Low priority tasks (e.g., SCHED_OTHER) can suffer starvation if tasks
with higher priority (e.g., SCHED_FIFO) monopolize CPU(s).
RT Throttling has been introduced a while ago as a (mostly debug)
countermeasure one can utilize to reserve some CPU time for low priority
tasks (usually background type of work, e.g. workqueues, timers, etc.).
It however has its own problems (see documentation) and the undesired
effect of unconditionally throttling FIFO tasks even when no lower
priority activity needs to run (there are mechanisms to fix this issue
as well, but, again, with their own problems).
Introduce deadline servers to service low priority tasks needs under
starvation conditions. Deadline servers are built extending SCHED_DEADLINE
implementation to allow 2-level scheduling (a sched_deadline entity
becomes a container for lower priority scheduling entities).
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Daniel Bristot de Oliveira <bristot@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/4968601859d920335cf85822eb573a5f179f04b8.1699095159.git.bristot@kernel.org
Now that trace_sched_stat_runtime() no longer takes a vruntime
argument, the task specific bits are identical between
update_curr_common() and update_curr().
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tracing the runtime delta makes sense, observer can sum over time.
Tracing the absolute vruntime makes less sense, inconsistent:
absolute-vs-delta, but also vruntime delta can be computed from
runtime delta.
Removing the vruntime thing also makes the two tracepoint sites
identical, allowing to unify the code in a later patch.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Since the RB-tree is now sorted by deadline, let's first try the
leftmost entity which has the earliest virtual deadline. I've done
some benchmarks to see its effectiveness.
All the benchmarks are done inside a normal cpu cgroup in a clean
environment with cpu turbo disabled, on a dual-CPU Intel Xeon(R)
Platinum 8260 with 2 NUMA nodes each of which has 24C/48T.
hackbench: process/thread + pipe/socket + 1/2/4/8 groups
netperf: TCP/UDP + STREAM/RR + 24/48/72/96/192 threads
tbench: loopback 24/48/72/96/192 threads
schbench: 1/2/4/8 mthreads
direct: cfs_rq has only one entity
parity: RUN_TO_PARITY
fast: O(1) fastpath
slow: heap search
(%) direct parity fast slow
hackbench 92.95 2.02 4.91 0.12
netperf 68.08 6.60 24.18 1.14
tbench 67.55 11.22 20.61 0.62
schbench 69.91 2.65 25.73 1.71
The above results indicate that this fastpath really makes task
selection more efficient.
Signed-off-by: Abel Wu <wuyun.abel@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20231115033647.80785-4-wuyun.abel@bytedance.com
Since commit fc137c0dda ("sched/numa: enhance vma scanning logic")
NUMA Balancing allows updating PTEs to trap NUMA hinting faults if the
task had previously accessed VMA. However unconditional scan of VMAs are
allowed during initial phase of VMA creation until process's
mm numa_scan_seq reaches 2 even though current task had not accessed VMA.
Rationale:
- Without initial scan subsequent PTE update may never happen.
- Give fair opportunity to all the VMAs to be scanned and subsequently
understand the access pattern of all the VMAs.
But it has a corner case where, if a VMA is created after some time,
process's mm numa_scan_seq could be already greater than 2.
For e.g., values of mm numa_scan_seq when VMAs are created by running
mmtest autonuma benchmark briefly looks like:
start_seq=0 : 459
start_seq=2 : 138
start_seq=3 : 144
start_seq=4 : 8
start_seq=8 : 1
start_seq=9 : 1
This results in no unconditional PTE updates for those VMAs created after
some time.
Fix:
- Note down the initial value of mm numa_scan_seq in per VMA start_seq.
- Allow unconditional scan till start_seq + 2.
Result:
SUT: AMD EPYC Milan with 2 NUMA nodes 256 cpus.
base kernel: upstream 6.6-rc6 with Mels patches [1] applied.
kernbench
========== base patched %gain
Amean elsp-128 165.09 ( 0.00%) 164.78 * 0.19%*
Duration User 41404.28 41375.08
Duration System 9862.22 9768.48
Duration Elapsed 519.87 518.72
Ops NUMA PTE updates 1041416.00 831536.00
Ops NUMA hint faults 263296.00 220966.00
Ops NUMA pages migrated 258021.00 212769.00
Ops AutoNUMA cost 1328.67 1114.69
autonumabench
NUMA01_THREADLOCAL
==================
Amean elsp-NUMA01_THREADLOCAL 81.79 (0.00%) 67.74 * 17.18%*
Duration User 54832.73 47379.67
Duration System 75.00 185.75
Duration Elapsed 576.72 476.09
Ops NUMA PTE updates 394429.00 11121044.00
Ops NUMA hint faults 1001.00 8906404.00
Ops NUMA pages migrated 288.00 2998694.00
Ops AutoNUMA cost 7.77 44666.84
Signed-off-by: Raghavendra K T <raghavendra.kt@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Link: https://lore.kernel.org/r/2ea7cbce80ac7c62e90cbfb9653a7972f902439f.1697816692.git.raghavendra.kt@amd.com
should_we_balance is called for the decision to do load-balancing.
When sched ticks invoke this function, only one CPU should return
true. However, in the current code, two CPUs can return true. The
following situation, where b means busy and i means idle, is an
example, because CPU 0 and CPU 2 return true.
[0, 1] [2, 3]
b b i b
This fix checks if there exists an idle CPU with busy sibling(s)
after looking for a CPU on an idle core. If some idle CPUs with busy
siblings are found, just the first one should do load-balancing.
Fixes: b1bfeab9b0 ("sched/fair: Consider the idle state of the whole core for load balance")
Signed-off-by: Keisuke Nishimura <keisuke.nishimura@inria.fr>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Reviewed-by: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lkml.kernel.org/r/20231031133821.1570861-1-keisuke.nishimura@inria.fr
519fabc7aa ("psi: remove 500ms min window size limitation for
triggers") breaks unprivileged psi polling on cgroups.
Historically, we had a privilege check for polling in the open() of a
pressure file in /proc, but were erroneously missing it for the open()
of cgroup pressure files.
When unprivileged polling was introduced in d82caa2735 ("sched/psi:
Allow unprivileged polling of N*2s period"), it needed to filter
privileges depending on the exact polling parameters, and as such
moved the CAP_SYS_RESOURCE check from the proc open() callback to
psi_trigger_create(). Both the proc files as well as cgroup files go
through this during write(). This implicitly added the missing check
for privileges required for HT polling for cgroups.
When 519fabc7aa ("psi: remove 500ms min window size limitation for
triggers") followed right after to remove further restrictions on the
RT polling window, it incorrectly assumed the cgroup privilege check
was still missing and added it to the cgroup open(), mirroring what we
used to do for proc files in the past.
As a result, unprivileged poll requests that would be supported now
get rejected when opening the cgroup pressure file for writing.
Remove the cgroup open() check. psi_trigger_create() handles it.
Fixes: 519fabc7aa ("psi: remove 500ms min window size limitation for triggers")
Reported-by: Luca Boccassi <bluca@debian.org>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Luca Boccassi <bluca@debian.org>
Acked-by: Suren Baghdasaryan <surenb@google.com>
Cc: stable@vger.kernel.org # 6.5+
Link: https://lore.kernel.org/r/20231026164114.2488682-1-hannes@cmpxchg.org
vruntime of the (on_rq && !0-lag) entity needs to be adjusted when
it gets re-weighted, and the calculations can be simplified based
on the fact that re-weight won't change the w-average of all the
entities. Please check the proofs in comments.
But adjusting vruntime can also cause position change in RB-tree
hence require re-queue to fix up which might be costly. This might
be avoided by deferring adjustment to the time the entity actually
leaves tree (dequeue/pick), but that will negatively affect task
selection and probably not good enough either.
Fixes: 147f3efaa2 ("sched/fair: Implement an EEVDF-like scheduling policy")
Signed-off-by: Abel Wu <wuyun.abel@bytedance.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20231107090510.71322-2-wuyun.abel@bytedance.com
Pull probes fixes from Masami Hiramatsu:
- Documentation update: Add a note about argument and return value
fetching is the best effort because it depends on the type.
- objpool: Fix to make internal global variables static in
test_objpool.c.
- kprobes: Unify kprobes_exceptions_nofify() prototypes. There are the
same prototypes in asm/kprobes.h for some architectures, but some of
them are missing the prototype and it causes a warning. So move the
prototype into linux/kprobes.h.
- tracing: Fix to check the tracepoint event and return event at
parsing stage. The tracepoint event doesn't support %return but if
$retval exists, it will be converted to %return silently. This finds
that case and rejects it.
- tracing: Fix the order of the descriptions about the parameters of
__kprobe_event_gen_cmd_start() to be consistent with the argument
list of the function.
* tag 'probes-fixes-v6.7-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
tracing/kprobes: Fix the order of argument descriptions
tracing: fprobe-event: Fix to check tracepoint event and return
kprobes: unify kprobes_exceptions_nofify() prototypes
lib: test_objpool: make global variables static
Documentation: tracing: Add a note about argument and retval access
Pull dma-mapping fixes from Christoph Hellwig:
- don't leave pages decrypted for DMA in encrypted memory setups linger
around on failure (Petr Tesarik)
- fix an out of bounds access in the new dynamic swiotlb code (Petr
Tesarik)
- fix dma_addressing_limited for systems with weird physical memory
layouts (Jia He)
* tag 'dma-mapping-6.7-2023-11-10' of git://git.infradead.org/users/hch/dma-mapping:
swiotlb: fix out-of-bounds TLB allocations with CONFIG_SWIOTLB_DYNAMIC
dma-mapping: fix dma_addressing_limited() if dma_range_map can't cover all system RAM
dma-mapping: move dma_addressing_limited() out of line
swiotlb: do not free decrypted pages if dynamic
Fix to check the tracepoint event is not valid with $retval.
The commit 08c9306fc2 ("tracing/fprobe-event: Assume fprobe is
a return event by $retval") introduced automatic return probe
conversion with $retval. But since tracepoint event does not
support return probe, $retval is not acceptable.
Without this fix, ftracetest, tprobe_syntax_errors.tc fails;
[22] Tracepoint probe event parser error log check [FAIL]
----
# tail 22-tprobe_syntax_errors.tc-log.mRKroL
+ ftrace_errlog_check trace_fprobe t kfree ^$retval dynamic_events
+ printf %s t kfree
+ wc -c
+ pos=8
+ printf %s t kfree ^$retval
+ tr -d ^
+ command=t kfree $retval
+ echo Test command: t kfree $retval
Test command: t kfree $retval
+ echo
----
So 't kfree $retval' should fail (tracepoint doesn't support
return probe) but passed it.
Link: https://lore.kernel.org/all/169944555933.45057.12831706585287704173.stgit@devnote2/
Fixes: 08c9306fc2 ("tracing/fprobe-event: Assume fprobe is a return event by $retval")
Cc: stable@vger.kernel.org
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Pull networking fixes from Jakub Kicinski:
"Including fixes from netfilter and bpf.
Current release - regressions:
- sched: fix SKB_NOT_DROPPED_YET splat under debug config
Current release - new code bugs:
- tcp:
- fix usec timestamps with TCP fastopen
- fix possible out-of-bounds reads in tcp_hash_fail()
- fix SYN option room calculation for TCP-AO
- tcp_sigpool: fix some off by one bugs
- bpf: fix compilation error without CGROUPS
- ptp:
- ptp_read() should not release queue
- fix tsevqs corruption
Previous releases - regressions:
- llc: verify mac len before reading mac header
Previous releases - always broken:
- bpf:
- fix check_stack_write_fixed_off() to correctly spill imm
- fix precision tracking for BPF_ALU | BPF_TO_BE | BPF_END
- check map->usercnt after timer->timer is assigned
- dsa: lan9303: consequently nested-lock physical MDIO
- dccp/tcp: call security_inet_conn_request() after setting IP addr
- tg3: fix the TX ring stall due to incorrect full ring handling
- phylink: initialize carrier state at creation
- ice: fix direction of VF rules in switchdev mode
Misc:
- fill in a bunch of missing MODULE_DESCRIPTION()s, more to come"
* tag 'net-6.7-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (84 commits)
net: ti: icss-iep: fix setting counter value
ptp: fix corrupted list in ptp_open
ptp: ptp_read should not release queue
net_sched: sch_fq: better validate TCA_FQ_WEIGHTS and TCA_FQ_PRIOMAP
net: kcm: fill in MODULE_DESCRIPTION()
net/sched: act_ct: Always fill offloading tuple iifidx
netfilter: nat: fix ipv6 nat redirect with mapped and scoped addresses
netfilter: xt_recent: fix (increase) ipv6 literal buffer length
ipvs: add missing module descriptions
netfilter: nf_tables: remove catchall element in GC sync path
netfilter: add missing module descriptions
drivers/net/ppp: use standard array-copy-function
net: enetc: shorten enetc_setup_xdp_prog() error message to fit NETLINK_MAX_FMTMSG_LEN
virtio/vsock: Fix uninit-value in virtio_transport_recv_pkt()
r8169: respect userspace disabling IFF_MULTICAST
selftests/bpf: get trusted cgrp from bpf_iter__cgroup directly
bpf: Let verifier consider {task,cgroup} is trusted in bpf_iter_reg
net: phylink: initialize carrier state at creation
test/vsock: add dobule bind connect test
test/vsock: refactor vsock_accept
...
Pull RCU fixes from Frederic Weisbecker:
- Fix a lock inversion between scheduler and RCU introduced in
v6.2-rc4. The scenario could trigger on any user of RCU_NOCB
(mostly Android but also nohz_full)
- Fix PF_IDLE semantic changes introduced in v6.6-rc3 breaking
some RCU-Tasks and RCU-Tasks-Trace expectations as to what
exactly is an idle task. This resulted in potential spurious
stalls and warnings.
* tag 'rcu-fixes-v6.7' of git://git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-dynticks:
rcu/tasks-trace: Handle new PF_IDLE semantics
rcu/tasks: Handle new PF_IDLE semantics
rcu: Introduce rcu_cpu_online()
rcu: Break rcu_node_0 --> &rq->__lock order
Pull kgdb updates from Daniel Thompson:
"Just two patches for you this time!
- During a panic, flush the console before entering kgdb.
This makes things a little easier to comprehend, especially if an
NMI backtrace was triggered on all CPUs just before we enter the
panic routines
- Correcting a couple of misleading (a.k.a. plain wrong) comments"
* tag 'kgdb-6.7-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/danielt/linux:
kdb: Corrects comment for kdballocenv
kgdb: Flush console before entering kgdb on panic
BTF_TYPE_SAFE_TRUSTED(struct bpf_iter__task) in verifier.c wanted to
teach BPF verifier that bpf_iter__task -> task is a trusted ptr. But it
doesn't work well.
The reason is, bpf_iter__task -> task would go through btf_ctx_access()
which enforces the reg_type of 'task' is ctx_arg_info->reg_type, and in
task_iter.c, we actually explicitly declare that the
ctx_arg_info->reg_type is PTR_TO_BTF_ID_OR_NULL.
Actually we have a previous case like this[1] where PTR_TRUSTED is added to
the arg flag for map_iter.
This patch sets ctx_arg_info->reg_type is PTR_TO_BTF_ID_OR_NULL |
PTR_TRUSTED in task_reg_info.
Similarly, bpf_cgroup_reg_info -> cgroup is also PTR_TRUSTED since we are
under the protection of cgroup_mutex and we would check cgroup_is_dead()
in __cgroup_iter_seq_show().
This patch is to improve the user experience of the newly introduced
bpf_iter_css_task kfunc before hitting the mainline. The Fixes tag is
pointing to the commit introduced the bpf_iter_css_task kfunc.
Link[1]:https://lore.kernel.org/all/20230706133932.45883-3-aspsk@isovalent.com/
Fixes: 9c66dc94b6 ("bpf: Introduce css_task open-coded iterator kfuncs")
Signed-off-by: Chuyi Zhou <zhouchuyi@bytedance.com>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20231107132204.912120-2-zhouchuyi@bytedance.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
There is an unusual case that the range map covers right up to the top
of system RAM, but leaves a hole somewhere lower down. Then it prevents
the nvme device dma mapping in the checking path of phys_to_dma() and
causes the hangs at boot.
E.g. On an Armv8 Ampere server, the dsdt ACPI table is:
Method (_DMA, 0, Serialized) // _DMA: Direct Memory Access
{
Name (RBUF, ResourceTemplate ()
{
QWordMemory (ResourceConsumer, PosDecode, MinFixed,
MaxFixed, Cacheable, ReadWrite,
0x0000000000000000, // Granularity
0x0000000000000000, // Range Minimum
0x00000000FFFFFFFF, // Range Maximum
0x0000000000000000, // Translation Offset
0x0000000100000000, // Length
,, , AddressRangeMemory, TypeStatic)
QWordMemory (ResourceConsumer, PosDecode, MinFixed,
MaxFixed, Cacheable, ReadWrite,
0x0000000000000000, // Granularity
0x0000006010200000, // Range Minimum
0x000000602FFFFFFF, // Range Maximum
0x0000000000000000, // Translation Offset
0x000000001FE00000, // Length
,, , AddressRangeMemory, TypeStatic)
QWordMemory (ResourceConsumer, PosDecode, MinFixed,
MaxFixed, Cacheable, ReadWrite,
0x0000000000000000, // Granularity
0x00000060F0000000, // Range Minimum
0x00000060FFFFFFFF, // Range Maximum
0x0000000000000000, // Translation Offset
0x0000000010000000, // Length
,, , AddressRangeMemory, TypeStatic)
QWordMemory (ResourceConsumer, PosDecode, MinFixed,
MaxFixed, Cacheable, ReadWrite,
0x0000000000000000, // Granularity
0x0000007000000000, // Range Minimum
0x000003FFFFFFFFFF, // Range Maximum
0x0000000000000000, // Translation Offset
0x0000039000000000, // Length
,, , AddressRangeMemory, TypeStatic)
})
But the System RAM ranges are:
cat /proc/iomem |grep -i ram
90000000-91ffffff : System RAM
92900000-fffbffff : System RAM
880000000-fffffffff : System RAM
8800000000-bff5990fff : System RAM
bff59d0000-bff5a4ffff : System RAM
bff8000000-bfffffffff : System RAM
So some RAM ranges are out of dma_range_map.
Fix it by checking whether each of the system RAM resources can be
properly encompassed within the dma_range_map.
Signed-off-by: Jia He <justin.he@arm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
