Re: [bug report] rdma_rxe: WARNING: inconsistent lock state, inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage

[Bart Van Assche <bvanassche@acm.org>]

On 4/7/22 14:06, Bob Pearson wrote:
> On 4/7/22 14:49, Bart Van Assche wrote:
>> On 4/7/22 12:15, Robert Pearson wrote:
>>> I would say it is very possible. There was a period when the
>>> pool locks were switched to _bh spinlocks but that was later
>>> reversed back to _irqsave locks which cleaned up some failures. I
>>> don't know which version Yi Zhang was using. The root cause of
>>> this bug was caused by librdmacm making verbs API calls while
>>> holding _irqsave locks which I didn't figure out until later.
>> 
>> I can reproduce the issue Yi reported with kernel v5.18-rc1. Please
>> let me know if you need any additional information to reproduce
>> this issue.
> 
> From your note I can't figure out what you are running. Can you give
> me more details about the failing test case?

Hi Bob,

In Yi's email I found the following:

run blktests srp/001

That test case can be run as follows:

git clone https://github.com/osandov/blktests && cd blktests && make && 
sudo ./check -q srp/001

Thanks,

Bart.