lkml.org 
[lkml]   [2017]   [May]   [17]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH v5 3/3] perf: cavium: Support transmit-link PMU counters
Date
Add support for the transmit-link (TLK) PMU counters found
on Caviums SOCs with an interconnect.

Signed-off-by: Jan Glauber <jglauber@cavium.com>
---
drivers/edac/thunderx_edac.c | 7 ++
drivers/perf/cavium_pmu.c | 223 +++++++++++++++++++++++++++++++++++++++-
include/linux/perf/cavium_pmu.h | 1 +
3 files changed, 230 insertions(+), 1 deletion(-)

diff --git a/drivers/edac/thunderx_edac.c b/drivers/edac/thunderx_edac.c
index 8de4faf..d3da6d3 100644
--- a/drivers/edac/thunderx_edac.c
+++ b/drivers/edac/thunderx_edac.c
@@ -1496,6 +1496,10 @@ static int thunderx_ocx_probe(struct pci_dev *pdev,

writeq(OCX_COM_INT_ENA_ALL, ocx->regs + OCX_COM_INT_ENA_W1S);

+ if (IS_ENABLED(CONFIG_CAVIUM_PMU))
+ ocx->pmu_data = cvm_pmu_probe(pdev, ocx->regs,
+ CVM_PMU_TLK);
+
return 0;
err_free:
edac_device_free_ctl_info(edac_dev);
@@ -1509,6 +1513,9 @@ static void thunderx_ocx_remove(struct pci_dev *pdev)
struct thunderx_ocx *ocx = edac_dev->pvt_info;
int i;

+ if (IS_ENABLED(CONFIG_CAVIUM_PMU))
+ cvm_pmu_remove(pdev, ocx->pmu_data, CVM_PMU_TLK);
+
writeq(OCX_COM_INT_ENA_ALL, ocx->regs + OCX_COM_INT_ENA_W1C);

for (i = 0; i < OCX_INTS; i++) {
diff --git a/drivers/perf/cavium_pmu.c b/drivers/perf/cavium_pmu.c
index 9881dc8..379c08a 100644
--- a/drivers/perf/cavium_pmu.c
+++ b/drivers/perf/cavium_pmu.c
@@ -21,6 +21,7 @@
#include <linux/perf/cavium_pmu.h>

struct list_head cvm_pmu_lmcs;
+struct list_head cvm_pmu_tlks;

/*
* Common Cavium PMU stuff
@@ -367,11 +368,228 @@ static void cvm_pmu_lmc_remove(struct pci_dev *pdev, void *pmu_data)
kfree(lmc);
}

+/*
+ * CCPI interface controller (OCX) Transmit link (TLK) counters:
+ * - per-unit control
+ * - writable
+ * - one PCI device with multiple TLK units
+ */
+
+#define TLK_NR_UNITS 3
+#define TLK_UNIT_OFFSET 0x2000
+#define TLK_START_ADDR 0x10000
+#define TLK_STAT_CTL_OFFSET 0x40
+#define TLK_STAT_OFFSET 0x400
+
+#define TLK_STAT_ENABLE_BIT BIT(0)
+#define TLK_STAT_RESET_BIT BIT(1)
+
+#define CVM_PMU_TLK_EVENT_ATTR(_name, _id) \
+ &((struct perf_pmu_events_attr[]) { \
+ { \
+ __ATTR(_name, S_IRUGO, cvm_pmu_event_sysfs_show, NULL), \
+ 0, \
+ "tlk_event=" __stringify(_id), \
+ } \
+ })[0].attr.attr
+
+static void cvm_pmu_tlk_enable_pmu(struct pmu *pmu)
+{
+ struct cvm_pmu_dev *pmu_dev = container_of(pmu, struct cvm_pmu_dev, pmu);
+
+ /* enable all counters */
+ writeb(TLK_STAT_ENABLE_BIT, pmu_dev->map + TLK_STAT_CTL_OFFSET);
+}
+
+static void cvm_pmu_tlk_disable_pmu(struct pmu *pmu)
+{
+ struct cvm_pmu_dev *pmu_dev = container_of(pmu, struct cvm_pmu_dev, pmu);
+
+ /* disable all counters */
+ writeb(0, pmu_dev->map + TLK_STAT_CTL_OFFSET);
+}
+
+static int cvm_pmu_tlk_add(struct perf_event *event, int flags)
+{
+ struct hw_perf_event *hwc = &event->hw;
+
+ return cvm_pmu_add(event, flags, TLK_STAT_CTL_OFFSET,
+ TLK_STAT_OFFSET + hwc->config * 8);
+}
+
+PMU_FORMAT_ATTR(tlk_event, "config:0-5");
+
+static struct attribute *cvm_pmu_tlk_format_attr[] = {
+ &format_attr_tlk_event.attr,
+ NULL,
+};
+
+static struct attribute_group cvm_pmu_tlk_format_group = {
+ .name = "format",
+ .attrs = cvm_pmu_tlk_format_attr,
+};
+
+static struct attribute *cvm_pmu_tlk_events_attr[] = {
+ CVM_PMU_TLK_EVENT_ATTR(idle_cnt, 0x00),
+ CVM_PMU_TLK_EVENT_ATTR(data_cnt, 0x01),
+ CVM_PMU_TLK_EVENT_ATTR(sync_cnt, 0x02),
+ CVM_PMU_TLK_EVENT_ATTR(retry_cnt, 0x03),
+ CVM_PMU_TLK_EVENT_ATTR(err_cnt, 0x04),
+ CVM_PMU_TLK_EVENT_ATTR(mat0_cnt, 0x08),
+ CVM_PMU_TLK_EVENT_ATTR(mat1_cnt, 0x09),
+ CVM_PMU_TLK_EVENT_ATTR(mat2_cnt, 0x0a),
+ CVM_PMU_TLK_EVENT_ATTR(mat3_cnt, 0x0b),
+ CVM_PMU_TLK_EVENT_ATTR(vc0_cmd, 0x10),
+ CVM_PMU_TLK_EVENT_ATTR(vc1_cmd, 0x11),
+ CVM_PMU_TLK_EVENT_ATTR(vc2_cmd, 0x12),
+ CVM_PMU_TLK_EVENT_ATTR(vc3_cmd, 0x13),
+ CVM_PMU_TLK_EVENT_ATTR(vc4_cmd, 0x14),
+ CVM_PMU_TLK_EVENT_ATTR(vc5_cmd, 0x15),
+ CVM_PMU_TLK_EVENT_ATTR(vc0_pkt, 0x20),
+ CVM_PMU_TLK_EVENT_ATTR(vc1_pkt, 0x21),
+ CVM_PMU_TLK_EVENT_ATTR(vc2_pkt, 0x22),
+ CVM_PMU_TLK_EVENT_ATTR(vc3_pkt, 0x23),
+ CVM_PMU_TLK_EVENT_ATTR(vc4_pkt, 0x24),
+ CVM_PMU_TLK_EVENT_ATTR(vc5_pkt, 0x25),
+ CVM_PMU_TLK_EVENT_ATTR(vc6_pkt, 0x26),
+ CVM_PMU_TLK_EVENT_ATTR(vc7_pkt, 0x27),
+ CVM_PMU_TLK_EVENT_ATTR(vc8_pkt, 0x28),
+ CVM_PMU_TLK_EVENT_ATTR(vc9_pkt, 0x29),
+ CVM_PMU_TLK_EVENT_ATTR(vc10_pkt, 0x2a),
+ CVM_PMU_TLK_EVENT_ATTR(vc11_pkt, 0x2b),
+ CVM_PMU_TLK_EVENT_ATTR(vc12_pkt, 0x2c),
+ CVM_PMU_TLK_EVENT_ATTR(vc13_pkt, 0x2d),
+ CVM_PMU_TLK_EVENT_ATTR(vc0_con, 0x30),
+ CVM_PMU_TLK_EVENT_ATTR(vc1_con, 0x31),
+ CVM_PMU_TLK_EVENT_ATTR(vc2_con, 0x32),
+ CVM_PMU_TLK_EVENT_ATTR(vc3_con, 0x33),
+ CVM_PMU_TLK_EVENT_ATTR(vc4_con, 0x34),
+ CVM_PMU_TLK_EVENT_ATTR(vc5_con, 0x35),
+ CVM_PMU_TLK_EVENT_ATTR(vc6_con, 0x36),
+ CVM_PMU_TLK_EVENT_ATTR(vc7_con, 0x37),
+ CVM_PMU_TLK_EVENT_ATTR(vc8_con, 0x38),
+ CVM_PMU_TLK_EVENT_ATTR(vc9_con, 0x39),
+ CVM_PMU_TLK_EVENT_ATTR(vc10_con, 0x3a),
+ CVM_PMU_TLK_EVENT_ATTR(vc11_con, 0x3b),
+ CVM_PMU_TLK_EVENT_ATTR(vc12_con, 0x3c),
+ CVM_PMU_TLK_EVENT_ATTR(vc13_con, 0x3d),
+ NULL,
+};
+
+static struct attribute_group cvm_pmu_tlk_events_group = {
+ .name = "events",
+ .attrs = cvm_pmu_tlk_events_attr,
+};
+static const struct attribute_group *cvm_pmu_tlk_attr_groups[] = {
+ &cvm_pmu_attr_group,
+ &cvm_pmu_tlk_format_group,
+ &cvm_pmu_tlk_events_group,
+ NULL,
+};
+
+static bool cvm_pmu_tlk_event_valid(u64 config)
+{
+ if (config < ARRAY_SIZE(cvm_pmu_tlk_events_attr))
+ return true;
+
+ return false;
+}
+
+static void *cvm_pmu_tlk_probe_unit(struct pci_dev *pdev, void __iomem *regs,
+ int nr)
+{
+ struct cvm_pmu_dev *tlk;
+ int ret = -ENOMEM;
+ char name[12];
+
+ tlk = kzalloc(sizeof(*tlk), GFP_KERNEL);
+ if (!tlk)
+ goto fail_nomem;
+
+ memset(name, 0, 12);
+ snprintf(name, 12, "ocx_tlk%d", nr);
+
+ list_add(&tlk->entry, &cvm_pmu_tlks);
+
+ tlk->pdev = pdev;
+ tlk->map = regs + TLK_START_ADDR + nr * TLK_UNIT_OFFSET;
+ tlk->num_counters = ARRAY_SIZE(cvm_pmu_tlk_events_attr) - 1;
+ tlk->pmu = (struct pmu) {
+ .name = name,
+ .task_ctx_nr = perf_invalid_context,
+ .pmu_enable = cvm_pmu_tlk_enable_pmu,
+ .pmu_disable = cvm_pmu_tlk_disable_pmu,
+ .event_init = cvm_pmu_event_init,
+ .add = cvm_pmu_tlk_add,
+ .del = cvm_pmu_del,
+ .start = cvm_pmu_start,
+ .stop = cvm_pmu_stop,
+ .read = cvm_pmu_read,
+ .attr_groups = cvm_pmu_tlk_attr_groups,
+ };
+
+ cpuhp_state_add_instance_nocalls(CPUHP_AP_PERF_ARM_CVM_ONLINE,
+ &tlk->cpuhp_node);
+
+ /*
+ * perf PMU is CPU dependent so pick a random CPU and migrate away
+ * if it goes offline.
+ */
+ cpumask_set_cpu(smp_processor_id(), &tlk->active_mask);
+
+ ret = perf_pmu_register(&tlk->pmu, tlk->pmu.name, -1);
+ if (ret)
+ goto fail_hp;
+
+ tlk->event_valid = cvm_pmu_tlk_event_valid;
+ return tlk;
+
+fail_hp:
+ cpuhp_state_remove_instance(CPUHP_AP_PERF_ARM_CVM_ONLINE,
+ &tlk->cpuhp_node);
+ kfree(tlk);
+fail_nomem:
+ return ERR_PTR(ret);
+}
+
+static void *cvm_pmu_tlk_probe(struct pci_dev *pdev, void __iomem *regs)
+{
+ struct cvm_pmu_dev *tlk;
+ int i;
+
+ for (i = 0; i < TLK_NR_UNITS; i++) {
+ tlk = cvm_pmu_tlk_probe_unit(pdev, regs, i);
+ if (PTR_ERR(tlk))
+ continue;
+ dev_info(&pdev->dev, "Enabled %s PMU with %d counters\n",
+ tlk->pmu.name, tlk->num_counters);
+ }
+ return &cvm_pmu_tlks;
+}
+
+static void cvm_pmu_tlk_remove(struct pci_dev *pdev, void *pmu_data)
+{
+ struct list_head *l, *tmp;
+ struct cvm_pmu_dev *tlk;
+
+ list_for_each_safe(l, tmp, &cvm_pmu_tlks) {
+ tlk = list_entry(l, struct cvm_pmu_dev, entry);
+
+ list_del(&tlk->entry);
+ cpuhp_state_remove_instance(CPUHP_AP_PERF_ARM_CVM_ONLINE,
+ &tlk->cpuhp_node);
+ perf_pmu_unregister(&tlk->pmu);
+ kfree(tlk);
+ }
+}
+
void *cvm_pmu_probe(struct pci_dev *pdev, void __iomem *regs, int type)
{
switch (type) {
case CVM_PMU_LMC:
return cvm_pmu_lmc_probe(pdev, regs);
+ case CVM_PMU_TLK:
+ return cvm_pmu_tlk_probe(pdev, regs);
}
return NULL;
}
@@ -382,6 +600,8 @@ void cvm_pmu_remove(struct pci_dev *pdev, void *pmu_data, int type)
switch (type) {
case CVM_PMU_LMC:
return cvm_pmu_lmc_remove(pdev, pmu_data);
+ case CVM_PMU_TLK:
+ return cvm_pmu_tlk_remove(pdev, pmu_data);
}
}
EXPORT_SYMBOL_GPL(cvm_pmu_remove);
@@ -389,6 +609,7 @@ EXPORT_SYMBOL_GPL(cvm_pmu_remove);
static int __init cvm_pmu_init(void)
{
INIT_LIST_HEAD(&cvm_pmu_lmcs);
+ INIT_LIST_HEAD(&cvm_pmu_tlks);

return cpuhp_setup_state_multi(CPUHP_AP_PERF_ARM_CVM_ONLINE,
"perf/arm/cvm:online", NULL,
@@ -397,7 +618,7 @@ static int __init cvm_pmu_init(void)

static void __exit cvm_pmu_exit(void)
{
- if (list_empty(&cvm_pmu_lmcs))
+ if (list_empty(&cvm_pmu_lmcs) && list_empty(&cvm_pmu_tlks))
cpuhp_remove_state(CPUHP_AP_PERF_ARM_CVM_ONLINE);
}

diff --git a/include/linux/perf/cavium_pmu.h b/include/linux/perf/cavium_pmu.h
index 7b13346..78f4d57 100644
--- a/include/linux/perf/cavium_pmu.h
+++ b/include/linux/perf/cavium_pmu.h
@@ -9,6 +9,7 @@

enum cvm_pmu_type {
CVM_PMU_LMC,
+ CVM_PMU_TLK,
};

/* maximum number of parallel hardware counters for all pmu types */
--
2.9.0.rc0.21.g7777322
\
 
 \ /
  Last update: 2017-05-17 10:32    [W:0.115 / U:0.704 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site