This patch adds support for the proposed type match relation to
relo_core where it is shared between userspace and kernel. It plumbs
through both kernel-side and libbpf-side support.
The matching relation is defined as follows (copy from source):
- modifiers and typedefs are stripped (and, hence, effectively ignored)
- generally speaking types need to be of same kind (struct vs. struct, union
vs. union, etc.)
- exceptions are struct/union behind a pointer which could also match a
forward declaration of a struct or union, respectively, and enum vs.
enum64 (see below)
Then, depending on type:
- integers:
- match if size and signedness match
- arrays & pointers:
- target types are recursively matched
- structs & unions:
- local members need to exist in target with the same name
- for each member we recursively check match unless it is already behind a
pointer, in which case we only check matching names and compatible kind
- enums:
- local variants have to have a match in target by symbolic name (but not
numeric value)
- size has to match (but enum may match enum64 and vice versa)
- function pointers:
- number and position of arguments in local type has to match target
- for each argument and the return value we recursively check match
Signed-off-by: Daniel Müller <deso@posteo.net>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20220628160127.607834-5-deso@posteo.net
Use try_cmpxchg instead of cmpxchg (*ptr, old, new) != old in
set_nr_{and_not,if}_polling. x86 cmpxchg returns success in ZF flag,
so this change saves a compare after cmpxchg.
The definition of cmpxchg based fetch_or was changed in the
same way as atomic_fetch_##op definitions were changed
in e6790e4b5d.
Also declare these two functions as inline to ensure inlining. In the
case of set_nr_and_not_polling, the compiler (gcc) tries to outsmart
itself by constructing the boolean return value with logic operations
on the fetched value, and these extra operations enlarge the function
over the inlining threshold value.
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220629151552.6015-1-ubizjak@gmail.com
4feee7d126 previously added per-task forced idle accounting. This patch
extends this to also include cgroups.
rstat is used for cgroup accounting, except for the root, which uses
kcpustat in order to bypass the need for doing an rstat flush when
reading root stats.
Only cgroup v2 is supported. Similar to the task accounting, the cgroup
accounting requires that schedstats is enabled.
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lkml.kernel.org/r/20220629211426.3329954-1-joshdon@google.com
Currently shrinkers are anonymous objects. For debugging purposes they
can be identified by count/scan function names, but it's not always
useful: e.g. for superblock's shrinkers it's nice to have at least an
idea of to which superblock the shrinker belongs.
This commit adds names to shrinkers. register_shrinker() and
prealloc_shrinker() functions are extended to take a format and arguments
to master a name.
In some cases it's not possible to determine a good name at the time when
a shrinker is allocated. For such cases shrinker_debugfs_rename() is
provided.
The expected format is:
<subsystem>-<shrinker_type>[:<instance>]-<id>
For some shrinkers an instance can be encoded as (MAJOR:MINOR) pair.
After this change the shrinker debugfs directory looks like:
$ cd /sys/kernel/debug/shrinker/
$ ls
dquota-cache-16 sb-devpts-28 sb-proc-47 sb-tmpfs-42
mm-shadow-18 sb-devtmpfs-5 sb-proc-48 sb-tmpfs-43
mm-zspool:zram0-34 sb-hugetlbfs-17 sb-pstore-31 sb-tmpfs-44
rcu-kfree-0 sb-hugetlbfs-33 sb-rootfs-2 sb-tmpfs-49
sb-aio-20 sb-iomem-12 sb-securityfs-6 sb-tracefs-13
sb-anon_inodefs-15 sb-mqueue-21 sb-selinuxfs-22 sb-xfs:vda1-36
sb-bdev-3 sb-nsfs-4 sb-sockfs-8 sb-zsmalloc-19
sb-bpf-32 sb-pipefs-14 sb-sysfs-26 thp-deferred_split-10
sb-btrfs:vda2-24 sb-proc-25 sb-tmpfs-1 thp-zero-9
sb-cgroup2-30 sb-proc-39 sb-tmpfs-27 xfs-buf:vda1-37
sb-configfs-23 sb-proc-41 sb-tmpfs-29 xfs-inodegc:vda1-38
sb-dax-11 sb-proc-45 sb-tmpfs-35
sb-debugfs-7 sb-proc-46 sb-tmpfs-40
[roman.gushchin@linux.dev: fix build warnings]
Link: https://lkml.kernel.org/r/Yr+ZTnLb9lJk6fJO@castle
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lkml.kernel.org/r/20220601032227.4076670-4-roman.gushchin@linux.dev
Signed-off-by: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Cc: Dave Chinner <dchinner@redhat.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Kent Overstreet <kent.overstreet@gmail.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Daniel Borkmann says:
====================
pull-request: bpf 2022-07-02
We've added 7 non-merge commits during the last 14 day(s) which contain
a total of 6 files changed, 193 insertions(+), 86 deletions(-).
The main changes are:
1) Fix clearing of page contiguity when unmapping XSK pool, from Ivan Malov.
2) Two verifier fixes around bounds data propagation, from Daniel Borkmann.
3) Fix fprobe sample module's parameter descriptions, from Masami Hiramatsu.
4) General BPF maintainer entry revamp to better scale patch reviews.
* https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
bpf, selftests: Add verifier test case for jmp32's jeq/jne
bpf, selftests: Add verifier test case for imm=0,umin=0,umax=1 scalar
bpf: Fix insufficient bounds propagation from adjust_scalar_min_max_vals
bpf: Fix incorrect verifier simulation around jmp32's jeq/jne
xsk: Clear page contiguity bit when unmapping pool
bpf, docs: Better scale maintenance of BPF subsystem
fprobe, samples: Add module parameter descriptions
====================
Link: https://lore.kernel.org/r/20220701230121.10354-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Most in-kernel tests (such as KUnit tests) are not supposed to run on
production systems: they may do deliberately illegal things to trigger
errors, and have security implications (for example, KUnit assertions
will often deliberately leak kernel addresses).
Add a new taint type, TAINT_TEST to signal that a test has been run.
This will be printed as 'N' (originally for kuNit, as every other
sensible letter was taken.)
This should discourage people from running these tests on production
systems, and to make it easier to tell if tests have been run
accidentally (by loading the wrong configuration, etc.)
Acked-by: Luis Chamberlain <mcgrof@kernel.org>
Reviewed-by: Brendan Higgins <brendanhiggins@google.com>
Signed-off-by: David Gow <davidgow@google.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
When CONFIG_MODULE_UNLOAD is not selected, 'exit' is
set but never used.
It is not possible to replace the #ifdef CONFIG_MODULE_UNLOAD by
IS_ENABLED(CONFIG_MODULE_UNLOAD) because mod->exit doesn't exist
when CONFIG_MODULE_UNLOAD is not selected.
And because of the rcu_read_lock_sched() section it is not easy
to regroup everything in a single #ifdef. Let's regroup partially
and add missing #ifdef to completely opt out the use of
'exit' when CONFIG_MODULE_UNLOAD is not selected.
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
cppcheck reports the following warnings:
kernel/module/main.c:1455:26: warning: Redundant assignment of 'mod->core_layout.size' to itself. [selfAssignment]
mod->core_layout.size = strict_align(mod->core_layout.size);
^
kernel/module/main.c:1489:26: warning: Redundant assignment of 'mod->init_layout.size' to itself. [selfAssignment]
mod->init_layout.size = strict_align(mod->init_layout.size);
^
kernel/module/main.c:1493:26: warning: Redundant assignment of 'mod->init_layout.size' to itself. [selfAssignment]
mod->init_layout.size = strict_align(mod->init_layout.size);
^
kernel/module/main.c:1504:26: warning: Redundant assignment of 'mod->init_layout.size' to itself. [selfAssignment]
mod->init_layout.size = strict_align(mod->init_layout.size);
^
kernel/module/main.c:1459:26: warning: Redundant assignment of 'mod->data_layout.size' to itself. [selfAssignment]
mod->data_layout.size = strict_align(mod->data_layout.size);
^
kernel/module/main.c:1463:26: warning: Redundant assignment of 'mod->data_layout.size' to itself. [selfAssignment]
mod->data_layout.size = strict_align(mod->data_layout.size);
^
kernel/module/main.c:1467:26: warning: Redundant assignment of 'mod->data_layout.size' to itself. [selfAssignment]
mod->data_layout.size = strict_align(mod->data_layout.size);
^
This is due to strict_align() being a no-op when
CONFIG_STRICT_MODULE_RWX is not selected.
Transform strict_align() macro into an inline function. It will
allow type checking and avoid the selfAssignment warning.
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
The commit 91fb02f315 ("module: Move kallsyms support into a separate
file") changed from using strlcpy() to using strscpy() which created a
buffer overflow. That happened because:
1) an incorrect value was passed as the buffer length
2) strscpy() (unlike strlcpy()) may copy beyond the length of the
input string when copying word-by-word.
The assumption was that because it was already known that the strings
being copied would fit in the space available, it was not necessary
to correctly set the buffer length. strscpy() breaks that assumption
because although it will not touch bytes beyond the given buffer length
it may write bytes beyond the input string length when writing
word-by-word.
The result of the buffer overflow is to corrupt the symbol type
information that follows. e.g.
$ sudo cat -v /proc/kallsyms | grep '\^' | head
ffffffffc0615000 ^@ rfcomm_session_get [rfcomm]
ffffffffc061c060 ^@ session_list [rfcomm]
ffffffffc06150d0 ^@ rfcomm_send_frame [rfcomm]
ffffffffc0615130 ^@ rfcomm_make_uih [rfcomm]
ffffffffc07ed58d ^@ bnep_exit [bnep]
ffffffffc07ec000 ^@ bnep_rx_control [bnep]
ffffffffc07ec1a0 ^@ bnep_session [bnep]
ffffffffc07e7000 ^@ input_leds_event [input_leds]
ffffffffc07e9000 ^@ input_leds_handler [input_leds]
ffffffffc07e7010 ^@ input_leds_disconnect [input_leds]
Notably, the null bytes (represented above by ^@) can confuse tools.
Fix by correcting the buffer length.
Fixes: 91fb02f315 ("module: Move kallsyms support into a separate file")
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
Kuee reported a corner case where the tnum becomes constant after the call
to __reg_bound_offset(), but the register's bounds are not, that is, its
min bounds are still not equal to the register's max bounds.
This in turn allows to leak pointers through turning a pointer register as
is into an unknown scalar via adjust_ptr_min_max_vals().
Before:
func#0 @0
0: R1=ctx(off=0,imm=0,umax=0,var_off=(0x0; 0x0)) R10=fp(off=0,imm=0,umax=0,var_off=(0x0; 0x0))
0: (b7) r0 = 1 ; R0_w=scalar(imm=1,umin=1,umax=1,var_off=(0x1; 0x0))
1: (b7) r3 = 0 ; R3_w=scalar(imm=0,umax=0,var_off=(0x0; 0x0))
2: (87) r3 = -r3 ; R3_w=scalar()
3: (87) r3 = -r3 ; R3_w=scalar()
4: (47) r3 |= 32767 ; R3_w=scalar(smin=-9223372036854743041,umin=32767,var_off=(0x7fff; 0xffffffffffff8000),s32_min=-2147450881)
5: (75) if r3 s>= 0x0 goto pc+1 ; R3_w=scalar(umin=9223372036854808575,var_off=(0x8000000000007fff; 0x7fffffffffff8000),s32_min=-2147450881,u32_min=32767)
6: (95) exit
from 5 to 7: R0=scalar(imm=1,umin=1,umax=1,var_off=(0x1; 0x0)) R1=ctx(off=0,imm=0,umax=0,var_off=(0x0; 0x0)) R3=scalar(umin=32767,umax=9223372036854775807,var_off=(0x7fff; 0x7fffffffffff8000),s32_min=-2147450881) R10=fp(off=0,imm=0,umax=0,var_off=(0x0; 0x0))
7: (d5) if r3 s<= 0x8000 goto pc+1 ; R3=scalar(umin=32769,umax=9223372036854775807,var_off=(0x7fff; 0x7fffffffffff8000),s32_min=-2147450881,u32_min=32767)
8: (95) exit
from 7 to 9: R0=scalar(imm=1,umin=1,umax=1,var_off=(0x1; 0x0)) R1=ctx(off=0,imm=0,umax=0,var_off=(0x0; 0x0)) R3=scalar(umin=32767,umax=32768,var_off=(0x7fff; 0x8000)) R10=fp(off=0,imm=0,umax=0,var_off=(0x0; 0x0))
9: (07) r3 += -32767 ; R3_w=scalar(imm=0,umax=1,var_off=(0x0; 0x0)) <--- [*]
10: (95) exit
What can be seen here is that R3=scalar(umin=32767,umax=32768,var_off=(0x7fff;
0x8000)) after the operation R3 += -32767 results in a 'malformed' constant, that
is, R3_w=scalar(imm=0,umax=1,var_off=(0x0; 0x0)). Intersecting with var_off has
not been done at that point via __update_reg_bounds(), which would have improved
the umax to be equal to umin.
Refactor the tnum <> min/max bounds information flow into a reg_bounds_sync()
helper and use it consistently everywhere. After the fix, bounds have been
corrected to R3_w=scalar(imm=0,umax=0,var_off=(0x0; 0x0)) and thus the register
is regarded as a 'proper' constant scalar of 0.
After:
func#0 @0
0: R1=ctx(off=0,imm=0,umax=0,var_off=(0x0; 0x0)) R10=fp(off=0,imm=0,umax=0,var_off=(0x0; 0x0))
0: (b7) r0 = 1 ; R0_w=scalar(imm=1,umin=1,umax=1,var_off=(0x1; 0x0))
1: (b7) r3 = 0 ; R3_w=scalar(imm=0,umax=0,var_off=(0x0; 0x0))
2: (87) r3 = -r3 ; R3_w=scalar()
3: (87) r3 = -r3 ; R3_w=scalar()
4: (47) r3 |= 32767 ; R3_w=scalar(smin=-9223372036854743041,umin=32767,var_off=(0x7fff; 0xffffffffffff8000),s32_min=-2147450881)
5: (75) if r3 s>= 0x0 goto pc+1 ; R3_w=scalar(umin=9223372036854808575,var_off=(0x8000000000007fff; 0x7fffffffffff8000),s32_min=-2147450881,u32_min=32767)
6: (95) exit
from 5 to 7: R0=scalar(imm=1,umin=1,umax=1,var_off=(0x1; 0x0)) R1=ctx(off=0,imm=0,umax=0,var_off=(0x0; 0x0)) R3=scalar(umin=32767,umax=9223372036854775807,var_off=(0x7fff; 0x7fffffffffff8000),s32_min=-2147450881) R10=fp(off=0,imm=0,umax=0,var_off=(0x0; 0x0))
7: (d5) if r3 s<= 0x8000 goto pc+1 ; R3=scalar(umin=32769,umax=9223372036854775807,var_off=(0x7fff; 0x7fffffffffff8000),s32_min=-2147450881,u32_min=32767)
8: (95) exit
from 7 to 9: R0=scalar(imm=1,umin=1,umax=1,var_off=(0x1; 0x0)) R1=ctx(off=0,imm=0,umax=0,var_off=(0x0; 0x0)) R3=scalar(umin=32767,umax=32768,var_off=(0x7fff; 0x8000)) R10=fp(off=0,imm=0,umax=0,var_off=(0x0; 0x0))
9: (07) r3 += -32767 ; R3_w=scalar(imm=0,umax=0,var_off=(0x0; 0x0)) <--- [*]
10: (95) exit
Fixes: b03c9f9fdc ("bpf/verifier: track signed and unsigned min/max values")
Reported-by: Kuee K1r0a <liulin063@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20220701124727.11153-2-daniel@iogearbox.net
Kuee reported a quirk in the jmp32's jeq/jne simulation, namely that the
register value does not match expectations for the fall-through path. For
example:
Before fix:
0: R1=ctx(off=0,imm=0) R10=fp0
0: (b7) r2 = 0 ; R2_w=P0
1: (b7) r6 = 563 ; R6_w=P563
2: (87) r2 = -r2 ; R2_w=Pscalar()
3: (87) r2 = -r2 ; R2_w=Pscalar()
4: (4c) w2 |= w6 ; R2_w=Pscalar(umin=563,umax=4294967295,var_off=(0x233; 0xfffffdcc),s32_min=-2147483085) R6_w=P563
5: (56) if w2 != 0x8 goto pc+1 ; R2_w=P571 <--- [*]
6: (95) exit
R0 !read_ok
After fix:
0: R1=ctx(off=0,imm=0) R10=fp0
0: (b7) r2 = 0 ; R2_w=P0
1: (b7) r6 = 563 ; R6_w=P563
2: (87) r2 = -r2 ; R2_w=Pscalar()
3: (87) r2 = -r2 ; R2_w=Pscalar()
4: (4c) w2 |= w6 ; R2_w=Pscalar(umin=563,umax=4294967295,var_off=(0x233; 0xfffffdcc),s32_min=-2147483085) R6_w=P563
5: (56) if w2 != 0x8 goto pc+1 ; R2_w=P8 <--- [*]
6: (95) exit
R0 !read_ok
As can be seen on line 5 for the branch fall-through path in R2 [*] is that
given condition w2 != 0x8 is false, verifier should conclude that r2 = 8 as
upper 32 bit are known to be zero. However, verifier incorrectly concludes
that r2 = 571 which is far off.
The problem is it only marks false{true}_reg as known in the switch for JE/NE
case, but at the end of the function, it uses {false,true}_{64,32}off to
update {false,true}_reg->var_off and they still hold the prior value of
{false,true}_reg->var_off before it got marked as known. The subsequent
__reg_combine_32_into_64() then propagates this old var_off and derives new
bounds. The information between min/max bounds on {false,true}_reg from
setting the register to known const combined with the {false,true}_reg->var_off
based on the old information then derives wrong register data.
Fix it by detangling the BPF_JEQ/BPF_JNE cases and updating relevant
{false,true}_{64,32}off tnums along with the register marking to known
constant.
Fixes: 3f50f132d8 ("bpf: Verifier, do explicit ALU32 bounds tracking")
Reported-by: Kuee K1r0a <liulin063@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20220701124727.11153-1-daniel@iogearbox.net
drivers/net/ethernet/microchip/sparx5/sparx5_switchdev.c
9c5de246c1 ("net: sparx5: mdb add/del handle non-sparx5 devices")
fbb89d02e3 ("net: sparx5: Allow mdb entries to both CPU and ports")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
add proc_dointvec_ms_jiffies_minmax to fit read msecs value to jiffies
with a limited range of values
Signed-off-by: Yuwei Wang <wangyuweihx@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
I don't see how to make it nice without introducing btf id lists
for the hooks where these helpers are allowed. Some LSM hooks
work on the locked sockets, some are triggering early and
don't grab any locks, so have two lists for now:
1. LSM hooks which trigger under socket lock - minority of the hooks,
but ideal case for us, we can expose existing BTF-based helpers
2. LSM hooks which trigger without socket lock, but they trigger
early in the socket creation path where it should be safe to
do setsockopt without any locks
3. The rest are prohibited. I'm thinking that this use-case might
be a good gateway to sleeping lsm cgroup hooks in the future.
We can either expose lock/unlock operations (and add tracking
to the verifier) or have another set of bpf_setsockopt
wrapper that grab the locks and might sleep.
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-7-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
We have two options:
1. Treat all BPF_LSM_CGROUP the same, regardless of attach_btf_id
2. Treat BPF_LSM_CGROUP+attach_btf_id as a separate hook point
I was doing (2) in the original patch, but switching to (1) here:
* bpf_prog_query returns all attached BPF_LSM_CGROUP programs
regardless of attach_btf_id
* attach_btf_id is exported via bpf_prog_info
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-6-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Allow attaching to lsm hooks in the cgroup context.
Attaching to per-cgroup LSM works exactly like attaching
to other per-cgroup hooks. New BPF_LSM_CGROUP is added
to trigger new mode; the actual lsm hook we attach to is
signaled via existing attach_btf_id.
For the hooks that have 'struct socket' or 'struct sock' as its first
argument, we use the cgroup associated with that socket. For the rest,
we use 'current' cgroup (this is all on default hierarchy == v2 only).
Note that for some hooks that work on 'struct sock' we still
take the cgroup from 'current' because some of them work on the socket
that hasn't been properly initialized yet.
Behind the scenes, we allocate a shim program that is attached
to the trampoline and runs cgroup effective BPF programs array.
This shim has some rudimentary ref counting and can be shared
between several programs attaching to the same lsm hook from
different cgroups.
Note that this patch bloats cgroup size because we add 211
cgroup_bpf_attach_type(s) for simplicity sake. This will be
addressed in the subsequent patch.
Also note that we only add non-sleepable flavor for now. To enable
sleepable use-cases, bpf_prog_run_array_cg has to grab trace rcu,
shim programs have to be freed via trace rcu, cgroup_bpf.effective
should be also trace-rcu-managed + maybe some other changes that
I'm not aware of.
Reviewed-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/20220628174314.1216643-4-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
find_energy_efficient_cpu() integrates a margin to protect tasks from
bouncing back and forth from a CPU to another. This margin is set as being
6% of the total current energy estimated on the system. This however does
not work for two reasons:
1. The energy estimation is not a good absolute value:
compute_energy() used in feec() is a good estimation for task placement as
it allows to compare the energy with and without a task. The computed
delta will give a good overview of the cost for a certain task placement.
It, however, doesn't work as an absolute estimation for the total energy
of the system. First it adds the contribution to idle CPUs into the
energy, second it mixes util_avg with util_est values. util_avg contains
the near history for a CPU usage, it doesn't tell at all what the current
utilization is. A system that has been quite busy in the near past will
hold a very high energy and then a high margin preventing any task
migration to a lower capacity CPU, wasting energy. It even creates a
negative feedback loop: by holding the tasks on a less efficient CPU, the
margin contributes in keeping the energy high.
2. The margin handicaps small tasks:
On a system where the workload is composed mostly of small tasks (which is
often the case on Android), the overall energy will be high enough to
create a margin none of those tasks can cross. On a Pixel4, a small
utilization of 5% on all the CPUs creates a global estimated energy of 140
joules, as per the Energy Model declaration of that same device. This
means, after applying the 6% margin that any migration must save more than
8 joules to happen. No task with a utilization lower than 40 would then be
able to migrate away from the biggest CPU of the system.
The 6% of the overall system energy was brought by the following patch:
(eb92692b25 sched/fair: Speed-up energy-aware wake-ups)
It was previously 6% of the prev_cpu energy. Also, the following one
made this margin value conditional on the clusters where the task fits:
(8d4c97c105 sched/fair: Only compute base_energy_pd if necessary)
We could simply revert that margin change to what it was, but the original
version didn't have strong grounds neither and as demonstrated in (1.) the
estimated energy isn't a good absolute value. Instead, removing it
completely. It is indeed, made possible by recent changes that improved
energy estimation comparison fairness (sched/fair: Remove task_util from
effective utilization in feec()) (PM: EM: Increase energy calculation
precision) and task utilization stabilization (sched/fair: Decay task
util_avg during migration)
Without a margin, we could have feared bouncing between CPUs. But running
LISA's eas_behaviour test coverage on three different platforms (Hikey960,
RB-5 and DB-845) showed no issue.
Removing the energy margin enables more energy-optimized placements for a
more energy efficient system.
Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>
Signed-off-by: Vincent Donnefort <vdonnefort@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Link: https://lkml.kernel.org/r/20220621090414.433602-8-vdonnefort@google.com
The energy estimation in find_energy_efficient_cpu() (feec()) relies on
the computation of the effective utilization for each CPU of a perf domain
(PD). This effective utilization is then used as an estimation of the busy
time for this pd. The function effective_cpu_util() which gives this value,
scales the utilization relative to IRQ pressure on the CPU to take into
account that the IRQ time is hidden from the task clock. The IRQ scaling is
as follow:
effective_cpu_util = irq + (cpu_cap - irq)/cpu_cap * util
Where util is the sum of CFS/RT/DL utilization, cpu_cap the capacity of
the CPU and irq the IRQ avg time.
If now we take as an example a task placement which doesn't raise the OPP
on the candidate CPU, we can write the energy delta as:
delta = OPPcost/cpu_cap * (effective_cpu_util(cpu_util + task_util) -
effective_cpu_util(cpu_util))
= OPPcost/cpu_cap * (cpu_cap - irq)/cpu_cap * task_util
We end-up with an energy delta depending on the IRQ avg time, which is a
problem: first the time spent on IRQs by a CPU has no effect on the
additional energy that would be consumed by a task. Second, we don't want
to favour a CPU with a higher IRQ avg time value.
Nonetheless, we need to take the IRQ avg time into account. If a task
placement raises the PD's frequency, it will increase the energy cost for
the entire time where the CPU is busy. A solution is to only use
effective_cpu_util() with the CPU contribution part. The task contribution
is added separately and scaled according to prev_cpu's IRQ time.
No change for the FREQUENCY_UTIL component of the energy estimation. We
still want to get the actual frequency that would be selected after the
task placement.
Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>
Signed-off-by: Vincent Donnefort <vdonnefort@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Link: https://lkml.kernel.org/r/20220621090414.433602-7-vdonnefort@google.com
The Perf Domain (PD) cpumask (struct em_perf_domain.cpus) stays
invariant after Energy Model creation, i.e. it is not updated after
CPU hotplug operations.
That's why the PD mask is used in conjunction with the cpu_online_mask
(or Sched Domain cpumask). Thereby the cpu_online_mask is fetched
multiple times (in compute_energy()) during a run-queue selection
for a task.
cpu_online_mask may change during this time which can lead to wrong
energy calculations.
To be able to avoid this, use the select_rq_mask per-cpu cpumask to
create a cpumask out of PD cpumask and cpu_online_mask and pass it
through the function calls of the EAS run-queue selection path.
The PD cpumask for max_spare_cap_cpu/compute_prev_delta selection
(find_energy_efficient_cpu()) is now ANDed not only with the SD mask
but also with the cpu_online_mask. This is fine since this cpumask
has to be in syc with the one used for energy computation
(compute_energy()).
An exclusive cpuset setup with at least one asymmetric CPU capacity
island (hence the additional AND with the SD cpumask) is the obvious
exception here.
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Link: https://lkml.kernel.org/r/20220621090414.433602-6-vdonnefort@google.com
On 21/06/2022 11:04, Vincent Donnefort wrote:
> From: Dietmar Eggemann <dietmar.eggemann@arm.com>
https://lkml.kernel.org/r/202206221253.ZVyGQvPX-lkp@intel.com discovered
that this patch doesn't build anymore (on tip sched/core or linux-next)
because of commit f5b2eeb499 ("sched/fair: Consider CPU affinity when
allowing NUMA imbalance in find_idlest_group()").
New version of [PATCH v11 4/7] sched/fair: Rename select_idle_mask to
select_rq_mask below.
-- >8 --
Decouple the name of the per-cpu cpumask select_idle_mask from its usage
in select_idle_[cpu/capacity]() of the CFS run-queue selection
(select_task_rq_fair()).
This is to support the reuse of this cpumask in the Energy Aware
Scheduling (EAS) path (find_energy_efficient_cpu()) of the CFS run-queue
selection.
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Link: https://lkml.kernel.org/r/250691c7-0e2b-05ab-bedf-b245c11d9400@arm.com
effective_cpu_util() already has a `int cpu' parameter which allows to
retrieve the CPU capacity scale factor (or maximum CPU capacity) inside
this function via an arch_scale_cpu_capacity(cpu).
A lot of code calling effective_cpu_util() (or the shim
sched_cpu_util()) needs the maximum CPU capacity, i.e. it will call
arch_scale_cpu_capacity() already.
But not having to pass it into effective_cpu_util() will make the EAS
wake-up code easier, especially when the maximum CPU capacity reduced
by the thermal pressure is passed through the EAS wake-up functions.
Due to the asymmetric CPU capacity support of arm/arm64 architectures,
arch_scale_cpu_capacity(int cpu) is a per-CPU variable read access via
per_cpu(cpu_scale, cpu) on such a system.
On all other architectures it is a a compile-time constant
(SCHED_CAPACITY_SCALE).
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Link: https://lkml.kernel.org/r/20220621090414.433602-4-vdonnefort@google.com
Before being migrated to a new CPU, a task sees its PELT values
synchronized with rq last_update_time. Once done, that same task will also
have its sched_avg last_update_time reset. This means the time between
the migration and the last clock update will not be accounted for in
util_avg and a discontinuity will appear. This issue is amplified by the
PELT clock scaling. It takes currently one tick after the CPU being idle
to let clock_pelt catching up clock_task.
This is especially problematic for asymmetric CPU capacity systems which
need stable util_avg signals for task placement and energy estimation.
Ideally, this problem would be solved by updating the runqueue clocks
before the migration. But that would require taking the runqueue lock
which is quite expensive [1]. Instead estimate the missing time and update
the task util_avg with that value.
To that end, we need sched_clock_cpu() but it is a costly function. Limit
the usage to the case where the source CPU is idle as we know this is when
the clock is having the biggest risk of being outdated.
See comment in migrate_se_pelt_lag() for more details about how the PELT
value is estimated. Notice though this estimation doesn't take into account
IRQ and Paravirt time.
[1] https://lkml.kernel.org/r/20190709115759.10451-1-chris.redpath@arm.com
Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>
Signed-off-by: Vincent Donnefort <vdonnefort@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Link: https://lkml.kernel.org/r/20220621090414.433602-3-vdonnefort@google.com
Introducing macro helpers u64_u32_{store,load}() to factorize lockless
accesses to u64 variables for 32-bits architectures.
Users are for now cfs_rq.min_vruntime and sched_avg.last_update_time. To
accommodate the later where the copy lies outside of the structure
(cfs_rq.last_udpate_time_copy instead of sched_avg.last_update_time_copy),
use the _copy() version of those helpers.
Those new helpers encapsulate smp_rmb() and smp_wmb() synchronization and
therefore, have a small penalty for 32-bits machines in set_task_rq_fair()
and init_cfs_rq().
Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>
Signed-off-by: Vincent Donnefort <vdonnefort@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Lukasz Luba <lukasz.luba@arm.com>
Link: https://lkml.kernel.org/r/20220621090414.433602-2-vdonnefort@google.com
Sometimes we want to know an accurate number of samples even if it's
lost. Currenlty PERF_RECORD_LOST is generated for a ring-buffer which
might be shared with other events. So it's hard to know per-event
lost count.
Add event->lost_samples field and PERF_FORMAT_LOST to retrieve it from
userspace.
Original-patch-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220616180623.1358843-1-namhyung@kernel.org
[Problem Statement]
select_idle_cpu() might spend too much time searching for an idle CPU,
when the system is overloaded.
The following histogram is the time spent in select_idle_cpu(),
when running 224 instances of netperf on a system with 112 CPUs
per LLC domain:
@usecs:
[0] 533 | |
[1] 5495 | |
[2, 4) 12008 | |
[4, 8) 239252 | |
[8, 16) 4041924 |@@@@@@@@@@@@@@ |
[16, 32) 12357398 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ |
[32, 64) 14820255 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
[64, 128) 13047682 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ |
[128, 256) 8235013 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@ |
[256, 512) 4507667 |@@@@@@@@@@@@@@@ |
[512, 1K) 2600472 |@@@@@@@@@ |
[1K, 2K) 927912 |@@@ |
[2K, 4K) 218720 | |
[4K, 8K) 98161 | |
[8K, 16K) 37722 | |
[16K, 32K) 6715 | |
[32K, 64K) 477 | |
[64K, 128K) 7 | |
netperf latency usecs:
=======
case load Lat_99th std%
TCP_RR thread-224 257.39 ( 0.21)
The time spent in select_idle_cpu() is visible to netperf and might have a negative
impact.
[Symptom analysis]
The patch [1] from Mel Gorman has been applied to track the efficiency
of select_idle_sibling. Copy the indicators here:
SIS Search Efficiency(se_eff%):
A ratio expressed as a percentage of runqueues scanned versus
idle CPUs found. A 100% efficiency indicates that the target,
prev or recent CPU of a task was idle at wakeup. The lower the
efficiency, the more runqueues were scanned before an idle CPU
was found.
SIS Domain Search Efficiency(dom_eff%):
Similar, except only for the slower SIS
patch.
SIS Fast Success Rate(fast_rate%):
Percentage of SIS that used target, prev or
recent CPUs.
SIS Success rate(success_rate%):
Percentage of scans that found an idle CPU.
The test is based on Aubrey's schedtests tool, including netperf, hackbench,
schbench and tbench.
Test on vanilla kernel:
schedstat_parse.py -f netperf_vanilla.log
case load se_eff% dom_eff% fast_rate% success_rate%
TCP_RR 28 threads 99.978 18.535 99.995 100.000
TCP_RR 56 threads 99.397 5.671 99.964 100.000
TCP_RR 84 threads 21.721 6.818 73.632 100.000
TCP_RR 112 threads 12.500 5.533 59.000 100.000
TCP_RR 140 threads 8.524 4.535 49.020 100.000
TCP_RR 168 threads 6.438 3.945 40.309 99.999
TCP_RR 196 threads 5.397 3.718 32.320 99.982
TCP_RR 224 threads 4.874 3.661 25.775 99.767
UDP_RR 28 threads 99.988 17.704 99.997 100.000
UDP_RR 56 threads 99.528 5.977 99.970 100.000
UDP_RR 84 threads 24.219 6.992 76.479 100.000
UDP_RR 112 threads 13.907 5.706 62.538 100.000
UDP_RR 140 threads 9.408 4.699 52.519 100.000
UDP_RR 168 threads 7.095 4.077 44.352 100.000
UDP_RR 196 threads 5.757 3.775 35.764 99.991
UDP_RR 224 threads 5.124 3.704 28.748 99.860
schedstat_parse.py -f schbench_vanilla.log
(each group has 28 tasks)
case load se_eff% dom_eff% fast_rate% success_rate%
normal 1 mthread 99.152 6.400 99.941 100.000
normal 2 mthreads 97.844 4.003 99.908 100.000
normal 3 mthreads 96.395 2.118 99.917 99.998
normal 4 mthreads 55.288 1.451 98.615 99.804
normal 5 mthreads 7.004 1.870 45.597 61.036
normal 6 mthreads 3.354 1.346 20.777 34.230
normal 7 mthreads 2.183 1.028 11.257 21.055
normal 8 mthreads 1.653 0.825 7.849 15.549
schedstat_parse.py -f hackbench_vanilla.log
(each group has 28 tasks)
case load se_eff% dom_eff% fast_rate% success_rate%
process-pipe 1 group 99.991 7.692 99.999 100.000
process-pipe 2 groups 99.934 4.615 99.997 100.000
process-pipe 3 groups 99.597 3.198 99.987 100.000
process-pipe 4 groups 98.378 2.464 99.958 100.000
process-pipe 5 groups 27.474 3.653 89.811 99.800
process-pipe 6 groups 20.201 4.098 82.763 99.570
process-pipe 7 groups 16.423 4.156 77.398 99.316
process-pipe 8 groups 13.165 3.920 72.232 98.828
process-sockets 1 group 99.977 5.882 99.999 100.000
process-sockets 2 groups 99.927 5.505 99.996 100.000
process-sockets 3 groups 99.397 3.250 99.980 100.000
process-sockets 4 groups 79.680 4.258 98.864 99.998
process-sockets 5 groups 7.673 2.503 63.659 92.115
process-sockets 6 groups 4.642 1.584 58.946 88.048
process-sockets 7 groups 3.493 1.379 49.816 81.164
process-sockets 8 groups 3.015 1.407 40.845 75.500
threads-pipe 1 group 99.997 0.000 100.000 100.000
threads-pipe 2 groups 99.894 2.932 99.997 100.000
threads-pipe 3 groups 99.611 4.117 99.983 100.000
threads-pipe 4 groups 97.703 2.624 99.937 100.000
threads-pipe 5 groups 22.919 3.623 87.150 99.764
threads-pipe 6 groups 18.016 4.038 80.491 99.557
threads-pipe 7 groups 14.663 3.991 75.239 99.247
threads-pipe 8 groups 12.242 3.808 70.651 98.644
threads-sockets 1 group 99.990 6.667 99.999 100.000
threads-sockets 2 groups 99.940 5.114 99.997 100.000
threads-sockets 3 groups 99.469 4.115 99.977 100.000
threads-sockets 4 groups 87.528 4.038 99.400 100.000
threads-sockets 5 groups 6.942 2.398 59.244 88.337
threads-sockets 6 groups 4.359 1.954 49.448 87.860
threads-sockets 7 groups 2.845 1.345 41.198 77.102
threads-sockets 8 groups 2.871 1.404 38.512 74.312
schedstat_parse.py -f tbench_vanilla.log
case load se_eff% dom_eff% fast_rate% success_rate%
loopback 28 threads 99.976 18.369 99.995 100.000
loopback 56 threads 99.222 7.799 99.934 100.000
loopback 84 threads 19.723 6.819 70.215 100.000
loopback 112 threads 11.283 5.371 55.371 99.999
loopback 140 threads 0.000 0.000 0.000 0.000
loopback 168 threads 0.000 0.000 0.000 0.000
loopback 196 threads 0.000 0.000 0.000 0.000
loopback 224 threads 0.000 0.000 0.000 0.000
According to the test above, if the system becomes busy, the
SIS Search Efficiency(se_eff%) drops significantly. Although some
benchmarks would finally find an idle CPU(success_rate% = 100%), it is
doubtful whether it is worth it to search the whole LLC domain.
[Proposal]
It would be ideal to have a crystal ball to answer this question:
How many CPUs must a wakeup path walk down, before it can find an idle
CPU? Many potential metrics could be used to predict the number.
One candidate is the sum of util_avg in this LLC domain. The benefit
of choosing util_avg is that it is a metric of accumulated historic
activity, which seems to be smoother than instantaneous metrics
(such as rq->nr_running). Besides, choosing the sum of util_avg
would help predict the load of the LLC domain more precisely, because
SIS_PROP uses one CPU's idle time to estimate the total LLC domain idle
time.
In summary, the lower the util_avg is, the more select_idle_cpu()
should scan for idle CPU, and vice versa. When the sum of util_avg
in this LLC domain hits 85% or above, the scan stops. The reason to
choose 85% as the threshold is that this is the imbalance_pct(117)
when a LLC sched group is overloaded.
Introduce the quadratic function:
y = SCHED_CAPACITY_SCALE - p * x^2
and y'= y / SCHED_CAPACITY_SCALE
x is the ratio of sum_util compared to the CPU capacity:
x = sum_util / (llc_weight * SCHED_CAPACITY_SCALE)
y' is the ratio of CPUs to be scanned in the LLC domain,
and the number of CPUs to scan is calculated by:
nr_scan = llc_weight * y'
Choosing quadratic function is because:
[1] Compared to the linear function, it scans more aggressively when the
sum_util is low.
[2] Compared to the exponential function, it is easier to calculate.
[3] It seems that there is no accurate mapping between the sum of util_avg
and the number of CPUs to be scanned. Use heuristic scan for now.
For a platform with 112 CPUs per LLC, the number of CPUs to scan is:
sum_util% 0 5 15 25 35 45 55 65 75 85 86 ...
scan_nr 112 111 108 102 93 81 65 47 25 1 0 ...
For a platform with 16 CPUs per LLC, the number of CPUs to scan is:
sum_util% 0 5 15 25 35 45 55 65 75 85 86 ...
scan_nr 16 15 15 14 13 11 9 6 3 0 0 ...
Furthermore, to minimize the overhead of calculating the metrics in
select_idle_cpu(), borrow the statistics from periodic load balance.
As mentioned by Abel, on a platform with 112 CPUs per LLC, the
sum_util calculated by periodic load balance after 112 ms would
decay to about 0.5 * 0.5 * 0.5 * 0.7 = 8.75%, thus bringing a delay
in reflecting the latest utilization. But it is a trade-off.
Checking the util_avg in newidle load balance would be more frequent,
but it brings overhead - multiple CPUs write/read the per-LLC shared
variable and introduces cache contention. Tim also mentioned that,
it is allowed to be non-optimal in terms of scheduling for the
short-term variations, but if there is a long-term trend in the load
behavior, the scheduler can adjust for that.
When SIS_UTIL is enabled, the select_idle_cpu() uses the nr_scan
calculated by SIS_UTIL instead of the one from SIS_PROP. As Peter and
Mel suggested, SIS_UTIL should be enabled by default.
This patch is based on the util_avg, which is very sensitive to the
CPU frequency invariance. There is an issue that, when the max frequency
has been clamp, the util_avg would decay insanely fast when
the CPU is idle. Commit addca28512 ("cpufreq: intel_pstate: Handle no_turbo
in frequency invariance") could be used to mitigate this symptom, by adjusting
the arch_max_freq_ratio when turbo is disabled. But this issue is still
not thoroughly fixed, because the current code is unaware of the user-specified
max CPU frequency.
[Test result]
netperf and tbench were launched with 25% 50% 75% 100% 125% 150%
175% 200% of CPU number respectively. Hackbench and schbench were launched
by 1, 2 ,4, 8 groups. Each test lasts for 100 seconds and repeats 3 times.
The following is the benchmark result comparison between
baseline:vanilla v5.19-rc1 and compare:patched kernel. Positive compare%
indicates better performance.
Each netperf test is a:
netperf -4 -H 127.0.1 -t TCP/UDP_RR -c -C -l 100
netperf.throughput
=======
case load baseline(std%) compare%( std%)
TCP_RR 28 threads 1.00 ( 0.34) -0.16 ( 0.40)
TCP_RR 56 threads 1.00 ( 0.19) -0.02 ( 0.20)
TCP_RR 84 threads 1.00 ( 0.39) -0.47 ( 0.40)
TCP_RR 112 threads 1.00 ( 0.21) -0.66 ( 0.22)
TCP_RR 140 threads 1.00 ( 0.19) -0.69 ( 0.19)
TCP_RR 168 threads 1.00 ( 0.18) -0.48 ( 0.18)
TCP_RR 196 threads 1.00 ( 0.16) +194.70 ( 16.43)
TCP_RR 224 threads 1.00 ( 0.16) +197.30 ( 7.85)
UDP_RR 28 threads 1.00 ( 0.37) +0.35 ( 0.33)
UDP_RR 56 threads 1.00 ( 11.18) -0.32 ( 0.21)
UDP_RR 84 threads 1.00 ( 1.46) -0.98 ( 0.32)
UDP_RR 112 threads 1.00 ( 28.85) -2.48 ( 19.61)
UDP_RR 140 threads 1.00 ( 0.70) -0.71 ( 14.04)
UDP_RR 168 threads 1.00 ( 14.33) -0.26 ( 11.16)
UDP_RR 196 threads 1.00 ( 12.92) +186.92 ( 20.93)
UDP_RR 224 threads 1.00 ( 11.74) +196.79 ( 18.62)
Take the 224 threads as an example, the SIS search metrics changes are
illustrated below:
vanilla patched
4544492 +237.5% 15338634 sched_debug.cpu.sis_domain_search.avg
38539 +39686.8% 15333634 sched_debug.cpu.sis_failed.avg
128300000 -87.9% 15551326 sched_debug.cpu.sis_scanned.avg
5842896 +162.7% 15347978 sched_debug.cpu.sis_search.avg
There is -87.9% less CPU scans after patched, which indicates lower overhead.
Besides, with this patch applied, there is -13% less rq lock contention
in perf-profile.calltrace.cycles-pp._raw_spin_lock.raw_spin_rq_lock_nested
.try_to_wake_up.default_wake_function.woken_wake_function.
This might help explain the performance improvement - Because this patch allows
the waking task to remain on the previous CPU, rather than grabbing other CPUs'
lock.
Each hackbench test is a:
hackbench -g $job --process/threads --pipe/sockets -l 1000000 -s 100
hackbench.throughput
=========
case load baseline(std%) compare%( std%)
process-pipe 1 group 1.00 ( 1.29) +0.57 ( 0.47)
process-pipe 2 groups 1.00 ( 0.27) +0.77 ( 0.81)
process-pipe 4 groups 1.00 ( 0.26) +1.17 ( 0.02)
process-pipe 8 groups 1.00 ( 0.15) -4.79 ( 0.02)
process-sockets 1 group 1.00 ( 0.63) -0.92 ( 0.13)
process-sockets 2 groups 1.00 ( 0.03) -0.83 ( 0.14)
process-sockets 4 groups 1.00 ( 0.40) +5.20 ( 0.26)
process-sockets 8 groups 1.00 ( 0.04) +3.52 ( 0.03)
threads-pipe 1 group 1.00 ( 1.28) +0.07 ( 0.14)
threads-pipe 2 groups 1.00 ( 0.22) -0.49 ( 0.74)
threads-pipe 4 groups 1.00 ( 0.05) +1.88 ( 0.13)
threads-pipe 8 groups 1.00 ( 0.09) -4.90 ( 0.06)
threads-sockets 1 group 1.00 ( 0.25) -0.70 ( 0.53)
threads-sockets 2 groups 1.00 ( 0.10) -0.63 ( 0.26)
threads-sockets 4 groups 1.00 ( 0.19) +11.92 ( 0.24)
threads-sockets 8 groups 1.00 ( 0.08) +4.31 ( 0.11)
Each tbench test is a:
tbench -t 100 $job 127.0.0.1
tbench.throughput
======
case load baseline(std%) compare%( std%)
loopback 28 threads 1.00 ( 0.06) -0.14 ( 0.09)
loopback 56 threads 1.00 ( 0.03) -0.04 ( 0.17)
loopback 84 threads 1.00 ( 0.05) +0.36 ( 0.13)
loopback 112 threads 1.00 ( 0.03) +0.51 ( 0.03)
loopback 140 threads 1.00 ( 0.02) -1.67 ( 0.19)
loopback 168 threads 1.00 ( 0.38) +1.27 ( 0.27)
loopback 196 threads 1.00 ( 0.11) +1.34 ( 0.17)
loopback 224 threads 1.00 ( 0.11) +1.67 ( 0.22)
Each schbench test is a:
schbench -m $job -t 28 -r 100 -s 30000 -c 30000
schbench.latency_90%_us
========
case load baseline(std%) compare%( std%)
normal 1 mthread 1.00 ( 31.22) -7.36 ( 20.25)*
normal 2 mthreads 1.00 ( 2.45) -0.48 ( 1.79)
normal 4 mthreads 1.00 ( 1.69) +0.45 ( 0.64)
normal 8 mthreads 1.00 ( 5.47) +9.81 ( 14.28)
*Consider the Standard Deviation, this -7.36% regression might not be valid.
Also, a OLTP workload with a commercial RDBMS has been tested, and there
is no significant change.
There were concerns that unbalanced tasks among CPUs would cause problems.
For example, suppose the LLC domain is composed of 8 CPUs, and 7 tasks are
bound to CPU0~CPU6, while CPU7 is idle:
CPU0 CPU1 CPU2 CPU3 CPU4 CPU5 CPU6 CPU7
util_avg 1024 1024 1024 1024 1024 1024 1024 0
Since the util_avg ratio is 87.5%( = 7/8 ), which is higher than 85%,
select_idle_cpu() will not scan, thus CPU7 is undetected during scan.
But according to Mel, it is unlikely the CPU7 will be idle all the time
because CPU7 could pull some tasks via CPU_NEWLY_IDLE.
lkp(kernel test robot) has reported a regression on stress-ng.sock on a
very busy system. According to the sched_debug statistics, it might be caused
by SIS_UTIL terminates the scan and chooses a previous CPU earlier, and this
might introduce more context switch, especially involuntary preemption, which
impacts a busy stress-ng. This regression has shown that, not all benchmarks
in every scenario benefit from idle CPU scan limit, and it needs further
investigation.
Besides, there is slight regression in hackbench's 16 groups case when the
LLC domain has 16 CPUs. Prateek mentioned that we should scan aggressively
in an LLC domain with 16 CPUs. Because the cost to search for an idle one
among 16 CPUs is negligible. The current patch aims to propose a generic
solution and only considers the util_avg. Something like the below could
be applied on top of the current patch to fulfill the requirement:
if (llc_weight <= 16)
nr_scan = nr_scan * 32 / llc_weight;
For LLC domain with 16 CPUs, the nr_scan will be expanded to 2 times large.
The smaller the CPU number this LLC domain has, the larger nr_scan will be
expanded. This needs further investigation.
There is also ongoing work[2] from Abel to filter out the busy CPUs during
wakeup, to further speed up the idle CPU scan. And it could be a following-up
optimization on top of this change.
Suggested-by: Tim Chen <tim.c.chen@intel.com>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Yicong Yang <yangyicong@hisilicon.com>
Tested-by: Mohini Narkhede <mohini.narkhede@intel.com>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://lore.kernel.org/r/20220612163428.849378-1-yu.c.chen@intel.com
sched_setattr(2) issues via kernel/sched/core.c:__sched_setscheduler()
a CAP_SYS_NICE audit event unconditionally, even when the requested
operation does not require that capability / is unprivileged, i.e. for
reducing niceness.
This is relevant in connection with SELinux, where a capability check
results in a policy decision and by default a denial message on
insufficient permission is issued.
It can lead to three undesired cases:
1. A denial message is generated, even in case the operation was an
unprivileged one and thus the syscall succeeded, creating noise.
2. To avoid the noise from 1. the policy writer adds a rule to ignore
those denial messages, hiding future syscalls, where the task
performs an actual privileged operation, leading to hidden limited
functionality of that task.
3. To avoid the noise from 1. the policy writer adds a rule to allow
the task the capability CAP_SYS_NICE, while it does not need it,
violating the principle of least privilege.
Conduct privilged/unprivileged categorization first and perform a
capable test (and at most once) only if needed.
Signed-off-by: Christian Göttsche <cgzones@googlemail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220615152505.310488-1-cgzones@googlemail.com
