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;
Message-ID: <03e82e1a1bcf516d01ca472546d8b31e468aba8b.camel@gmail.com>
Subject: Re: [PATCH 4/4] powerpc/pseries/iommu: Remove default DMA window
 before creating DDW
From:   Leonardo Bras <leobras.c@gmail.com>
To:     Alexey Kardashevskiy <aik@ozlabs.ru>,
        Michael Ellerman <mpe@ellerman.id.au>,
        Benjamin Herrenschmidt <benh@kernel.crashing.org>,
        Paul Mackerras <paulus@samba.org>,
        Thiago Jung Bauermann <bauerman@linux.ibm.com>,
        Ram Pai <linuxram@us.ibm.com>
Cc:     linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org
Date:   Mon, 22 Jun 2020 23:43:24 -0300
In-Reply-To: <cfb197e1-c608-71f9-5c98-c120a3496266@ozlabs.ru>
References: <20200619050619.266888-1-leobras.c@gmail.com>
         <20200619050619.266888-5-leobras.c@gmail.com>
         <ade15776-61d1-b660-db74-7aeba4eddfdf@ozlabs.ru>
         <4bf1d32da3d13a44e3c2e4b04f369fe52c24a023.camel@gmail.com>
         <afd1c5ac-d291-5281-1592-a345ee3c0c8c@ozlabs.ru>
         <205edd45b7bbf39d2fc1d2d75fd7e35336f109ac.camel@gmail.com>
         <cfb197e1-c608-71f9-5c98-c120a3496266@ozlabs.ru>
Organization: IBM
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.34.4 (3.34.4-1.fc31) 
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, 2020-06-23 at 12:35 +1000, Alexey Kardashevskiy wrote:
> > I am not sure if this is true in general, but in this device (SR-IOV
> > VF) I am testing it will return 0 windows if the default DMA window is
> > not deleted, and 1 after it's deleted.
> 
> Since pHyp can only create windows in "64bit space", now (I did not know
> until a few month back) I expect that thing to return "1" always no
> matter what happened to the default window. And removing the default
> window will only affect the maximum number of TCEs but not the number of
> possible windows.

Humm, something gone wrong then.

This patchset was developed mostly because on testing, DDW would never
be created because windows_available would always be 0.

I will take a deeper look in that.

Best regards,
Leonardo