linux.git/kernel/delayacct.c, branch v6.6.132

delayacct: track delays from IRQ/SOFTIRQ

2023-04-18T23:39:34+00:00

Delay accounting does not track the delay of IRQ/SOFTIRQ.  While
IRQ/SOFTIRQ could have obvious impact on some workloads productivity, such
as when workloads are running on system which is busy handling network
IRQ/SOFTIRQ.

Get the delay of IRQ/SOFTIRQ could help users to reduce such delay.  Such
as setting interrupt affinity or task affinity, using kernel thread for
NAPI etc.  This is inspired by "sched/psi: Add PSI_IRQ to track
IRQ/SOFTIRQ pressure"[1].  Also fix some code indent problems of older
code.

And update tools/accounting/getdelays.c:
    / # ./getdelays -p 156 -di
    print delayacct stats ON
    printing IO accounting
    PID     156

    CPU             count     real total  virtual total    delay total  delay average
                       15       15836008       16218149      275700790         18.380ms
    IO              count    delay total  delay average
                        0              0          0.000ms
    SWAP            count    delay total  delay average
                        0              0          0.000ms
    RECLAIM         count    delay total  delay average
                        0              0          0.000ms
    THRASHING       count    delay total  delay average
                        0              0          0.000ms
    COMPACT         count    delay total  delay average
                        0              0          0.000ms
    WPCOPY          count    delay total  delay average
                       36        7586118          0.211ms
    IRQ             count    delay total  delay average
                       42         929161          0.022ms

[1] commit 52b1364ba0b1("sched/psi: Add PSI_IRQ to track IRQ/SOFTIRQ pressure")

Link: https://lkml.kernel.org/r/202304081728353557233@zte.com.cn
Signed-off-by: Yang Yang 
Cc: Jiang Xuexin 
Cc: wangyong 
Cc: junhua huang 
Cc: Balbir Singh 
Cc: Ingo Molnar 
Cc: Jonathan Corbet 
Cc: Juri Lelli 
Cc: Peter Zijlstra 
Signed-off-by: Andrew Morton

delayacct: support re-entrance detection of thrashing accounting

2022-09-27T02:46:07+00:00

Once upon a time, we only support accounting thrashing of page cache. 
Then Joonsoo introduced workingset detection for anonymous pages and we
gained the ability to account thrashing of them[1].

For page cache thrashing accounting, there is no suitable place to do it
in fs level likes swap_readpage().  So we have to do it in
folio_wait_bit_common().

Then for anonymous pages thrashing accounting, we have to do it in both
swap_readpage() and folio_wait_bit_common().  This likes PSI, so we should
let thrashing accounting supports re-entrance detection.

This patch is to prepare complete thrashing accounting, and is based on
patch "filemap: make the accounting of thrashing more consistent".

[1] commit aae466b0052e ("mm/swap: implement workingset detection for anonymous LRU")

Link: https://lkml.kernel.org/r/20220815071134.74551-1-yang.yang29@zte.com.cn
Signed-off-by: Yang Yang 
Signed-off-by: CGEL ZTE 
Reviewed-by: Ran Xiaokai 
Reviewed-by: wangyong 
Acked-by: Joonsoo Kim 
Signed-off-by: Andrew Morton

delayacct: track delays from write-protect copy

2022-06-01T22:55:25+00:00

Delay accounting does not track the delay of write-protect copy.  When
tasks trigger many write-protect copys(include COW and unsharing of
anonymous pages[1]), it may spend a amount of time waiting for them.  To
get the delay of tasks in write-protect copy, could help users to evaluate
the impact of using KSM or fork() or GUP.

Also update tools/accounting/getdelays.c:

    / # ./getdelays -dl -p 231
    print delayacct stats ON
    listen forever
    PID     231

    CPU             count     real total  virtual total    delay total  delay average
                     6247     1859000000     2154070021     1674255063          0.268ms
    IO              count    delay total  delay average
                        0              0              0ms
    SWAP            count    delay total  delay average
                        0              0              0ms
    RECLAIM         count    delay total  delay average
                        0              0              0ms
    THRASHING       count    delay total  delay average
                        0              0              0ms
    COMPACT         count    delay total  delay average
                        3          72758              0ms
    WPCOPY          count    delay total  delay average
                     3635      271567604              0ms

[1] commit 31cc5bc4af70("mm: support GUP-triggered unsharing of anonymous pages")

Link: https://lkml.kernel.org/r/20220409014342.2505532-1-yang.yang29@zte.com.cn
Signed-off-by: Yang Yang 
Reviewed-by: David Hildenbrand 
Reviewed-by: Jiang Xuexin 
Reviewed-by: Ran Xiaokai 
Reviewed-by: wangyong 
Cc: Jonathan Corbet 
Cc: Balbir Singh 
Cc: Mike Kravetz 
Cc: Stephen Rothwell 
Signed-off-by: Andrew Morton

kernel/delayacct: move delayacct sysctls to its own file

2022-04-06T20:43:44+00:00

kernel/sysctl.c is a kitchen sink where everyone leaves their dirty
dishes, this makes it very difficult to maintain.

To help with this maintenance let's start by moving sysctls to places
where they actually belong.  The proc sysctl maintainers do not want to
know what sysctl knobs you wish to add for your own piece of code, we
just care about the core logic.

All filesystem syctls now get reviewed by fs folks. This commit
follows the commit of fs, move the delayacct sysctl to its own file,
kernel/delayacct.c.

Signed-off-by: tangmeng 
Signed-off-by: Luis Chamberlain

delayacct: track delays from memory compact

2022-01-20T06:52:55+00:00

Delay accounting does not track the delay of memory compact.  When there
is not enough free memory, tasks can spend a amount of their time
waiting for compact.

To get the impact of tasks in direct memory compact, measure the delay
when allocating memory through memory compact.

Also update tools/accounting/getdelays.c:

    / # ./getdelays_next  -di -p 304
    print delayacct stats ON
    printing IO accounting
    PID     304

    CPU             count     real total  virtual total    delay total  delay average
                      277      780000000      849039485       18877296          0.068ms
    IO              count    delay total  delay average
                        0              0              0ms
    SWAP            count    delay total  delay average
                        0              0              0ms
    RECLAIM         count    delay total  delay average
                        5    11088812685           2217ms
    THRASHING       count    delay total  delay average
                        0              0              0ms
    COMPACT         count    delay total  delay average
                        3          72758              0ms
    watch: read=0, write=0, cancelled_write=0

Link: https://lkml.kernel.org/r/1638619795-71451-1-git-send-email-wang.yong12@zte.com.cn
Signed-off-by: wangyong 
Reviewed-by: Jiang Xuexin 
Reviewed-by: Zhang Wenya 
Reviewed-by: Yang Yang 
Reviewed-by: Balbir Singh 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

delayacct: support swapin delay accounting for swapping without blkio

2022-01-20T06:52:55+00:00

Currently delayacct accounts swapin delay only for swapping that cause
blkio.  If we use zram for swapping, tools/accounting/getdelays can't
get any SWAP delay.

It's useful to get zram swapin delay information, for example to adjust
compress algorithm or /proc/sys/vm/swappiness.

Reference to PSI, it accounts any kind of swapping by doing its work in
swap_readpage(), no matter whether swapping causes blkio.  Let delayacct
do the similar work.

Link: https://lkml.kernel.org/r/20211112083813.8559-1-yang.yang29@zte.com.cn
Signed-off-by: Yang Yang 
Reported-by: Zeal Robot 
Cc: Balbir Singh 
Cc: Peter Zijlstra 
Cc: Ingo Molnar 
Cc: Johannes Weiner 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

delayacct: Add sysctl to enable at runtime

2021-05-12T09:43:25+00:00

Just like sched_schedstats, allow runtime enabling (and disabling) of
delayacct. This is useful if one forgot to add the delayacct boot time
option.

Signed-off-by: Peter Zijlstra (Intel) 
Link: https://lkml.kernel.org/r/YJkhebGJAywaZowX@hirez.programming.kicks-ass.net

delayacct: Default disabled

2021-05-12T09:43:25+00:00

Assuming this stuff isn't actually used much; disable it by default
and avoid allocating and tracking the task_delay_info structure.

taskstats is changed to still report the regular sched and sched_info
and only skip the missing task_delay_info fields instead of not
reporting anything.

Signed-off-by: Peter Zijlstra (Intel) 
Reviewed-by: Thomas Gleixner 
Reviewed-by: Ingo Molnar 
Link: https://lkml.kernel.org/r/20210505111525.308018373@infradead.org

delayacct: Add static_branch in scheduler hooks

2021-05-12T09:43:25+00:00

Cheaper when delayacct is disabled.

Signed-off-by: Peter Zijlstra (Intel) 
Reviewed-by: Thomas Gleixner 
Reviewed-by: Ingo Molnar 
Acked-by: Balbir Singh 
Acked-by: Johannes Weiner 
Link: https://lkml.kernel.org/r/20210505111525.248028369@infradead.org

sched: Simplify sched_info_on()

2021-05-12T09:43:24+00:00

The situation around sched_info is somewhat complicated, it is used by
sched_stats and delayacct and, indirectly, kvm.

If SCHEDSTATS=Y (but disabled by default) sched_info_on() is
unconditionally true -- this is the case for all distro kernel configs
I checked.

If for some reason SCHEDSTATS=N, but TASK_DELAY_ACCT=Y, then
sched_info_on() can return false when delayacct is disabled,
presumably because there would be no other users left; except kvm is.

Instead of complicating matters further by accurately accounting
sched_stat and kvm state, simply unconditionally enable when
SCHED_INFO=Y, matching the common distro case.

Signed-off-by: Peter Zijlstra (Intel) 
Reviewed-by: Thomas Gleixner 
Reviewed-by: Ingo Molnar 
Acked-by: Johannes Weiner 
Link: https://lkml.kernel.org/r/20210505111525.121458839@infradead.org