[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <91dd06dd-22bd-c50a-4a6f-f45334b13b4a@oracle.com>
Date: Mon, 24 Oct 2016 13:36:54 -0700
From: tndave <tushar.n.dave@...cle.com>
To: David Miller <davem@...emloft.net>
Cc: chris.hyser@...cle.com, sowmini.varadhan@...cle.com,
vgupta@...opsys.com, robin.murphy@....com, jroedel@...e.de,
egtvedt@...fundet.no, krzk@...nel.org, rkuo@...eaurora.org,
sparclinux@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 3/6] sparc64: Initialize iommu_map_table and iommu_pool
On 10/24/2016 10:47 AM, David Miller wrote:
> From: Tushar Dave <tushar.n.dave@...cle.com>
> Date: Mon, 10 Oct 2016 11:12:04 -0700
>
>> Like legacy IOMMU, use common iommu_map_table and iommu_pool for ATU.
>> This change initializes iommu_map_table and iommu_pool for ATU.
>>
>> Signed-off-by: Tushar Dave <tushar.n.dave@...cle.com>
>> Reviewed-by: chris hyser <chris.hyser@...cle.com>
>> Reviewed-by: Sowmini Varadhan <sowmini.varadhan@...cle.com>
>
> I suspect you are adding a bisection problem here.
>
> It looks to me like patch #2 starts setting up to use the ATU
> but these changes in #3 and later are necessary for it to even
> work properly.
>
> You can't leave the tree in an intermediate non-working state
> between changes like that.
I agree but even if you apply only patch #1, #2 things still work using
legacy IOMMU. (and the same is true for rest of the other patches i.e.
#3, #4, #5).
All the patches, except patch #6 , sets up needed data structures and
functions for ATU however ATU will only be used in action after you
apply patch #6 which actually enables 64bit DMA. Until then, things are
going to work using legacy IOMMU with 32bit DMA.
Thanks.
-Tushar
>
Powered by blists - more mailing lists