines
It does metric grouping on a combined list of events and will create a list of grouping strings as final results of the grouping method. These grouping strings will be used in the same mannor as existing metric grouping
manner
Quoted 25 lines
process.
This method will fall back to normal grouping when hardware aware grouping return with err so that perf stat still executes and returns with correct result.
Signed-off-by Weilin Wang <[email protected]>--- tools/perf/util/metricgroup.c | 210 +++++++++++++++++++++++++++++++++- tools/perf/util/metricgroup.h | 9 ++ 2 files changed, 218 insertions(+), 1 deletion(-)tools/perf/util/metricgroup.c +5
-0@@ -1432,6 +1432,101 @@ static int build_combined_expr_ctx(const struct list_head *metric_list,return ret; } +/** +* hw_aware_build_grouping - Build event groupings by reading counter +* requirement of the events and counter available on the system from +* pmu-events. +* @ctx: the event identifiers parsed from metrics.
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list, · line 1449
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+* @groupings: header to the list of final event grouping. +* @modifier: any modifiers added to the events. +*/ +static int hw_aware_build_grouping (struct expr_parse_ctx *ctx__maybe_unused , +struct list_head *groupings__maybe_unused , +const char *modifier__maybe_unused ) +{ +int ret =0 ; + +pr_debug ("This is a placeholder\n" );
I think the patch series still use a reversed way to organize the patches. So you have to insert such the useless place holder here.
I think a better way is to start the series from the essential blocks, then build more complex functions on top of the blocks, and finally add/enable the new feature.
Quoted 27 lines
tools/perf/util/metricgroup.c +27
-0+return ret; +} + +static void group_str_free (struct metricgroup__group_strs *g) +{ +if (!g) +return ; + +strbuf_release (&g->grouping_str); +free (g); +} + +static void metricgroup__free_grouping_strs (struct list_head + *grouping_strs) +{ +struct metricgroup__group_strs *g, *tmp; + +list_for_each_entry_safe (g, tmp, grouping_strs, nd) { +list_del_init (&g->nd); +group_str_free (g); + } +} + +/** +* hw_aware_parse_ids - Build the event string for the ids and parse them +* creating an evlist. The encoded metric_ids are decoded. Events are placed +* into groups based on event counter requirements and counter availabilities of
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list, · line 1486
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+* the system. +* @metric_no_merge: is metric sharing explicitly disabled. +* @fake_pmu: used when testing metrics not supported by the current CPU. +* @ids: the event identifiers parsed from a metric. +* @modifier: any modifiers added to the events. +* @out_evlist: the created list of events. +*/ +static int hw_aware_parse_ids (struct perf_pmu *fake_pmu, +struct expr_parse_ctx *ids,const char *modifier, +struct evlist **out_evlist)
The hw_aware_parse_ids() is very similar to parse_ids(). Can we factor a common function to avoid the duplication?
Quoted 48 lines
tools/perf/util/metricgroup.c +43
-0+{ +struct parse_events_error parse_error; +struct evlist *parsed_evlist; +LIST_HEAD (groupings); +struct metricgroup__group_strs *group; +int ret; + + *out_evlist =NULL ; + ret =hw_aware_build_grouping (ids, &groupings, modifier); +if (ret) { +metricgroup__free_grouping_strs (&groupings); +return ret; + } + + parsed_evlist =evlist__new (); +if (!parsed_evlist) { + ret = -ENOMEM ; + goto err_out; + } +list_for_each_entry (group, &groupings, nd) { +struct strbuf *events = &group->grouping_str; + +pr_debug ("Parsing metric events '%s'\n" , events->buf); +parse_events_error__init (&parse_error); + ret =__parse_events (parsed_evlist, events->buf,/*pmu_filter=*/ NULL , + &parse_error, fake_pmu,/*warn_if_reordered=*/ false); +if (ret) { +parse_events_error__print (&parse_error, events->buf); + goto err_out; + } + ret =decode_all_metric_ids (parsed_evlist, modifier); +if (ret) + goto err_out; + } + *out_evlist = parsed_evlist; + parsed_evlist =NULL ; +err_out: +parse_events_error__exit (&parse_error); +evlist__delete (parsed_evlist); +metricgroup__free_grouping_strs (&groupings); +return ret; +} +/** * parse_ids - Build the event string for the ids and parse them creating an * evlist. The encoded metric_ids are decoded. @@ -1520,6 +1615,114 @@ static int parse_ids(bool metric_no_merge, struct perf_pmu *fake_pmu,return ret;
tools/perf/util/metricgroup.c static int parse_ids(bool metric_no_merge, struct perf_pmu *fake_pmu, · line 1625
Quoted 10 lines
tools/perf/util/metricgroup.c +8
-0} +static int hw_aware_parse_groups (struct evlist *perf_evlist, +const char *pmu,const char *str, +bool metric_no_threshold, +const char *user_requested_cpu_list, +bool system_wide, +struct perf_pmu *fake_pmu, +struct rblist *metric_events_list, +const struct pmu_metrics_table *table)
The hw_aware_parse_groups() is very similar to parse_groups(). Please factor out a common function to avoid the duplication as well.
Thanks,
Kan
Quoted 147 lines
tools/perf/util/metricgroup.c +106
-1+{ +struct evlist *combined_evlist =NULL ; +LIST_HEAD (metric_list); +struct metric *m; +int ret; +bool metric_no_group = false; +bool metric_no_merge = false; + +if (metric_events_list->nr_entries ==0 ) +metricgroup__rblist_init (metric_events_list); + ret =metricgroup__add_metric_list (pmu, str, metric_no_group, metric_no_threshold, + user_requested_cpu_list, + system_wide, &metric_list, table); +if (ret) + goto out; + +/* Sort metrics from largest to smallest. */ +list_sort (NULL , &metric_list, metric_list_cmp); + +if (!metric_no_merge) { +struct expr_parse_ctx *combined =NULL ; + + ret =build_combined_expr_ctx (&metric_list, &combined); + +if (!ret && combined &&hashmap__size (combined->ids)) { + ret =hw_aware_parse_ids (fake_pmu, combined, +/*modifier=*/ NULL , + &combined_evlist); + } + +if (ret) + goto out; + +if (combined) +expr__ctx_free (combined); + } + +list_for_each_entry (m, &metric_list, nd) { +struct metric_expr *expr; +struct metric_event *me; +struct evsel **metric_events; + + ret =setup_metric_events (fake_pmu ?"all" : m->pmu, m->pctx->ids, + combined_evlist, &metric_events); +if (ret) { +pr_debug ("Cannot resolve IDs for %s: %s\n" , + m->metric_name, m->metric_expr); + goto out; + } + + me =metricgroup__lookup (metric_events_list, metric_events[0 ], true); + + expr =malloc (sizeof (struct metric_expr )); +if (!expr) { + ret = -ENOMEM ; +free (metric_events); + goto out; + } + + expr->metric_refs = m->metric_refs; + m->metric_refs =NULL ; + expr->metric_expr = m->metric_expr; +if (m->modifier) { +char *tmp; + +if (asprintf (&tmp,"%s:%s" , m->metric_name, m->modifier) <0 ) + expr->metric_name =NULL ; +else + expr->metric_name = tmp; + }else { + expr->metric_name =strdup (m->metric_name); + } + +if (!expr->metric_name) { + ret = -ENOMEM ; +free (metric_events); + goto out; + } + expr->metric_threshold = m->metric_threshold; + expr->metric_unit = m->metric_unit; + expr->metric_events = metric_events; + expr->runtime = m->pctx->sctx.runtime; +list_add (&expr->nd, &me->head); + } + +if (combined_evlist) { +evlist__splice_list_tail (perf_evlist, &combined_evlist->core.entries); +evlist__delete (combined_evlist); + } + +list_for_each_entry (m, &metric_list, nd) { +if (m->evlist) +evlist__splice_list_tail (perf_evlist, &m->evlist->core.entries); + } + +out: +metricgroup__free_metrics (&metric_list); +return ret; +} + static int parse_groups(struct evlist *perf_evlist, const char *pmu, const char *str, bool metric_no_group,@@ -1698,10 +1901,15 @@ int metricgroup__parse_groups(struct evlist *perf_evlist,if (!table)return -EINVAL ;if (hardware_aware_grouping) { +int ret;pr_debug ("Use hardware aware grouping instead of traditional metric grouping method\n" ); + ret =hw_aware_parse_groups (perf_evlist, pmu, str, + metric_no_threshold, user_requested_cpu_list, system_wide, +/*fake_pmu=*/ NULL , metric_events, table); +if (!ret) +return 0 ; } -return parse_groups (perf_evlist, pmu, str, metric_no_group, metric_no_merge, metric_no_threshold, user_requested_cpu_list, system_wide,/*fake_pmu=*/ NULL , metric_events, table);tools/perf/util/metricgroup.h +9
-0@@ -6,6 +6,7 @@#include <linux/rbtree.h> #include <stdbool.h> #include "pmu-events/pmu-events.h" +#include "strbuf.h" struct evlist ;struct evsel ;@@ -66,6 +67,14 @@ struct metric_expr {int runtime; }; +/** +* Each group is one node in the group string list. +*/ +struct metricgroup__group_strs { +struct list_head nd; +struct strbuf grouping_str; +}; +struct metric_event *metricgroup__lookup (struct rblist *metric_events,struct evsel *evsel,bool create);
RFC PATCH v2 3/17 perf pmu-events: Add functions in jevent.py to parse counter and event info for hardware aware grouping
In reply to [email protected]From: Weilin Wang <[email protected]>
These functions are added to parse event counter restrictions and counter availability info from json files so that the metric grouping method could do grouping based on the counter restriction of events and the counters that are available on the system.
--- tools/perf/pmu-events/jevents.py | 172 ++++++++++++++++++++++++++++- tools/perf/pmu-events/pmu-events.h | 25 ++++- 2 files changed, 189 insertions(+), 8 deletions(-)
tools/perf/pmu-events/jevents.py +168 -4
@@ -23,6 +23,8 @@ _metric_tables = [] _sys_metric_tables = [] # Mapping between sys event table names and sys metric table names. _sys_event_table_to_metric_table_mapping = {} +# List of regular PMU counter layout tables. +_pmu_layouts_tables = [] # Map from an event name to an architecture standard # JsonEvent. Architecture standard events are in json files in the top # f'{_args.starting_dir}/{_args.arch}' directory.
@@ -31,6 +33,9 @@ _arch_std_events = {} _pending_events = [] # Name of events table to be written out _pending_events_tblname = None +# PMU counter layout to write out when the table is closed +_pending_pmu_counts = [] # Name of PMU counter layout table to be written out +_pending_pmu_counts_tblname = None # Metrics to write out when the table is closed _pending_metrics = [] # Name of metrics table to be written out
@@ -47,10 +52,17 @@ _json_event_attributes = [ 'event', # Short things in alphabetical order. 'compat', 'deprecated', 'perpkg', 'unit', + # Counter this event could use + 'counter', # Longer things (the last won't be iterated over during decompress). 'long_desc' ] +# Attributes that are in pmu_unit_layout. +_json_layout_attributes = [ + 'pmu', 'desc', 'size', 'fixed_size' +] + # Attributes that are in pmu_metric rather than pmu_event. _json_metric_attributes = [ 'metric_name', 'metric_group', 'metric_expr', 'metric_threshold',
@@ -58,7 +70,9 @@ _json_metric_attributes = [ 'default_metricgroup_name', 'aggr_mode', 'event_grouping' ] # Attributes that are bools or enum int values, encoded as '0', '1',... -_json_enum_attributes = ['aggr_mode', 'deprecated', 'event_grouping', 'perpkg'] +_json_enum_attributes = ['aggr_mode', 'deprecated', 'event_grouping', 'perpkg', + 'size', 'fixed_size' +] def removesuffix(s: str, suffix: str) -> str: """Remove the suffix from a string
@@ -289,6 +303,7 @@ class JsonEvent: 'cpu_core': 'cpu_core', 'cpu_atom': 'cpu_atom', 'ali_drw': 'ali_drw', + 'Core': 'default_core', } return table[unit] if unit in table else f'uncore_{unit.lower()}'
@@ -314,6 +329,9 @@ class JsonEvent: if 'Errata' in jd: extra_desc += ' Spec update: ' + jd['Errata'] self.pmu = unit_to_pmu(jd.get('Unit')) + self.counter = jd.get('Counter') + self.size = jd.get('Size') + self.fixed_size = jd.get('FixedSize') filter = jd.get('Filter') self.unit = jd.get('ScaleUnit') self.perpkg = jd.get('PerPkg')
@@ -377,8 +395,16 @@ class JsonEvent: s += f'\t{attr} = {value},\n' return s + '}' - def build_c_string(self, metric: bool) -> str: + def build_c_string(self, metric: bool, layout: bool = False) -> str: s = '' + if layout: + for attr in _json_layout_attributes: + x = getattr(self, attr) + if attr in _json_enum_attributes: + s += x if x else '0' + else: + s += f'{x}\\000' if x else '\\000' + return s for attr in _json_metric_attributes if metric else _json_event_attributes: x = getattr(self, attr) if metric and x and attr == 'metric_expr':
@@ -393,10 +419,10 @@ class JsonEvent: s += f'{x}\\000' if x else '\\000' return s - def to_c_string(self, metric: bool) -> str: + def to_c_string(self, metric: bool, layout: bool = False) -> str: """Representation of the event as a C struct initializer.""" - s = self.build_c_string(metric) + s = self.build_c_string(metric, layout) return f'{{ { _bcs.offsets[s] } }}, /* {s} */\n'
@@ -433,6 +459,8 @@ def preprocess_arch_std_files(archpath: str) -> None: _arch_std_events[event.name.lower()] = event if event.metric_name: _arch_std_events[event.metric_name.lower()] = event + if event.size: + _arch_std_events[event.pmu.lower()] = event def add_events_table_entries(item: os.DirEntry, topic: str) -> None:
@@ -442,6 +470,8 @@ def add_events_table_entries(item: os.DirEntry, topic: str) -> None: _pending_events.append(e) if e.metric_name: _pending_metrics.append(e) + if e.size: + _pending_pmu_counts.append(e) def print_pending_events() -> None:
@@ -555,6 +585,33 @@ const struct pmu_table_entry {_pending_metrics_tblname}[] = {{ """) _args.output_file.write('};\n\n') +def print_pending_pmu_counts() -> None: + + def pmu_counts_cmp_key(j: JsonEvent) -> Tuple[bool, str, str]: + def fix_none(s: Optional[str]) -> str: + if s is None: + return '' + return s + + return (j.desc is not None, fix_none(j.pmu), fix_none(j.size)) + + global _pending_pmu_counts + if not _pending_pmu_counts: + return + + global _pending_pmu_counts_tblname + global pmu_layouts_tables + _pmu_layouts_tables.append(_pending_pmu_counts_tblname) + + _args.output_file.write( + f'static const struct compact_pmu_event {_pending_pmu_counts_tblname}[] = {{\n') + + for pmu_layout in sorted(_pending_pmu_counts, key=pmu_counts_cmp_key): + _args.output_file.write(pmu_layout.to_c_string(metric=False, layout=True)) + _pending_pmu_counts = [] + + _args.output_file.write('};\n\n') + def get_topic(topic: str) -> str: if topic.endswith('metrics.json'): return 'metrics'
@@ -595,6 +652,8 @@ def preprocess_one_file(parents: Sequence[str], item: os.DirEntry) -> None: if event.metric_name: _bcs.add(pmu_name, metric=True) _bcs.add(event.build_c_string(metric=True), metric=True) + if event.size: + _bcs.add(event.build_c_string(metric=False, layout=True), metric=False) def process_one_file(parents: Sequence[str], item: os.DirEntry) -> None: """Process a JSON file during the main walk."""
@@ -608,11 +667,14 @@ def process_one_file(parents: Sequence[str], item: os.DirEntry) -> None: if item.is_dir() and is_leaf_dir(item.path): print_pending_events() print_pending_metrics() + print_pending_pmu_counts() global _pending_events_tblname _pending_events_tblname = file_name_to_table_name('pmu_events_', parents, item.name) global _pending_metrics_tblname _pending_metrics_tblname = file_name_to_table_name('pmu_metrics_', parents, item.name) + global _pending_pmu_counts_tblname + _pending_pmu_counts_tblname = file_name_to_table_name('pmu_layouts_', parents, item.name) if item.name == 'sys': _sys_event_table_to_metric_table_mapping[_pending_events_tblname] = _pending_metrics_tblname
@@ -646,6 +708,12 @@ struct pmu_metrics_table { uint32_t num_pmus; }; +/* Struct used to make the PMU counter layout table implementation opaque to callers. */ +struct pmu_layouts_table { + const struct compact_pmu_event *entries; + size_t length; +}; + /* * Map a CPU to its table of PMU events. The CPU is identified by the * cpuid field, which is an arch-specific identifier for the CPU.
@@ -659,6 +727,7 @@ struct pmu_events_map { const char *cpuid; struct pmu_events_table event_table; struct pmu_metrics_table metric_table; + struct pmu_layouts_table layout_table; }; /*
@@ -703,6 +772,12 @@ const struct pmu_events_map pmu_events_map[] = { metric_size = '0' if event_size == '0' and metric_size == '0': continue + layout_tblname = file_name_to_table_name('pmu_layouts_', [], row[2].replace('/', '_')) + if layout_tblname in _pmu_layouts_tables: + layout_size = f'ARRAY_SIZE({layout_tblname})' + else: + layout_tblname = 'NULL' + layout_size = '0' cpuid = row[0].replace('\\', '\\\\') _args.output_file.write(f"""{{ \t.arch = "{arch}",
@@ -714,6 +789,10 @@ const struct pmu_events_map pmu_events_map[] = { \t.metric_table = {{ \t\t.pmus = {metric_tblname}, \t\t.num_pmus = {metric_size} +\t}}, +\t.layout_table = {{ +\t\t.entries = {layout_tblname}, +\t\t.length = {layout_size} \t}} }}, """)
@@ -724,6 +803,7 @@ const struct pmu_events_map pmu_events_map[] = { \t.cpuid = 0, \t.event_table = { 0, 0 }, \t.metric_table = { 0, 0 }, +\t.layout_table = { 0, 0 }, } }; """)
@@ -812,6 +892,24 @@ static void decompress_metric(int offset, struct pmu_metric *pm) _args.output_file.write('\twhile (*p++);') _args.output_file.write("""} +static void decompress_layout(int offset, struct pmu_layout *pm) +{ +\tconst char *p = &big_c_string[offset]; +""") + for attr in _json_layout_attributes: + _args.output_file.write(f'\n\tpm->{attr} = ') + if attr in _json_enum_attributes: + _args.output_file.write("*p - '0';\n") + else: + _args.output_file.write("(*p == '\\0' ? NULL : p);\n") + if attr == _json_layout_attributes[-1]: + continue + if attr in _json_enum_attributes: + _args.output_file.write('\tp++;') + else: + _args.output_file.write('\twhile (*p++);') + _args.output_file.write("""} + static int pmu_events_table__for_each_event_pmu(const struct pmu_events_table *table, const struct pmu_table_entry *pmu, pmu_event_iter_fn fn,
@@ -967,6 +1065,21 @@ int pmu_metrics_table__for_each_metric(const struct pmu_metrics_table *table, return 0; } +int pmu_layouts_table__for_each_layout(const struct pmu_layouts_table *table, + pmu_layout_iter_fn fn, + void *data) { + for (size_t i = 0; i < table->length; i++) { + struct pmu_layout pm; + int ret; + + decompress_layout(table->entries[i].offset, &pm); + ret = fn(&pm, data); + if (ret) + return ret; + } + return 0; +} + const struct pmu_events_table *perf_pmu__find_events_table(struct perf_pmu *pmu) { const struct pmu_events_table *table = NULL;
@@ -1031,6 +1144,33 @@ const struct pmu_metrics_table *perf_pmu__find_metrics_table(struct perf_pmu *pm return table; } +const struct pmu_layouts_table *perf_pmu__find_layouts_table(struct perf_pmu *pmu) +{ + const struct pmu_layouts_table *table = NULL; + char *cpuid = perf_pmu__getcpuid(pmu); + int i; + + /* on some platforms which uses cpus map, cpuid can be NULL for + * PMUs other than CORE PMUs. + */ + if (!cpuid) + return NULL; + + i = 0; + for (;;) { + const struct pmu_events_map *map = &pmu_events_map[i++]; + if (!map->arch) + break; + + if (!strcmp_cpuid_str(map->cpuid, cpuid)) { + table = &map->layout_table; + break; + } + } + free(cpuid); + return table; +} + const struct pmu_events_table *find_core_events_table(const char *arch, const char *cpuid) { for (const struct pmu_events_map *tables = &pmu_events_map[0];
@@ -1052,6 +1192,16 @@ const struct pmu_metrics_table *find_core_metrics_table(const char *arch, const } return NULL; } +const struct pmu_layouts_table *find_core_layouts_table(const char *arch, const char *cpuid) +{ + for (const struct pmu_events_map *tables = &pmu_events_map[0]; + tables->arch; + tables++) { + if (!strcmp(tables->arch, arch) && !strcmp_cpuid_str(tables->cpuid, cpuid)) + return &tables->layout_table; + } + return NULL; +} int pmu_for_each_core_event(pmu_event_iter_fn fn, void *data) {
@@ -1080,6 +1230,19 @@ int pmu_for_each_core_metric(pmu_metric_iter_fn fn, void *data) return 0; } +int pmu_for_each_core_layout(pmu_layout_iter_fn fn, void *data) +{ + for (const struct pmu_events_map *tables = &pmu_events_map[0]; + tables->arch; + tables++) { + int ret = pmu_layouts_table__for_each_layout(&tables->layout_table, fn, data); + + if (ret) + return ret; + } + return 0; +} + const struct pmu_events_table *find_sys_events_table(const char *name) { for (const struct pmu_sys_events *tables = &pmu_sys_event_tables[0];
@@ -1236,6 +1399,7 @@ struct pmu_table_entry { ftw(arch_path, [], process_one_file) print_pending_events() print_pending_metrics() + print_pending_pmu_counts() print_mapping_table(archs) print_system_mapping_table()
tools/perf/pmu-events/pmu-events.h +21 -4
@@ -45,6 +45,7 @@ struct pmu_event {const char *desc;const char *topic;const char *long_desc; +const char *counter;const char *pmu;const char *unit;bool perpkg;
@@ -67,8 +68,16 @@ struct pmu_metric {enum metric_event_groups event_grouping; }; +struct pmu_layout { +const char *pmu; +const char *desc; +int size; +int fixed_size; +}; +struct pmu_events_table ;struct pmu_metrics_table ; +struct pmu_layouts_table ; typedef int (*pmu_event_iter_fn)(const struct pmu_event *pe, const struct pmu_events_table *table,
@@ -78,15 +87,20 @@ typedef int (*pmu_metric_iter_fn)(const struct pmu_metric *pm,const struct pmu_metrics_table *table, void *data); +typedef int (*pmu_layout_iter_fn )(const struct pmu_layout *pm, +void *data); +int pmu_events_table__for_each_event (const struct pmu_events_table *table,struct perf_pmu *pmu,pmu_event_iter_fn fn,void *data);int pmu_events_table__find_event (const struct pmu_events_table *table, -struct perf_pmu *pmu, -const char *name, -pmu_event_iter_fn fn, -void *data); +struct perf_pmu *pmu, +const char *name, +pmu_event_iter_fn fn, +void *data); +int pmu_layouts_table__for_each_layout (const struct pmu_layouts_table *table,pmu_layout_iter_fn fn, +void *data);size_t pmu_events_table__num_events (const struct pmu_events_table *table,struct perf_pmu *pmu);
@@ -95,10 +109,13 @@ int pmu_metrics_table__for_each_metric(const struct pmu_metrics_table *table, pmconst struct pmu_events_table *perf_pmu__find_events_table (struct perf_pmu *pmu);const struct pmu_metrics_table *perf_pmu__find_metrics_table (struct perf_pmu *pmu); +const struct pmu_layouts_table *perf_pmu__find_layouts_table (struct perf_pmu *pmu);const struct pmu_events_table *find_core_events_table (const char *arch,const char *cpuid);const struct pmu_metrics_table *find_core_metrics_table (const char *arch,const char *cpuid); +const struct pmu_layouts_table *find_core_layouts_table (const char *arch,const char *cpuid);int pmu_for_each_core_event (pmu_event_iter_fn fn,void *data);int pmu_for_each_core_metric (pmu_metric_iter_fn fn,void *data); +int pmu_for_each_core_layout (pmu_layout_iter_fn fn,void *data);const struct pmu_events_table *find_sys_events_table (const char *name);const struct pmu_metrics_table *find_sys_metrics_table (const char *name);
Signature · 1 line
2.39.3
RFC PATCH v2 4/17 perf pmu-events: Add counter info into JSON files for SapphireRapids
In reply to [email protected]From: Weilin Wang <[email protected]>
The added new fields are required in the hardware aware grouping method. Currently, the format of added fields are temporary for experiments.
--- .../arch/x86/sapphirerapids/cache.json | 104 ++++ .../arch/x86/sapphirerapids/counter.json | 17 + .../x86/sapphirerapids/floating-point.json | 25 + .../arch/x86/sapphirerapids/frontend.json | 98 +++- .../arch/x86/sapphirerapids/memory.json | 44 ++ .../arch/x86/sapphirerapids/other.json | 40 ++ .../arch/x86/sapphirerapids/pipeline.json | 118 ++++ .../arch/x86/sapphirerapids/uncore-cache.json | 534 +++++++++++++++++- .../arch/x86/sapphirerapids/uncore-cxl.json | 56 ++ .../sapphirerapids/uncore-interconnect.json | 476 ++++++++++++++++ .../arch/x86/sapphirerapids/uncore-io.json | 373 ++++++++++++ .../x86/sapphirerapids/uncore-memory.json | 391 +++++++++++++ .../arch/x86/sapphirerapids/uncore-power.json | 24 + .../x86/sapphirerapids/virtual-memory.json | 20 + 14 files changed, 2304 insertions(+), 16 deletions(-) create mode 100644 tools/perf/pmu-events/arch/x86/sapphirerapids/counter.json
tools/perf/pmu-events/arch/x86/sapphirerapids/cache.json +104 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "L1D.HWPF_MISS", + "Counter": "0,1,2,3", "EventCode": "0x51", "EventName": "L1D.HWPF_MISS", "SampleAfterValue": "1000003",
@@ -8,6 +9,7 @@ }, { "BriefDescription": "Counts the number of cache lines replaced in L1 data cache.", + "Counter": "0,1,2,3", "EventCode": "0x51", "EventName": "L1D.REPLACEMENT", "PublicDescription": "Counts L1D data line replacements including opportunistic replacements, and replacements that require stall-for-replace or block-for-replace.",
@@ -16,6 +18,7 @@ }, { "BriefDescription": "Number of cycles a demand request has waited due to L1D Fill Buffer (FB) unavailability.", + "Counter": "0,1,2,3", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.FB_FULL", "PublicDescription": "Counts number of cycles a demand request has waited due to L1D Fill Buffer (FB) unavailability. Demand requests include cacheable/uncacheable demand load, store, lock or SW prefetch accesses.",
@@ -24,6 +27,7 @@ }, { "BriefDescription": "Number of phases a demand request has waited due to L1D Fill Buffer (FB) unavailability.", + "Counter": "0,1,2,3", "CounterMask": "1", "EdgeDetect": "1", "EventCode": "0x48",
@@ -34,6 +38,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event L1D_PEND_MISS.L2_STALLS", + "Counter": "0,1,2,3", "Deprecated": "1", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.L2_STALL",
@@ -42,6 +47,7 @@ }, { "BriefDescription": "Number of cycles a demand request has waited due to L1D due to lack of L2 resources.", + "Counter": "0,1,2,3", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.L2_STALLS", "PublicDescription": "Counts number of cycles a demand request has waited due to L1D due to lack of L2 resources. Demand requests include cacheable/uncacheable demand load, store, lock or SW prefetch accesses.",
@@ -50,6 +56,7 @@ }, { "BriefDescription": "Number of L1D misses that are outstanding", + "Counter": "0,1,2,3", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.PENDING", "PublicDescription": "Counts number of L1D misses that are outstanding in each cycle, that is each cycle the number of Fill Buffers (FB) outstanding required by Demand Reads. FB either is held by demand loads, or it is held by non-demand loads and gets hit at least once by demand. The valid outstanding interval is defined until the FB deallocation by one of the following ways: from FB allocation, if FB is allocated by demand from the demand Hit FB, if it is allocated by hardware or software prefetch. Note: In the L1D, a Demand Read contains cacheable or noncacheable demand loads, including ones causing cache-line splits and reads due to page walks resulted from any request type.",
@@ -58,6 +65,7 @@ }, { "BriefDescription": "Cycles with L1D load Misses outstanding.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.PENDING_CYCLES",
@@ -67,6 +75,7 @@ }, { "BriefDescription": "L2 cache lines filling L2", + "Counter": "0,1,2,3", "EventCode": "0x25", "EventName": "L2_LINES_IN.ALL", "PublicDescription": "Counts the number of L2 cache lines filling the L2. Counting does not cover rejects.",
@@ -75,6 +84,7 @@ }, { "BriefDescription": "L2_LINES_OUT.NON_SILENT", + "Counter": "0,1,2,3", "EventCode": "0x26", "EventName": "L2_LINES_OUT.NON_SILENT", "SampleAfterValue": "200003",
@@ -82,6 +92,7 @@ }, { "BriefDescription": "Non-modified cache lines that are silently dropped by L2 cache when triggered by an L2 cache fill.", + "Counter": "0,1,2,3", "EventCode": "0x26", "EventName": "L2_LINES_OUT.SILENT", "PublicDescription": "Counts the number of lines that are silently dropped by L2 cache when triggered by an L2 cache fill. These lines are typically in Shared or Exclusive state. A non-threaded event.",
@@ -90,6 +101,7 @@ }, { "BriefDescription": "Cache lines that have been L2 hardware prefetched but not used by demand accesses", + "Counter": "0,1,2,3", "EventCode": "0x26", "EventName": "L2_LINES_OUT.USELESS_HWPF", "PublicDescription": "Counts the number of cache lines that have been prefetched by the L2 hardware prefetcher but not used by demand access when evicted from the L2 cache",
@@ -98,6 +110,7 @@ }, { "BriefDescription": "All accesses to L2 cache [This event is alias to L2_RQSTS.REFERENCES]", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_REQUEST.ALL", "PublicDescription": "Counts all requests that were hit or true misses in L2 cache. True-miss excludes misses that were merged with ongoing L2 misses. [This event is alias to L2_RQSTS.REFERENCES]",
@@ -106,6 +119,7 @@ }, { "BriefDescription": "Read requests with true-miss in L2 cache. [This event is alias to L2_RQSTS.MISS]", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_REQUEST.MISS", "PublicDescription": "Counts read requests of any type with true-miss in the L2 cache. True-miss excludes L2 misses that were merged with ongoing L2 misses. [This event is alias to L2_RQSTS.MISS]",
@@ -114,6 +128,7 @@ }, { "BriefDescription": "L2 code requests", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_CODE_RD", "PublicDescription": "Counts the total number of L2 code requests.",
@@ -122,6 +137,7 @@ }, { "BriefDescription": "Demand Data Read access L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_DEMAND_DATA_RD", "PublicDescription": "Counts Demand Data Read requests accessing the L2 cache. These requests may hit or miss L2 cache. True-miss exclude misses that were merged with ongoing L2 misses. An access is counted once.",
@@ -130,6 +146,7 @@ }, { "BriefDescription": "Demand requests that miss L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_DEMAND_MISS", "PublicDescription": "Counts demand requests that miss L2 cache.",
@@ -138,6 +155,7 @@ }, { "BriefDescription": "Demand requests to L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_DEMAND_REFERENCES", "PublicDescription": "Counts demand requests to L2 cache.",
@@ -146,6 +164,7 @@ }, { "BriefDescription": "L2_RQSTS.ALL_HWPF", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_HWPF", "SampleAfterValue": "200003",
@@ -153,6 +172,7 @@ }, { "BriefDescription": "RFO requests to L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_RFO", "PublicDescription": "Counts the total number of RFO (read for ownership) requests to L2 cache. L2 RFO requests include both L1D demand RFO misses as well as L1D RFO prefetches.",
@@ -161,6 +181,7 @@ }, { "BriefDescription": "L2 cache hits when fetching instructions, code reads.", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.CODE_RD_HIT", "PublicDescription": "Counts L2 cache hits when fetching instructions, code reads.",
@@ -169,6 +190,7 @@ }, { "BriefDescription": "L2 cache misses when fetching instructions", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.CODE_RD_MISS", "PublicDescription": "Counts L2 cache misses when fetching instructions.",
@@ -177,6 +199,7 @@ }, { "BriefDescription": "Demand Data Read requests that hit L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.DEMAND_DATA_RD_HIT", "PublicDescription": "Counts the number of demand Data Read requests initiated by load instructions that hit L2 cache.",
@@ -185,6 +208,7 @@ }, { "BriefDescription": "Demand Data Read miss L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.DEMAND_DATA_RD_MISS", "PublicDescription": "Counts demand Data Read requests with true-miss in the L2 cache. True-miss excludes misses that were merged with ongoing L2 misses. An access is counted once.",
@@ -193,6 +217,7 @@ }, { "BriefDescription": "L2_RQSTS.HWPF_MISS", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.HWPF_MISS", "SampleAfterValue": "200003",
@@ -200,6 +225,7 @@ }, { "BriefDescription": "Read requests with true-miss in L2 cache. [This event is alias to L2_REQUEST.MISS]", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.MISS", "PublicDescription": "Counts read requests of any type with true-miss in the L2 cache. True-miss excludes L2 misses that were merged with ongoing L2 misses. [This event is alias to L2_REQUEST.MISS]",
@@ -208,6 +234,7 @@ }, { "BriefDescription": "All accesses to L2 cache [This event is alias to L2_REQUEST.ALL]", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.REFERENCES", "PublicDescription": "Counts all requests that were hit or true misses in L2 cache. True-miss excludes misses that were merged with ongoing L2 misses. [This event is alias to L2_REQUEST.ALL]",
@@ -216,6 +243,7 @@ }, { "BriefDescription": "RFO requests that hit L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.RFO_HIT", "PublicDescription": "Counts the RFO (Read-for-Ownership) requests that hit L2 cache.",
@@ -224,6 +252,7 @@ }, { "BriefDescription": "RFO requests that miss L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.RFO_MISS", "PublicDescription": "Counts the RFO (Read-for-Ownership) requests that miss L2 cache.",
@@ -232,6 +261,7 @@ }, { "BriefDescription": "SW prefetch requests that hit L2 cache.", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.SWPF_HIT", "PublicDescription": "Counts Software prefetch requests that hit the L2 cache. Accounts for PREFETCHNTA and PREFETCHT0/1/2 instructions when FB is not full.",
@@ -240,6 +270,7 @@ }, { "BriefDescription": "SW prefetch requests that miss L2 cache.", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.SWPF_MISS", "PublicDescription": "Counts Software prefetch requests that miss the L2 cache. Accounts for PREFETCHNTA and PREFETCHT0/1/2 instructions when FB is not full.",
@@ -248,6 +279,7 @@ }, { "BriefDescription": "Core-originated cacheable requests that missed L3 (Except hardware prefetches to the L3)", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0x2e", "EventName": "LONGEST_LAT_CACHE.MISS", "PublicDescription": "Counts core-originated cacheable requests that miss the L3 cache (Longest Latency cache). Requests include data and code reads, Reads-for-Ownership (RFOs), speculative accesses and hardware prefetches to the L1 and L2. It does not include hardware prefetches to the L3, and may not count other types of requests to the L3.",
@@ -256,6 +288,7 @@ }, { "BriefDescription": "Core-originated cacheable requests that refer to L3 (Except hardware prefetches to the L3)", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0x2e", "EventName": "LONGEST_LAT_CACHE.REFERENCE", "PublicDescription": "Counts core-originated cacheable requests to the L3 cache (Longest Latency cache). Requests include data and code reads, Reads-for-Ownership (RFOs), speculative accesses and hardware prefetches to the L1 and L2. It does not include hardware prefetches to the L3, and may not count other types of requests to the L3.",
@@ -264,6 +297,7 @@ }, { "BriefDescription": "Retired load instructions.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.ALL_LOADS",
@@ -274,6 +308,7 @@ }, { "BriefDescription": "Retired store instructions.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.ALL_STORES",
@@ -284,6 +319,7 @@ }, { "BriefDescription": "All retired memory instructions.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.ANY",
@@ -294,6 +330,7 @@ }, { "BriefDescription": "Retired load instructions with locked access.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.LOCK_LOADS",
@@ -304,6 +341,7 @@ }, { "BriefDescription": "Retired load instructions that split across a cacheline boundary.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.SPLIT_LOADS",
@@ -314,6 +352,7 @@ }, { "BriefDescription": "Retired store instructions that split across a cacheline boundary.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.SPLIT_STORES",
@@ -324,6 +363,7 @@ }, { "BriefDescription": "Retired load instructions that miss the STLB.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.STLB_MISS_LOADS",
@@ -334,6 +374,7 @@ }, { "BriefDescription": "Retired store instructions that miss the STLB.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.STLB_MISS_STORES",
@@ -344,6 +385,7 @@ }, { "BriefDescription": "Completed demand load uops that miss the L1 d-cache.", + "Counter": "0,1,2,3", "EventCode": "0x43", "EventName": "MEM_LOAD_COMPLETED.L1_MISS_ANY", "PublicDescription": "Number of completed demand load requests that missed the L1 data cache including shadow misses (FB hits, merge to an ongoing L1D miss)",
@@ -352,6 +394,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources were HitM responses from shared L3", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD",
@@ -362,6 +405,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources were L3 hit and cross-core snoop missed in on-pkg core cache.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS",
@@ -372,6 +416,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources were hits in L3 without snoops required", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_NONE",
@@ -382,6 +427,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources were L3 and cross-core snoop hits in on-pkg core cache", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_NO_FWD",
@@ -392,6 +438,7 @@ }, { "BriefDescription": "Retired load instructions which data sources missed L3 but serviced from local dram", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM",
@@ -402,6 +449,7 @@ }, { "BriefDescription": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM",
@@ -411,6 +459,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources was forwarded from a remote cache", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD",
@@ -421,6 +470,7 @@ }, { "BriefDescription": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM",
@@ -430,6 +480,7 @@ }, { "BriefDescription": "Retired load instructions with remote Intel(R) Optane(TM) DC persistent memory as the data source where the data request missed all caches.", + "Counter": "0,1,2,3", "EventCode": "0xd3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM", "PublicDescription": "Counts retired load instructions with remote Intel(R) Optane(TM) DC persistent memory as the data source and the data request missed L3.",
@@ -438,6 +489,7 @@ }, { "BriefDescription": "Retired instructions with at least 1 uncacheable load or lock.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd4", "EventName": "MEM_LOAD_MISC_RETIRED.UC",
@@ -448,6 +500,7 @@ }, { "BriefDescription": "Number of completed demand load requests that missed the L1, but hit the FB(fill buffer), because a preceding miss to the same cacheline initiated the line to be brought into L1, but data is not yet ready in L1.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.FB_HIT",
@@ -458,6 +511,7 @@ }, { "BriefDescription": "Retired load instructions with L1 cache hits as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L1_HIT",
@@ -468,6 +522,7 @@ }, { "BriefDescription": "Retired load instructions missed L1 cache as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L1_MISS",
@@ -478,6 +533,7 @@ }, { "BriefDescription": "Retired load instructions with L2 cache hits as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L2_HIT",
@@ -488,6 +544,7 @@ }, { "BriefDescription": "Retired load instructions missed L2 cache as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L2_MISS",
@@ -498,6 +555,7 @@ }, { "BriefDescription": "Retired load instructions with L3 cache hits as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L3_HIT",
@@ -508,6 +566,7 @@ }, { "BriefDescription": "Retired load instructions missed L3 cache as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L3_MISS",
@@ -518,6 +577,7 @@ }, { "BriefDescription": "Retired load instructions with local Intel(R) Optane(TM) DC persistent memory as the data source where the data request missed all caches.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.LOCAL_PMM",
@@ -528,6 +588,7 @@ }, { "BriefDescription": "MEM_STORE_RETIRED.L2_HIT", + "Counter": "0,1,2,3", "EventCode": "0x44", "EventName": "MEM_STORE_RETIRED.L2_HIT", "SampleAfterValue": "200003",
@@ -535,6 +596,7 @@ }, { "BriefDescription": "Retired memory uops for any access", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xe5", "EventName": "MEM_UOP_RETIRED.ANY", "PublicDescription": "Number of retired micro-operations (uops) for load or store memory accesses",
@@ -543,6 +605,7 @@ }, { "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -552,6 +615,7 @@ }, { "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that resulted in a snoop hit a modified line in another core's caches which forwarded the data.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -561,6 +625,7 @@ }, { "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_CODE_RD.SNC_CACHE.HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -570,6 +635,7 @@ }, { "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that either hit a non-modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_CODE_RD.SNC_CACHE.HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -579,6 +645,7 @@ }, { "BriefDescription": "Counts demand data reads that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -588,6 +655,7 @@ }, { "BriefDescription": "Counts demand data reads that resulted in a snoop hit a modified line in another core's caches which forwarded the data.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -597,6 +665,7 @@ }, { "BriefDescription": "Counts demand data reads that resulted in a snoop that hit in another core, which did not forward the data.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -606,6 +675,7 @@ }, { "BriefDescription": "Counts demand data reads that resulted in a snoop hit in another core's caches which forwarded the unmodified data to the requesting core.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -615,6 +685,7 @@ }, { "BriefDescription": "Counts demand data reads that were supplied by a cache on a remote socket where a snoop hit a modified line in another core's caches which forwarded the data.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.REMOTE_CACHE.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -624,6 +695,7 @@ }, { "BriefDescription": "Counts demand data reads that were supplied by a cache on a remote socket where a snoop hit in another core's caches which forwarded the unmodified data to the requesting core.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.REMOTE_CACHE.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -633,6 +705,7 @@ }, { "BriefDescription": "Counts demand data reads that hit a modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.SNC_CACHE.HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -642,6 +715,7 @@ }, { "BriefDescription": "Counts demand data reads that either hit a non-modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.SNC_CACHE.HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -651,6 +725,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_RFO.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -660,6 +735,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that resulted in a snoop hit a modified line in another core's caches which forwarded the data.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -669,6 +745,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_RFO.SNC_CACHE.HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -678,6 +755,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that either hit a non-modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_RFO.SNC_CACHE.HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -687,6 +765,7 @@ }, { "BriefDescription": "Counts hardware prefetches to the L3 only that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.HWPF_L3.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -696,6 +775,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -705,6 +785,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that resulted in a snoop hit a modified line in another core's caches which forwarded the data.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -714,6 +795,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that resulted in a snoop that hit in another core, which did not forward the data.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.L3_HIT.SNOOP_HIT_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -723,6 +805,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that resulted in a snoop hit in another core's caches which forwarded the unmodified data to the requesting core.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -732,6 +815,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by a cache on a remote socket where a snoop was sent and data was returned (Modified or Not Modified).", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.REMOTE_CACHE.SNOOP_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -741,6 +825,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by a cache on a remote socket where a snoop hit a modified line in another core's caches which forwarded the data.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.REMOTE_CACHE.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -750,6 +835,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by a cache on a remote socket where a snoop hit in another core's caches which forwarded the unmodified data to the requesting core.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.REMOTE_CACHE.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -759,6 +845,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that hit a modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.SNC_CACHE.HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -768,6 +855,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that either hit a non-modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.SNC_CACHE.HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -777,6 +865,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO), hardware prefetch RFOs (which bring data to L2), and software prefetches for exclusive ownership (PREFETCHW) that hit to a (M)odified cacheline in the L3 or snoop filter.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.RFO_TO_CORE.L3_HIT_M", "MSRIndex": "0x1a6,0x1a7",
@@ -786,6 +875,7 @@ }, { "BriefDescription": "Counts streaming stores that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.STREAMING_WR.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -795,6 +885,7 @@ }, { "BriefDescription": "OFFCORE_REQUESTS.ALL_REQUESTS", + "Counter": "0,1,2,3", "EventCode": "0x21", "EventName": "OFFCORE_REQUESTS.ALL_REQUESTS", "SampleAfterValue": "100003",
@@ -802,6 +893,7 @@ }, { "BriefDescription": "Demand and prefetch data reads", + "Counter": "0,1,2,3", "EventCode": "0x21", "EventName": "OFFCORE_REQUESTS.DATA_RD", "PublicDescription": "Counts the demand and prefetch data reads. All Core Data Reads include cacheable 'Demands' and L2 prefetchers (not L3 prefetchers). Counting also covers reads due to page walks resulted from any request type.",
@@ -810,6 +902,7 @@ }, { "BriefDescription": "Demand Data Read requests sent to uncore", + "Counter": "0,1,2,3", "EventCode": "0x21", "EventName": "OFFCORE_REQUESTS.DEMAND_DATA_RD", "PublicDescription": "Counts the Demand Data Read requests sent to uncore. Use it in conjunction with OFFCORE_REQUESTS_OUTSTANDING to determine average latency in the uncore.",
@@ -818,6 +911,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OFFCORE_REQUESTS_OUTSTANDING.DATA_RD", + "Counter": "0,1,2,3", "Deprecated": "1", "EventCode": "0x20", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD",
@@ -826,6 +920,7 @@ }, { "BriefDescription": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x20", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD",
@@ -834,6 +929,7 @@ }, { "BriefDescription": "Cycles where at least 1 outstanding demand data read request is pending.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x20", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_DATA_RD",
@@ -842,6 +938,7 @@ }, { "BriefDescription": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x20", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO",
@@ -850,6 +947,7 @@ }, { "BriefDescription": "OFFCORE_REQUESTS_OUTSTANDING.DATA_RD", + "Counter": "0,1,2,3", "EventCode": "0x20", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.DATA_RD", "SampleAfterValue": "1000003",
@@ -857,6 +955,7 @@ }, { "BriefDescription": "For every cycle, increments by the number of outstanding demand data read requests pending.", + "Counter": "0,1,2,3", "EventCode": "0x20", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD", "PublicDescription": "For every cycle, increments by the number of outstanding demand data read requests pending. Requests are considered outstanding from the time they miss the core's L2 cache until the transaction completion message is sent to the requestor.",
@@ -865,6 +964,7 @@ }, { "BriefDescription": "Counts bus locks, accounts for cache line split locks and UC locks.", + "Counter": "0,1,2,3", "EventCode": "0x2c", "EventName": "SQ_MISC.BUS_LOCK", "PublicDescription": "Counts the more expensive bus lock needed to enforce cache coherency for certain memory accesses that need to be done atomically. Can be created by issuing an atomic instruction (via the LOCK prefix) which causes a cache line split or accesses uncacheable memory.",
@@ -873,6 +973,7 @@ }, { "BriefDescription": "Number of PREFETCHNTA instructions executed.", + "Counter": "0,1,2,3", "EventCode": "0x40", "EventName": "SW_PREFETCH_ACCESS.NTA", "PublicDescription": "Counts the number of PREFETCHNTA instructions executed.",
@@ -881,6 +982,7 @@ }, { "BriefDescription": "Number of PREFETCHW instructions executed.", + "Counter": "0,1,2,3", "EventCode": "0x40", "EventName": "SW_PREFETCH_ACCESS.PREFETCHW", "PublicDescription": "Counts the number of PREFETCHW instructions executed.",
@@ -889,6 +991,7 @@ }, { "BriefDescription": "Number of PREFETCHT0 instructions executed.", + "Counter": "0,1,2,3", "EventCode": "0x40", "EventName": "SW_PREFETCH_ACCESS.T0", "PublicDescription": "Counts the number of PREFETCHT0 instructions executed.",
@@ -897,6 +1000,7 @@ }, { "BriefDescription": "Number of PREFETCHT1 or PREFETCHT2 instructions executed.", + "Counter": "0,1,2,3", "EventCode": "0x40", "EventName": "SW_PREFETCH_ACCESS.T1_T2", "PublicDescription": "Counts the number of PREFETCHT1 or PREFETCHT2 instructions executed.",
tools/perf/pmu-events/arch/x86/sapphirerapids/counter.json new file +17 -0
@@ -0,0 +1,17 @@ +[ + { + "Unit": "Core", + "Size": "8", + "FixedSize": "4" + }, + { + "Unit": "UBOX", + "Size": "2", + "FixedSize": "0" + }, + { + "Unit": "IRP", + "Size": "2", + "FixedSize": "0" + } +]
tools/perf/pmu-events/arch/x86/sapphirerapids/floating-point.json +25 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "ARITH.FPDIV_ACTIVE", + "Counter": "0,1,2,3,4,5,6,7", "CounterMask": "1", "EventCode": "0xb0", "EventName": "ARITH.FPDIV_ACTIVE",
@@ -9,6 +10,7 @@ }, { "BriefDescription": "Counts all microcode FP assists.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc1", "EventName": "ASSISTS.FP", "PublicDescription": "Counts all microcode Floating Point assists.",
@@ -17,6 +19,7 @@ }, { "BriefDescription": "ASSISTS.SSE_AVX_MIX", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc1", "EventName": "ASSISTS.SSE_AVX_MIX", "SampleAfterValue": "1000003",
@@ -24,6 +27,7 @@ }, { "BriefDescription": "FP_ARITH_DISPATCHED.PORT_0", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xb3", "EventName": "FP_ARITH_DISPATCHED.PORT_0", "SampleAfterValue": "2000003",
@@ -31,6 +35,7 @@ }, { "BriefDescription": "FP_ARITH_DISPATCHED.PORT_1", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xb3", "EventName": "FP_ARITH_DISPATCHED.PORT_1", "SampleAfterValue": "2000003",
@@ -38,6 +43,7 @@ }, { "BriefDescription": "FP_ARITH_DISPATCHED.PORT_5", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xb3", "EventName": "FP_ARITH_DISPATCHED.PORT_5", "SampleAfterValue": "2000003",
@@ -45,6 +51,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 128-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE", "PublicDescription": "Number of SSE/AVX computational 128-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -53,6 +60,7 @@ }, { "BriefDescription": "Number of SSE/AVX computational 128-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE", "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -61,6 +69,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE", "PublicDescription": "Number of SSE/AVX computational 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -69,6 +78,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 256-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE", "PublicDescription": "Number of SSE/AVX computational 256-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -77,6 +87,7 @@ }, { "BriefDescription": "Number of SSE/AVX computational 128-bit packed single and 256-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and packed double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.4_FLOPS", "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision and 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point and packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -85,6 +96,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE", "PublicDescription": "Number of SSE/AVX computational 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -93,6 +105,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 512-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 16 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE", "PublicDescription": "Number of SSE/AVX computational 512-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 16 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -101,6 +114,7 @@ }, { "BriefDescription": "Number of SSE/AVX computational 256-bit packed single precision and 512-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RSQRT14 RCP RCP14 DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.8_FLOPS", "PublicDescription": "Number of SSE/AVX computational 256-bit packed single precision and 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision and double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RSQRT14 RCP RCP14 DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -109,6 +123,7 @@ }, { "BriefDescription": "Number of SSE/AVX computational scalar floating-point instructions retired; some instructions will count twice as noted below. Applies to SSE* and AVX* scalar, double and single precision floating-point: ADD SUB MUL DIV MIN MAX RCP14 RSQRT14 RANGE SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform multiple calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.SCALAR", "PublicDescription": "Number of SSE/AVX computational scalar single precision and double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -117,6 +132,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational scalar double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "PublicDescription": "Number of SSE/AVX computational scalar double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -125,6 +141,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational scalar single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "PublicDescription": "Number of SSE/AVX computational scalar single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -133,6 +150,7 @@ }, { "BriefDescription": "Number of any Vector retired FP arithmetic instructions", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.VECTOR", "PublicDescription": "Number of any Vector retired FP arithmetic instructions. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -141,6 +159,7 @@ }, { "BriefDescription": "FP_ARITH_INST_RETIRED2.128B_PACKED_HALF", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xcf", "EventName": "FP_ARITH_INST_RETIRED2.128B_PACKED_HALF", "SampleAfterValue": "100003",
@@ -148,6 +167,7 @@ }, { "BriefDescription": "FP_ARITH_INST_RETIRED2.256B_PACKED_HALF", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xcf", "EventName": "FP_ARITH_INST_RETIRED2.256B_PACKED_HALF", "SampleAfterValue": "100003",
@@ -155,6 +175,7 @@ }, { "BriefDescription": "FP_ARITH_INST_RETIRED2.512B_PACKED_HALF", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xcf", "EventName": "FP_ARITH_INST_RETIRED2.512B_PACKED_HALF", "SampleAfterValue": "100003",
@@ -162,6 +183,7 @@ }, { "BriefDescription": "FP_ARITH_INST_RETIRED2.COMPLEX_SCALAR_HALF", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xcf", "EventName": "FP_ARITH_INST_RETIRED2.COMPLEX_SCALAR_HALF", "SampleAfterValue": "100003",
@@ -169,6 +191,7 @@ }, { "BriefDescription": "Number of all Scalar Half-Precision FP arithmetic instructions(1) retired - regular and complex.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xcf", "EventName": "FP_ARITH_INST_RETIRED2.SCALAR", "PublicDescription": "FP_ARITH_INST_RETIRED2.SCALAR",
@@ -177,6 +200,7 @@ }, { "BriefDescription": "FP_ARITH_INST_RETIRED2.SCALAR_HALF", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xcf", "EventName": "FP_ARITH_INST_RETIRED2.SCALAR_HALF", "SampleAfterValue": "100003",
@@ -184,6 +208,7 @@ }, { "BriefDescription": "Number of all Vector (also called packed) Half-Precision FP arithmetic instructions(1) retired.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xcf", "EventName": "FP_ARITH_INST_RETIRED2.VECTOR", "PublicDescription": "FP_ARITH_INST_RETIRED2.VECTOR",
tools/perf/pmu-events/arch/x86/sapphirerapids/frontend.json +92 -6
401 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/sapphirerapids/memory.json +44 -0
294 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/sapphirerapids/other.json +40 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "ASSISTS.PAGE_FAULT", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc1", "EventName": "ASSISTS.PAGE_FAULT", "SampleAfterValue": "1000003",
@@ -8,6 +9,7 @@ }, { "BriefDescription": "Counts the cycles where the AMX (Advance Matrix Extension) unit is busy performing an operation.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xb7", "EventName": "EXE.AMX_BUSY", "SampleAfterValue": "2000003",
@@ -15,6 +17,7 @@ }, { "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that have any type of response.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_CODE_RD.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7",
@@ -24,6 +27,7 @@ }, { "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that were supplied by DRAM.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_CODE_RD.DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -33,6 +37,7 @@ }, { "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that were supplied by DRAM attached to this socket, unless in Sub NUMA Cluster(SNC) Mode. In SNC Mode counts only those DRAM accesses that are controlled by the close SNC Cluster.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_CODE_RD.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -42,6 +47,7 @@ }, { "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that were supplied by DRAM on a distant memory controller of this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_CODE_RD.SNC_DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -51,6 +57,7 @@ }, { "BriefDescription": "Counts demand data reads that have any type of response.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7",
@@ -60,6 +67,7 @@ }, { "BriefDescription": "Counts demand data reads that were supplied by DRAM.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -69,6 +77,7 @@ }, { "BriefDescription": "Counts demand data reads that were supplied by DRAM attached to this socket, unless in Sub NUMA Cluster(SNC) Mode. In SNC Mode counts only those DRAM accesses that are controlled by the close SNC Cluster.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -78,6 +87,7 @@ }, { "BriefDescription": "Counts demand data reads that were supplied by PMM attached to this socket, whether or not in Sub NUMA Cluster(SNC) Mode. In SNC Mode counts PMM accesses that are controlled by the close or distant SNC Cluster.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.LOCAL_SOCKET_PMM", "MSRIndex": "0x1a6,0x1a7",
@@ -87,6 +97,7 @@ }, { "BriefDescription": "Counts demand data reads that were supplied by PMM.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.PMM", "MSRIndex": "0x1a6,0x1a7",
@@ -96,6 +107,7 @@ }, { "BriefDescription": "Counts demand data reads that were supplied by DRAM attached to another socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.REMOTE_DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -105,6 +117,7 @@ }, { "BriefDescription": "Counts demand data reads that were supplied by PMM attached to another socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.REMOTE_PMM", "MSRIndex": "0x1a6,0x1a7",
@@ -114,6 +127,7 @@ }, { "BriefDescription": "Counts demand data reads that were supplied by DRAM on a distant memory controller of this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_DATA_RD.SNC_DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -123,6 +137,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that have any type of response.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_RFO.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7",
@@ -132,6 +147,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that were supplied by DRAM.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_RFO.DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -141,6 +157,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that were supplied by DRAM attached to this socket, unless in Sub NUMA Cluster(SNC) Mode. In SNC Mode counts only those DRAM accesses that are controlled by the close SNC Cluster.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_RFO.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -150,6 +167,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that were supplied by DRAM on a distant memory controller of this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.DEMAND_RFO.SNC_DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -159,6 +177,7 @@ }, { "BriefDescription": "Counts data load hardware prefetch requests to the L1 data cache that have any type of response.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.HWPF_L1D.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7",
@@ -168,6 +187,7 @@ }, { "BriefDescription": "Counts hardware prefetches (which bring data to L2) that have any type of response.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.HWPF_L2.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7",
@@ -177,6 +197,7 @@ }, { "BriefDescription": "Counts hardware prefetches to the L3 only that have any type of response.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.HWPF_L3.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7",
@@ -186,6 +207,7 @@ }, { "BriefDescription": "Counts hardware prefetches to the L3 only that were not supplied by the local socket's L1, L2, or L3 caches and the cacheline was homed in a remote socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.HWPF_L3.REMOTE", "MSRIndex": "0x1a6,0x1a7",
@@ -195,6 +217,7 @@ }, { "BriefDescription": "Counts writebacks of modified cachelines and streaming stores that have any type of response.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.MODIFIED_WRITE.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7",
@@ -204,6 +227,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that have any type of response.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7",
@@ -213,6 +237,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by DRAM.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -222,6 +247,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by DRAM attached to this socket, unless in Sub NUMA Cluster(SNC) Mode. In SNC Mode counts only those DRAM accesses that are controlled by the close SNC Cluster.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -231,6 +257,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by DRAM attached to this socket, whether or not in Sub NUMA Cluster(SNC) Mode. In SNC Mode counts DRAM accesses that are controlled by the close or distant SNC Cluster.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.LOCAL_SOCKET_DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -240,6 +267,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by PMM attached to this socket, whether or not in Sub NUMA Cluster(SNC) Mode. In SNC Mode counts PMM accesses that are controlled by the close or distant SNC Cluster.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.LOCAL_SOCKET_PMM", "MSRIndex": "0x1a6,0x1a7",
@@ -249,6 +277,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were not supplied by the local socket's L1, L2, or L3 caches and were supplied by a remote socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.REMOTE", "MSRIndex": "0x1a6,0x1a7",
@@ -258,6 +287,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by DRAM attached to another socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.REMOTE_DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -267,6 +297,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by DRAM or PMM attached to another socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.REMOTE_MEMORY", "MSRIndex": "0x1a6,0x1a7",
@@ -276,6 +307,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by PMM attached to another socket.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.REMOTE_PMM", "MSRIndex": "0x1a6,0x1a7",
@@ -285,6 +317,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by DRAM on a distant memory controller of this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.READS_TO_CORE.SNC_DRAM", "MSRIndex": "0x1a6,0x1a7",
@@ -294,6 +327,7 @@ }, { "BriefDescription": "Counts streaming stores that have any type of response.", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.STREAMING_WR.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7",
@@ -303,6 +337,7 @@ }, { "BriefDescription": "Counts Demand RFOs, ItoM's, PREFECTHW's, Hardware RFO Prefetches to the L1/L2 and Streaming stores that likely resulted in a store to Memory (DRAM or PMM)", + "Counter": "0,1,2,3", "EventCode": "0x2A,0x2B", "EventName": "OCR.WRITE_ESTIMATE.MEMORY", "MSRIndex": "0x1a6,0x1a7",
@@ -312,6 +347,7 @@ }, { "BriefDescription": "Cycles when Reservation Station (RS) is empty for the thread.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xa5", "EventName": "RS.EMPTY", "PublicDescription": "Counts cycles during which the reservation station (RS) is empty for this logical processor. This is usually caused when the front-end pipeline runs into starvation periods (e.g. branch mispredictions or i-cache misses)",
@@ -320,6 +356,7 @@ }, { "BriefDescription": "Counts end of periods where the Reservation Station (RS) was empty.", + "Counter": "0,1,2,3,4,5,6,7", "CounterMask": "1", "EdgeDetect": "1", "EventCode": "0xa5",
@@ -331,6 +368,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event RS.EMPTY_COUNT", + "Counter": "0,1,2,3,4,5,6,7", "CounterMask": "1", "Deprecated": "1", "EdgeDetect": "1",
@@ -342,6 +380,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event RS.EMPTY", + "Counter": "0,1,2,3,4,5,6,7", "Deprecated": "1", "EventCode": "0xa5", "EventName": "RS_EMPTY.CYCLES",
@@ -350,6 +389,7 @@ }, { "BriefDescription": "Cycles the uncore cannot take further requests", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x2d", "EventName": "XQ.FULL_CYCLES",
tools/perf/pmu-events/arch/x86/sapphirerapids/pipeline.json +118 -0
820 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-cache.json +524 -10
284 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-cxl.json +56 -0
56 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-interconnect.json +476 -0
476 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-io.json +373 -0
373 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-memory.json +391 -0
391 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-power.json +24 -0
Lines of this file are not shown. View the raw message.
tools/perf/pmu-events/arch/x86/sapphirerapids/virtual-memory.json +20 -0
Lines of this file are not shown. View the raw message.
The rest of this diff is not shown. View the raw message.
Signature · 1 line
2.39.3
Part of the diff is not shown here. View the raw message.
RFC PATCH v2 5/17 perf pmu-events: Add event counter data for Cascadelakex
In reply to [email protected]From: Weilin Wang <[email protected]>
The added new fields are required in this new grouping method. Currently, the format of added fields are temporary for experiments.
--- .../arch/x86/cascadelakex/cache.json | 1237 ++++++++++++++++ .../arch/x86/cascadelakex/counter.json | 17 + .../arch/x86/cascadelakex/floating-point.json | 16 + .../arch/x86/cascadelakex/frontend.json | 68 + .../arch/x86/cascadelakex/memory.json | 751 ++++++++++ .../arch/x86/cascadelakex/other.json | 168 +++ .../arch/x86/cascadelakex/pipeline.json | 102 ++ .../arch/x86/cascadelakex/uncore-cache.json | 1138 +++++++++++++++ .../x86/cascadelakex/uncore-interconnect.json | 1272 +++++++++++++++++ .../arch/x86/cascadelakex/uncore-io.json | 394 +++++ .../arch/x86/cascadelakex/uncore-memory.json | 509 +++++++ .../arch/x86/cascadelakex/uncore-power.json | 25 + .../arch/x86/cascadelakex/virtual-memory.json | 28 + 13 files changed, 5725 insertions(+) create mode 100644 tools/perf/pmu-events/arch/x86/cascadelakex/counter.json
tools/perf/pmu-events/arch/x86/cascadelakex/cache.json +1237 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "L1D data line replacements", + "Counter": "0,1,2,3", "EventCode": "0x51", "EventName": "L1D.REPLACEMENT", "PublicDescription": "Counts L1D data line replacements including opportunistic replacements, and replacements that require stall-for-replace or block-for-replace.",
@@ -9,6 +10,7 @@ }, { "BriefDescription": "Number of times a request needed a FB entry but there was no entry available for it. That is the FB unavailability was dominant reason for blocking the request. A request includes cacheable/uncacheable demands that is load, store or SW prefetch.", + "Counter": "0,1,2,3", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.FB_FULL", "PublicDescription": "Number of times a request needed a FB (Fill Buffer) entry but there was no entry available for it. A request includes cacheable/uncacheable demands that are load, store or SW prefetch instructions.",
@@ -17,6 +19,7 @@ }, { "BriefDescription": "L1D miss outstandings duration in cycles", + "Counter": "0,1,2,3", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.PENDING", "PublicDescription": "Counts duration of L1D miss outstanding, that is each cycle number of Fill Buffers (FB) outstanding required by Demand Reads. FB either is held by demand loads, or it is held by non-demand loads and gets hit at least once by demand. The valid outstanding interval is defined until the FB deallocation by one of the following ways: from FB allocation, if FB is allocated by demand from the demand Hit FB, if it is allocated by hardware or software prefetch.Note: In the L1D, a Demand Read contains cacheable or noncacheable demand loads, including ones causing cache-line splits and reads due to page walks resulted from any request type.",
@@ -25,6 +28,7 @@ }, { "BriefDescription": "Cycles with L1D load Misses outstanding.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.PENDING_CYCLES",
@@ -35,6 +39,7 @@ { "AnyThread": "1", "BriefDescription": "Cycles with L1D load Misses outstanding from any thread on physical core.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.PENDING_CYCLES_ANY",
@@ -43,6 +48,7 @@ }, { "BriefDescription": "L2 cache lines filling L2", + "Counter": "0,1,2,3", "EventCode": "0xF1", "EventName": "L2_LINES_IN.ALL", "PublicDescription": "Counts the number of L2 cache lines filling the L2. Counting does not cover rejects.",
@@ -51,6 +57,7 @@ }, { "BriefDescription": "Counts the number of lines that are evicted by L2 cache when triggered by an L2 cache fill. Those lines can be either in modified state or clean state. Modified lines may either be written back to L3 or directly written to memory and not allocated in L3. Clean lines may either be allocated in L3 or dropped", + "Counter": "0,1,2,3", "EventCode": "0xF2", "EventName": "L2_LINES_OUT.NON_SILENT", "PublicDescription": "Counts the number of lines that are evicted by L2 cache when triggered by an L2 cache fill. Those lines can be either in modified state or clean state. Modified lines may either be written back to L3 or directly written to memory and not allocated in L3. Clean lines may either be allocated in L3 or dropped.",
@@ -59,6 +66,7 @@ }, { "BriefDescription": "Counts the number of lines that are silently dropped by L2 cache when triggered by an L2 cache fill. These lines are typically in Shared state. A non-threaded event.", + "Counter": "0,1,2,3", "EventCode": "0xF2", "EventName": "L2_LINES_OUT.SILENT", "SampleAfterValue": "200003",
@@ -66,6 +74,7 @@ }, { "BriefDescription": "Counts the number of lines that have been hardware prefetched but not used and now evicted by L2 cache", + "Counter": "0,1,2,3", "EventCode": "0xF2", "EventName": "L2_LINES_OUT.USELESS_HWPF", "SampleAfterValue": "200003",
@@ -73,6 +82,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event L2_LINES_OUT.USELESS_HWPF", + "Counter": "0,1,2,3", "Deprecated": "1", "EventCode": "0xF2", "EventName": "L2_LINES_OUT.USELESS_PREF",
@@ -81,6 +91,7 @@ }, { "BriefDescription": "L2 code requests", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_CODE_RD", "PublicDescription": "Counts the total number of L2 code requests.",
@@ -89,6 +100,7 @@ }, { "BriefDescription": "Demand Data Read requests", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_DEMAND_DATA_RD", "PublicDescription": "Counts the number of demand Data Read requests (including requests from L1D hardware prefetchers). These loads may hit or miss L2 cache. Only non rejected loads are counted.",
@@ -97,6 +109,7 @@ }, { "BriefDescription": "Demand requests that miss L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_DEMAND_MISS", "PublicDescription": "Demand requests that miss L2 cache.",
@@ -105,6 +118,7 @@ }, { "BriefDescription": "Demand requests to L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_DEMAND_REFERENCES", "PublicDescription": "Demand requests to L2 cache.",
@@ -113,6 +127,7 @@ }, { "BriefDescription": "Requests from the L1/L2/L3 hardware prefetchers or Load software prefetches", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_PF", "PublicDescription": "Counts the total number of requests from the L2 hardware prefetchers.",
@@ -121,6 +136,7 @@ }, { "BriefDescription": "RFO requests to L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_RFO", "PublicDescription": "Counts the total number of RFO (read for ownership) requests to L2 cache. L2 RFO requests include both L1D demand RFO misses as well as L1D RFO prefetches.",
@@ -129,6 +145,7 @@ }, { "BriefDescription": "L2 cache hits when fetching instructions, code reads.", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.CODE_RD_HIT", "PublicDescription": "Counts L2 cache hits when fetching instructions, code reads.",
@@ -137,6 +154,7 @@ }, { "BriefDescription": "L2 cache misses when fetching instructions", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.CODE_RD_MISS", "PublicDescription": "Counts L2 cache misses when fetching instructions.",
@@ -145,6 +163,7 @@ }, { "BriefDescription": "Demand Data Read requests that hit L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.DEMAND_DATA_RD_HIT", "PublicDescription": "Counts the number of demand Data Read requests, initiated by load instructions, that hit L2 cache",
@@ -153,6 +172,7 @@ }, { "BriefDescription": "Demand Data Read miss L2, no rejects", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.DEMAND_DATA_RD_MISS", "PublicDescription": "Counts the number of demand Data Read requests that miss L2 cache. Only not rejected loads are counted.",
@@ -161,6 +181,7 @@ }, { "BriefDescription": "All requests that miss L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.MISS", "PublicDescription": "All requests that miss L2 cache.",
@@ -169,6 +190,7 @@ }, { "BriefDescription": "Requests from the L1/L2/L3 hardware prefetchers or Load software prefetches that hit L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.PF_HIT", "PublicDescription": "Counts requests from the L1/L2/L3 hardware prefetchers or Load software prefetches that hit L2 cache.",
@@ -177,6 +199,7 @@ }, { "BriefDescription": "Requests from the L1/L2/L3 hardware prefetchers or Load software prefetches that miss L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.PF_MISS", "PublicDescription": "Counts requests from the L1/L2/L3 hardware prefetchers or Load software prefetches that miss L2 cache.",
@@ -185,6 +208,7 @@ }, { "BriefDescription": "All L2 requests", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.REFERENCES", "PublicDescription": "All L2 requests.",
@@ -193,6 +217,7 @@ }, { "BriefDescription": "RFO requests that hit L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.RFO_HIT", "PublicDescription": "Counts the RFO (Read-for-Ownership) requests that hit L2 cache.",
@@ -201,6 +226,7 @@ }, { "BriefDescription": "RFO requests that miss L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.RFO_MISS", "PublicDescription": "Counts the RFO (Read-for-Ownership) requests that miss L2 cache.",
@@ -209,6 +235,7 @@ }, { "BriefDescription": "L2 writebacks that access L2 cache", + "Counter": "0,1,2,3", "EventCode": "0xF0", "EventName": "L2_TRANS.L2_WB", "PublicDescription": "Counts L2 writebacks that access L2 cache.",
@@ -217,6 +244,7 @@ }, { "BriefDescription": "Core-originated cacheable demand requests missed L3", + "Counter": "0,1,2,3", "Errata": "SKL057", "EventCode": "0x2E", "EventName": "LONGEST_LAT_CACHE.MISS",
@@ -226,6 +254,7 @@ }, { "BriefDescription": "Core-originated cacheable demand requests that refer to L3", + "Counter": "0,1,2,3", "Errata": "SKL057", "EventCode": "0x2E", "EventName": "LONGEST_LAT_CACHE.REFERENCE",
@@ -235,6 +264,7 @@ }, { "BriefDescription": "Retired load instructions.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD0", "EventName": "MEM_INST_RETIRED.ALL_LOADS",
@@ -245,6 +275,7 @@ }, { "BriefDescription": "Retired store instructions.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD0", "EventName": "MEM_INST_RETIRED.ALL_STORES",
@@ -255,6 +286,7 @@ }, { "BriefDescription": "All retired memory instructions.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD0", "EventName": "MEM_INST_RETIRED.ANY",
@@ -265,6 +297,7 @@ }, { "BriefDescription": "Retired load instructions with locked access.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD0", "EventName": "MEM_INST_RETIRED.LOCK_LOADS",
@@ -274,6 +307,7 @@ }, { "BriefDescription": "Retired load instructions that split across a cacheline boundary.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD0", "EventName": "MEM_INST_RETIRED.SPLIT_LOADS",
@@ -284,6 +318,7 @@ }, { "BriefDescription": "Retired store instructions that split across a cacheline boundary.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD0", "EventName": "MEM_INST_RETIRED.SPLIT_STORES",
@@ -294,6 +329,7 @@ }, { "BriefDescription": "Retired load instructions that miss the STLB.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD0", "EventName": "MEM_INST_RETIRED.STLB_MISS_LOADS",
@@ -304,6 +340,7 @@ }, { "BriefDescription": "Retired store instructions that miss the STLB.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD0", "EventName": "MEM_INST_RETIRED.STLB_MISS_STORES",
@@ -314,6 +351,7 @@ }, { "BriefDescription": "Retired load instructions which data sources were L3 and cross-core snoop hits in on-pkg core cache", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT",
@@ -324,6 +362,7 @@ }, { "BriefDescription": "Retired load instructions which data sources were HitM responses from shared L3", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM",
@@ -334,6 +373,7 @@ }, { "BriefDescription": "Retired load instructions which data sources were L3 hit and cross-core snoop missed in on-pkg core cache.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS",
@@ -343,6 +383,7 @@ }, { "BriefDescription": "Retired load instructions which data sources were hits in L3 without snoops required", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_NONE",
@@ -353,6 +394,7 @@ }, { "BriefDescription": "Retired load instructions which data sources missed L3 but serviced from local dram", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM",
@@ -363,6 +405,7 @@ }, { "BriefDescription": "Retired load instructions which data sources missed L3 but serviced from remote dram", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM",
@@ -372,6 +415,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources was forwarded from a remote cache", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD",
@@ -381,6 +425,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources was remote HITM", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM",
@@ -391,6 +436,7 @@ }, { "BriefDescription": "Retired load instructions with remote Intel(R) Optane(TM) DC persistent memory as the data source where the data request missed all caches.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM",
@@ -401,6 +447,7 @@ }, { "BriefDescription": "Retired instructions with at least 1 uncacheable load or lock.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD4", "EventName": "MEM_LOAD_MISC_RETIRED.UC",
@@ -410,6 +457,7 @@ }, { "BriefDescription": "Retired load instructions which data sources were load missed L1 but hit FB due to preceding miss to the same cache line with data not ready", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD1", "EventName": "MEM_LOAD_RETIRED.FB_HIT",
@@ -420,6 +468,7 @@ }, { "BriefDescription": "Retired load instructions with L1 cache hits as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD1", "EventName": "MEM_LOAD_RETIRED.L1_HIT",
@@ -430,6 +479,7 @@ }, { "BriefDescription": "Retired load instructions missed L1 cache as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD1", "EventName": "MEM_LOAD_RETIRED.L1_MISS",
@@ -440,6 +490,7 @@ }, { "BriefDescription": "Retired load instructions with L2 cache hits as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD1", "EventName": "MEM_LOAD_RETIRED.L2_HIT",
@@ -450,6 +501,7 @@ }, { "BriefDescription": "Retired load instructions missed L2 cache as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD1", "EventName": "MEM_LOAD_RETIRED.L2_MISS",
@@ -460,6 +512,7 @@ }, { "BriefDescription": "Retired load instructions with L3 cache hits as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD1", "EventName": "MEM_LOAD_RETIRED.L3_HIT",
@@ -470,6 +523,7 @@ }, { "BriefDescription": "Retired load instructions missed L3 cache as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD1", "EventName": "MEM_LOAD_RETIRED.L3_MISS",
@@ -480,6 +534,7 @@ }, { "BriefDescription": "Retired load instructions with local Intel(R) Optane(TM) DC persistent memory as the data source where the data request missed all caches.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xD1", "EventName": "MEM_LOAD_RETIRED.LOCAL_PMM",
@@ -490,6 +545,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT.ANY_SNOOP OCR.ALL_DATA_RD.L3_HIT.ANY_SNOOP OCR.ALL_DATA_RD.L3_HIT.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -499,6 +555,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT.HITM_OTHER_CORE OCR.ALL_DATA_RD.L3_HIT.HITM_OTHER_CORE OCR.ALL_DATA_RD.L3_HIT.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -508,6 +565,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD OCR.ALL_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD OCR.ALL_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -517,6 +575,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.ALL_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.ALL_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -526,6 +585,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT.NO_SNOOP_NEEDED OCR.ALL_DATA_RD.L3_HIT.NO_SNOOP_NEEDED OCR.ALL_DATA_RD.L3_HIT.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -535,6 +595,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -544,6 +605,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT.SNOOP_MISS OCR.ALL_DATA_RD.L3_HIT.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -553,6 +615,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT.SNOOP_NONE OCR.ALL_DATA_RD.L3_HIT.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -562,6 +625,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_E.ANY_SNOOP OCR.ALL_DATA_RD.L3_HIT_E.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_E.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -571,6 +635,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_E.HITM_OTHER_CORE OCR.ALL_DATA_RD.L3_HIT_E.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_E.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -580,6 +645,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_FWD OCR.ALL_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -589,6 +655,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_NO_FWD OCR.ALL_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -598,6 +665,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_E.NO_SNOOP_NEEDED OCR.ALL_DATA_RD.L3_HIT_E.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_E.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -607,6 +675,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_E.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_E.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -616,6 +685,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_E.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_E.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -625,6 +695,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_F.ANY_SNOOP OCR.ALL_DATA_RD.L3_HIT_F.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_F.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -634,6 +705,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_F.HITM_OTHER_CORE OCR.ALL_DATA_RD.L3_HIT_F.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_F.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -643,6 +715,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_FWD OCR.ALL_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -652,6 +725,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_NO_FWD OCR.ALL_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -661,6 +735,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_F.NO_SNOOP_NEEDED OCR.ALL_DATA_RD.L3_HIT_F.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_F.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -670,6 +745,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_F.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_F.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -679,6 +755,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_F.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_F.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -688,6 +765,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_M.ANY_SNOOP OCR.ALL_DATA_RD.L3_HIT_M.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_M.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -697,6 +775,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_M.HITM_OTHER_CORE OCR.ALL_DATA_RD.L3_HIT_M.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_M.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -706,6 +785,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_FWD OCR.ALL_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -715,6 +795,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_NO_FWD OCR.ALL_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -724,6 +805,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_M.NO_SNOOP_NEEDED OCR.ALL_DATA_RD.L3_HIT_M.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_M.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -733,6 +815,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_M.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_M.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -742,6 +825,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_M.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_M.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -751,6 +835,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_S.ANY_SNOOP OCR.ALL_DATA_RD.L3_HIT_S.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_S.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -760,6 +845,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_S.HITM_OTHER_CORE OCR.ALL_DATA_RD.L3_HIT_S.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_S.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -769,6 +855,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_FWD OCR.ALL_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -778,6 +865,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_NO_FWD OCR.ALL_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -787,6 +875,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_S.NO_SNOOP_NEEDED OCR.ALL_DATA_RD.L3_HIT_S.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_S.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -796,6 +885,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_S.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_S.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -805,6 +895,7 @@ }, { "BriefDescription": "OCR.ALL_DATA_RD.L3_HIT_S.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_DATA_RD.L3_HIT_S.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -814,6 +905,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT.ANY_SNOOP OCR.ALL_PF_DATA_RD.L3_HIT.ANY_SNOOP OCR.ALL_PF_DATA_RD.L3_HIT.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -823,6 +915,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT.HITM_OTHER_CORE OCR.ALL_PF_DATA_RD.L3_HIT.HITM_OTHER_CORE OCR.ALL_PF_DATA_RD.L3_HIT.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -832,6 +925,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD OCR.ALL_PF_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD OCR.ALL_PF_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -841,6 +935,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.ALL_PF_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.ALL_PF_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -850,6 +945,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT.NO_SNOOP_NEEDED OCR.ALL_PF_DATA_RD.L3_HIT.NO_SNOOP_NEEDED OCR.ALL_PF_DATA_RD.L3_HIT.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -859,6 +955,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -868,6 +965,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT.SNOOP_MISS OCR.ALL_PF_DATA_RD.L3_HIT.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -877,6 +975,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT.SNOOP_NONE OCR.ALL_PF_DATA_RD.L3_HIT.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -886,6 +985,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_E.ANY_SNOOP OCR.ALL_PF_DATA_RD.L3_HIT_E.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_E.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -895,6 +995,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_E.HITM_OTHER_CORE OCR.ALL_PF_DATA_RD.L3_HIT_E.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_E.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -904,6 +1005,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_FWD OCR.ALL_PF_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -913,6 +1015,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_NO_FWD OCR.ALL_PF_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -922,6 +1025,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_E.NO_SNOOP_NEEDED OCR.ALL_PF_DATA_RD.L3_HIT_E.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_E.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -931,6 +1035,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_E.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_E.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -940,6 +1045,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_E.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_E.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -949,6 +1055,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_F.ANY_SNOOP OCR.ALL_PF_DATA_RD.L3_HIT_F.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_F.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -958,6 +1065,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_F.HITM_OTHER_CORE OCR.ALL_PF_DATA_RD.L3_HIT_F.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_F.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -967,6 +1075,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_FWD OCR.ALL_PF_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -976,6 +1085,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_NO_FWD OCR.ALL_PF_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -985,6 +1095,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_F.NO_SNOOP_NEEDED OCR.ALL_PF_DATA_RD.L3_HIT_F.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_F.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -994,6 +1105,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_F.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_F.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1003,6 +1115,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_F.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_F.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1012,6 +1125,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_M.ANY_SNOOP OCR.ALL_PF_DATA_RD.L3_HIT_M.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_M.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1021,6 +1135,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_M.HITM_OTHER_CORE OCR.ALL_PF_DATA_RD.L3_HIT_M.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_M.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1030,6 +1145,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_FWD OCR.ALL_PF_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1039,6 +1155,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_NO_FWD OCR.ALL_PF_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1048,6 +1165,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_M.NO_SNOOP_NEEDED OCR.ALL_PF_DATA_RD.L3_HIT_M.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_M.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1057,6 +1175,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_M.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_M.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1066,6 +1185,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_M.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_M.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1075,6 +1195,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_S.ANY_SNOOP OCR.ALL_PF_DATA_RD.L3_HIT_S.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_S.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1084,6 +1205,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_S.HITM_OTHER_CORE OCR.ALL_PF_DATA_RD.L3_HIT_S.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_S.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1093,6 +1215,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_FWD OCR.ALL_PF_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1102,6 +1225,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_NO_FWD OCR.ALL_PF_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1111,6 +1235,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_S.NO_SNOOP_NEEDED OCR.ALL_PF_DATA_RD.L3_HIT_S.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_S.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1120,6 +1245,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_S.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_S.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1129,6 +1255,7 @@ }, { "BriefDescription": "OCR.ALL_PF_DATA_RD.L3_HIT_S.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_DATA_RD.L3_HIT_S.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1138,6 +1265,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT.ANY_SNOOP OCR.ALL_PF_RFO.L3_HIT.ANY_SNOOP OCR.ALL_PF_RFO.L3_HIT.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1147,6 +1275,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT.HITM_OTHER_CORE OCR.ALL_PF_RFO.L3_HIT.HITM_OTHER_CORE OCR.ALL_PF_RFO.L3_HIT.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1156,6 +1285,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT.HIT_OTHER_CORE_FWD OCR.ALL_PF_RFO.L3_HIT.HIT_OTHER_CORE_FWD OCR.ALL_PF_RFO.L3_HIT.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1165,6 +1295,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.ALL_PF_RFO.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.ALL_PF_RFO.L3_HIT.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1174,6 +1305,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT.NO_SNOOP_NEEDED OCR.ALL_PF_RFO.L3_HIT.NO_SNOOP_NEEDED OCR.ALL_PF_RFO.L3_HIT.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1183,6 +1315,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT.SNOOP_HIT_WITH_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1192,6 +1325,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT.SNOOP_MISS OCR.ALL_PF_RFO.L3_HIT.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1201,6 +1335,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT.SNOOP_NONE OCR.ALL_PF_RFO.L3_HIT.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1210,6 +1345,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_E.ANY_SNOOP OCR.ALL_PF_RFO.L3_HIT_E.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_E.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1219,6 +1355,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_E.HITM_OTHER_CORE OCR.ALL_PF_RFO.L3_HIT_E.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_E.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1228,6 +1365,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_E.HIT_OTHER_CORE_FWD OCR.ALL_PF_RFO.L3_HIT_E.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_E.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1237,6 +1375,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_E.HIT_OTHER_CORE_NO_FWD OCR.ALL_PF_RFO.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1246,6 +1385,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_E.NO_SNOOP_NEEDED OCR.ALL_PF_RFO.L3_HIT_E.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_E.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1255,6 +1395,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_E.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_E.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1264,6 +1405,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_E.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_E.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1273,6 +1415,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_F.ANY_SNOOP OCR.ALL_PF_RFO.L3_HIT_F.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_F.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1282,6 +1425,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_F.HITM_OTHER_CORE OCR.ALL_PF_RFO.L3_HIT_F.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_F.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1291,6 +1435,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_F.HIT_OTHER_CORE_FWD OCR.ALL_PF_RFO.L3_HIT_F.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_F.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1300,6 +1445,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_F.HIT_OTHER_CORE_NO_FWD OCR.ALL_PF_RFO.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1309,6 +1455,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_F.NO_SNOOP_NEEDED OCR.ALL_PF_RFO.L3_HIT_F.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_F.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1318,6 +1465,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_F.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_F.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1327,6 +1475,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_F.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_F.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1336,6 +1485,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_M.ANY_SNOOP OCR.ALL_PF_RFO.L3_HIT_M.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_M.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1345,6 +1495,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_M.HITM_OTHER_CORE OCR.ALL_PF_RFO.L3_HIT_M.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_M.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1354,6 +1505,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_M.HIT_OTHER_CORE_FWD OCR.ALL_PF_RFO.L3_HIT_M.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_M.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1363,6 +1515,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_M.HIT_OTHER_CORE_NO_FWD OCR.ALL_PF_RFO.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1372,6 +1525,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_M.NO_SNOOP_NEEDED OCR.ALL_PF_RFO.L3_HIT_M.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_M.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1381,6 +1535,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_M.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_M.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1390,6 +1545,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_M.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_M.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1399,6 +1555,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_S.ANY_SNOOP OCR.ALL_PF_RFO.L3_HIT_S.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_S.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1408,6 +1565,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_S.HITM_OTHER_CORE OCR.ALL_PF_RFO.L3_HIT_S.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_S.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1417,6 +1575,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_S.HIT_OTHER_CORE_FWD OCR.ALL_PF_RFO.L3_HIT_S.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_S.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1426,6 +1585,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_S.HIT_OTHER_CORE_NO_FWD OCR.ALL_PF_RFO.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1435,6 +1595,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_S.NO_SNOOP_NEEDED OCR.ALL_PF_RFO.L3_HIT_S.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_S.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1444,6 +1605,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_S.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_S.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1453,6 +1615,7 @@ }, { "BriefDescription": "OCR.ALL_PF_RFO.L3_HIT_S.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_PF_RFO.L3_HIT_S.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1462,6 +1625,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT.ANY_SNOOP OCR.ALL_READS.L3_HIT.ANY_SNOOP OCR.ALL_READS.L3_HIT.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1471,6 +1635,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT.HITM_OTHER_CORE OCR.ALL_READS.L3_HIT.HITM_OTHER_CORE OCR.ALL_READS.L3_HIT.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1480,6 +1645,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT.HIT_OTHER_CORE_FWD OCR.ALL_READS.L3_HIT.HIT_OTHER_CORE_FWD OCR.ALL_READS.L3_HIT.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1489,6 +1655,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.ALL_READS.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.ALL_READS.L3_HIT.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1498,6 +1665,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT.NO_SNOOP_NEEDED OCR.ALL_READS.L3_HIT.NO_SNOOP_NEEDED OCR.ALL_READS.L3_HIT.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1507,6 +1675,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT.SNOOP_HIT_WITH_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1516,6 +1685,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT.SNOOP_MISS OCR.ALL_READS.L3_HIT.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1525,6 +1695,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT.SNOOP_NONE OCR.ALL_READS.L3_HIT.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1534,6 +1705,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_E.ANY_SNOOP OCR.ALL_READS.L3_HIT_E.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_E.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1543,6 +1715,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_E.HITM_OTHER_CORE OCR.ALL_READS.L3_HIT_E.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_E.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1552,6 +1725,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_E.HIT_OTHER_CORE_FWD OCR.ALL_READS.L3_HIT_E.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_E.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1561,6 +1735,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_E.HIT_OTHER_CORE_NO_FWD OCR.ALL_READS.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1570,6 +1745,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_E.NO_SNOOP_NEEDED OCR.ALL_READS.L3_HIT_E.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_E.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1579,6 +1755,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_E.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_E.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1588,6 +1765,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_E.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_E.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1597,6 +1775,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_F.ANY_SNOOP OCR.ALL_READS.L3_HIT_F.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_F.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1606,6 +1785,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_F.HITM_OTHER_CORE OCR.ALL_READS.L3_HIT_F.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_F.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1615,6 +1795,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_F.HIT_OTHER_CORE_FWD OCR.ALL_READS.L3_HIT_F.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_F.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1624,6 +1805,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_F.HIT_OTHER_CORE_NO_FWD OCR.ALL_READS.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1633,6 +1815,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_F.NO_SNOOP_NEEDED OCR.ALL_READS.L3_HIT_F.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_F.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1642,6 +1825,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_F.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_F.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1651,6 +1835,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_F.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_F.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1660,6 +1845,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_M.ANY_SNOOP OCR.ALL_READS.L3_HIT_M.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_M.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1669,6 +1855,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_M.HITM_OTHER_CORE OCR.ALL_READS.L3_HIT_M.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_M.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1678,6 +1865,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_M.HIT_OTHER_CORE_FWD OCR.ALL_READS.L3_HIT_M.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_M.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1687,6 +1875,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_M.HIT_OTHER_CORE_NO_FWD OCR.ALL_READS.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1696,6 +1885,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_M.NO_SNOOP_NEEDED OCR.ALL_READS.L3_HIT_M.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_M.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1705,6 +1895,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_M.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_M.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1714,6 +1905,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_M.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_M.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1723,6 +1915,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_S.ANY_SNOOP OCR.ALL_READS.L3_HIT_S.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_S.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1732,6 +1925,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_S.HITM_OTHER_CORE OCR.ALL_READS.L3_HIT_S.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_S.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1741,6 +1935,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_S.HIT_OTHER_CORE_FWD OCR.ALL_READS.L3_HIT_S.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_S.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1750,6 +1945,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_S.HIT_OTHER_CORE_NO_FWD OCR.ALL_READS.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1759,6 +1955,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_S.NO_SNOOP_NEEDED OCR.ALL_READS.L3_HIT_S.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_S.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1768,6 +1965,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_S.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_S.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1777,6 +1975,7 @@ }, { "BriefDescription": "OCR.ALL_READS.L3_HIT_S.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_READS.L3_HIT_S.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1786,6 +1985,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT.ANY_SNOOP OCR.ALL_RFO.L3_HIT.ANY_SNOOP OCR.ALL_RFO.L3_HIT.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1795,6 +1995,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT.HITM_OTHER_CORE OCR.ALL_RFO.L3_HIT.HITM_OTHER_CORE OCR.ALL_RFO.L3_HIT.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1804,6 +2005,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT.HIT_OTHER_CORE_FWD OCR.ALL_RFO.L3_HIT.HIT_OTHER_CORE_FWD OCR.ALL_RFO.L3_HIT.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1813,6 +2015,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.ALL_RFO.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.ALL_RFO.L3_HIT.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1822,6 +2025,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT.NO_SNOOP_NEEDED OCR.ALL_RFO.L3_HIT.NO_SNOOP_NEEDED OCR.ALL_RFO.L3_HIT.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1831,6 +2035,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT.SNOOP_HIT_WITH_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1840,6 +2045,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT.SNOOP_MISS OCR.ALL_RFO.L3_HIT.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1849,6 +2055,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT.SNOOP_NONE OCR.ALL_RFO.L3_HIT.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1858,6 +2065,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_E.ANY_SNOOP OCR.ALL_RFO.L3_HIT_E.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_E.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1867,6 +2075,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_E.HITM_OTHER_CORE OCR.ALL_RFO.L3_HIT_E.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_E.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1876,6 +2085,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_E.HIT_OTHER_CORE_FWD OCR.ALL_RFO.L3_HIT_E.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_E.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1885,6 +2095,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_E.HIT_OTHER_CORE_NO_FWD OCR.ALL_RFO.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1894,6 +2105,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_E.NO_SNOOP_NEEDED OCR.ALL_RFO.L3_HIT_E.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_E.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1903,6 +2115,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_E.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_E.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1912,6 +2125,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_E.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_E.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1921,6 +2135,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_F.ANY_SNOOP OCR.ALL_RFO.L3_HIT_F.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_F.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1930,6 +2145,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_F.HITM_OTHER_CORE OCR.ALL_RFO.L3_HIT_F.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_F.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -1939,6 +2155,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_F.HIT_OTHER_CORE_FWD OCR.ALL_RFO.L3_HIT_F.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_F.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1948,6 +2165,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_F.HIT_OTHER_CORE_NO_FWD OCR.ALL_RFO.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -1957,6 +2175,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_F.NO_SNOOP_NEEDED OCR.ALL_RFO.L3_HIT_F.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_F.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -1966,6 +2185,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_F.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_F.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -1975,6 +2195,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_F.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_F.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -1984,6 +2205,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_M.ANY_SNOOP OCR.ALL_RFO.L3_HIT_M.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_M.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -1993,6 +2215,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_M.HITM_OTHER_CORE OCR.ALL_RFO.L3_HIT_M.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_M.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2002,6 +2225,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_M.HIT_OTHER_CORE_FWD OCR.ALL_RFO.L3_HIT_M.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_M.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2011,6 +2235,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_M.HIT_OTHER_CORE_NO_FWD OCR.ALL_RFO.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2020,6 +2245,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_M.NO_SNOOP_NEEDED OCR.ALL_RFO.L3_HIT_M.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_M.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2029,6 +2255,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_M.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_M.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2038,6 +2265,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_M.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_M.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2047,6 +2275,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_S.ANY_SNOOP OCR.ALL_RFO.L3_HIT_S.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_S.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2056,6 +2285,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_S.HITM_OTHER_CORE OCR.ALL_RFO.L3_HIT_S.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_S.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2065,6 +2295,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_S.HIT_OTHER_CORE_FWD OCR.ALL_RFO.L3_HIT_S.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_S.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2074,6 +2305,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_S.HIT_OTHER_CORE_NO_FWD OCR.ALL_RFO.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2083,6 +2315,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_S.NO_SNOOP_NEEDED OCR.ALL_RFO.L3_HIT_S.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_S.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2092,6 +2325,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_S.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_S.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2101,6 +2335,7 @@ }, { "BriefDescription": "OCR.ALL_RFO.L3_HIT_S.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.ALL_RFO.L3_HIT_S.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2110,6 +2345,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT.ANY_SNOOP OCR.DEMAND_CODE_RD.L3_HIT.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2119,6 +2355,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT.HITM_OTHER_CORE OCR.DEMAND_CODE_RD.L3_HIT.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2128,6 +2365,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT.HIT_OTHER_CORE_FWD OCR.DEMAND_CODE_RD.L3_HIT.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2137,6 +2375,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.DEMAND_CODE_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2146,6 +2385,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT.NO_SNOOP_NEEDED OCR.DEMAND_CODE_RD.L3_HIT.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2155,6 +2395,7 @@ }, { "BriefDescription": "Counts all demand code reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2164,6 +2405,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2173,6 +2415,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2182,6 +2425,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_E.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_E.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2191,6 +2435,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_E.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_E.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2200,6 +2445,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_E.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_E.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2209,6 +2455,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2218,6 +2465,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_E.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_E.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2227,6 +2475,7 @@ }, { "BriefDescription": "Counts all demand code reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_E.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2236,6 +2485,7 @@ }, { "BriefDescription": "Counts all demand code reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_E.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2245,6 +2495,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_F.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_F.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2254,6 +2505,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_F.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_F.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2263,6 +2515,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_F.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_F.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2272,6 +2525,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2281,6 +2535,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_F.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_F.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2290,6 +2545,7 @@ }, { "BriefDescription": "Counts all demand code reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_F.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2299,6 +2555,7 @@ }, { "BriefDescription": "Counts all demand code reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_F.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2308,6 +2565,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_M.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_M.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2317,6 +2575,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_M.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_M.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2326,6 +2585,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_M.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_M.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2335,6 +2595,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2344,6 +2605,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_M.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_M.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2353,6 +2615,7 @@ }, { "BriefDescription": "Counts all demand code reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_M.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2362,6 +2625,7 @@ }, { "BriefDescription": "Counts all demand code reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_M.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2371,6 +2635,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_S.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_S.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2380,6 +2645,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_S.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_S.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2389,6 +2655,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_S.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_S.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2398,6 +2665,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2407,6 +2675,7 @@ }, { "BriefDescription": "Counts all demand code reads OCR.DEMAND_CODE_RD.L3_HIT_S.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_S.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2416,6 +2685,7 @@ }, { "BriefDescription": "Counts all demand code reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_S.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2425,6 +2695,7 @@ }, { "BriefDescription": "Counts all demand code reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT_S.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2434,6 +2705,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT.ANY_SNOOP OCR.DEMAND_DATA_RD.L3_HIT.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2443,6 +2715,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE OCR.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2452,6 +2725,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD OCR.DEMAND_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2461,6 +2735,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.DEMAND_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2470,6 +2745,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT.NO_SNOOP_NEEDED OCR.DEMAND_DATA_RD.L3_HIT.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2479,6 +2755,7 @@ }, { "BriefDescription": "Counts demand data reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2488,6 +2765,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2497,6 +2775,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2506,6 +2785,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_E.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_E.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2515,6 +2795,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_E.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_E.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2524,6 +2805,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2533,6 +2815,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2542,6 +2825,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_E.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_E.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2551,6 +2835,7 @@ }, { "BriefDescription": "Counts demand data reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_E.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2560,6 +2845,7 @@ }, { "BriefDescription": "Counts demand data reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_E.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2569,6 +2855,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_F.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_F.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2578,6 +2865,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_F.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_F.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2587,6 +2875,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2596,6 +2885,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2605,6 +2895,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_F.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_F.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2614,6 +2905,7 @@ }, { "BriefDescription": "Counts demand data reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_F.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2623,6 +2915,7 @@ }, { "BriefDescription": "Counts demand data reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_F.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2632,6 +2925,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_M.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_M.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2641,6 +2935,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_M.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_M.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2650,6 +2945,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2659,6 +2955,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2668,6 +2965,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_M.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_M.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2677,6 +2975,7 @@ }, { "BriefDescription": "Counts demand data reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_M.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2686,6 +2985,7 @@ }, { "BriefDescription": "Counts demand data reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_M.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2695,6 +2995,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_S.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_S.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2704,6 +3005,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_S.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_S.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2713,6 +3015,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2722,6 +3025,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2731,6 +3035,7 @@ }, { "BriefDescription": "Counts demand data reads OCR.DEMAND_DATA_RD.L3_HIT_S.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_S.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2740,6 +3045,7 @@ }, { "BriefDescription": "Counts demand data reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_S.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2749,6 +3055,7 @@ }, { "BriefDescription": "Counts demand data reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT_S.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2758,6 +3065,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT.ANY_SNOOP OCR.DEMAND_RFO.L3_HIT.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2767,6 +3075,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT.HITM_OTHER_CORE OCR.DEMAND_RFO.L3_HIT.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2776,6 +3085,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT.HIT_OTHER_CORE_FWD OCR.DEMAND_RFO.L3_HIT.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2785,6 +3095,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.DEMAND_RFO.L3_HIT.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2794,6 +3105,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT.NO_SNOOP_NEEDED OCR.DEMAND_RFO.L3_HIT.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2803,6 +3115,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs)", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2812,6 +3125,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2821,6 +3135,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2830,6 +3145,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_E.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_E.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2839,6 +3155,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_E.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_E.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2848,6 +3165,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_E.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_E.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2857,6 +3175,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2866,6 +3185,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_E.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_E.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2875,6 +3195,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs)", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_E.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2884,6 +3205,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs)", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_E.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2893,6 +3215,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_F.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_F.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2902,6 +3225,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_F.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_F.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2911,6 +3235,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_F.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_F.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2920,6 +3245,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2929,6 +3255,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_F.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_F.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -2938,6 +3265,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs)", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_F.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -2947,6 +3275,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs)", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_F.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -2956,6 +3285,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_M.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_M.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -2965,6 +3295,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_M.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_M.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -2974,6 +3305,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_M.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_M.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2983,6 +3315,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -2992,6 +3325,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_M.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_M.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3001,6 +3335,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs)", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_M.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3010,6 +3345,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs)", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_M.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3019,6 +3355,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_S.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_S.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3028,6 +3365,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_S.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_S.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3037,6 +3375,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_S.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_S.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3046,6 +3385,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3055,6 +3395,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs) OCR.DEMAND_RFO.L3_HIT_S.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_S.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3064,6 +3405,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs)", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_S.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3073,6 +3415,7 @@ }, { "BriefDescription": "Counts all demand data writes (RFOs)", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT_S.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3082,6 +3425,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT.ANY_SNOOP OCR.OTHER.L3_HIT.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3091,6 +3435,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT.HITM_OTHER_CORE OCR.OTHER.L3_HIT.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3100,6 +3445,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT.HIT_OTHER_CORE_FWD OCR.OTHER.L3_HIT.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3109,6 +3455,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.OTHER.L3_HIT.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3118,6 +3465,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT.NO_SNOOP_NEEDED OCR.OTHER.L3_HIT.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3127,6 +3475,7 @@ }, { "BriefDescription": "Counts any other requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3136,6 +3485,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3145,6 +3495,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3154,6 +3505,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_E.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_E.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3163,6 +3515,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_E.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_E.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3172,6 +3525,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_E.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_E.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3181,6 +3535,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3190,6 +3545,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_E.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_E.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3199,6 +3555,7 @@ }, { "BriefDescription": "Counts any other requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_E.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3208,6 +3565,7 @@ }, { "BriefDescription": "Counts any other requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_E.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3217,6 +3575,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_F.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_F.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3226,6 +3585,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_F.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_F.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3235,6 +3595,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_F.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_F.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3244,6 +3605,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3253,6 +3615,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_F.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_F.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3262,6 +3625,7 @@ }, { "BriefDescription": "Counts any other requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_F.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3271,6 +3635,7 @@ }, { "BriefDescription": "Counts any other requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_F.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3280,6 +3645,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_M.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_M.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3289,6 +3655,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_M.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_M.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3298,6 +3665,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_M.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_M.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3307,6 +3675,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3316,6 +3685,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_M.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_M.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3325,6 +3695,7 @@ }, { "BriefDescription": "Counts any other requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_M.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3334,6 +3705,7 @@ }, { "BriefDescription": "Counts any other requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_M.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3343,6 +3715,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_S.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_S.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3352,6 +3725,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_S.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_S.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3361,6 +3735,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_S.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_S.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3370,6 +3745,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3379,6 +3755,7 @@ }, { "BriefDescription": "Counts any other requests OCR.OTHER.L3_HIT_S.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_S.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3388,6 +3765,7 @@ }, { "BriefDescription": "Counts any other requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_S.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3397,6 +3775,7 @@ }, { "BriefDescription": "Counts any other requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.OTHER.L3_HIT_S.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3406,6 +3785,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT.ANY_SNOOP OCR.PF_L1D_AND_SW.L3_HIT.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3415,6 +3795,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT.HITM_OTHER_CORE OCR.PF_L1D_AND_SW.L3_HIT.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3424,6 +3805,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT.HIT_OTHER_CORE_FWD OCR.PF_L1D_AND_SW.L3_HIT.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3433,6 +3815,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.PF_L1D_AND_SW.L3_HIT.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3442,6 +3825,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT.NO_SNOOP_NEEDED OCR.PF_L1D_AND_SW.L3_HIT.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3451,6 +3835,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3460,6 +3845,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3469,6 +3855,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3478,6 +3865,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_E.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_E.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3487,6 +3875,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_E.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_E.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3496,6 +3885,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_E.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_E.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3505,6 +3895,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3514,6 +3905,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_E.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_E.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3523,6 +3915,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_E.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3532,6 +3925,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_E.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3541,6 +3935,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_F.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_F.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3550,6 +3945,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_F.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_F.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3559,6 +3955,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_F.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_F.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3568,6 +3965,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_F.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3577,6 +3975,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_F.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_F.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3586,6 +3985,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_F.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3595,6 +3995,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_F.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3604,6 +4005,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_M.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_M.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3613,6 +4015,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_M.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_M.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3622,6 +4025,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_M.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_M.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3631,6 +4035,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_M.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3640,6 +4045,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_M.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_M.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3649,6 +4055,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_M.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3658,6 +4065,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_M.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3667,6 +4075,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_S.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_S.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3676,6 +4085,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_S.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_S.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3685,6 +4095,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_S.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_S.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3694,6 +4105,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_S.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3703,6 +4115,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests OCR.PF_L1D_AND_SW.L3_HIT_S.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_S.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3712,6 +4125,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_S.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3721,6 +4135,7 @@ }, { "BriefDescription": "Counts L1 data cache hardware prefetch requests and software prefetch requests", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L1D_AND_SW.L3_HIT_S.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3730,6 +4145,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads OCR.PF_L2_DATA_RD.L3_HIT.ANY_SNOOP OCR.PF_L2_DATA_RD.L3_HIT.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3739,6 +4155,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads OCR.PF_L2_DATA_RD.L3_HIT.HITM_OTHER_CORE OCR.PF_L2_DATA_RD.L3_HIT.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3748,6 +4165,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads OCR.PF_L2_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD OCR.PF_L2_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3757,6 +4175,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads OCR.PF_L2_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD OCR.PF_L2_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3766,6 +4185,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads OCR.PF_L2_DATA_RD.L3_HIT.NO_SNOOP_NEEDED OCR.PF_L2_DATA_RD.L3_HIT.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3775,6 +4195,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3784,6 +4205,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads OCR.PF_L2_DATA_RD.L3_HIT.SNOOP_MISS", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7",
@@ -3793,6 +4215,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads OCR.PF_L2_DATA_RD.L3_HIT.SNOOP_NONE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT.SNOOP_NONE", "MSRIndex": "0x1a6,0x1a7",
@@ -3802,6 +4225,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads OCR.PF_L2_DATA_RD.L3_HIT_E.ANY_SNOOP", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT_E.ANY_SNOOP", "MSRIndex": "0x1a6,0x1a7",
@@ -3811,6 +4235,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads OCR.PF_L2_DATA_RD.L3_HIT_E.HITM_OTHER_CORE", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT_E.HITM_OTHER_CORE", "MSRIndex": "0x1a6,0x1a7",
@@ -3820,6 +4245,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads OCR.PF_L2_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3829,6 +4255,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads OCR.PF_L2_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT_E.HIT_OTHER_CORE_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -3838,6 +4265,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads OCR.PF_L2_DATA_RD.L3_HIT_E.NO_SNOOP_NEEDED", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PF_L2_DATA_RD.L3_HIT_E.NO_SNOOP_NEEDED", "MSRIndex": "0x1a6,0x1a7",
@@ -3847,6 +4275,7 @@ }, { "BriefDescription": "Counts prefetch (that bring data to L2) data reads", + "Counter": "0,1,2,3",
Lines of this file are not shown. View the raw message.
tools/perf/pmu-events/arch/x86/cascadelakex/counter.json new file +17 -0
Lines of this file are not shown. View the raw message.
tools/perf/pmu-events/arch/x86/cascadelakex/floating-point.json +16 -0
Lines of this file are not shown. View the raw message.
tools/perf/pmu-events/arch/x86/cascadelakex/frontend.json +68 -0
68 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/cascadelakex/memory.json +751 -0
751 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/cascadelakex/other.json +168 -0
168 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/cascadelakex/pipeline.json +102 -0
102 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/cascadelakex/uncore-cache.json +1138 -0
1,138 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/cascadelakex/uncore-interconnect.json +1272 -0
1,272 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/cascadelakex/uncore-io.json +394 -0
394 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/cascadelakex/uncore-memory.json +509 -0
509 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/cascadelakex/uncore-power.json +25 -0
Lines of this file are not shown. View the raw message.
tools/perf/pmu-events/arch/x86/cascadelakex/virtual-memory.json +28 -0
Lines of this file are not shown. View the raw message.
The rest of this diff is not shown. View the raw message.
Signature · 1 line
2.39.3
Part of the diff is not shown here. View the raw message.
RFC PATCH v2 6/17 perf pmu-events: Add event counter data for Icelakex
In reply to [email protected]From: Weilin Wang <[email protected]>
The added new fields are required in this new grouping method. Currently, the format of added fields are temporary for experiments.
--- .../pmu-events/arch/x86/icelakex/cache.json | 98 + .../pmu-events/arch/x86/icelakex/counter.json | 17 + .../arch/x86/icelakex/floating-point.json | 13 + .../arch/x86/icelakex/frontend.json | 55 + .../pmu-events/arch/x86/icelakex/memory.json | 53 + .../pmu-events/arch/x86/icelakex/other.json | 52 + .../arch/x86/icelakex/pipeline.json | 92 + .../arch/x86/icelakex/uncore-cache.json | 965 ++++++++++ .../x86/icelakex/uncore-interconnect.json | 1667 +++++++++++++++++ .../arch/x86/icelakex/uncore-io.json | 966 ++++++++++ .../arch/x86/icelakex/uncore-memory.json | 186 ++ .../arch/x86/icelakex/uncore-power.json | 26 + .../arch/x86/icelakex/virtual-memory.json | 22 + 13 files changed, 4212 insertions(+) create mode 100644 tools/perf/pmu-events/arch/x86/icelakex/counter.json
tools/perf/pmu-events/arch/x86/icelakex/cache.json +98 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "Counts the number of cache lines replaced in L1 data cache.", + "Counter": "0,1,2,3", "EventCode": "0x51", "EventName": "L1D.REPLACEMENT", "PublicDescription": "Counts L1D data line replacements including opportunistic replacements, and replacements that require stall-for-replace or block-for-replace.",
@@ -9,6 +10,7 @@ }, { "BriefDescription": "Number of cycles a demand request has waited due to L1D Fill Buffer (FB) unavailability.", + "Counter": "0,1,2,3", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.FB_FULL", "PublicDescription": "Counts number of cycles a demand request has waited due to L1D Fill Buffer (FB) unavailability. Demand requests include cacheable/uncacheable demand load, store, lock or SW prefetch accesses.",
@@ -17,6 +19,7 @@ }, { "BriefDescription": "Number of phases a demand request has waited due to L1D Fill Buffer (FB) unavailability.", + "Counter": "0,1,2,3", "CounterMask": "1", "EdgeDetect": "1", "EventCode": "0x48",
@@ -27,6 +30,7 @@ }, { "BriefDescription": "Number of cycles a demand request has waited due to L1D due to lack of L2 resources.", + "Counter": "0,1,2,3", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.L2_STALL", "PublicDescription": "Counts number of cycles a demand request has waited due to L1D due to lack of L2 resources. Demand requests include cacheable/uncacheable demand load, store, lock or SW prefetch accesses.",
@@ -35,6 +39,7 @@ }, { "BriefDescription": "Number of L1D misses that are outstanding", + "Counter": "0,1,2,3", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.PENDING", "PublicDescription": "Counts number of L1D misses that are outstanding in each cycle, that is each cycle the number of Fill Buffers (FB) outstanding required by Demand Reads. FB either is held by demand loads, or it is held by non-demand loads and gets hit at least once by demand. The valid outstanding interval is defined until the FB deallocation by one of the following ways: from FB allocation, if FB is allocated by demand from the demand Hit FB, if it is allocated by hardware or software prefetch. Note: In the L1D, a Demand Read contains cacheable or noncacheable demand loads, including ones causing cache-line splits and reads due to page walks resulted from any request type.",
@@ -43,6 +48,7 @@ }, { "BriefDescription": "Cycles with L1D load Misses outstanding.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.PENDING_CYCLES",
@@ -52,6 +58,7 @@ }, { "BriefDescription": "L2 cache lines filling L2", + "Counter": "0,1,2,3", "EventCode": "0xF1", "EventName": "L2_LINES_IN.ALL", "PublicDescription": "Counts the number of L2 cache lines filling the L2. Counting does not cover rejects.",
@@ -60,6 +67,7 @@ }, { "BriefDescription": "Cache lines that are evicted by L2 cache when triggered by an L2 cache fill.", + "Counter": "0,1,2,3", "EventCode": "0xF2", "EventName": "L2_LINES_OUT.NON_SILENT", "PublicDescription": "Counts the number of lines that are evicted by the L2 cache due to L2 cache fills. Evicted lines are delivered to the L3, which may or may not cache them, according to system load and priorities.",
@@ -68,6 +76,7 @@ }, { "BriefDescription": "Non-modified cache lines that are silently dropped by L2 cache when triggered by an L2 cache fill.", + "Counter": "0,1,2,3", "EventCode": "0xF2", "EventName": "L2_LINES_OUT.SILENT", "PublicDescription": "Counts the number of lines that are silently dropped by L2 cache when triggered by an L2 cache fill. These lines are typically in Shared or Exclusive state. A non-threaded event.",
@@ -76,6 +85,7 @@ }, { "BriefDescription": "L2 code requests", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_CODE_RD", "PublicDescription": "Counts the total number of L2 code requests.",
@@ -84,6 +94,7 @@ }, { "BriefDescription": "Demand Data Read requests", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_DEMAND_DATA_RD", "PublicDescription": "Counts the number of demand Data Read requests (including requests from L1D hardware prefetchers). These loads may hit or miss L2 cache. Only non rejected loads are counted.",
@@ -92,6 +103,7 @@ }, { "BriefDescription": "Demand requests that miss L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_DEMAND_MISS", "PublicDescription": "Counts demand requests that miss L2 cache.",
@@ -100,6 +112,7 @@ }, { "BriefDescription": "RFO requests to L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_RFO", "PublicDescription": "Counts the total number of RFO (read for ownership) requests to L2 cache. L2 RFO requests include both L1D demand RFO misses as well as L1D RFO prefetches.",
@@ -108,6 +121,7 @@ }, { "BriefDescription": "L2 cache hits when fetching instructions, code reads.", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.CODE_RD_HIT", "PublicDescription": "Counts L2 cache hits when fetching instructions, code reads.",
@@ -116,6 +130,7 @@ }, { "BriefDescription": "L2 cache misses when fetching instructions", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.CODE_RD_MISS", "PublicDescription": "Counts L2 cache misses when fetching instructions.",
@@ -124,6 +139,7 @@ }, { "BriefDescription": "Demand Data Read requests that hit L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.DEMAND_DATA_RD_HIT", "PublicDescription": "Counts the number of demand Data Read requests initiated by load instructions that hit L2 cache.",
@@ -132,6 +148,7 @@ }, { "BriefDescription": "Demand Data Read miss L2, no rejects", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.DEMAND_DATA_RD_MISS", "PublicDescription": "Counts the number of demand Data Read requests that miss L2 cache. Only not rejected loads are counted.",
@@ -140,6 +157,7 @@ }, { "BriefDescription": "RFO requests that hit L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.RFO_HIT", "PublicDescription": "Counts the RFO (Read-for-Ownership) requests that hit L2 cache.",
@@ -148,6 +166,7 @@ }, { "BriefDescription": "RFO requests that miss L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.RFO_MISS", "PublicDescription": "Counts the RFO (Read-for-Ownership) requests that miss L2 cache.",
@@ -156,6 +175,7 @@ }, { "BriefDescription": "SW prefetch requests that hit L2 cache.", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.SWPF_HIT", "PublicDescription": "Counts Software prefetch requests that hit the L2 cache. Accounts for PREFETCHNTA and PREFETCHT0/1/2 instructions when FB is not full.",
@@ -164,6 +184,7 @@ }, { "BriefDescription": "SW prefetch requests that miss L2 cache.", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.SWPF_MISS", "PublicDescription": "Counts Software prefetch requests that miss the L2 cache. Accounts for PREFETCHNTA and PREFETCHT0/1/2 instructions when FB is not full.",
@@ -172,6 +193,7 @@ }, { "BriefDescription": "L2 writebacks that access L2 cache", + "Counter": "0,1,2,3", "EventCode": "0xF0", "EventName": "L2_TRANS.L2_WB", "PublicDescription": "Counts L2 writebacks that access L2 cache.",
@@ -180,6 +202,7 @@ }, { "BriefDescription": "Core-originated cacheable requests that missed L3 (Except hardware prefetches to the L3)", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0x2e", "EventName": "LONGEST_LAT_CACHE.MISS", "PublicDescription": "Counts core-originated cacheable requests that miss the L3 cache (Longest Latency cache). Requests include data and code reads, Reads-for-Ownership (RFOs), speculative accesses and hardware prefetches to the L1 and L2. It does not include hardware prefetches to the L3, and may not count other types of requests to the L3.",
@@ -188,6 +211,7 @@ }, { "BriefDescription": "Core-originated cacheable requests that refer to L3 (Except hardware prefetches to the L3)", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0x2e", "EventName": "LONGEST_LAT_CACHE.REFERENCE", "PublicDescription": "Counts core-originated cacheable requests to the L3 cache (Longest Latency cache). Requests include data and code reads, Reads-for-Ownership (RFOs), speculative accesses and hardware prefetches to the L1 and L2. It does not include hardware prefetches to the L3, and may not count other types of requests to the L3.",
@@ -196,6 +220,7 @@ }, { "BriefDescription": "Retired load instructions.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.ALL_LOADS",
@@ -206,6 +231,7 @@ }, { "BriefDescription": "Retired store instructions.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.ALL_STORES",
@@ -216,6 +242,7 @@ }, { "BriefDescription": "All retired memory instructions.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.ANY",
@@ -226,6 +253,7 @@ }, { "BriefDescription": "Retired load instructions with locked access.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.LOCK_LOADS",
@@ -236,6 +264,7 @@ }, { "BriefDescription": "Retired load instructions that split across a cacheline boundary.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.SPLIT_LOADS",
@@ -246,6 +275,7 @@ }, { "BriefDescription": "Retired store instructions that split across a cacheline boundary.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.SPLIT_STORES",
@@ -256,6 +286,7 @@ }, { "BriefDescription": "Retired load instructions that miss the STLB.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.STLB_MISS_LOADS",
@@ -266,6 +297,7 @@ }, { "BriefDescription": "Retired store instructions that miss the STLB.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.STLB_MISS_STORES",
@@ -276,6 +308,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources were HitM responses from shared L3", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD",
@@ -286,6 +319,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event MEM_LOAD_L3_HIT_RETIRED.XSNP_NO_FWD", + "Counter": "0,1,2,3", "Data_LA": "1", "Deprecated": "1", "EventCode": "0xd2",
@@ -296,6 +330,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD", + "Counter": "0,1,2,3", "Data_LA": "1", "Deprecated": "1", "EventCode": "0xd2",
@@ -306,6 +341,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources were L3 hit and cross-core snoop missed in on-pkg core cache.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS",
@@ -316,6 +352,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources were hits in L3 without snoops required", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_NONE",
@@ -326,6 +363,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources were L3 and cross-core snoop hits in on-pkg core cache", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_NO_FWD",
@@ -336,6 +374,7 @@ }, { "BriefDescription": "Retired load instructions which data sources missed L3 but serviced from local dram", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM",
@@ -346,6 +385,7 @@ }, { "BriefDescription": "Retired load instructions which data sources missed L3 but serviced from remote dram", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM",
@@ -355,6 +395,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources was forwarded from a remote cache", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD",
@@ -365,6 +406,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources was remote HITM", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM",
@@ -375,6 +417,7 @@ }, { "BriefDescription": "Retired load instructions with remote Intel(R) Optane(TM) DC persistent memory as the data source where the data request missed all caches.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd3", "EventName": "MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM",
@@ -385,6 +428,7 @@ }, { "BriefDescription": "Retired instructions with at least 1 uncacheable load or Bus Lock.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd4", "EventName": "MEM_LOAD_MISC_RETIRED.UC",
@@ -395,6 +439,7 @@ }, { "BriefDescription": "Number of completed demand load requests that missed the L1, but hit the FB(fill buffer), because a preceding miss to the same cacheline initiated the line to be brought into L1, but data is not yet ready in L1.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.FB_HIT",
@@ -405,6 +450,7 @@ }, { "BriefDescription": "Retired load instructions with L1 cache hits as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L1_HIT",
@@ -415,6 +461,7 @@ }, { "BriefDescription": "Retired load instructions missed L1 cache as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L1_MISS",
@@ -425,6 +472,7 @@ }, { "BriefDescription": "Retired load instructions with L2 cache hits as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L2_HIT",
@@ -435,6 +483,7 @@ }, { "BriefDescription": "Retired load instructions missed L2 cache as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L2_MISS",
@@ -445,6 +494,7 @@ }, { "BriefDescription": "Retired load instructions with L3 cache hits as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L3_HIT",
@@ -455,6 +505,7 @@ }, { "BriefDescription": "Retired load instructions missed L3 cache as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L3_MISS",
@@ -465,6 +516,7 @@ }, { "BriefDescription": "Retired load instructions with local Intel(R) Optane(TM) DC persistent memory as the data source where the data request missed all caches.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.LOCAL_PMM",
@@ -475,6 +527,7 @@ }, { "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -484,6 +537,7 @@ }, { "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that resulted in a snoop hit a modified line in another core's caches which forwarded the data.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -493,6 +547,7 @@ }, { "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that hit a modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.SNC_CACHE.HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -502,6 +557,7 @@ }, { "BriefDescription": "Counts demand instruction fetches and L1 instruction cache prefetches that either hit a non-modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_CODE_RD.SNC_CACHE.HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -511,6 +567,7 @@ }, { "BriefDescription": "Counts demand data reads that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -520,6 +577,7 @@ }, { "BriefDescription": "Counts demand data reads that resulted in a snoop hit a modified line in another core's caches which forwarded the data.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -529,6 +587,7 @@ }, { "BriefDescription": "Counts demand data reads that resulted in a snoop that hit in another core, which did not forward the data.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -538,6 +597,7 @@ }, { "BriefDescription": "Counts demand data reads that resulted in a snoop hit in another core's caches which forwarded the unmodified data to the requesting core.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -547,6 +607,7 @@ }, { "BriefDescription": "Counts demand data reads that were supplied by a cache on a remote socket where a snoop hit a modified line in another core's caches which forwarded the data.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.REMOTE_CACHE.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -556,6 +617,7 @@ }, { "BriefDescription": "Counts demand data reads that were supplied by a cache on a remote socket where a snoop hit in another core's caches which forwarded the unmodified data to the requesting core.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.REMOTE_CACHE.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -565,6 +627,7 @@ }, { "BriefDescription": "Counts demand data reads that hit a modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.SNC_CACHE.HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -574,6 +637,7 @@ }, { "BriefDescription": "Counts demand data reads that either hit a non-modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.SNC_CACHE.HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -583,6 +647,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -592,6 +657,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that resulted in a snoop hit a modified line in another core's caches which forwarded the data.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -601,6 +667,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.SNC_CACHE.HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -610,6 +677,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that either hit a non-modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.SNC_CACHE.HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -619,6 +687,7 @@ }, { "BriefDescription": "Counts L1 data cache prefetch requests and software prefetches (except PREFETCHW) that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.HWPF_L1D_AND_SWPF.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -628,6 +697,7 @@ }, { "BriefDescription": "Counts hardware prefetches to the L3 only that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.HWPF_L3.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -637,6 +707,7 @@ }, { "BriefDescription": "Counts hardware and software prefetches to all cache levels that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.PREFETCHES.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -646,6 +717,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.READS_TO_CORE.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -655,6 +727,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that resulted in a snoop hit a modified line in another core's caches which forwarded the data.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.READS_TO_CORE.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -664,6 +737,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that resulted in a snoop that hit in another core, which did not forward the data.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.READS_TO_CORE.L3_HIT.SNOOP_HIT_NO_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -673,6 +747,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that resulted in a snoop hit in another core's caches which forwarded the unmodified data to the requesting core.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.READS_TO_CORE.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -682,6 +757,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by a cache on a remote socket where a snoop was sent and data was returned (Modified or Not Modified).", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.READS_TO_CORE.REMOTE_CACHE.SNOOP_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -691,6 +767,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by a cache on a remote socket where a snoop hit a modified line in another core's caches which forwarded the data.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.READS_TO_CORE.REMOTE_CACHE.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -700,6 +777,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that were supplied by a cache on a remote socket where a snoop hit in another core's caches which forwarded the unmodified data to the requesting core.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.READS_TO_CORE.REMOTE_CACHE.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -709,6 +787,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that hit a modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.READS_TO_CORE.SNC_CACHE.HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -718,6 +797,7 @@ }, { "BriefDescription": "Counts all (cacheable) data read, code read and RFO requests including demands and prefetches to the core caches (L1 or L2) that either hit a non-modified line in a distant L3 Cache or were snooped from a distant core's L1/L2 caches on this socket when the system is in SNC (sub-NUMA cluster) mode.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.READS_TO_CORE.SNC_CACHE.HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -727,6 +807,7 @@ }, { "BriefDescription": "Counts streaming stores that hit in the L3 or were snooped from another core's caches on the same socket.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.STREAMING_WR.L3_HIT", "MSRIndex": "0x1a6,0x1a7",
@@ -736,6 +817,7 @@ }, { "BriefDescription": "Demand and prefetch data reads", + "Counter": "0,1,2,3", "EventCode": "0xB0", "EventName": "OFFCORE_REQUESTS.ALL_DATA_RD", "PublicDescription": "Counts the demand and prefetch data reads. All Core Data Reads include cacheable 'Demands' and L2 prefetchers (not L3 prefetchers). Counting also covers reads due to page walks resulted from any request type.",
@@ -744,6 +826,7 @@ }, { "BriefDescription": "Counts memory transactions sent to the uncore.", + "Counter": "0,1,2,3", "EventCode": "0xB0", "EventName": "OFFCORE_REQUESTS.ALL_REQUESTS", "PublicDescription": "Counts memory transactions sent to the uncore including requests initiated by the core, all L3 prefetches, reads resulting from page walks, and snoop responses.",
@@ -752,6 +835,7 @@ }, { "BriefDescription": "Counts cacheable and non-cacheable code reads to the core.", + "Counter": "0,1,2,3", "EventCode": "0xb0", "EventName": "OFFCORE_REQUESTS.DEMAND_CODE_RD", "PublicDescription": "Counts both cacheable and non-cacheable code reads to the core.",
@@ -760,6 +844,7 @@ }, { "BriefDescription": "Demand Data Read requests sent to uncore", + "Counter": "0,1,2,3", "EventCode": "0xb0", "EventName": "OFFCORE_REQUESTS.DEMAND_DATA_RD", "PublicDescription": "Counts the Demand Data Read requests sent to uncore. Use it in conjunction with OFFCORE_REQUESTS_OUTSTANDING to determine average latency in the uncore.",
@@ -768,6 +853,7 @@ }, { "BriefDescription": "Demand RFO requests including regular RFOs, locks, ItoM", + "Counter": "0,1,2,3", "EventCode": "0xb0", "EventName": "OFFCORE_REQUESTS.DEMAND_RFO", "PublicDescription": "Counts the demand RFO (read for ownership) requests including regular RFOs, locks, ItoM.",
@@ -776,6 +862,7 @@ }, { "BriefDescription": "For every cycle, increments by the number of outstanding data read requests pending.", + "Counter": "0,1,2,3", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD", "PublicDescription": "For every cycle, increments by the number of outstanding data read requests pending. Data read requests include cacheable demand reads and L2 prefetches, but do not include RFOs, code reads or prefetches to the L3. Reads due to page walks resulting from any request type will also be counted. Requests are considered outstanding from the time they miss the core's L2 cache until the transaction completion message is sent to the requestor.",
@@ -784,6 +871,7 @@ }, { "BriefDescription": "Cycles where at least 1 outstanding data read request is pending.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD",
@@ -793,6 +881,7 @@ }, { "BriefDescription": "Cycles with outstanding code read requests pending.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_CODE_RD",
@@ -802,6 +891,7 @@ }, { "BriefDescription": "Cycles where at least 1 outstanding Demand RFO request is pending.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO",
@@ -811,6 +901,7 @@ }, { "BriefDescription": "For every cycle, increments by the number of outstanding code read requests pending.", + "Counter": "0,1,2,3", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_CODE_RD", "PublicDescription": "For every cycle, increments by the number of outstanding code read requests pending. Code Read requests include both cacheable and non-cacheable Code Reads. Requests are considered outstanding from the time they miss the core's L2 cache until the transaction completion message is sent to the requestor.",
@@ -819,6 +910,7 @@ }, { "BriefDescription": "For every cycle, increments by the number of outstanding demand data read requests pending.", + "Counter": "0,1,2,3", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD", "PublicDescription": "For every cycle, increments by the number of outstanding demand data read requests pending. Requests are considered outstanding from the time they miss the core's L2 cache until the transaction completion message is sent to the requestor.",
@@ -827,6 +919,7 @@ }, { "BriefDescription": "Counts bus locks, accounts for cache line split locks and UC locks.", + "Counter": "0,1,2,3", "EventCode": "0xF4", "EventName": "SQ_MISC.BUS_LOCK", "PublicDescription": "Counts the more expensive bus lock needed to enforce cache coherency for certain memory accesses that need to be done atomically. Can be created by issuing an atomic instruction (via the LOCK prefix) which causes a cache line split or accesses uncacheable memory.",
@@ -835,6 +928,7 @@ }, { "BriefDescription": "Cycles the queue waiting for offcore responses is full.", + "Counter": "0,1,2,3", "EventCode": "0xf4", "EventName": "SQ_MISC.SQ_FULL", "PublicDescription": "Counts the cycles for which the thread is active and the queue waiting for responses from the uncore cannot take any more entries.",
@@ -843,6 +937,7 @@ }, { "BriefDescription": "Number of PREFETCHNTA instructions executed.", + "Counter": "0,1,2,3", "EventCode": "0x32", "EventName": "SW_PREFETCH_ACCESS.NTA", "PublicDescription": "Counts the number of PREFETCHNTA instructions executed.",
@@ -851,6 +946,7 @@ }, { "BriefDescription": "Number of PREFETCHW instructions executed.", + "Counter": "0,1,2,3", "EventCode": "0x32", "EventName": "SW_PREFETCH_ACCESS.PREFETCHW", "PublicDescription": "Counts the number of PREFETCHW instructions executed.",
@@ -859,6 +955,7 @@ }, { "BriefDescription": "Number of PREFETCHT0 instructions executed.", + "Counter": "0,1,2,3", "EventCode": "0x32", "EventName": "SW_PREFETCH_ACCESS.T0", "PublicDescription": "Counts the number of PREFETCHT0 instructions executed.",
@@ -867,6 +964,7 @@ }, { "BriefDescription": "Number of PREFETCHT1 or PREFETCHT2 instructions executed.", + "Counter": "0,1,2,3", "EventCode": "0x32", "EventName": "SW_PREFETCH_ACCESS.T1_T2", "PublicDescription": "Counts the number of PREFETCHT1 or PREFETCHT2 instructions executed.",
tools/perf/pmu-events/arch/x86/icelakex/counter.json new file +17 -0
@@ -0,0 +1,17 @@ +[ + { + "Unit": "Core", + "Size": "8", + "FixedSize": "4" + }, + { + "Unit": "UBOX", + "Size": "2", + "FixedSize": "0" + }, + { + "Unit": "IRP", + "Size": "2", + "FixedSize": "0" + } +]
tools/perf/pmu-events/arch/x86/icelakex/floating-point.json +13 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "Counts all microcode FP assists.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc1", "EventName": "ASSISTS.FP", "PublicDescription": "Counts all microcode Floating Point assists.",
@@ -9,6 +10,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 128-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE", "PublicDescription": "Number of SSE/AVX computational 128-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -17,6 +19,7 @@ }, { "BriefDescription": "Number of SSE/AVX computational 128-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE", "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -25,6 +28,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE", "PublicDescription": "Number of SSE/AVX computational 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -33,6 +37,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 256-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE", "PublicDescription": "Number of SSE/AVX computational 256-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -41,6 +46,7 @@ }, { "BriefDescription": "Number of SSE/AVX computational 128-bit packed single and 256-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and packed double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.4_FLOPS", "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision and 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point and packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -49,6 +55,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE", "PublicDescription": "Number of SSE/AVX computational 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -57,6 +64,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 512-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 16 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE", "PublicDescription": "Number of SSE/AVX computational 512-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 16 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -65,6 +73,7 @@ }, { "BriefDescription": "Number of SSE/AVX computational 256-bit packed single precision and 512-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RSQRT14 RCP RCP14 DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.8_FLOPS", "PublicDescription": "Number of SSE/AVX computational 256-bit packed single precision and 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision and double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RSQRT14 RCP RCP14 DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -73,6 +82,7 @@ }, { "BriefDescription": "Number of SSE/AVX computational scalar floating-point instructions retired; some instructions will count twice as noted below. Applies to SSE* and AVX* scalar, double and single precision floating-point: ADD SUB MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform multiple calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.SCALAR", "PublicDescription": "Number of SSE/AVX computational scalar single precision and double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -81,6 +91,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational scalar double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "PublicDescription": "Number of SSE/AVX computational scalar double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -89,6 +100,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational scalar single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "PublicDescription": "Number of SSE/AVX computational scalar single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -97,6 +109,7 @@ }, { "BriefDescription": "Number of any Vector retired FP arithmetic instructions", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.VECTOR", "SampleAfterValue": "1000003",
tools/perf/pmu-events/arch/x86/icelakex/frontend.json +55 -0
351 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/icelakex/memory.json +53 -0
355 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/icelakex/other.json +52 -0
364 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/icelakex/pipeline.json +92 -0
644 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/icelakex/uncore-cache.json +965 -0
491 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/icelakex/uncore-interconnect.json +1667 -0
1,667 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/icelakex/uncore-io.json +966 -0
966 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/icelakex/uncore-memory.json +186 -0
186 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/icelakex/uncore-power.json +26 -0
Lines of this file are not shown. View the raw message.
tools/perf/pmu-events/arch/x86/icelakex/virtual-memory.json +22 -0
Lines of this file are not shown. View the raw message.
The rest of this diff is not shown. View the raw message.
Signature · 1 line
2.39.3
Part of the diff is not shown here. View the raw message.
RFC PATCH v2 7/17 perf stat: Add functions to set counter bitmaps for hardware-grouping method
In reply to [email protected]From: Weilin Wang <[email protected]>
Add metricgroup__event_info data structure to represent an event in the metric grouping context; the list of counters and the PMU name an event should be collected with.
Add functions to parse event counter info from pmu-events and generate a list of metricgroup__event_info data to prepare grouping.
--- tools/perf/util/metricgroup.c | 196 +++++++++++++++++++++++++++++++++- tools/perf/util/metricgroup.h | 27 +++++ 2 files changed, 220 insertions(+), 3 deletions(-)
tools/perf/util/metricgroup.c +193 -3
@@ -1432,6 +1432,182 @@ static int build_combined_expr_ctx(const struct list_head *metric_list,return ret; } +/** +* set_counter_bitmap - The counter bit mapping: [8-15,0-7], e.g. the GP0 is the +* 8th bit and GP7 is the 1st bit in this 16-bits bitmap. It is helpful for +* assigning GP4-7 before GP0-3 because some events can be collected using GP0-3 +* only on some platforms. +*/ +static int set_counter_bitmap (int pos,unsigned long *bitmap) +{ +if (pos >=NR_COUNTERS || pos <0 ) +return -EINVAL ; +if (pos <=7 ) + pos =TRANSFER_FIRST_BYTE (pos); +else + pos =TRANSFER_SEC_BYTE (pos); + *bitmap |=1ul << pos; +return 0 ; +} + +static int parse_fixed_counter (const char *counter, +unsigned long *bitmap, +bool *fixed) +{ +int ret = -ENOENT ; +//TODO: this pattern is different on some other platforms +const char *pattern ="Fixed counter " ; +int pos =0 ; + +if (!strncmp (counter, pattern,strlen (pattern))) { + pos =atoi (counter +strlen (pattern)); + ret =set_counter_bitmap (pos, bitmap); +if (ret) +return ret; + *fixed = true; +return 0 ; + } +return ret; +} + +/** +* parse_counter - Parse event counter info from pmu-events and set up bitmap +* accordingly. +* +* @counter: counter info string to be parsed. +* @bitmap: bitmap to set based on counter info parsed. +* @fixed: is set to true if the event uses fixed counter. +*/ +static int parse_counter (const char *counter, +unsigned long *bitmap, +bool *fixed) +{ +int ret =0 ; +char *p; +char *tok; +int pos =0 ; + + ret =parse_fixed_counter (counter, bitmap, fixed); +// ret==0 means matched with fixed counter +if (ret ==0 ) +return ret; + + p =strdup (counter); + tok =strtok (p,"," ); +if (!tok) +return -ENOENT ; + +while (tok) { + pos =atoi (tok); + ret =set_counter_bitmap (pos, bitmap); +if (ret) +return ret; + tok =strtok (NULL ,"," ); + } +return 0 ; +} + +static struct metricgroup__event_info *event_info__new (const char *name, +const char *pmu_name, +const char *counter, +bool free_counter) +{ +int ret =0 ; +char *bit_buf =malloc (NR_COUNTERS ); +bool fixed_counter = false; +struct metricgroup__event_info *e; + + e =zalloc (sizeof (*e)); +if (!e) +return NULL ; +if (!pmu_name) + pmu_name ="core" ; + + e->name = name; + e->free_counter = free_counter; + e->pmu_name =strdup (pmu_name); +if (free_counter) { + ret =set_counter_bitmap (0 , e->counters); +if (ret) +return NULL ; + }else { + ret =parse_counter (counter, e->counters, &fixed_counter); +if (ret) +return NULL ; + e->fixed_counter = fixed_counter; + } + +bitmap_scnprintf (e->counters,NR_COUNTERS , bit_buf,NR_COUNTERS ); +pr_debug ("Event %s requires pmu %s counter: %s bitmap %s, [pmu=%s]\n" , + e->name, e->pmu_name, counter, bit_buf, pmu_name); + +return e; +} + +struct metricgroup__add_metric_event_data { +struct list_head *list; +/* pure event name, exclude umask and other info*/ +const char *event_name; +/* event name and umask if applicable*/ +const char *event_id; +}; + +static int metricgroup__add_metric_event_callback (const struct pmu_event *pe, +const struct pmu_events_table *table__maybe_unused , +void *data) +{ +struct metricgroup__event_info *event; +struct metricgroup__add_metric_event_data *d = data; + +if (!strcasecmp (pe->name, d->event_name)) { + event =event_info__new (d->event_id, pe->pmu, pe->counter,/*free_counter=*/ false); +if (!event) +return -ENOMEM ; +list_add (&event->nd, d->list); + } + +return 0 ; +} + +/** +* get_metricgroup_events - Find counter requirement of events from the +* pmu_events table +* @full_id: the full event identifiers. +* @table: pmu_events table that is searched for event data. +* @event_info_list: the list that the new event counter info added to. +*/ +static int get_metricgroup_events (const char *full_id, +const struct pmu_events_table *table, +struct list_head *event_info_list) +{ +LIST_HEAD (list); +int ret =0 ; +const char *id; +const char *rsep, *sep =strchr (full_id,'@' ); + +if (sep) { + rsep =strchr (full_id,',' ); + id =strndup (sep +1 , rsep - sep -1 ); +if (ret) + goto out; + }else { + id = full_id; + } + { +struct metricgroup__add_metric_event_data data = { + .list = &list, + .event_name = id, + .event_id = full_id, + }; + ret =pmu_events_table_for_each_event (table, + metricgroup__add_metric_event_callback, &data); + } + +out: +list_splice (&list, event_info_list); +return ret; +} +/** * hw_aware_build_grouping - Build event groupings by reading counter * requirement of the events and counter available on the system from
@@ -1445,9 +1621,25 @@ static int hw_aware_build_grouping(struct expr_parse_ctx *ctx __maybe_unused, const char *modifier __maybe_unused) {int ret =0 ; +struct hashmap_entry *cur; +LIST_HEAD (pmu_info_list); +LIST_HEAD (event_info_list); +size_t bkt; +const struct pmu_events_table *etable =pmu_events_table__find (); + +#define RETURN_IF_NON_ZERO (x) do { if (x) return x; } while (0) +hashmap__for_each_entry (ctx->ids, cur, bkt) { +const char *id = cur->pkey; + +pr_debug ("found event %s\n" , id); + + ret =get_metricgroup_events (id, etable, &event_info_list); +if (ret) +return ret; + } -pr_debug ("This is a placeholder\n" );return ret; +#undef RETURN_IF_NON_ZERO }static void group_str_free (struct metricgroup__group_strs *g)
@@ -1521,8 +1713,6 @@ static int hw_aware_parse_ids(struct perf_pmu *fake_pmu, *out_evlist = parsed_evlist; parsed_evlist =NULL ; err_out: -parse_events_error__exit (&parse_error); -evlist__delete (parsed_evlist);metricgroup__free_grouping_strs (&groupings);return ret; }
tools/perf/util/metricgroup.h +27 -0
@@ -5,6 +5,7 @@#include <linux/list.h> #include <linux/rbtree.h> #include <stdbool.h> +#include <linux/bitmap.h> #include "pmu-events/pmu-events.h" #include "strbuf.h"
@@ -67,6 +68,32 @@ struct metric_expr {int runtime; }; +/* Maximum number of counters per PMU*/ +#define NR_COUNTERS 16 +/* +* Transfer bit position in the bitmap to ensure start assigning counter from +* the last GP counter to the first. +* bit15 <---> bit0 +* [GP8-GP15] [GP0-GP7] +*/ +#define TRANSFER_FIRST_BYTE (pos) (7 - pos) +#define TRANSFER_SEC_BYTE (pos) (23 - pos) + +/** +* An event used in a metric. This info is for metric grouping. +*/ +struct metricgroup__event_info { +struct list_head nd; +/** The name of the event. */ +const char *name; +/** The name of the pmu the event be collected on. */ +const char *pmu_name; +bool fixed_counter; +bool free_counter; +/** The counters the event allowed to be collected on. */ + DECLARE_BITMAP(counters, NR_COUNTERS); +}; +/** * Each group is one node in the group string list. */
Signature · 1 line
2.39.3
Re: [RFC PATCH v2 07/17] perf stat: Add functions to set counter bitmaps for hardware-grouping method
Quoted 20 lines · On 2023-10-13 9:51 p.m., [email protected] wrote:
From: Weilin Wang <[email protected]>
Add
metricgroup__event_infodata structure to represent an event in the metric grouping context; the list of counters and the PMU name an event should be collected with.Add functions to parse event counter info from pmu-events and generate a list of
metricgroup__event_infodata to prepare grouping.Signed-off-by Weilin Wang <[email protected]>--- tools/perf/util/metricgroup.c | 196 +++++++++++++++++++++++++++++++++- tools/perf/util/metricgroup.h | 27 +++++ 2 files changed, 220 insertions(+), 3 deletions(-)tools/perf/util/metricgroup.c +0
-0
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list, · line 1439
Quoted 8 lines
tools/perf/util/metricgroup.c +5
-0@@ -1432,6 +1432,182 @@ static int build_combined_expr_ctx(const struct list_head *metric_list,return ret; } +/** +* set_counter_bitmap - The counter bit mapping: [8-15,0-7], e.g. the GP0 is the +* 8th bit and GP7 is the 1st bit in this 16-bits bitmap. It is helpful for +* assigning GP4-7 before GP0-3 because some events can be collected using GP0-3 +* only on some platforms.
The bitmap looks weird. Can we use the normal bitmap and always search from the last set bit to find the available counter?
Quoted 42 lines
tools/perf/util/metricgroup.c +42
-0+*/ +static int set_counter_bitmap (int pos,unsigned long *bitmap) +{ +if (pos >=NR_COUNTERS || pos <0 ) +return -EINVAL ; +if (pos <=7 ) + pos =TRANSFER_FIRST_BYTE (pos); +else + pos =TRANSFER_SEC_BYTE (pos); + *bitmap |=1ul << pos; +return 0 ; +} + +static int parse_fixed_counter (const char *counter, +unsigned long *bitmap, +bool *fixed) +{ +int ret = -ENOENT ; +//TODO: this pattern is different on some other platforms +const char *pattern ="Fixed counter " ; +int pos =0 ; + +if (!strncmp (counter, pattern,strlen (pattern))) { + pos =atoi (counter +strlen (pattern)); + ret =set_counter_bitmap (pos, bitmap); +if (ret) +return ret; + *fixed = true; +return 0 ; + } +return ret; +} + +/** +* parse_counter - Parse event counter info from pmu-events and set up bitmap +* accordingly. +* +* @counter: counter info string to be parsed. +* @bitmap: bitmap to set based on counter info parsed. +* @fixed: is set to true if the event uses fixed counter. +*/ +static int parse_counter(const char *counter,
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list, · line 1491
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+unsigned long *bitmap, + bool *fixed) +{ + int ret = 0; +char *p; +char *tok; +int pos =0 ; + + ret =parse_fixed_counter (counter, bitmap, fixed); +// ret==0 means matched with fixed counter
Move the comments to the above of parse_fixed_counter(). Uses /**/.
Quoted 103 lines
tools/perf/util/metricgroup.c +103
-0+if (ret ==0 ) +return ret; + + p =strdup (counter); + tok =strtok (p,"," ); +if (!tok) +return -ENOENT ; + +while (tok) { + pos =atoi (tok); + ret =set_counter_bitmap (pos, bitmap); +if (ret) +return ret; + tok =strtok (NULL ,"," ); + } +return 0 ; +} + +static struct metricgroup__event_info *event_info__new (const char *name, +const char *pmu_name, +const char *counter, +bool free_counter) +{ +int ret =0 ; +char *bit_buf =malloc (NR_COUNTERS ); +bool fixed_counter = false; +struct metricgroup__event_info *e; + + e =zalloc (sizeof (*e)); +if (!e) +return NULL ; +if (!pmu_name) + pmu_name ="core" ; + + e->name = name; + e->free_counter = free_counter; + e->pmu_name =strdup (pmu_name); +if (free_counter) { + ret =set_counter_bitmap (0 , e->counters); +if (ret) +return NULL ; + }else { + ret =parse_counter (counter, e->counters, &fixed_counter); +if (ret) +return NULL ; + e->fixed_counter = fixed_counter; + } + +bitmap_scnprintf (e->counters,NR_COUNTERS , bit_buf,NR_COUNTERS ); +pr_debug ("Event %s requires pmu %s counter: %s bitmap %s, [pmu=%s]\n" , + e->name, e->pmu_name, counter, bit_buf, pmu_name); + +return e; +} + +struct metricgroup__add_metric_event_data { +struct list_head *list; +/* pure event name, exclude umask and other info*/ +const char *event_name; +/* event name and umask if applicable*/ +const char *event_id; +}; + +static int metricgroup__add_metric_event_callback (const struct pmu_event *pe, +const struct pmu_events_table *table__maybe_unused , +void *data) +{ +struct metricgroup__event_info *event; +struct metricgroup__add_metric_event_data *d = data; + +if (!strcasecmp (pe->name, d->event_name)) { + event =event_info__new (d->event_id, pe->pmu, pe->counter,/*free_counter=*/ false); +if (!event) +return -ENOMEM ; +list_add (&event->nd, d->list); + } + +return 0 ; +} + +/** +* get_metricgroup_events - Find counter requirement of events from the +* pmu_events table +* @full_id: the full event identifiers. +* @table: pmu_events table that is searched for event data. +* @event_info_list: the list that the new event counter info added to. +*/ +static int get_metricgroup_events (const char *full_id, +const struct pmu_events_table *table, +struct list_head *event_info_list) +{ +LIST_HEAD (list); +int ret =0 ; +const char *id; +const char *rsep, *sep =strchr (full_id,'@' ); + +if (sep) { + rsep =strchr (full_id,',' ); + id =strndup (sep +1 , rsep - sep -1 ); +if (ret) + goto out; + }else { + id = full_id;
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list, · line 1604
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+ } + { +struct metricgroup__add_metric_event_data data = { + .list = &list, + .event_name = id, + .event_id = full_id, + }; + ret =pmu_events_table_for_each_event (table, + metricgroup__add_metric_event_callback, &data); + }
Please remove the useless {}
Thanks,
Kan
Quoted 90 lines
tools/perf/util/metricgroup.c +23
-3+ +out: +list_splice (&list, event_info_list); +return ret; +} +/** * hw_aware_build_grouping - Build event groupings by reading counter * requirement of the events and counter available on the system from @@ -1445,9 +1621,25 @@ static int hw_aware_build_grouping(struct expr_parse_ctx *ctx __maybe_unused, const char *modifier __maybe_unused) {int ret =0 ; +struct hashmap_entry *cur; +LIST_HEAD (pmu_info_list); +LIST_HEAD (event_info_list); +size_t bkt; +const struct pmu_events_table *etable =pmu_events_table__find (); + +#define RETURN_IF_NON_ZERO (x) do { if (x) return x; } while (0) +hashmap__for_each_entry (ctx->ids, cur, bkt) { +const char *id = cur->pkey; + +pr_debug ("found event %s\n" , id); + + ret =get_metricgroup_events (id, etable, &event_info_list); +if (ret) +return ret; + } -pr_debug ("This is a placeholder\n" );return ret; +#undef RETURN_IF_NON_ZERO }static void group_str_free (struct metricgroup__group_strs *g)@@ -1521,8 +1713,6 @@ static int hw_aware_parse_ids(struct perf_pmu *fake_pmu, *out_evlist = parsed_evlist; parsed_evlist =NULL ; err_out: -parse_events_error__exit (&parse_error); -evlist__delete (parsed_evlist);metricgroup__free_grouping_strs (&groupings);return ret; }tools/perf/util/metricgroup.h +27
-0@@ -5,6 +5,7 @@#include <linux/list.h> #include <linux/rbtree.h> #include <stdbool.h> +#include <linux/bitmap.h> #include "pmu-events/pmu-events.h" #include "strbuf.h" @@ -67,6 +68,32 @@ struct metric_expr {int runtime; }; +/* Maximum number of counters per PMU*/ +#define NR_COUNTERS 16 +/* +* Transfer bit position in the bitmap to ensure start assigning counter from +* the last GP counter to the first. +* bit15 <---> bit0 +* [GP8-GP15] [GP0-GP7] +*/ +#define TRANSFER_FIRST_BYTE (pos) (7 - pos) +#define TRANSFER_SEC_BYTE (pos) (23 - pos) + +/** +* An event used in a metric. This info is for metric grouping. +*/ +struct metricgroup__event_info { +struct list_head nd; +/** The name of the event. */ +const char *name; +/** The name of the pmu the event be collected on. */ +const char *pmu_name; +bool fixed_counter; +bool free_counter; +/** The counters the event allowed to be collected on. */ + DECLARE_BITMAP(counters, NR_COUNTERS); +}; +/** * Each group is one node in the group string list. */
RE: [RFC PATCH v2 07/17] perf stat: Add functions to set counter bitmaps for hardware-grouping method
Quoted 48 lines
-----Original Message-----
From: Liang, Kan <[email protected]>
Sent: Monday, October 30, 2023 8:32 AM
To: Wang, Weilin <[email protected]>; Rogers, Ian <[email protected]>; Peter Zijlstra <[email protected]>; Ingo Molnar <[email protected]>; Arnaldo Carvalho de Melo <[email protected]>; Alexander Shishkin <[email protected]>; Jiri Olsa <[email protected]>; Namhyung Kim <[email protected]>; Hunter, Adrian <[email protected]>
Cc: [email protected]; [email protected]; Taylor, Perry <[email protected]>; Alt, Samantha <[email protected]>; Biggers, Caleb <[email protected]>; Mark Rutland <[email protected]>; Yang Jihong <[email protected]>
Subject: Re: [RFC PATCH v2 07/17] perf stat: Add functions to set counter bitmaps for hardware-grouping methodQuoted 6 lines · On 2023-10-13 9:51 p.m., [email protected] wrote:
From: Weilin Wang <[email protected]>
Add
metricgroup__event_infodata structure to represent an event in the metric grouping context; the list of counters and the PMU name an event should be collected with.Quoted 6 lines
Add functions to parse event counter info from pmu-events and generate a list of
metricgroup__event_infodata to prepare grouping.Signed-off-by Weilin Wang <[email protected]>--- tools/perf/util/metricgroup.c | 196+++++++++++++++++++++++++++++++++-
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const
Quoted 8 lines
tools/perf/util/metricgroup.h | 27 +++++ 2 files changed, 220 insertions(+), 3 deletions(-)tools/perf/util/metricgroup.c +0
-0@@ -1432,6 +1432,182 @@ static int build_combined_expr_ctx(conststruct
list_head*metric_list,tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 5 lines
tools/perf/util/metricgroup.c +2
-0return ret; } +/** +* set_counter_bitmap - The counter bit mapping: [8-15,0-7], e.g. the GP0
Quoted 10 lines
is the
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const · line 1438
Quoted 2 lines
tools/perf/util/metricgroup.c +2
-0+* 8th bit and GP7 is the 1st bit in this 16-bits bitmap. It is helpful for +* assigning GP4-7 before GP0-3 because some events can be collected using GP0-3
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list, · line 1439
Quoted 1 line
tools/perf/util/metricgroup.c +1
-0+* only on some platforms. The bitmap looks weird. Can we use the normal bitmap and always search from the last set bit to find the available counter?
Hi Kan,
Yes, I just want to ensure to start using the highest indexed counter first. Could you please let me know what is the best way to search from the last set bit?
Thanks,
Weilin
Quoted 274 lines
Quoted 25 lines
tools/perf/util/metricgroup.c +25
-0+*/ +static int set_counter_bitmap (int pos,unsigned long *bitmap) +{ +if (pos >=NR_COUNTERS || pos <0 ) +return -EINVAL ; +if (pos <=7 ) + pos =TRANSFER_FIRST_BYTE (pos); +else + pos =TRANSFER_SEC_BYTE (pos); + *bitmap |=1ul << pos; +return 0 ; +} + +static int parse_fixed_counter (const char *counter, +unsigned long *bitmap, +bool *fixed) +{ +int ret = -ENOENT ; +//TODO: this pattern is different on some other platforms +const char *pattern ="Fixed counter " ; +int pos =0 ; + +if (!strncmp (counter, pattern,strlen (pattern))) { + pos =atoi (counter +strlen (pattern)); + ret =set_counter_bitmap (pos, bitmap);tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+if (ret) +return ret; + *fixed = true; +return 0 ; + } +return ret; +} + +/** +* parse_counter - Parse event counter info from pmu-events and set up bitmap
Quoted 7 lines
tools/perf/util/metricgroup.c +7
-0+* accordingly. +* +* @counter: counter info string to be parsed. +* @bitmap: bitmap to set based on counter info parsed. +* @fixed: is set to true if the event uses fixed counter. +*/ +static int parse_counter(const char *counter,tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list, · line 1491
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+unsigned long *bitmap, + bool *fixed) +{ + int ret = 0; +char *p; +char *tok; +int pos =0 ; + + ret =parse_fixed_counter (counter, bitmap, fixed); +// ret==0 means matched with fixed counter Move the comments to the above of
parse_fixed_counter(). Uses /**/.Quoted 9 lines
tools/perf/util/metricgroup.c +9
-0+if (ret ==0 ) +return ret; + + p =strdup (counter); + tok =strtok (p,"," ); +if (!tok) +return -ENOENT ; + +while (tok) {tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+ pos =atoi (tok); + ret =set_counter_bitmap (pos, bitmap); +if (ret) +return ret; + tok =strtok (NULL ,"," ); + } +return 0 ; +} + +static struct metricgroup__event_info *event_info__new(const char*name,
Quoted 20 lines
tools/perf/util/metricgroup.c +20
-0+const char *pmu_name, + const char *counter, + bool free_counter) +{ + int ret = 0; +char *bit_buf =malloc (NR_COUNTERS ); +bool fixed_counter = false; +struct metricgroup__event_info *e; + + e =zalloc (sizeof (*e)); +if (!e) +return NULL ; +if (!pmu_name) + pmu_name ="core" ; + + e->name = name; + e->free_counter = free_counter; + e->pmu_name =strdup (pmu_name); +if (free_counter) { + ret =set_counter_bitmap (0 , e->counters);tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+if (ret) +return NULL ; + }else { + ret =parse_counter (counter, e->counters, &fixed_counter); +if (ret) +return NULL ; + e->fixed_counter = fixed_counter; + } + + bitmap_scnprintf(e->counters, NR_COUNTERS, bit_buf,
NR_COUNTERS);tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const · line 1541
Quoted 1 line
tools/perf/util/metricgroup.c +1
-0+ pr_debug("Event %s requires pmu %s counter: %s bitmap %s,[pmu=%s]\n",
Quoted 4 lines
tools/perf/util/metricgroup.c +4
-0+ e->name, e->pmu_name, counter, bit_buf, pmu_name); + +return e; +}tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+ +struct metricgroup__add_metric_event_data { +struct list_head *list; +/* pure event name, exclude umask and other info*/ +const char *event_name; +/* event name and umask if applicable*/ +const char *event_id; +}; + +static int metricgroup__add_metric_event_callback(const struct pmu_event*pe,
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const · line 1556
Quoted 1 line
tools/perf/util/metricgroup.c +1
-0+ const struct
pmu_events_table*table__maybe_unused,tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 7 lines
tools/perf/util/metricgroup.c +7
-0+ void *data) +{ +struct metricgroup__event_info *event; +struct metricgroup__add_metric_event_data *d = data; + +if (!strcasecmp (pe->name, d->event_name)) { + event = event_info__new(d->event_id, pe->pmu, pe->counter,/*free_counter=*/false);
Quoted 30 lines
tools/perf/util/metricgroup.c +30
-0+if (!event) +return -ENOMEM ; +list_add (&event->nd, d->list); + } + +return 0 ; +} + +/** +* get_metricgroup_events - Find counter requirement of events from the +* pmu_events table +* @full_id: the full event identifiers. +* @table: pmu_events table that is searched for event data. +* @event_info_list: the list that the new event counter info added to. +*/ +static int get_metricgroup_events (const char *full_id, +const struct pmu_events_table *table, +struct list_head *event_info_list) +{ +LIST_HEAD (list); +int ret =0 ; +const char *id; +const char *rsep, *sep =strchr (full_id,'@' ); + +if (sep) { + rsep =strchr (full_id,',' ); + id =strndup (sep +1 , rsep - sep -1 ); +if (ret) + goto out; + }else {tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+ id = full_id; + } + { +struct metricgroup__add_metric_event_data data = { + .list = &list, + .event_name = id, + .event_id = full_id, + }; + ret = pmu_events_table_for_each_event(table, + metricgroup__add_metric_event_callback,&data);
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const · line 1604
Quoted 1 line
tools/perf/util/metricgroup.c +1
-0+ }Please remove the useless {}
Thanks,
Kantools/perf/util/metricgroup.c static int hw_aware_build_grouping(struct
Quoted 10 lines
tools/perf/util/metricgroup.c +6
-0+ +out: +list_splice (&list, event_info_list); +return ret; +} +/** * hw_aware_build_grouping - Build event groupings by reading counter * requirement of the events and counter available on the system from @@ -1445,9 +1621,25 @@ static int hw_aware_build_grouping(struct
expr_parse_ctx*ctx__maybe_unused,tools/perf/util/metricgroup.c static int hw_aware_parse_ids(struct perf_pmu
Quoted 27 lines
tools/perf/util/metricgroup.c +17
-1const char *modifier __maybe_unused) {int ret =0 ; +struct hashmap_entry *cur; +LIST_HEAD (pmu_info_list); +LIST_HEAD (event_info_list); +size_t bkt; +const struct pmu_events_table *etable =pmu_events_table__find (); + +#define RETURN_IF_NON_ZERO (x) do { if (x) return x; } while (0) +hashmap__for_each_entry (ctx->ids, cur, bkt) { +const char *id = cur->pkey; + +pr_debug ("found event %s\n" , id); + + ret =get_metricgroup_events (id, etable, &event_info_list); +if (ret) +return ret; + } -pr_debug ("This is a placeholder\n" );return ret; +#undef RETURN_IF_NON_ZERO }static void group_str_free (struct metricgroup__group_strs *g)@@ -1521,8 +1713,6 @@ static int hw_aware_parse_ids(struct perf_pmu
*fake_pmu,Quoted 53 lines
tools/perf/util/metricgroup.c +0
-2*out_evlist = parsed_evlist; parsed_evlist =NULL ; err_out: -parse_events_error__exit (&parse_error); -evlist__delete (parsed_evlist);metricgroup__free_grouping_strs (&groupings);return ret; }tools/perf/util/metricgroup.h +27
-0@@ -5,6 +5,7 @@#include <linux/list.h> #include <linux/rbtree.h> #include <stdbool.h> +#include <linux/bitmap.h> #include "pmu-events/pmu-events.h" #include "strbuf.h" @@ -67,6 +68,32 @@ struct metric_expr {int runtime; }; +/* Maximum number of counters per PMU*/ +#define NR_COUNTERS 16 +/* +* Transfer bit position in the bitmap to ensure start assigning counter from +* the last GP counter to the first. +* bit15 <---> bit0 +* [GP8-GP15] [GP0-GP7] +*/ +#define TRANSFER_FIRST_BYTE (pos) (7 - pos) +#define TRANSFER_SEC_BYTE (pos) (23 - pos) + +/** +* An event used in a metric. This info is for metric grouping. +*/ +struct metricgroup__event_info { +struct list_head nd; +/** The name of the event. */ +const char *name; +/** The name of the pmu the event be collected on. */ +const char *pmu_name; +bool fixed_counter; +bool free_counter; +/** The counters the event allowed to be collected on. */ + DECLARE_BITMAP(counters, NR_COUNTERS); +}; +/** * Each group is one node in the group string list. */
Re: [RFC PATCH v2 07/17] perf stat: Add functions to set counter bitmaps for hardware-grouping method
Quoted 59 lines · Weilin
Quoted 48 lines
-----Original Message-----
From: Liang, Kan <[email protected]>
Sent: Monday, October 30, 2023 8:32 AM
To: Wang, Weilin <[email protected]>; Rogers, Ian <[email protected]>; Peter Zijlstra <[email protected]>; Ingo Molnar <[email protected]>; Arnaldo Carvalho de Melo <[email protected]>; Alexander Shishkin <[email protected]>; Jiri Olsa <[email protected]>; Namhyung Kim <[email protected]>; Hunter, Adrian <[email protected]>
Cc: [email protected]; [email protected]; Taylor, Perry <[email protected]>; Alt, Samantha <[email protected]>; Biggers, Caleb <[email protected]>; Mark Rutland <[email protected]>; Yang Jihong <[email protected]>
Subject: Re: [RFC PATCH v2 07/17] perf stat: Add functions to set counter bitmaps for hardware-grouping methodQuoted 6 lines · On 2023-10-13 9:51 p.m., [email protected] wrote:
From: Weilin Wang <[email protected]>
Add
metricgroup__event_infodata structure to represent an event in the metric grouping context; the list of counters and the PMU name an event should be collected with.Quoted 6 lines
Add functions to parse event counter info from pmu-events and generate a list of
metricgroup__event_infodata to prepare grouping.Signed-off-by Weilin Wang <[email protected]>--- tools/perf/util/metricgroup.c | 196+++++++++++++++++++++++++++++++++-
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const
Quoted 8 lines
tools/perf/util/metricgroup.h | 27 +++++ 2 files changed, 220 insertions(+), 3 deletions(-)tools/perf/util/metricgroup.c +0
-0@@ -1432,6 +1432,182 @@ static int build_combined_expr_ctx(conststruct
list_head*metric_list,tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 5 lines
tools/perf/util/metricgroup.c +2
-0return ret; } +/** +* set_counter_bitmap - The counter bit mapping: [8-15,0-7], e.g. the GP0 Quoted 10 lines
is the
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const · line 1438
Quoted 2 lines
tools/perf/util/metricgroup.c +2
-0+* 8th bit and GP7 is the 1st bit in this 16-bits bitmap. It is helpful for +* assigning GP4-7 before GP0-3 because some events can be collected using GP0-3
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list, · line 1439
Quoted 1 line
tools/perf/util/metricgroup.c +1
-0+* only on some platforms. The bitmap looks weird. Can we use the normal bitmap and always search from the last set bit to find the available counter?
Quoted 5 lines
Hi Kan,
Yes, I just want to ensure to start using the highest indexed counter first. Could you please let me know what is the best way to search from the last set bit?
I didn't find any existing helper that does reverse search from the last set bit. But it should not be hard to implement one.
An alternative way would start the search from the middle, e.g., for_each_set_bit_wrap(), which should be what you try to achieve here.
Thanks,
Kan
Quoted 278 lines
Thanks,
WeilinQuoted 274 lines
Quoted 25 lines
tools/perf/util/metricgroup.c +25
-0+*/ +static int set_counter_bitmap (int pos,unsigned long *bitmap) +{ +if (pos >=NR_COUNTERS || pos <0 ) +return -EINVAL ; +if (pos <=7 ) + pos =TRANSFER_FIRST_BYTE (pos); +else + pos =TRANSFER_SEC_BYTE (pos); + *bitmap |=1ul << pos; +return 0 ; +} + +static int parse_fixed_counter (const char *counter, +unsigned long *bitmap, +bool *fixed) +{ +int ret = -ENOENT ; +//TODO: this pattern is different on some other platforms +const char *pattern ="Fixed counter " ; +int pos =0 ; + +if (!strncmp (counter, pattern,strlen (pattern))) { + pos =atoi (counter +strlen (pattern)); + ret =set_counter_bitmap (pos, bitmap);tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+if (ret) +return ret; + *fixed = true; +return 0 ; + } +return ret; +} + +/** +* parse_counter - Parse event counter info from pmu-events and set up bitmap
Quoted 7 lines
tools/perf/util/metricgroup.c +7
-0+* accordingly. +* +* @counter: counter info string to be parsed. +* @bitmap: bitmap to set based on counter info parsed. +* @fixed: is set to true if the event uses fixed counter. +*/ +static int parse_counter(const char *counter,tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list, · line 1491
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+unsigned long *bitmap, + bool *fixed) +{ + int ret = 0; +char *p; +char *tok; +int pos =0 ; + + ret =parse_fixed_counter (counter, bitmap, fixed); +// ret==0 means matched with fixed counter Move the comments to the above of
parse_fixed_counter(). Uses /**/.Quoted 9 lines
tools/perf/util/metricgroup.c +9
-0+if (ret ==0 ) +return ret; + + p =strdup (counter); + tok =strtok (p,"," ); +if (!tok) +return -ENOENT ; + +while (tok) {tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+ pos =atoi (tok); + ret =set_counter_bitmap (pos, bitmap); +if (ret) +return ret; + tok =strtok (NULL ,"," ); + } +return 0 ; +} + +static struct metricgroup__event_info *event_info__new(const char*name,
Quoted 20 lines
tools/perf/util/metricgroup.c +20
-0+const char *pmu_name, + const char *counter, + bool free_counter) +{ + int ret = 0; +char *bit_buf =malloc (NR_COUNTERS ); +bool fixed_counter = false; +struct metricgroup__event_info *e; + + e =zalloc (sizeof (*e)); +if (!e) +return NULL ; +if (!pmu_name) + pmu_name ="core" ; + + e->name = name; + e->free_counter = free_counter; + e->pmu_name =strdup (pmu_name); +if (free_counter) { + ret =set_counter_bitmap (0 , e->counters);tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+if (ret) +return NULL ; + }else { + ret =parse_counter (counter, e->counters, &fixed_counter); +if (ret) +return NULL ; + e->fixed_counter = fixed_counter; + } + + bitmap_scnprintf(e->counters, NR_COUNTERS, bit_buf,
NR_COUNTERS);tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const · line 1541
Quoted 1 line
tools/perf/util/metricgroup.c +1
-0+ pr_debug("Event %s requires pmu %s counter: %s bitmap %s,[pmu=%s]\n",
Quoted 4 lines
tools/perf/util/metricgroup.c +4
-0+ e->name, e->pmu_name, counter, bit_buf, pmu_name); + +return e; +}tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+ +struct metricgroup__add_metric_event_data { +struct list_head *list; +/* pure event name, exclude umask and other info*/ +const char *event_name; +/* event name and umask if applicable*/ +const char *event_id; +}; + +static int metricgroup__add_metric_event_callback(const struct pmu_event*pe,
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const · line 1556
Quoted 1 line
tools/perf/util/metricgroup.c +1
-0+ const struct
pmu_events_table*table__maybe_unused,tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 7 lines
tools/perf/util/metricgroup.c +7
-0+ void *data) +{ +struct metricgroup__event_info *event; +struct metricgroup__add_metric_event_data *d = data; + +if (!strcasecmp (pe->name, d->event_name)) { + event = event_info__new(d->event_id, pe->pmu, pe->counter,/*free_counter=*/false);
Quoted 30 lines
tools/perf/util/metricgroup.c +30
-0+if (!event) +return -ENOMEM ; +list_add (&event->nd, d->list); + } + +return 0 ; +} + +/** +* get_metricgroup_events - Find counter requirement of events from the +* pmu_events table +* @full_id: the full event identifiers. +* @table: pmu_events table that is searched for event data. +* @event_info_list: the list that the new event counter info added to. +*/ +static int get_metricgroup_events (const char *full_id, +const struct pmu_events_table *table, +struct list_head *event_info_list) +{ +LIST_HEAD (list); +int ret =0 ; +const char *id; +const char *rsep, *sep =strchr (full_id,'@' ); + +if (sep) { + rsep =strchr (full_id,',' ); + id =strndup (sep +1 , rsep - sep -1 ); +if (ret) + goto out; + }else {tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const struct list_head *metric_list,
Quoted 10 lines
tools/perf/util/metricgroup.c +10
-0+ id = full_id; + } + { +struct metricgroup__add_metric_event_data data = { + .list = &list, + .event_name = id, + .event_id = full_id, + }; + ret = pmu_events_table_for_each_event(table, + metricgroup__add_metric_event_callback,&data);
tools/perf/util/metricgroup.c static int build_combined_expr_ctx(const · line 1604
Quoted 1 line
tools/perf/util/metricgroup.c +1
-0+ }Please remove the useless {}
Thanks,
Kantools/perf/util/metricgroup.c static int hw_aware_build_grouping(struct
Quoted 10 lines
tools/perf/util/metricgroup.c +6
-0+ +out: +list_splice (&list, event_info_list); +return ret; +} +/** * hw_aware_build_grouping - Build event groupings by reading counter * requirement of the events and counter available on the system from @@ -1445,9 +1621,25 @@ static int hw_aware_build_grouping(struct
expr_parse_ctx*ctx__maybe_unused,tools/perf/util/metricgroup.c static int hw_aware_parse_ids(struct perf_pmu
Quoted 27 lines
tools/perf/util/metricgroup.c +17
-1const char *modifier __maybe_unused) {int ret =0 ; +struct hashmap_entry *cur; +LIST_HEAD (pmu_info_list); +LIST_HEAD (event_info_list); +size_t bkt; +const struct pmu_events_table *etable =pmu_events_table__find (); + +#define RETURN_IF_NON_ZERO (x) do { if (x) return x; } while (0) +hashmap__for_each_entry (ctx->ids, cur, bkt) { +const char *id = cur->pkey; + +pr_debug ("found event %s\n" , id); + + ret =get_metricgroup_events (id, etable, &event_info_list); +if (ret) +return ret; + } -pr_debug ("This is a placeholder\n" );return ret; +#undef RETURN_IF_NON_ZERO }static void group_str_free (struct metricgroup__group_strs *g)@@ -1521,8 +1713,6 @@ static int hw_aware_parse_ids(struct perf_pmu
*fake_pmu,Quoted 53 lines
tools/perf/util/metricgroup.c +0
-2*out_evlist = parsed_evlist; parsed_evlist =NULL ; err_out: -parse_events_error__exit (&parse_error); -evlist__delete (parsed_evlist);metricgroup__free_grouping_strs (&groupings);return ret; }tools/perf/util/metricgroup.h +27
-0@@ -5,6 +5,7 @@#include <linux/list.h> #include <linux/rbtree.h> #include <stdbool.h> +#include <linux/bitmap.h> #include "pmu-events/pmu-events.h" #include "strbuf.h" @@ -67,6 +68,32 @@ struct metric_expr {int runtime; }; +/* Maximum number of counters per PMU*/ +#define NR_COUNTERS 16 +/* +* Transfer bit position in the bitmap to ensure start assigning counter from +* the last GP counter to the first. +* bit15 <---> bit0 +* [GP8-GP15] [GP0-GP7] +*/ +#define TRANSFER_FIRST_BYTE (pos) (7 - pos) +#define TRANSFER_SEC_BYTE (pos) (23 - pos) + +/** +* An event used in a metric. This info is for metric grouping. +*/ +struct metricgroup__event_info { +struct list_head nd; +/** The name of the event. */ +const char *name; +/** The name of the pmu the event be collected on. */ +const char *pmu_name; +bool fixed_counter; +bool free_counter; +/** The counters the event allowed to be collected on. */ + DECLARE_BITMAP(counters, NR_COUNTERS); +}; +/** * Each group is one node in the group string list. */
RFC PATCH v2 8/17 perf stat: Add functions to get counter info
In reply to [email protected]From: Weilin Wang <[email protected]>
Add data structure metricgroup__pmu_counters to represent hardware counters available in the system.
Add functions to parse pmu-events and create the list of pmu_info_list to hold the counter information of the system.
Add functions to free pmu_info_list and event_info_list before exit grouping for hardware-grouping method
This method would fall back to normal grouping when event json files do not support hardware aware grouping.
--- tools/perf/util/metricgroup.c | 85 +++++++++++++++++++++++++++++++++-- tools/perf/util/metricgroup.h | 15 +++++++ tools/perf/util/pmu.c | 5 +++ tools/perf/util/pmu.h | 1 + 4 files changed, 103 insertions(+), 3 deletions(-)
tools/perf/util/metricgroup.c +82 -3
@@ -1507,6 +1507,27 @@ static int parse_counter(const char *counter,return 0 ; } +static void metricgroup__free_event_info (struct list_head + *event_info_list) +{ +struct metricgroup__event_info *e, *tmp; + +list_for_each_entry_safe (e, tmp, event_info_list, nd) { +list_del_init (&e->nd); +free (e); + } +} + +static void metricgroup__free_pmu_info (struct list_head *pmu_info_list) +{ +struct metricgroup__pmu_counters *p, *tmp; + +list_for_each_entry_safe (p, tmp, pmu_info_list, nd) { +list_del_init (&p->nd); +free (p); + } +} + static struct metricgroup__event_info *event_info__new(const char *name, const char *pmu_name, const char *counter,
@@ -1525,7 +1546,7 @@ static struct metricgroup__event_info *event_info__new(const char *name, e->name = name; e->free_counter = free_counter; - e->pmu_name =strdup (pmu_name); + e->pmu_name = pmu_name;if (free_counter) { ret =set_counter_bitmap (0 , e->counters);if (ret)
@@ -1560,6 +1581,8 @@ static int metricgroup__add_metric_event_callback(const struct pmu_event *pe,struct metricgroup__add_metric_event_data *d = data;if (!strcasecmp (pe->name, d->event_name)) { +if (!pe->counter) +return -EINVAL ; event =event_info__new (d->event_id, pe->pmu, pe->counter,/*free_counter=*/ false);if (!event)return -ENOMEM ;
@@ -1599,7 +1622,7 @@ static int get_metricgroup_events(const char *full_id, .event_name = id, .event_id = full_id, }; - ret =pmu_events_table_for_each_event (table, + ret =pmu_events_table__for_each_event (table,/*pmu=*/ NULL , metricgroup__add_metric_event_callback, &data); }
@@ -1608,6 +1631,57 @@ static int get_metricgroup_events(const char *full_id,return ret; } +static struct metricgroup__pmu_counters *pmu_layout__new (const struct pmu_layout *pl) +{ +struct metricgroup__pmu_counters *l; + + l =zalloc (sizeof (*l)); + +if (!l) +return NULL ; + + l->name = pl->pmu; + l->size = pl->size; + l->fixed_size = pl->fixed_size; +pr_debug ("create new pmu_layout: [pmu]=%s, [gp_size]=%ld, [fixed_size]=%ld\n" , + l->name, l->size, l->fixed_size); +return l; +} + +static int metricgroup__add_pmu_layout_callback (const struct pmu_layout *pl, +void *data) +{ +struct metricgroup__pmu_counters *pmu; +struct list_head *d = data; +int ret =0 ; + + pmu =pmu_layout__new (pl); +if (!pmu) +return -ENOMEM ; +list_add (&pmu->nd, d); +return ret; +} + +/** +* get_pmu_counter_layouts - Find counter info of the architecture from +* the pmu_layouts table +* @pmu_info_list: the list that the new counter info of a pmu is added to. +* @table: pmu_layouts table that is searched for counter info. +*/ +static int get_pmu_counter_layouts (struct list_head *pmu_info_list, +const struct pmu_layouts_table + *table) +{ +LIST_HEAD (list); +int ret; + + ret =pmu_layouts_table__for_each_layout (table, + metricgroup__add_pmu_layout_callback, &list); + +list_splice (&list, pmu_info_list); +return ret; +} +/** * hw_aware_build_grouping - Build event groupings by reading counter * requirement of the events and counter available on the system from
@@ -1626,6 +1700,7 @@ static int hw_aware_build_grouping(struct expr_parse_ctx *ctx __maybe_unused,LIST_HEAD (event_info_list);size_t bkt;const struct pmu_events_table *etable =pmu_events_table__find (); +const struct pmu_layouts_table *ltable =pmu_layouts_table__find ();#define RETURN_IF_NON_ZERO (x) do { if (x) return x; } while (0)hashmap__for_each_entry (ctx->ids, cur, bkt) {
@@ -1635,9 +1710,13 @@ static int hw_aware_build_grouping(struct expr_parse_ctx *ctx __maybe_unused, ret =get_metricgroup_events (id, etable, &event_info_list);if (ret) -return ret; + goto err_out; } + ret =get_pmu_counter_layouts (&pmu_info_list, ltable); +err_out: +metricgroup__free_event_info (&event_info_list); +metricgroup__free_pmu_info (&pmu_info_list);return ret;#undef RETURN_IF_NON_ZERO }
tools/perf/util/metricgroup.h +15 -0
@@ -94,6 +94,21 @@ struct metricgroup__event_info {DECLARE_BITMAP (counters,NR_COUNTERS ); }; +/** +* A node is the counter availability of a pmu. +* This info is built up at the beginning from JSON file and +* used as a reference in metric grouping process. +*/ +struct metricgroup__pmu_counters { +struct list_head nd; +/** The name of the pmu the event collected on. */ +const char *name; +//DECLARE_BITMAP(counter_bits, NR_COUNTERS); +/** The number of gp counters in the pmu. */ +size_t size; +size_t fixed_size; +}; +/** * Each group is one node in the group string list. */
tools/perf/util/pmu.c +5 -0
@@ -813,6 +813,11 @@ __weak const struct pmu_metrics_table *pmu_metrics_table__find(void)return perf_pmu__find_metrics_table (NULL ); } +__weak const struct pmu_layouts_table *pmu_layouts_table__find (void ) +{ +return perf_pmu__find_layouts_table (NULL ); +} +/** * perf_pmu__match_ignoring_suffix - Does the pmu_name match tok ignoring any * trailing suffix? The Suffix must be in form
tools/perf/util/pmu.h +1 -0
@@ -240,6 +240,7 @@ void pmu_add_cpu_aliases_table(struct perf_pmu *pmu,char *perf_pmu__getcpuid (struct perf_pmu *pmu);const struct pmu_events_table *pmu_events_table__find (void );const struct pmu_metrics_table *pmu_metrics_table__find (void ); +const struct pmu_layouts_table *pmu_layouts_table__find (void );int perf_pmu__convert_scale (const char *scale,char **end,double *sval);
Signature · 1 line
2.39.3
RFC PATCH v2 9/17 perf stat: Add functions to create new group and assign events into groups for hardware-grouping method
In reply to [email protected]From: Weilin Wang <[email protected]>
Add struct metricgroup__pmu_group_list to hold the lists of groups from different PMUs. Each PMU has one separate list.
Add struct metricgroup__group as one node (one group in the grouping result) of the metricgroup__pmu_group_list. It uses two bitmaps to log counter availabilities(gp counters and fixed counters).
Add functions to create group and assign event into the groups based on the event restrictions (struct metricgroup__event_info) and counter availability (pmu_info_list and bitmaps). New group is inserted into the list of groups.
Add functions to handle counter bitmaps. Add functions do find and insert operations to handle inserting event into groups.
Add function to fill all bits of one counter bitmap. Add functions to create new groups when no counter is available in all the existing groups.
--- tools/lib/bitmap.c | 20 +++ tools/perf/util/metricgroup.c | 254 ++++++++++++++++++++++++++++++++++ tools/perf/util/metricgroup.h | 37 +++++ 3 files changed, 311 insertions(+)
tools/lib/bitmap.c +20 -0
@@ -100,3 +100,23 @@ bool __bitmap_intersects(const unsigned long *bitmap1,return true;return false; } + +void bitmap_clear (unsigned long *map,unsigned int start,int len) +{ +unsigned long *p = map +BIT_WORD (start); +const unsigned int size = start + len; +int bits_to_clear =BITS_PER_LONG - (start %BITS_PER_LONG ); +unsigned long mask_to_clear =BITMAP_FIRST_WORD_MASK (start); + +while (len - bits_to_clear >=0 ) { + *p &= ~mask_to_clear; + len -= bits_to_clear; + bits_to_clear =BITS_PER_LONG ; + mask_to_clear = ~0UL ; + p++; + } +if (len) { + mask_to_clear &=BITMAP_LAST_WORD_MASK (size); + *p &= ~mask_to_clear; + } +}
tools/perf/util/metricgroup.c +254 -0
278 lines not on this page · raw message
tools/perf/util/metricgroup.h +37 -0
@@ -109,6 +109,43 @@ struct metricgroup__pmu_counters {size_t fixed_size; }; +/** +* A list of groups for this pmu. +* This is updated during the grouping. +*/ +struct metricgroup__pmu_group_list { +struct list_head nd; +/** The name of the pmu(/core) the events collected on. */ +const char *pmu_name; +/** The number of gp counters in the pmu(/core). */ +size_t size; +/** The number of fixed counters in the pmu(/core) if applicable. */ +size_t fixed_size; +/** Head to the list of groups using this pmu(/core)*/ +struct list_head group_head; +}; + +/** +* This is one node in the metricgroup__pmu_group_list. +* It represents on group. +*/ +struct metricgroup__group { +struct list_head nd; +/** The bitmaps represent availability of the counters. +* They are updated once the corresponding counter is used by +* an event (event inserted into the group). +*/ + DECLARE_BITMAP(gp_counters, NR_COUNTERS); + DECLARE_BITMAP(fixed_counters, NR_COUNTERS); +/** Head to the list of event names in this group*/ +struct list_head event_head; +}; + +struct metricgroup__group_events { +struct list_head nd; +const char *event_name; +}; +/** * Each group is one node in the group string list. */
Signature · 1 line
2.39.3
RFC PATCH v2 10/17 perf stat: Add build string function and topdown events handling in hardware-grouping
In reply to [email protected]From: Weilin Wang <[email protected]>
Add the function to generate final grouping strings. This function is very similar to the existing metricgroup__build_event_string() function. The difference is that the input data includes a list of grouping lists.
--- tools/perf/util/metricgroup.c | 96 +++++++++++++++++++++++++++++++++-- 1 file changed, 93 insertions(+), 3 deletions(-)
tools/perf/util/metricgroup.c +93 -3
@@ -1899,6 +1899,96 @@ static int assign_event_grouping(struct metricgroup__event_info *e,return ret; } +static int hw_aware_metricgroup__build_event_string (struct list_head *group_strs, +const char *modifier, +struct list_head *groups) +{ +struct metricgroup__pmu_group_list *p; +struct metricgroup__group *g; +struct metricgroup__group_events *ge; +bool no_group = true; +int ret =0 ; + +#define RETURN_IF_NON_ZERO (x) do { if (x) return x; } while (0) + +list_for_each_entry (p, groups, nd) { +list_for_each_entry (g, &p->group_head, nd) { +struct strbuf *events; +struct metricgroup__group_strs *new_group_str = +malloc (sizeof (struct metricgroup__group_strs )); + +if (!new_group_str) +return -ENOMEM ; +strbuf_init (&new_group_str->grouping_str,0 ); + events = &new_group_str->grouping_str; + ret =strbuf_addch (events,'{' ); +RETURN_IF_NON_ZERO (ret); + no_group = true; +list_for_each_entry (ge, &g->event_head, nd) { +const char *sep, *rsep, *id = ge->event_name; + +pr_debug ("found event %s\n" , id); + +/* Separate events with commas and open the group if necessary. */ +if (!no_group) { + ret =strbuf_addch (events,',' ); +RETURN_IF_NON_ZERO (ret); + } +/* +* Encode the ID as an event string. Add a qualifier for +* metric_id that is the original name except with characters +* that parse-events can't parse replaced. For example, +* 'msr@tsc@' gets added as msr/tsc,metric-id=msr!3tsc!3/ +*/ + sep =strchr (id,'@' ); +if (sep) { + ret =strbuf_add (events, id, sep - id); +RETURN_IF_NON_ZERO (ret); + ret =strbuf_addch (events,'/' ); +RETURN_IF_NON_ZERO (ret); + rsep =strrchr (sep,'@' ); + ret =strbuf_add (events, sep +1 , rsep - sep -1 ); +RETURN_IF_NON_ZERO (ret); + ret =strbuf_addstr (events,",metric-id=" ); +RETURN_IF_NON_ZERO (ret); + sep = rsep; + }else { + sep =strchr (id,':' ); +if (sep) { + ret =strbuf_add (events, id, sep - id); +RETURN_IF_NON_ZERO (ret); + }else { + ret =strbuf_addstr (events, id); +RETURN_IF_NON_ZERO (ret); + } + ret =strbuf_addstr (events,"/metric-id=" ); +RETURN_IF_NON_ZERO (ret); + } + ret =encode_metric_id (events, id); +RETURN_IF_NON_ZERO (ret); + ret =strbuf_addstr (events,"/" ); +RETURN_IF_NON_ZERO (ret); + +if (sep) { + ret =strbuf_addstr (events, sep +1 ); +RETURN_IF_NON_ZERO (ret); + } +if (modifier) { + ret =strbuf_addstr (events, modifier); +RETURN_IF_NON_ZERO (ret); + } + no_group = false; + } + ret =strbuf_addf (events,"}:W" ); +RETURN_IF_NON_ZERO (ret); +pr_debug ("events-buf: %s\n" , events->buf); +list_add_tail (&new_group_str->nd, group_strs); + } + } +return ret; +#undef RETURN_IF_NON_ZERO +} +/** * create_grouping - Create a list of groups and place all the events of * event_info_list into these groups.
@@ -1910,8 +2000,8 @@ static int assign_event_grouping(struct metricgroup__event_info *e,*/ static int create_grouping (struct list_head *pmu_info_list,struct list_head *event_info_list, -struct list_head *groupings__maybe_unused , -const char *modifier__maybe_unused ) +struct list_head *groupings, +const char *modifier) {int ret =0 ;struct metricgroup__event_info *e;
@@ -1927,6 +2017,7 @@ static int create_grouping(struct list_head *pmu_info_list,if (ret) goto out; } + ret =hw_aware_metricgroup__build_event_string (groupings, modifier, &groups); out:metricgroup__free_group_list (&groups);return ret;
@@ -1957,7 +2048,6 @@ static int hw_aware_build_grouping(struct expr_parse_ctx *ctx __maybe_unused,const char *id = cur->pkey;pr_debug ("found event %s\n" , id); - ret =get_metricgroup_events (id, etable, &event_info_list);if (ret) goto err_out;
Signature · 1 line
2.39.3
RFC PATCH v2 11/17 perf stat: Add function to handle special events in hardware-grouping
In reply to [email protected]From: Weilin Wang <[email protected]>
There are some special events like topdown events and TSC that are not decribed in pmu-event JSON files. Add support to handle this type of events. This should be considered as a temporary solution because including these events in JSON files would be a better solution.
--- tools/perf/util/metricgroup.c | 25 +++++++++++++++++++++++-- tools/perf/util/metricgroup.h | 15 +++++++++++++++ 2 files changed, 38 insertions(+), 2 deletions(-)
tools/perf/util/metricgroup.c +23 -2
@@ -2023,6 +2023,15 @@ static int create_grouping(struct list_head *pmu_info_list,return ret; }; +static bool is_special_event (const char *id) +{ +for (int i =0 ; i <SPECIAL_EVENT_MAX ; i++) { +if (!strncmp (id, special_event_names[i],strlen (special_event_names[i]))) +return true; + } +return false; +} +/** * hw_aware_build_grouping - Build event groupings by reading counter * requirement of the events and counter available on the system from
@@ -2047,7 +2056,17 @@ static int hw_aware_build_grouping(struct expr_parse_ctx *ctx __maybe_unused,hashmap__for_each_entry (ctx->ids, cur, bkt) {const char *id = cur->pkey; -pr_debug ("found event %s\n" , id); +if (is_special_event (id)) { +struct metricgroup__event_info *event; + + event =event_info__new (id,"default_core" ,"0" , +/*free_counter=*/ true); +if (!event) + goto err_out; + +list_add (&event->nd, &event_info_list); +continue ; + } ret =get_metricgroup_events (id, etable, &event_info_list);if (ret) goto err_out;
@@ -2519,8 +2538,10 @@ int metricgroup__parse_groups(struct evlist *perf_evlist, ret =hw_aware_parse_groups (perf_evlist, pmu, str, metric_no_threshold, user_requested_cpu_list, system_wide,/*fake_pmu=*/ NULL , metric_events, table); -if (!ret) +if (!ret) { +pr_info ("Hardware aware grouping completed\n" );return 0 ; + } } return parse_groups(perf_evlist, pmu, str, metric_no_group, metric_no_merge,
tools/perf/util/metricgroup.h +15 -0
@@ -79,6 +79,21 @@ struct metric_expr {#define TRANSFER_FIRST_BYTE (pos) (7 - pos)#define TRANSFER_SEC_BYTE (pos) (23 - pos) +/* Special events that are not described in pmu-event JSON files. +* topdown-* and TSC use dedicated registers, set as free +* counter for grouping purpose +*/ +enum special_events { +TOPDOWN =0 , +TSC =1 , +SPECIAL_EVENT_MAX , +}; + +static const char *const special_event_names[SPECIAL_EVENT_MAX ] = { +"topdown-" , +"TSC" , +}; +/** * An event used in a metric. This info is for metric grouping. */
Signature · 1 line
2.39.3
RFC PATCH v2 12/17 perf stat: Add function to combine metrics for hardware-grouping
In reply to [email protected]From: Weilin Wang <[email protected]>
This function is very similar to the existing build_combined_expr_ctx(). Should be able to reuse current function instead of adding a new one. Will fix this later.
--- tools/perf/util/metricgroup.c | 51 +++++++++++++++++++++++++++++++---- 1 file changed, 46 insertions(+), 5 deletions(-)
tools/perf/util/metricgroup.c +46 -5
@@ -1432,10 +1432,50 @@ static int build_combined_expr_ctx(const struct list_head *metric_list,return ret; } +/** +* hw_aware_build_combined_expr_ctx - Make an expr_parse_ctx with all !group_events +* metric IDs, as the IDs are held in a set, +* duplicates will be removed. +* @metric_list: List to take metrics from. +* @combined: Out argument for result. +*/ +static int hw_aware_build_combined_expr_ctx (const struct list_head *metric_list, +struct expr_parse_ctx **combined) +{ +struct hashmap_entry *cur; +size_t bkt; +struct metric *m; +char *dup; +int ret; + + *combined =expr__ctx_new (); +if (!*combined) +return -ENOMEM ; + +list_for_each_entry (m, metric_list, nd) { +hashmap__for_each_entry (m->pctx->ids, cur, bkt) { +pr_debug2 ("metric: %s\n" , m->metric_name); + dup =strdup (cur->pkey); +if (!dup) { + ret = -ENOMEM ; + goto err_out; + } + ret =expr__add_id (*combined, dup); +if (ret) + goto err_out; + } + } +return 0 ; +err_out: +expr__ctx_free (*combined); + *combined =NULL ; +return ret; +} +/** * set_counter_bitmap - The counter bit mapping: [8-15,0-7], e.g. the GP0 is the -* 8th bit and GP7 is the 1st bit in this 16-bits bitmap. It is helpful for -* assigning GP4-7 before GP0-3 because some events can be collected using GP0-3 +* 8th bit and GP7 is the 1st bit in this 16-bits bitmap. The is helpful to +* assign GP4-7 before GP0-3 because some events can be collected using GP0-3 * only on some platforms. */ static int set_counter_bitmap (int pos,unsigned long *bitmap)
@@ -1722,8 +1762,7 @@ static int metricgroup__add_pmu_layout_callback(const struct pmu_layout *pl,* @table: pmu_layouts table that is searched for counter info. */ static int get_pmu_counter_layouts (struct list_head *pmu_info_list, -const struct pmu_layouts_table - *table) +const struct pmu_layouts_table *table) {LIST_HEAD (list);int ret;
@@ -2155,6 +2194,8 @@ static int hw_aware_parse_ids(struct perf_pmu *fake_pmu, *out_evlist = parsed_evlist; parsed_evlist =NULL ; err_out: +parse_events_error__exit (&parse_error); +evlist__delete (parsed_evlist);metricgroup__free_grouping_strs (&groupings);return ret; }
@@ -2277,7 +2318,7 @@ static int hw_aware_parse_groups(struct evlist *perf_evlist,if (!metric_no_merge) {struct expr_parse_ctx *combined =NULL ; - ret =build_combined_expr_ctx (&metric_list, &combined); + ret =hw_aware_build_combined_expr_ctx (&metric_list, &combined);if (!ret && combined &&hashmap__size (combined->ids)) { ret = hw_aware_parse_ids(fake_pmu, combined,
Signature · 1 line
2.39.3
RFC PATCH v2 13/17 perf stat: Handle taken alone in hardware-grouping
In reply to [email protected]From: Weilin Wang <[email protected]>
Add taken alone into consideration when grouping. Only one taken alone event is supported per group.
--- tools/perf/pmu-events/jevents.py | 7 +++++-- tools/perf/pmu-events/pmu-events.h | 1 + tools/perf/util/metricgroup.c | 18 +++++++++++++----- tools/perf/util/metricgroup.h | 2 ++ 4 files changed, 21 insertions(+), 7 deletions(-)
tools/perf/pmu-events/jevents.py +5 -2
@@ -55,7 +55,9 @@ _json_event_attributes = [ # Counter this event could use 'counter', # Longer things (the last won't be iterated over during decompress). - 'long_desc' + 'long_desc', + # Taken alone event could not be collected in the same group with other taken alone event + 'taken_alone' ] # Attributes that are in pmu_unit_layout.
@@ -71,7 +73,7 @@ _json_metric_attributes = [ ] # Attributes that are bools or enum int values, encoded as '0', '1',... _json_enum_attributes = ['aggr_mode', 'deprecated', 'event_grouping', 'perpkg', - 'size', 'fixed_size' + 'size', 'fixed_size', 'taken_alone' ] def removesuffix(s: str, suffix: str) -> str:
@@ -330,6 +332,7 @@ class JsonEvent: extra_desc += ' Spec update: ' + jd['Errata'] self.pmu = unit_to_pmu(jd.get('Unit')) self.counter = jd.get('Counter') + self.taken_alone = jd.get('TakenAlone') self.size = jd.get('Size') self.fixed_size = jd.get('FixedSize') filter = jd.get('Filter')
tools/perf/pmu-events/pmu-events.h +1 -0
@@ -50,6 +50,7 @@ struct pmu_event {const char *unit;bool perpkg;bool deprecated; +bool taken_alone; };struct pmu_metric {
tools/perf/util/metricgroup.c +13 -5
@@ -1624,6 +1624,7 @@ static void metricgroup__free_pmu_info(struct list_head *pmu_info_list)static struct metricgroup__event_info *event_info__new (const char *name,const char *pmu_name,const char *counter, +bool taken_alone,bool free_counter) {int ret =0 ;
@@ -1638,6 +1639,7 @@ static struct metricgroup__event_info *event_info__new(const char *name, pmu_name ="core" ; e->name = name; + e->taken_alone = taken_alone; e->free_counter = free_counter; e->pmu_name = pmu_name;if (free_counter) {
@@ -1676,7 +1678,8 @@ static int metricgroup__add_metric_event_callback(const struct pmu_event *pe,if (!strcasecmp (pe->name, d->event_name)) {if (!pe->counter)return -EINVAL ; - event =event_info__new (d->event_id, pe->pmu, pe->counter,/*free_counter=*/ false); + event =event_info__new (d->event_id, pe->pmu, pe->counter, + pe->taken_alone,/*free_counter=*/ false);if (!event)return -ENOMEM ;list_add (&event->nd, d->list);
@@ -1799,6 +1802,8 @@ static int find_and_set_counters(struct metricgroup__event_info *e,int ret;unsigned long find_bit =0 ; +if (e->taken_alone && current_group->taken_alone) +return -ENOSPC ;if (e->free_counter)return 0 ;if (e->fixed_counter) {
@@ -1833,11 +1838,13 @@ static int _insert_event(struct metricgroup__event_info *e,list_add (&event->nd, &group->event_head); else list_add_tail(&event->nd, &group->event_head); +if (e->taken_alone) + group->taken_alone = true;return 0 ; }/** -* Insert the new_group node at the end of the group list. +* Initialize the new group and insert it to the end of the group list. */ static int insert_new_group(struct list_head *head, struct metricgroup__group *new_group,
@@ -1847,6 +1854,7 @@ static int insert_new_group(struct list_head *head,INIT_LIST_HEAD (&new_group->event_head);fill_counter_bitmap (new_group->gp_counters,0 , size);fill_counter_bitmap (new_group->fixed_counters,0 , fixed_size); + new_group->taken_alone = false;list_add_tail (&new_group->nd, head);return 0 ; }
@@ -2050,8 +2058,8 @@ static int create_grouping(struct list_head *pmu_info_list,//TODO: for each new core group, we should consider to add events that uses fixed counters list_for_each_entry (e, event_info_list, nd) {bitmap_scnprintf (e->counters,NR_COUNTERS , bit_buf,NR_COUNTERS ); -pr_debug ("Event name %s, [pmu]=%s, [counters]=%s\n" , e->name, - e->pmu_name, bit_buf); +pr_debug ("Event name %s, [pmu]=%s, [counters]=%s, [taken_alone]=%d\n" , + e->name, e->pmu_name, bit_buf, e->taken_alone); ret =assign_event_grouping (e, pmu_info_list, &groups);if (ret) goto out;
@@ -2098,7 +2106,7 @@ static int hw_aware_build_grouping(struct expr_parse_ctx *ctx __maybe_unused,if (is_special_event (id)) {struct metricgroup__event_info *event; - event =event_info__new (id,"default_core" ,"0" , + event =event_info__new (id,"default_core" ,"0" , false,/*free_counter=*/ true);if (!event) goto err_out;
tools/perf/util/metricgroup.h +2 -0
@@ -105,6 +105,7 @@ struct metricgroup__event_info {const char *pmu_name;bool fixed_counter;bool free_counter; +bool taken_alone;/** The counters the event allowed to be collected on. */ DECLARE_BITMAP (counters,NR_COUNTERS ); };
@@ -154,6 +155,7 @@ struct metricgroup__group {DECLARE_BITMAP (fixed_counters,NR_COUNTERS );/** Head to the list of event names in this group*/ struct list_head event_head; +bool taken_alone; };struct metricgroup__group_events {
Signature · 1 line
2.39.3
RFC PATCH v2 14/17 perf stat: Handle NMI in hardware-grouping
In reply to [email protected]From: Weilin Wang <[email protected]>
Add an easy nmi watchdog support in grouping. When nmi watchdog is enabled, we reduce the total num of events could be assigned to one group by 1. A more efficient solution will be added in later.
--- tools/perf/util/metricgroup.c | 4 ++++ 1 file changed, 4 insertions(+)
tools/perf/util/metricgroup.c +4 -0
@@ -1852,6 +1852,10 @@ static int insert_new_group(struct list_head *head, size_t fixed_size) {INIT_LIST_HEAD (&new_group->event_head); +if (sysctl__nmi_watchdog_enabled ()) { +pr_debug ("NMI watchdog is enabled\n" ); + size -=1 ; + }fill_counter_bitmap (new_group->gp_counters,0 , size);fill_counter_bitmap (new_group->fixed_counters,0 , fixed_size); new_group->taken_alone = false;
Signature · 1 line
2.39.3
RFC PATCH v2 15/17 perf stat: Code refactoring in hardware-grouping
In reply to [email protected]From: Weilin Wang <[email protected]>
Decouple the step to generate final grouping strings out from the build_grouping step so that we could do single metric grouping and then merge groups if needed later.
--- tools/perf/util/metricgroup.c | 50 +++++++++++++++++------------------ 1 file changed, 25 insertions(+), 25 deletions(-)
tools/perf/util/metricgroup.c +25 -25
@@ -1801,9 +1801,10 @@ static int find_and_set_counters(struct metricgroup__event_info *e, {int ret;unsigned long find_bit =0 ; - -if (e->taken_alone && current_group->taken_alone) +if (e->taken_alone && current_group->taken_alone) { +pr_debug ("current group with taken alone event already\n" );return -ENOSPC ; + }if (e->free_counter)return 0 ;if (e->fixed_counter) {
@@ -1922,7 +1923,8 @@ static int assign_event_grouping(struct metricgroup__event_info *e,list_for_each_entry (g, groups, nd) {if (!strcasecmp (g->pmu_name, e->pmu_name)) { -pr_debug ("found group for event %s in pmu %s\n" , e->name, g->pmu_name); +pr_debug ("found group header for event %s in pmu %s\n" , + e->name, g->pmu_name); pmu_group_head = g;break ; }
@@ -2051,26 +2053,22 @@ static int hw_aware_metricgroup__build_event_string(struct list_head *group_strs*/ static int create_grouping (struct list_head *pmu_info_list,struct list_head *event_info_list, -struct list_head *groupings, -const char *modifier) +struct list_head *grouping) {int ret =0 ;struct metricgroup__event_info *e; -LIST_HEAD (groups);char *bit_buf =malloc (NR_COUNTERS ); -//TODO: for each new core group, we should consider to add events that uses fixed counters +//TODO: for each new core group, we could consider to add events that +//uses fixed counters list_for_each_entry (e, event_info_list, nd) {bitmap_scnprintf (e->counters,NR_COUNTERS , bit_buf,NR_COUNTERS );pr_debug ("Event name %s, [pmu]=%s, [counters]=%s, [taken_alone]=%d\n" , e->name, e->pmu_name, bit_buf, e->taken_alone); - ret =assign_event_grouping (e, pmu_info_list, &groups); + ret =assign_event_grouping (e, pmu_info_list, grouping);if (ret) - goto out; +return ret; } - ret =hw_aware_metricgroup__build_event_string (groupings, modifier, &groups); -out: -metricgroup__free_group_list (&groups);return ret; };
@@ -2091,9 +2089,8 @@ static bool is_special_event(const char *id)* @groupings: header to the list of final event grouping. * @modifier: any modifiers added to the events. */ -static int hw_aware_build_grouping (struct expr_parse_ctx *ctx__maybe_unused , -struct list_head *groupings__maybe_unused , -const char *modifier__maybe_unused ) +static int hw_aware_build_grouping (struct expr_parse_ctx *ctx, +struct list_head *grouping) {int ret =0 ;struct hashmap_entry *cur;
@@ -2125,8 +2122,7 @@ static int hw_aware_build_grouping(struct expr_parse_ctx *ctx __maybe_unused, ret =get_pmu_counter_layouts (&pmu_info_list, ltable);if (ret) goto err_out; - ret =create_grouping (&pmu_info_list, &event_info_list, groupings, - modifier); + ret =create_grouping (&pmu_info_list, &event_info_list, grouping); err_out:metricgroup__free_event_info (&event_info_list);
@@ -2172,23 +2168,25 @@ static int hw_aware_parse_ids(struct perf_pmu *fake_pmu, {struct parse_events_error parse_error;struct evlist *parsed_evlist; -LIST_HEAD (groupings); +LIST_HEAD (grouping_str); +LIST_HEAD (grouping);struct metricgroup__group_strs *group;int ret; *out_evlist =NULL ; - ret =hw_aware_build_grouping (ids, &groupings, modifier); -if (ret) { -metricgroup__free_grouping_strs (&groupings); -return ret; - } + ret =hw_aware_build_grouping (ids, &grouping); +if (ret) + goto out; + ret =hw_aware_metricgroup__build_event_string (&grouping_str, modifier, &grouping); +if (ret) + goto out; parsed_evlist =evlist__new ();if (!parsed_evlist) { ret = -ENOMEM ; goto err_out; } -list_for_each_entry (group, &groupings, nd) { +list_for_each_entry (group, &grouping_str, nd) {struct strbuf *events = &group->grouping_str;pr_debug ("Parsing metric events '%s'\n" , events->buf);
@@ -2208,7 +2206,9 @@ static int hw_aware_parse_ids(struct perf_pmu *fake_pmu, err_out:parse_events_error__exit (&parse_error);evlist__delete (parsed_evlist); -metricgroup__free_grouping_strs (&groupings); +out: +metricgroup__free_group_list (&grouping); +metricgroup__free_grouping_strs (&grouping_str);return ret; }
Signature · 1 line
2.39.3
RFC PATCH v2 16/17 perf stat: Add tool events support in hardware-grouping
In reply to [email protected]From: Weilin Wang <[email protected]>
Add tool events into default_core grouping strings if find tool events so that metrics use tool events could be correctly calculated. Need this step to support TopdownL4-L5.
--- tools/perf/util/metricgroup.c | 49 ++++++++++++++++++++++++++++++++++- 1 file changed, 48 insertions(+), 1 deletion(-)
tools/perf/util/metricgroup.c +48 -1
@@ -1391,6 +1391,35 @@ static void find_tool_events(const struct list_head *metric_list, } } +/** +* get_tool_event_str - Generate and return a string with all the used tool +* event names. +*/ +static int get_tool_event_str (struct strbuf *events, +const bool tool_events[PERF_TOOL_MAX ], +bool *has_tool_event) +{ +int i =0 ; +int ret; + +perf_tool_event__for_each_event (i) { +if (tool_events[i]) { +const char *tmp =strdup (perf_tool_event__to_str (i)); + +if (!tmp) +return -ENOMEM ; + *has_tool_event = true; + ret =strbuf_addstr (events,"," ); +if (ret) +return ret; + ret =strbuf_addstr (events, tmp); +if (ret) +return ret; + } + } +return 0 ; +} +/** * build_combined_expr_ctx - Make an expr_parse_ctx with all !group_events * metric IDs, as the IDs are held in a set,
@@ -1954,6 +1983,7 @@ static int assign_event_grouping(struct metricgroup__event_info *e,static int hw_aware_metricgroup__build_event_string (struct list_head *group_strs,const char *modifier, +const bool tool_events[PERF_TOOL_MAX ],struct list_head *groups) {struct metricgroup__pmu_group_list *p;
@@ -1961,8 +1991,12 @@ static int hw_aware_metricgroup__build_event_string(struct list_head *group_strsstruct metricgroup__group_events *ge;bool no_group = true;int ret =0 ; +struct strbuf tool_event_str =STRBUF_INIT ; +bool has_tool_event = false;#define RETURN_IF_NON_ZERO (x) do { if (x) return x; } while (0) + ret =get_tool_event_str (&tool_event_str, tool_events, &has_tool_event); +RETURN_IF_NON_ZERO (ret);list_for_each_entry (p, groups, nd) {list_for_each_entry (g, &p->group_head, nd) {
@@ -2034,6 +2068,12 @@ static int hw_aware_metricgroup__build_event_string(struct list_head *group_strs } ret =strbuf_addf (events,"}:W" );RETURN_IF_NON_ZERO (ret); + +if (!strcmp (p->pmu_name,"default_core" ) && has_tool_event) { + ret =strbuf_addstr (events, tool_event_str.buf); +RETURN_IF_NON_ZERO (ret); + } +pr_debug ("events-buf: %s\n" , events->buf);list_add_tail (&new_group_str->nd, group_strs); }
@@ -2119,6 +2159,7 @@ static int hw_aware_build_grouping(struct expr_parse_ctx *ctx,if (ret) goto err_out; } + ret =get_pmu_counter_layouts (&pmu_info_list, ltable);if (ret) goto err_out;
@@ -2164,6 +2205,7 @@ static void metricgroup__free_grouping_strs(struct list_head*/ static int hw_aware_parse_ids (struct perf_pmu *fake_pmu,struct expr_parse_ctx *ids,const char *modifier, +const bool tool_events[PERF_TOOL_MAX ],struct evlist **out_evlist) {struct parse_events_error parse_error;
@@ -2177,7 +2219,8 @@ static int hw_aware_parse_ids(struct perf_pmu *fake_pmu, ret =hw_aware_build_grouping (ids, &grouping);if (ret) goto out; - ret =hw_aware_metricgroup__build_event_string (&grouping_str, modifier, &grouping); + ret =hw_aware_metricgroup__build_event_string (&grouping_str, modifier, + tool_events, &grouping);if (ret) goto out;
@@ -2312,6 +2355,7 @@ static int hw_aware_parse_groups(struct evlist *perf_evlist,struct evlist *combined_evlist =NULL ;LIST_HEAD (metric_list);struct metric *m; +bool tool_events[PERF_TOOL_MAX ] = {false};int ret;bool metric_no_group = false;bool metric_no_merge = false;
@@ -2330,11 +2374,14 @@ static int hw_aware_parse_groups(struct evlist *perf_evlist,if (!metric_no_merge) {struct expr_parse_ctx *combined =NULL ; +find_tool_events (&metric_list, tool_events); + ret =hw_aware_build_combined_expr_ctx (&metric_list, &combined);if (!ret && combined &&hashmap__size (combined->ids)) { ret =hw_aware_parse_ids (fake_pmu, combined,/*modifier=*/ NULL , + tool_events, &combined_evlist); }
Signature · 1 line
2.39.3
RFC PATCH v2 17/17 perf pmu-events: Add event counter data for Tigerlake
In reply to [email protected]From: Weilin Wang <[email protected]>
The added new fields are required in this new grouping method. Currently, the format of added fields are temporary for experiments.
This platform is not tested yet.
--- .../pmu-events/arch/x86/tigerlake/cache.json | 65 +++++++++++++ .../arch/x86/tigerlake/counter.json | 7 ++ .../arch/x86/tigerlake/floating-point.json | 13 +++ .../arch/x86/tigerlake/frontend.json | 56 +++++++++++ .../pmu-events/arch/x86/tigerlake/memory.json | 31 ++++++ .../pmu-events/arch/x86/tigerlake/other.json | 4 + .../arch/x86/tigerlake/pipeline.json | 96 +++++++++++++++++++ .../x86/tigerlake/uncore-interconnect.json | 11 +++ .../arch/x86/tigerlake/uncore-memory.json | 6 ++ .../arch/x86/tigerlake/uncore-other.json | 1 + .../arch/x86/tigerlake/virtual-memory.json | 20 ++++ 11 files changed, 310 insertions(+) create mode 100644 tools/perf/pmu-events/arch/x86/tigerlake/counter.json
tools/perf/pmu-events/arch/x86/tigerlake/cache.json +65 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "Counts the number of cache lines replaced in L1 data cache.", + "Counter": "0,1,2,3", "EventCode": "0x51", "EventName": "L1D.REPLACEMENT", "PublicDescription": "Counts L1D data line replacements including opportunistic replacements, and replacements that require stall-for-replace or block-for-replace.",
@@ -9,6 +10,7 @@ }, { "BriefDescription": "Number of cycles a demand request has waited due to L1D Fill Buffer (FB) unavailability.", + "Counter": "0,1,2,3", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.FB_FULL", "PublicDescription": "Counts number of cycles a demand request has waited due to L1D Fill Buffer (FB) unavailability. Demand requests include cacheable/uncacheable demand load, store, lock or SW prefetch accesses.",
@@ -17,6 +19,7 @@ }, { "BriefDescription": "Number of phases a demand request has waited due to L1D Fill Buffer (FB) unavailability.", + "Counter": "0,1,2,3", "CounterMask": "1", "EdgeDetect": "1", "EventCode": "0x48",
@@ -27,6 +30,7 @@ }, { "BriefDescription": "Number of cycles a demand request has waited due to L1D due to lack of L2 resources.", + "Counter": "0,1,2,3", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.L2_STALL", "PublicDescription": "Counts number of cycles a demand request has waited due to L1D due to lack of L2 resources. Demand requests include cacheable/uncacheable demand load, store, lock or SW prefetch accesses.",
@@ -35,6 +39,7 @@ }, { "BriefDescription": "Number of L1D misses that are outstanding", + "Counter": "0,1,2,3", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.PENDING", "PublicDescription": "Counts number of L1D misses that are outstanding in each cycle, that is each cycle the number of Fill Buffers (FB) outstanding required by Demand Reads. FB either is held by demand loads, or it is held by non-demand loads and gets hit at least once by demand. The valid outstanding interval is defined until the FB deallocation by one of the following ways: from FB allocation, if FB is allocated by demand from the demand Hit FB, if it is allocated by hardware or software prefetch. Note: In the L1D, a Demand Read contains cacheable or noncacheable demand loads, including ones causing cache-line splits and reads due to page walks resulted from any request type.",
@@ -43,6 +48,7 @@ }, { "BriefDescription": "Cycles with L1D load Misses outstanding.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x48", "EventName": "L1D_PEND_MISS.PENDING_CYCLES",
@@ -52,6 +58,7 @@ }, { "BriefDescription": "L2 cache lines filling L2", + "Counter": "0,1,2,3", "EventCode": "0xf1", "EventName": "L2_LINES_IN.ALL", "PublicDescription": "Counts the number of L2 cache lines filling the L2. Counting does not cover rejects.",
@@ -60,6 +67,7 @@ }, { "BriefDescription": "Modified cache lines that are evicted by L2 cache when triggered by an L2 cache fill.", + "Counter": "0,1,2,3", "EventCode": "0xf2", "EventName": "L2_LINES_OUT.NON_SILENT", "PublicDescription": "Counts the number of lines that are evicted by L2 cache when triggered by an L2 cache fill. Those lines are in Modified state. Modified lines are written back to L3",
@@ -68,6 +76,7 @@ }, { "BriefDescription": "Non-modified cache lines that are silently dropped by L2 cache when triggered by an L2 cache fill.", + "Counter": "0,1,2,3", "EventCode": "0xf2", "EventName": "L2_LINES_OUT.SILENT", "PublicDescription": "Counts the number of lines that are silently dropped by L2 cache when triggered by an L2 cache fill. These lines are typically in Shared or Exclusive state. A non-threaded event.",
@@ -76,6 +85,7 @@ }, { "BriefDescription": "L2 code requests", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_CODE_RD", "PublicDescription": "Counts the total number of L2 code requests.",
@@ -84,6 +94,7 @@ }, { "BriefDescription": "Demand Data Read access L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_DEMAND_DATA_RD", "PublicDescription": "Counts Demand Data Read requests accessing the L2 cache. These requests may hit or miss L2 cache. True-miss exclude misses that were merged with ongoing L2 misses. An access is counted once.",
@@ -92,6 +103,7 @@ }, { "BriefDescription": "RFO requests to L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.ALL_RFO", "PublicDescription": "Counts the total number of RFO (read for ownership) requests to L2 cache. L2 RFO requests include both L1D demand RFO misses as well as L1D RFO prefetches.",
@@ -100,6 +112,7 @@ }, { "BriefDescription": "L2 cache hits when fetching instructions, code reads.", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.CODE_RD_HIT", "PublicDescription": "Counts L2 cache hits when fetching instructions, code reads.",
@@ -108,6 +121,7 @@ }, { "BriefDescription": "L2 cache misses when fetching instructions", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.CODE_RD_MISS", "PublicDescription": "Counts L2 cache misses when fetching instructions.",
@@ -116,6 +130,7 @@ }, { "BriefDescription": "Demand Data Read requests that hit L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.DEMAND_DATA_RD_HIT", "PublicDescription": "Counts the number of demand Data Read requests initiated by load instructions that hit L2 cache.",
@@ -124,6 +139,7 @@ }, { "BriefDescription": "Demand Data Read miss L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.DEMAND_DATA_RD_MISS", "PublicDescription": "Counts demand Data Read requests with true-miss in the L2 cache. True-miss excludes misses that were merged with ongoing L2 misses. An access is counted once.",
@@ -132,6 +148,7 @@ }, { "BriefDescription": "Read requests with true-miss in L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.MISS", "PublicDescription": "Counts read requests of any type with true-miss in the L2 cache. True-miss excludes L2 misses that were merged with ongoing L2 misses.",
@@ -140,6 +157,7 @@ }, { "BriefDescription": "All accesses to L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.REFERENCES", "PublicDescription": "Counts all requests that were hit or true misses in L2 cache. True-miss excludes misses that were merged with ongoing L2 misses.",
@@ -148,6 +166,7 @@ }, { "BriefDescription": "RFO requests that hit L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.RFO_HIT", "PublicDescription": "Counts the RFO (Read-for-Ownership) requests that hit L2 cache.",
@@ -156,6 +175,7 @@ }, { "BriefDescription": "RFO requests that miss L2 cache", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.RFO_MISS", "PublicDescription": "Counts the RFO (Read-for-Ownership) requests that miss L2 cache.",
@@ -164,6 +184,7 @@ }, { "BriefDescription": "SW prefetch requests that hit L2 cache.", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.SWPF_HIT", "PublicDescription": "Counts Software prefetch requests that hit the L2 cache. Accounts for PREFETCHNTA and PREFETCHT0/1/2 instructions when FB is not full.",
@@ -172,6 +193,7 @@ }, { "BriefDescription": "SW prefetch requests that miss L2 cache.", + "Counter": "0,1,2,3", "EventCode": "0x24", "EventName": "L2_RQSTS.SWPF_MISS", "PublicDescription": "Counts Software prefetch requests that miss the L2 cache. Accounts for PREFETCHNTA and PREFETCHT0/1/2 instructions when FB is not full.",
@@ -180,6 +202,7 @@ }, { "BriefDescription": "L2 writebacks that access L2 cache", + "Counter": "0,1,2,3", "EventCode": "0xf0", "EventName": "L2_TRANS.L2_WB", "PublicDescription": "Counts L2 writebacks that access L2 cache.",
@@ -188,6 +211,7 @@ }, { "BriefDescription": "Cycles when L1D is locked", + "Counter": "0,1,2,3", "EventCode": "0x63", "EventName": "LOCK_CYCLES.CACHE_LOCK_DURATION", "PublicDescription": "This event counts the number of cycles when the L1D is locked. It is a superset of the 0x1 mask (BUS_LOCK_CLOCKS.BUS_LOCK_DURATION).",
@@ -196,6 +220,7 @@ }, { "BriefDescription": "Core-originated cacheable requests that missed L3 (Except hardware prefetches to the L3)", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0x2e", "EventName": "LONGEST_LAT_CACHE.MISS", "PublicDescription": "Counts core-originated cacheable requests that miss the L3 cache (Longest Latency cache). Requests include data and code reads, Reads-for-Ownership (RFOs), speculative accesses and hardware prefetches to the L1 and L2. It does not include hardware prefetches to the L3, and may not count other types of requests to the L3.",
@@ -204,6 +229,7 @@ }, { "BriefDescription": "Retired load instructions.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.ALL_LOADS",
@@ -214,6 +240,7 @@ }, { "BriefDescription": "Retired store instructions.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.ALL_STORES",
@@ -224,6 +251,7 @@ }, { "BriefDescription": "All retired memory instructions.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.ANY",
@@ -234,6 +262,7 @@ }, { "BriefDescription": "Retired load instructions with locked access.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.LOCK_LOADS",
@@ -244,6 +273,7 @@ }, { "BriefDescription": "Retired load instructions that split across a cacheline boundary.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.SPLIT_LOADS",
@@ -254,6 +284,7 @@ }, { "BriefDescription": "Retired store instructions that split across a cacheline boundary.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.SPLIT_STORES",
@@ -264,6 +295,7 @@ }, { "BriefDescription": "Retired load instructions that miss the STLB.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.STLB_MISS_LOADS",
@@ -274,6 +306,7 @@ }, { "BriefDescription": "Retired store instructions that miss the STLB.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd0", "EventName": "MEM_INST_RETIRED.STLB_MISS_STORES",
@@ -284,6 +317,7 @@ }, { "BriefDescription": "Snoop hit a modified(HITM) or clean line(HIT_W_FWD) in another on-pkg core which forwarded the data back due to a retired load instruction.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD",
@@ -294,6 +328,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources were L3 hit and cross-core snoop missed in on-pkg core cache.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS",
@@ -304,6 +339,7 @@ }, { "BriefDescription": "Retired load instructions whose data sources were hits in L3 without snoops required", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_NONE",
@@ -314,6 +350,7 @@ }, { "BriefDescription": "Snoop hit without forwarding in another on-pkg core due to a retired load instruction, data was supplied by the L3.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd2", "EventName": "MEM_LOAD_L3_HIT_RETIRED.XSNP_NO_FWD",
@@ -324,6 +361,7 @@ }, { "BriefDescription": "Retired instructions with at least 1 uncacheable load or lock.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd4", "EventName": "MEM_LOAD_MISC_RETIRED.UC",
@@ -334,6 +372,7 @@ }, { "BriefDescription": "Number of completed demand load requests that missed the L1, but hit the FB(fill buffer), because a preceding miss to the same cacheline initiated the line to be brought into L1, but data is not yet ready in L1.", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.FB_HIT",
@@ -344,6 +383,7 @@ }, { "BriefDescription": "Retired load instructions with L1 cache hits as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L1_HIT",
@@ -354,6 +394,7 @@ }, { "BriefDescription": "Retired load instructions missed L1 cache as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L1_MISS",
@@ -364,6 +405,7 @@ }, { "BriefDescription": "Retired load instructions with L2 cache hits as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L2_HIT",
@@ -374,6 +416,7 @@ }, { "BriefDescription": "Retired load instructions missed L2 cache as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L2_MISS",
@@ -384,6 +427,7 @@ }, { "BriefDescription": "Retired load instructions with L3 cache hits as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L3_HIT",
@@ -394,6 +438,7 @@ }, { "BriefDescription": "Retired load instructions missed L3 cache as data sources", + "Counter": "0,1,2,3", "Data_LA": "1", "EventCode": "0xd1", "EventName": "MEM_LOAD_RETIRED.L3_MISS",
@@ -404,6 +449,7 @@ }, { "BriefDescription": "Counts demand data reads that hit a cacheline in the L3 where a snoop hit in another cores caches, data forwarding is required as the data is modified.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -413,6 +459,7 @@ }, { "BriefDescription": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7",
@@ -422,6 +469,7 @@ }, { "BriefDescription": "Counts demand reads for ownership (RFO) requests and software prefetches for exclusive ownership (PREFETCHW) that hit a cacheline in the L3 where a snoop hit in another cores caches, data forwarding is required as the data is modified.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7",
@@ -431,6 +479,7 @@ }, { "BriefDescription": "Demand and prefetch data reads", + "Counter": "0,1,2,3", "EventCode": "0xb0", "EventName": "OFFCORE_REQUESTS.ALL_DATA_RD", "PublicDescription": "Counts the demand and prefetch data reads. All Core Data Reads include cacheable 'Demands' and L2 prefetchers (not L3 prefetchers). Counting also covers reads due to page walks resulted from any request type.",
@@ -439,6 +488,7 @@ }, { "BriefDescription": "Any memory transaction that reached the SQ.", + "Counter": "0,1,2,3", "EventCode": "0xb0", "EventName": "OFFCORE_REQUESTS.ALL_REQUESTS", "PublicDescription": "Counts memory transactions reached the super queue including requests initiated by the core, all L3 prefetches, page walks, etc..",
@@ -447,6 +497,7 @@ }, { "BriefDescription": "Demand Data Read requests sent to uncore", + "Counter": "0,1,2,3", "EventCode": "0xb0", "EventName": "OFFCORE_REQUESTS.DEMAND_DATA_RD", "PublicDescription": "Counts the Demand Data Read requests sent to uncore. Use it in conjunction with OFFCORE_REQUESTS_OUTSTANDING to determine average latency in the uncore.",
@@ -455,6 +506,7 @@ }, { "BriefDescription": "Demand RFO requests including regular RFOs, locks, ItoM", + "Counter": "0,1,2,3", "EventCode": "0xb0", "EventName": "OFFCORE_REQUESTS.DEMAND_RFO", "PublicDescription": "Counts the demand RFO (read for ownership) requests including regular RFOs, locks, ItoM.",
@@ -463,6 +515,7 @@ }, { "BriefDescription": "Offcore outstanding cacheable Core Data Read transactions in SuperQueue (SQ), queue to uncore", + "Counter": "0,1,2,3", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD", "PublicDescription": "Counts the number of offcore outstanding cacheable Core Data Read transactions in the super queue every cycle. A transaction is considered to be in the Offcore outstanding state between L2 miss and transaction completion sent to requestor (SQ de-allocation). See corresponding Umask under OFFCORE_REQUESTS.",
@@ -471,6 +524,7 @@ }, { "BriefDescription": "Cycles when offcore outstanding cacheable Core Data Read transactions are present in SuperQueue (SQ), queue to uncore.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD",
@@ -480,6 +534,7 @@ }, { "BriefDescription": "Cycles when offcore outstanding Demand Data Read transactions are present in SuperQueue (SQ), queue to uncore", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_DATA_RD",
@@ -489,6 +544,7 @@ }, { "BriefDescription": "Cycles with offcore outstanding demand rfo reads transactions in SuperQueue (SQ), queue to uncore.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO",
@@ -498,6 +554,7 @@ }, { "BriefDescription": "Demand Data Read transactions pending for off-core. Highly correlated.", + "Counter": "0,1,2,3", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD", "PublicDescription": "Counts the number of off-core outstanding Demand Data Read transactions every cycle. A transaction is considered to be in the Off-core outstanding state between L2 cache miss and data-return to the core.",
@@ -506,6 +563,7 @@ }, { "BriefDescription": "Cycles with at least 6 offcore outstanding Demand Data Read transactions in uncore queue.", + "Counter": "0,1,2,3", "CounterMask": "6", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD_GE_6",
@@ -514,6 +572,7 @@ }, { "BriefDescription": "Store Read transactions pending for off-core. Highly correlated.", + "Counter": "0,1,2,3", "EventCode": "0x60", "EventName": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_RFO", "PublicDescription": "Counts the number of off-core outstanding read-for-ownership (RFO) store transactions every cycle. An RFO transaction is considered to be in the Off-core outstanding state between L2 cache miss and transaction completion.",
@@ -522,6 +581,7 @@ }, { "BriefDescription": "Counts bus locks, accounts for cache line split locks and UC locks.", + "Counter": "0,1,2,3", "EventCode": "0xf4", "EventName": "SQ_MISC.BUS_LOCK", "PublicDescription": "Counts the more expensive bus lock needed to enforce cache coherency for certain memory accesses that need to be done atomically. Can be created by issuing an atomic instruction (via the LOCK prefix) which causes a cache line split or accesses uncacheable memory.",
@@ -530,6 +590,7 @@ }, { "BriefDescription": "Cycles the superQ cannot take any more entries.", + "Counter": "0,1,2,3", "EventCode": "0xf4", "EventName": "SQ_MISC.SQ_FULL", "PublicDescription": "Counts the cycles for which the thread is active and the superQ cannot take any more entries.",
@@ -538,6 +599,7 @@ }, { "BriefDescription": "Number of PREFETCHNTA instructions executed.", + "Counter": "0,1,2,3", "EventCode": "0x32", "EventName": "SW_PREFETCH_ACCESS.NTA", "PublicDescription": "Counts the number of PREFETCHNTA instructions executed.",
@@ -546,6 +608,7 @@ }, { "BriefDescription": "Number of PREFETCHW instructions executed.", + "Counter": "0,1,2,3", "EventCode": "0x32", "EventName": "SW_PREFETCH_ACCESS.PREFETCHW", "PublicDescription": "Counts the number of PREFETCHW instructions executed.",
@@ -554,6 +617,7 @@ }, { "BriefDescription": "Number of PREFETCHT0 instructions executed.", + "Counter": "0,1,2,3", "EventCode": "0x32", "EventName": "SW_PREFETCH_ACCESS.T0", "PublicDescription": "Counts the number of PREFETCHT0 instructions executed.",
@@ -562,6 +626,7 @@ }, { "BriefDescription": "Number of PREFETCHT1 or PREFETCHT2 instructions executed.", + "Counter": "0,1,2,3", "EventCode": "0x32", "EventName": "SW_PREFETCH_ACCESS.T1_T2", "PublicDescription": "Counts the number of PREFETCHT1 or PREFETCHT2 instructions executed.",
tools/perf/pmu-events/arch/x86/tigerlake/counter.json new file +7 -0
@@ -0,0 +1,7 @@ +[ + { + "Unit": "Core", + "Size": "8", + "FixedSize": "4" + } +]
tools/perf/pmu-events/arch/x86/tigerlake/floating-point.json +13 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "Counts all microcode FP assists.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc1", "EventName": "ASSISTS.FP", "PublicDescription": "Counts all microcode Floating Point assists.",
@@ -9,6 +10,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 128-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE", "PublicDescription": "Number of SSE/AVX computational 128-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -17,6 +19,7 @@ }, { "BriefDescription": "Number of SSE/AVX computational 128-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE", "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -25,6 +28,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE", "PublicDescription": "Number of SSE/AVX computational 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 4 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -33,6 +37,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 256-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE", "PublicDescription": "Number of SSE/AVX computational 256-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RCP DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -41,6 +46,7 @@ }, { "BriefDescription": "Number of SSE/AVX computational 128-bit packed single and 256-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and packed double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.4_FLOPS", "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision and 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point and packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -49,6 +55,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE", "PublicDescription": "Number of SSE/AVX computational 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -57,6 +64,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational 512-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 16 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE", "PublicDescription": "Number of SSE/AVX computational 512-bit packed single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 16 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -65,6 +73,7 @@ }, { "BriefDescription": "Number of SSE/AVX computational 256-bit packed single precision and 512-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RSQRT14 RCP RCP14 DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.8_FLOPS", "PublicDescription": "Number of SSE/AVX computational 256-bit packed single precision and 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision and double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RSQRT14 RCP RCP14 DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -73,6 +82,7 @@ }, { "BriefDescription": "Number of SSE/AVX computational scalar floating-point instructions retired; some instructions will count twice as noted below. Applies to SSE* and AVX* scalar, double and single precision floating-point: ADD SUB MUL DIV MIN MAX RCP14 RSQRT14 RANGE SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform multiple calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.SCALAR", "PublicDescription": "Number of SSE/AVX computational scalar single precision and double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -81,6 +91,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational scalar double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "PublicDescription": "Number of SSE/AVX computational scalar double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -89,6 +100,7 @@ }, { "BriefDescription": "Counts number of SSE/AVX computational scalar single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "PublicDescription": "Number of SSE/AVX computational scalar single precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SSE* and AVX* scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.",
@@ -97,6 +109,7 @@ }, { "BriefDescription": "Number of any Vector retired FP arithmetic instructions", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.VECTOR", "SampleAfterValue": "1000003",
tools/perf/pmu-events/arch/x86/tigerlake/frontend.json +56 -0
358 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/tigerlake/memory.json +31 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "Execution stalls while L3 cache miss demand load is outstanding.", + "Counter": "0,1,2,3", "CounterMask": "6", "EventCode": "0xa3", "EventName": "CYCLE_ACTIVITY.STALLS_L3_MISS",
@@ -9,6 +10,7 @@ }, { "BriefDescription": "Number of machine clears due to memory ordering conflicts.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc3", "EventName": "MACHINE_CLEARS.MEMORY_ORDERING", "PublicDescription": "Counts the number of Machine Clears detected dye to memory ordering. Memory Ordering Machine Clears may apply when a memory read may not conform to the memory ordering rules of the x86 architecture",
@@ -17,6 +19,7 @@ }, { "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 128 cycles.", + "Counter": "0,1,2,3,4,5,6,7", "Data_LA": "1", "EventCode": "0xcd", "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_128",
@@ -25,10 +28,12 @@ "PEBS": "2", "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 128 cycles. Reported latency may be longer than just the memory latency.", "SampleAfterValue": "1009", + "TakenAlone": "1", "UMask": "0x1" }, { "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 16 cycles.", + "Counter": "0,1,2,3,4,5,6,7", "Data_LA": "1", "EventCode": "0xcd", "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_16",
@@ -37,10 +42,12 @@ "PEBS": "2", "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 16 cycles. Reported latency may be longer than just the memory latency.", "SampleAfterValue": "20011", + "TakenAlone": "1", "UMask": "0x1" }, { "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 256 cycles.", + "Counter": "0,1,2,3,4,5,6,7", "Data_LA": "1", "EventCode": "0xcd", "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_256",
@@ -49,10 +56,12 @@ "PEBS": "2", "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 256 cycles. Reported latency may be longer than just the memory latency.", "SampleAfterValue": "503", + "TakenAlone": "1", "UMask": "0x1" }, { "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 32 cycles.", + "Counter": "0,1,2,3,4,5,6,7", "Data_LA": "1", "EventCode": "0xcd", "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_32",
@@ -61,10 +70,12 @@ "PEBS": "2", "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 32 cycles. Reported latency may be longer than just the memory latency.", "SampleAfterValue": "100007", + "TakenAlone": "1", "UMask": "0x1" }, { "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 4 cycles.", + "Counter": "0,1,2,3,4,5,6,7", "Data_LA": "1", "EventCode": "0xcd", "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_4",
@@ -73,10 +84,12 @@ "PEBS": "2", "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 4 cycles. Reported latency may be longer than just the memory latency.", "SampleAfterValue": "100003", + "TakenAlone": "1", "UMask": "0x1" }, { "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 512 cycles.", + "Counter": "0,1,2,3,4,5,6,7", "Data_LA": "1", "EventCode": "0xcd", "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_512",
@@ -85,10 +98,12 @@ "PEBS": "2", "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 512 cycles. Reported latency may be longer than just the memory latency.", "SampleAfterValue": "101", + "TakenAlone": "1", "UMask": "0x1" }, { "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 64 cycles.", + "Counter": "0,1,2,3,4,5,6,7", "Data_LA": "1", "EventCode": "0xcd", "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_64",
@@ -97,10 +112,12 @@ "PEBS": "2", "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 64 cycles. Reported latency may be longer than just the memory latency.", "SampleAfterValue": "2003", + "TakenAlone": "1", "UMask": "0x1" }, { "BriefDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 8 cycles.", + "Counter": "0,1,2,3,4,5,6,7", "Data_LA": "1", "EventCode": "0xcd", "EventName": "MEM_TRANS_RETIRED.LOAD_LATENCY_GT_8",
@@ -109,10 +126,12 @@ "PEBS": "2", "PublicDescription": "Counts randomly selected loads when the latency from first dispatch to completion is greater than 8 cycles. Reported latency may be longer than just the memory latency.", "SampleAfterValue": "50021", + "TakenAlone": "1", "UMask": "0x1" }, { "BriefDescription": "Demand Data Read requests who miss L3 cache", + "Counter": "0,1,2,3", "EventCode": "0xb0", "EventName": "OFFCORE_REQUESTS.L3_MISS_DEMAND_DATA_RD", "PublicDescription": "Demand Data Read requests who miss L3 cache.",
@@ -121,6 +140,7 @@ }, { "BriefDescription": "Number of times an RTM execution aborted.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc9", "EventName": "RTM_RETIRED.ABORTED", "PublicDescription": "Counts the number of times RTM abort was triggered.",
@@ -129,6 +149,7 @@ }, { "BriefDescription": "Number of times an RTM execution aborted due to none of the previous 4 categories (e.g. interrupt)", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc9", "EventName": "RTM_RETIRED.ABORTED_EVENTS", "PublicDescription": "Counts the number of times an RTM execution aborted due to none of the previous 4 categories (e.g. interrupt).",
@@ -137,6 +158,7 @@ }, { "BriefDescription": "Number of times an RTM execution aborted due to various memory events (e.g. read/write capacity and conflicts)", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc9", "EventName": "RTM_RETIRED.ABORTED_MEM", "PublicDescription": "Counts the number of times an RTM execution aborted due to various memory events (e.g. read/write capacity and conflicts).",
@@ -145,6 +167,7 @@ }, { "BriefDescription": "Number of times an RTM execution aborted due to incompatible memory type", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc9", "EventName": "RTM_RETIRED.ABORTED_MEMTYPE", "PublicDescription": "Counts the number of times an RTM execution aborted due to incompatible memory type.",
@@ -153,6 +176,7 @@ }, { "BriefDescription": "Number of times an RTM execution aborted due to HLE-unfriendly instructions", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc9", "EventName": "RTM_RETIRED.ABORTED_UNFRIENDLY", "PublicDescription": "Counts the number of times an RTM execution aborted due to HLE-unfriendly instructions.",
@@ -161,6 +185,7 @@ }, { "BriefDescription": "Number of times an RTM execution successfully committed", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc9", "EventName": "RTM_RETIRED.COMMIT", "PublicDescription": "Counts the number of times RTM commit succeeded.",
@@ -169,6 +194,7 @@ }, { "BriefDescription": "Number of times an RTM execution started.", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0xc9", "EventName": "RTM_RETIRED.START", "PublicDescription": "Counts the number of times we entered an RTM region. Does not count nested transactions.",
@@ -177,6 +203,7 @@ }, { "BriefDescription": "Counts the number of times a class of instructions that may cause a transactional abort was executed inside a transactional region", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0x5d", "EventName": "TX_EXEC.MISC2", "PublicDescription": "Counts Unfriendly TSX abort triggered by a vzeroupper instruction.",
@@ -185,6 +212,7 @@ }, { "BriefDescription": "Number of times an instruction execution caused the transactional nest count supported to be exceeded", + "Counter": "0,1,2,3,4,5,6,7", "EventCode": "0x5d", "EventName": "TX_EXEC.MISC3", "PublicDescription": "Counts Unfriendly TSX abort triggered by a nest count that is too deep.",
@@ -193,6 +221,7 @@ }, { "BriefDescription": "Speculatively counts the number of TSX aborts due to a data capacity limitation for transactional reads", + "Counter": "0,1,2,3", "EventCode": "0x54", "EventName": "TX_MEM.ABORT_CAPACITY_READ", "PublicDescription": "Speculatively counts the number of Transactional Synchronization Extensions (TSX) aborts due to a data capacity limitation for transactional reads",
@@ -201,6 +230,7 @@ }, { "BriefDescription": "Speculatively counts the number of TSX aborts due to a data capacity limitation for transactional writes.", + "Counter": "0,1,2,3", "EventCode": "0x54", "EventName": "TX_MEM.ABORT_CAPACITY_WRITE", "PublicDescription": "Speculatively counts the number of Transactional Synchronization Extensions (TSX) aborts due to a data capacity limitation for transactional writes.",
@@ -209,6 +239,7 @@ }, { "BriefDescription": "Number of times a transactional abort was signaled due to a data conflict on a transactionally accessed address", + "Counter": "0,1,2,3", "EventCode": "0x54", "EventName": "TX_MEM.ABORT_CONFLICT", "PublicDescription": "Counts the number of times a TSX line had a cache conflict.",
tools/perf/pmu-events/arch/x86/tigerlake/other.json +4 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "Core cycles where the core was running in a manner where Turbo may be clipped to the Non-AVX turbo schedule.", + "Counter": "0,1,2,3", "EventCode": "0x28", "EventName": "CORE_POWER.LVL0_TURBO_LICENSE", "PublicDescription": "Counts Core cycles where the core was running with power-delivery for baseline license level 0. This includes non-AVX codes, SSE, AVX 128-bit, and low-current AVX 256-bit codes.",
@@ -9,6 +10,7 @@ }, { "BriefDescription": "Core cycles where the core was running in a manner where Turbo may be clipped to the AVX2 turbo schedule.", + "Counter": "0,1,2,3", "EventCode": "0x28", "EventName": "CORE_POWER.LVL1_TURBO_LICENSE", "PublicDescription": "Counts Core cycles where the core was running with power-delivery for license level 1. This includes high current AVX 256-bit instructions as well as low current AVX 512-bit instructions.",
@@ -17,6 +19,7 @@ }, { "BriefDescription": "Core cycles where the core was running in a manner where Turbo may be clipped to the AVX512 turbo schedule.", + "Counter": "0,1,2,3", "EventCode": "0x28", "EventName": "CORE_POWER.LVL2_TURBO_LICENSE", "PublicDescription": "Core cycles where the core was running with power-delivery for license level 2 (introduced in Skylake Server microarchtecture). This includes high current AVX 512-bit instructions.",
@@ -25,6 +28,7 @@ }, { "BriefDescription": "Counts streaming stores that have any type of response.", + "Counter": "0,1,2,3", "EventCode": "0xB7, 0xBB", "EventName": "OCR.STREAMING_WR.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7",
tools/perf/pmu-events/arch/x86/tigerlake/pipeline.json +96 -0
672 lines not on this page · raw message
tools/perf/pmu-events/arch/x86/tigerlake/uncore-interconnect.json +11 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "UNC_ARB_COH_TRK_REQUESTS.ALL", + "Counter": "0,1", "EventCode": "0x84", "EventName": "UNC_ARB_COH_TRK_REQUESTS.ALL", "PerPkg": "1",
@@ -9,6 +10,7 @@ }, { "BriefDescription": "Each cycle counts number of any coherent request at memory controller that were issued by any core.", + "Counter": "0", "EventCode": "0x85", "EventName": "UNC_ARB_DAT_OCCUPANCY.ALL", "PerPkg": "1",
@@ -17,6 +19,7 @@ }, { "BriefDescription": "Each cycle counts number of coherent reads pending on data return from memory controller that were issued by any core.", + "Counter": "0", "EventCode": "0x85", "EventName": "UNC_ARB_DAT_OCCUPANCY.RD", "PerPkg": "1",
@@ -25,6 +28,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event UNC_ARB_REQ_TRK_REQUEST.DRD", + "Counter": "0,1", "EventCode": "0x81", "EventName": "UNC_ARB_DAT_REQUESTS.RD", "PerPkg": "1",
@@ -33,6 +37,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event UNC_ARB_DAT_OCCUPANCY.ALL", + "Counter": "0", "EventCode": "0x85", "EventName": "UNC_ARB_IFA_OCCUPANCY.ALL", "PerPkg": "1",
@@ -41,6 +46,7 @@ }, { "BriefDescription": "Each cycle count number of 'valid' coherent Data Read entries . Such entry is defined as valid when it is allocated till deallocation. Doesn't include prefetches [This event is alias to UNC_ARB_TRK_OCCUPANCY.RD]", + "Counter": "0", "EventCode": "0x80", "EventName": "UNC_ARB_REQ_TRK_OCCUPANCY.DRD", "PerPkg": "1",
@@ -49,6 +55,7 @@ }, { "BriefDescription": "Number of all coherent Data Read entries. Doesn't include prefetches [This event is alias to UNC_ARB_TRK_REQUESTS.RD]", + "Counter": "0,1", "EventCode": "0x81", "EventName": "UNC_ARB_REQ_TRK_REQUEST.DRD", "PerPkg": "1",
@@ -57,6 +64,7 @@ }, { "BriefDescription": "Each cycle count number of all outgoing valid entries in ReqTrk. Such entry is defined as valid from it's allocation in ReqTrk till deallocation. Accounts for Coherent and non-coherent traffic.", + "Counter": "0", "EventCode": "0x80", "EventName": "UNC_ARB_TRK_OCCUPANCY.ALL", "PerPkg": "1",
@@ -65,6 +73,7 @@ }, { "BriefDescription": "Each cycle count number of 'valid' coherent Data Read entries . Such entry is defined as valid when it is allocated till deallocation. Doesn't include prefetches [This event is alias to UNC_ARB_REQ_TRK_OCCUPANCY.DRD]", + "Counter": "0", "EventCode": "0x80", "EventName": "UNC_ARB_TRK_OCCUPANCY.RD", "PerPkg": "1",
@@ -73,6 +82,7 @@ }, { "BriefDescription": "UNC_ARB_TRK_REQUESTS.ALL", + "Counter": "0,1", "EventCode": "0x81", "EventName": "UNC_ARB_TRK_REQUESTS.ALL", "PerPkg": "1",
@@ -81,6 +91,7 @@ }, { "BriefDescription": "Number of all coherent Data Read entries. Doesn't include prefetches [This event is alias to UNC_ARB_REQ_TRK_REQUEST.DRD]", + "Counter": "0,1", "EventCode": "0x81", "EventName": "UNC_ARB_TRK_REQUESTS.RD", "PerPkg": "1",
tools/perf/pmu-events/arch/x86/tigerlake/uncore-memory.json +6 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "Counts every read (RdCAS) issued by the Memory Controller to DRAM (sum of all channels). All requests result in 64 byte data transfers from DRAM.", + "Counter": "1", "EventCode": "0xff", "EventName": "UNC_MC0_RDCAS_COUNT_FREERUN", "PerPkg": "1",
@@ -9,6 +10,7 @@ }, { "BriefDescription": "Counts every 64B read and write request entering the Memory Controller to DRAM (sum of all channels). Each write request counts as a new request incrementing this counter. However, same cache line write requests (both full and partial) are combined to a single 64 byte data transfer to DRAM.", + "Counter": "0", "EventCode": "0xff", "EventName": "UNC_MC0_TOTAL_REQCOUNT_FREERUN", "PerPkg": "1",
@@ -17,6 +19,7 @@ }, { "BriefDescription": "Counts every write (WrCAS) issued by the Memory Controller to DRAM (sum of all channels). All requests result in 64 byte data transfers from DRAM.", + "Counter": "2", "EventCode": "0xff", "EventName": "UNC_MC0_WRCAS_COUNT_FREERUN", "PerPkg": "1",
@@ -25,6 +28,7 @@ }, { "BriefDescription": "Counts every read (RdCAS) issued by the Memory Controller to DRAM (sum of all channels). All requests result in 64 byte data transfers from DRAM.", + "Counter": "4", "EventCode": "0xff", "EventName": "UNC_MC1_RDCAS_COUNT_FREERUN", "PerPkg": "1",
@@ -33,6 +37,7 @@ }, { "BriefDescription": "Counts every 64B read and write request entering the Memory Controller to DRAM (sum of all channels). Each write request counts as a new request incrementing this counter. However, same cache line write requests (both full and partial) are combined to a single 64 byte data transfer to DRAM.", + "Counter": "3", "EventCode": "0xff", "EventName": "UNC_MC1_TOTAL_REQCOUNT_FREERUN", "PerPkg": "1",
@@ -41,6 +46,7 @@ }, { "BriefDescription": "Counts every write (WrCAS) issued by the Memory Controller to DRAM (sum of all channels). All requests result in 64 byte data transfers from DRAM.", + "Counter": "5", "EventCode": "0xff", "EventName": "UNC_MC1_WRCAS_COUNT_FREERUN", "PerPkg": "1",
tools/perf/pmu-events/arch/x86/tigerlake/uncore-other.json +1 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "UNC_CLOCK.SOCKET", + "Counter": "Fixed", "EventCode": "0xff", "EventName": "UNC_CLOCK.SOCKET", "PerPkg": "1",
tools/perf/pmu-events/arch/x86/tigerlake/virtual-memory.json +20 -0
@@ -1,6 +1,7 @@ [ { "BriefDescription": "Loads that miss the DTLB and hit the STLB.", + "Counter": "0,1,2,3", "EventCode": "0x08", "EventName": "DTLB_LOAD_MISSES.STLB_HIT", "PublicDescription": "Counts loads that miss the DTLB (Data TLB) and hit the STLB (Second level TLB).",
@@ -9,6 +10,7 @@ }, { "BriefDescription": "Cycles when at least one PMH is busy with a page walk for a demand load.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x08", "EventName": "DTLB_LOAD_MISSES.WALK_ACTIVE",
@@ -18,6 +20,7 @@ }, { "BriefDescription": "Load miss in all TLB levels causes a page walk that completes. (All page sizes)", + "Counter": "0,1,2,3", "EventCode": "0x08", "EventName": "DTLB_LOAD_MISSES.WALK_COMPLETED", "PublicDescription": "Counts completed page walks (all page sizes) caused by demand data loads. This implies it missed in the DTLB and further levels of TLB. The page walk can end with or without a fault.",
@@ -26,6 +29,7 @@ }, { "BriefDescription": "Page walks completed due to a demand data load to a 2M/4M page.", + "Counter": "0,1,2,3", "EventCode": "0x08", "EventName": "DTLB_LOAD_MISSES.WALK_COMPLETED_2M_4M", "PublicDescription": "Counts completed page walks (2M/4M sizes) caused by demand data loads. This implies address translations missed in the DTLB and further levels of TLB. The page walk can end with or without a fault.",
@@ -34,6 +38,7 @@ }, { "BriefDescription": "Page walks completed due to a demand data load to a 4K page.", + "Counter": "0,1,2,3", "EventCode": "0x08", "EventName": "DTLB_LOAD_MISSES.WALK_COMPLETED_4K", "PublicDescription": "Counts completed page walks (4K sizes) caused by demand data loads. This implies address translations missed in the DTLB and further levels of TLB. The page walk can end with or without a fault.",
@@ -42,6 +47,7 @@ }, { "BriefDescription": "Number of page walks outstanding for a demand load in the PMH each cycle.", + "Counter": "0,1,2,3", "EventCode": "0x08", "EventName": "DTLB_LOAD_MISSES.WALK_PENDING", "PublicDescription": "Counts the number of page walks outstanding for a demand load in the PMH (Page Miss Handler) each cycle.",
@@ -50,6 +56,7 @@ }, { "BriefDescription": "Stores that miss the DTLB and hit the STLB.", + "Counter": "0,1,2,3", "EventCode": "0x49", "EventName": "DTLB_STORE_MISSES.STLB_HIT", "PublicDescription": "Counts stores that miss the DTLB (Data TLB) and hit the STLB (2nd Level TLB).",
@@ -58,6 +65,7 @@ }, { "BriefDescription": "Cycles when at least one PMH is busy with a page walk for a store.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x49", "EventName": "DTLB_STORE_MISSES.WALK_ACTIVE",
@@ -67,6 +75,7 @@ }, { "BriefDescription": "Store misses in all TLB levels causes a page walk that completes. (All page sizes)", + "Counter": "0,1,2,3", "EventCode": "0x49", "EventName": "DTLB_STORE_MISSES.WALK_COMPLETED", "PublicDescription": "Counts completed page walks (all page sizes) caused by demand data stores. This implies it missed in the DTLB and further levels of TLB. The page walk can end with or without a fault.",
@@ -75,6 +84,7 @@ }, { "BriefDescription": "Page walks completed due to a demand data store to a 2M/4M page.", + "Counter": "0,1,2,3", "EventCode": "0x49", "EventName": "DTLB_STORE_MISSES.WALK_COMPLETED_2M_4M", "PublicDescription": "Counts page walks completed due to demand data stores whose address translations missed in the TLB and were mapped to 2M/4M pages. The page walks can end with or without a page fault.",
@@ -83,6 +93,7 @@ }, { "BriefDescription": "Page walks completed due to a demand data store to a 4K page.", + "Counter": "0,1,2,3", "EventCode": "0x49", "EventName": "DTLB_STORE_MISSES.WALK_COMPLETED_4K", "PublicDescription": "Counts page walks completed due to demand data stores whose address translations missed in the TLB and were mapped to 4K pages. The page walks can end with or without a page fault.",
@@ -91,6 +102,7 @@ }, { "BriefDescription": "Number of page walks outstanding for a store in the PMH each cycle.", + "Counter": "0,1,2,3", "EventCode": "0x49", "EventName": "DTLB_STORE_MISSES.WALK_PENDING", "PublicDescription": "Counts the number of page walks outstanding for a store in the PMH (Page Miss Handler) each cycle.",
@@ -99,6 +111,7 @@ }, { "BriefDescription": "Instruction fetch requests that miss the ITLB and hit the STLB.", + "Counter": "0,1,2,3", "EventCode": "0x85", "EventName": "ITLB_MISSES.STLB_HIT", "PublicDescription": "Counts instruction fetch requests that miss the ITLB (Instruction TLB) and hit the STLB (Second-level TLB).",
@@ -107,6 +120,7 @@ }, { "BriefDescription": "Cycles when at least one PMH is busy with a page walk for code (instruction fetch) request.", + "Counter": "0,1,2,3", "CounterMask": "1", "EventCode": "0x85", "EventName": "ITLB_MISSES.WALK_ACTIVE",
@@ -116,6 +130,7 @@ }, { "BriefDescription": "Code miss in all TLB levels causes a page walk that completes. (All page sizes)", + "Counter": "0,1,2,3", "EventCode": "0x85", "EventName": "ITLB_MISSES.WALK_COMPLETED", "PublicDescription": "Counts completed page walks (all page sizes) caused by a code fetch. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB. The page walk can end with or without a fault.",
@@ -124,6 +139,7 @@ }, { "BriefDescription": "Code miss in all TLB levels causes a page walk that completes. (2M/4M)", + "Counter": "0,1,2,3", "EventCode": "0x85", "EventName": "ITLB_MISSES.WALK_COMPLETED_2M_4M", "PublicDescription": "Counts completed page walks (2M/4M page sizes) caused by a code fetch. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB. The page walk can end with or without a fault.",
@@ -132,6 +148,7 @@ }, { "BriefDescription": "Code miss in all TLB levels causes a page walk that completes. (4K)", + "Counter": "0,1,2,3", "EventCode": "0x85", "EventName": "ITLB_MISSES.WALK_COMPLETED_4K", "PublicDescription": "Counts completed page walks (4K page sizes) caused by a code fetch. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB. The page walk can end with or without a fault.",
@@ -140,6 +157,7 @@ }, { "BriefDescription": "Number of page walks outstanding for an outstanding code request in the PMH each cycle.", + "Counter": "0,1,2,3", "EventCode": "0x85", "EventName": "ITLB_MISSES.WALK_PENDING", "PublicDescription": "Counts the number of page walks outstanding for an outstanding code (instruction fetch) request in the PMH (Page Miss Handler) each cycle.",
@@ -148,6 +166,7 @@ }, { "BriefDescription": "DTLB flush attempts of the thread-specific entries", + "Counter": "0,1,2,3", "EventCode": "0xbd", "EventName": "TLB_FLUSH.DTLB_THREAD", "PublicDescription": "Counts the number of DTLB flush attempts of the thread-specific entries.",
@@ -156,6 +175,7 @@ }, { "BriefDescription": "STLB flush attempts", + "Counter": "0,1,2,3", "EventCode": "0xbd", "EventName": "TLB_FLUSH.STLB_ANY", "PublicDescription": "Counts the number of any STLB flush attempts (such as entire, VPID, PCID, InvPage, CR3 write, etc.).",
Signature · 1 line
2.39.3