[Bug 2086920] Re: [SRU] fio 3.36-1build2 in Ubuntu Noble crashes on riscv64
Simon Chopin
2086920 at bugs.launchpad.net
Wed Dec 11 11:17:40 UTC 2024
** Changed in: fio (Ubuntu Noble)
Status: New => In Progress
--
You received this bug notification because you are a member of Ubuntu
Sponsors, which is subscribed to the bug report.
https://bugs.launchpad.net/bugs/2086920
Title:
[SRU] fio 3.36-1build2 in Ubuntu Noble crashes on riscv64
Status in fio package in Ubuntu:
Fix Released
Status in fio source package in Noble:
In Progress
Bug description:
[ Impact ]
fio 3.36-1build2 in Ubuntu Noble crashes on riscv64
$ fio -name test --size=1M --rw=read --bs=4k
test: (g=0): rw=read, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=psync, iodepth=1
fio-3.36
Illegal instruction (core dumped)
apport-retrace --gdb shows that the crash occurs when executing
rdcycle a3
This is the call stack:
#4 0x0000555e549d3a74 in fio_time_init () at time.c:149
#3 0x0000555e549c1dc0 in fio_clock_init () at gettime.c:416
#2 0x0000555e549c106e in calibrate_cpu_clock () at gettime.c:268
#1 get_cycles_per_msec () at gettime.c:246
#0 get_cpu_clock () at arch/arch-riscv64.h:19
A patch resolving the issue is available at
https://github.com/axboe/fio/commit/577b85ec41b5560a81a4fb09d1ec7760bcacc49f
ProblemType: Bug
DistroRelease: Ubuntu 24.04
Package: fio 3.36-1build2
Uname: Linux 6.6.21-5-premier riscv64
ApportVersion: 2.28.1-0ubuntu3.1
Architecture: riscv64
CasperMD5CheckResult: unknown
CloudArchitecture: riscv64
CloudBuildName: server
CloudID: nocloud
CloudName: unknown
CloudPlatform: nocloud
CloudSerial: 20240423
CloudSubPlatform: seed-dir (/var/lib/cloud/seed/nocloud-net)
Date: Thu Nov 7 17:37:00 2024
SourcePackage: fio
UpgradeStatus: No upgrade log present (probably fresh install)
[ Test Plan ]
On a riscv64 system run
fio -name test --size=1M --rw=read --bs=4k
This should provide output similar too
$ fio -name test --size=1M --rw=read --bs=4k
test: (g=0): rw=read, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=psync, iodepth=1
fio-3.36
Starting 1 process
test: Laying out IO file (1 file / 1MiB)
test: (groupid=0, jobs=1): err= 0: pid=2593: Thu Nov 7 17:42:07 2024
read: IOPS=36.6k, BW=143MiB/s (150MB/s)(1024KiB/7msec)
clat (usec): min=7, max=672, avg=21.28, stdev=74.34
lat (usec): min=7, max=673, avg=21.74, stdev=74.43
clat percentiles (usec):
| 1.00th=[ 8], 5.00th=[ 8], 10.00th=[ 8], 20.00th=[ 9],
| 30.00th=[ 9], 40.00th=[ 9], 50.00th=[ 9], 60.00th=[ 9],
| 70.00th=[ 9], 80.00th=[ 9], 90.00th=[ 10], 95.00th=[ 30],
| 99.00th=[ 441], 99.50th=[ 611], 99.90th=[ 676], 99.95th=[ 676],
| 99.99th=[ 676]
lat (usec) : 10=93.75%, 20=1.17%, 50=0.78%, 100=1.17%, 500=2.34%
lat (usec) : 750=0.78%
cpu : usr=0.00%, sys=83.33%, ctx=2, majf=0, minf=26
IO depths : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
issued rwts: total=256,0,0,0 short=0,0,0,0 dropped=0,0,0,0
latency : target=0, window=0, percentile=100.00%, depth=1
Run status group 0 (all jobs):
READ: bw=143MiB/s (150MB/s), 143MiB/s-143MiB/s (150MB/s-150MB/s), io=1024KiB (1049kB), run=7-7msec
Disk stats (read/write):
sda: ios=0/0, sectors=0/0, merge=0/0, ticks=0/0, in_queue=0, util=0.00%
[ Where problems could occur ]
The suggested patch only affects riscv64. It affects the measurement
of time.
This could result in inaccurate measurements.
[ Other Info ]
n/a
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/fio/+bug/2086920/+subscriptions
More information about the Ubuntu-sponsors
mailing list