Thread (46 messages) 46 messages, 5 authors, 2022-12-04

Re: [RFC PATCH v2 3/6] perf vendor events arm64: Add cache metrics for neoverse-n2

From: Xing Zhengjun <hidden>
Date: 2022-11-14 08:36:08
Also in: linux-perf-users, lkml


On 11/14/2022 3:41 PM, Jing Zhang wrote:
quoted hunk ↗ jump to hunk
Add cache related metrics.

Signed-off-by: Jing Zhang <renyu.zj@linux.alibaba.com>
---
  .../arch/arm64/arm/neoverse-n2/metrics.json        | 77 ++++++++++++++++++++++
  1 file changed, 77 insertions(+)
diff --git a/tools/perf/pmu-events/arch/arm64/arm/neoverse-n2/metrics.json b/tools/perf/pmu-events/arch/arm64/arm/neoverse-n2/metrics.json
index 324ca12..1690ef6 100644
--- a/tools/perf/pmu-events/arch/arm64/arm/neoverse-n2/metrics.json
+++ b/tools/perf/pmu-events/arch/arm64/arm/neoverse-n2/metrics.json
@@ -54,5 +54,82 @@
          "BriefDescription": "The rate of DTLB Walks to the overall TLB lookups",
          "MetricGroup": "TLB",
          "MetricName": "dtlb_walk_rate"
+    },
+    {
+        "MetricExpr": "L1I_CACHE_REFILL / INST_RETIRED * 1000",
+        "PublicDescription": "The rate of L1 I-Cache misses per kilo instructions",
+        "BriefDescription": "The rate of L1 I-Cache misses per kilo instructions",
+        "MetricGroup": "Cache",
+        "MetricName": "l1i_cache_mpki"
+    },
+    {
+        "MetricExpr": "L1I_CACHE_REFILL / L1I_CACHE",
+        "PublicDescription": "The rate of L1 I-Cache misses to the overall L1 I-Cache",
+        "BriefDescription": "The rate of L1 I-Cache misses to the overall L1 I-Cache",
+        "MetricGroup": "Cache",
+        "MetricName": "l1i_cache_miss_rate"
+    },
+    {
+        "MetricExpr": "L1D_CACHE_REFILL / INST_RETIRED * 1000",
+        "PublicDescription": "The rate of L1 D-Cache misses per kilo instructions",
+        "BriefDescription": "The rate of L1 D-Cache misses per kilo instructions",
+        "MetricGroup": "Cache",
+        "MetricName": "l1d_cache_mpki"
+    },
+    {
+        "MetricExpr": "L1D_CACHE_REFILL / L1D_CACHE",
+        "PublicDescription": "The rate of L1 D-Cache misses to the overall L1 D-Cache",
+        "BriefDescription": "The rate of L1 D-Cache misses to the overall L1 D-Cache",
+        "MetricGroup": "Cache",
+        "MetricName": "l1d_cache_miss_rate"
+    },
+    {
+        "MetricExpr": "L2D_CACHE_REFILL / INST_RETIRED * 1000",
+        "PublicDescription": "The rate of L2 D-Cache misses per kilo instructions",
+        "BriefDescription": "The rate of L2 D-Cache misses per kilo instructions",
+        "MetricGroup": "Cache",
+        "MetricName": "l2d_cache_mpki"
+    },
+    {
+        "MetricExpr": "L2D_CACHE_REFILL / L2D_CACHE",
+        "PublicDescription": "The rate of L2 D-Cache misses to the overall L2 D-Cache",
+        "BriefDescription": "The rate of L2 D-Cache misses to the overall L2 D-Cache",
+        "MetricGroup": "Cache",
+        "MetricName": "l2d_cache_miss_rate"
+    },
+    {
+        "MetricExpr": "L3D_CACHE_REFILL / INST_RETIRED * 1000",
+        "PublicDescription": "The rate of L3 D-Cache misses per kilo instructions",
+        "BriefDescription": "The rate of L3 D-Cache misses per kilo instructions",
+        "MetricGroup": "Cache",
+        "MetricName": "l3d_cache_mpki"
+    },
+    {
+        "MetricExpr": "L3D_CACHE_REFILL / L3D_CACHE",
+        "PublicDescription": "The rate of L3 D-Cache misses to the overall L3 D-Cache",
+        "BriefDescription": "The rate of L3 D-Cache misses to the overall L3 D-Cache",
+        "MetricGroup": "Cache",
+        "MetricName": "l3d_cache_miss_rate"
+    },
+    {
+        "MetricExpr": "LL_CACHE_MISS_RD / INST_RETIRED * 1000",
+        "PublicDescription": "The rate of LL Cache read misses per kilo instructions",
+        "BriefDescription": "The rate of LL Cache read misses per kilo instructions",
+        "MetricGroup": "Cache",
+        "MetricName": "ll_cache_read_mpki"
+    },
+    {
+        "MetricExpr": "LL_CACHE_MISS_RD / LL_CACHE_RD",
+        "PublicDescription": "The rate of LL Cache read misses to the overall LL Cache read",
+        "BriefDescription": "The rate of LL Cache read misses to the overall LL Cache read",
+        "MetricGroup": "Cache",
+        "MetricName": "ll_cache_read_miss_rate"
+    },
+    {
+        "MetricExpr": "(LL_CACHE_RD - LL_CACHE_MISS_RD) / LL_CACHE_RD",
+        "PublicDescription": "The rate of LL Cache read hit to the overall LL Cache read",
+        "BriefDescription": "The rate of LL Cache read hit to the overall LL Cache read",
+        "MetricGroup": "Cache",
+        "MetricName": "ll_cache_read_hit_rate"
      }
  ]
\ No newline at end of file
It is better to fix this by adding a newline at the end of the file.

-- 
Zhengjun Xing

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help