[dpdk-dev] [PATCH v2 0/6] NXP DPAA2: Refactor bus scan/probe code

Shreyansh Jain shreyansh.jain at nxp.com
Mon Sep 11 16:25:53 CEST 2017


On Monday 11 September 2017 07:36 PM, Ferruh Yigit wrote:
> On 8/25/2017 11:19 AM, Shreyansh Jain wrote:
>> Change Log:
>> ~~~~~~~~~~~
>>   v2:
>>     - Minor updates for logging (removed some logs and changed others
>>       to make it cleaner when application starts)
>>
>> Brief:
>> ~~~~~~
>>
>>   -- v1 is at [3] --
>>
>> In [1], during the IOVA Mapping patch set [2] discussion, it was
>> observed that DPAA2 scan was actually doing work meant for probing.
>>
>> This patchset demarcates the roles of FSLMC bus scan and probe
>> functions much more clearly than before:
>>
>> 1. scan now only add devices into a list
>>   unlike previously, scan doesn't initialize the devices using the VFIO
>>   operations. Now, scan would only add the devices onto a local device
>>   list after marking their type and filling in device name.
>>
>> 2. probe would now perform VFIO operations
>>   in dpaa2, for the device added in list, an initialization needs
>>   to be done so as to enable the devices - before actually API calls
>>   can be served. Probe function now initializes the devices as well
>>   as links then to the Eth/Crypto drivers.
>>
>> 3. Refactoring some VFIO code
>>   This patch improves the overall code contained within DPAA2 bus
>>   for DPAA2 VFIO layer.
>>
>> [1] http://dpdk.org/ml/archives/dev/2017-July/071270.html
>> [2] http://dpdk.org/ml/archives/dev/2017-July/070833.html
>> [3] http://dpdk.org/ml/archives/dev/2017-August/073011.html
>>
>> Shreyansh Jain (6):
>>    bus/fslmc: support only single group and container
>>    bus/fslmc: introduce new device type enumerator
>>    crypto/dpaa2_sec: update driver type field
>>    net/dpaa2: update driver type field
>>    drivers: refactor DPAA2 object definition
>>    bus/fslmc: refactor scan and probe functions
> 
> Series Reviewed-by: Ferruh Yigit <ferruh.yigit at intel.com>
> 
> I guess this set needs to go in to the main tree, because of both
> - as far as I get IOVA patchset breaks dpaa2, so this should go in
> before that patchset
> - This patchset has both crypto and net patches, hard to get by sub-trees
> 

I agree.
I don't think much patches have come on 17.11 tree yet - still, I can 
send a rebased set again.
Thanks a lot for taking out time for this.

-
Shreyansh


More information about the dev mailing list