kernel-fxtec-pro1x/drivers/devfreq/governor_memlat.c
Santosh Mardi 4088fdc46d PM / devfreq: memlat: Add writeback percentage as a lat condition
Certain L3 workloads are latency sensitive to cache snoop traffic and
this traffic is not directly quantified in the existing memlat scheme.

Use L2 writeback percentage as a metric to identify snoop traffic to
improve memlat.

Change-Id: I9d43375d96de5a199c6a87c55e5c1079549b23ce
Signed-off-by: Santosh Mardi <gsantosh@codeaurora.org>
2020-06-17 10:12:10 -07:00

585 lines
13 KiB
C

// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright (c) 2015-2019, The Linux Foundation. All rights reserved.
*/
#define pr_fmt(fmt) "mem_lat: " fmt
#include <linux/kernel.h>
#include <linux/sizes.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/io.h>
#include <linux/delay.h>
#include <linux/ktime.h>
#include <linux/time.h>
#include <linux/err.h>
#include <linux/errno.h>
#include <linux/mutex.h>
#include <linux/interrupt.h>
#include <linux/platform_device.h>
#include <linux/device.h>
#include <linux/of.h>
#include <linux/devfreq.h>
#include "governor.h"
#include "governor_memlat.h"
#include <trace/events/power.h>
struct memlat_node {
unsigned int ratio_ceil;
unsigned int stall_floor;
unsigned int wb_pct_thres;
unsigned int wb_filter_ratio;
bool mon_started;
bool already_zero;
struct list_head list;
void *orig_data;
struct memlat_hwmon *hw;
struct devfreq_governor *gov;
struct attribute_group *attr_grp;
unsigned long resume_freq;
};
static LIST_HEAD(memlat_list);
static DEFINE_MUTEX(list_lock);
static int memlat_use_cnt;
static int compute_use_cnt;
static DEFINE_MUTEX(state_lock);
#define show_attr(name) \
static ssize_t show_##name(struct device *dev, \
struct device_attribute *attr, char *buf) \
{ \
struct devfreq *df = to_devfreq(dev); \
struct memlat_node *hw = df->data; \
return scnprintf(buf, PAGE_SIZE, "%u\n", hw->name); \
}
#define store_attr(name, _min, _max) \
static ssize_t store_##name(struct device *dev, \
struct device_attribute *attr, const char *buf, \
size_t count) \
{ \
struct devfreq *df = to_devfreq(dev); \
struct memlat_node *hw = df->data; \
int ret; \
unsigned int val; \
ret = kstrtouint(buf, 10, &val); \
if (ret) \
return ret; \
val = max(val, _min); \
val = min(val, _max); \
hw->name = val; \
return count; \
}
#define gov_attr(__attr, min, max) \
show_attr(__attr) \
store_attr(__attr, min, max) \
static DEVICE_ATTR(__attr, 0644, show_##__attr, store_##__attr)
static ssize_t freq_map_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
struct devfreq *df = to_devfreq(dev);
struct memlat_node *n = df->data;
struct core_dev_map *map = n->hw->freq_map;
unsigned int cnt = 0;
cnt += scnprintf(buf, PAGE_SIZE, "Core freq (MHz)\tDevice BW\n");
while (map->core_mhz && cnt < PAGE_SIZE) {
cnt += scnprintf(buf + cnt, PAGE_SIZE - cnt, "%15u\t%9u\n",
map->core_mhz, map->target_freq);
map++;
}
if (cnt < PAGE_SIZE)
cnt += scnprintf(buf + cnt, PAGE_SIZE - cnt, "\n");
return cnt;
}
static DEVICE_ATTR_RO(freq_map);
static unsigned long core_to_dev_freq(struct memlat_node *node,
unsigned long coref)
{
struct memlat_hwmon *hw = node->hw;
struct core_dev_map *map = hw->freq_map;
unsigned long freq = 0;
if (!map)
goto out;
while (map->core_mhz && map->core_mhz < coref)
map++;
if (!map->core_mhz)
map--;
freq = map->target_freq;
out:
pr_debug("freq: %lu -> dev: %lu\n", coref, freq);
return freq;
}
static struct memlat_node *find_memlat_node(struct devfreq *df)
{
struct memlat_node *node, *found = NULL;
mutex_lock(&list_lock);
list_for_each_entry(node, &memlat_list, list)
if (node->hw->dev == df->dev.parent ||
node->hw->of_node == df->dev.parent->of_node) {
found = node;
break;
}
mutex_unlock(&list_lock);
return found;
}
static int start_monitor(struct devfreq *df)
{
struct memlat_node *node = df->data;
struct memlat_hwmon *hw = node->hw;
struct device *dev = df->dev.parent;
int ret;
ret = hw->start_hwmon(hw);
if (ret) {
dev_err(dev, "Unable to start HW monitor! (%d)\n", ret);
return ret;
}
if (!hw->should_ignore_df_monitor)
devfreq_monitor_start(df);
node->mon_started = true;
return 0;
}
static void stop_monitor(struct devfreq *df)
{
struct memlat_node *node = df->data;
struct memlat_hwmon *hw = node->hw;
node->mon_started = false;
if (!hw->should_ignore_df_monitor)
devfreq_monitor_stop(df);
hw->stop_hwmon(hw);
}
static int gov_start(struct devfreq *df)
{
int ret = 0;
struct device *dev = df->dev.parent;
struct memlat_node *node;
struct memlat_hwmon *hw;
node = find_memlat_node(df);
if (!node) {
dev_err(dev, "Unable to find HW monitor!\n");
return -ENODEV;
}
hw = node->hw;
hw->df = df;
node->orig_data = df->data;
df->data = node;
ret = start_monitor(df);
if (ret)
goto err_start;
ret = sysfs_create_group(&df->dev.kobj, node->attr_grp);
if (ret)
goto err_sysfs;
return 0;
err_sysfs:
stop_monitor(df);
err_start:
df->data = node->orig_data;
node->orig_data = NULL;
hw->df = NULL;
return ret;
}
static int gov_suspend(struct devfreq *df)
{
struct memlat_node *node = df->data;
unsigned long prev_freq = df->previous_freq;
node->mon_started = false;
devfreq_monitor_suspend(df);
mutex_lock(&df->lock);
update_devfreq(df);
mutex_unlock(&df->lock);
node->resume_freq = max(prev_freq, 1UL);
return 0;
}
static int gov_resume(struct devfreq *df)
{
struct memlat_node *node = df->data;
mutex_lock(&df->lock);
update_devfreq(df);
mutex_unlock(&df->lock);
node->resume_freq = 0;
devfreq_monitor_resume(df);
node->mon_started = true;
return 0;
}
static void gov_stop(struct devfreq *df)
{
struct memlat_node *node = df->data;
struct memlat_hwmon *hw = node->hw;
sysfs_remove_group(&df->dev.kobj, node->attr_grp);
stop_monitor(df);
df->data = node->orig_data;
node->orig_data = NULL;
hw->df = NULL;
}
static int devfreq_memlat_get_freq(struct devfreq *df,
unsigned long *freq)
{
int i, lat_dev = 0;
struct memlat_node *node = df->data;
struct memlat_hwmon *hw = node->hw;
unsigned long max_freq = 0;
unsigned int ratio;
/*
* node->resume_freq is set to 0 at the end of resume (after the update)
* and is set to df->prev_freq at the end of suspend (after the update).
* This function will be called as part of the update_devfreq call in
* both scenarios. As a result, this block will cause a 0 vote during
* suspend and a vote for df->prev_freq during resume.
*/
if (!node->mon_started) {
*freq = node->resume_freq;
return 0;
}
hw->get_cnt(hw);
for (i = 0; i < hw->num_cores; i++) {
ratio = hw->core_stats[i].inst_count;
if (hw->core_stats[i].mem_count)
ratio /= hw->core_stats[i].mem_count;
if (!hw->core_stats[i].freq)
continue;
trace_memlat_dev_meas(dev_name(df->dev.parent),
hw->core_stats[i].id,
hw->core_stats[i].inst_count,
hw->core_stats[i].mem_count,
hw->core_stats[i].freq,
hw->core_stats[i].stall_pct,
hw->core_stats[i].wb_pct, ratio);
if (((ratio <= node->ratio_ceil
&& hw->core_stats[i].stall_pct >= node->stall_floor) ||
(hw->core_stats[i].wb_pct >= node->wb_pct_thres
&& ratio <= node->wb_filter_ratio))
&& (hw->core_stats[i].freq > max_freq)) {
lat_dev = i;
max_freq = hw->core_stats[i].freq;
}
}
if (max_freq)
max_freq = core_to_dev_freq(node, max_freq);
if (max_freq || !node->already_zero) {
trace_memlat_dev_update(dev_name(df->dev.parent),
hw->core_stats[lat_dev].id,
hw->core_stats[lat_dev].inst_count,
hw->core_stats[lat_dev].mem_count,
hw->core_stats[lat_dev].freq,
max_freq);
}
node->already_zero = !max_freq;
*freq = max_freq;
return 0;
}
gov_attr(ratio_ceil, 1U, 20000U);
gov_attr(stall_floor, 0U, 100U);
gov_attr(wb_pct_thres, 0U, 100U);
gov_attr(wb_filter_ratio, 0U, 50000U);
static struct attribute *memlat_dev_attr[] = {
&dev_attr_ratio_ceil.attr,
&dev_attr_stall_floor.attr,
&dev_attr_wb_pct_thres.attr,
&dev_attr_wb_filter_ratio.attr,
&dev_attr_freq_map.attr,
NULL,
};
static struct attribute *compute_dev_attr[] = {
&dev_attr_freq_map.attr,
NULL,
};
static struct attribute_group memlat_dev_attr_group = {
.name = "mem_latency",
.attrs = memlat_dev_attr,
};
static struct attribute_group compute_dev_attr_group = {
.name = "compute",
.attrs = compute_dev_attr,
};
#define MIN_MS 0U
#define MAX_MS 500U
static int devfreq_memlat_ev_handler(struct devfreq *df,
unsigned int event, void *data)
{
int ret;
unsigned int sample_ms;
struct memlat_node *node;
struct memlat_hwmon *hw;
switch (event) {
case DEVFREQ_GOV_START:
sample_ms = df->profile->polling_ms;
sample_ms = max(MIN_MS, sample_ms);
sample_ms = min(MAX_MS, sample_ms);
df->profile->polling_ms = sample_ms;
ret = gov_start(df);
if (ret)
return ret;
dev_dbg(df->dev.parent,
"Enabled Memory Latency governor\n");
break;
case DEVFREQ_GOV_STOP:
gov_stop(df);
dev_dbg(df->dev.parent,
"Disabled Memory Latency governor\n");
break;
case DEVFREQ_GOV_SUSPEND:
ret = gov_suspend(df);
if (ret) {
dev_err(df->dev.parent,
"Unable to suspend memlat governor (%d)\n",
ret);
return ret;
}
dev_dbg(df->dev.parent, "Suspended memlat governor\n");
break;
case DEVFREQ_GOV_RESUME:
ret = gov_resume(df);
if (ret) {
dev_err(df->dev.parent,
"Unable to resume memlat governor (%d)\n",
ret);
return ret;
}
dev_dbg(df->dev.parent, "Resumed memlat governor\n");
break;
case DEVFREQ_GOV_INTERVAL:
node = df->data;
hw = node->hw;
sample_ms = *(unsigned int *)data;
sample_ms = max(MIN_MS, sample_ms);
sample_ms = min(MAX_MS, sample_ms);
if (hw->request_update_ms)
hw->request_update_ms(hw, sample_ms);
if (!hw->should_ignore_df_monitor)
devfreq_interval_update(df, &sample_ms);
break;
}
return 0;
}
static struct devfreq_governor devfreq_gov_memlat = {
.name = "mem_latency",
.get_target_freq = devfreq_memlat_get_freq,
.event_handler = devfreq_memlat_ev_handler,
};
static struct devfreq_governor devfreq_gov_compute = {
.name = "compute",
.get_target_freq = devfreq_memlat_get_freq,
.event_handler = devfreq_memlat_ev_handler,
};
#define NUM_COLS 2
static struct core_dev_map *init_core_dev_map(struct device *dev,
struct device_node *of_node,
char *prop_name)
{
int len, nf, i, j;
u32 data;
struct core_dev_map *tbl;
int ret;
if (!of_node)
of_node = dev->of_node;
if (!of_find_property(of_node, prop_name, &len))
return NULL;
len /= sizeof(data);
if (len % NUM_COLS || len == 0)
return NULL;
nf = len / NUM_COLS;
tbl = devm_kzalloc(dev, (nf + 1) * sizeof(struct core_dev_map),
GFP_KERNEL);
if (!tbl)
return NULL;
for (i = 0, j = 0; i < nf; i++, j += 2) {
ret = of_property_read_u32_index(of_node, prop_name, j,
&data);
if (ret)
return NULL;
tbl[i].core_mhz = data / 1000;
ret = of_property_read_u32_index(of_node, prop_name, j + 1,
&data);
if (ret)
return NULL;
tbl[i].target_freq = data;
pr_debug("Entry%d CPU:%u, Dev:%u\n", i, tbl[i].core_mhz,
tbl[i].target_freq);
}
tbl[i].core_mhz = 0;
return tbl;
}
static struct memlat_node *register_common(struct device *dev,
struct memlat_hwmon *hw)
{
struct memlat_node *node;
struct device_node *of_child;
if (!hw->dev && !hw->of_node)
return ERR_PTR(-EINVAL);
node = devm_kzalloc(dev, sizeof(*node), GFP_KERNEL);
if (!node)
return ERR_PTR(-ENOMEM);
node->ratio_ceil = 10;
node->wb_pct_thres = 100;
node->wb_filter_ratio = 25000;
node->hw = hw;
if (hw->get_child_of_node) {
of_child = hw->get_child_of_node(dev);
hw->freq_map = init_core_dev_map(dev, of_child,
"qcom,core-dev-table");
} else {
hw->freq_map = init_core_dev_map(dev, NULL,
"qcom,core-dev-table");
}
if (!hw->freq_map) {
dev_err(dev, "Couldn't find the core-dev freq table!\n");
return ERR_PTR(-EINVAL);
}
mutex_lock(&list_lock);
list_add_tail(&node->list, &memlat_list);
mutex_unlock(&list_lock);
return node;
}
int register_compute(struct device *dev, struct memlat_hwmon *hw)
{
struct memlat_node *node;
int ret = 0;
node = register_common(dev, hw);
if (IS_ERR(node)) {
ret = PTR_ERR(node);
goto out;
}
mutex_lock(&state_lock);
node->gov = &devfreq_gov_compute;
node->attr_grp = &compute_dev_attr_group;
if (!compute_use_cnt)
ret = devfreq_add_governor(&devfreq_gov_compute);
if (!ret)
compute_use_cnt++;
mutex_unlock(&state_lock);
out:
if (!ret)
dev_info(dev, "Compute governor registered.\n");
else
dev_err(dev, "Compute governor registration failed!\n");
return ret;
}
int register_memlat(struct device *dev, struct memlat_hwmon *hw)
{
struct memlat_node *node;
int ret = 0;
node = register_common(dev, hw);
if (IS_ERR(node)) {
ret = PTR_ERR(node);
goto out;
}
mutex_lock(&state_lock);
node->gov = &devfreq_gov_memlat;
node->attr_grp = &memlat_dev_attr_group;
if (!memlat_use_cnt)
ret = devfreq_add_governor(&devfreq_gov_memlat);
if (!ret)
memlat_use_cnt++;
mutex_unlock(&state_lock);
out:
if (!ret)
dev_info(dev, "Memory Latency governor registered.\n");
else
dev_err(dev, "Memory Latency governor registration failed!\n");
return ret;
}
MODULE_DESCRIPTION("HW monitor based dev DDR bandwidth voting driver");
MODULE_LICENSE("GPL v2");