[dpdk-dev] [PATCH] crypto/dpaa2_sec: remove iova conversion for fle address

Shreyansh Jain shreyansh.jain at nxp.com
Wed May 9 14:29:14 CEST 2018


On Wednesday 09 May 2018 05:00 PM, De Lara Guarch, Pablo wrote:
> Hi,
> 
>> -----Original Message-----
>> From: Shreyansh Jain [mailto:shreyansh.jain at nxp.com]
>> Sent: Wednesday, May 9, 2018 12:40 PM
>> To: De Lara Guarch, Pablo <pablo.de.lara.guarch at intel.com>
>> Cc: Akhil Goyal <akhil.goyal at nxp.com>; dev at dpdk.org; Hemant Agrawal
>> <hemant.agrawal at nxp.com>
>> Subject: Re: [dpdk-dev] [PATCH] crypto/dpaa2_sec: remove iova conversion for
>> fle address
>>
>> On Wednesday 09 May 2018 03:54 PM, De Lara Guarch, Pablo wrote:
>>>
>>>
>>>> -----Original Message-----
>>>> From: Akhil Goyal [mailto:akhil.goyal at nxp.com]
>>>> Sent: Wednesday, May 9, 2018 10:14 AM
>>>> To: Shreyansh Jain <shreyansh.jain at nxp.com>; De Lara Guarch, Pablo
>>>> <pablo.de.lara.guarch at intel.com>
>>>> Cc: akhil.goyal at nxp.com; dev at dpdk.org; Hemant Agrawal
>>>> <hemant.agrawal at nxp.com>
>>>> Subject: Re: [dpdk-dev] [PATCH] crypto/dpaa2_sec: remove iova
>>>> conversion for fle address
>>>>
>>>> On 5/9/2018 2:56 PM, Shreyansh Jain wrote:
>>>>> From: Hemant Agrawal <hemant.agrawal at nxp.com>
>>>>>
>>>>> fle is already in virtual addressing mode - no need to perform
>>>>> address conversion for it.
>>>>>
>>>>> Fixes: 8d1f3a5d751b ("crypto/dpaa2_sec: support crypto operation")
>>>>> Cc: akhil.goyal at nxp.com
>>>>>
>>>>> Signed-off-by: Hemant Agrawal <hemant.agrawal at nxp.com>
>>>
>>> ...
>>>
>>>>>
>>>> Acked-by: Akhil Goyal <akhil.goyal at nxp.com>
>>>
>>> Added Cc: stable at dpdk.org (try remembering this for next time, please).
>>
>> Thanks for doing it this time. I'll take care of this next time.
> 
> No worries. Actually, gcc 32 bits is broken:
> 
> drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c:1264:7:
> error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]
>    op = (struct rte_crypto_op *)DPAA2_GET_FLE_ADDR((fle - 1));
> 
> Is this PMD supposed to be compatible for 32-bit targets?

Oh! This is bad. I will send you an update on this in a few minutes.
Ideally, the code was written for 64bit in mind - though, it should 
still be able to work (at least compile) for 32 bit.

> 
> Thanks,
> Pablo
> 
>>
>>>
>>> Applied to dpdk-next-crypto.
>>> Thanks,
>>>
>>> Pablo
>>>
> 



More information about the dev mailing list