On 3/1/21 5:56 AM, Anatoly Pugachev wrote:
On Mon, Mar 1, 2021 at 9:09 AM Rob Gardner <rob.gardner@xxxxxxxxxx> wrote:
is_no_fault_exception() has two bugs which were discovered via random
opcode testing with stress-ng. Both are caused by improper filtering
of opcodes.
Rob, tested on my ldom, works perfectly now...
$ uname -a
Linux ttip 5.12.0-rc1-dirty #195 SMP Mon Mar 1 15:46:15 MSK 2021
sparc64 GNU/Linux
$ stress-ng --opcode 1 --timeout 60 --metrics-brief
stress-ng: info: [945] dispatching hogs: 1 opcode
stress-ng: info: [945] successful run completed in 60.00s (1 min, 0.00 secs)
stress-ng: info: [945] stressor bogo ops real time usr time
sys time bogo ops/s bogo ops/s
stress-ng: info: [945] (secs) (secs)
(secs) (real time) (usr+sys time)
stress-ng: info: [945] opcode 17847 60.00 27.45
34.03 297.45 290.29
Thank you for a quick fix.
You're welcome. Please add your "tested-by" if you like.
Rob