lkml.org 
[lkml]   [2016]   [Jul]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH 06/16] gpu: ipu-v3: Add ipu_set_vdi_src_mux()
From
Date


On 07/15/2016 05:48 AM, Philipp Zabel wrote:
> Am Donnerstag, den 07.07.2016, 16:03 -0700 schrieb Steve Longerbeam:
>> Adds ipu_set_vdi_src_mux() that selects the VDIC input
>> (from CSI or memory).
>>
>> Signed-off-by: Steve Longerbeam <steve_longerbeam@mentor.com>
>> ---
>> drivers/gpu/ipu-v3/ipu-common.c | 20 ++++++++++++++++++++
>> include/video/imx-ipu-v3.h | 1 +
>> 2 files changed, 21 insertions(+)
>>
>> diff --git a/drivers/gpu/ipu-v3/ipu-common.c b/drivers/gpu/ipu-v3/ipu-common.c
>> index 6d1676e..374100e 100644
>> --- a/drivers/gpu/ipu-v3/ipu-common.c
>> +++ b/drivers/gpu/ipu-v3/ipu-common.c
>> @@ -730,6 +730,26 @@ void ipu_set_ic_src_mux(struct ipu_soc *ipu, int csi_id, bool vdi)
>> }
>> EXPORT_SYMBOL_GPL(ipu_set_ic_src_mux);
>>
>> +/*
>> + * Set the source for the VDIC. Selects either from CSI[01] or memory.
>> + */
>> +void ipu_set_vdi_src_mux(struct ipu_soc *ipu, bool csi)
>> +{
>> + unsigned long flags;
>> + u32 val;
>> +
>> + spin_lock_irqsave(&ipu->lock, flags);
>> +
>> + val = ipu_cm_read(ipu, IPU_FS_PROC_FLOW1);
>> + val &= ~(0x3 << 28);
>> + if (csi)
>> + val |= (0x01 << 28);
> This doesn't set a bus mux, or does it? As I understand it, this
> configures the FSU to trigger the VDIC conversion from the CSI direct
> channel sync signals (IC channel CB7). As such I believe this belongs
> together with the other channel linking code, maybe even in ipu-fsu.c.

It can't be folded into ipu_idmac_link(), since that links idmac channels.

I don't know if this should be characterized as a mux either.
But it does seem to have a similar function to setting the IC
source (ipu_set_ic_src_mux()), which configures the FSU to
trigger the IC conversion to start from CSI or VDIC sync signals.

But it is a bit of a mystery why the VDIC source bits were placed in
IPU_FS_PROC_FLOW1 instead of IPU_CONF.


>
> Also, some #defines would be nice, for example:
>
> #define FS_VDI_SRC_SEL_MASK (0x3 << 28)
> #define FS_VDI_SRC_SEL_OFFSET 28
> #define FS_VDI_SRC_SEL_CSI_DIRECT 0x1
> #define FS_VDI_SRC_SEL_VDOA 0x2

Ok I will do that in next version.

Steve

\
 
 \ /
  Last update: 2016-07-16 00:01    [W:0.066 / U:1.932 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site