[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Performance measurement
From: |
sterbl . ta |
Subject: |
Performance measurement |
Date: |
Wed, 1 Apr 2020 23:33:26 +0200 (CEST) |
Sorry I am not sure if this is the right place to ask but this is the only mailing list that most relevant. Apologize if this is not, and please let me know which place I should check. Thank you.
So I try replacing qemu startup command in How to use section with mine below
qemu-system-x86_64 -enable-kvm \
-drive "file=disk.img,if=virtio" -drive "file=seed.img,if=virtio,format=raw" \
-device virtio-net-pci,netdev=net00 --netdev type=user,id=net00 \
-m 512 -nographic
perf record command generates qemu_perf.data file without a problem.
But the problem raises when trying to run the command
perf script -s qemu-boot-time/perf-script/qemu-perf-script.py -i \
qemu_perf.data
It throws
in trace_begin
Traceback (most recent call last):
File "qemu-boot-time/perf-script/qemu-perf-script.py", line 195, in kvm__kvm_entry
if (events.traces[pid].qemu_init_end != 0):
AttributeError: 'collections.defaultdict' object has no attribute 'qemu_init_end'
Fatal Python error: problem in Python trace event handler
Even after export env variable PERF_EXEC_PATH to /path/to/linux-source-5.4/tools/perf (where Core.py, Utils.py files located in linux kernel source 5.4)
So my questions:
How can I fix this?
Otherwise in qemu, any command or anyway that I can measure the boot time of a specific image created?
Many thanks for advice.
- Performance measurement,
sterbl . ta <=