Compare commits
4 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
2b2ee8f3e6 | ||
|
1d2c0e5c66 | ||
|
|
ee6cc74e03 | ||
|
|
e9bc6b34e4 |
@@ -3,7 +3,7 @@
|
||||
# Linux distribution you are using, options are "Arch", "Ubuntu", "Debian", "Fedora", "Suse", "Gentoo", "Generic".
|
||||
# It is automatically set to "Arch" when using PKGBUILD.
|
||||
# If left empty, the script will prompt
|
||||
_distro=""
|
||||
_distro="Arch"
|
||||
|
||||
# Kernel Version - Options are "5.4", and from "5.7" to "5.19"
|
||||
# you can also set a specific kernel version, e.g. "6.0-rc4" or "5.10.51",
|
||||
@@ -46,7 +46,7 @@ CUSTOM_GCC_PATH=""
|
||||
CUSTOM_LLVM_PATH=""
|
||||
|
||||
# Set to true to bypass makepkg.conf and use all available threads for compilation. False will respect your makepkg.conf options.
|
||||
_force_all_threads="true"
|
||||
_force_all_threads="false"
|
||||
|
||||
# Set to true to prevent ccache from being used and set CONFIG_GCC_PLUGINS=y (which needs to be disabled for ccache to work properly)
|
||||
_noccache="false"
|
||||
@@ -60,10 +60,10 @@ _modprobeddb="false"
|
||||
_modprobeddb_db_path=~/.config/modprobed.db
|
||||
|
||||
# Set to "1" to call make menuconfig, "2" to call make nconfig, "3" to call make xconfig, before building the kernel. Set to false to disable and skip the prompt.
|
||||
_menunconfig=""
|
||||
_menunconfig="false"
|
||||
|
||||
# Set to true to generate a kernel config fragment from your changes in menuconfig/nconfig. Set to false to disable and skip the prompt.
|
||||
_diffconfig=""
|
||||
_diffconfig="false"
|
||||
|
||||
# Set to the file name where the generated config fragment should be written to. Only used if _diffconfig is active.
|
||||
_diffconfig_name=""
|
||||
@@ -97,11 +97,11 @@ _STRIP="true"
|
||||
# LEAVE AN EMPTY VALUE TO BE PROMPTED ABOUT FOLLOWING OPTIONS AT BUILD TIME
|
||||
|
||||
# CPU scheduler - Options are "upds" (TkG's Undead PDS), "pds", "bmq", "muqss", "cacule", "tt", "bore" or "cfs" (kernel's default)
|
||||
_cpusched=""
|
||||
_cpusched="pds"
|
||||
|
||||
# Compiler to use - Options are "gcc" or "llvm".
|
||||
# For advanced users.
|
||||
_compiler=""
|
||||
_compiler="gcc"
|
||||
|
||||
# Force the use of the LLVM Integrated Assembler whether using LLVM, LTO or not.
|
||||
# Set to "1" to enable.
|
||||
@@ -131,7 +131,7 @@ _preempt_rt_force=""
|
||||
# For BMQ: 0: No yield.
|
||||
# 1: Deboost and requeue task. (Default)
|
||||
# 2: Set rq skip task.
|
||||
_sched_yield_type=""
|
||||
_sched_yield_type="0"
|
||||
|
||||
# Round Robin interval is the longest duration two tasks with the same nice level will be delayed for. When CPU time is requested by a task, it receives a time slice equal
|
||||
# to the rr_interval in addition to a virtual deadline. When using yield_type 2, a low value can help offset the disadvantages of rescheduling a process that has yielded.
|
||||
@@ -139,7 +139,7 @@ _sched_yield_type=""
|
||||
# PDS default: 4ms"
|
||||
# BMQ default: 2ms"
|
||||
# Set to "1" for 2ms, "2" for 4ms, "3" for 6ms, "4" for 8ms, or "default" to keep the chosen scheduler defaults.
|
||||
_rr_interval=""
|
||||
_rr_interval="2"
|
||||
|
||||
# Set to "true" to disable FUNCTION_TRACER/GRAPH_TRACER, lowering overhead but limiting debugging and analyzing of kernel functions - Kernel default is "false"
|
||||
_ftracedisable="false"
|
||||
@@ -154,10 +154,10 @@ _misc_adds="true"
|
||||
# Full tickless can give higher performances in case you use isolation of CPUs for tasks
|
||||
# and it works only when using the nohz_full kernel parameter, otherwise behaves like idle.
|
||||
# Just tickless idle perform better for most platforms.
|
||||
_tickless=""
|
||||
_tickless="2"
|
||||
|
||||
# Set to "true" to use ACS override patch - https://wiki.archlinux.org/index.php/PCI_passthrough_via_OVMF#Bypassing_the_IOMMU_groups_.28ACS_override_patch.29 - Kernel default is "false"
|
||||
_acs_override=""
|
||||
_acs_override="false"
|
||||
|
||||
# Set to "true" to add Bcache filesystem support. You'll have to install bcachefs-tools-git from AUR for utilities - https://bcachefs.org/ - If in doubt, set to "false"
|
||||
# This can be buggy and isn't recommended on a production machine, also enabling this option will not allow you to enable MGLRU.
|
||||
@@ -168,13 +168,13 @@ _bcachefs="false"
|
||||
_winesync="false"
|
||||
|
||||
# Set to "true" to enable Binder and Ashmem, the kernel modules required to use the android emulator Anbox. ! This doesn't apply to 5.4.y !
|
||||
_anbox=""
|
||||
_anbox="false"
|
||||
|
||||
# A selection of patches from Zen/Liquorix kernel and additional tweaks for a better gaming experience (ZENIFY) - Default is "true"
|
||||
_zenify="true"
|
||||
|
||||
# compiler optimization level - 1. Optimize for performance (-O2); 2. Optimize harder (-O3); 3. Optimize for size (-Os) - Kernel default is "1"
|
||||
_compileroptlevel="1"
|
||||
_compileroptlevel="2"
|
||||
|
||||
# CPU compiler optimizations - Defaults to prompt at kernel config if left empty
|
||||
# AMD CPUs : "k8" "k8sse3" "k10" "barcelona" "bobcat" "jaguar" "bulldozer" "piledriver" "steamroller" "excavator" "zen" "zen2" "zen3" "zen4" (zen3 opt support depends on GCC11) (zen4 opt support depends on GCC13)
|
||||
@@ -188,7 +188,7 @@ _compileroptlevel="1"
|
||||
# - "generic_v2" (depends on GCC11 - to share the package between machines with different CPU µarch supporting at least x86-64-v2
|
||||
# - "generic_v3" (depends on GCC11 - to share the package between machines with different CPU µarch supporting at least x86-64-v3
|
||||
# - "generic_v4" (depends on GCC11 - to share the package between machines with different CPU µarch supporting at least x86-64-v4
|
||||
_processor_opt=""
|
||||
_processor_opt="skylake"
|
||||
|
||||
# CacULE only - Enable Response Driven Balancer, an experimental load balancer for CacULE
|
||||
_cacule_rdb="false"
|
||||
@@ -201,13 +201,13 @@ _cacule_rdb_interval="19"
|
||||
_tt_high_hz="false"
|
||||
|
||||
# MuQSS and PDS only - SMT (Hyperthreading) aware nice priority and policy support (SMT_NICE) - Kernel default is "true" - You can disable this on non-SMT/HT CPUs for lower overhead
|
||||
_smt_nice=""
|
||||
_smt_nice="true"
|
||||
|
||||
# Trust the CPU manufacturer to initialize Linux's CRNG (RANDOM_TRUST_CPU) - Kernel default is "false"
|
||||
_random_trust_cpu="true"
|
||||
|
||||
# Timer frequency - "100" "250" "300" "500" "750" "1000" ("2000" is available for cacule cpusched only) - More options available in kernel config prompt when left empty depending on selected cpusched with the default option pointed with a ">" (2000 for cacule, 100 for muqss and 1000 for other cpu schedulers)
|
||||
_timer_freq=""
|
||||
_timer_freq="500"
|
||||
|
||||
# Default CPU governor - "performance", "ondemand", "schedutil" or leave empty for default (schedutil)
|
||||
_default_cpu_gov="ondemand"
|
||||
|
||||
@@ -358,3 +358,263 @@ index c0cd1b9..af1e2fb 100644
|
||||
--
|
||||
https://clearlinux.org
|
||||
|
||||
From 676c2dc63592f52b716515573a3a825582a371e9 Mon Sep 17 00:00:00 2001
|
||||
From: Arjan van de Ven <arjan@linux.intel.com>
|
||||
Date: Sat, 8 Dec 2018 18:21:32 +0000
|
||||
Subject: [PATCH 1/9] x86/vdso: Use lfence instead of rep and nop
|
||||
|
||||
Signed-off-by: Alexandre Frade <kernel@xanmod.org>
|
||||
---
|
||||
arch/x86/include/asm/vdso/processor.h | 2 +-
|
||||
1 file changed, 1 insertion(+), 1 deletion(-)
|
||||
|
||||
diff --git a/arch/x86/include/asm/vdso/processor.h b/arch/x86/include/asm/vdso/processor.h
|
||||
index 57b1a7034c64..e2c45674f989 100644
|
||||
--- a/arch/x86/include/asm/vdso/processor.h
|
||||
+++ b/arch/x86/include/asm/vdso/processor.h
|
||||
@@ -10,7 +10,7 @@
|
||||
/* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
|
||||
static __always_inline void rep_nop(void)
|
||||
{
|
||||
- asm volatile("rep; nop" ::: "memory");
|
||||
+ asm volatile("lfence" ::: "memory");
|
||||
}
|
||||
|
||||
static __always_inline void cpu_relax(void)
|
||||
--
|
||||
2.39.1
|
||||
|
||||
From 48dc9669f8db68adc480ffc2698ed8204440e45b Mon Sep 17 00:00:00 2001
|
||||
From: Arjan van de Ven <arjan@linux.intel.com>
|
||||
Date: Thu, 13 Dec 2018 01:00:49 +0000
|
||||
Subject: [PATCH 2/9] sched/wait: Do accept() in LIFO order for cache
|
||||
efficiency
|
||||
|
||||
Signed-off-by: Alexandre Frade <kernel@xanmod.org>
|
||||
---
|
||||
include/linux/wait.h | 2 ++
|
||||
kernel/sched/wait.c | 24 ++++++++++++++++++++++++
|
||||
net/ipv4/inet_connection_sock.c | 2 +-
|
||||
3 files changed, 27 insertions(+), 1 deletion(-)
|
||||
|
||||
diff --git a/include/linux/wait.h b/include/linux/wait.h
|
||||
index a0307b516b09..edc21128f387 100644
|
||||
--- a/include/linux/wait.h
|
||||
+++ b/include/linux/wait.h
|
||||
@@ -165,6 +165,7 @@ static inline bool wq_has_sleeper(struct wait_queue_head *wq_head)
|
||||
|
||||
extern void add_wait_queue(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry);
|
||||
extern void add_wait_queue_exclusive(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry);
|
||||
+extern void add_wait_queue_exclusive_lifo(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry);
|
||||
extern void add_wait_queue_priority(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry);
|
||||
extern void remove_wait_queue(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry);
|
||||
|
||||
@@ -1192,6 +1193,7 @@ do { \
|
||||
*/
|
||||
void prepare_to_wait(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry, int state);
|
||||
bool prepare_to_wait_exclusive(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry, int state);
|
||||
+void prepare_to_wait_exclusive_lifo(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry, int state);
|
||||
long prepare_to_wait_event(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry, int state);
|
||||
void finish_wait(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry);
|
||||
long wait_woken(struct wait_queue_entry *wq_entry, unsigned mode, long timeout);
|
||||
diff --git a/kernel/sched/wait.c b/kernel/sched/wait.c
|
||||
index 133b74730738..1647fb8662eb 100644
|
||||
--- a/kernel/sched/wait.c
|
||||
+++ b/kernel/sched/wait.c
|
||||
@@ -47,6 +47,17 @@ void add_wait_queue_priority(struct wait_queue_head *wq_head, struct wait_queue_
|
||||
}
|
||||
EXPORT_SYMBOL_GPL(add_wait_queue_priority);
|
||||
|
||||
+void add_wait_queue_exclusive_lifo(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry)
|
||||
+{
|
||||
+ unsigned long flags;
|
||||
+
|
||||
+ wq_entry->flags |= WQ_FLAG_EXCLUSIVE;
|
||||
+ spin_lock_irqsave(&wq_head->lock, flags);
|
||||
+ __add_wait_queue(wq_head, wq_entry);
|
||||
+ spin_unlock_irqrestore(&wq_head->lock, flags);
|
||||
+}
|
||||
+EXPORT_SYMBOL(add_wait_queue_exclusive_lifo);
|
||||
+
|
||||
void remove_wait_queue(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry)
|
||||
{
|
||||
unsigned long flags;
|
||||
@@ -293,6 +304,19 @@ prepare_to_wait_exclusive(struct wait_queue_head *wq_head, struct wait_queue_ent
|
||||
}
|
||||
EXPORT_SYMBOL(prepare_to_wait_exclusive);
|
||||
|
||||
+void prepare_to_wait_exclusive_lifo(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry, int state)
|
||||
+{
|
||||
+ unsigned long flags;
|
||||
+
|
||||
+ wq_entry->flags |= WQ_FLAG_EXCLUSIVE;
|
||||
+ spin_lock_irqsave(&wq_head->lock, flags);
|
||||
+ if (list_empty(&wq_entry->entry))
|
||||
+ __add_wait_queue(wq_head, wq_entry);
|
||||
+ set_current_state(state);
|
||||
+ spin_unlock_irqrestore(&wq_head->lock, flags);
|
||||
+}
|
||||
+EXPORT_SYMBOL(prepare_to_wait_exclusive_lifo);
|
||||
+
|
||||
void init_wait_entry(struct wait_queue_entry *wq_entry, int flags)
|
||||
{
|
||||
wq_entry->flags = flags;
|
||||
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
|
||||
index f2c43f67187d..9885bfb429a2 100644
|
||||
--- a/net/ipv4/inet_connection_sock.c
|
||||
+++ b/net/ipv4/inet_connection_sock.c
|
||||
@@ -606,7 +606,7 @@ static int inet_csk_wait_for_connect(struct sock *sk, long timeo)
|
||||
* having to remove and re-insert us on the wait queue.
|
||||
*/
|
||||
for (;;) {
|
||||
- prepare_to_wait_exclusive(sk_sleep(sk), &wait,
|
||||
+ prepare_to_wait_exclusive_lifo(sk_sleep(sk), &wait,
|
||||
TASK_INTERRUPTIBLE);
|
||||
release_sock(sk);
|
||||
if (reqsk_queue_empty(&icsk->icsk_accept_queue))
|
||||
--
|
||||
2.39.1
|
||||
|
||||
|
||||
|
||||
From afa213811c5490906caf394b20bb4b616fc6f12a Mon Sep 17 00:00:00 2001
|
||||
From: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
|
||||
Date: Thu, 25 Aug 2022 15:55:26 -0700
|
||||
Subject: [PATCH 3/9] sched/fair: Simplify asym_packing logic for SMT sched
|
||||
groups
|
||||
|
||||
When the destination CPU is an SMT sibling and idle, it can only help the
|
||||
busiest group if all of its other SMT siblings are also idle. Otherwise,
|
||||
there is not increase in throughput.
|
||||
|
||||
It does not matter whether the busiest group has SMT siblings. Simply
|
||||
check if there are any tasks running on the local group before proceeding.
|
||||
|
||||
Cc: Ben Segall <bsegall@google.com>
|
||||
Cc: Daniel Bristot de Oliveira <bristot@redhat.com>
|
||||
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
|
||||
Cc: Len Brown <len.brown@intel.com>
|
||||
Cc: Mel Gorman <mgorman@suse.de>
|
||||
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
|
||||
Cc: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
|
||||
Cc: Steven Rostedt <rostedt@goodmis.org>
|
||||
Cc: Tim C. Chen <tim.c.chen@intel.com>
|
||||
Cc: Valentin Schneider <vschneid@redhat.com>
|
||||
Cc: x86@kernel.org
|
||||
Cc: linux-kernel@vger.kernel.org
|
||||
Reviewed-by: Len Brown <len.brown@intel.com>
|
||||
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
|
||||
Signed-off-by: Alexandre Frade <kernel@xanmod.org>
|
||||
---
|
||||
kernel/sched/fair.c | 29 +++++++++--------------------
|
||||
1 file changed, 9 insertions(+), 20 deletions(-)
|
||||
|
||||
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
|
||||
index 345cc5e9fa6e..60f9690a5626 100644
|
||||
--- a/kernel/sched/fair.c
|
||||
+++ b/kernel/sched/fair.c
|
||||
@@ -8921,12 +8921,10 @@ static bool asym_smt_can_pull_tasks(int dst_cpu, struct sd_lb_stats *sds,
|
||||
struct sched_group *sg)
|
||||
{
|
||||
#ifdef CONFIG_SCHED_SMT
|
||||
- bool local_is_smt, sg_is_smt;
|
||||
+ bool local_is_smt;
|
||||
int sg_busy_cpus;
|
||||
|
||||
local_is_smt = sds->local->flags & SD_SHARE_CPUCAPACITY;
|
||||
- sg_is_smt = sg->flags & SD_SHARE_CPUCAPACITY;
|
||||
-
|
||||
sg_busy_cpus = sgs->group_weight - sgs->idle_cpus;
|
||||
|
||||
if (!local_is_smt) {
|
||||
@@ -8947,25 +8945,16 @@ static bool asym_smt_can_pull_tasks(int dst_cpu, struct sd_lb_stats *sds,
|
||||
return sched_asym_prefer(dst_cpu, sg->asym_prefer_cpu);
|
||||
}
|
||||
|
||||
- /* @dst_cpu has SMT siblings. */
|
||||
-
|
||||
- if (sg_is_smt) {
|
||||
- int local_busy_cpus = sds->local->group_weight -
|
||||
- sds->local_stat.idle_cpus;
|
||||
- int busy_cpus_delta = sg_busy_cpus - local_busy_cpus;
|
||||
-
|
||||
- if (busy_cpus_delta == 1)
|
||||
- return sched_asym_prefer(dst_cpu, sg->asym_prefer_cpu);
|
||||
-
|
||||
- return false;
|
||||
- }
|
||||
-
|
||||
/*
|
||||
- * @sg does not have SMT siblings. Ensure that @sds::local does not end
|
||||
- * up with more than one busy SMT sibling and only pull tasks if there
|
||||
- * are not busy CPUs (i.e., no CPU has running tasks).
|
||||
+ * @dst_cpu has SMT siblings. When both @dst_cpu and the busiest core
|
||||
+ * have one or more busy siblings, moving tasks between them results
|
||||
+ * in the same throughput. Only if all the siblings of @dst_cpu are
|
||||
+ * idle throughput can increase.
|
||||
+ *
|
||||
+ * If the difference in the number of busy CPUs is two or more, let
|
||||
+ * find_busiest_group() take care of it.
|
||||
*/
|
||||
- if (!sds->local_stat.sum_nr_running)
|
||||
+ if (sg_busy_cpus == 1 && !sds->local_stat.sum_nr_running)
|
||||
return sched_asym_prefer(dst_cpu, sg->asym_prefer_cpu);
|
||||
|
||||
return false;
|
||||
--
|
||||
2.39.1
|
||||
|
||||
|
||||
|
||||
From a1f627fd10ced4f5eeae678bc4ba96ea7fa01d7e Mon Sep 17 00:00:00 2001
|
||||
From: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
|
||||
Date: Thu, 25 Aug 2022 15:55:28 -0700
|
||||
Subject: [PATCH 4/9] sched/fair: Let lower-priority CPUs do active balancing
|
||||
|
||||
When more than one SMT siblings of a physical core are busy, an idle CPU
|
||||
of lower priority can help.
|
||||
|
||||
Indicate that the low priority CPU can do active balancing from the high-
|
||||
priority CPU only if they belong to separate cores.
|
||||
|
||||
Cc: Ben Segall <bsegall@google.com>
|
||||
Cc: Daniel Bristot de Oliveira <bristot@redhat.com>
|
||||
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
|
||||
Cc: Len Brown <len.brown@intel.com>
|
||||
Cc: Mel Gorman <mgorman@suse.de>
|
||||
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
|
||||
Cc: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
|
||||
Cc: Steven Rostedt <rostedt@goodmis.org>
|
||||
Cc: Tim C. Chen <tim.c.chen@intel.com>
|
||||
Cc: Valentin Schneider <vschneid@redhat.com>
|
||||
Cc: x86@kernel.org
|
||||
Cc: linux-kernel@vger.kernel.org
|
||||
Reviewed-by: Len Brown <len.brown@intel.com>
|
||||
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
|
||||
Signed-off-by: Alexandre Frade <kernel@xanmod.org>
|
||||
---
|
||||
kernel/sched/fair.c | 7 ++++++-
|
||||
1 file changed, 6 insertions(+), 1 deletion(-)
|
||||
|
||||
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
|
||||
index 60f9690a5626..67b0eacad0e9 100644
|
||||
--- a/kernel/sched/fair.c
|
||||
+++ b/kernel/sched/fair.c
|
||||
@@ -10176,9 +10176,14 @@ asym_active_balance(struct lb_env *env)
|
||||
* ASYM_PACKING needs to force migrate tasks from busy but
|
||||
* lower priority CPUs in order to pack all tasks in the
|
||||
* highest priority CPUs.
|
||||
+ *
|
||||
+ * If the busy CPU has higher priority but is an SMT sibling
|
||||
+ * in which other SMT siblings are also busy, a lower-priority
|
||||
+ * CPU in a separate core can help.
|
||||
*/
|
||||
return env->idle != CPU_NOT_IDLE && (env->sd->flags & SD_ASYM_PACKING) &&
|
||||
- sched_asym_prefer(env->dst_cpu, env->src_cpu);
|
||||
+ (sched_asym_prefer(env->dst_cpu, env->src_cpu) ||
|
||||
+ !(env->sd->flags & SD_SHARE_CPUCAPACITY));
|
||||
}
|
||||
|
||||
static inline bool
|
||||
--
|
||||
2.39.1
|
||||
|
||||
@@ -820,3 +820,31 @@ index a0b0397e29ee4c..87a983a356530c 100644
|
||||
spin_unlock(&zone->lock);
|
||||
return allocated;
|
||||
}
|
||||
|
||||
From 6329525a0fa10cd13f39b76948b1296150f75c95 Mon Sep 17 00:00:00 2001
|
||||
From: Alexandre Frade <kernel@xanmod.org>
|
||||
Date: Mon, 29 Aug 2022 16:47:26 +0000
|
||||
Subject: [PATCH 14/16] XANMOD: Makefile: Disable GCC vectorization on trees
|
||||
|
||||
Signed-off-by: Alexandre Frade <kernel@xanmod.org>
|
||||
---
|
||||
Makefile | 3 +++
|
||||
1 file changed, 3 insertions(+)
|
||||
|
||||
diff --git a/Makefile b/Makefile
|
||||
index 3f6628780eb2..35a5ae1ede42 100644
|
||||
--- a/Makefile
|
||||
+++ b/Makefile
|
||||
@@ -1069,6 +1069,9 @@ endif
|
||||
KBUILD_CFLAGS-$(call gcc-min-version, 90100) += -Wno-alloc-size-larger-than
|
||||
KBUILD_CFLAGS += $(KBUILD_CFLAGS-y) $(CONFIG_CC_IMPLICIT_FALLTHROUGH)
|
||||
|
||||
+# disable GCC vectorization on trees
|
||||
+KBUILD_CFLAGS += $(call cc-option, -fno-tree-vectorize)
|
||||
+
|
||||
# disable invalid "can't wrap" optimizations for signed / pointers
|
||||
KBUILD_CFLAGS += -fno-strict-overflow
|
||||
|
||||
--
|
||||
2.39.1
|
||||
|
||||
|
||||
@@ -20,3 +20,265 @@ index 6b3b59cc51d6..2a0072192c3d 100644
|
||||
|
||||
int sched_thermal_decay_shift;
|
||||
static int __init setup_sched_thermal_decay_shift(char *str)
|
||||
|
||||
From 5d5b708e3731e135ea7ae168571ad78d883e63e8 Mon Sep 17 00:00:00 2001
|
||||
From: Alexandre Frade <kernel@xanmod.org>
|
||||
Date: Wed, 1 Feb 2023 10:17:47 +0000
|
||||
Subject: [PATCH 02/16] XANMOD: fair: Remove all energy efficiency functions
|
||||
|
||||
Signed-off-by: Alexandre Frade <kernel@xanmod.org>
|
||||
---
|
||||
kernel/sched/fair.c | 224 +-------------------------------------------
|
||||
1 file changed, 3 insertions(+), 221 deletions(-)
|
||||
|
||||
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
|
||||
index 0f8736991427..345cc5e9fa6e 100644
|
||||
--- a/kernel/sched/fair.c
|
||||
+++ b/kernel/sched/fair.c
|
||||
@@ -19,6 +19,9 @@
|
||||
*
|
||||
* Adaptive scheduling granularity, math enhancements by Peter Zijlstra
|
||||
* Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra
|
||||
+ *
|
||||
+ * Remove energy efficiency functions by Alexandre Frade
|
||||
+ * (C) 2021 Alexandre Frade <kernel@xanmod.org>
|
||||
*/
|
||||
#include <linux/energy_model.h>
|
||||
#include <linux/mmap_lock.h>
|
||||
@@ -7136,219 +7139,6 @@ eenv_pd_max_util(struct energy_env *eenv, struct cpumask *pd_cpus,
|
||||
return min(max_util, eenv->cpu_cap);
|
||||
}
|
||||
|
||||
-/*
|
||||
- * compute_energy(): Use the Energy Model to estimate the energy that @pd would
|
||||
- * consume for a given utilization landscape @eenv. When @dst_cpu < 0, the task
|
||||
- * contribution is ignored.
|
||||
- */
|
||||
-static inline unsigned long
|
||||
-compute_energy(struct energy_env *eenv, struct perf_domain *pd,
|
||||
- struct cpumask *pd_cpus, struct task_struct *p, int dst_cpu)
|
||||
-{
|
||||
- unsigned long max_util = eenv_pd_max_util(eenv, pd_cpus, p, dst_cpu);
|
||||
- unsigned long busy_time = eenv->pd_busy_time;
|
||||
-
|
||||
- if (dst_cpu >= 0)
|
||||
- busy_time = min(eenv->pd_cap, busy_time + eenv->task_busy_time);
|
||||
-
|
||||
- return em_cpu_energy(pd->em_pd, max_util, busy_time, eenv->cpu_cap);
|
||||
-}
|
||||
-
|
||||
-/*
|
||||
- * find_energy_efficient_cpu(): Find most energy-efficient target CPU for the
|
||||
- * waking task. find_energy_efficient_cpu() looks for the CPU with maximum
|
||||
- * spare capacity in each performance domain and uses it as a potential
|
||||
- * candidate to execute the task. Then, it uses the Energy Model to figure
|
||||
- * out which of the CPU candidates is the most energy-efficient.
|
||||
- *
|
||||
- * The rationale for this heuristic is as follows. In a performance domain,
|
||||
- * all the most energy efficient CPU candidates (according to the Energy
|
||||
- * Model) are those for which we'll request a low frequency. When there are
|
||||
- * several CPUs for which the frequency request will be the same, we don't
|
||||
- * have enough data to break the tie between them, because the Energy Model
|
||||
- * only includes active power costs. With this model, if we assume that
|
||||
- * frequency requests follow utilization (e.g. using schedutil), the CPU with
|
||||
- * the maximum spare capacity in a performance domain is guaranteed to be among
|
||||
- * the best candidates of the performance domain.
|
||||
- *
|
||||
- * In practice, it could be preferable from an energy standpoint to pack
|
||||
- * small tasks on a CPU in order to let other CPUs go in deeper idle states,
|
||||
- * but that could also hurt our chances to go cluster idle, and we have no
|
||||
- * ways to tell with the current Energy Model if this is actually a good
|
||||
- * idea or not. So, find_energy_efficient_cpu() basically favors
|
||||
- * cluster-packing, and spreading inside a cluster. That should at least be
|
||||
- * a good thing for latency, and this is consistent with the idea that most
|
||||
- * of the energy savings of EAS come from the asymmetry of the system, and
|
||||
- * not so much from breaking the tie between identical CPUs. That's also the
|
||||
- * reason why EAS is enabled in the topology code only for systems where
|
||||
- * SD_ASYM_CPUCAPACITY is set.
|
||||
- *
|
||||
- * NOTE: Forkees are not accepted in the energy-aware wake-up path because
|
||||
- * they don't have any useful utilization data yet and it's not possible to
|
||||
- * forecast their impact on energy consumption. Consequently, they will be
|
||||
- * placed by find_idlest_cpu() on the least loaded CPU, which might turn out
|
||||
- * to be energy-inefficient in some use-cases. The alternative would be to
|
||||
- * bias new tasks towards specific types of CPUs first, or to try to infer
|
||||
- * their util_avg from the parent task, but those heuristics could hurt
|
||||
- * other use-cases too. So, until someone finds a better way to solve this,
|
||||
- * let's keep things simple by re-using the existing slow path.
|
||||
- */
|
||||
-static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
|
||||
-{
|
||||
- struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_rq_mask);
|
||||
- unsigned long prev_delta = ULONG_MAX, best_delta = ULONG_MAX;
|
||||
- unsigned long p_util_min = uclamp_is_used() ? uclamp_eff_value(p, UCLAMP_MIN) : 0;
|
||||
- unsigned long p_util_max = uclamp_is_used() ? uclamp_eff_value(p, UCLAMP_MAX) : 1024;
|
||||
- struct root_domain *rd = this_rq()->rd;
|
||||
- int cpu, best_energy_cpu, target = -1;
|
||||
- struct sched_domain *sd;
|
||||
- struct perf_domain *pd;
|
||||
- struct energy_env eenv;
|
||||
-
|
||||
- rcu_read_lock();
|
||||
- pd = rcu_dereference(rd->pd);
|
||||
- if (!pd || READ_ONCE(rd->overutilized))
|
||||
- goto unlock;
|
||||
-
|
||||
- /*
|
||||
- * Energy-aware wake-up happens on the lowest sched_domain starting
|
||||
- * from sd_asym_cpucapacity spanning over this_cpu and prev_cpu.
|
||||
- */
|
||||
- sd = rcu_dereference(*this_cpu_ptr(&sd_asym_cpucapacity));
|
||||
- while (sd && !cpumask_test_cpu(prev_cpu, sched_domain_span(sd)))
|
||||
- sd = sd->parent;
|
||||
- if (!sd)
|
||||
- goto unlock;
|
||||
-
|
||||
- target = prev_cpu;
|
||||
-
|
||||
- sync_entity_load_avg(&p->se);
|
||||
- if (!uclamp_task_util(p, p_util_min, p_util_max))
|
||||
- goto unlock;
|
||||
-
|
||||
- eenv_task_busy_time(&eenv, p, prev_cpu);
|
||||
-
|
||||
- for (; pd; pd = pd->next) {
|
||||
- unsigned long util_min = p_util_min, util_max = p_util_max;
|
||||
- unsigned long cpu_cap, cpu_thermal_cap, util;
|
||||
- unsigned long cur_delta, max_spare_cap = 0;
|
||||
- unsigned long rq_util_min, rq_util_max;
|
||||
- unsigned long prev_spare_cap = 0;
|
||||
- int max_spare_cap_cpu = -1;
|
||||
- unsigned long base_energy;
|
||||
-
|
||||
- cpumask_and(cpus, perf_domain_span(pd), cpu_online_mask);
|
||||
-
|
||||
- if (cpumask_empty(cpus))
|
||||
- continue;
|
||||
-
|
||||
- /* Account thermal pressure for the energy estimation */
|
||||
- cpu = cpumask_first(cpus);
|
||||
- cpu_thermal_cap = arch_scale_cpu_capacity(cpu);
|
||||
- cpu_thermal_cap -= arch_scale_thermal_pressure(cpu);
|
||||
-
|
||||
- eenv.cpu_cap = cpu_thermal_cap;
|
||||
- eenv.pd_cap = 0;
|
||||
-
|
||||
- for_each_cpu(cpu, cpus) {
|
||||
- struct rq *rq = cpu_rq(cpu);
|
||||
-
|
||||
- eenv.pd_cap += cpu_thermal_cap;
|
||||
-
|
||||
- if (!cpumask_test_cpu(cpu, sched_domain_span(sd)))
|
||||
- continue;
|
||||
-
|
||||
- if (!cpumask_test_cpu(cpu, p->cpus_ptr))
|
||||
- continue;
|
||||
-
|
||||
- util = cpu_util_next(cpu, p, cpu);
|
||||
- cpu_cap = capacity_of(cpu);
|
||||
-
|
||||
- /*
|
||||
- * Skip CPUs that cannot satisfy the capacity request.
|
||||
- * IOW, placing the task there would make the CPU
|
||||
- * overutilized. Take uclamp into account to see how
|
||||
- * much capacity we can get out of the CPU; this is
|
||||
- * aligned with sched_cpu_util().
|
||||
- */
|
||||
- if (uclamp_is_used() && !uclamp_rq_is_idle(rq)) {
|
||||
- /*
|
||||
- * Open code uclamp_rq_util_with() except for
|
||||
- * the clamp() part. Ie: apply max aggregation
|
||||
- * only. util_fits_cpu() logic requires to
|
||||
- * operate on non clamped util but must use the
|
||||
- * max-aggregated uclamp_{min, max}.
|
||||
- */
|
||||
- rq_util_min = uclamp_rq_get(rq, UCLAMP_MIN);
|
||||
- rq_util_max = uclamp_rq_get(rq, UCLAMP_MAX);
|
||||
-
|
||||
- util_min = max(rq_util_min, p_util_min);
|
||||
- util_max = max(rq_util_max, p_util_max);
|
||||
- }
|
||||
- if (!util_fits_cpu(util, util_min, util_max, cpu))
|
||||
- continue;
|
||||
-
|
||||
- lsub_positive(&cpu_cap, util);
|
||||
-
|
||||
- if (cpu == prev_cpu) {
|
||||
- /* Always use prev_cpu as a candidate. */
|
||||
- prev_spare_cap = cpu_cap;
|
||||
- } else if (cpu_cap > max_spare_cap) {
|
||||
- /*
|
||||
- * Find the CPU with the maximum spare capacity
|
||||
- * among the remaining CPUs in the performance
|
||||
- * domain.
|
||||
- */
|
||||
- max_spare_cap = cpu_cap;
|
||||
- max_spare_cap_cpu = cpu;
|
||||
- }
|
||||
- }
|
||||
-
|
||||
- if (max_spare_cap_cpu < 0 && prev_spare_cap == 0)
|
||||
- continue;
|
||||
-
|
||||
- eenv_pd_busy_time(&eenv, cpus, p);
|
||||
- /* Compute the 'base' energy of the pd, without @p */
|
||||
- base_energy = compute_energy(&eenv, pd, cpus, p, -1);
|
||||
-
|
||||
- /* Evaluate the energy impact of using prev_cpu. */
|
||||
- if (prev_spare_cap > 0) {
|
||||
- prev_delta = compute_energy(&eenv, pd, cpus, p,
|
||||
- prev_cpu);
|
||||
- /* CPU utilization has changed */
|
||||
- if (prev_delta < base_energy)
|
||||
- goto unlock;
|
||||
- prev_delta -= base_energy;
|
||||
- best_delta = min(best_delta, prev_delta);
|
||||
- }
|
||||
-
|
||||
- /* Evaluate the energy impact of using max_spare_cap_cpu. */
|
||||
- if (max_spare_cap_cpu >= 0 && max_spare_cap > prev_spare_cap) {
|
||||
- cur_delta = compute_energy(&eenv, pd, cpus, p,
|
||||
- max_spare_cap_cpu);
|
||||
- /* CPU utilization has changed */
|
||||
- if (cur_delta < base_energy)
|
||||
- goto unlock;
|
||||
- cur_delta -= base_energy;
|
||||
- if (cur_delta < best_delta) {
|
||||
- best_delta = cur_delta;
|
||||
- best_energy_cpu = max_spare_cap_cpu;
|
||||
- }
|
||||
- }
|
||||
- }
|
||||
- rcu_read_unlock();
|
||||
-
|
||||
- if (best_delta < prev_delta)
|
||||
- target = best_energy_cpu;
|
||||
-
|
||||
- return target;
|
||||
-
|
||||
-unlock:
|
||||
- rcu_read_unlock();
|
||||
-
|
||||
- return target;
|
||||
-}
|
||||
-
|
||||
/*
|
||||
* select_task_rq_fair: Select target runqueue for the waking task in domains
|
||||
* that have the relevant SD flag set. In practice, this is SD_BALANCE_WAKE,
|
||||
@@ -7376,14 +7166,6 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
|
||||
lockdep_assert_held(&p->pi_lock);
|
||||
if (wake_flags & WF_TTWU) {
|
||||
record_wakee(p);
|
||||
-
|
||||
- if (sched_energy_enabled()) {
|
||||
- new_cpu = find_energy_efficient_cpu(p, prev_cpu);
|
||||
- if (new_cpu >= 0)
|
||||
- return new_cpu;
|
||||
- new_cpu = prev_cpu;
|
||||
- }
|
||||
-
|
||||
want_affine = !wake_wide(p) && cpumask_test_cpu(cpu, p->cpus_ptr);
|
||||
}
|
||||
|
||||
--
|
||||
2.39.1
|
||||
|
||||
@@ -88,3 +88,31 @@ index 6b423eebfd5d..61e3271675d6 100644
|
||||
#define MICRO_FREQUENCY_MIN_SAMPLE_RATE (10000)
|
||||
#define MIN_FREQUENCY_UP_THRESHOLD (1)
|
||||
#define MAX_FREQUENCY_UP_THRESHOLD (100)
|
||||
|
||||
From cba31b19f8c38696b13ba48e0e8b6dbe747d6bae Mon Sep 17 00:00:00 2001
|
||||
From: Alexandre Frade <admfrade@gmail.com>
|
||||
Date: Mon, 29 Jan 2018 17:31:25 +0000
|
||||
Subject: [PATCH 10/16] XANMOD: mm/vmscan: vm_swappiness = 30 decreases the
|
||||
amount of swapping
|
||||
|
||||
Signed-off-by: Alexandre Frade <admfrade@gmail.com>
|
||||
Signed-off-by: Alexandre Frade <kernel@xanmod.org>
|
||||
---
|
||||
mm/vmscan.c | 2 +-
|
||||
1 file changed, 1 insertion(+), 1 deletion(-)
|
||||
|
||||
diff --git a/mm/vmscan.c b/mm/vmscan.c
|
||||
index 5b7b8d4f5297..549684b29418 100644
|
||||
--- a/mm/vmscan.c
|
||||
+++ b/mm/vmscan.c
|
||||
@@ -190,7 +190,7 @@ struct scan_control {
|
||||
/*
|
||||
* From 0 .. 200. Higher means more swappy.
|
||||
*/
|
||||
-int vm_swappiness = 60;
|
||||
+int vm_swappiness = 30;
|
||||
|
||||
static void set_task_reclaim_state(struct task_struct *task,
|
||||
struct reclaim_state *rs)
|
||||
--
|
||||
2.39.1
|
||||
|
||||
Reference in New Issue
Block a user