Received-SPF: pass (google.com: domain of linux-kernel+bounces-131858-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
Message-ID: <e565670b-d31e-48fa-a873-6c68dba5b04b@linuxfoundation.org>
Date: Thu, 4 Apr 2024 10:54:12 -0600
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [BUG] seltests/iommu: runaway ./iommufd consuming 99% CPU after a
 failed assert()
To: Mirsad Todorovac <mirsad.todorovac@alu.unizg.hr>,
 Joao Martins <joao.m.martins@oracle.com>, Jason Gunthorpe <jgg@nvidia.com>
Cc: iommu@lists.linux.dev, Kevin Tian <kevin.tian@intel.com>,
 Shuah Khan <shuah@kernel.org>, linux-kselftest@vger.kernel.org,
 linux-kernel@vger.kernel.org, Shuah Khan <skhan@linuxfoundation.org>
References: <0d5c3b29-fc5b-41d9-9556-5ce94262dac8@alu.unizg.hr>
 <20240319135852.GA393211@nvidia.com>
 <a692d5d7-11d5-4c1b-9abc-208d2194ccde@alu.unizg.hr>
 <cdc9c46b-1bad-41cd-8f98-38cc2171186a@oracle.com>
 <20240325135207.GC6245@nvidia.com>
 <f8e03425-98cf-4076-8959-d85eda846bab@oracle.com>
 <08a13b79-e94a-4f80-96e9-ce223d928b3e@alu.unizg.hr>
Content-Language: en-US
From: Shuah Khan <skhan@linuxfoundation.org>
In-Reply-To: <08a13b79-e94a-4f80-96e9-ce223d928b3e@alu.unizg.hr>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

On 3/27/24 14:04, Mirsad Todorovac wrote:
> 
> 
> On 3/27/24 11:41, Joao Martins wrote:
>> On 25/03/2024 13:52, Jason Gunthorpe wrote:
>>> On Mon, Mar 25, 2024 at 12:17:28PM +0000, Joao Martins wrote:
>>>>> However, I am not smart enough to figure out why ...
>>>>>
>>>>> Apparently, from the source, mmap() fails to allocate pages on the desired address:
>>>>>
>>>>>    1746         assert((uintptr_t)self->buffer % HUGEPAGE_SIZE == 0);
>>>>>    1747         vrc = mmap(self->buffer, variant->buffer_size, PROT_READ |
>>>>> PROT_WRITE,
>>>>>    1748                    mmap_flags, -1, 0);
>>>>> → 1749         assert(vrc == self->buffer);
>>>>>    1750
>>>>>
>>>>> But I am not that deep into the source to figure our what was intended and what
>>>>> went
>>>>> wrong :-/
>>>>
>>>> I can SKIP() the test rather assert() in here if it helps. Though there are
>>>> other tests that fail if no hugetlb pages are reserved.
>>>>
>>>> But I am not sure if this is problem here as the initial bug email had an
>>>> enterily different set of failures? Maybe all you need is an assert() and it
>>>> gets into this state?
>>>
>>> I feel like there is something wrong with the kselftest framework,
>>> there should be some way to fail the setup/teardown operations without
>>> triggering an infinite loop :(
>>
>> I am now wondering if the problem is the fact that we have an assert() in the
>> middle of FIXTURE_{TEST,SETUP} whereby we should be having ASSERT_TRUE() (or any
>> other kselftest macro that). The expect/assert macros from kselftest() don't do
>> asserts and it looks like we are failing mid tests in the assert().
>>
>> Maybe it is OK for setup_sizes(), but maybe not OK for the rest (i.e. during the
>> actual setup / tests). I can throw a patch there to see if this helps Mirsad.
> 
> Well, we are in the job of making the kernel better and as bug free as we can.
> 
> Maybe we should not delve too much into detail: is this a kernel bug, or the kselftest
> program bug?
> 
> Some people already mentioned that I might have sysctl variable problems. I don't see
> what the mmap() HUGEPAGE allocation at fixed address was meant to prove?
> 

I applied fix to this problem to linux-kselftest fixes branch for next rc.
Please give it a try.

thanks,
-- Shuah