This repository has been archived by the owner on Jun 18, 2024. It is now read-only.
Implement ops.dump(), dump_cpu() and dump_task().
#195
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Implement ops.dump(), dump_cpu() and dump_task() which allows the BPF
scheduler to include scheduler specific information in debug dump.
The below example output from scx_qmap shows full dump of qmap's FIFO queues
and per-CPU and per-task contexts.
$ scx_qmap -e 3
stats : enq=9 dsp=0 delta=9 reenq=5 deq=9 core=0 exp=0
cpuperf: cur min/avg/max=0/0/0 target min/avg/max=0/0/0
DEBUG DUMP
================================================================================
scx_qmap[1627] triggered exit kind 1025:
scx_bpf_error (test triggering error)
Backtrace:
scx_bpf_error_bstr+0x60/0x80
bpf_prog_5ab320c9dc9e846d_qmap_enqueue+0xd1/0x283
QMAP FIFO[0]:
QMAP FIFO[1]:
QMAP FIFO[2]: 1581 1583 1584 1585 1586 1588 1593 1594 1595
QMAP FIFO[3]:
QMAP FIFO[4]:
CPU states
----------
CPU 0 : nr_run=2 flags=0x0 cpu_rel=1 ops_qseq=4 pnt_seq=12
curr=stress[1590] class=ext_sched_class
QMAP: dsp_idx=0 dsp_cnt=0 avg_weight=0 cpuperf_target=0
*R stress[1590] +0ms
scx_state/flags=3/0x5 dsq_flags=0x0 ops_state/qseq=0/0
sticky/holding_cpu=-1/-1 dsq_id=(n/a)
cpus=ff
QMAP: force_local=0 core_sched_seq=0
R stress[1585] +0ms
scx_state/flags=3/0x1 dsq_flags=0x0 ops_state/qseq=2/3
sticky/holding_cpu=-1/-1 dsq_id=(n/a)
cpus=ff
QMAP: force_local=0 core_sched_seq=3
irqentry_exit+0x31/0x80
sysvec_apic_timer_interrupt+0x44/0x80
asm_sysvec_apic_timer_interrupt+0x1b/0x20
...
Byte-Lab
approved these changes
May 2, 2024
Comment on lines
4524
to
+4526
| seq_buf_printf(s, "\n %c%c %s[%d] %+ldms\n", | ||
| marker, task_state_to_char(p), p->comm, p->pid, | ||
| jiffies_delta_msecs(p->scx.runnable_at, now)); | ||
| jiffies_delta_msecs(p->scx.runnable_at, dctx->at_jiffies)); |
There was a problem hiding this comment.
A bit tangential to this change, but would it be useful to also print p->scx.dsq_vtime? If a task hasn't been scheduled for a while, it seems like vtime could be a common culprit. Though with this change, maybe that's not necessary anymore.
There was a problem hiding this comment.
Yeah, I'm gonna do more work on the dump stuff. Will add vtime.
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Implement ops.dump(), dump_cpu() and dump_task() which allows the BPF
scheduler to include scheduler specific information in debug dump.
The below example output from scx_qmap shows full dump of qmap's FIFO queues
and per-CPU and per-task contexts.