This reverts commit eb994f2728.
The hooks android_rvh_attach_entity_load_avg,
android_rvh_detach_entity_load_avg, android_rvh_update_load_avg,
android_rvh_remove_entity_load_avg, and android_rvh_update_blocked_fair
are not used by any vendor, so remove it to help with merge issues with
future LTS releases.
If they are needed by any real user, it can easily be reverted to add
them back and then the symbols should be added to the abi list at the
same time to prevent it from being removed again later.
Bug: 203756332
Bug: 201260585
Cc: Rick Yiu <rickyiu@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Ibb6f0d2b71e830092f8d3452fdf34c5d1975e71e
To avoid changing the visibiliy of data types when including
hook definition headers remove header file inclusions from
the hook definition header files.
Instead, the hook definition headers should just have forward
declarations that don't require full definition.
To provide full definitions of the types for the KMI, the
headers that define the types should be included by the
source file that instantiates the hooks - normally
vendor_hooks.c.
Since the KMI is frozen, some of the inclusions are still
required to preserve the CRC associated with symbols. Keep
these inclusions under #ifdef __GENKSYMS__.
Bug: 233047575
Signed-off-by: Todd Kjos <tkjos@google.com>
Change-Id: Ibc1173eb4b07fcec21c7abd8e0ab1950b3fb5b34
Change how vendor hooks are instantiated to promote
more complete structure definition in the ABI XML
description without complicating hook definition for
partners.
We don't want to force partners to include all
headers as part of the hook definition in include/trace/hooks/
since that causes extra headers to be included in
source files that runs the risk of changing visibility
resulting in CRC changes to KMI symbols. Instead continue
the practice of using forward declarations in the hook
header files.
Instead of instantiating hook tracepoints globally in
drivers/android/vendor_hooks.c, use subsystem-specific
vendor_hooks.c. This avoids namespace collisions between
internal header files and limits the exposure to the internal
headers to the instantiation, not the call sites.
In this patch, all of the scheduler related hooks are
instantiated in kernel/sched/vendor_hooks.c which can
cleanly include scheduler-related header files to
provide full type visibility.
Since this is being done on a KMI frozen branch, there
are some __GENKSYMS__ tricks needed to preserve the
existing CRC as we increase the number of fully typed
structs. These tricks won't be needed when this is
introduced on a non-KMI-frozen branch.
With this patch, the number of undefined types was
reduced from 108 to 68.
Bug: 233047575
Signed-off-by: Todd Kjos <tkjos@google.com>
Change-Id: Ife5a66c2968de73e3f6d05840411310611e2e175
Vendor may have need to track rt util.
Bug: 201261299
Signed-off-by: Rick Yiu <rickyiu@google.com>
Change-Id: I2f4e5142c6bc8574ee3558042e1fb0dae13b702d
Vendor may have the need to implement their own util tracking.
Bug: 201260585
Signed-off-by: Rick Yiu <rickyiu@google.com>
Change-Id: I973902e6ff82a85ecd029ac5a78692d629df1ebe
These hooks will do the following works:
a) record the time of the process in various states
b) Make corresponding optimization strategies in different hooks
Bug: 205938967
Signed-off-by: Liujie Xie <xieliujie@oppo.com>
Change-Id: Ia3c47bbf0aadd17337ce18fd910343b1b8c3ef93
(cherry picked from commit a61d61bab7c3d221d5d2250eacab2e4e9f59b252)
Signed-off-by: Carlos Llamas <cmllamas@google.com>
add vendor hooks for cpu affinity to support oem's feature.
Bug: 183674818
Signed-off-by: lijianzhong <lijianzhong@xiaomi.com>
Change-Id: I3402abec4d9faa08f564409bfb8db8d7902f3aa2
add vendor hooks to compute new cpu freq for oem feature.
Bug: 235925535
Signed-off-by: heshuai1 <heshuai1@xiaomi.com>
Change-Id: I2b8e1f76f3b9792148f153190b862face679ebbd
Add new vendor hook when set affinity to detect task status
for performance tuning.
Bug: 236775946
Change-Id: I407637c85e2ea93585877312f090981fee848979
Signed-off-by: Jing-Ting Wu <Jing-Ting.Wu@mediatek.com>
The inclusion of the internal kernel/cgroup/cgroup-internal.h header
broke some androidci builds.
Bug: 233047575
Fixes: d5f809c84d ("ANDROID: GKI: use public type definitions in vendor hooks")
Change-Id: Iacca4b8518b9a9c45c927a58621383e66022cb5a
Signed-off-by: Giuliano Procida <gprocida@google.com>
After 7b26719a77 ("ANDROID: GKI: use internal type definitions in
vendor hooks") we stopped providing a forward declaration of 'struct
uclamp_se' and instead relied on pulling its definition from
linux/sched.h. the 'uclamp_se' structure is conditionally defined based
upon CONFIG_UCLAMP_TASK therefore causing a build error to show up for
the android_rvh_uclamp_eff_get trace point.
Fix this by providing a forward declaration of 'struct uclamp_se' like
before.
Fixes: 7b26719a77 ("ANDROID: GKI: use internal type definitions in vendor hooks")
Signed-off-by: Florian Fainelli <florian.fainelli@broadcom.com>
Change-Id: Iaa803d149150c1703435f9cfa52ea1406a9521fd
This change is intended to reduce the chance of a missed ABI break
involving vendor hooks.
The following types are now fully defined in ABI XML.
* `struct binder_transaction`
* `struct binder_thread`
* `struct binder_proc`
* `struct printk_ringbuffer`
* `struct printk_record`
symbol '__alloc_pages' changed
CRC changed from 0x6497cbd to 0x230fd8e5
symbol '__find_vma' changed
CRC changed from 0xf11b619f to 0xb0069a61
symbol '__free_pages' changed
CRC changed from 0xba5eeb68 to 0xdc394265
... 195 omitted; 198 symbols have only CRC changes
Bug: 233047575
Change-Id: I807227f6161f0a44ec3657dc8cdb8a3fdffc40b3
Signed-off-by: Giuliano Procida <gprocida@google.com>
This change is intended to reduce the chance of a missed ABI break
involving vendor hooks.
The following type is now fully defined in ABI XML.
* `struct timekeeper`
symbol '__traceiter_android_rvh_account_irq' changed
CRC changed from 0x18e0dccf to 0x4e82c102
symbol '__traceiter_android_rvh_after_dequeue_task' changed
CRC changed from 0x7efc419b to 0x1c0066e1
symbol '__traceiter_android_rvh_after_enqueue_task' changed
CRC changed from 0xd6f0f603 to 0xf7eaafb8
... 75 omitted; 78 symbols have only CRC changes
Bug: 233047575
Change-Id: I8c04eede1b9d9e788abce95cb31dbd19916774ef
Signed-off-by: Giuliano Procida <gprocida@google.com>
Add a hook in account_process_tick, which help us to get information
about the high load task and the cpu they running on.
Bug: 183260319
Change-Id: I54162ce3c65bd69e08d2d4747e4d4883efe4c442
Signed-off-by: Liujie Xie <xieliujie@oppo.com>
(cherry picked from commit 63399b4e2e25b0a1522deb724f1de269ac4c8f8b)
Add hook to dup_task_struct for vendor data fields initialisation.
Bug: 188004638
Change-Id: I4b58604ee822fb8d1e0cc37bec72e820e7318427
Signed-off-by: Liangliang Li <liliangliang@vivo.com>
(cherry picked from commit f66d96b14aab5051fdf6b5054d87362c17a7b365)
Add a hook in irqtime_account_process_tick, which helps to get
information about the high load task.
Bug: 187904818
Change-Id: I644f7d66b09d047ca6b0a0fbd2915a6387c8c007
Signed-off-by: Liangliang Li <liliangliang@vivo.com>
(cherry picked from commit fe580539f6cec43ddb0d6ecfd39aa2f4e45754ca)
Allow the vendor module to know the target cpu for better decisions on
whether to enforce __ttwu_queue_wakelist() based wakeup.
Bug: 234483895
Change-Id: Ic27054a5f6adc040fa3cadbd57d37608bf353c5f
Signed-off-by: Abhijeet Dharmapurikar <quic_adharmap@quicinc.com>
Currently the code has single hook for tracking irqs. However
modules need to deduce start and end of the irq.
Create separate hooks for irq start and end since the
cputime has already figured it out.
Bug: 231341763
Change-Id: Ie0dd503b283d83f69d01171ebd1cd6127c3bafd0
Signed-off-by: Shaleen Agrawal <quic_shalagra@quicinc.com>
Signed-off-by: Ashay Jaiswal <quic_ashayj@quicinc.com>
Currently, the enqueue and dequeue tracehooks pass in the flags
parameter, however, the after tracehooks that follow do not.
Bug: 226570047
Change-Id: I51cb50054562893271e5d3efd7c6bd028977622d
Signed-off-by: Shaleen Agrawal <quic_shalagra@quicinc.com>
Vendor module needs access to the task structure
through is_cpu_allowed tracehook to better
assess whether the chosen cpu can be allowed.
Update the existing is_cpu_allowed tracehook to
pass the task struct of the task currently being
handled.
Bug: 228392842
Change-Id: I882b593ccb77da0755e076c7e636db304ee74b42
Signed-off-by: Stephen Dickey <quic_dickey@quicinc.com>
Just as sched_getaffinity updates the affinity based upon the
active_mask, the vendor hook needs to be able to constrain
the task's affinity.
Introduce a hook to sched_getaffinity such that vendor modules
are able to update the affinity.
Bug: 229133948
Change-Id: I5c501e9204d4fcc6688f675a41be60ef5a2d1075
Signed-off-by: Stephen Dickey <quic_dickey@quicinc.com>
Module code would like to hold some locks when affinity is being updated
for 32 bit task exec.
Create pre and post tracepoints in force_compatible_cpus_allowed_ptr()
Bug: 187917024
Change-Id: I95bff9f4d5b5d37c1d5440acbd6857d2855c2b43
Signed-off-by: Abhijeet Dharmapurikar <adharmap@codeaurora.org>
Signed-off-by: Shaleen Agrawal <shalagra@codeaurora.org>
Signed-off-by: Stephen Dickey <quic_dickey@quicinc.com>
When vendor hooks are added to a file that previously didn't have any
vendor hooks, we end up indirectly including linux/tracepoint.h. This
causes some data types that used to be opaque (forward declared) to the
code to become visible to the code.
Modversions correctly catches this change in visibility, but we don't
really care about the data types made visible when linux/tracepoint.h is
included. So, hide this from modversions in the central vendor_hooks.h file
instead of having to fix this on a case by case basis.
This change itself will cause a one time CRC breakage/churn because it's
fixing the existing vendor hook headers, but should reduce unnecessary CRC
churns in the future.
To avoid future pointless CRC churn, vendor hook header files that include
vendor_hooks.h should not include linux/tracepoint.h directly.
Bug: 227513263
Bug: 226140073
Signed-off-by: Saravana Kannan <saravanak@google.com>
Change-Id: Ia88e6af11dd94fe475c464eb30a6e5e1e24c938b
android_rvh_effective_cpu_util:
To perform vendor-specific cpu util, it is used in EAS/schedutil/thermal.
The effective_cpu_util would be called when thermal calc the dynamic power,
it's non-atomic context, so set the hook be restricted.
Bug: 226686099
Test: build pass
Signed-off-by: Xuewen Yan <xuewen.yan@unisoc.com>
Change-Id: I6fd77f44ca4328f5ef37d96989aa2e08d65e29bb
android_vh_do_wake_up_sync:
To modify the mode value of __wake_up_sync_key
android_vh_set_wake_flags:
To modify the wake flag from a module
Bug: 181743516
Signed-off-by: Namkyu Kim <namkyu78.kim@samsung.com>
Change-Id: I972e2469c3f139373d21f1e8c85974763388a693
(cherry picked from commit 97368fc2dcc29777e8d3d637d0afdef90e611763)
(cherry picked from commit 0d0f0c5020bc425c9a51c8d17b16ca831c2598fb)
Vendor may have its own estimated utilization.
Bug: 170508405
Signed-off-by: Rick Yiu <rickyiu@google.com>
Change-Id: I6055907de75ace4586c3ad854d40f42e3bf40147
Signed-off-by: Will McVicker <willmcvicker@google.com>
Vendor may have their own behavior for determing uclamp value. Also
uclamp_eff_get() is used in uclamp_rq_inc_id() when task is enqueued,
and it is contained in uclamp_eff_value(), there is no much benefit to
hook uclamp_eff_value(), so remove the hook in uclamp_eff_value().
Bug: 170507972
Bug: 180859906
Signed-off-by: Rick Yiu <rickyiu@google.com>
Change-Id: Ibf7c043188a77c40baf5b4040b198a4197236089
Signed-off-by: Will McVicker <willmcvicker@google.com>
Currently, cpu_overutilized uses the threshold 80% of cpu capacity, Add
vendor hook so that vendor could define other values.
Bug: 176722431
Signed-off-by: Rick Yiu <rickyiu@google.com>
Change-Id: Ic10bd8ebb40f7321cfa10e33ac020e3347ca6550
(cherry picked from commit b199c6e8795791e008cfff6e214676857cd0651c)
Signed-off-by: J. Avila <elavila@google.com>
(cherry picked from commit f5998fbf2dbbcdda953018d95150b394b02b286b)
Allow module to control behavior of get_nohz_timer_target.
Bug: 205164003
Change-Id: I38cb201ebf06db7bbce0d6cb68dbbe3729355be8
Signed-off-by: Stephen Dickey <quic_dickey@quicinc.com>
To support the replacement of pause, is_cpu_allowed is the best
place to hook into the code to restrict CPUs for a module based
implementation. This restricts select_fallback_rq, select_task_rq,
and __migate_task, to ensure the cpu is allowed.
Include a hook in is_cpu_allowed to allow the module to control
which cpu is allowed during a migration event.
Bug: 205164003
Change-Id: I665e4d39318079bdb99bd248969ecb9eb528f9df
Signed-off-by: Stephen Dickey <quic_dickey@quicinc.com>
Add android_rvh_find_new_ilb to select a next ilb cpu for vendors.
Bug: 190228983
Change-Id: Iba1a0cd9cdc22dcf628dd33f8d838fe513a4818f
Signed-off-by: Choonghoon Park <choong.park@samsung.com>
A vendor hook is added in post_init_entity_util_avg before
a new cfs task's util is attached to cfs_rq's util so that
vendors can gather and modify se's information to modify
scheduling behavior and DVFS as they want.
trace_android_rvh_new_task_stats is not a proper hook because
it is called after the task's util is attached to cfs_rq's util,
which means updating cfs_rq's sched_avg and DVFS request are done.
Bug: 184219858
Signed-off-by: Choonghoon Park <choong.park@samsung.com>
Change-Id: I2deaa93297f8464895978496c9838cdffaa35b7f
Vendor scheduler may not want to rebalance between domains in some case.
New vendor hook is added for vendor to decide whether rebalance domains.
Bug: 176952463
Change-Id: Ie2edbd4b11679786096dd0170496904ae180583c
Signed-off-by: Park Bumgyu <bumgyu.park@samsung.com>
When a task yields, it relinquishes the cpu and
scheduler is tasked to find another task.
However our vendor scheduler logic implementation
could return the same task leading to a loop where
the yielded task gets to run back, so add hook point
in do_sched_yield() for vendor can do some work
before task is scheduled.
Bug: 205804537
Change-Id: I6528c3f4b0ee360559ef9c97cb1eb2b2d1357870
Signed-off-by: Tengfei Fan <quic_tengfan@quicinc.com>
Signed-off-by: Sai Harshini Nimmala <quic_snimmala@quicinc.com>
place_entity() vendor hook is meant to tweak vruntime by vendor
modules as needed, but with current form of the hook that is not
possible as vruntime is passed by it's value. Fix it by switching
to pass by reference.
Bug: 175448877
Change-Id: Ibb51592f94da31019fa98a6767d080ec61daafe6
Signed-off-by: Satya Durga Srinivasu Prabhala <satyap@codeaurora.org>
Vendor architectures may contain CPUs running on the same clock line
which contain different capacities. Add a tracehook in this path to
allow vendor modules to skip implicit check to prevent crashes.
Bug: 206602617
Change-Id: Ica01a214689607b8d79b370c20bc9a8c44ca2117
Signed-off-by: Shaleen Agrawal <shalagra@codeaurora.org>
While overriding sugov behavior, it is also necessary to know the
runqueue on which the task is going to be enqueued.
Bug: 171598214
Change-Id: I1491a2be34a6e615c5d4bc68e095647744870233
Signed-off-by: Shaleen Agrawal <shalagra@codeaurora.org>
Restricted vendor hook to modify the cpu selected in rto_next_cpu,
which is needed for the implementation of CPU Pause.
Bug: 205164003
Change-Id: I0dc675e54f7f116d538840262fbb0ba6d28246f4
Signed-off-by: Stephen Dickey <quic_dickey@quicinc.com>
It is necessary to update the valid mask to implement CPU Pause
from a vendor module. Introduce a hook to allow this.
Bug: 205164003
Change-Id: Ic99b74d17f361453b044e115a54698e566db13b6
Signed-off-by: Stephen Dickey <quic_dickey@quicinc.com>
android_rvh_check_preempt_wakeup hook is in place to allow vendor
modules to force the running task preemption by the waking task.
Update the tracepoint to accept another input to not preempting
the current running task. The hook is moved further down so that
it can be updated to pass the sched_entity structure corresponding
to waking and running tasks in the next patch.
Bug: 184575210
Change-Id: Id4f45ba2819802636b6b86ed34c124771d0d69eb
Signed-off-by: Pavankumar Kondeti <quic_pkondeti@quicinc.com>
Signed-off-by: Shaleen Agrawal <shalagra@codeaurora.org>
Add a restricted vendor hook to notify that a cpu controller
cgroup is online.
Bug: 200103201
Change-Id: I7d37f38c24ce146eabb4716a959aee703d71926e
Signed-off-by: Pavankumar Kondeti <quic_pkondeti@quicinc.com>
Signed-off-by: Ashay Jaiswal <quic_ashayj@quicinc.com>
