krd/linux
1
0
Fork 0
Code Activity

add 3rd party/custom patches

Browse Source 
3rd patchs (in alphabetical order):
- bbr3
- ntsync5
- openwrt
- pf-kernel
- xanmod
- zen

no configuration changes for now
This commit is contained in:
Konstantin Demin
2024-10-29 05:12:06 +03:00
parent 8082dfeaca
commit 8cbaf1dea2
186 changed files with 43626 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

180
debian/patches/patchset-pf/cpuidle/0001-cpuidle-menu-Remove-iowait-influence.patch vendored Normal file
View File

@@ -0,0 +1,180 @@
From d31e903a364802c068ff23bdd448cc70eda71a7c Mon Sep 17 00:00:00 2001
From: Christian Loehle <christian.loehle@arm.com>
Date: Thu, 5 Sep 2024 10:26:38 +0100
Subject: cpuidle: menu: Remove iowait influence
Remove CPU iowaiters influence on idle state selection.
Remove the menu notion of performance multiplier which increased with
the number of tasks that went to iowait sleep on this CPU and haven't
woken up yet.
Relying on iowait for cpuidle is problematic for a few reasons:
1. There is no guarantee that an iowaiting task will wake up on the
same CPU.
2. The task being in iowait says nothing about the idle duration, we
could be selecting shallower states for a long time.
3. The task being in iowait doesn't always imply a performance hit
with increased latency.
4. If there is such a performance hit, the number of iowaiting tasks
doesn't directly correlate.
5. The definition of iowait altogether is vague at best, it is
sprinkled across kernel code.
Signed-off-by: Christian Loehle <christian.loehle@arm.com>
---
drivers/cpuidle/governors/menu.c | 76 ++++----------------------------
1 file changed, 9 insertions(+), 67 deletions(-)
--- a/drivers/cpuidle/governors/menu.c
+++ b/drivers/cpuidle/governors/menu.c
@@ -19,7 +19,7 @@
#include "gov.h"
-#define BUCKETS 12
+#define BUCKETS 6
#define INTERVAL_SHIFT 3
#define INTERVALS (1UL << INTERVAL_SHIFT)
#define RESOLUTION 1024
@@ -29,12 +29,11 @@
/*
* Concepts and ideas behind the menu governor
*
- * For the menu governor, there are 3 decision factors for picking a C
+ * For the menu governor, there are 2 decision factors for picking a C
* state:
* 1) Energy break even point
- * 2) Performance impact
- * 3) Latency tolerance (from pmqos infrastructure)
- * These three factors are treated independently.
+ * 2) Latency tolerance (from pmqos infrastructure)
+ * These two factors are treated independently.
*
* Energy break even point
* -----------------------
@@ -75,30 +74,6 @@
* intervals and if the stand deviation of these 8 intervals is below a
* threshold value, we use the average of these intervals as prediction.
*
- * Limiting Performance Impact
- * ---------------------------
- * C states, especially those with large exit latencies, can have a real
- * noticeable impact on workloads, which is not acceptable for most sysadmins,
- * and in addition, less performance has a power price of its own.
- *
- * As a general rule of thumb, menu assumes that the following heuristic
- * holds:
- * The busier the system, the less impact of C states is acceptable
- *
- * This rule-of-thumb is implemented using a performance-multiplier:
- * If the exit latency times the performance multiplier is longer than
- * the predicted duration, the C state is not considered a candidate
- * for selection due to a too high performance impact. So the higher
- * this multiplier is, the longer we need to be idle to pick a deep C
- * state, and thus the less likely a busy CPU will hit such a deep
- * C state.
- *
- * Currently there is only one value determining the factor:
- * 10 points are added for each process that is waiting for IO on this CPU.
- * (This value was experimentally determined.)
- * Utilization is no longer a factor as it was shown that it never contributed
- * significantly to the performance multiplier in the first place.
- *
*/
struct menu_device {
@@ -112,19 +87,10 @@ struct menu_device {
int interval_ptr;
};
-static inline int which_bucket(u64 duration_ns, unsigned int nr_iowaiters)
+static inline int which_bucket(u64 duration_ns)
{
int bucket = 0;
- /*
- * We keep two groups of stats; one with no
- * IO pending, one without.
- * This allows us to calculate
- * E(duration)|iowait
- */
- if (nr_iowaiters)
- bucket = BUCKETS/2;
-
if (duration_ns < 10ULL * NSEC_PER_USEC)
return bucket;
if (duration_ns < 100ULL * NSEC_PER_USEC)
@@ -138,19 +104,6 @@ static inline int which_bucket(u64 durat
return bucket + 5;
}
-/*
- * Return a multiplier for the exit latency that is intended
- * to take performance requirements into account.
- * The more performance critical we estimate the system
- * to be, the higher this multiplier, and thus the higher
- * the barrier to go to an expensive C state.
- */
-static inline int performance_multiplier(unsigned int nr_iowaiters)
-{
- /* for IO wait tasks (per cpu!) we add 10x each */
- return 1 + 10 * nr_iowaiters;
-}
-
static DEFINE_PER_CPU(struct menu_device, menu_devices);
static void menu_update(struct cpuidle_driver *drv, struct cpuidle_device *dev);
@@ -258,8 +211,6 @@ static int menu_select(struct cpuidle_dr
struct menu_device *data = this_cpu_ptr(&menu_devices);
s64 latency_req = cpuidle_governor_latency_req(dev->cpu);
u64 predicted_ns;
- u64 interactivity_req;
- unsigned int nr_iowaiters;
ktime_t delta, delta_tick;
int i, idx;
@@ -268,8 +219,6 @@ static int menu_select(struct cpuidle_dr
data->needs_update = 0;
}
- nr_iowaiters = nr_iowait_cpu(dev->cpu);
-
/* Find the shortest expected idle interval. */
predicted_ns = get_typical_interval(data) * NSEC_PER_USEC;
if (predicted_ns > RESIDENCY_THRESHOLD_NS) {
@@ -283,7 +232,7 @@ static int menu_select(struct cpuidle_dr
}
data->next_timer_ns = delta;
- data->bucket = which_bucket(data->next_timer_ns, nr_iowaiters);
+ data->bucket = which_bucket(data->next_timer_ns);
/* Round up the result for half microseconds. */
timer_us = div_u64((RESOLUTION * DECAY * NSEC_PER_USEC) / 2 +
@@ -301,7 +250,7 @@ static int menu_select(struct cpuidle_dr
*/
data->next_timer_ns = KTIME_MAX;
delta_tick = TICK_NSEC / 2;
- data->bucket = which_bucket(KTIME_MAX, nr_iowaiters);
+ data->bucket = which_bucket(KTIME_MAX);
}
if (unlikely(drv->state_count <= 1 || latency_req == 0) ||
@@ -328,15 +277,8 @@ static int menu_select(struct cpuidle_dr
*/
if (predicted_ns < TICK_NSEC)
predicted_ns = data->next_timer_ns;
- } else {
- /*
- * Use the performance multiplier and the user-configurable
- * latency_req to determine the maximum exit latency.
- */
- interactivity_req = div64_u64(predicted_ns,
- performance_multiplier(nr_iowaiters));
- if (latency_req > interactivity_req)
- latency_req = interactivity_req;
+ } else if (latency_req > predicted_ns) {
+ latency_req = predicted_ns;
}
/*

58
debian/patches/patchset-pf/cpuidle/0002-cpuidle-Prefer-teo-over-menu-governor.patch vendored Normal file
View File

@@ -0,0 +1,58 @@
From 3f840a42780323a4437dd1a417488d141c33af15 Mon Sep 17 00:00:00 2001
From: Christian Loehle <christian.loehle@arm.com>
Date: Thu, 5 Sep 2024 10:26:39 +0100
Subject: cpuidle: Prefer teo over menu governor
Since menu no longer has the interactivity boost teo works better
overall, so make it the default.
Signed-off-by: Christian Loehle <christian.loehle@arm.com>
---
drivers/cpuidle/Kconfig | 5 +----
drivers/cpuidle/governors/menu.c | 2 +-
drivers/cpuidle/governors/teo.c | 2 +-
3 files changed, 3 insertions(+), 6 deletions(-)
--- a/drivers/cpuidle/Kconfig
+++ b/drivers/cpuidle/Kconfig
@@ -5,7 +5,7 @@ config CPU_IDLE
bool "CPU idle PM support"
default y if ACPI || PPC_PSERIES
select CPU_IDLE_GOV_LADDER if (!NO_HZ && !NO_HZ_IDLE)
- select CPU_IDLE_GOV_MENU if (NO_HZ || NO_HZ_IDLE) && !CPU_IDLE_GOV_TEO
+ select CPU_IDLE_GOV_TEO if (NO_HZ || NO_HZ_IDLE) && !CPU_IDLE_GOV_MENU
help
CPU idle is a generic framework for supporting software-controlled
idle processor power management. It includes modular cross-platform
@@ -30,9 +30,6 @@ config CPU_IDLE_GOV_TEO
This governor implements a simplified idle state selection method
focused on timer events and does not do any interactivity boosting.
- Some workloads benefit from using it and it generally should be safe
- to use. Say Y here if you are not happy with the alternatives.
-
config CPU_IDLE_GOV_HALTPOLL
bool "Haltpoll governor (for virtualized systems)"
depends on KVM_GUEST
--- a/drivers/cpuidle/governors/menu.c
+++ b/drivers/cpuidle/governors/menu.c
@@ -508,7 +508,7 @@ static int menu_enable_device(struct cpu
static struct cpuidle_governor menu_governor = {
.name = "menu",
- .rating = 20,
+ .rating = 19,
.enable = menu_enable_device,
.select = menu_select,
.reflect = menu_reflect,
--- a/drivers/cpuidle/governors/teo.c
+++ b/drivers/cpuidle/governors/teo.c
@@ -537,7 +537,7 @@ static int teo_enable_device(struct cpui
static struct cpuidle_governor teo_governor = {
.name = "teo",
- .rating = 19,
+ .rating = 20,
.enable = teo_enable_device,
.select = teo_select,
.reflect = teo_reflect,

39
debian/patches/patchset-pf/cpuidle/0003-TEST-cpufreq-schedutil-Linear-iowait-boost-step.patch vendored Normal file
View File

@@ -0,0 +1,39 @@
From ca8c9368b6f28ef625716b03aa930acfb8afe158 Mon Sep 17 00:00:00 2001
From: Christian Loehle <christian.loehle@arm.com>
Date: Thu, 5 Sep 2024 10:26:40 +0100
Subject: TEST: cpufreq/schedutil: Linear iowait boost step
In preparation for capping iowait boost make the steps linear as
opposed to doubling.
Signed-off-by: Christian Loehle <christian.loehle@arm.com>
---
kernel/sched/cpufreq_schedutil.c | 9 ++++-----
1 file changed, 4 insertions(+), 5 deletions(-)
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -267,7 +267,8 @@ static void sugov_iowait_boost(struct su
/* Double the boost at each request */
if (sg_cpu->iowait_boost) {
sg_cpu->iowait_boost =
- min_t(unsigned int, sg_cpu->iowait_boost << 1, SCHED_CAPACITY_SCALE);
+ min_t(unsigned int,
+ sg_cpu->iowait_boost + IOWAIT_BOOST_MIN, SCHED_CAPACITY_SCALE);
return;
}
@@ -308,11 +309,9 @@ static unsigned long sugov_iowait_apply(
/*
* No boost pending; reduce the boost value.
*/
- sg_cpu->iowait_boost >>= 1;
- if (sg_cpu->iowait_boost < IOWAIT_BOOST_MIN) {
- sg_cpu->iowait_boost = 0;
+ sg_cpu->iowait_boost -= IOWAIT_BOOST_MIN;
+ if (!sg_cpu->iowait_boost)
return 0;
- }
}
sg_cpu->iowait_boost_pending = false;

106
debian/patches/patchset-pf/cpuidle/0004-TEST-cpufreq-schedutil-iowait-boost-cap-sysfs.patch vendored Normal file
View File

@@ -0,0 +1,106 @@
From 33f05bd16a4ac2f6f36c9eb88016e2375dcb597c Mon Sep 17 00:00:00 2001
From: Christian Loehle <christian.loehle@arm.com>
Date: Thu, 5 Sep 2024 10:26:41 +0100
Subject: TEST: cpufreq/schedutil: iowait boost cap sysfs
Add a knob to cap applied iowait_boost per sysfs.
This is to test for potential regressions.
Signed-off-by: Christian Loehle <christian.loehle@arm.com>
---
kernel/sched/cpufreq_schedutil.c | 38 ++++++++++++++++++++++++++++++++
1 file changed, 38 insertions(+)
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -11,6 +11,7 @@
struct sugov_tunables {
struct gov_attr_set attr_set;
unsigned int rate_limit_us;
+ unsigned int iowait_boost_cap;
};
struct sugov_policy {
@@ -35,6 +36,8 @@ struct sugov_policy {
bool limits_changed;
bool need_freq_update;
+
+ unsigned int iowait_boost_cap;
};
struct sugov_cpu {
@@ -316,6 +319,9 @@ static unsigned long sugov_iowait_apply(
sg_cpu->iowait_boost_pending = false;
+ if (sg_cpu->iowait_boost > sg_cpu->sg_policy->iowait_boost_cap)
+ sg_cpu->iowait_boost = sg_cpu->sg_policy->iowait_boost_cap;
+
/*
* sg_cpu->util is already in capacity scale; convert iowait_boost
* into the same scale so we can compare.
@@ -554,6 +560,14 @@ static ssize_t rate_limit_us_show(struct
return sprintf(buf, "%u\n", tunables->rate_limit_us);
}
+
+static ssize_t iowait_boost_cap_show(struct gov_attr_set *attr_set, char *buf)
+{
+ struct sugov_tunables *tunables = to_sugov_tunables(attr_set);
+
+ return sprintf(buf, "%u\n", tunables->iowait_boost_cap);
+}
+
static ssize_t
rate_limit_us_store(struct gov_attr_set *attr_set, const char *buf, size_t count)
{
@@ -572,10 +586,30 @@ rate_limit_us_store(struct gov_attr_set
return count;
}
+static ssize_t
+iowait_boost_cap_store(struct gov_attr_set *attr_set, const char *buf, size_t count)
+{
+ struct sugov_tunables *tunables = to_sugov_tunables(attr_set);
+ struct sugov_policy *sg_policy;
+ unsigned int iowait_boost_cap;
+
+ if (kstrtouint(buf, 10, &iowait_boost_cap))
+ return -EINVAL;
+
+ tunables->iowait_boost_cap = iowait_boost_cap;
+
+ list_for_each_entry(sg_policy, &attr_set->policy_list, tunables_hook)
+ sg_policy->iowait_boost_cap = iowait_boost_cap;
+
+ return count;
+}
+
static struct governor_attr rate_limit_us = __ATTR_RW(rate_limit_us);
+static struct governor_attr iowait_boost_cap = __ATTR_RW(iowait_boost_cap);
static struct attribute *sugov_attrs[] = {
&rate_limit_us.attr,
+ &iowait_boost_cap.attr,
NULL
};
ATTRIBUTE_GROUPS(sugov);
@@ -765,6 +799,8 @@ static int sugov_init(struct cpufreq_pol
tunables->rate_limit_us = cpufreq_policy_transition_delay_us(policy);
+ tunables->iowait_boost_cap = SCHED_CAPACITY_SCALE;
+
policy->governor_data = sg_policy;
sg_policy->tunables = tunables;
@@ -834,6 +870,8 @@ static int sugov_start(struct cpufreq_po
sg_policy->limits_changed = false;
sg_policy->cached_raw_freq = 0;
+ sg_policy->iowait_boost_cap = SCHED_CAPACITY_SCALE;
+
sg_policy->need_freq_update = cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS);
if (policy_is_shared(policy))

325
debian/patches/patchset-pf/cpuidle/0005-cpufreq-schedutil-Remove-iowait-boost.patch vendored Normal file
View File

@@ -0,0 +1,325 @@
From 33eb6c08d7c615fad308001921c7b1148cbccfde Mon Sep 17 00:00:00 2001
From: Christian Loehle <christian.loehle@arm.com>
Date: Thu, 5 Sep 2024 10:26:42 +0100
Subject: cpufreq/schedutil: Remove iowait boost
iowait boost in schedutil was introduced by
commit ("21ca6d2c52f8 cpufreq: schedutil: Add iowait boosting").
with it more or less following intel_pstate's approach to increase
frequency after an iowait wakeup.
Behaviour that is piggy-backed onto iowait boost is problematic
due to a lot of reasons, so remove it.
For schedutil specifically these are some of the reasons:
1. Boosting is applied even in scenarios where it doesn't improve
throughput.
2. The boost is not accounted for in EAS: a) feec() will only consider
the actual task utilization for task placement, but another CPU might
be more energy-efficient at that capacity than the boosted one.)
b) When placing a non-IO task while a CPU is boosted compute_energy()
assumes a lower OPP than what is actually applied. This leads to
wrong EAS decisions.
3. Actual IO heavy workloads are hardly distinguished from infrequent
in_iowait wakeups.
4. The boost isn't accounted for in task placement.
5. The boost isn't associated with a task, it therefore lingers on the
rq even after the responsible task has migrated / stopped.
6. The boost isn't associated with a task, it therefore needs to ramp
up again when migrated.
7. Since schedutil doesn't know which task is getting woken up,
multiple unrelated in_iowait tasks lead to boosting.
8. Boosting is hard to control with UCLAMP_MAX (which is only active
when the task is on the rq, which for boosted tasks is usually not
the case for most of the time).
One benefit of schedutil specifically is the reliance on the
scheduler's utilization signals, which have evolved a lot since it's
original introduction. Some cases that benefitted from iowait boosting
in the past can now be covered by e.g. util_est.
Signed-off-by: Christian Loehle <christian.loehle@arm.com>
---
kernel/sched/cpufreq_schedutil.c | 181 +------------------------------
1 file changed, 3 insertions(+), 178 deletions(-)
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -6,12 +6,9 @@
* Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
*/
-#define IOWAIT_BOOST_MIN (SCHED_CAPACITY_SCALE / 8)
-
struct sugov_tunables {
struct gov_attr_set attr_set;
unsigned int rate_limit_us;
- unsigned int iowait_boost_cap;
};
struct sugov_policy {
@@ -36,8 +33,6 @@ struct sugov_policy {
bool limits_changed;
bool need_freq_update;
-
- unsigned int iowait_boost_cap;
};
struct sugov_cpu {
@@ -45,10 +40,6 @@ struct sugov_cpu {
struct sugov_policy *sg_policy;
unsigned int cpu;
- bool iowait_boost_pending;
- unsigned int iowait_boost;
- u64 last_update;
-
unsigned long util;
unsigned long bw_min;
@@ -198,137 +189,15 @@ unsigned long sugov_effective_cpu_perf(i
return max(min, max);
}
-static void sugov_get_util(struct sugov_cpu *sg_cpu, unsigned long boost)
+static void sugov_get_util(struct sugov_cpu *sg_cpu)
{
unsigned long min, max, util = cpu_util_cfs_boost(sg_cpu->cpu);
util = effective_cpu_util(sg_cpu->cpu, util, &min, &max);
- util = max(util, boost);
sg_cpu->bw_min = min;
sg_cpu->util = sugov_effective_cpu_perf(sg_cpu->cpu, util, min, max);
}
-/**
- * sugov_iowait_reset() - Reset the IO boost status of a CPU.
- * @sg_cpu: the sugov data for the CPU to boost
- * @time: the update time from the caller
- * @set_iowait_boost: true if an IO boost has been requested
- *
- * The IO wait boost of a task is disabled after a tick since the last update
- * of a CPU. If a new IO wait boost is requested after more then a tick, then
- * we enable the boost starting from IOWAIT_BOOST_MIN, which improves energy
- * efficiency by ignoring sporadic wakeups from IO.
- */
-static bool sugov_iowait_reset(struct sugov_cpu *sg_cpu, u64 time,
- bool set_iowait_boost)
-{
- s64 delta_ns = time - sg_cpu->last_update;
-
- /* Reset boost only if a tick has elapsed since last request */
- if (delta_ns <= TICK_NSEC)
- return false;
-
- sg_cpu->iowait_boost = set_iowait_boost ? IOWAIT_BOOST_MIN : 0;
- sg_cpu->iowait_boost_pending = set_iowait_boost;
-
- return true;
-}
-
-/**
- * sugov_iowait_boost() - Updates the IO boost status of a CPU.
- * @sg_cpu: the sugov data for the CPU to boost
- * @time: the update time from the caller
- * @flags: SCHED_CPUFREQ_IOWAIT if the task is waking up after an IO wait
- *
- * Each time a task wakes up after an IO operation, the CPU utilization can be
- * boosted to a certain utilization which doubles at each "frequent and
- * successive" wakeup from IO, ranging from IOWAIT_BOOST_MIN to the utilization
- * of the maximum OPP.
- *
- * To keep doubling, an IO boost has to be requested at least once per tick,
- * otherwise we restart from the utilization of the minimum OPP.
- */
-static void sugov_iowait_boost(struct sugov_cpu *sg_cpu, u64 time,
- unsigned int flags)
-{
- bool set_iowait_boost = flags & SCHED_CPUFREQ_IOWAIT;
-
- /* Reset boost if the CPU appears to have been idle enough */
- if (sg_cpu->iowait_boost &&
- sugov_iowait_reset(sg_cpu, time, set_iowait_boost))
- return;
-
- /* Boost only tasks waking up after IO */
- if (!set_iowait_boost)
- return;
-
- /* Ensure boost doubles only one time at each request */
- if (sg_cpu->iowait_boost_pending)
- return;
- sg_cpu->iowait_boost_pending = true;
-
- /* Double the boost at each request */
- if (sg_cpu->iowait_boost) {
- sg_cpu->iowait_boost =
- min_t(unsigned int,
- sg_cpu->iowait_boost + IOWAIT_BOOST_MIN, SCHED_CAPACITY_SCALE);
- return;
- }
-
- /* First wakeup after IO: start with minimum boost */
- sg_cpu->iowait_boost = IOWAIT_BOOST_MIN;
-}
-
-/**
- * sugov_iowait_apply() - Apply the IO boost to a CPU.
- * @sg_cpu: the sugov data for the cpu to boost
- * @time: the update time from the caller
- * @max_cap: the max CPU capacity
- *
- * A CPU running a task which woken up after an IO operation can have its
- * utilization boosted to speed up the completion of those IO operations.
- * The IO boost value is increased each time a task wakes up from IO, in
- * sugov_iowait_apply(), and it's instead decreased by this function,
- * each time an increase has not been requested (!iowait_boost_pending).
- *
- * A CPU which also appears to have been idle for at least one tick has also
- * its IO boost utilization reset.
- *
- * This mechanism is designed to boost high frequently IO waiting tasks, while
- * being more conservative on tasks which does sporadic IO operations.
- */
-static unsigned long sugov_iowait_apply(struct sugov_cpu *sg_cpu, u64 time,
- unsigned long max_cap)
-{
- /* No boost currently required */
- if (!sg_cpu->iowait_boost)
- return 0;
-
- /* Reset boost if the CPU appears to have been idle enough */
- if (sugov_iowait_reset(sg_cpu, time, false))
- return 0;
-
- if (!sg_cpu->iowait_boost_pending) {
- /*
- * No boost pending; reduce the boost value.
- */
- sg_cpu->iowait_boost -= IOWAIT_BOOST_MIN;
- if (!sg_cpu->iowait_boost)
- return 0;
- }
-
- sg_cpu->iowait_boost_pending = false;
-
- if (sg_cpu->iowait_boost > sg_cpu->sg_policy->iowait_boost_cap)
- sg_cpu->iowait_boost = sg_cpu->sg_policy->iowait_boost_cap;
-
- /*
- * sg_cpu->util is already in capacity scale; convert iowait_boost
- * into the same scale so we can compare.
- */
- return (sg_cpu->iowait_boost * max_cap) >> SCHED_CAPACITY_SHIFT;
-}
-
#ifdef CONFIG_NO_HZ_COMMON
static bool sugov_cpu_is_busy(struct sugov_cpu *sg_cpu)
{
@@ -356,18 +225,12 @@ static inline bool sugov_update_single_c
u64 time, unsigned long max_cap,
unsigned int flags)
{
- unsigned long boost;
-
- sugov_iowait_boost(sg_cpu, time, flags);
- sg_cpu->last_update = time;
-
ignore_dl_rate_limit(sg_cpu);
if (!sugov_should_update_freq(sg_cpu->sg_policy, time))
return false;
- boost = sugov_iowait_apply(sg_cpu, time, max_cap);
- sugov_get_util(sg_cpu, boost);
+ sugov_get_util(sg_cpu);
return true;
}
@@ -468,11 +331,8 @@ static unsigned int sugov_next_freq_shar
for_each_cpu(j, policy->cpus) {
struct sugov_cpu *j_sg_cpu = &per_cpu(sugov_cpu, j);
- unsigned long boost;
-
- boost = sugov_iowait_apply(j_sg_cpu, time, max_cap);
- sugov_get_util(j_sg_cpu, boost);
+ sugov_get_util(j_sg_cpu);
util = max(j_sg_cpu->util, util);
}
@@ -488,9 +348,6 @@ sugov_update_shared(struct update_util_d
raw_spin_lock(&sg_policy->update_lock);
- sugov_iowait_boost(sg_cpu, time, flags);
- sg_cpu->last_update = time;
-
ignore_dl_rate_limit(sg_cpu);
if (sugov_should_update_freq(sg_policy, time)) {
@@ -560,14 +417,6 @@ static ssize_t rate_limit_us_show(struct
return sprintf(buf, "%u\n", tunables->rate_limit_us);
}
-
-static ssize_t iowait_boost_cap_show(struct gov_attr_set *attr_set, char *buf)
-{
- struct sugov_tunables *tunables = to_sugov_tunables(attr_set);
-
- return sprintf(buf, "%u\n", tunables->iowait_boost_cap);
-}
-
static ssize_t
rate_limit_us_store(struct gov_attr_set *attr_set, const char *buf, size_t count)
{
@@ -586,30 +435,10 @@ rate_limit_us_store(struct gov_attr_set
return count;
}
-static ssize_t
-iowait_boost_cap_store(struct gov_attr_set *attr_set, const char *buf, size_t count)
-{
- struct sugov_tunables *tunables = to_sugov_tunables(attr_set);
- struct sugov_policy *sg_policy;
- unsigned int iowait_boost_cap;
-
- if (kstrtouint(buf, 10, &iowait_boost_cap))
- return -EINVAL;
-
- tunables->iowait_boost_cap = iowait_boost_cap;
-
- list_for_each_entry(sg_policy, &attr_set->policy_list, tunables_hook)
- sg_policy->iowait_boost_cap = iowait_boost_cap;
-
- return count;
-}
-
static struct governor_attr rate_limit_us = __ATTR_RW(rate_limit_us);
-static struct governor_attr iowait_boost_cap = __ATTR_RW(iowait_boost_cap);
static struct attribute *sugov_attrs[] = {
&rate_limit_us.attr,
- &iowait_boost_cap.attr,
NULL
};
ATTRIBUTE_GROUPS(sugov);
@@ -799,8 +628,6 @@ static int sugov_init(struct cpufreq_pol
tunables->rate_limit_us = cpufreq_policy_transition_delay_us(policy);
- tunables->iowait_boost_cap = SCHED_CAPACITY_SCALE;
-
policy->governor_data = sg_policy;
sg_policy->tunables = tunables;
@@ -870,8 +697,6 @@ static int sugov_start(struct cpufreq_po
sg_policy->limits_changed = false;
sg_policy->cached_raw_freq = 0;
- sg_policy->iowait_boost_cap = SCHED_CAPACITY_SCALE;
-
sg_policy->need_freq_update = cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS);
if (policy_is_shared(policy))

113
debian/patches/patchset-pf/cpuidle/0006-cpufreq-intel_pstate-Remove-iowait-boost.patch vendored Normal file
View File

@@ -0,0 +1,113 @@
From af7bbb59c2411e985a5d79173af5686337b4af9b Mon Sep 17 00:00:00 2001
From: Christian Loehle <christian.loehle@arm.com>
Date: Thu, 5 Sep 2024 10:26:43 +0100
Subject: cpufreq: intel_pstate: Remove iowait boost
Analogous to schedutil, remove iowait boost for the same reasons.
Signed-off-by: Christian Loehle <christian.loehle@arm.com>
---
drivers/cpufreq/intel_pstate.c | 50 ++--------------------------------
1 file changed, 3 insertions(+), 47 deletions(-)
--- a/drivers/cpufreq/intel_pstate.c
+++ b/drivers/cpufreq/intel_pstate.c
@@ -191,7 +191,6 @@ struct global_params {
* @policy: CPUFreq policy value
* @update_util: CPUFreq utility callback information
* @update_util_set: CPUFreq utility callback is set
- * @iowait_boost: iowait-related boost fraction
* @last_update: Time of the last update.
* @pstate: Stores P state limits for this CPU
* @vid: Stores VID limits for this CPU
@@ -245,7 +244,6 @@ struct cpudata {
struct acpi_processor_performance acpi_perf_data;
bool valid_pss_table;
#endif
- unsigned int iowait_boost;
s16 epp_powersave;
s16 epp_policy;
s16 epp_default;
@@ -2136,28 +2134,7 @@ static inline void intel_pstate_update_u
{
cpu->sample.time = time;
- if (cpu->sched_flags & SCHED_CPUFREQ_IOWAIT) {
- bool do_io = false;
-
- cpu->sched_flags = 0;
- /*
- * Set iowait_boost flag and update time. Since IO WAIT flag
- * is set all the time, we can't just conclude that there is
- * some IO bound activity is scheduled on this CPU with just
- * one occurrence. If we receive at least two in two
- * consecutive ticks, then we treat as boost candidate.
- */
- if (time_before64(time, cpu->last_io_update + 2 * TICK_NSEC))
- do_io = true;
-
- cpu->last_io_update = time;
-
- if (do_io)
- intel_pstate_hwp_boost_up(cpu);
-
- } else {
- intel_pstate_hwp_boost_down(cpu);
- }
+ intel_pstate_hwp_boost_down(cpu);
}
static inline void intel_pstate_update_util_hwp(struct update_util_data *data,
@@ -2240,9 +2217,6 @@ static inline int32_t get_target_pstate(
busy_frac = div_fp(sample->mperf << cpu->aperf_mperf_shift,
sample->tsc);
- if (busy_frac < cpu->iowait_boost)
- busy_frac = cpu->iowait_boost;
-
sample->busy_scaled = busy_frac * 100;
target = READ_ONCE(global.no_turbo) ?
@@ -2303,7 +2277,7 @@ static void intel_pstate_adjust_pstate(s
sample->aperf,
sample->tsc,
get_avg_frequency(cpu),
- fp_toint(cpu->iowait_boost * 100));
+ 0);
}
static void intel_pstate_update_util(struct update_util_data *data, u64 time,
@@ -2317,24 +2291,6 @@ static void intel_pstate_update_util(str
return;
delta_ns = time - cpu->last_update;
- if (flags & SCHED_CPUFREQ_IOWAIT) {
- /* Start over if the CPU may have been idle. */
- if (delta_ns > TICK_NSEC) {
- cpu->iowait_boost = ONE_EIGHTH_FP;
- } else if (cpu->iowait_boost >= ONE_EIGHTH_FP) {
- cpu->iowait_boost <<= 1;
- if (cpu->iowait_boost > int_tofp(1))
- cpu->iowait_boost = int_tofp(1);
- } else {
- cpu->iowait_boost = ONE_EIGHTH_FP;
- }
- } else if (cpu->iowait_boost) {
- /* Clear iowait_boost if the CPU may have been idle. */
- if (delta_ns > TICK_NSEC)
- cpu->iowait_boost = 0;
- else
- cpu->iowait_boost >>= 1;
- }
cpu->last_update = time;
delta_ns = time - cpu->sample.time;
if ((s64)delta_ns < INTEL_PSTATE_SAMPLING_INTERVAL)
@@ -2832,7 +2788,7 @@ static void intel_cpufreq_trace(struct c
sample->aperf,
sample->tsc,
get_avg_frequency(cpu),
- fp_toint(cpu->iowait_boost * 100));
+ 0);
}
static void intel_cpufreq_hwp_update(struct cpudata *cpu, u32 min, u32 max,

42
debian/patches/patchset-pf/cpuidle/0007-cpufreq-Remove-SCHED_CPUFREQ_IOWAIT-update.patch vendored Normal file
View File

@@ -0,0 +1,42 @@
From fd1e0723b0a7ad140d2bf7cd9154997d5ece2b37 Mon Sep 17 00:00:00 2001
From: Christian Loehle <christian.loehle@arm.com>
Date: Thu, 5 Sep 2024 10:26:44 +0100
Subject: cpufreq: Remove SCHED_CPUFREQ_IOWAIT update
Neither intel_pstate nor schedutil care for the flag anymore, so
remove the update and flag definition.
Signed-off-by: Christian Loehle <christian.loehle@arm.com>
---
include/linux/sched/cpufreq.h | 2 --
kernel/sched/fair.c | 8 --------
2 files changed, 10 deletions(-)
--- a/include/linux/sched/cpufreq.h
+++ b/include/linux/sched/cpufreq.h
@@ -8,8 +8,6 @@
* Interface between cpufreq drivers and the scheduler:
*/
-#define SCHED_CPUFREQ_IOWAIT (1U << 0)
-
#ifdef CONFIG_CPU_FREQ
struct cpufreq_policy;
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6768,14 +6768,6 @@ enqueue_task_fair(struct rq *rq, struct
*/
util_est_enqueue(&rq->cfs, p);
- /*
- * If in_iowait is set, the code below may not trigger any cpufreq
- * utilization updates, so do it here explicitly with the IOWAIT flag
- * passed.
- */
- if (p->in_iowait)
- cpufreq_update_util(rq, SCHED_CPUFREQ_IOWAIT);
-
for_each_sched_entity(se) {
if (se->on_rq)
break;

55
debian/patches/patchset-pf/cpuidle/0008-io_uring-Do-not-set-iowait-before-sleeping.patch vendored Normal file
View File

@@ -0,0 +1,55 @@
From 30cdb8d7d06f51bb86142c537ea05bd01c31bb40 Mon Sep 17 00:00:00 2001
From: Christian Loehle <christian.loehle@arm.com>
Date: Thu, 5 Sep 2024 10:26:45 +0100
Subject: io_uring: Do not set iowait before sleeping
Setting in_iowait was introduced in commit
8a796565cec3 ("io_uring: Use io_schedule* in cqring wait")
to tackle a perf regression that was caused by menu taking iowait into
account for synchronous IO and thus not selecting deeper states like in
the io_uring counterpart.
That behaviour is gone, so the workaround can be removed.
Signed-off-by: Christian Loehle <christian.loehle@arm.com>
---
io_uring/io_uring.c | 17 -----------------
1 file changed, 17 deletions(-)
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -2359,15 +2359,6 @@ int io_run_task_work_sig(struct io_ring_
return 0;
}
-static bool current_pending_io(void)
-{
- struct io_uring_task *tctx = current->io_uring;
-
- if (!tctx)
- return false;
- return percpu_counter_read_positive(&tctx->inflight);
-}
-
/* when returns >0, the caller should retry */
static inline int io_cqring_wait_schedule(struct io_ring_ctx *ctx,
struct io_wait_queue *iowq)
@@ -2385,19 +2376,11 @@ static inline int io_cqring_wait_schedul
if (unlikely(io_should_wake(iowq)))
return 0;
- /*
- * Mark us as being in io_wait if we have pending requests, so cpufreq
- * can take into account that the task is waiting for IO - turns out
- * to be important for low QD IO.
- */
- if (current_pending_io())
- current->in_iowait = 1;
ret = 0;
if (iowq->timeout == KTIME_MAX)
schedule();
else if (!schedule_hrtimeout(&iowq->timeout, HRTIMER_MODE_ABS))
ret = -ETIME;
- current->in_iowait = 0;
return ret;
}