linux-kernel - Re: [PATCH 02/16] gpu: ipu-cpmem: Add ipu_cpmem_set_uv

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Wed, 13 Jul 2016 15:54:18 -0700
From:	Steve Longerbeam <steve_longerbeam@...tor.com>
To:	Philipp Zabel <p.zabel@...gutronix.de>,
	Steve Longerbeam <slongerbeam@...il.com>
CC:	<dri-devel@...ts.freedesktop.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 02/16] gpu: ipu-cpmem: Add ipu_cpmem_set_uv_offset()

Hi Philipp,

On 07/08/2016 10:34 AM, Philipp Zabel wrote:
> Am Donnerstag, den 07.07.2016, 16:03 -0700 schrieb Steve Longerbeam:
>> Adds ipu_cpmem_set_uv_offset(), to set planar U/V offsets.
>>
>> Signed-off-by: Steve Longerbeam <steve_longerbeam@...tor.com>
>> ---
>>  drivers/gpu/ipu-v3/ipu-cpmem.c | 7 +++++++
>>  include/video/imx-ipu-v3.h     | 1 +
>>  2 files changed, 8 insertions(+)
>>
>> diff --git a/drivers/gpu/ipu-v3/ipu-cpmem.c b/drivers/gpu/ipu-v3/ipu-cpmem.c
>> index 6494a4d..a36c35e 100644
>> --- a/drivers/gpu/ipu-v3/ipu-cpmem.c
>> +++ b/drivers/gpu/ipu-v3/ipu-cpmem.c
>> @@ -253,6 +253,13 @@ void ipu_cpmem_set_buffer(struct ipuv3_channel *ch, int bufnum, dma_addr_t buf)
>>  }
>>  EXPORT_SYMBOL_GPL(ipu_cpmem_set_buffer);
>>  
>> +void ipu_cpmem_set_uv_offset(struct ipuv3_channel *ch, u32 u_off, u32 v_off)
>> +{
>> +	ipu_ch_param_write_field(ch, IPU_FIELD_UBO, u_off / 8);
>> +	ipu_ch_param_write_field(ch, IPU_FIELD_VBO, v_off / 8);
>> +}
>> +EXPORT_SYMBOL_GPL(ipu_cpmem_set_uv_offset);
>> +
> I'd prefer to avoid too much duplication in the API. Is there a reason
> you can't use ipu_cpmem_set_yuv_planar_full?

I could, but it would be cumbersome. ipu_cpmem_set_uv_offset() is being
called when kicking off a new tile conversion, and the stride length would
have to be calculated all over again if I were to use ipu_cpmem_set_yuv_planar_ful().
When kicking off a new tile conversion, the stride doesn't change, only the
tile's buffer address, and U/V offsets for planar, need to be updated.

Steve