[PATCH v2 0/2] perf jevents: Add IOMMU PTW & device entry metrics

[PATCH v2 0/2] perf jevents: Add IOMMU PTW & device entry metrics

[Jacob Lefkowitz via B4 Relay]

Adds AMD specific metrics for the IOMMU with respect to page table
walks and device table entires. The metrics cover ratios, rates, and
percentages with respect to any event which might be triggered during
a page table walk or device table entry lookup.

The metrics target Zen 2+ processors and use the standard amd_iommu
PMU events. All metrics are tested with 'sleep 10' and behave as
expected. An example output of sleep 10 is provided within each
individual change. 

The changes are split into two patches:
1. perf jevents: add iommu page table walk metrics
2. perf jevents: add amd specific device table entry metrics

The specific metris added are:
- lpm_iommu_dte_miss_rate
- lpm_iommu_dte_lookups
- lpm_iommu_ptw_reads_per_tlb_miss
- lpm_iommu_ptw_nested_read_ratio
- lpm_iommu_ptw_guest_read_ratio
- lpm_iommu_trans_per_sec

Signed-off-by: Jacob Lefkowitz <jlefkowi@akamai.com>
---
Changes in v2:
- Drop min() wrappers d_ratio() for the IOMMU ratio metrics: min(NaN, 1)
  evaluated to 1, reporting a false 100% on sockets with no amd_counts
- Did not test with previously mentioned custom workload and did not add
  to the workloads collection. The workload stressed the CPU TLB,
  but the IOMMU only translates device DMA, so the load never reaches
  the IOMMU.
- Remove unused min import
- Remove global declaration of '_zen_model' within each function
- Link to v1: https://lore.kernel.org/r/20261029-perf-tools-next-v1-0-a75a2b64a15e@akamai.com

---
Jacob Lefkowitz (2):
      perf jevents: add iommu page table walk metrics
      perf jevents: add amd specific device table entry metrics

 tools/perf/pmu-events/amd_metrics.py | 75 ++++++++++++++++++++++++++++++++++++
 1 file changed, 75 insertions(+)
---
base-commit: 1c90f0a3c8a1c4e38f9edd97af9e20848805a7fb
change-id: 20261028-perf-tools-next-8fd1d433316a

Best regards,
-- 
Jacob Lefkowitz <jlefkowi@akamai.com>

[PATCH v2 1/2] perf jevents: add iommu page table walk metrics

[Jacob Lefkowitz via B4 Relay]

From: Jacob Lefkowitz <jlefkowi@akamai.com>

Adds page table walk metrics specfically to perf jevents for AMD
platforms to enhance observability into IOMMU overhead. The metrics
are supported on Zen 2 and neweer processors. They are implemented
using the amd_iommu PMU events.

The list of metrics added here are:
- lpm_iommu_ptw_reads_per_tlb_miss
- lpm_iommu_ptw_nested_read_ratio
- lpm_iommu_ptw_guest_read_ratio
- lpm_iommu_trans_per_sec

The changes here have been tested with 'sleep 10' as the workload:

perf stat -M iommuptw --per-socket --metric-only -j -a -- sleep 10
{"socket" : "S0", "counters" : 8,
"reads/tlb_miss lpm_iommu_ptw_reads_per_tlb_miss" : "0.5",
"%  lpm_iommu_ptw_nested_read_ratio" : "none",
"%  lpm_iommu_ptw_guest_read_ratio" : "0.0",
"trans/s  lpm_iommu_trans_per_sec" : "971.5"}
{"socket" : "S1", "counters" : 0,
"reads/tlb_miss  lpm_iommu_ptw_reads_per_tlb_miss" : "none",
"%  lpm_iommu_ptw_nested_read_ratio" : "none",
"%  lpm_iommu_ptw_guest_read_ratio" : "none,
"trans/s  lpm_iommu_trans_per_sec" : "none"}

Signed-off-by: Jacob Lefkowitz <jlefkowi@akamai.com>
---
 tools/perf/pmu-events/amd_metrics.py | 45 ++++++++++++++++++++++++++++++++++++
 1 file changed, 45 insertions(+)

diff --git a/tools/perf/pmu-events/amd_metrics.py b/tools/perf/pmu-events/amd_metrics.py
index 42569b9153f4..452b88033492 100755
--- a/tools/perf/pmu-events/amd_metrics.py
+++ b/tools/perf/pmu-events/amd_metrics.py
@@ -318,6 +318,50 @@ def AmdIotlb() -> Optional[MetricGroup]:
     )
 
 
+def AmdIommuptw() -> Optional[MetricGroup]:
+    if _zen_model < 2:
+        return None
+
+    walks = Event("amd_iommu/mem_iommu_tlb_pte_mis/") + Event(
+        "amd_iommu/mem_iommu_tlb_pde_mis/"
+    )
+    pt_rd = Event("amd_iommu/page_tbl_read_tot/")
+    pt_rd_nst = Event("amd_iommu/page_tbl_read_nst/")
+    pt_rd_gst = Event("amd_iommu/page_tbl_read_gst/")
+    trans = Event("amd_iommu/mem_trans_total/")
+
+    return MetricGroup(
+        "iommuptw",
+        [
+            Metric(
+                "lpm_iommu_ptw_reads_per_tlb_miss",
+                "Average IOMMU page table reads per IOTLB miss",
+                d_ratio(pt_rd, walks),
+                "reads/tlb_miss"
+            ),
+            Metric(
+                "lpm_iommu_ptw_nested_read_ratio",
+                "Share of IOMMU page table reads from nested (host) tables",
+                d_ratio(pt_rd_nst, pt_rd),
+                "100%"
+            ),
+            Metric(
+                "lpm_iommu_ptw_guest_read_ratio",
+                "Share of IOMMU page table reads from guest tables",
+                d_ratio(pt_rd_gst, pt_rd),
+                "100%"
+            ),
+            Metric(
+                "lpm_iommu_trans_per_sec",
+                "IOMMU translation requests per second",
+                d_ratio(trans, interval_sec),
+                "trans/s"
+            )
+        ],
+        description="IOMMU page table walk metrics"
+    )
+
+
 def AmdItlb():
     l2h = Event("bp_l1_tlb_miss_l2_tlb_hit", "bp_l1_tlb_miss_l2_hit")
     l2m = Event("bp_l1_tlb_miss_l2_tlb_miss.all", "l2_itlb_misses",)
@@ -526,6 +570,7 @@ def main() -> None:
         AmdCtxSw(),
         AmdDtlb(),
         AmdIotlb(),
+        AmdIommuptw(),
         AmdItlb(),
         AmdLdSt(),
         AmdUpc(),

-- 
2.53.0

Re: [PATCH v2 1/2] perf jevents: add iommu page table walk metrics

[sashiko-bot]

> From: Jacob Lefkowitz <jlefkowi@akamai.com>
> 
> Adds page table walk metrics specfically to perf jevents for AMD
> platforms to enhance observability into IOMMU overhead. The metrics
> are supported on Zen 2 and neweer processors. They are implemented
> using the amd_iommu PMU events.
> 
> The list of metrics added here are:
> - lpm_iommu_ptw_reads_per_tlb_miss
> - lpm_iommu_ptw_nested_read_ratio
> - lpm_iommu_ptw_guest_read_ratio
> - lpm_iommu_trans_per_sec
> 
> The changes here have been tested with 'sleep 10' as the workload:
> 
> perf stat -M iommuptw --per-socket --metric-only -j -a -- sleep 10
> {"socket" : "S0", "counters" : 8,
> "reads/tlb_miss lpm_iommu_ptw_reads_per_tlb_miss" : "0.5",
> "%  lpm_iommu_ptw_nested_read_ratio" : "none",
> "%  lpm_iommu_ptw_guest_read_ratio" : "0.0",
> "trans/s  lpm_iommu_trans_per_sec" : "971.5"}
> {"socket" : "S1", "counters" : 0,
> "reads/tlb_miss  lpm_iommu_ptw_reads_per_tlb_miss" : "none",
> "%  lpm_iommu_ptw_nested_read_ratio" : "none",
> "%  lpm_iommu_ptw_guest_read_ratio" : "none,
> "trans/s  lpm_iommu_trans_per_sec" : "none"}
> 
> Signed-off-by: Jacob Lefkowitz <jlefkowi@akamai.com>

Sashiko has reviewed this patch and found no issues.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261102-perf-tools-next-v2-0-4d237f886710@akamai.com?part=1

[PATCH v2 2/2] perf jevents: add amd specific device table entry metrics

[Jacob Lefkowitz via B4 Relay]

From: Jacob Lefkowitz <jlefkowi@akamai.com>

Adds AMD specific device table entry metrics to perf jevents.
These metrics are only allowed on Zen 2+ processors and are
implemented using the 'amd_iommu' PMU events.

The list of metrics added are:
- lpm_iommu_dte_miss_rate
- lpm_iommu_dte_lookups

The metrics were tested with the standard 'sleep' workload:

perf stat -M iommudte --per-socket --metric-only -j -a -- sleep 10
{"socket" : "S0", "counters" : 8,
"lookups/s  lpm_iommu_dte_lookups" : "4117.3",
"%  lpm_iommu_dte_miss_rate" : "0.0"}
{"socket" : "S1", "counters" : 0,
"lookups/s  lpm_iommu_dte_lookups" : "none",
"%  lpm_iommu_dte_miss_rate" : "none"}

Signed-off-by: Jacob Lefkowitz <jlefkowi@akamai.com>
---
 tools/perf/pmu-events/amd_metrics.py | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/tools/perf/pmu-events/amd_metrics.py b/tools/perf/pmu-events/amd_metrics.py
index 452b88033492..fd0bd9f16529 100755
--- a/tools/perf/pmu-events/amd_metrics.py
+++ b/tools/perf/pmu-events/amd_metrics.py
@@ -318,6 +318,35 @@ def AmdIotlb() -> Optional[MetricGroup]:
     )
 
 
+def AmdIommudte() -> Optional[MetricGroup]:
+    if _zen_model < 2:
+        return None
+
+    dte_hit = Event("amd_iommu/mem_dte_hit/")
+    dte_mis = Event("amd_iommu/mem_dte_mis/")
+    lookups = dte_hit + dte_mis
+
+    return MetricGroup(
+        "iommudte",
+        [
+            Metric(
+                "lpm_iommu_dte_miss_rate",
+                "IOMMU device table cache miss rate",
+                d_ratio(dte_mis, lookups),
+                "100%"
+            ),
+            Metric(
+                "lpm_iommu_dte_lookups",
+                "IOMMU device table cache lookups",
+                d_ratio(lookups, interval_sec),
+                "lookups/s"
+            )
+        ],
+        description="IOMMU Device Table Cache Metrics"
+    )
+
+
+
 def AmdIommuptw() -> Optional[MetricGroup]:
     if _zen_model < 2:
         return None
@@ -570,6 +599,7 @@ def main() -> None:
         AmdCtxSw(),
         AmdDtlb(),
         AmdIotlb(),
+        AmdIommudte(),
         AmdIommuptw(),
         AmdItlb(),
         AmdLdSt(),

-- 
2.53.0

Re: [PATCH v2 2/2] perf jevents: add amd specific device table entry metrics

[sashiko-bot]

> From: Jacob Lefkowitz <jlefkowi@akamai.com>
> 
> Adds AMD specific device table entry metrics to perf jevents.
> These metrics are only allowed on Zen 2+ processors and are
> implemented using the 'amd_iommu' PMU events.
> 
> The list of metrics added are:
> - lpm_iommu_dte_miss_rate
> - lpm_iommu_dte_lookups
> 
> The metrics were tested with the standard 'sleep' workload:
> 
> perf stat -M iommudte --per-socket --metric-only -j -a -- sleep 10
> {"socket" : "S0", "counters" : 8,
> "lookups/s  lpm_iommu_dte_lookups" : "4117.3",
> "%  lpm_iommu_dte_miss_rate" : "0.0"}
> {"socket" : "S1", "counters" : 0,
> "lookups/s  lpm_iommu_dte_lookups" : "none",
> "%  lpm_iommu_dte_miss_rate" : "none"}
> 
> Signed-off-by: Jacob Lefkowitz <jlefkowi@akamai.com>

Sashiko has reviewed this patch and found no issues.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261102-perf-tools-next-v2-0-4d237f886710@akamai.com?part=2