Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH v5 10/11] powerpc/pseries/iommu: Make use of DDW for
 indirect mapping
To:     Alexey Kardashevskiy <aik@ozlabs.ru>,
        Leonardo Bras <leobras.c@gmail.com>,
        Michael Ellerman <mpe@ellerman.id.au>,
        Benjamin Herrenschmidt <benh@kernel.crashing.org>,
        Paul Mackerras <paulus@samba.org>,
        David Gibson <david@gibson.dropbear.id.au>,
        kernel test robot <lkp@intel.com>,
        Nicolin Chen <nicoleotsuka@gmail.com>
Cc:     linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org
References: <20210716082755.428187-1-leobras.c@gmail.com>
 <20210716082755.428187-11-leobras.c@gmail.com>
 <b98f696a-ed64-4c9e-ccb6-549ae8bc7fd6@linux.ibm.com>
 <8dfb28d5-b654-746c-03d8-aeee3d438240@ozlabs.ru>
From:   Frederic Barrat <fbarrat@linux.ibm.com>
Message-ID: <994051df-73b3-4dad-76aa-1a03d9afaf6d@linux.ibm.com>
Date:   Wed, 21 Jul 2021 17:04:34 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.11.0
MIME-Version: 1.0
In-Reply-To: <8dfb28d5-b654-746c-03d8-aeee3d438240@ozlabs.ru>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Precedence: bulk


On 21/07/2021 05:32, Alexey Kardashevskiy wrote:
>>> +        struct iommu_table *newtbl;
>>> +        int i;
>>> +
>>> +        for (i = 0; i < ARRAY_SIZE(pci->phb->mem_resources); i++) {
>>> +            const unsigned long mask = IORESOURCE_MEM_64 | 
>>> IORESOURCE_MEM;
>>> +
>>> +            /* Look for MMIO32 */
>>> +            if ((pci->phb->mem_resources[i].flags & mask) == 
>>> IORESOURCE_MEM)
>>> +                break;
>>> +        }
>>> +
>>> +        if (i == ARRAY_SIZE(pci->phb->mem_resources))
>>> +            goto out_del_list;
>>
>>
>> So we exit and do nothing if there's no MMIO32 bar?
>> Isn't the intent just to figure out the MMIO32 area to reserve it when 
>> init'ing the table? In which case we could default to 0,0
>>
>> I'm actually not clear why we are reserving this area on pseries.
> 
> 
> 
> If we do not reserve it, then the iommu code will allocate DMA pages 
> from there and these addresses are MMIO32 from the kernel pov at least. 
> I saw crashes when (I think) a device tried DMAing to the top 2GB of the 
> bus space which happened to be a some other device's BAR.


hmmm... then figuring out the correct range needs more work. We could 
have more than one MMIO32 bar. And they don't have to be adjacent. I 
don't see that we are reserving any range on the initial table though 
(on pseries).

   Fred