Re: [herbert-cryptodev-2.6:master] [rhashtable] e1d3422c95: stress-ng.syscall.ops_per_sec 98.9% regression

[Tejun Heo <tj@kernel.org>]

On Fri, Dec 27, 2024 at 11:10:11AM +0800, kernel test robot wrote:
> 
> 
> Hello,
> 
> kernel test robot noticed a 98.9% regression of stress-ng.syscall.ops_per_sec on:
> 
> 
> commit: e1d3422c95f003eba241c176adfe593c33e8a8f6 ("rhashtable: Fix potential deadlock by moving schedule_work outside lock")
> https://git.kernel.org/cgit/linux/kernel/git/herbert/cryptodev-2.6.git master
> 
> testcase: stress-ng
> config: x86_64-rhel-9.4
> compiler: gcc-12
> test machine: 224 threads 2 sockets Intel(R) Xeon(R) Platinum 8480CTDX (Sapphire Rapids) with 256G memory
> parameters:
> 
> 	nr_threads: 100%
> 	testtime: 60s
> 	test: syscall
> 	cpufreq_governor: performance

Hmm... the only meaningful behavior difference would be that after the
patch, rht_grow_above_75() test is done regardless of the return value while
before it was done only when the return value is zero. Breno, can you please
look into whether this report is valid and whether restoring the NULL check
makes it go away?

Thanks.

-- 
tejun